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 .
Response to Arguments
Applicant's arguments filed 10/07/2025 have been fully considered but they are not persuasive.
Regarding the arguments concerning the simulation versus real world operation, this is not persuasive, the simulation is how Dicle predicts what would happen should it implement a trajectory. I.e. Dicle takes in the current perception/sensing data, run simulations (predicts what would happen into the future) based on that sensor data and possible paths the vehicle could take and based the results thereof (how other objects in the area would respond and the collisions risks thereof). Based on these simulations the safest (lowest risk) path is selected and then implemented (driven in the real-world).
Please see figure 7 posted below which makes clear the operation Dicle.
PNG
media_image1.png
544
560
media_image1.png
Greyscale
As can be seen in figure 7, in step 702 Dicle obtain the current sensor data ([0153]) i.e. obtains realworld data via sensors (output of the perception stack) of the autonomous vehicle of Dicle. In 704 Dicle determines the possible routes (e.g. turn left, turn right, go forwards, etc) ([0155]); Through steps 706-710 the simulation steps and risk evaluations are occurring Dicle then (in steps 706-710) predicts further simulated states (i.e. what would happen in the future should Dicle implement a possible route). It evaluates (for collision risk) those simulate trajectories. Selects (step 712) the safest simulated trajectory. And then in step 714 implements that trajectory/action in the real world ([0176])
As to the arguments concerning the calculation of the probability i.e. teachings of [0122] this argument is not persuasive.
Provided is [0122] and the surrounding paragraphs, the pertinent sections have been highlighted:
“
[0120] In certain cases, the scene evaluation system 504 can vary the path/trajectory of the objects during the simulation as indicated by the object simulation policy. In some such cases, the scene evaluation system can select actions (by the objects) that correspond to an object responding to the actions of the vehicle 200. For example, the scene evaluation system 504 can determine that the objects will react to the vehicle's actions, such as a pedestrian slowing down or speeding up in a crosswalk if the vehicle 200 accelerates, or another vehicle decelerating if the vehicle 200 decelerates, etc.
[0121] In some cases, the scene evaluation system 504 can simulate the actions of the objects in the vehicle scene based on their classification or type as indicated by the object simulation policy. For example, the scene evaluation system 504 can simulate the actions of a pedestrian different from the actions of a vehicle or bicycle. As a non-limiting example, the object simulation policy can assign a higher probability that a pedestrian will change direction or stop as compared to a vehicle or bicycle. In some cases, a pedestrian may have a “stop” action within a particular time period (e.g., 500 ms), whereas a vehicle may not (e.g., it may not be able to stop within that particular time period given its current velocity).
[0122] In simulating actions for the different objects, the scene evaluation system 504 can treat each object as an independent actor (e.g., each object may act independent of the other objects in the scene) as indicated by the object simulation policy and/or simulation policy. For example, the scene evaluation system 504 may estimate that different pedestrians will act differently based on their size (e.g., child may act differently than adult, such as by being more likely to accelerate or change direction in a short time period (e.g., <1 sec.)), location (e.g., the actions of pedestrians on a crosswalk may be dependent on the vehicle's actions (e.g., pedestrian is likely to speed up/slow down depending on what the vehicle 200 does), whereas the actions of pedestrians on a sidewalk may be independent of the vehicle's actions), speed (e.g., pedestrians running may be less likely to stop/move in the opposite direction (or otherwise make a significant change in direction (e.g., >60 degrees) within a particular time period (e.g., <500 ms) than pedestrians that are walking). Similarly, the scene evaluation system can take into account the features of other vehicles (e.g., size, position relative to vehicle 200, velocity, etc.) or bicycles (e.g., size, position relative to vehicle 200, velocity, etc.) in simulating their actions. For example, the scene evaluation system 504 may treat vehicles traveling in the same direction and in front of the vehicle 200 as acting independent of the vehicle's actions but may treat vehicles traveling in the same direction and behind (or to the side) as taking actions dependent on the vehicle's actions.
