TRAJECTORY PLANNING FOR AUTONOMOUS VEHICLE WITH PARTICLE SWARM OPTIMIZATION

Application No. 18/617,570

• Filed: Mar 26, 2024
• Examiner: PARK, KYLE S
• Art Unit: 3666
• Tech Center: 3600 — Transportation & Electronic Commerce
• Assignee: Honda Motor Co. Ltd.
• OA Round: 3 (Non-Final)

Office Action

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 .

Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 01/27/2026 has been entered.

Status of the Claims
This action is in response to the applicant’s amendment/response and RCE of January 27, 2026.
Claims 1-11 and 13-20 are pending and have been considered as follows.

Response to Arguments
Applicant’s arguments/amendments with respect to the rejection of claims under 35 USC § 103 have been fully considered and are not persuasive. As to amended claim 1, Applicant argues, broadly, that “Applicant contends that Zhang does not remedy the shortcomings of Subramanian with respect to claim 1, and that the cited references therefore do not support a prima facie case of obviousness with respect to claim 1. The Office Action cites paragraphs 3-4 and 56-58 of Zhang for allegedly disclosing a particle swarm optimization algorithm based on self-cognition and social influence in simulating the foraging behavior of bird flocks. However, even if the simulating the foraging behavior of bird flocks of Zhang are somehow taken to be a particle swarm optimization algorithm based on self-cognition and social influence, Zhang does not disclose acceleration coefficients reflecting self-cognition and social influence. As such, claim 1 should be found to define over the cited references.” Accordingly, Applicant argues that ZHANG is silent as to the following claim limitations:
“generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm based on acceleration coefficients reflecting self-cognition and social influence”.

The Examiner respectfully disagrees. ZHANG renders obvious the claim limitations at issue. ZHANG explicitly discloses that particle swarm optimization is a random search algorithm based on group collaboration proposed by simulating the foraging behavior of bird flocks. In the path planning of drones, each particle represents a planned path. The algorithm is initialized with a number of random particles. In each iteration, the particles tend to the optimal particle position in the population and the individual historical optimal position, realizing the path optimization process (see at least paragraphs 3-4 and 56-58). Further, ZHANG explicitly discloses that calculating the path cost of each particle, select the particle position with the lowest global path cost value as g_best, select the particle position with the lowest path cost value in each group as gr_i,best, and record the lowest historical path cost value position of each particle in the iteration process as p_i,best. The particle position update formula of the improved particle swarm algorithm is: Vi (t + 1) = ω Vi (t) + c1r1 (pi, best-pi) + c2r2 (gbest-pi) + c3r3 (gri, best-pi) pi (t + 1) = pi (t) + Vi (t + 1). Wherein, t represents the number of iterations; V_i(t) represents the velocity of the ith particle; p_i=p_i(t) represents the position of the ith particle; gr_i,best represents the optimal particle position of the ith particle corresponding to the group; ω represents the inertia weight; c₁, c₂, c₃ represent the acceleration factors corresponding to different parts; r₁, r₂, r₃ are all real numbers between 0 and 1 (see at least paragraphs 73-77. See also at least paragraphs 104-115). Therefore, the Examiner respectfully submits that the mapping of ZHANG to Applicant’s claimed invention is appropriate. Accordingly, the claim rejections under § 103 are maintained.

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.

