DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This is a Final Action on the Merits. Claims 1-6, 8-16, and 18-22 are currently pending and are addressed below.
Response to Amendments
The amendment filed on February 3rd, 2026 has been considered and entered. Accordingly, claims 1, 12, and 20 have been amended.
Response to Arguments
The applicant’s arguments with respect to claims 1-6, 8-16, and 18-22 have been considered but are moot in view of the newly formulated grounds of rejections necessitated by the applicant’s amendments.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 6, 8, 12, 18, and 20-22 are rejected under 35 U.S.C. 103 as being unpatentable over Lerner (US 20200349833 A1) (“Lerner”) in view of Fowe (US 20200286372 A1) (“Fowe”) in view of Zhu (US 10754339 B2) (“Zhu”) in view of Mulcahy (US 20200143669 A1) (“Mulcahy”).
With respect to claim 1, Lerner teaches a computer-implemented method for generating speed profiles for autonomous vehicles, the computer-implemented method comprising:
identifying a road segment associated with a safety risk1 profile based at least in part on road condition data related to the road segment (See at least Lerner FIG. 4 “405” and Paragraph 64 “The process 400 beings at operation 405, where multiple vehicle sensors are employed to continuously monitor the current road conditions, in real-time. The monitoring of operation 405 can involve the sensors capturing real-time data pertaining to a driving environment surrounding a vehicle, such as a portion of a roadway. Thus, the real-time data can be analyzed to determine, or otherwise estimate, real-time conditions in which the vehicle is currently traveling. The road conditions detected at operation 405 may impact an appropriate speed for the vehicle, such as traffic speed, traffic volume, congestion, weather conditions, and the like. Alternatively, monitoring at operation 405 can be triggered by an event (rather than continuous), such as a time interval or traveling in a certain location. As a result of monitoring at operation 405, the vehicle has an awareness of the current, or real-time, road conditions, including any significant changes.”);
obtaining probe data from one or more probe apparatuses traveling along the road segment during an interval of time (See at least Lerner FIG. 4 “415” and Paragraph 67 “At operation 415, a vehicle can collect federated real-time data from multiple communication points, such as nearby vehicles, infrastructure devices, and road condition services. The federated real-time data can be used to optimize the dynamic speed limit predicted at operation 401. That is, supplementing real-time data obtained directly by the vehicle at operation 405 with federated real-time data collected during operation 415 can improve the accuracy of the models, thus improving the dynamic speed limit prediction. For example, a plurality of vehicles traveling on the same road can communicate real-time imagery obtained from their respective cameras to the vehicle performing process 400, via V2V technology for instance. Referring back to the example of a vehicle approaching rain, real-time images collected from the plurality of vehicles can further indicate that there is rain in the same vicinity. Thus, the federated learning approach supports a consensus amongst the vehicles that rain is indeed present. Analyzing the federated real-time data collected at operation 415 can serve to confirm (or deny) the current road conditions, as monitored by the vehicle.”);
generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the data (See at least Lerner FIG. 4 “420” and Paragraphs 68-69 “Thereafter, at operation 420, the initial predicted dynamic speed limit can be optimized using the collected federated real-time data. In some embodiments, operation 420 can involve applying the newly collected federated real-time data to the machine learning algorithm(s) and/or model(s) for predicting an optimized dynamic speed limit. In some cases, operation 420 can include employing an optimization algorithm, which generates an optimized value of the initially predicted dynamic speed limit. In an embodiment, operation 420 involves verifying the dynamic speed limit predicted at operation 410 using the federated real-time data collected during operation 415. Verifying can be generally described as determining whether there is consensus between the real-time data obtained directly from the vehicle (e.g., operation 405) and the federated real-time data collected from the multiple communication points (e.g., operation 420). In the case where the real-time data converges, it may indicate that the initial dynamic speed limit was appropriately predicted for the current road conditions, and thus can be employed in operating the vehicle …”); and
Lerner fails to explicitly disclose generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the probe data, wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed; providing an indication of the speed profile to an electronic control unit of one or more autonomous vehicles to facilitate navigation of the one or more autonomous vehicles along the road segment.
Fowe teaches providing an indication of the speed profile to an electronic control unit of one or more autonomous vehicles to facilitate navigation of the one or more autonomous vehicles along the road segment (See at least Fowe FIG. 6 “570” and Claim 1 “A mapping system comprising … provide for at least one of navigational instructions or autonomous vehicle control based on the lane-level speed profile for the road segment.” and Paragraph 11 “Causing the apparatus to provide for at least one of navigational instruction or autonomous vehicle control based on the lane-level speed profile for the road segment may include causing the apparatus to identify a desired speed for the road segment based upon a purpose of travel for a vehicle; identify a lane of the road segment corresponding to the desired speed for the road segment; and provide instructions for directing travel of a vehicle in the identified lane” | Paragraph 59 “At 560, a lane-level speed profile for the road segment is generated. Navigational instruction or autonomous vehicle control is provided at 570 using the lane-level speed profile for the road segment.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner to include providing an indication of the speed profile to an electronic control unit of one or more autonomous vehicles to facilitate navigation of the one or more autonomous vehicles along the road segment, as taught by Fowe as disclosed above, in order to ensure safe vehicle traversal (Fowe Paragraph 1 “An example embodiment of the present invention relates to determining lane level speed profiles, and more particularly, to using historical vehicle speed data to establish speed profiles on a lane level of granularity for road segments and for a series of road segments.”).
Lerner in view of Fowe fail to explicitly disclose generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the probe data, wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed.
Zhu teaches generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the vehicle data and generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the time-space mapping data (See at least Zhu Claim 1 “generate a driving trajectory for an autonomous driving vehicle (ADV), the method comprising: calculating a first trajectory based on a map and a route information; generating a path profile based on the first trajectory, traffic rules, and an obstacle information describing one or more obstacles perceived by the ADV; generating a speed profile based on the path profile, wherein the speed profile includes, for each of the obstacles, a decision to yield or overtake the obstacle; performing a quadratic programming optimization on the path profile and the speed profile to identify an optimal path with optimal speeds, including optimizing a path cost function using quadratic programming to generate a station-lateral map based on the path profile and optimizing a speed cost function using quadratic programming to generate a two-dimensional station-time graph indicative of a distance travelled with respect to time based on the speed profile; generating a second trajectory based on the optimal path profile and the optimal speeds; and driving the ADV autonomously according to the second trajectory.” | Col. 10 line 55 – Col. 11 line 2 “In one embodiment, decision module 304 generates a rough speed profile (as part of path/speed profiles 313) based on the generated rough path profile. The rough speed profile indicates the best speed at a particular point in time controlling the ADV. Similar to the rough path profile, candidate speeds at different points in time are iterated using dynamic programming to find speed candidates (e.g., speed up or slow down) with a lowest speed cost based on cost functions, as part of costs functions 315 of FIG. 3A, in view of obstacles perceived by the ADV. The rough speed profile decides whether the ADV should overtake or avoid an obstacle, and to the left or right of the obstacle. In one embodiment, the rough speed profile includes a station-time (ST) graph (as part of SL maps/ST graphs 314). Station-time graph indicates a distance travelled with respect to time” | Col. 13 lines 32-44 “Station-time graphs 531 can include the station-time (ST) graph generated by ST graphs generator 515 of speed decision process 405. Speed planning process or speed planning module 523 can use a rough speed profile (e.g., a station-time graph) and results from path planning process 407 as initial constraints to calculate an optimal station-time curve. Sequence smoother 533 can apply a smoothing algorithm (such as B-spline or regression) to the time sequence of points. Speed costs module 535 can recalculate the ST graph with a speed cost function, as part of cost functions 315 of FIG. 3A, to optimize a total cost for movement candidates (e.g., speed up/slow down) at different points in time.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe to include generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the probe data and generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the time-space mapping data, as taught by Zhu as disclosed above, such that the generation of a time-space mapping is based in least in part on the probe data, in order to ensure an accurate determination of the vehicle speed profile (Zhu “The system optimizes one or more reference lines based on the paths and speeds decisions as trajectories to plan when and where the car should be at a particular point in time”).