[0123] In certain cases, the scene evaluation system 504 can use a trained neural network to estimate or select the actions of the objects in the vehicle scene as indicated by the object simulation policy. The trained neural network may use different classifications of objects to simulate the actions the objects may take. For example, the trained neural network can select an action for vehicles based on learned characteristics of vehicles. Similarly, the trained neural network can select an action for bicycles and/or pedestrians based on learned characteristics of the bicycles and pedestrians. “
As can be seen from [0121] of Dicle “As a non-limiting example, the object simulation policy can assign a higher probability that a pedestrian will change direction or stop as compared to a vehicle or bicycle.”
A probability is assigned (that a pedestrian will change directions in response to the vehicle’s action), i.e. A probability of the response to a given trajectory/simulation states being explored. The following sections [0122] is teaching basically a further modification (raising/lowering of the class probability) based on the state (e.g. running versus walking) and/or location (e.g. on sidewalk). As such the applicant’s arguments concerning the determining. Sections [0121]-[0123] are already cited as part of the rejections for the (probability of) the object altering its path.
As such the grounds for rejection over Dicle are maintained. An updated rejection to reflect the amended claim language appears below.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1, 3-10 and 12-20 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by US 20230322266 A1, “VEHICLE ACTION SELECTION BASED ON SIMULATED STATES”, Dicle et al.
Regarding Claim 1, Dicle et al teaches “A computer-implemented method, comprising: generating a perception output, using a perception stack of an autonomous vehicle (AV), wherein the perception output identifies at least one dynamic entity in an environment;”( [0153] At block 702, the planning system 404 obtains scene data associated with a scene of a vehicle 200. As described herein, the scene data can include vehicle data associated with the vehicle 200 and/or object data associated with one or more objects in the scene. The data can be generated from one or more sensors from sensor data associated a sensor suite of the with the perception system 402, sensor data from the localization system 406, and/or one or more sensors in or around the vehicle 200 that are specific to the vehicle 200. The scene data can include data related to the position, orientation, heading, velocity, acceleration, of the vehicle 200 or objects in the scene, the amount of acceleration or deceleration of the vehicle 200, steering wheel position of the vehicle 200, etc.);” receiving, by a prediction module, the perception output, and generating by the prediction module, a projected dynamic entity path of the at least one dynamic entity;”(see [0090]-[0092] Here teaches the detecting/sending of sensing information to generate the scene which includes the information of other (dynamic) entities around the vehicle + [0118] the simulations are based in part of the dynamic entities predicted trajectories);”determining, by the prediction module, a plurality of projected trajectories for the autonomous vehicle (AV);”( [0096] The simulation policy (or policies) 510 can include one or more policies to indicate which vehicle actions to simulate (e.g., vehicle action policy), which simulated states to generate, explore, and expand (e.g., state selection policy), how to simulate the vehicle during a trajectory simulation (e.g., vehicle simulation policy), how to simulate objects in the scene during a trajectory simulation (e.g., object simulation policy), when to end a trajectory simulation (e.g., end state policy), how to evaluate a trajectory simulation (e.g., trajectory evaluation policy), and/or when to end the state/trajectory simulations (e.g., simulation termination policy). Accordingly, the scene evaluation system 504 can use the simulation policy 510 to evaluate the simulated trajectories and simulated scene states and determine the scores 512 for the simulated trajectories and simulated states. Although reference herein may be made to one or more policies or sub-policies of the simulation policy 510, it will be understood that the policies individually or in the aggregate can be referred to as the simulation policy 510.” Here teaches the simulating of the autonomous vehicle’s potential/possible actions (trajectories) based on the simulated scene(s) (i.e. the simulations are of multiple potential trajectories in order to determine/select a projected trajectory);” and calculating, by the prediction module, a risk metric for each potential trajectory of the AV based on the perception output and the projected dynamic entity path of the at least one dynamic entity;,”( [0098] The action selector 506 can select an action to be taken by the vehicle 200 based on the results of the scene evaluation system. In some cases, the action selector 506 can select the action based on a vehicle planning policy. The vehicle planning policy can indicate how the action selector 506 is to evaluate the results for the scene evaluation system. In some cases, the evaluation of the results can include a review of the scores associated with one or more simulated states or simulated trajectories. In certain cases, the evaluation can include a review of the scores of simulated states that result from an action taken from the scene state (also referred to herein as child states or first-level simulated states). In some such cases, the action selector 506 can select an action that corresponds to one of the first-level simulated states. In certain cases, the action selector 506 can select that action that corresponds to the first-level simulated state with the highest score, highest aggregate score, highest average score, highest single score, or for which the most trajectories have been simulated, etc.
