A system and method are provided and include a subject vehicle having vehicle actuation systems and vehicle sensors. A planning system includes a global route planner module, an inference module, a motion planner module, and a trajectory follower module. The inference module receives a route from th
A system and method are provided and include a subject vehicle having vehicle actuation systems and vehicle sensors. A planning system includes a global route planner module, an inference module, a motion planner module, and a trajectory follower module. The inference module receives a route from the global route planner module and dynamic obstacles data from a perception system and determines a total cost for different sets of motions associated with different trajectories for traveling along the received route. The total cost includes an inferred cost based on a probability of the associated set of motions having an increased or decreased cost based on the dynamic obstacles data. The motion planner selects a particular set of motions based on the total costs and generates a smooth trajectory for the vehicle. The trajectory follower module controls the vehicle actuation systems based on the smooth trajectory.
대표청구항▼
1. A system comprising: a vehicle having at least one vehicle actuation system and at least one vehicle sensor, the at least one vehicle actuation system including at least one of a steering system, a braking system, and a throttle system, and the at least one vehicle sensor including at least one o
1. A system comprising: a vehicle having at least one vehicle actuation system and at least one vehicle sensor, the at least one vehicle actuation system including at least one of a steering system, a braking system, and a throttle system, and the at least one vehicle sensor including at least one of a vehicle speed sensor, a vehicle acceleration sensor, an image sensor, a Lidar sensor, a radar sensor, a stereo sensor, an ultrasonic sensor, a global positioning system, and an inertial measurement unit;a perception system that generates dynamic obstacles data based on information received from the at least one vehicle sensor, the dynamic obstacles data including at least one of a current location, a size, a current estimated trajectory, a current estimated velocity, and a current estimated acceleration/deceleration of an object;a planning system having at least one processor and at least one memory storing computer-executable instructions that, when executed by the at least one processor, configure the at least one processor to implement a global route planner module, an inference module, a motion planner module, and a trajectory follower module, wherein:the global route planner module receives an inputted destination and generates a route to the inputted destination;the inference module receives the route from the global route planner module and the dynamic obstacles data from the perception system and determines a total cost for each set of motions of a plurality of sets of motions associated with different trajectories for traveling along the received route, the total cost including at least one associated cost and an inferred cost for the associated set of motions, the inferred cost being based on a probability of the associated set of motions having an increased or decreased cost based on the dynamic obstacles data;the motion planner module receives the total cost for each set of motions of the plurality of sets of motions, selects a particular set of motions from the plurality of sets of motions based on the total cost for each set of motions, and generates a smooth trajectory for the vehicle; andthe trajectory follower module controls the at least one vehicle actuation system based on the smooth trajectory. 2. The system of claim 1, wherein the inference module generates a posterior probability of the inferred cost being high based on the dynamic obstacles data. 3. The system of claim 1, wherein the dynamic obstacles data indicates that at least one preceding vehicle to the vehicle is changing lanes or has changed lanes from a current lane of travel to another lane and wherein the inference module increases the inferred cost associated with the set of motions for traveling in the current lane of travel based on the dynamic obstacles data indicating that the at least one preceding vehicle is changing lanes or has changed lanes. 4. The system of claim 1, wherein the dynamic obstacles data indicates that vehicles in a first lane of travel are traveling faster than vehicles in a second lane of travel and wherein the inference module performs at least one of increasing the inferred cost for a first set of motions associated with the first lane of travel and decreasing the inferred cost for a second set of motions associated with the second lane of travel based on the dynamic obstacles data indicating that vehicles in the first lane of travel are traveling faster than vehicles in the second lane of travel. 5. The system of claim 1, wherein the dynamic obstacles data indicates that a secondary vehicle is traveling in a first lane of a parking lot and wherein the inference module performs at least one of increasing the inferred cost for a set of motions associated with the first lane of the parking lot and decreasing the inferred cost for a set of motions associated with a second lane of the parking lot based on the dynamic obstacles data indicating that the secondary vehicle is traveling in the first lane of the parking lot. 6. The system of claim 1, wherein the at least one associated cost for each set of motions includes a cost-to-goal based on at least one of a total distance to the inputted destination and a total estimated travel time to the inputted destination. 