Lerner in view of Fowe in view of Zhu fail to explicitly disclose wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed.
Mulcahy teaches wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity (See at least Mulcahy FIGS. 1-6 and Paragraph 5 “A space-time diagram is generated based upon the location data received from the vehicles. Location data of the road segment and/or surrounding road segments may be used to generate the space-time diagram. A first axis of the space-time diagram represents distance along the road segment(s) and a second axis of the space-time diagram represents time. Location data associated with vehicles traveling the road segments(s) is plotted within the space-time diagram. In this way, the space-time diagram represents locations of cars along the road segment(s) over time, which is referred to as vehicle trajectories that are indicative of vehicle speeds. A vehicle trajectory of a vehicle corresponds to location data of the vehicle plotted within the space-time diagram. The more horizontal a vehicle trajectory (e.g., the more parallel to the first axis representing distance along the road segment(s)), the faster a vehicle is traveling. The more vertical a vehicle trajectory (e.g., the more parallel to the second axis representing time), the slower the vehicle is traveling. Also, colors may be assigned to vehicle trajectories based upon ranges of vehicle speeds, such as a green color (or any other color) for free flowing vehicles that are traveling close to the speed limit and a red color (or any other color) for congested flow vehicles traveling well below the speed limit”); determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed (See at least Mulcahy FIGS. 1-6 and Paragraphs 44-51 “At 105, the convolutional neural network 706 is used to process the space-time diagram 704 to identify a slowdown 707, if present. The convolutional neural network 706 is trained and configured to identify, extract, and evaluate features of space-time diagrams in order to identify slowdowns represented by the space-time diagrams. The features may correspond to shapes of trajectories, angles/slopes of trajectories, counts or percentages of trajectories having certain features, locations of trajectories, changes in slope of trajectories, colors of trajectories, and/or a wide variety of other features that are indicative of slowdowns being illustrated within space-time diagrams or not. In this way, the convolutional neural network 706 processes the space-time diagram 704 to identify the slowdown 707, if present. In an embodiment, the space-time diagram 704 is processed using the convolutional neural network 706 to output a first probability that the space-time diagram 704 illustrates a slowdown. The convolutional neural network 706 may also output a second probability that the space-time diagram 704 does not illustrate a slowdown. The convolutional neural network 706 is trained to utilize image recognition functionality to evaluate features of space-time diagrams to output probabilities that the space-time diagrams illustrate slowdowns based upon the features (e.g., space-time diagrams labeled as illustrating and not illustrating slowdowns are used to train the convolutional neural network to identify features indicative of slowdowns being illustrated or not within space-time diagrams). The features may correspond to shapes of trajectories, angles/slopes of trajectories, counts or percentages of trajectories having certain features, locations of trajectories, changes in slope of trajectories, colors of trajectories, and/or a wide variety of other features that are indicative of slowdowns being illustrated within space-time diagrams or not. In an embodiment, at 106, a regression convolutional neural network 706 is used to identify one or more transitions points 708 depicted by the space-time diagram 704. The transition points 708 are where free-flowing vehicle speeds transition to congested vehicle speeds. In particular, the regression convolutional neural network 706 is trained to identify pixels within the space-time diagram 704 that represent transition points (e.g., a back of queue), such as based upon features corresponding to changes in slopes of trajectories and locations of the changes in slope. In an example, the regression convolutional neural network 706 identifies a time series of transition points. A Kalman filter is executed upon the time series of transition points to identify an accurate current transition point and to predict future transition point locations. With having known transition point locations, vehicle speeds are separated into free flow and congested speed clusters. If a difference in median speed of the clusters is greater than a threshold speed, then the slowdown is categorized as a dangerous slowdown or other slowdown category based upon other threshold speeds … At 108, a notification of the slowdown is constructed and transmitted over a network to a computing device associated with a driver of a vehicle that is to travel the road segment during the slowdown. In an embodiment, the notification is transmitted to computing devices of vehicles within a threshold distance of a location of the slowdown. The notification may be transmitted to computing devices associated with vehicles that are traveling routes that will encounter the slowdown during a predicted duration of the slowdown. The notification may be constructed to comprise a description of the slowdown (e.g., a number of vehicles affected), an alternative route for avoiding the slowdown, a relatively precise location of the slowdown (e.g., the back of queue location), a current distance of a vehicle to the location of the slowdown, a timeframe of the slowdown, a predicted timeframe of the slowdown dissipating, a predicted future location of the slowdown (e.g., a location of the back of queue by the time the vehicle will be within a threshold distance of the slowdown), etc. Various visual and audible notifications/alerts may be provided by the notification through the computing device.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu to include wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed, as taught by Mulcahy as disclosed above, in order to ensure optimal vehicle speeds for road segments (Mulcahy Paragraph 51 “In this way, drivers are provided with adequate notice of slowdowns so that drivers have ample time to safely react to the slowdowns”).
With respect to claim 6, Lerner in view of Fowe in view of Zhu in view of Mulcahy teach obtaining the probe data comprises obtaining sensor data from the one or more probe apparatuses traveling along the road segment associated with the safety risk profile (See at least Lerner Paragraph 67 “That is, supplementing real-time data obtained directly by the vehicle at operation 405 with federated real-time data collected during operation 415 can improve the accuracy of the models, thus improving the dynamic speed limit prediction. For example, a plurality of vehicles traveling on the same road can communicate real-time imagery obtained from their respective cameras to the vehicle performing process 400, via V2V technology for instance”).
With respect to claim 8, and similarly claim 18 Lerner in view of Fowe in view of Zhu in view of Mulcahy teach configuring a speed setting for the one or more autonomous vehicles based at least in part on the speed profile (See at least Lerner FIG. 4 “425” and Paragraph 71 “Next, at operation 425, a driving operation can be performed using the optimized dynamic speed limit. Operation 425 can involve automatically performing the driving operation, where the driving operation causes the vehicle to move at a driving speed that is approximately equal to the predicted dynamic speed limit. In an autonomous vehicle, the vehicle can perform fully automated functions to operate the vehicle in accordance with the dynamic speed limit”).