[0099] It will be understood that the action selector 506 can select an action based on any number of criteria. In some cases, the action selector 506 can select an action based on a safety or comfort score. The safety and/or comfort score can be based on one or more safety threshold and/or comfort thresholds (similar to those described herein with reference to the vehicle action policy). In some such cases, the action selector 506 can select an action that is determined to be the safest or most comfortable and/or use these features to select an action from a number of similarly scored actions/simulated states.” Each simulation/scene (with associated potential trajectory) is assessed to be given a risk score + see figure 7 in particular steps 704-710 (paragraphs [0156]-[0168]);” wherein the risk metric comprises an unrealized risk score and wherein the unrealized risk score is based on a probability of future collision between the AV and the at least one dynamic entity along the corresponding potential trajectory for the AV and wherein the probability of future collision between the AV and the at least one dynamic entity includes determining a probability that the at least one dynamic entity will alter the dynamic entity path to avoid a collision;”( [0122] gives examples of object (dynamic entity) simulation policies which include/teaches that the simulation accounts for/associates a probability that an object will change its behavior (change path to avoid collision) based in part of the object’s classification, location, speed, etc) from the phrasing “More” or “less” “likely” it is known that the responsive behaviors/changes in the moving objects (pedestrians) paths are probability based + [0128] The thresholds can vary for the different features. For example, the threshold for a feature related to colliding with an object, a determination that the trajectory would (or would likely, e.g., >50%) result in a collision with an object would cause the scene evaluation system 504 to classify the trajectory as unsuccessful (or give it a low or failing score). As another example, the threshold for veering left or right may be based on the velocity of the vehicle 200. For example, a relatively larger degree turn threshold may be used at lower velocities and relatively smaller degree turn threshold may be used at higher velocities given that the centrifugal forces at the lower velocity will be lower. Accordingly, in certain cases, the scene evaluation system 504 can compare an estimated centrifugal force along a trajectory with a threshold centrifugal force to evaluate the simulated trajectory 621A.” Here teaches that the safety score/feature is based on/relates to the collision probability + [0120] In certain cases, the scene evaluation system 504 can vary the path/trajectory of the objects during the simulation as indicated by the object simulation policy. In some such cases, the scene evaluation system can select actions (by the objects) that correspond to an object responding to the actions of the vehicle 200. For example, the scene evaluation system 504 can determine that the objects will react to the vehicle's actions, such as a pedestrian slowing down or speeding up in a crosswalk if the vehicle 200 accelerates, or another vehicle decelerating if the vehicle 200 decelerates, etc.” Here teaches that the simulation of other object’s behaviors (and thus their corresponding predicted trajectories) includes their reactions of the autonomous vehicle, and in this example gives of a pedestrian adjusting (slowing up/down) and/or other vehicles decelerating to avoid the autonomous vehicle, thus the resulting safety scores/collision probabilities account for that other objects can/would avoid the autonomous vehicle + see [0121]-[0123] which teach that the simulation/response of other objects (i.e. teachings of [0120]) are probability based include an objects ability to change direction/response (i.e. “successfully divert”; [0121] teaches assigning a probability that a pedestrian will change its path) in response to the vehicle’s actions and the examples of the types of response are implicitly/include changing the objects path (e.g. speeding up/slowing down) to avoid hitting the vehicle. + [0128] the risk score/collision score is based on the simulated responses (diverting) of other objects, thus this score/risk is includes determining/accounting for if the responses of the other objects are “successful” in avoiding collision.) ;” identifying a potential trajectory for the AV having a lowest risk score from the plurality of potential trajectories, and selecting the potential trajectory for the AV as the projected trajectory for the AV”( ([0098]-[0100] Selected actions (AV’s potential path) are/include selectin the “safest” action (lowest risk) and those selected actions are used/implemented by the AV 9selected as projected path) + [0174] the AV selects/implements the action (potential trajectory) based on criteria which earlier in [0099] is known to include/be the safest action/path);” and operating the AV, using a control stack, along the projected trajectory”( [0176] At block 714, causes the vehicle 200 to perform the action. In some cases, the selected action can correspond to the action associated with the selected simulated state (e.g., the action that if taken by the vehicle 200 results in the simulated state). In certain cases, the planning system communicates the selected action to the control system 408, which controls the vehicle 200 to execute the action.)