7. The system of claim 1, wherein the at least one associated cost for each set of motions includes a sum of a plurality of motion-costs associated with the associated set of motions. 8. The system of claim 1, wherein the at least one associated cost for each set of motions includes a collision-cost based on a probability of an anticipated collision. 9. The system of claim 1, further comprising a communication system configured to communicate with at least one of another vehicle, a cloud computing device, and an infrastructure location having an associated communication system, wherein the inference module receives information from the communication system and additionally determined the inferred cost based on the information from the communication system. 10. A method comprising: receiving, with a planning system of a vehicle, an inputted destination, the vehicle having at least one vehicle actuation system and at least one vehicle sensor, the at least one vehicle actuation system including at least one of a steering system, a braking system, and a throttle system and the at least one vehicle sensor including at least one of a vehicle speed sensor, a vehicle acceleration sensor, an image sensor, a Lidar sensor, a radar sensor, a stereo sensor, an ultrasonic sensor, a global positioning system, and an inertial measurement unit, the planning system including at least one processor and at least one memory storing computer-executable instructions that, when executed by the at least one processor, configure the at least one processor to implement a global route planner module, an inference module, a motion planner module, and a trajectory follower module;generating, with the global route planner module, a route to the inputted destination;generating, with a perception system, dynamic obstacles data based on information received from the at least one vehicle sensor, the dynamic obstacles data including at least one of a current location, a size, a current estimated trajectory, a current estimated velocity, and a current estimated acceleration/deceleration of an object;receiving, with an inference module of the planning system, the route from the global route planner module and the dynamic obstacles data from the perception system;determining, with the inference module, a total cost for each set of motions of a plurality of sets of motions associated with different trajectories for traveling along the received route, the total cost including at least one associated cost and an inferred cost, the inferred cost being based on a probability of the set of motions having an increased or decreased cost based on the dynamic obstacles data;receiving, with a motion planner module of the planning system, the total cost for each set of motions;selecting, with the motion planner module, a particular set of motions from the plurality of sets of motions based on the total cost for each set of motions;generating, with the motion planner module, a smooth trajectory for the vehicle based on the particular set of motions; andcontrolling, with a trajectory follower module of the planning system, the at least one vehicle actuation system based on the smooth trajectory. 11. The method of claim 10, further comprising generating, with the inference module, a posterior probability of the inferred cost being high based on the dynamic obstacles data. 12. The method of claim 10, wherein the dynamic obstacles data indicates that at least one preceding vehicle to the vehicle is changing lanes or has changed lanes from a current lane of travel to another lane, the method further comprising increasing, with the inference module, the inferred cost associated with the set of motions for traveling in the current lane of travel based on the dynamic obstacles data indicating that the at least one preceding vehicle is changing lanes or has changed lanes. 13. The method of claim 10, wherein the dynamic obstacles data indicates that vehicles in a first lane of travel are traveling faster than vehicles in a second lane of travel, the method further comprising performing, with the inference module, at least one of increasing the inferred cost for a first set of motions associated with the first lane of travel and decreasing the inferred cost for a second set of motions associated with the second lane of travel based on the dynamic obstacles data indicating that vehicles in the first lane of travel are traveling faster than vehicles in the second lane of travel. 14. The method of claim 10, wherein the dynamic obstacles data indicates that a secondary vehicle is traveling in a first lane of a parking lot, the method further comprising performing at least one of increasing the inferred cost for a first set of motions associated with the first lane of the parking lot and decreasing the inferred cost for a second set of motions associated with a second lane of the parking lot based on the dynamic obstacles data indicating that the secondary vehicle is traveling in the first lane of the parking lot. 15. The method of claim 10, wherein the at least one associated cost for each set of motions includes a cost-to-goal based on at least one of a total distance to the inputted destination and a total estimated travel time to the inputted destination. 