DETAILED ACTION
Response to Amendment
This Action is responsive to Applicant’s response filed on 10/23/2025. All claims are still pending
in the present application. This Action is made FINAL.
Response to Arguments
During prosecution, claim scope not solely on the basis of claim language, but also on giving claims their broadest reasonable construction in light of the specification as it would be interpreted by one of ordinary skill in the art. In re Am. Acad. of Sci. Tech. Ctr., 367 F.3d 1359, 1364 (Fed. Cir. 2004). See also Superguide Corp. v. DirecTV Enterprises, Inc., 358 F.3d 870, 875 (Fed. Cir. 2004) (“Though understanding the claim language may be aided by explanations contained in the written description, it is important not to import into a claim limitations that are not part of the claim.”). Additionally, “[t]hough understanding the claim language may be aided by the explanations contained in the written description, it is important not to import into a claim limitations that are not a part of the claim. For example, a particular embodiment appearing in the written description may not be read into a claim when the claim language is broader than the embodiment.” Superguide Corp. v. DirecTV Enterprises, Inc., 358 F.3d 870, 875 (Fed. Cir. 2004).
In regards to Argument(s) 1, Applicant(s) state(s) that, Ollis fails to teach or suggest a controller configured to "perform semantic segmentation of the image, match the image, on which the semantic segmentation is performed, with the point cloud and configured to detect a vehicle and a traffic line from the captured image, and track the detected vehicle and the detected traffic line," as recited in claim 1. Instead, Ollis predicts object behavior based on the object's own state (current location; current speed and/or acceleration; current heading; current pose; current shape). See, Para. [0090] of Ollis. In other words, Ollis predicts each object's trajectory based on its own state and environmental context., (Emphasis added, Remarks, page 7-8).
With respect to the 35 USC 103 rejection, Applicant’s arguments have been considered but they are not persuasive. Examiner respectfully disagrees. While Ollis may not explicitly use the term “semantic segmentation,” Ollis discloses perception modules that classify and distinguish different types of objects and roadway features from sensor data, including vehicles, roadway boundaries, traffic signals, pedestrians, and obstacles. Specifically, Ollis teaches that perception detection identifies and classifies objects in the surrounding environment using camera and LiDAR data, thereby assigning semantic meaning to image regions corresponding to different object classes (See Ollis Paragraph [0102], which describes detecting vehicles, roadway boundaries, traffic signals, and other objects). Such classification of image data into semantically meaningful categories constitutes semantic segmentation under the broadest reasonable interpretation of the claims, or at minimum renders the claimed semantic segmentation obvious when considered in combination with Goel.
As a result, the argued features are written such that they read upon the cited references. Therefore, the previous rejection still applies.
In regards to Argument(s) 2, Applicant(s) state(s) that, Goel fails to disclose or suggest at least several limitations of claim 1. Notably, Goel does not teach or suggest: (i) matching semantic-segmented image data with point cloud data; (ii) detecting and tracking traffic lines; or (iii) predicting the driving path of a target vehicle based on extracted feature information from surrounding vehicles, therefore, the rejection of 35 U.S.C. 103 should be removed, (Emphasis added, Remarks, page 7-8).
With respect to the 35 USC 103 rejection, Applicant’s arguments (i) have been considered but they are not persuasive. Examiner respectfully disagrees. Applicant’s argument improperly construes the claimed “matching” as requiring an explicit registration or alignment algorithm. Claim 1 does not recite any particular matching technique or mathematical registration process. Goel expressly teaches projecting pixels or regions derived from 2D image data into three-dimensional (3D) space to generate a point cloud representation, thereby associating image regions with corresponding 3D points. For example, Goel discloses that “a 3D point cloud may be generated by projecting pixels from the 2D image into 3D space” (See Goel Paragraph [0054]; and See Figure 1), which necessarily results in an association between image derived regions and corresponding point cloud data. This association meets the claimed “matching” under a broadest reasonable interpretation. Therefore, Ollis in view of Goel teaches or renders obvious matching a semantically processed image with a point cloud.