Regarding Claim 3, Dicle et al teaches “The computer-implemented method of claim 2, further comprising: identifying, by the prediction module, the risk metric corresponding to the projected trajectory among the plurality of projected trajectories having the lowest unrealized risk score.”( [0098] The action selector 506 can select an action to be taken by the vehicle 200 based on the results of the scene evaluation system. In some cases, the action selector 506 can select the action based on a vehicle planning policy. The vehicle planning policy can indicate how the action selector 506 is to evaluate the results for the scene evaluation system. In some cases, the evaluation of the results can include a review of the scores associated with one or more simulated states or simulated trajectories. In certain cases, the evaluation can include a review of the scores of simulated states that result from an action taken from the scene state (also referred to herein as child states or first-level simulated states). In some such cases, the action selector 506 can select an action that corresponds to one of the first-level simulated states. In certain cases, the action selector 506 can select that action that corresponds to the first-level simulated state with the highest score, highest aggregate score, highest average score, highest single score, or for which the most trajectories have been simulated, etc.
[0099] It will be understood that the action selector 506 can select an action based on any number of criteria. In some cases, the action selector 506 can select an action based on a safety or comfort score. The safety and/or comfort score can be based on one or more safety threshold and/or comfort thresholds (similar to those described herein with reference to the vehicle action policy). In some such cases, the action selector 506 can select an action that is determined to be the safest or most comfortable and/or use these features to select an action from a number of similarly scored actions/simulated states. [0098] teaches selecting/implementing the “highest score” and from [0099] it is known that the score can be a safety (inverse of collision risk) score thus selecting the highest safety score trajectory is the selecting of the lowest collision risk route/trajectory which naturally requires the identification of the “highest” score trajectory in order to implement it)
Regarding Claim 4, Dicle et al teaches “The computer-implemented method of claim 2, further comprising: providing, by the prediction module, the risk metric having the lowest unrealized risk score to the planning module to cause the planning module to select a navigation path based on the risk metric.”( [0098] The action selector 506 can select an action to be taken by the vehicle 200 based on the results of the scene evaluation system. In some cases, the action selector 506 can select the action based on a vehicle planning policy. The vehicle planning policy can indicate how the action selector 506 is to evaluate the results for the scene evaluation system. In some cases, the evaluation of the results can include a review of the scores associated with one or more simulated states or simulated trajectories. In certain cases, the evaluation can include a review of the scores of simulated states that result from an action taken from the scene state (also referred to herein as child states or first-level simulated states). In some such cases, the action selector 506 can select an action that corresponds to one of the first-level simulated states. In certain cases, the action selector 506 can select that action that corresponds to the first-level simulated state with the highest score, highest aggregate score, highest average score, highest single score, or for which the most trajectories have been simulated, etc.