With respect to claim 12, Lerner teaches an apparatus comprising processing circuitry and at least one memory including computer program code instructions, the computer program code instructions configured to (See at least Lerner Paragraph 48 “As alluded to above, vehicle 120 may include an electronic control unit 50. Electronic control unit 50 may include circuitry to control various aspects of the vehicle operation. Electronic control unit 50 may include, for example, a microcomputer that includes a one or more processing units (e.g., microprocessors), memory storage (e.g., RAM, ROM, etc.), and I/O devices.”), when executed by the processing circuity, cause the apparatus to:
identifying a road segment associated with a safety risk2 profile based at least in part on road condition data related to the road segment (See at least Lerner FIG. 4 “405” and Paragraph 64 “The process 400 beings at operation 405, where multiple vehicle sensors are employed to continuously monitor the current road conditions, in real-time. The monitoring of operation 405 can involve the sensors capturing real-time data pertaining to a driving environment surrounding a vehicle, such as a portion of a roadway. Thus, the real-time data can be analyzed to determine, or otherwise estimate, real-time conditions in which the vehicle is currently traveling. The road conditions detected at operation 405 may impact an appropriate speed for the vehicle, such as traffic speed, traffic volume, congestion, weather conditions, and the like. Alternatively, monitoring at operation 405 can be triggered by an event (rather than continuous), such as a time interval or traveling in a certain location. As a result of monitoring at operation 405, the vehicle has an awareness of the current, or real-time, road conditions, including any significant changes.”);
obtaining probe data from one or more probe apparatuses traveling along the road segment during an interval of time (See at least Lerner FIG. 4 “415” and Paragraph 67 “At operation 415, a vehicle can collect federated real-time data from multiple communication points, such as nearby vehicles, infrastructure devices, and road condition services. The federated real-time data can be used to optimize the dynamic speed limit predicted at operation 401. That is, supplementing real-time data obtained directly by the vehicle at operation 405 with federated real-time data collected during operation 415 can improve the accuracy of the models, thus improving the dynamic speed limit prediction. For example, a plurality of vehicles traveling on the same road can communicate real-time imagery obtained from their respective cameras to the vehicle performing process 400, via V2V technology for instance. Referring back to the example of a vehicle approaching rain, real-time images collected from the plurality of vehicles can further indicate that there is rain in the same vicinity. Thus, the federated learning approach supports a consensus amongst the vehicles that rain is indeed present. Analyzing the federated real-time data collected at operation 415 can serve to confirm (or deny) the current road conditions, as monitored by the vehicle.”);
generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the data (See at least Lerner FIG. 4 “420” and Paragraphs 68-69 “Thereafter, at operation 420, the initial predicted dynamic speed limit can be optimized using the collected federated real-time data. In some embodiments, operation 420 can involve applying the newly collected federated real-time data to the machine learning algorithm(s) and/or model(s) for predicting an optimized dynamic speed limit. In some cases, operation 420 can include employing an optimization algorithm, which generates an optimized value of the initially predicted dynamic speed limit. In an embodiment, operation 420 involves verifying the dynamic speed limit predicted at operation 410 using the federated real-time data collected during operation 415. Verifying can be generally described as determining whether there is consensus between the real-time data obtained directly from the vehicle (e.g., operation 405) and the federated real-time data collected from the multiple communication points (e.g., operation 420). In the case where the real-time data converges, it may indicate that the initial dynamic speed limit was appropriately predicted for the current road conditions, and thus can be employed in operating the vehicle …”); and
Lerner fails to explicitly disclose generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the probe data, wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed; providing an indication of the speed profile to an electronic control unit of one or more autonomous vehicles to facilitate navigation of the one or more autonomous vehicles along the road segment.
Fowe teaches providing an indication of the speed profile to an electronic control unit of one or more autonomous vehicles to facilitate navigation of the one or more autonomous vehicles along the road segment (See at least Fowe FIG. 6 “570” and Claim 1 “A mapping system comprising … provide for at least one of navigational instructions or autonomous vehicle control based on the lane-level speed profile for the road segment.” and Paragraph 11 “Causing the apparatus to provide for at least one of navigational instruction or autonomous vehicle control based on the lane-level speed profile for the road segment may include causing the apparatus to identify a desired speed for the road segment based upon a purpose of travel for a vehicle; identify a lane of the road segment corresponding to the desired speed for the road segment; and provide instructions for directing travel of a vehicle in the identified lane” | Paragraph 59 “At 560, a lane-level speed profile for the road segment is generated. Navigational instruction or autonomous vehicle control is provided at 570 using the lane-level speed profile for the road segment.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner to include providing an indication of the speed profile to an electronic control unit of one or more autonomous vehicles to facilitate navigation of the one or more autonomous vehicles along the road segment, as taught by Fowe as disclosed above, in order to ensure safe vehicle traversal (Fowe Paragraph 1 “An example embodiment of the present invention relates to determining lane level speed profiles, and more particularly, to using historical vehicle speed data to establish speed profiles on a lane level of granularity for road segments and for a series of road segments.”).
Lerner in view of Fowe fail to explicitly disclose generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the probe data, wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed.
Zhu teaches generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the vehicle data and generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the time-space mapping data (See at least Zhu Claim 1 “generate a driving trajectory for an autonomous driving vehicle (ADV), the method comprising: calculating a first trajectory based on a map and a route information; generating a path profile based on the first trajectory, traffic rules, and an obstacle information describing one or more obstacles perceived by the ADV; generating a speed profile based on the path profile, wherein the speed profile includes, for each of the obstacles, a decision to yield or overtake the obstacle; performing a quadratic programming optimization on the path profile and the speed profile to identify an optimal path with optimal speeds, including optimizing a path cost function using quadratic programming to generate a station-lateral map based on the path profile and optimizing a speed cost function using quadratic programming to generate a two-dimensional station-time graph indicative of a distance travelled with respect to time based on the speed profile; generating a second trajectory based on the optimal path profile and the optimal speeds; and driving the ADV autonomously according to the second trajectory.” | Col. 10 line 55 – Col. 11 line 2 “In one embodiment, decision module 304 generates a rough speed profile (as part of path/speed profiles 313) based on the generated rough path profile. The rough speed profile indicates the best speed at a particular point in time controlling the ADV. Similar to the rough path profile, candidate speeds at different points in time are iterated using dynamic programming to find speed candidates (e.g., speed up or slow down) with a lowest speed cost based on cost functions, as part of costs functions 315 of FIG. 3A, in view of obstacles perceived by the ADV. The rough speed profile decides whether the ADV should overtake or avoid an obstacle, and to the left or right of the obstacle. In one embodiment, the rough speed profile includes a station-time (ST) graph (as part of SL maps/ST graphs 314). Station-time graph indicates a distance travelled with respect to time” | Col. 13 lines 32-44 “Station-time graphs 531 can include the station-time (ST) graph generated by ST graphs generator 515 of speed decision process 405. Speed planning process or speed planning module 523 can use a rough speed profile (e.g., a station-time graph) and results from path planning process 407 as initial constraints to calculate an optimal station-time curve. Sequence smoother 533 can apply a smoothing algorithm (such as B-spline or regression) to the time sequence of points. Speed costs module 535 can recalculate the ST graph with a speed cost function, as part of cost functions 315 of FIG. 3A, to optimize a total cost for movement candidates (e.g., speed up/slow down) at different points in time.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe to include generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the probe data and generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the time-space mapping data, as taught by Zhu as disclosed above, such that the generation of a time-space mapping is based in least in part on the probe data, in order to ensure an accurate determination of the vehicle speed profile (Zhu “The system optimizes one or more reference lines based on the paths and speeds decisions as trajectories to plan when and where the car should be at a particular point in time”).
Lerner in view of Fowe in view of Zhu fail to explicitly disclose wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed.