Claim(s) 1, 13, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Subramanian et al., US 2022/0176995 A1, hereinafter referred to as Subramanian, in view of ZHAO et al., CN 112109704 A, hereinafter referred to as ZHAO, and further in view of ZHANG et al., CN 115657725 A, hereinafter referred to as ZHANG, respectively.
As to claim 1, Subramanian teaches a method for generating and controlling a trajectory of an autonomous vehicle, comprising:
generating an initial trajectory of the autonomous vehicle (see at least Abstract regarding sensor data describing an environment of an autonomous vehicle and an initial travel path for the autonomous vehicle through the environment can be obtained. See also at least paragraphs 23 and 30-31 regarding the onboard computing system of an autonomous vehicle (e.g., vehicle computing system) can obtain an initial travel path describing a path for the autonomous vehicle to travel from the first position to the second position, Subramanian);
estimating future positions of other vehicles in an environment surrounding the autonomous vehicle using a trained neural network model (see at least paragraphs 27-28 regarding receiving sensor data about the environment, perceive objects within the vehicle's surrounding environment (e.g., other vehicles), predict the motion of the objects within the surrounding environment, generate trajectories based on the sensor data and/or perceived/predicted motion of the objects, and, based on the trajectory, transmit control signals to a vehicle control system to enable the autonomous vehicle to progress to a target destination. See also at least paragraphs 88-90 regarding the vehicle computing system 110 can be configured to predict a motion of the object(s) within the surrounding environment of the vehicle 105. For instance, the vehicle computing system 110 can generate prediction data 175B associated with such object(s). The prediction data 175B can be indicative of one or more predicted future locations of each respective object. … The vehicle computing system 110 can utilize one or more algorithms and/or machine-learned model(s) that are configured to predict the future motion of object(s) based at least in part on the sensor data 155, the perception data 175A, map data 160, and/or other data. This can include, for example, one or more neural networks trained to predict the motion of the object(s) within the surrounding environment of the vehicle 105 based at least in part on the past and/or current state(s) of those objects as well as the environment in which the objects are located (e.g., the lane boundary in which it is travelling, etc.), Subramanian); and
transmitting control signals to a vehicle actuator system to cause the vehicle actuator system to autonomously control the autonomous vehicle to travel according to the trajectory (see at least paragraph 66 regarding the vehicle computing system can control the motion of the autonomous vehicle based, at least in part, on the selected trajectory. For instance, once a trajectory has been chosen, the vehicle computing system can transmit the motion to a vehicle controller. The vehicle controller can use the selected trajectory to generate one or more motion controls for the autonomous vehicle. … The vehicle controller can transmit those motion controls to the autonomous vehicle to be executed and follow the selected trajectory, Subramanian).
Subramanian does not explicitly teach generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm based on acceleration coefficients reflecting self-cognition and social influence; fitting a polynomial curve to the dynamically feasible trajectory; or converting the polynomial curve into reference waypoints and generating the trajectory based on the reference waypoints.
However, ZHAO teaches generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm (see at least paragraph 5 regarding in order to ensure that the vehicle accurately avoids obstacle vehicles, it is first necessary to predict the movement trajectories of its own vehicle and the obstacle vehicles, so as to provide a basis for collision avoidance path planning. See also at least paragraphs 124-131. See also at least paragraphs 133-137 regarding based on the three-degree-of-freedom vehicle dynamics model and combined with the CTRA kinematic model, the vehicle's driving trajectory is accurately predicted in the short-term domain. Build an LSTM neural network to accurately predict the driving trajectories of the vehicle and the obstacle vehicle in the long term, and output the predicted data of the driving trajectories of the vehicle and the obstacle vehicle. Based on the particle swarm algorithm, the weights of the driving trajectory prediction results based on the three-degree-of-freedom vehicle dynamics model and the driving trajectory prediction results based on the LSTM neural network are optimized online. The prediction results of the short-term and long-term domains are integrated according to the optimized weights to accurately predict the vehicle's driving trajectory); fitting a polynomial curve to the dynamically feasible trajectory (see at least paragraphs 124-131 regarding Step 4: Integrate grid division, artificial potential field method and high-order polynomial curve fitting method to build a vehicle dynamic safety path planning model. See also at least paragraphs 175-199 regarding according to the position of each node in the collision avoidance path, a smooth collision avoidance path is obtained by using high-order polynomial curve fitting); and converting the polynomial curve into reference waypoints (see at least paragraphs 124-131 regarding Step 4: Integrate grid division, artificial potential field method and high-order polynomial curve fitting method to build a vehicle dynamic safety path planning model. See also at least FIG. 6 and paragraphs 175-199 regarding according to the position of each node in the collision avoidance path, a smooth collision avoidance path is obtained by using high-order polynomial curve fitting. The high-order polynomial curve fitting uses a quintic polynomial curve to perform segmented fitting on the preliminary collision avoidance path and ensures smooth transition between each segment of the curve) and generating the trajectory based on the reference waypoints (see at least FIG. 6 and paragraphs 124-131 regarding Step 5: Based on the driving trajectory prediction data of the vehicle and the obstacle vehicle obtained in step 3, the vehicle dynamic safety path planning model is used to obtain the optimal collision avoidance path 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 use the system of ZHAO which teaches generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm; fitting a polynomial curve to the dynamically feasible trajectory; and converting the polynomial curve into reference waypoints and generating the trajectory based on the reference waypoints with the system of Subramanian as both systems are directed to a system and method for generating an optimal trajectory based on the an environment of a vehicle, and one of ordinary skill in the art would have recognized the established utility of generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm; fitting a polynomial curve to the dynamically feasible trajectory; and converting the polynomial curve into reference waypoints and generating the trajectory based on the reference waypoints and would have predictably applied it to improve the system of Subramanian.