As a result, the argued features are written such that they read upon the cited references. Therefore, the previous rejection still applies.
With respect to the 35 USC 103 rejection, Applicant’s arguments (ii) have been considered but they are not persuasive. Examiner respectfully disagrees. Ollis teaches detecting roadway features, including roadway boundaries, as part of its perception and prediction system. Specifically, Ollis discloses that detected objects may include “roadway boundaries” along with vehicles and other roadway features (See Ollis Paragraph [0102]). Traffic lines are a type of roadway boundary. Furthermore, Ollis teaches maintaining state information of detected objects over time, including location, speed, and heading, for use in trajectory prediction (See Ollis Paragraph [0103] – Paragraph [0104]). Maintaining such state information over time constitutes tracking under the broadest reasonable interpretation. In addition, Goel teaches performing image-based object detection by semantically processing image regions and projecting image-derived features into a three-dimensional (3D) point cloud representation, thereby improving the identification of roadway features from image and depth data (see Goel Paragraph [0054] and Figure 1). When combined, Ollis provides the detection and tracking of roadway features, including traffic lines, while Goel provides enhanced perception techniques for accurately identifying such features. Accordingly, Ollis in view of Goel teaches or renders obvious detecting and tracking traffic lines as recited in claim 1.
As a result, the argued features are written such that they read upon the cited references. Therefore, the previous rejection still applies.
With respect to the 35 USC 103 rejection, Applicant’s arguments (iii) have been considered but they are not persuasive. Examiner respectfully disagrees. Applicant’s argument mischaracterizes Ollis. Ollis explicitly teaches predicting future trajectories of multiple detected objects based on perception information of the surrounding environment, including the states of other detected objects. Ollis states that the prediction system “may predict the future locations, trajectories, and/or actions of such objects perceived in the environment, based at least in part on perception information (e.g., the state data for each object) […] and/or any other data related to […] the surrounding environment, and/or relationship(s)” (See Ollis Paragraph [0104]). Thus, Ollis predicts driving paths using feature information of multiple surrounding vehicles, as recited in claim 1. Additionally, Goel teaches extracting three-dimensional representations of multiple objects from image data by semantically processing image regions and projecting image-derived features into a point cloud representation (See Goel Paragraph [0054] and Figure 1), thereby providing accurate feature information for multiple surrounding objects. When combined, Ollis uses feature information of surrounding vehicles to perform trajectory prediction, while Goel provides enhanced perception techniques for accurately extracting such feature information from image and depth data. Accordingly, Ollis in view of Goel teaches or renders obvious predicting a driving path based on feature information of surrounding vehicles.
As a result, the argued features are written such that they read upon the cited references. Therefore, the previous rejection still applies.
Office Action Summary
Claim(s) 3-4 and 12-13 is/are cancelled.
Claim(s) 1-2 and 10-11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ollis et al (US 2023/0278581 A1) in view of Goel (US 2023/0033177 A1).
Claim(s) 5 and 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ollis et al (US 2023/0278581 A1) in view of Goel (US 2023/0033177 A1), further in view of Takamatsu et al (US 2016/0090084 A1).
Claim(s) 6-9 and 15-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ollis et al (US 2023/0278581 A1) in view of Goel (US 2023/0033177 A1), further in view of Wang (US 2023/0211660 A1).
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.
Claim(s) 1-2 and 10-11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ollis et al (US 2023/0278581 A1) in view of Goel (US 2023/0033177 A1).