[0099] It will be understood that the action selector 506 can select an action based on any number of criteria. In some cases, the action selector 506 can select an action based on a safety or comfort score. The safety and/or comfort score can be based on one or more safety threshold and/or comfort thresholds (similar to those described herein with reference to the vehicle action policy). In some such cases, the action selector 506 can select an action that is determined to be the safest or most comfortable and/or use these features to select an action from a number of similarly scored actions/simulated states.” [0098] teaches selecting/implementing the “highest score” and from [0099] it is known that the score can be a safety (inverse of collision risk) score thus selecting the highest safety score trajectory is the selecting of the lowest collision risk route/trajectory)
Regarding Claim 5, Dicle et al teaches “The computer-implemented method of claim 1, wherein determining the projected trajectory further comprises: determining a location of the AV; and computing the projected trajectory based on the location of the AV and a navigation intent of the AV.”([0020] + [0063] trajectories are based in part on the current state (location) of the AV and its destination/final state (i.e. navigation intent))
Regarding Claim 6, Dicle et al teaches “The computer-implemented method of claim 1, wherein the risk metric is based on kinematic characteristics of the at least one dynamic entity.”( ([0092] “The state data associated with the objects (individually or collectively referred to as object state data) in the scene can be obtained from the perception system 402 (or other source) and be based on data obtained from a sensor suite that includes one or more devices such as cameras 202a, LiDAR sensors 202b, radar sensors 202c, and microphones 202d, or communications with the objects (e.g., wireless communications with other vehicles), etc. As described herein, the object state data can include any one or any combination of, acceleration, velocity, position (relative to vehicle 200 or absolute/geographic), orientation/heading, classification, or size, etc., of the objects.” + [0096] The simulation policy (or policies) 510 can include one or more policies to indicate which vehicle actions to simulate (e.g., vehicle action policy), which simulated states to generate, explore, and expand (e.g., state selection policy), how to simulate the vehicle during a trajectory simulation (e.g., vehicle simulation policy), how to simulate objects in the scene during a trajectory simulation (e.g., object simulation policy), when to end a trajectory simulation (e.g., end state policy), how to evaluate a trajectory simulation (e.g., trajectory evaluation policy), and/or when to end the state/trajectory simulations (e.g., simulation termination policy). Accordingly, the scene evaluation system 504 can use the simulation policy 510 to evaluate the simulated trajectories and simulated scene states and determine the scores 512 for the simulated trajectories and simulated states. Although reference herein may be made to one or more policies or sub-policies of the simulation policy 510, it will be understood that the policies individually or in the aggregate can be referred to as the simulation policy 510.” Here teaches that the predict (object) trajectories are simulated/evaluated + [0097] teaches the safety score of the trajectories can be evaluated based/in regards to the (predicted) behavior of other objects)
Regarding Claim 7, Dicle et al teaches “The computer-implemented method of claim 1, wherein the risk metric is used to calculate a new trajectory for the AV.”( ([0099] It will be understood that the action selector 506 can select an action based on any number of criteria. In some cases, the action selector 506 can select an action based on a safety or comfort score. The safety and/or comfort score can be based on one or more safety threshold and/or comfort thresholds (similar to those described herein with reference to the vehicle action policy). In some such cases, the action selector 506 can select an action that is determined to be the safest or most comfortable and/or use these features to select an action from a number of similarly scored actions/simulated states.” Here the safety score of the simulated trajectories is used to select (calculate/implement) the new trajectory)
Regarding Claim 8, it is a system (processor) equivalent to the method claim 1 above. It has the same grounds of rejection.
Regarding Claim 9, Dicle et al teaches “The system of claim 8, wherein the perception output is received from a perception module of an AV stack.”( [0092] The state data associated with the objects (individually or collectively referred to as object state data) in the scene can be obtained from the perception system 402 (or other source) and be based on data obtained from a sensor suite that includes one or more devices such as cameras 202a, LiDAR sensors 202b, radar sensors 202c, and microphones 202d, or communications with the objects (e.g., wireless communications with other vehicles), etc. As described herein, the object state data can include any one or any combination of, acceleration, velocity, position (relative to vehicle 200 or absolute/geographic), orientation/heading, classification, or size, etc., of the objects.)
Regarding Claim 10, Dicle et al teaches “The system of claim 8, wherein the perception output is based on sensor data collected by one or more environmental sensors of the AV.”( [0092] The state data associated with the objects (individually or collectively referred to as object state data) in the scene can be obtained from the perception system 402 (or other source) and be based on data obtained from a sensor suite that includes one or more devices such as cameras 202a, LiDAR sensors 202b, radar sensors 202c, and microphones 202d, or communications with the objects (e.g., wireless communications with other vehicles), etc. As described herein, the object state data can include any one or any combination of, acceleration, velocity, position (relative to vehicle 200 or absolute/geographic), orientation/heading, classification, or size, etc., of the objects.)