16. The method of claim 10, wherein the at least one associated cost for each set of motions includes a sum of a plurality of motion-costs associated with the associated set of motions. 17. The method of claim 10, wherein the at least one associated cost for each set of motions includes a collision-cost based on a probability of an anticipated collision. 18. The method of claim 10, further comprising: receiving information with a communication system configured to communicate with at least one of another vehicle, a cloud computing device, and an infrastructure location having an associated communication system; anddetermining, with the inference module, the inferred cost based on the information from the communication system. 19. A system comprising: a vehicle having a plurality of vehicle actuation systems and a plurality of vehicle sensors, the plurality of vehicle actuation systems including a steering system, a braking system, and a throttle system and the plurality of vehicle sensors including a global positioning system and inertial measurement unit and at least one of a vehicle speed sensor, a vehicle acceleration sensor, an image sensor, a Lidar sensor, a radar sensor, a stereo sensor, and an ultrasonic sensor;a map database storing map data for a geographic area in which the vehicle is traveling;a vehicle information database storing vehicle information indicating at least one of a vehicle model, a vehicle size, a vehicle wheelbase, a vehicle mass, and a vehicle turning radius of the vehicle;a motion primitive database storing a listing of motion primitives, each corresponding to a discretized smooth path that can be traversed by the vehicle over a predetermined time interval;a traffic rules database storing traffic rules associated with the geographic area in which the vehicle is traveling;a communication system configured to communicate with at least one other vehicle and receive information related to at least one of a warning of an accident, a driving hazard, an obstacle, a traffic pattern, a location of the at least one other vehicle, a traffic signal location, and a traffic signal timing;a perception system configured to generate dynamic obstacles data, static obstacles data, and road geometry data based on information received from the plurality of vehicle sensors, the dynamic obstacles data including at least one of a current location, a size, a current estimated trajectory, a current estimated velocity, and a current estimated acceleration/deceleration of an object, the static obstacles data including information about static obstacles, and the road geometry data including information about a road that the vehicle is traveling on;a planning system having at least one processor and at least one memory storing computer-executable instructions that, when executed by the at least one processor, configure the at least one processor to implement a global route planner module, an inference module, a motion planner module, and a trajectory follower module, wherein:the global route planner module receives an inputted destination and generates a route to the inputted destination based on the map data from the map database and based on traffic information from the global positioning system and inertial measurement unit;the perception system generates localization/inertial data corresponding to a current location and orientation of the vehicle based on information received from the plurality of vehicle sensors;the inference module receives the route from the global route planner module, the dynamic obstacles data from the perception system, and the information from the communication system, and determines a total cost for each set of motions of a plurality of sets of motions associated with different trajectories for traveling along the route based on at least one associated cost and an inferred cost for each set of motions, the inferred cost being based on a probability of the associated set of motions having an increased or decreased cost based on the dynamic obstacles data and based on the information from the communication system;the planning system (i) receives the total cost for each set of motions of the plurality of motions, the map data from the map database, the vehicle information from the vehicle information database, the listing of motion primitives from the motion primitive database, the traffic rules from the traffic rules database, the information from the communication system, the dynamic obstacles data from the perception system, the static obstacles data from the perception system, the road geometry data from the perception system, and localization/inertial data from the perception system, (ii) selects a particular set of motions from the plurality of sets of motions based on the total cost for each set of motions, and (iii) generates a smooth trajectory for the vehicle based on the particular set of motions; andthe trajectory follower module controls the plurality of vehicle actuation system based on the smooth trajectory.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (5)
Ferguson, David I.; Zhu, Jiajun, Actively modifying a field of view of an autonomous vehicle in view of constraints.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.