Regarding claim(s) 1, Ollis teaches an apparatus for predicting a driving path of a vehicle, the apparatus comprising:
a camera configured to capture an image around an ego vehicle (Figure 2; and Paragraph [0070]: “such sensors may include, without limitation, a laser detection and ranging (LiDAR) system […] one or more cameras (e.g., visible spectrum cameras, infrared cameras, etc.) […] The sensor data can include information that describes the location of objects within the surrounding environment of AV 102, information about the environment itself, information about the motion of AV 102, information about a route of AV 102, and/or the like”);
a light detection and ranging (LiDAR) sensor configured to generate a point cloud around the ego vehicle (Figure 2; and Paragraph [0064]: “sensor data may include light detection and ranging (LiDAR) point cloud maps (e.g., map point data, etc.) associated with a geographic location (e.g., a location in three-dimensional space relative to the LiDAR system of a mapping vehicle in one or more roadways) of a number of points (e.g., a point cloud) that correspond to objects that have reflected a ranging laser of one or more mapping vehicles at the geographic location (e.g. an object such as a vehicle, bicycle, pedestrian, etc. in the roadway). As an example, sensor data may include LiDAR point cloud data that represents objects in the roadway, such as, other vehicles, pedestrians, cones, debris, and/or the like”); and
a controller configured to detect pieces of feature information of respective vehicles located around the ego vehicle based on the image and the point cloud and configured to predict a driving path of a target vehicle based on the pieces of feature information of the respective vehicles (Paragraph [0102] – Paragraph [0104]: “For example, prediction system 316 may process sensor data (e.g., from LiDAR, RADAR, camera images, etc.) in order to identify objects and/or features in and around the geospatial area of the autonomous vehicle. Detected objects may include traffic signals, roadway boundaries, vehicles, pedestrians, obstacles in the roadway, and/or the like. Perception detection 302 may use known object recognition and detection algorithms, video tracking algorithms, or computer vision algorithms (e.g., tracking objects frame-to-frame iteratively over a number of time periods, etc.) to perceive an environment of AV 102 [...] perception detection 302 may also determine, for one or more identified objects in the environment, a current state of the object. The state information may include, without limitation, for each object: current location; current speed and/or acceleration; current heading; current orientation; size/footprint; type (e.g., vehicle vs. pedestrian vs. bicycle vs. static object or obstacle); and/or other state information [...] Prediction system 316 may predict the future locations, trajectories, and/or actions of such objects perceived in the environment, based at least in part on perception information (e.g., the state data for each object) received from perception detection 302, the location information received from location system 312, sensor data, and/or any other data related to a past and/or current state of an object, the autonomous vehicle, the surrounding environment, and/or relationship(s)”); and
track the detected vehicle and the detected traffic line (Paragraph [0103]: “perception detection 302 may also determine, for one or more identified objects in the environment, a current state of the object […] current location; current speed and/or acceleration; current heading; current orientation; size/footprint; type (e.g., vehicle vs. pedestrian vs. bicycle vs. static object or obstacle) […]”).
Ollie fails to teach wherein the controller is further configured to: perform semantic segmentation of the image, match the image, on which the semantic segmentation is performed, with the point cloud and configured to detect a vehicle and a traffic line from the captured image.