Mulcahy teaches wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity (See at least Mulcahy FIGS. 1-6 and Paragraph 5 “A space-time diagram is generated based upon the location data received from the vehicles. Location data of the road segment and/or surrounding road segments may be used to generate the space-time diagram. A first axis of the space-time diagram represents distance along the road segment(s) and a second axis of the space-time diagram represents time. Location data associated with vehicles traveling the road segments(s) is plotted within the space-time diagram. In this way, the space-time diagram represents locations of cars along the road segment(s) over time, which is referred to as vehicle trajectories that are indicative of vehicle speeds. A vehicle trajectory of a vehicle corresponds to location data of the vehicle plotted within the space-time diagram. The more horizontal a vehicle trajectory (e.g., the more parallel to the first axis representing distance along the road segment(s)), the faster a vehicle is traveling. The more vertical a vehicle trajectory (e.g., the more parallel to the second axis representing time), the slower the vehicle is traveling. Also, colors may be assigned to vehicle trajectories based upon ranges of vehicle speeds, such as a green color (or any other color) for free flowing vehicles that are traveling close to the speed limit and a red color (or any other color) for congested flow vehicles traveling well below the speed limit”); determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed (See at least Mulcahy FIGS. 1-6 and Paragraphs 44-51 “At 105, the convolutional neural network 706 is used to process the space-time diagram 704 to identify a slowdown 707, if present. The convolutional neural network 706 is trained and configured to identify, extract, and evaluate features of space-time diagrams in order to identify slowdowns represented by the space-time diagrams. The features may correspond to shapes of trajectories, angles/slopes of trajectories, counts or percentages of trajectories having certain features, locations of trajectories, changes in slope of trajectories, colors of trajectories, and/or a wide variety of other features that are indicative of slowdowns being illustrated within space-time diagrams or not. In this way, the convolutional neural network 706 processes the space-time diagram 704 to identify the slowdown 707, if present. In an embodiment, the space-time diagram 704 is processed using the convolutional neural network 706 to output a first probability that the space-time diagram 704 illustrates a slowdown. The convolutional neural network 706 may also output a second probability that the space-time diagram 704 does not illustrate a slowdown. The convolutional neural network 706 is trained to utilize image recognition functionality to evaluate features of space-time diagrams to output probabilities that the space-time diagrams illustrate slowdowns based upon the features (e.g., space-time diagrams labeled as illustrating and not illustrating slowdowns are used to train the convolutional neural network to identify features indicative of slowdowns being illustrated or not within space-time diagrams). The features may correspond to shapes of trajectories, angles/slopes of trajectories, counts or percentages of trajectories having certain features, locations of trajectories, changes in slope of trajectories, colors of trajectories, and/or a wide variety of other features that are indicative of slowdowns being illustrated within space-time diagrams or not. In an embodiment, at 106, a regression convolutional neural network 706 is used to identify one or more transitions points 708 depicted by the space-time diagram 704. The transition points 708 are where free-flowing vehicle speeds transition to congested vehicle speeds. In particular, the regression convolutional neural network 706 is trained to identify pixels within the space-time diagram 704 that represent transition points (e.g., a back of queue), such as based upon features corresponding to changes in slopes of trajectories and locations of the changes in slope. In an example, the regression convolutional neural network 706 identifies a time series of transition points. A Kalman filter is executed upon the time series of transition points to identify an accurate current transition point and to predict future transition point locations. With having known transition point locations, vehicle speeds are separated into free flow and congested speed clusters. If a difference in median speed of the clusters is greater than a threshold speed, then the slowdown is categorized as a dangerous slowdown or other slowdown category based upon other threshold speeds … At 108, a notification of the slowdown is constructed and transmitted over a network to a computing device associated with a driver of a vehicle that is to travel the road segment during the slowdown. In an embodiment, the notification is transmitted to computing devices of vehicles within a threshold distance of a location of the slowdown. The notification may be transmitted to computing devices associated with vehicles that are traveling routes that will encounter the slowdown during a predicted duration of the slowdown. The notification may be constructed to comprise a description of the slowdown (e.g., a number of vehicles affected), an alternative route for avoiding the slowdown, a relatively precise location of the slowdown (e.g., the back of queue location), a current distance of a vehicle to the location of the slowdown, a timeframe of the slowdown, a predicted timeframe of the slowdown dissipating, a predicted future location of the slowdown (e.g., a location of the back of queue by the time the vehicle will be within a threshold distance of the slowdown), etc. Various visual and audible notifications/alerts may be provided by the notification through the computing device.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu to include wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed, as taught by Mulcahy as disclosed above, in order to ensure optimal vehicle speeds for road segments (Mulcahy Paragraph 51 “In this way, drivers are provided with adequate notice of slowdowns so that drivers have ample time to safely react to the slowdowns”).
With respect to claim 20, Lerner teaches a computer program product comprising at least one non-transitory computer-readable storage medium having computer-executable program code instructions stored therein (See at least Lerner Paragraph 79 “In this document, the terms “computer program medium” and “computer usable medium” are used to generally refer to transitory or non-transitory media. Such media may be, e.g., memory 508, storage unit 520, media 514, and channel 528. These and other various forms of computer program media or computer usable media may be involved in carrying one or more sequences of one or more instructions to a processing device for execution”), the computer-executable program code instructions comprising program code instructions to:
identifying a road segment associated with a safety risk3 profile based at least in part on road condition data related to the road segment (See at least Lerner FIG. 4 “405” and Paragraph 64 “The process 400 beings at operation 405, where multiple vehicle sensors are employed to continuously monitor the current road conditions, in real-time. The monitoring of operation 405 can involve the sensors capturing real-time data pertaining to a driving environment surrounding a vehicle, such as a portion of a roadway. Thus, the real-time data can be analyzed to determine, or otherwise estimate, real-time conditions in which the vehicle is currently traveling. The road conditions detected at operation 405 may impact an appropriate speed for the vehicle, such as traffic speed, traffic volume, congestion, weather conditions, and the like. Alternatively, monitoring at operation 405 can be triggered by an event (rather than continuous), such as a time interval or traveling in a certain location. As a result of monitoring at operation 405, the vehicle has an awareness of the current, or real-time, road conditions, including any significant changes.”);
obtaining probe data from one or more probe apparatuses traveling along the road segment during an interval of time (See at least Lerner FIG. 4 “415” and Paragraph 67 “At operation 415, a vehicle can collect federated real-time data from multiple communication points, such as nearby vehicles, infrastructure devices, and road condition services. The federated real-time data can be used to optimize the dynamic speed limit predicted at operation 401. That is, supplementing real-time data obtained directly by the vehicle at operation 405 with federated real-time data collected during operation 415 can improve the accuracy of the models, thus improving the dynamic speed limit prediction. For example, a plurality of vehicles traveling on the same road can communicate real-time imagery obtained from their respective cameras to the vehicle performing process 400, via V2V technology for instance. Referring back to the example of a vehicle approaching rain, real-time images collected from the plurality of vehicles can further indicate that there is rain in the same vicinity. Thus, the federated learning approach supports a consensus amongst the vehicles that rain is indeed present. Analyzing the federated real-time data collected at operation 415 can serve to confirm (or deny) the current road conditions, as monitored by the vehicle.”);
generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the data (See at least Lerner FIG. 4 “420” and Paragraphs 68-69 “Thereafter, at operation 420, the initial predicted dynamic speed limit can be optimized using the collected federated real-time data. In some embodiments, operation 420 can involve applying the newly collected federated real-time data to the machine learning algorithm(s) and/or model(s) for predicting an optimized dynamic speed limit. In some cases, operation 420 can include employing an optimization algorithm, which generates an optimized value of the initially predicted dynamic speed limit. In an embodiment, operation 420 involves verifying the dynamic speed limit predicted at operation 410 using the federated real-time data collected during operation 415. Verifying can be generally described as determining whether there is consensus between the real-time data obtained directly from the vehicle (e.g., operation 405) and the federated real-time data collected from the multiple communication points (e.g., operation 420). In the case where the real-time data converges, it may indicate that the initial dynamic speed limit was appropriately predicted for the current road conditions, and thus can be employed in operating the vehicle …”); and
Lerner fails to explicitly disclose generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the probe data, wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed; providing an indication of the speed profile to an electronic control unit of one or more autonomous vehicles to facilitate navigation of the one or more autonomous vehicles along the road segment.