Regarding Claim 12, Dicle et al teaches “The system of claim 8, wherein determining the projected trajectory further comprises: determining a location of the AV; and computing the projected trajectory based on the location of the AV and a navigation intent of the AV.” ( [0063] In some embodiments, planning system 404 receives data associated with a destination and generates data associated with at least one route (e.g., routes 106) along which a vehicle (e.g., vehicles 102) can travel along toward a destination. In some embodiments, planning system 404 periodically or continuously receives data from perception system 402 (e.g., data associated with the classification of physical objects, described above) and planning system 404 updates the at least one trajectory or generates at least one different trajectory based on the data generated by perception system 402. In some embodiments, planning system 404 receives data associated with an updated position of a vehicle (e.g., vehicles 102) from localization system 406 and planning system 404 updates the at least one trajectory or generates at least one different trajectory based on the data generated by localization system 406.” The trajectory/trajectories are generated based on the destination (intent) )
Regarding Claim 13, Dicle et al teaches “The system of claim 8, wherein the risk metric is based on kinematic characteristics of the at least one dynamic entity.”( ([0092] “The state data associated with the objects (individually or collectively referred to as object state data) in the scene can be obtained from the perception system 402 (or other source) and be based on data obtained from a sensor suite that includes one or more devices such as cameras 202a, LiDAR sensors 202b, radar sensors 202c, and microphones 202d, or communications with the objects (e.g., wireless communications with other vehicles), etc. As described herein, the object state data can include any one or any combination of, acceleration, velocity, position (relative to vehicle 200 or absolute/geographic), orientation/heading, classification, or size, etc., of the objects.” + [0096] The simulation policy (or policies) 510 can include one or more policies to indicate which vehicle actions to simulate (e.g., vehicle action policy), which simulated states to generate, explore, and expand (e.g., state selection policy), how to simulate the vehicle during a trajectory simulation (e.g., vehicle simulation policy), how to simulate objects in the scene during a trajectory simulation (e.g., object simulation policy), when to end a trajectory simulation (e.g., end state policy), how to evaluate a trajectory simulation (e.g., trajectory evaluation policy), and/or when to end the state/trajectory simulations (e.g., simulation termination policy). Accordingly, the scene evaluation system 504 can use the simulation policy 510 to evaluate the simulated trajectories and simulated scene states and determine the scores 512 for the simulated trajectories and simulated states. Although reference herein may be made to one or more policies or sub-policies of the simulation policy 510, it will be understood that the policies individually or in the aggregate can be referred to as the simulation policy 510.” Here teaches that the predict (object) trajectories are simulated/evaluated + [0097] teaches the safety score of the trajectories can be evalulated based/in regards to the (predicted) behavior of other objects)
Regarding Claim 14, Dicle et al teaches “The system of claim 8, wherein the risk metric is used to calculate a new trajectory for the AV.”( [0099] It will be understood that the action selector 506 can select an action based on any number of criteria. In some cases, the action selector 506 can select an action based on a safety or comfort score. The safety and/or comfort score can be based on one or more safety threshold and/or comfort thresholds (similar to those described herein with reference to the vehicle action policy). In some such cases, the action selector 506 can select an action that is determined to be the safest or most comfortable and/or use these features to select an action from a number of similarly scored actions/simulated states.” Here the safety score of the simulated trajectories is used to select (calculate/implement) the new trajectory)
Regarding Claim 15, it is a non-transitory computer readable medium equivalent to claim 1 above. It has the same grounds of rejection.
Regarding Claim 16, Dicle et al teaches “The non-transitory computer-readable storage medium of The non-transitory computer-readable storage medium of wherein the perception output is received from a perception module of an AV stack.”( [0092] The state data associated with the objects (individually or collectively referred to as object state data) in the scene can be obtained from the perception system 402 (or other source) and be based on data obtained from a sensor suite that includes one or more devices such as cameras 202a, LiDAR sensors 202b, radar sensors 202c, and microphones 202d, or communications with the objects (e.g., wireless communications with other vehicles), etc. As described herein, the object state data can include any one or any combination of, acceleration, velocity, position (relative to vehicle 200 or absolute/geographic), orientation/heading, classification, or size, etc., of the objects.)
Regarding Claim 17, Dicle et al teaches “he non-transitory computer-readable storage medium of claim 15, wherein the perception output is based on sensor data collected by one or more environmental sensors of the AV.”( [0092] The state data associated with the objects (individually or collectively referred to as object state data) in the scene can be obtained from the perception system 402 (or other source) and be based on data obtained from a sensor suite that includes one or more devices such as cameras 202a, LiDAR sensors 202b, radar sensors 202c, and microphones 202d, or communications with the objects (e.g., wireless communications with other vehicles), etc. As described herein, the object state data can include any one or any combination of, acceleration, velocity, position (relative to vehicle 200 or absolute/geographic), orientation/heading, classification, or size, etc., of the objects.)