Subramanian, as modified by ZHAO, does not explicitly teach generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm based on acceleration coefficients reflecting self-cognition and social influence.
However, such matter is taught by ZHANG (see at least paragraphs 3-4 and 56-58 regarding particle swarm optimization is a random search algorithm based on group collaboration proposed by simulating the foraging behavior of bird flocks. In the path planning of drones, each particle represents a planned path. The algorithm is initialized with a number of random particles. In each iteration, the particles tend to the optimal particle position in the population and the individual historical optimal position, realizing the path optimization process. See also at least paragraphs 73-77 regarding calculating the path cost of each particle, select the particle position with the lowest global path cost value as g_best, select the particle position with the lowest path cost value in each group as gr_i,best, and record the lowest historical path cost value position of each particle in the iteration process as p_i,best. The particle position update formula of the improved particle swarm algorithm is: Vi (t + 1) = ω Vi (t) + c1r1 (pi, best-pi) + c2r2 (gbest-pi) + c3r3 (gri, best-pi) pi (t + 1) = pi (t) + Vi (t + 1). Wherein, t represents the number of iterations; V_i(t) represents the velocity of the ith particle; p_i=p_i(t) represents the position of the ith particle; gr_i,best represents the optimal particle position of the ith particle corresponding to the group; ω represents the inertia weight; c₁, c₂, c₃ represent the acceleration factors corresponding to different parts; r₁, r₂, r₃ are all real numbers between 0 and 1. See also at least paragraphs 99-115).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of ZHANG which teaches generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm based on acceleration coefficients reflecting self-cognition and social influence with the system of Subramanian, as modified by ZHAO, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm based on acceleration coefficients reflecting self-cognition and social influence and would have predictably applied it to improve the system of Subramanian as modified by ZHAO.
As to claim 13, Examiner notes claim 13 recites similar limitations to claim 1 and is rejected under the same rational.
As to claim 20, Subramanian teaches a vehicle capable of autonomous driving, comprising:
a vehicle sensor system (see at least FIG. 1, Subramanian);
a vehicle actuator system (see at least FIG. 1, Subramanian); and
a vehicle electronic control unit in communication with the vehicle sensor system and the vehicle actuator system, the vehicle electronic control unit being programmed to (see at least FIG. 1, Subramanian):
generate an initial trajectory of the vehicle for autonomously driving the vehicle (see at least Abstract regarding sensor data describing an environment of an autonomous vehicle and an initial travel path for the autonomous vehicle through the environment can be obtained. See also at least paragraphs 23 and 30-31 regarding the onboard computing system of an autonomous vehicle (e.g., vehicle computing system) can obtain an initial travel path describing a path for the autonomous vehicle to travel from the first position to the second position, Subramanian);
estimate, based on received sensor data from the vehicle sensor system, future positions of other vehicles in an environment surrounding the vehicle using a trained neural network model (see at least paragraphs 27-28 regarding receiving sensor data about the environment, perceive objects within the vehicle's surrounding environment (e.g., other vehicles), predict the motion of the objects within the surrounding environment, generate trajectories based on the sensor data and/or perceived/predicted motion of the objects, and, based on the trajectory, transmit control signals to a vehicle control system to enable the autonomous vehicle to progress to a target destination. See also at least paragraphs 88-90 regarding the vehicle computing system 110 can be configured to predict a motion of the object(s) within the surrounding environment of the vehicle 105. For instance, the vehicle computing system 110 can generate prediction data 175B associated with such object(s). The prediction data 175B can be indicative of one or more predicted future locations of each respective object. … The vehicle computing system 110 can utilize one or more algorithms and/or machine-learned model(s) that are configured to predict the future motion of object(s) based at least in part on the sensor data 155, the perception data 175A, map data 160, and/or other data. This can include, for example, one or more neural networks trained to predict the motion of the object(s) within the surrounding environment of the vehicle 105 based at least in part on the past and/or current state(s) of those objects as well as the environment in which the objects are located (e.g., the lane boundary in which it is travelling, etc.), Subramanian); and
transmit control signals to the vehicle actuator system to cause the vehicle actuator system to control the vehicle to travel according to the trajectory (see at least paragraph 66 regarding the vehicle computing system can control the motion of the autonomous vehicle based, at least in part, on the selected trajectory. For instance, once a trajectory has been chosen, the vehicle computing system can transmit the motion to a vehicle controller. The vehicle controller can use the selected trajectory to generate one or more motion controls for the autonomous vehicle. … The vehicle controller can transmit those motion controls to the autonomous vehicle to be executed and follow the selected trajectory, Subramanian), and
the vehicle actuator system being configured to drive the vehicle based on the control signals transmitted by the electronic control unit (see at least paragraph 66 regarding the vehicle computing system can control the motion of the autonomous vehicle based, at least in part, on the selected trajectory. For instance, once a trajectory has been chosen, the vehicle computing system can transmit the motion to a vehicle controller. The vehicle controller can use the selected trajectory to generate one or more motion controls for the autonomous vehicle. … The vehicle controller can transmit those motion controls to the autonomous vehicle to be executed and follow the selected trajectory. See also at least paragraph 88 regarding the autonomy computing system 140 can communicate with the one or more vehicle control systems 150 to operate the vehicle 105 according to the motion plan (e.g., via the vehicle interface 145, etc.), Subramanian).