Fowe teaches providing an indication of the speed profile to an electronic control unit of one or more autonomous vehicles to facilitate navigation of the one or more autonomous vehicles along the road segment (See at least Fowe FIG. 6 “570” and Claim 1 “A mapping system comprising … provide for at least one of navigational instructions or autonomous vehicle control based on the lane-level speed profile for the road segment.” and Paragraph 11 “Causing the apparatus to provide for at least one of navigational instruction or autonomous vehicle control based on the lane-level speed profile for the road segment may include causing the apparatus to identify a desired speed for the road segment based upon a purpose of travel for a vehicle; identify a lane of the road segment corresponding to the desired speed for the road segment; and provide instructions for directing travel of a vehicle in the identified lane” | Paragraph 59 “At 560, a lane-level speed profile for the road segment is generated. Navigational instruction or autonomous vehicle control is provided at 570 using the lane-level speed profile for the road segment.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner to include providing an indication of the speed profile to an electronic control unit of one or more autonomous vehicles to facilitate navigation of the one or more autonomous vehicles along the road segment, as taught by Fowe as disclosed above, in order to ensure safe vehicle traversal (Fowe Paragraph 1 “An example embodiment of the present invention relates to determining lane level speed profiles, and more particularly, to using historical vehicle speed data to establish speed profiles on a lane level of granularity for road segments and for a series of road segments.”).
Lerner in view of Fowe fail to explicitly disclose generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the probe data, wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed.
Zhu teaches generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the vehicle data and generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the time-space mapping data (See at least Zhu Claim 1 “generate a driving trajectory for an autonomous driving vehicle (ADV), the method comprising: calculating a first trajectory based on a map and a route information; generating a path profile based on the first trajectory, traffic rules, and an obstacle information describing one or more obstacles perceived by the ADV; generating a speed profile based on the path profile, wherein the speed profile includes, for each of the obstacles, a decision to yield or overtake the obstacle; performing a quadratic programming optimization on the path profile and the speed profile to identify an optimal path with optimal speeds, including optimizing a path cost function using quadratic programming to generate a station-lateral map based on the path profile and optimizing a speed cost function using quadratic programming to generate a two-dimensional station-time graph indicative of a distance travelled with respect to time based on the speed profile; generating a second trajectory based on the optimal path profile and the optimal speeds; and driving the ADV autonomously according to the second trajectory.” | Col. 10 line 55 – Col. 11 line 2 “In one embodiment, decision module 304 generates a rough speed profile (as part of path/speed profiles 313) based on the generated rough path profile. The rough speed profile indicates the best speed at a particular point in time controlling the ADV. Similar to the rough path profile, candidate speeds at different points in time are iterated using dynamic programming to find speed candidates (e.g., speed up or slow down) with a lowest speed cost based on cost functions, as part of costs functions 315 of FIG. 3A, in view of obstacles perceived by the ADV. The rough speed profile decides whether the ADV should overtake or avoid an obstacle, and to the left or right of the obstacle. In one embodiment, the rough speed profile includes a station-time (ST) graph (as part of SL maps/ST graphs 314). Station-time graph indicates a distance travelled with respect to time” | Col. 13 lines 32-44 “Station-time graphs 531 can include the station-time (ST) graph generated by ST graphs generator 515 of speed decision process 405. Speed planning process or speed planning module 523 can use a rough speed profile (e.g., a station-time graph) and results from path planning process 407 as initial constraints to calculate an optimal station-time curve. Sequence smoother 533 can apply a smoothing algorithm (such as B-spline or regression) to the time sequence of points. Speed costs module 535 can recalculate the ST graph with a speed cost function, as part of cost functions 315 of FIG. 3A, to optimize a total cost for movement candidates (e.g., speed up/slow down) at different points in time.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe to include generating a time-space mapping that maps a position of vehicles along the road segment as a function of time and velocity based on at least in part on the probe data and generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the time-space mapping data, as taught by Zhu as disclosed above, such that the generation of a time-space mapping is based in least in part on the probe data, in order to ensure an accurate determination of the vehicle speed profile (Zhu “The system optimizes one or more reference lines based on the paths and speeds decisions as trajectories to plan when and where the car should be at a particular point in time”).
Lerner in view of Fowe in view of Zhu fail to explicitly disclose wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed.
Mulcahy teaches wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity (See at least Mulcahy FIGS. 1-6 and Paragraph 5 “A space-time diagram is generated based upon the location data received from the vehicles. Location data of the road segment and/or surrounding road segments may be used to generate the space-time diagram. A first axis of the space-time diagram represents distance along the road segment(s) and a second axis of the space-time diagram represents time. Location data associated with vehicles traveling the road segments(s) is plotted within the space-time diagram. In this way, the space-time diagram represents locations of cars along the road segment(s) over time, which is referred to as vehicle trajectories that are indicative of vehicle speeds. A vehicle trajectory of a vehicle corresponds to location data of the vehicle plotted within the space-time diagram. The more horizontal a vehicle trajectory (e.g., the more parallel to the first axis representing distance along the road segment(s)), the faster a vehicle is traveling. The more vertical a vehicle trajectory (e.g., the more parallel to the second axis representing time), the slower the vehicle is traveling. Also, colors may be assigned to vehicle trajectories based upon ranges of vehicle speeds, such as a green color (or any other color) for free flowing vehicles that are traveling close to the speed limit and a red color (or any other color) for congested flow vehicles traveling well below the speed limit”); determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed (See at least Mulcahy FIGS. 1-6 and Paragraphs 44-51 “At 105, the convolutional neural network 706 is used to process the space-time diagram 704 to identify a slowdown 707, if present. The convolutional neural network 706 is trained and configured to identify, extract, and evaluate features of space-time diagrams in order to identify slowdowns represented by the space-time diagrams. The features may correspond to shapes of trajectories, angles/slopes of trajectories, counts or percentages of trajectories having certain features, locations of trajectories, changes in slope of trajectories, colors of trajectories, and/or a wide variety of other features that are indicative of slowdowns being illustrated within space-time diagrams or not. In this way, the convolutional neural network 706 processes the space-time diagram 704 to identify the slowdown 707, if present. In an embodiment, the space-time diagram 704 is processed using the convolutional neural network 706 to output a first probability that the space-time diagram 704 illustrates a slowdown. The convolutional neural network 706 may also output a second probability that the space-time diagram 704 does not illustrate a slowdown. The convolutional neural network 706 is trained to utilize image recognition functionality to evaluate features of space-time diagrams to output probabilities that the space-time diagrams illustrate slowdowns based upon the features (e.g., space-time diagrams labeled as illustrating and not illustrating slowdowns are used to train the convolutional neural network to identify features indicative of slowdowns being illustrated or not within space-time diagrams). The features may correspond to shapes of trajectories, angles/slopes of trajectories, counts or percentages of trajectories