Regarding Claim 18, Dicle et al teaches “The non-transitory computer-readable storage medium of claim 17, wherein the one or more environmental sensors comprises one or more of: a Light Detection and Ranging (LiDAR) sensor, a camera sensor, and a radar sensor.”( [0092] The state data associated with the objects (individually or collectively referred to as object state data) in the scene can be obtained from the perception system 402 (or other source) and be based on data obtained from a sensor suite that includes one or more devices such as cameras 202a, LiDAR sensors 202b, radar sensors 202c, and microphones 202d, or communications with the objects (e.g., wireless communications with other vehicles), etc. As described herein, the object state data can include any one or any combination of, acceleration, velocity, position (relative to vehicle 200 or absolute/geographic), orientation/heading, classification, or size, etc., of the objects.)
Regarding Claim 19, Dicle et al teaches “The non-transitory computer-readable storage medium of The non-transitory computer-readable storage medium of wherein determining the projected trajectory further comprises: determining a location of the AV; and computing the projected trajectory based on the location of the AV and a navigation intent of the AV.”( [0063] In some embodiments, planning system 404 receives data associated with a destination and generates data associated with at least one route (e.g., routes 106) along which a vehicle (e.g., vehicles 102) can travel along toward a destination. In some embodiments, planning system 404 periodically or continuously receives data from perception system 402 (e.g., data associated with the classification of physical objects, described above) and planning system 404 updates the at least one trajectory or generates at least one different trajectory based on the data generated by perception system 402. In some embodiments, planning system 404 receives data associated with an updated position of a vehicle (e.g., vehicles 102) from localization system 406 and planning system 404 updates the at least one trajectory or generates at least one different trajectory based on the data generated by localization system 406.” The trajectory/trajectories are generated based on the destination (intent) )
Regarding Claim 20, Dicle et al teaches “The non-transitory computer-readable storage medium of claim 15, wherein the risk metric is based on kinematic characteristics of the at least one dynamic entity.” ([0092] “The state data associated with the objects (individually or collectively referred to as object state data) in the scene can be obtained from the perception system 402 (or other source) and be based on data obtained from a sensor suite that includes one or more devices such as cameras 202a, LiDAR sensors 202b, radar sensors 202c, and microphones 202d, or communications with the objects (e.g., wireless communications with other vehicles), etc. As described herein, the object state data can include any one or any combination of, acceleration, velocity, position (relative to vehicle 200 or absolute/geographic), orientation/heading, classification, or size, etc., of the objects.” + [0096] The simulation policy (or policies) 510 can include one or more policies to indicate which vehicle actions to simulate (e.g., vehicle action policy), which simulated states to generate, explore, and expand (e.g., state selection policy), how to simulate the vehicle during a trajectory simulation (e.g., vehicle simulation policy), how to simulate objects in the scene during a trajectory simulation (e.g., object simulation policy), when to end a trajectory simulation (e.g., end state policy), how to evaluate a trajectory simulation (e.g., trajectory evaluation policy), and/or when to end the state/trajectory simulations (e.g., simulation termination policy). Accordingly, the scene evaluation system 504 can use the simulation policy 510 to evaluate the simulated trajectories and simulated scene states and determine the scores 512 for the simulated trajectories and simulated states. Although reference herein may be made to one or more policies or sub-policies of the simulation policy 510, it will be understood that the policies individually or in the aggregate can be referred to as the simulation policy 510.” Here teaches that the predict (object) trajectories are simulated/evaluated + [0097] teaches the safety score of the trajectories can be evaluated based/in regards to the (predicted) behavior of other objects)
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 KENNETH MICHAEL DUNNE whose telephone number is (571)270-7392. The examiner can normally be reached Mon-Thurs 8:30-6:30.
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, Navid Z Mehdizadeh can be reached at (571) 272-7691. 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.
/KENNETH M DUNNE/Primary Examiner, Art Unit 3669
Prosecution Insights