Subramanian does not explicitly teach generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm based on acceleration coefficients reflecting self-cognition and social influence; fitting a polynomial curve to the dynamically feasible trajectory; or converting the polynomial curve into reference waypoints and generating a trajectory for the vehicle based on the reference waypoints.
However, ZHAO teaches generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm (see at least paragraph 5 regarding in order to ensure that the vehicle accurately avoids obstacle vehicles, it is first necessary to predict the movement trajectories of its own vehicle and the obstacle vehicles, so as to provide a basis for collision avoidance path planning. See also at least paragraphs 124-131. See also at least paragraphs 133-137 regarding based on the three-degree-of-freedom vehicle dynamics model and combined with the CTRA kinematic model, the vehicle's driving trajectory is accurately predicted in the short-term domain. Build an LSTM neural network to accurately predict the driving trajectories of the vehicle and the obstacle vehicle in the long term, and output the predicted data of the driving trajectories of the vehicle and the obstacle vehicle. Based on the particle swarm algorithm, the weights of the driving trajectory prediction results based on the three-degree-of-freedom vehicle dynamics model and the driving trajectory prediction results based on the LSTM neural network are optimized online. The prediction results of the short-term and long-term domains are integrated according to the optimized weights to accurately predict the vehicle's driving trajectory); fitting a polynomial curve to the dynamically feasible trajectory (see at least paragraphs 124-131 regarding Step 4: Integrate grid division, artificial potential field method and high-order polynomial curve fitting method to build a vehicle dynamic safety path planning model. See also at least paragraphs 175-199 regarding according to the position of each node in the collision avoidance path, a smooth collision avoidance path is obtained by using high-order polynomial curve fitting); and converting the polynomial curve into reference waypoints (see at least paragraphs 124-131 regarding Step 4: Integrate grid division, artificial potential field method and high-order polynomial curve fitting method to build a vehicle dynamic safety path planning model. See also at least FIG. 6 and paragraphs 175-199 regarding according to the position of each node in the collision avoidance path, a smooth collision avoidance path is obtained by using high-order polynomial curve fitting. The high-order polynomial curve fitting uses a quintic polynomial curve to perform segmented fitting on the preliminary collision avoidance path and ensures smooth transition between each segment of the curve) and generating a trajectory for the vehicle based on the reference waypoints (see at least FIG. 6 and paragraphs 124-131 regarding Step 5: Based on the driving trajectory prediction data of the vehicle and the obstacle vehicle obtained in step 3, the vehicle dynamic safety path planning model is used to obtain the optimal collision avoidance path 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 use the system of ZHAO which teaches generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm; fitting a polynomial curve to the dynamically feasible trajectory; and converting the polynomial curve into reference waypoints and generating a trajectory for the vehicle based on the reference waypoints with the system of Subramanian as both systems are directed to a system and method for generating an optimal trajectory based on the an environment of a vehicle, and one of ordinary skill in the art would have recognized the established utility of generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm; fitting a polynomial curve to the dynamically feasible trajectory; and converting the polynomial curve into reference waypoints and generating a trajectory for the vehicle based on the reference waypoints and would have predictably applied it to improve the system of Subramanian.
Subramanian, as modified by ZHAO, does not explicitly teach generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm based on acceleration coefficients reflecting self-cognition and social influence.
However, such matter is taught by ZHANG (see at least paragraphs 3-4 and 56-58 regarding particle swarm optimization is a random search algorithm based on group collaboration proposed by simulating the foraging behavior of bird flocks. In the path planning of drones, each particle represents a planned path. The algorithm is initialized with a number of random particles. In each iteration, the particles tend to the optimal particle position in the population and the individual historical optimal position, realizing the path optimization process. See also at least paragraphs 73-77 regarding calculating the path cost of each particle, select the particle position with the lowest global path cost value as g_best, select the particle position with the lowest path cost value in each group as gr_i,best, and record the lowest historical path cost value position of each particle in the iteration process as p_i,best. The particle position update formula of the improved particle swarm algorithm is: Vi (t + 1) = ω Vi (t) + c1r1 (pi, best-pi) + c2r2 (gbest-pi) + c3r3 (gri, best-pi) pi (t + 1) = pi (t) + Vi (t + 1). Wherein, t represents the number of iterations; V_i(t) represents the velocity of the ith particle; p_i=p_i(t) represents the position of the ith particle; gr_i,best represents the optimal particle position of the ith particle corresponding to the group; ω represents the inertia weight; c₁, c₂, c₃ represent the acceleration factors corresponding to different parts; r₁, r₂, r₃ are all real numbers between 0 and 1. See also at least paragraphs 99-115).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of ZHANG which teaches generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm based on acceleration coefficients reflecting self-cognition and social influence with the system of Subramanian, as modified by ZHAO, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of generating a dynamically feasible trajectory based on the initial trajectory and the estimated future positions of the other vehicles using a particle swarm optimization algorithm based on acceleration coefficients reflecting self-cognition and social influence and would have predictably applied it to improve the system of Subramanian as modified by ZHAO.