However, Goel teaches wherein the controller is further configured to: perform semantic segmentation of the image (Paragraph [0016]: “some or all operations of the example process 100 may be performed by an image-based object detector 102 executing on one or more computing devices. Process 100, and various other examples herein, may be described in reference to performing object detection functionalities (e.g., object identification, classification, instance segmentation, semantic segmentation, and/or object tracking) by an autonomous vehicle operating within an environment”), match the image, on which the semantic segmentation is performed (Paragraph [0029]: “the image-based object detector 102 may generate both a point cloud and a 3D grid in operation 112 corresponding to the same image data 106 (or same region within the image data 106), and may perform both the object detection techniques in the operation 116 based on the point cloud and the object detection techniques in operation 120 based on the 3D grid. Although a neural network 118 and convolutions 122 are shown as illustrative examples, the 3D object detection performed in operation 116 and/or operation 120 can include various combinations of 3D object detection techniques and/or other image analysis functionality, including but not limited to object detection or identification, object classification, instance segmentation, semantic segmentation, object tracking, feature extraction, and/or transformation”; and Paragraph [0054]: “the image-based object detector 102 may perform a processing loop configured to project each pixel within a region of interest in the 2D image data, into a 3D point cloud […]”), with the point cloud and configured to detect a vehicle and a traffic line from the captured image (Paragraph [0021]: “The image-based object detector 102 may use models and algorithms trained based on lidar and/or radar ground truth data, and/or may be configured based on heuristic-based rules regarding object size averages and distributions (e.g., average lengths for cars, trucks, and bicycles, average height and height distributions for pedestrians, standard sizes for traffic signs, lane markings, sidewalk widths, etc.)”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to modify to perform semantic segmentation of the image section of Ollis to incorporate the use to perform semantic segmentation of the image of Goel and one of ordinary skill in the art would have recognized that the results of the combination were predictable. One could look to Goel to include to perform semantic segmentation of the image section, for example an autonomous vehicle may include an image-based object detector, implemented within one or more on-board vehicle systems (e.g., perception, prediction, planning, etc.), which may be used to improve the object detection capabilities of the autonomous vehicle. In such examples, based on the object detection performed by the image-based object detector (and/or other vehicle systems), the autonomous vehicle may generate a trajectory for controlling the vehicle and navigating within its current environment. The autonomous vehicle also may use the outputs from the image-based object detector to activate a secondary vehicle controller, a remote teleoperations computing device, and/or may engage or disengage certain autonomous driving features, based on the detection and analyses of the objects in the environment.
Regarding claim(s) 2 and 11, Ollis as modified by Goel teaches the apparatus of claim 1, where Ollis teaches wherein the pieces of feature information of the respective vehicles include at least one of positions, speeds, heading angles, heading angle change rates, or driving lanes of the respective vehicles, or any combination thereof (Paragraph [0103]: “perception detection 302 may also determine, for one or more identified objects in the environment, a current state of the object. The state information may include, without limitation, for each object: current location; current speed and/or acceleration; current heading; current orientation; size/footprint; type (e.g., vehicle vs. pedestrian vs. bicycle vs. static object or obstacle); and/or other state information”).
Regarding claim(s) 10, Ollis teaches a method for predicting a driving path of a vehicle, the method comprising:
capturing, by a camera sensor, an image around an ego vehicle (Figure 2; and Paragraph [0070]: “such sensors may include, without limitation, a laser detection and ranging (LiDAR) system […] one or more cameras (e.g., visible spectrum cameras, infrared cameras, etc.) […] The sensor data can include information that describes the location of objects within the surrounding environment of AV 102, information about the environment itself, information about the motion of AV 102, information about a route of AV 102, and/or the like”);
generating, by a light detection and ranging (LiDAR) sensor, a point cloud around the ego vehicle (Figure 2; and Paragraph [0064]: “sensor data may include light detection and ranging (LiDAR) point cloud maps (e.g., map point data, etc.) associated with a geographic location (e.g., a location in three-dimensional space relative to the LiDAR system of a mapping vehicle in one or more roadways) of a number of points (e.g., a point cloud) that correspond to objects that have reflected a ranging laser of one or more mapping vehicles at the geographic location (e.g. an object such as a vehicle, bicycle, pedestrian, etc. in the roadway). As an example, sensor data may include LiDAR point cloud data that represents objects in the roadway, such as, other vehicles, pedestrians, cones, debris, and/or the like”);
detecting, by a controller, pieces of feature information of respective vehicles located around the ego vehicle based on the image and the point cloud (Figure 2; Paragraph [0064]; and Paragraph [0070]); and