having certain features, locations of trajectories, changes in slope of trajectories, colors of trajectories, and/or a wide variety of other features that are indicative of slowdowns being illustrated within space-time diagrams or not. In an embodiment, at 106, a regression convolutional neural network 706 is used to identify one or more transitions points 708 depicted by the space-time diagram 704. The transition points 708 are where free-flowing vehicle speeds transition to congested vehicle speeds. In particular, the regression convolutional neural network 706 is trained to identify pixels within the space-time diagram 704 that represent transition points (e.g., a back of queue), such as based upon features corresponding to changes in slopes of trajectories and locations of the changes in slope. In an example, the regression convolutional neural network 706 identifies a time series of transition points. A Kalman filter is executed upon the time series of transition points to identify an accurate current transition point and to predict future transition point locations. With having known transition point locations, vehicle speeds are separated into free flow and congested speed clusters. If a difference in median speed of the clusters is greater than a threshold speed, then the slowdown is categorized as a dangerous slowdown or other slowdown category based upon other threshold speeds … At 108, a notification of the slowdown is constructed and transmitted over a network to a computing device associated with a driver of a vehicle that is to travel the road segment during the slowdown. In an embodiment, the notification is transmitted to computing devices of vehicles within a threshold distance of a location of the slowdown. The notification may be transmitted to computing devices associated with vehicles that are traveling routes that will encounter the slowdown during a predicted duration of the slowdown. The notification may be constructed to comprise a description of the slowdown (e.g., a number of vehicles affected), an alternative route for avoiding the slowdown, a relatively precise location of the slowdown (e.g., the back of queue location), a current distance of a vehicle to the location of the slowdown, a timeframe of the slowdown, a predicted timeframe of the slowdown dissipating, a predicted future location of the slowdown (e.g., a location of the back of queue by the time the vehicle will be within a threshold distance of the slowdown), etc. Various visual and audible notifications/alerts may be provided by the notification through the computing device.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu to include wherein the time-space mapping comprises two or more visual indicator patterns for two or more vehicle paths, and the two or more visual indicator patterns are configured based on a degree of velocity; determining a respective statistically-derived speed for the two or more vehicle paths within the time-space mapping to generate a plurality of statistically-derived speeds for the road segment during the interval of time; selecting, from the plurality of statistically-derived speeds associated with the time-space mapping, a statistically-derived speed associated with the time-space mapping; generating a speed profile for modeling velocity of autonomous vehicles along the road segment based at least in part on the statistically derived speed, as taught by Mulcahy as disclosed above, in order to ensure optimal vehicle speeds for road segments (Mulcahy Paragraph 51 “In this way, drivers are provided with adequate notice of slowdowns so that drivers have ample time to safely react to the slowdowns”).
With respect to claim 21, Lerner in view of Fowe in view of Zhu in view of Mulcahy teach that generating the speed profile comprises updating speed limit information for the speed profile based at least in part on the time-space mapping (See at least Zhu Col. 19 lines 32-42 “In one embodiment, calculating a speed cost using a speed cost function includes calculating an individual speed cost based on a speed in view of a speed limit at each point and calculating a delta speed cost representing a cost to change speed between two adjacent points, where the speed cost is calculated based on the individual speed costs and the delta speed costs of all points along the trajectory. In another embodiment, processing logic further calculates an acceleration cost based on an acceleration of each point of the trajectory, where the speed cost is calculated further based on the acceleration cost of each point.”).
With respect to claim 22, Lerner in view of Fowe in view of Zhu in view of Mulcahy teach that generating the speed profile comprises determining a recommended speed for the speed profile based at least in part on the statistically-derived speed associated with the time-space mapping (See at least Zhu Col. 13 lines 30-58 “Speed planning process 409 or speed planning module 523 includes station-time graphs 531, sequence smoother 533, and speed costs module 535. Station-time graphs 531 can include the station-time (ST) graph generated by ST graphs generator 515 of speed decision process 405. Speed planning process or speed planning module 523 can use a rough speed profile (e.g., a station-time graph) and results from path planning process 407 as initial constraints to calculate an optimal station-time curve. Sequence smoother 533 can apply a smoothing algorithm (such as B-spline or regression) to the time sequence of points. Speed costs module 535 can recalculate the ST graph with a speed cost function, as part of cost functions 315 of FIG. 3A, to optimize a total cost for movement candidates (e.g., speed up/slow down) at different points in time. For example, in one embodiment, a total speed cost function can be … where the speeds cost are summed over all time progression points, speed′ denotes an acceleration value or a cost to change speed between two adjacent points, speed″ denotes a jerk value, or a derivative of the acceleration value or a cost to change a change of speed between two adjacent points, and distance denotes a distance from the ST point to the destination location. Here, speed costs module 535 calculates a station-time graph by minimizing the speed cost using quadratic programming optimization, for example, by QP module 540.”).
Claims 2-5, 11, 13-16, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Lerner (US 20200349833 A1) (“Lerner”) in view of Fowe (US 20200286372 A1) (“Fowe”) in view of Zhu (US 10754339 B2) (“Zhu”) in view of Mulcahy (US 20200143669 A1) (“Mulcahy”) further in view of Haque (US 20200189579 A1) (“Haque”).
With respect to claim 2, and similarly claim 13, Lerner in view of Fowe in view of Zhu in view of Mulcahy teaches identifying a road segment associated with a safety risk profile (See at least Lerner FIG. 4 “405” and Paragraph 64).
Lerner in view of Fowe in view of Zhu in view of Mulcahy fail to explicitly disclose identifying the road segment associated with the safety risk profile based at least in part on traffic incident data4 related to the road segment.
Haque teaches identifying the road segment associated with the safety risk profile based at least in part on traffic incident data5 related to the road segment (See at least Haque Paragraph 55 “For example, in some embodiments, the feature determination module 406 can analyze sensor data obtained by the sensor data module 404 to identify objects detected on or along the road segment being categorized. When identifying features such as objects, the feature determination module 406 can apply generally known object detection and recognition techniques. The identified objects can include, for example, pedestrians, vehicles, lane markings, curbs, trees, animals, debris, etc. In some embodiments, the feature determination module 406 can determine respective attributes for each of the identified objects. For example, upon detecting a pedestrian, the feature determination module 406 can determine attributes related to the pedestrian”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu in view of Mulcahy to include identifying the road segment associated with the safety risk profile based at least in part on traffic incident data related to the road segment, as taught by Haque as disclosed above, in order to ensure safe vehicle traversal in road segments (Haque Paragraph 2 “The present technology relates to the field of autonomous vehicles. More particularly, the present technology relates to systems, apparatus, and methods for determining similarities between road segments.”).
With respect to claim 3, and similarly claim 14, Lerner in view of Fowe in view of Zhu in view of Mulcahy teaches identifying a road segment associated with a safety risk profile (See at least Lerner FIG. 4 “405” and Paragraph 64).
Lerner in view of Fowe in view of Zhu in view of Mulcahy fail to explicitly disclose identifying the road segment associated with the safety risk profile based at least in part on hazard warning data6 related to the road segment.