Claim(s) 2, 3, 6, 9, 14, 15, 18, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Subramanian et al., US 2022/0176995 A1, hereinafter referred to as Subramanian, in view of ZHAO et al., CN 112109704 A, hereinafter referred to as ZHAO, in view of ZHANG et al., CN 115657725 A, hereinafter referred to as ZHANG, and further in view of JING et al., CN 117744366 A, hereinafter referred to as JING, respectively.
As to claim 2, Subramanian, as modified by ZHAO and ZHANG, does not explicitly teach uniformly and randomly initializing a steering angle sequence and a velocity for each of a plurality of particles, wherein the steering angle sequence and the velocity of each of the plurality of particles are uniformly randomly initialized in a range determined based on reference waypoints derived from the initial trajectory; iteratively updating the velocity, the steering angle sequence, and a position for each of the plurality of particles; or calculating a cost value for each of the plurality of particles at each iteration of updating using a cost function.
However, JING teaches uniformly and randomly initializing a steering angle sequence and a velocity for each of a plurality of particles, wherein the steering angle sequence and the velocity of each of the plurality of particles are uniformly randomly initialized in a range determined based on reference waypoints derived from the initial trajectory (see at least paragraphs 127-139 regarding assume that the algorithm selected is Particle Swarm Optimization (PSO). In PSO, each "particle" represents a parameter combination. The initial particle position may be set randomly in parameter space, while the speed and steering angle are random); iteratively updating the velocity, the steering angle sequence, and a position for each of the plurality of particles (see at least paragraphs 127-139 regarding assume that the algorithm selected is Particle Swarm Optimization (PSO). In PSO, each "particle" represents a parameter combination. The initial particle position may be set randomly in parameter space, while the speed and steering angle are random. Based on the returned scores, the algorithm updates the velocity and position (i.e. parameter combination) of each particle. In this step, the algorithm is assumed to infer that increasing speed and steering angles may result in a better score); and calculating a cost value for each of the plurality of particles at each iteration of updating using a cost function (see at least paragraphs 127-139 regarding based on the returned scores, the algorithm updates the velocity and position (i.e. parameter combination) of each particle. Repeat the above evaluation, cost calculation and score return steps. The optimization algorithm updates the parameter combination again. This process will continue to cycle until a certain termination condition is met, such as the score improvement for several consecutive rounds is not obvious, the predetermined maximum number of iterations is reached, or a parameter combination that satisfies the edge scenario is found).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of JING which teaches uniformly and randomly initializing a steering angle sequence and a velocity for each of a plurality of particles, wherein the steering angle sequence and the velocity of each of the plurality of particles are uniformly randomly initialized in a range determined based on reference waypoints derived from the initial trajectory; iteratively updating the velocity, the steering angle sequence, and a position for each of the plurality of particles; and calculating a cost value for each of the plurality of particles at each iteration of updating using a cost function with the system of Subramanian, as modified by ZHAO and ZHANG, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of uniformly and randomly initializing a steering angle sequence and a velocity for each of a plurality of particles, wherein the steering angle sequence and the velocity of each of the plurality of particles are uniformly randomly initialized in a range determined based on reference waypoints derived from the initial trajectory; iteratively updating the velocity, the steering angle sequence, and a position for each of the plurality of particles; and calculating a cost value for each of the plurality of particles at each iteration of updating using a cost function and would have predictably applied it to improve the system of Subramanian as modified by ZHAO and ZHANG.
As to claim 3, Subramanian, as modified by ZHAO, does not explicitly teach wherein the dynamically feasible trajectory is generated based on a particle among the plurality of particles having a minimum cost value.
However, such matter is taught by ZHANG (see at least paragraphs 22-24 regarding Calculate the path cost of each particle, select the particle position with the lowest global path cost value as g_best, select the particle position with the lowest path cost value in each group as gr_i,best, and record the lowest historical path cost value position of each particle in the iteration process as p_i,best. Step 43, iteratively execute step 42 until the maximum number of iterations or the maximum calculation time is met, then stop the iteration; select the particle with the lowest path cost value in the population, and the planned path it represents is the flight path of the mother platform and the daughter aircraft).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of ZHANG which teaches wherein the dynamically feasible trajectory is generated based on a particle among the plurality of particles having a minimum cost value with the system of Subramanian, as modified by ZHAO, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of having wherein the dynamically feasible trajectory is generated based on a particle among the plurality of particles having a minimum cost value and would have predictably applied it to improve the system of Subramanian as modified by ZHAO.
As to claim 6, Subramanian, as modified by ZHAO, does not explicitly teach selecting, as the dynamically feasible trajectory, a particle among the plurality of particles which has a minimum cost value.
However, such matter is taught by ZHANG (see at least paragraphs 22-24 regarding Calculate the path cost of each particle, select the particle position with the lowest global path cost value as g_best, select the particle position with the lowest path cost value in each group as gr_i,best, and record the lowest historical path cost value position of each particle in the iteration process as p_i,best. Step 43, iteratively execute step 42 until the maximum number of iterations or the maximum calculation time is met, then stop the iteration; select the particle with the lowest path cost value in the population, and the planned path it represents is the flight path of the mother platform and the daughter aircraft).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of ZHANG which teaches selecting, as the dynamically feasible trajectory, a particle among the plurality of particles which has a minimum cost value with the system of Subramanian, as modified by ZHAO, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of selecting, as the dynamically feasible trajectory, a particle among the plurality of particles which has a minimum cost value and would have predictably applied it to improve the system of Subramanian as modified by ZHAO.