predicting, by the controller, driving path of a target vehicle based on the pieces of feature information of the respective vehicles (Paragraph [0102] – Paragraph [0104]: “For example, prediction system 316 may process sensor data (e.g., from LiDAR, RADAR, camera images, etc.) in order to identify objects and/or features in and around the geospatial area of the autonomous vehicle. Detected objects may include traffic signals, roadway boundaries, vehicles, pedestrians, obstacles in the roadway, and/or the like. Perception detection 302 may use known object recognition and detection algorithms, video tracking algorithms, or computer vision algorithms (e.g., tracking objects frame-to-frame iteratively over a number of time periods, etc.) to perceive an environment of AV 102 [...] perception detection 302 may also determine, for one or more identified objects in the environment, a current state of the object. The state information may include, without limitation, for each object: current location; current speed and/or acceleration; current heading; current orientation; size/footprint; type (e.g., vehicle vs. pedestrian vs. bicycle vs. static object or obstacle); and/or other state information [...] Prediction system 316 may predict the future locations, trajectories, and/or actions of such objects perceived in the environment, based at least in part on perception information (e.g., the state data for each object) received from perception detection 302, the location information received from location system 312, sensor data, and/or any other data related to a past and/or current state of an object, the autonomous vehicle, the surrounding environment, and/or relationship(s)”); and
tracking, by the controller, the detected vehicle and the detected traffic line (Paragraph [0103]: “perception detection 302 may also determine, for one or more identified objects in the environment, a current state of the object […] current location; current speed and/or acceleration; current heading; current orientation; size/footprint; type (e.g., vehicle vs. pedestrian vs. bicycle vs. static object or obstacle) […]”).
Ollie fails to teaches wherein detecting the pieces of feature information of the respective vehicles includes: performing, by the controller, semantic segmentation of the image, matching, by the controller, the image on which the semantic segmentation is performed, with the point cloud and detecting a vehicle and a traffic line from the captured image, and tracking, by the controller, the detected vehicle and the detected traffic line.
However, Goel teaches wherein detecting the pieces of feature information of the respective vehicles includes: performing, by the controller, semantic segmentation of the image (Paragraph [0016]: “some or all operations of the example process 100 may be performed by an image-based object detector 102 executing on one or more computing devices. Process 100, and various other examples herein, may be described in reference to performing object detection functionalities (e.g., object identification, classification, instance segmentation, semantic segmentation, and/or object tracking) by an autonomous vehicle operating within an environment”), matching, by the controller, the image on which the semantic segmentation is performed (Paragraph [0029]: “the image-based object detector 102 may generate both a point cloud and a 3D grid in operation 112 corresponding to the same image data 106 (or same region within the image data 106), and may perform both the object detection techniques in the operation 116 based on the point cloud and the object detection techniques in operation 120 based on the 3D grid. Although a neural network 118 and convolutions 122 are shown as illustrative examples, the 3D object detection performed in operation 116 and/or operation 120 can include various combinations of 3D object detection techniques and/or other image analysis functionality, including but not limited to object detection or identification, object classification, instance segmentation, semantic segmentation, object tracking, feature extraction, and/or transformation”; and Paragraph [0054]: “the image-based object detector 102 may perform a processing loop configured to project each pixel within a region of interest in the 2D image data, into a 3D point cloud […]”), with the point cloud and detecting a vehicle and a traffic line from the captured image (Paragraph [0021]: “The image-based object detector 102 may use models and algorithms trained based on lidar and/or radar ground truth data, and/or may be configured based on heuristic-based rules regarding object size averages and distributions (e.g., average lengths for cars, trucks, and bicycles, average height and height distributions for pedestrians, standard sizes for traffic signs, lane markings, sidewalk widths, etc.)”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to modify to perform semantic segmentation of the image section of Ollis to incorporate the use to perform semantic segmentation of the image of Goel and one of ordinary skill in the art would have recognized that the results of the combination were predictable. One could look to Goel to include to perform semantic segmentation of the image section, for example an autonomous vehicle may include an image-based object detector, implemented within one or more on-board vehicle systems (e.g., perception, prediction, planning, etc.), which may be used to improve the object detection capabilities of the autonomous vehicle. In such examples, based on the object detection performed by the image-based object detector (and/or other vehicle systems), the autonomous vehicle may generate a trajectory for controlling the vehicle and navigating within its current environment. The autonomous vehicle also may use the outputs from the image-based object detector to activate a secondary vehicle controller, a remote teleoperations computing device, and/or may engage or disengage certain autonomous driving features, based on the detection and analyses of the objects in the environment.