Haque teaches identifying the road segment associated with the safety risk profile based at least in part on hazard warning data7 related to the road segment (See at least Haque Paragraphs 57-58 “The feature determination module 406 can also determine contextual features that correspond to a road segment being categorized. As mentioned, the contextual features can be used to determine (or predict) scenarios for the road segment. In some embodiments, such contextual features may be determined from sensor data obtained from, for example, the sensor data module 404, external data sources (e.g., weather data, etc.) … For example, in some embodiments, the feature determination module 406 can determine a respective calendar date, day of week, and time of day during which the sensor data was captured. In some embodiments, the feature determination module 406 can determine weather conditions (e.g., clear skies, overcast, fog, rain, sleet, snow, etc.) encountered while navigating the road segment based on the sensor data. The scenario determination module 408 can be configured to determine (or predict) scenarios for a road segment being categorized. For example, the scenario determination module 408 can determine (or predict) scenarios for a road segment based on features determined for the road segment by the feature determination module 406 … In some embodiments, a road segment can be associated with a scenario when all of the features associated with the road segment match features associated with the scenario. For example, assume a first scenario for “School Bus Stopping” is associated with features of a school bus with active hazard lights along with the presence of a stop sign. Assume further that sensor data for a road segment indicates the presence of features corresponding to a school bus with its hazard lights in use and the presence of a stop sign”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu in view of Mulcahy to include identifying the road segment associated with the safety risk profile based at least in part on hazard warning data related to the road segment, as taught by Haque as disclosed above, in order to ensure safe vehicle traversal in road segments (Haque Paragraph 2 “The present technology relates to the field of autonomous vehicles. More particularly, the present technology relates to systems, apparatus, and methods for determining similarities between road segments.”).
With respect to claim 4, and similarly claim 15, Lerner in view of Fowe in view of Zhu in view of Mulcahy teaches identifying a road segment associated with a safety risk profile (See at least Lerner FIG. 4 “405” and Paragraph 64).
Lerner in view of Fowe in view of Zhu in view of Mulcahy fail to explicitly disclose identifying the road segment associated with the safety risk profile based at least in part on weather condition data8 related to the road segment.
Haque teaches identifying the road segment associated with the safety risk profile based at least in part on weather condition data9 related to the road segment (See at least Haque Paragraphs 57 “The feature determination module 406 can also determine contextual features that correspond to a road segment being categorized. As mentioned, the contextual features can be used to determine (or predict) scenarios for the road segment. In some embodiments, such contextual features may be determined from sensor data obtained from, for example, the sensor data module 404, external data sources (e.g., weather data, etc.) … For example, in some embodiments, the feature determination module 406 can determine a respective calendar date, day of week, and time of day during which the sensor data was captured. In some embodiments, the feature determination module 406 can determine weather conditions (e.g., clear skies, overcast, fog, rain, sleet, snow, etc.) encountered while navigating the road segment based on the sensor data.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu in view of Mulcahy to include identifying the road segment associated with the safety risk profile based at least in part on weather condition data related to the road segment, as taught by Haque as disclosed above, in order to ensure safe vehicle traversal in road segments (Haque Paragraph 2 “The present technology relates to the field of autonomous vehicles. More particularly, the present technology relates to systems, apparatus, and methods for determining similarities between road segments.”).
With respect to claim 5, and similarly claim 16, Lerner in view of Fowe in view of Zhu in view of Mulcahy teaches identifying a road segment associated with a safety risk profile (See at least Lerner FIG. 4 “405” and Paragraph 64).
Lerner in view of Fowe in view of Zhu in view of Mulcahy fail to explicitly disclose identifying the road segment associated with the safety risk profile based at least in part on high-definition (HD) map data10 related to the road segment.
Haque teaches identifying the road segment associated with the safety risk profile based at least in part on high-definition (HD) map data11 related to the road segment (See at least Haque Paragraphs 46-47 “For example, the segment similarity module 306 may determine a similarity between a first road segment and a second road segment based on a comparison of their features (e.g., road features, etc.). In some embodiments, road features may include sampled (or collected) information describing objects associated with a given road segment as well as any permanent and ephemeral features associated with the road segment. Other examples of road features include geographic attributes (e.g., a shape or path of a road segment—straight, curved, etc.), metadata associated with the road segment (e.g., map features, zoning, surrounding businesses, census tracts, etc.), and detailed sensor data related to the configuration of the road segment (e.g., lane types, lane widths, existence of stop signs, etc.) … In some embodiments, once a threshold level of similarity or matching between a classified road segment and an unclassified road segment is determined, the information mapping module 308 can determine information associated with the classified road segment. The information mapping module 308 can then associate the determined information with the unclassified road segment. For example, in some embodiments, the classified road segment may be associated with a risk profile that is based on scenario exposure rates for the classified road segment. In this example, the information mapping module 308 can associate the risk profile with the unclassified road segment based upon the threshold similarity determination between the classified road segment and the unclassified road segment”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu in view of Mulcahy to include identifying the road segment associated with the safety risk profile based at least in part on high-definition (HD) map data related to the road segment, as taught by Haque as disclosed above, in order to ensure safe vehicle traversal in road segments (Haque Paragraph 2 “The present technology relates to the field of autonomous vehicles. More particularly, the present technology relates to systems, apparatus, and methods for determining similarities between road segments.”).
With respect to claim 11, and similarly claim 19, Lerner in view of Fowe in view of Zhu in view of Mulcahy fail to explicitly disclose configuring an autonomous driving level for the one or more autonomous vehicles based at least in part on the speed profile.
Haque, however, teaches configuring an autonomous driving level for the one or more autonomous vehicles based at least in part on the speed profile (See at least Haque Paragraph 39 “For example, a risk profile for a road segment type may be associated with generation of instructions or commands that cause a vehicle to navigate autonomously or semi-autonomously in accordance with the risk profile, such as decreasing its speed to a pre-defined speed limit when navigating road segments that have been categorized as the road segment type”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu in view of Mulcahy to include configuring an autonomous driving level for the one or more autonomous vehicles based at least in part on the speed profile, as taught by Haque as disclosed above, in order to ensure safe vehicle traversal in road segments (Haque Paragraph 2 “The present technology relates to the field of autonomous vehicles. More particularly, the present technology relates to systems, apparatus, and methods for determining similarities between road segments.”).
Claims 9-10 are rejected under 35 U.S.C. 103 as being unpatentable over Lerner (US 20200349833 A1) (“Lerner”) in view of Fowe (US 20200286372 A1) (“Fowe”) in view of Zhu (US 10754339 B2) (“Zhu”) in view of Mulcahy (US 20200143669 A1) (“Mulcahy”) further in view of Gordon (US 20230095539 A1) (“Gordon”).
With respect to claim 9, Lerner in view of Fowe in view of Zhu in view of Mulcahy fail to explicitly disclose initiating a lane change along the road segment for the one or more autonomous vehicles based at least in part on the speed profile.