Subramanian, as modified by ZHAO and ZHANG, does not explicitly teach for each of a plurality of particles, calculating a cost value using a cost function.
However, such matter is taught by JING (see at least paragraphs 127-139 regarding assume that the algorithm selected is Particle Swarm Optimization (PSO). In PSO, each "particle" represents a parameter combination. The initial particle position may be set randomly in parameter space, while the speed and steering angle are random. Based on the returned scores, the algorithm updates the velocity and position (i.e. parameter combination) of each particle. In this step, the algorithm is assumed to infer that increasing speed and steering angles may result in a better score. Repeat the above evaluation, cost calculation and score return steps. The optimization algorithm updates the parameter combination again. This process will continue to cycle until a certain termination condition is met, such as the score improvement for several consecutive rounds is not obvious, the predetermined maximum number of iterations is reached, or a parameter combination that satisfies the edge scenario is found).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of JING which teaches calculating a cost value using a cost function for each of a plurality of particles with the system of Subramanian, as modified by ZHAO and ZHANG, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of calculating a cost value using a cost function for each of a plurality of particles and would have predictably applied it to improve the system of Subramanian as modified by ZHAO and ZHANG.
As to claim 9, Subramanian, as modified by ZHAO and ZHANG, does not explicitly teach generating the dynamically feasible trajectory using the particle swarm optimization algorithm further comprises evaluating the cost value for each of the plurality of particles against cost values for each of the plurality of particles calculated at prior iterations; or iteratively updating the velocity, the steering angle sequence, and the position for each of the plurality of particles includes updating the velocity and the position based on at least one of a minimum cost value calculated at all prior iterations for each of the plurality of particles and a minimum cost value calculated at all prior iterations for all of the plurality of particles.
However, JING teaches generating the dynamically feasible trajectory using the particle swarm optimization algorithm further comprises evaluating the cost value for each of the plurality of particles against cost values for each of the plurality of particles calculated at prior iterations (see at least paragraphs 127-139 regarding assume that the algorithm selected is Particle Swarm Optimization (PSO). In PSO, each "particle" represents a parameter combination. The initial particle position may be set randomly in parameter space, while the speed and steering angle are random. Based on the returned scores, the algorithm updates the velocity and position (i.e. parameter combination) of each particle. In this step, the algorithm is assumed to infer that increasing speed and steering angles may result in a better score. Repeat the above evaluation, cost calculation and score return steps); and iteratively updating the velocity, the steering angle sequence, and the position for each of the plurality of particles includes updating the velocity and the position based on at least one of a minimum cost value calculated at all prior iterations for each of the plurality of particles and a minimum cost value calculated at all prior iterations for all of the plurality of particles (see at least paragraphs 127-139 regarding assume that the algorithm selected is Particle Swarm Optimization (PSO). In PSO, each "particle" represents a parameter combination. The initial particle position may be set randomly in parameter space, while the speed and steering angle are random. Based on the returned scores, the algorithm updates the velocity and position (i.e. parameter combination) of each particle. In this step, the algorithm is assumed to infer that increasing speed and steering angles may result in a better score. Repeat the above evaluation, cost calculation and score return steps. The optimization algorithm updates the parameter combination again. This process will continue to cycle until a certain termination condition is met, such as the score improvement for several consecutive rounds is not obvious, the predetermined maximum number of iterations is reached, or a parameter combination that satisfies the edge scenario is found).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of JING which teaches generating the dynamically feasible trajectory using the particle swarm optimization algorithm further comprises evaluating the cost value for each of the plurality of particles against cost values for each of the plurality of particles calculated at prior iterations; and iteratively updating the velocity, the steering angle sequence, and the position for each of the plurality of particles includes updating the velocity and the position based on at least one of a minimum cost value calculated at all prior iterations for each of the plurality of particles and a minimum cost value calculated at all prior iterations for all of the plurality of particles with the system of Subramanian, as modified by ZHAO and ZHANG, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of generating the dynamically feasible trajectory using the particle swarm optimization algorithm further comprises evaluating the cost value for each of the plurality of particles against cost values for each of the plurality of particles calculated at prior iterations; and iteratively updating the velocity, the steering angle sequence, and the position for each of the plurality of particles includes updating the velocity and the position based on at least one of a minimum cost value calculated at all prior iterations for each of the plurality of particles and a minimum cost value calculated at all prior iterations for all of the plurality of particles and would have predictably applied it to improve the system of Subramanian as modified by ZHAO and ZHANG.
As to claim 14, Examiner notes claim 14 recites similar limitations to claim 2 and is rejected under the same rational.
As to claim 15, Examiner notes claim 15 recites similar limitations to claim 3 and is rejected under the same rational.
As to claim 18, Examiner notes claim 18 recites similar limitations to claim 9 and is rejected under the same rational.
As to claim 19, Examiner notes claim 19 recites similar limitations to claim 6 and is rejected under the same rational.