Claim(s) 5 and 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ollis et al (US 2023/0278581 A1) in view of Goel (US 2023/0033177 A1), further in view of Takamatsu et al (US 2016/0090084 A1).
Regarding claim(s) 5 and 14, Ollis as modified by Goel teaches the apparatus of claim 1, but do not specifically teach wherein the controller is configured to track the detected vehicle and the detected traffic line based on a motion and measurement model (MAMM). However, Takamatsu teaches wherein the controller is configured to track the detected vehicle and the detected traffic line based on a motion and measurement model (MAMM) (Paragraph [0025]: “Additionally, the sensor system 16 is capable of determining the distance from the left, right, front and rear of the vehicle 10 to a road boundary 28 or other stationary or moving objects. For example, the sensor system 16 is capable of detecting the road boundary 28, such as a curb, lane marker, etc., or other stationary or moving objects to the left and right of the vehicle 10. Additionally, the sensor system 16 can include internal sensors capable of determining the steering wheel angle, the steering wheel angular speed and the vehicle speed along the road 26. Based on this information, the controller 14 is capable of calculating the relative position, relative speed, angle of the vehicle 10 relative to the road boundary 28, and estimated future position of the host vehicle 10”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to modify to track the detected vehicle and the detected traffic line based on a motion and measurement model section of Ollis and Goel to incorporate the use to track the detected vehicle and the detected traffic line based on a motion and measurement model section of Takamatsu and one of ordinary skill in the art would have recognized that the results of the combination were predictable. One could look to Takamatsu to include to track the detected vehicle and the detected traffic line based on a motion and measurement model section. The driver assistance system may accurately determine the location of the host vehicle on an electronic map, also any additional information including the current or predicted vehicle position and any past vehicle position or any other suitable information.
Claim(s) 6-9 and 15-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ollis et al (US 2023/0278581 A1) in view of Goel (US 2023/0033177 A1), further in view of Wang (US 2023/0211660 A1).
Regarding claim(s) 6 and 15, Ollis as modified by Goel teaches the apparatus of claim 1, but do not specifically teach further comprising: a storage configured to store a transformer network, training of which is completed. However, Wang teaches a storage configured to store a transformer network, training of which is completed (Paragraph [0024]; and Paragraph [0079]: “Vehicle sensors may collect data during a plurality of driving trips in which a vehicle is driven by a human driver. This data may be transmitted to a server that may store the data as training data”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to modify a storage configured to store a transformer network, training of which is completed section of Ollis to incorporate the use of a storage configured to store a transformer network, training of which is completed section of Wang and one of ordinary skill in the art would have recognized that the results of the combination were predictable. One could look to Wang to include a storage configured to store a transformer network, training of which is completed. The transformer network may then predict a future vehicle acceleration that the driver would perform if they were driving the vehicle manually without the P-ACC system. The P-ACC system may then cause the vehicle to match the predicted acceleration. As such, the P-ACC system may cause the vehicle to automatically drive in a manner that matches how the driver would drive the vehicle during manual driving, thereby making the driver more comfortable with the P-ACC system.
Regarding claim(s) 7 and 16, Ollis as modified by Goel and Wang teaches the apparatus of claim 6, where Wang teaches wherein the controller is configured to predict the driving path of the target vehicle based on the transformer network (Paragraph [0024]: “After sufficient data associated with the driver is collected, the data may be input to a transformer network as training data and the transformer network may be trained based on the training data to predict the vehicle acceleration at a future time step based on the data associated with a current time step. That is, the transformer network may be trained to predict the driver’s driving behavior in a variety of driving situations.”).