Gordon, however, teaches initiating a lane change along the road segment for the one or more autonomous vehicles based at least in part on the speed profile (See at least Gordon Paragraphs 41-42 “In some exemplary embodiments, vehicle flow may be measured by unique vehicle count in the traffic stream on a road segment in a specific time period. The value may be assumed to be proportional to the actual vehicle flow. Using the Greenshield Traffic Flow model in a novel way, an optimized speed flow boundary may be generated to minimize the collection of hard braking events (10 or more) within the boundary … The exemplary system and method may be used for controlling a relation between a vehicle and traffic flow over a road segment and for determining the property of a road segment based on connected vehicle data, established road profile data and road segment deviation from road profile. In addition, the system may be used as an input for navigation of a vehicle based on (i) selecting an alternate route for the vehicle (ii) recommending a target lane or making a lane change for a vehicle, (iii) change a headway dynamic for the vehicle, (iv) changing the speed of a vehicle, (v) changing the speed profile of the vehicle.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu in view of Mulcahy to include initiating a lane change along the road segment for the one or more autonomous vehicles based at least in part on the speed profile, as taught by Gordon as disclosed above, in order to ensure safe vehicle traversal in road segments (Gordon Paragraph 1 “The present disclosure relates generally to a system for predicting traffic flow disruption risk and providing traffic flow information to motor vehicles and infrastructure. More specifically, aspects of the present disclosure relate to systems, methods and devices for determining a traffic flow risk for a speed flow pair in response to a number of hard braking events associated with that speed flow pair in a roadway segment.”).
With respect to claim 10, Lerner in view of Fowe in view of Zhu in view of Mulcahy fail to explicitly disclose determining a navigation route along the road segment for the one or more autonomous vehicles based at least in part on the speed profile.
Gordon, however, teaches determining a navigation route along the road segment for the one or more autonomous vehicles based at least in part on the speed profile (See at least Gordon Paragraphs 41-42 “In some exemplary embodiments, vehicle flow may be measured by unique vehicle count in the traffic stream on a road segment in a specific time period. The value may be assumed to be proportional to the actual vehicle flow. Using the Greenshield Traffic Flow model in a novel way, an optimized speed flow boundary may be generated to minimize the collection of hard braking events (10 or more) within the boundary … The exemplary system and method may be used for controlling a relation between a vehicle and traffic flow over a road segment and for determining the property of a road segment based on connected vehicle data, established road profile data and road segment deviation from road profile. In addition, the system may be used as an input for navigation of a vehicle based on (i) selecting an alternate route for the vehicle (ii) recommending a target lane or making a lane change for a vehicle, (iii) change a headway dynamic for the vehicle, (iv) changing the speed of a vehicle, (v) changing the speed profile of the vehicle.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Lerner in view of Fowe in view of Zhu in view of Mulcahy to include determining a navigation route along the road segment for the one or more autonomous vehicles based at least in part on the speed profile, as taught by Gordon as disclosed above, in order to ensure safe vehicle traversal in road segments (Gordon Paragraph 1 “The present disclosure relates generally to a system for predicting traffic flow disruption risk and providing traffic flow information to motor vehicles and infrastructure. More specifically, aspects of the present disclosure relate to systems, methods and devices for determining a traffic flow risk for a speed flow pair in response to a number of hard braking events associated with that speed flow pair in a roadway segment.”).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to IBRAHIM ABDOALATIF ALSOMAIRY whose telephone number is (571)272-5653. The examiner can normally be reached M-F 7:30-5:30.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Faris Almatrahi can be reached at 313-446-4821. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/IBRAHIM ABDOALATIF ALSOMAIRY/Examiner, Art Unit 3667 /KENNETH J MALKOWSKI/Primary Examiner, Art Unit 3667
1 There is no limiting definition as to what constitutes a “safety risk”, however paragraph 48 of the published specification states that it may “provide a prediction as to a likely safety risk situation related to a hazard warning, increased traffic, certain weather conditions (e.g., clement weather, blizzard situations, etc.), a traffic incident, lack of cellular network coverage, and/or missing information for autonomous driving perception (e.g., a lack of autonomous vehicle data from autonomous vehicles) for the road segment”
2 There is no limiting definition as to what constitutes a “safety risk”, however paragraph 48 of the published specification states that it may “provide a prediction as to a likely safety risk situation related to a hazard warning, increased traffic, certain weather conditions (e.g., clement weather, blizzard situations, etc.), a traffic incident, lack of cellular network coverage, and/or missing information for autonomous driving perception (e.g., a lack of autonomous vehicle data from autonomous vehicles) for the road segment”
3 There is no limiting definition as to what constitutes a “safety risk”, however paragraph 48 of the published specification states that it may “provide a prediction as to a likely safety risk situation related to a hazard warning, increased traffic, certain weather conditions (e.g., clement weather, blizzard situations, etc.), a traffic incident, lack of cellular network coverage, and/or missing information for autonomous driving perception (e.g., a lack of autonomous vehicle data from autonomous vehicles) for the road segment”
4 There is no limiting definition as to what constitutes “traffic incident data”, however paragraph 64 of the published specification states that it “can include information regarding one or more traffic incidents such as, for example, real-time traffic data related to the road segment, one or more traffic accidents related to the road segment, one or more traffic jams related to the road segment, one or more road construction incidents related to the road segment, one or more traffic light incidents related to the road segment, one or more high pedestrian traffic incidents related to the road segment … and/or one or more other traffic incidents related to the road segment”
5 There is no limiting definition as to what constitutes “traffic incident data”, however paragraph 64 of the published specification states that it “can include information regarding one or more traffic incidents such as, for example, real-time traffic data related to the road segment, one or more traffic accidents related to the road segment, one or more traffic jams related to the road segment, one or more road construction incidents related to the road segment, one or more traffic light incidents related to the road segment, one or more high pedestrian traffic incidents related to the road segment … and/or one or more other traffic incidents related to the road segment”
6 There is no limiting definition as to what constitutes “hazard warning data”, however paragraph 65 of the published specification states that “The hazard warning data can include information regarding one or more hazard warning conditions related to the road segment such as, for example, an ice warning condition related to the road segment, a heavy rain warning condition related to the road segment, a hydroplane warning condition related to the road segment … a hazard light condition related to one or more autonomous vehicles traveling along the road segment … and/or one or more other hazard warning conditions related to the road segment”
7 There is no limiting definition as to what constitutes “hazard warning data”, however paragraph 65 of the published specification states that “The hazard warning data can include information regarding one or more hazard warning conditions related to the road segment such as, for example, an ice warning condition related to the road segment, a heavy rain warning condition related to the road segment, a hydroplane warning condition related to the road segment … a hazard light condition related to one or more autonomous vehicles traveling along the road segment … and/or one or more other hazard warning conditions related to the road segment”
8 There is no limiting definition as to what constitutes “weather condition data”, however paragraph 66 of the published specification states that “For example, the weather condition data can include a real-time weather condition related to the geographic location or geographic region for the road segment … a real-time precipitation condition related to the geographic location or geographic region for the road segment … and/or one or more other real-time weather conditions related to a geographic location or geographic region for the road segment”
9 There is no limiting definition as to what constitutes “weather condition data”, however paragraph 66 of the published specification states that “For example, the weather condition data can include a real-time weather condition related to the geographic location or geographic region for the road segment … a real-time precipitation condition related to the geographic location or geographic region for the road segment … and/or one or more other real-time weather conditions related to a geographic location or geographic region for the road segment”
10 There is no limiting definition as to what constitutes “high-definition (HD) map data”, however paragraph 66 of the published specification states that “can include node data, road segment data, link data, point of interest (POI) data, historical road condition data, historical traffic data, autonomous driving data, and/or other map data related to the road segment …”
11 There is no limiting definition as to what constitutes “high-definition (HD) map data”, however paragraph 66 of the published specification states that “can include node data, road segment data, link data, point of interest (POI) data, historical road condition data, historical traffic data, autonomous driving data, and/or other map data related to the road segment …”
Prosecution Insights