Claim(s) 4, 5, 7, 8, 10, 11, 16, and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Subramanian et al., US 2022/0176995 A1, hereinafter referred to as Subramanian, in view of ZHAO et al., CN 112109704 A, hereinafter referred to as ZHAO, in view of ZHANG et al., CN 115657725 A, hereinafter referred to as ZHANG, in view of JING et al., CN 117744366 A, hereinafter referred to as JING, and further in view of RAMAMOORTHY et al., US 2023/0042431 A1, hereinafter referred to as RAMAMOORTHY, respectively.
As to claim 4, Subramanian, as modified by ZHAO, ZHANG, and JING, does not explicitly teach wherein the cost function is set to penalize deviations from a current trajectory, penalize deviations from a current heading, penalize violations of safety metrics, reward increases in driving comfort, and reward maintenance of a lane-center position.
However, such matter is taught by RAMAMOORTHY (see at least paragraphs 14-20 regarding the reward of each trajectory and the reward difference of each goal may be determined using a reward function that rewards reduced travel time whilst penalizing unsafe trajectories. The reward function may also penalize other reward factor(s) such as lack of comfort (e.g. one or more of longitudinal jerk, latitudinal jerk, and path curvature). Each trajectory may take the form of a sequence of states. A state may encode location information at a particular time instance but may also encode motion information such as one or more of (instantaneous) speed, heading, acceleration (magnitude and/or direction), jerk (first time derivative of acceleration) etc. More generally, a trajectory may encode spatial information (path component) but also motion information (motion component). See also at least paragraphs 59-61 and 143 regarding both leverage a velocity smoothing function 224 for smoothing computed trajectories and a reward function 226 for evaluating trajectories in relation to particular goals using a metric-based approach, in which certain factors (such as progression towards a desired goal) are rewarded and others (such as lack of safety, comfort etc.) are penalized).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of RAMAMOORTHY which teaches wherein the cost function is set to penalize deviations from a current trajectory, penalize deviations from a current heading, penalize violations of safety metrics, reward increases in driving comfort, and reward maintenance of a lane-center position with the system of Subramanian, as modified by ZHAO, ZHANG, and JING, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of having wherein the cost function is set to penalize deviations from a current trajectory, penalize deviations from a current heading, penalize violations of safety metrics, reward increases in driving comfort, and reward maintenance of a lane-center position and would have predictably applied it to improve the system of Subramanian as modified by ZHAO, ZHANG, and JING.
As to claim 5, Subramanian teaches wherein the violations of safety metrics are determined based on an estimated proximity of the autonomous vehicle and any of the other vehicles in the environment surrounding the autonomous vehicle (see at least paragraphs 88-90 regarding the vehicle computing system 110 can be configured to predict a motion of the object(s) within the surrounding environment of the vehicle 105. For instance, the vehicle computing system 110 can generate prediction data 175B associated with such object(s). The prediction data 175B can be indicative of one or more predicted future locations of each respective object. … The vehicle computing system 110 can utilize one or more algorithms and/or machine-learned model(s) that are configured to predict the future motion of object(s) based at least in part on the sensor data 155, the perception data 175A, map data 160, and/or other data. This can include, for example, one or more neural networks trained to predict the motion of the object(s) within the surrounding environment of the vehicle 105 based at least in part on the past and/or current state(s) of those objects as well as the environment in which the objects are located (e.g., the lane boundary in which it is travelling, etc.), Subramanian).
As to claim 7, Subramanian, as modified by ZHAO, ZHANG, and JING, does not explicitly teach wherein the cost function is set to penalize deviations from a current trajectory, penalize deviations from a current heading, penalize violations of safety metrics, reward increases in driving comfort, and reward maintenance of a lane-center position.
However, such matter is taught by RAMAMOORTHY (see at least paragraphs 14-20 regarding the reward of each trajectory and the reward difference of each goal may be determined using a reward function that rewards reduced travel time whilst penalizing unsafe trajectories. The reward function may also penalize other reward factor(s) such as lack of comfort (e.g. one or more of longitudinal jerk, latitudinal jerk, and path curvature). Each trajectory may take the form of a sequence of states. A state may encode location information at a particular time instance but may also encode motion information such as one or more of (instantaneous) speed, heading, acceleration (magnitude and/or direction), jerk (first time derivative of acceleration) etc. More generally, a trajectory may encode spatial information (path component) but also motion information (motion component). See also at least paragraphs 59-61 and 143 regarding both leverage a velocity smoothing function 224 for smoothing computed trajectories and a reward function 226 for evaluating trajectories in relation to particular goals using a metric-based approach, in which certain factors (such as progression towards a desired goal) are rewarded and others (such as lack of safety, comfort etc.) are penalized).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of RAMAMOORTHY which teaches wherein the cost function is set to penalize deviations from a current trajectory, penalize deviations from a current heading, penalize violations of safety metrics, reward increases in driving comfort, and reward maintenance of a lane-center position with the system of Subramanian, as modified by ZHAO, ZHANG, and JING, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of having wherein the cost function is set to penalize deviations from a current trajectory, penalize deviations from a current heading, penalize violations of safety metrics, reward increases in driving comfort, and reward maintenance of a lane-center position and would have predictably applied it to improve the system of Subramanian as modified by ZHAO, ZHANG, and JING.