Regarding claim(s) 8 and 17, Ollis as modified by Goel and Wang teaches the apparatus of claim 7, where Wang teaches wherein the transformer network is configured to predict positions of the respective vehicles at a future time point based on input vectors of the respective vehicles at a past time point and input vectors of the respective vehicles at a current time point (Paragraph [0024]: “Each time that a driver goes on a driving trip, this type of vectorized data about road geometries, other vehicles on the road, and other data may be collected [...] the data may be input to a transformer network as training data and the transformer network may be trained based on the training data to predict the vehicle acceleration at a future time step based on the data associated with a current time step. That is, the transformer network may be trained to predict the driver’s driving behavior in a variety of driving situations”; and Paragraph [0062]: “the transformer training module 322 may train the transformer network 600 to predict future trajectories of road agents based on training data comprising past driving data collected by the ego vehicle 104”).
Regarding claim(s) 9 and 18, Ollis as modified by Goel and Wang teaches the apparatus of claim 7, where Wang teaches wherein the transformer network is configured to encode pieces of space information of the respective vehicles with respect to driving lanes of the respective vehicles (Paragraph [0030]: “In the illustrated example, the P-ACC server 102 comprises a cloud computing device. In some examples, the P-ACC server 102 may comprise a road-side unit (RSU) positioned near the road 108. In these examples, the system 100 may include any number of RSUs spaced along the road 108 such that each RSU covers a different service area”).
Relevant Prior Art Directed to State of Art
Terazawa (US 2023/0147535 A1) are relevant prior art not applied in the rejection(s) above. Terazawa discloses a vehicle position estimation device mounted on a vehicle and estimating a current vehicle position, the vehicle position estimation device comprising a control unit configured by at least one processor, wherein the control unit includes: a localization unit performing a process for specifying a position of the vehicle on a map based on (i) position information of a landmark detected based on an image frame captured by a front camera and (ii) position information of the landmark registered in the map; and an adverse environment determination unit determining whether a surrounding environment of the vehicle is an adverse environment based on at least one of (i) information output from a sensor equipped to the vehicle or (ii) information output from a communication device equipped to the vehicle, the adverse environment causing a deterioration in an accuracy of object recognition that is performed using the image frame, when the vehicle does not travel in an overtaking lane or an acceleration lane, the control unit outputs, to a vehicle control module that automatically controls a traveling speed of the vehicle, a deceleration request signal to restrict the traveling speed of the vehicle in response to the adverse environment determination unit determining that the surrounding environment of the vehicle is the adverse environment, and when the vehicle travels in an overtaking lane or an acceleration lane, the control unit cancels output of the deceleration request signal in response to the adverse environment determination unit determining that the surrounding environment of the vehicle is the adverse environment.
Lee et al (US 2023/0077393 A1) are relevant prior art not applied in the rejection(s) above. Lee discloses a vehicle comprising: a sensor unit configured to sense a three-dimensional (3D) space; a memory storing one or more instructions; and a processor configured to execute the one or more instructions to: acquire point cloud data for the sensed 3D space, distinguish individual object areas from the acquired point cloud data, acquire object information of an object identified by using an object classification model, track the sensed 3D space using the object information and information related to an object acquired from the individual object areas, and control driving of the vehicle based on information related to the tracked 3D space.
Conclusion
THIS ACTION IS MADE FINAL. 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 JONGBONG NAH whose telephone number is (571)272-1361. The examiner can normally be reached M - F: 7:30am - 4:30pm.
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, Edward Urban can be reached on 571-272-7899. 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.
/JONGBONG NAH/Examiner, Art Unit 2674
/ONEAL R MISTRY/Supervisory Patent Examiner, Art Unit 2674