As to claim 8, Subramanian teaches wherein the violations of safety metrics are determined based on an estimated proximity of the autonomous vehicle and any of the other vehicles in the environment surrounding the autonomous vehicle (see at least paragraphs 88-90 regarding the vehicle computing system 110 can be configured to predict a motion of the object(s) within the surrounding environment of the vehicle 105. For instance, the vehicle computing system 110 can generate prediction data 175B associated with such object(s). The prediction data 175B can be indicative of one or more predicted future locations of each respective object. … The vehicle computing system 110 can utilize one or more algorithms and/or machine-learned model(s) that are configured to predict the future motion of object(s) based at least in part on the sensor data 155, the perception data 175A, map data 160, and/or other data. This can include, for example, one or more neural networks trained to predict the motion of the object(s) within the surrounding environment of the vehicle 105 based at least in part on the past and/or current state(s) of those objects as well as the environment in which the objects are located (e.g., the lane boundary in which it is travelling, etc.), Subramanian).
As to claim 10, Subramanian, as modified by ZHAO, ZHANG, and JING, does not explicitly teach wherein the cost function is set to penalize deviations from a current trajectory, penalize deviations from a current heading, penalize violations of safety metrics, reward increases in driving comfort, and reward maintenance of a lane-center position.
However, such matter is taught by RAMAMOORTHY (see at least paragraphs 14-20 regarding the reward of each trajectory and the reward difference of each goal may be determined using a reward function that rewards reduced travel time whilst penalizing unsafe trajectories. The reward function may also penalize other reward factor(s) such as lack of comfort (e.g. one or more of longitudinal jerk, latitudinal jerk, and path curvature). Each trajectory may take the form of a sequence of states. A state may encode location information at a particular time instance but may also encode motion information such as one or more of (instantaneous) speed, heading, acceleration (magnitude and/or direction), jerk (first time derivative of acceleration) etc. More generally, a trajectory may encode spatial information (path component) but also motion information (motion component). See also at least paragraphs 59-61 and 143 regarding both leverage a velocity smoothing function 224 for smoothing computed trajectories and a reward function 226 for evaluating trajectories in relation to particular goals using a metric-based approach, in which certain factors (such as progression towards a desired goal) are rewarded and others (such as lack of safety, comfort etc.) are penalized).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the system of RAMAMOORTHY which teaches wherein the cost function is set to penalize deviations from a current trajectory, penalize deviations from a current heading, penalize violations of safety metrics, reward increases in driving comfort, and reward maintenance of a lane-center position with the system of Subramanian, as modified by ZHAO, ZHANG, and JING, as both systems are directed to a system and method for generating the path planning based on the an external environment of the vehicle, and one of ordinary skill in the art would have recognized the established utility of having wherein the cost function is set to penalize deviations from a current trajectory, penalize deviations from a current heading, penalize violations of safety metrics, reward increases in driving comfort, and reward maintenance of a lane-center position and would have predictably applied it to improve the system of Subramanian as modified by ZHAO, ZHANG, and JING.
As to claim 11, Subramanian teaches wherein the violations of safety metrics are determined based on an estimated proximity of the autonomous vehicle and any of the other vehicles in the environment surrounding the autonomous vehicle (see at least paragraphs 88-90 regarding the vehicle computing system 110 can be configured to predict a motion of the object(s) within the surrounding environment of the vehicle 105. For instance, the vehicle computing system 110 can generate prediction data 175B associated with such object(s). The prediction data 175B can be indicative of one or more predicted future locations of each respective object. … The vehicle computing system 110 can utilize one or more algorithms and/or machine-learned model(s) that are configured to predict the future motion of object(s) based at least in part on the sensor data 155, the perception data 175A, map data 160, and/or other data. This can include, for example, one or more neural networks trained to predict the motion of the object(s) within the surrounding environment of the vehicle 105 based at least in part on the past and/or current state(s) of those objects as well as the environment in which the objects are located (e.g., the lane boundary in which it is travelling, etc.), Subramanian).
As to claim 16, Examiner notes claim 16 recites similar limitations to claim 4 and is rejected under the same rational.
As to claim 17, Examiner notes claim 17 recites similar limitations to claim 5 and is rejected under the same rational.

Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
YAO et al. (CN 109275094 A) regarding a system for determining a target parameter based on particle swarm optimization algorithm.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KYLE S. PARK whose telephone number is [phone]. The examiner can normally be reached Mon-Thurs 9:00AM-5:00PM.
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, Anne M ANTONUCCI can be reached at [phone]. 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.

/K.S.P./Examiner, Art Unit 3666

/ANNE MARIE ANTONUCCI/Supervisory Patent Examiner, Art Unit 3666

Prosecution Timeline

• Mar 26, 2024: Application Filed
• Jul 24, 2025: Non-Final Rejection — §103
• Oct 07, 2025: Response Filed
• Oct 21, 2025: Final Rejection — §103
• Dec 23, 2025: Response after Non-Final Action
• Jan 27, 2026: Request for Continued Examination
• Feb 20, 2026: Response after Non-Final Action
• Feb 20, 2026: Non-Final Rejection — §103 (current)