System and method for generating simulated vehicles with configured behaviors for analyzing autonomous vehicle motion planners
A system and method for generating simulated vehicles with configured behaviors for analyzing autonomous vehicle motion planners are disclosed. A particular embodiment includes: obtaining configuration instructions and data for each of a plurality of simulated vehicles, a specific driving behavior for each of the plurality of simulated vehicles corresponding to perception data obtained from perception data sensors; and generating a plurality of trajectories and acceleration profiles to transition each of the plurality of simulated vehicles from a current position and speed to a corresponding target position and target speed, the target position and the target speed corresponding to the specific driving behavior for each of the plurality of simulated vehicles.
1. A system comprising: a data processor;
a simulated vehicle configuration module, executable by the data processor, to obtain configuration instructions and data for each of a plurality of simulated vehicles, a specific driving behavior for each of the plurality of simulated vehicles corresponding to perception data obtained from perception data sensors; and
a trajectory generator to generate a plurality of trajectories and acceleration profiles to transition each of the plurality of simulated vehicles from a current position and speed to a corresponding target position and target speed, the target position and the target speed corresponding to the specific driving behavior for each of the plurality of simulated vehicles.
2. The system of claim 1 wherein the perception data sensors being of a type selected from the group consisting of: image generating devices, light amplification by stimulated emission of radiation (laser) devices, light detection and ranging (LIDAR) devices, global positioning system (GPS) devices, sound navigation and ranging (sonar) devices, radio detection and ranging (radar) devices, and distance measuring systems.
3. The system of claim 1 wherein the perception data represents real-world traffic environment information related to locations, routings, scenarios, and driver behaviors being monitored.
4. The system of claim 1 wherein the specific driving behavior for each of the plurality of simulated vehicles comprises at least one simulated vehicle with an aggressive driving behavior and at least one simulated vehicle with a conservative driving behavior, wherein aggressive driving behavior corresponds to any of: a simulated vehicle that frequently changes lanes; a simulated vehicle that exhibits steep acceleration and deceleration rates; and a simulated vehicle that travels close to other neighboring vehicles, further wherein conservative driving behavior corresponds to any of: a simulated vehicle that infrequently changes lanes; a simulated vehicle that exhibits moderate acceleration and deceleration rates; and a simulated vehicle that maintains a greater distance from other neighboring vehicles.
5. The system of claim 1 wherein a rule-based process and corresponding data structures are used to generate the target position and target speed corresponding to the specific behavior of each simulated vehicle based the specific driving behavior for each of a plurality of simulated vehicles.
6. The system of claim 1 wherein the vehicle simulation module is further configured to generate a target heading for each of the plurality of simulated vehicles.
7. The system of claim 1 wherein the trajectory generator is further configured to generate a plurality of waypoints for each of the plurality of simulated vehicles, the waypoints representing movement and behavior of each simulated vehicle in a simulation environment.
8. The system of claim 1 wherein the plurality of trajectories and acceleration profiles for each of the plurality of simulated vehicles is used for analyzing a control system of an autonomous vehicle.
9. A method comprising:
obtaining configuration instructions and data for each of a plurality of simulated vehicles, a specific driving behavior for each of the plurality of simulated vehicles corresponding to perception data obtained from perception data sensors; and
generating a plurality of trajectories and acceleration profiles to transition each of the plurality of simulated vehicles from a current position and speed to a corresponding target position and target speed, the target position and the target speed corresponding to the specific driving behavior for each of the plurality of simulated vehicles.
10. The method of claim 9 wherein the perception data sensors are installed in or on a moving test vehicle being navigated through pre-defined routings.
11. The method of claim 9 wherein the perception data comprises vehicle image data or traffic data, wherein the traffic data comprises roadway data, environmental data, or distance data.
12. The method of claim 9 wherein the specific driving behavior for each of the plurality of simulated vehicles comprises at least one simulated vehicle with a first driving behavior and at least one simulated vehicle with a second driving behavior, wherein the at least one simulated vehicle with the first driving behavior changes lanes more frequently than the at least one simulated vehicle with the second driving behavior.
13. The method of claim 9 wherein the configuration instructions and data representing at least one simulated vehicle of the plurality of simulated vehicles with a first driving behavior and at least one simulated vehicle of the plurality of simulated vehicles with a second driving behavior, wherein a probability that the at least one simulated vehicle with the first driving behavior changes lanes, passes maneuver, turns sharply, or stops suddenly, is higher than a probability that the at least one simulated vehicle with the second driving behavior changes lanes, passes maneuver, turns sharply, or stops suddenly.
14. The method of claim 9 wherein the configuration instructions and data representing at least one simulated vehicle of the plurality of simulated vehicles with a first driving behavior and at least one simulated vehicle of the plurality of simulated vehicles with a second driving behavior, wherein a rate of changing acceleration of the at least one simulated vehicle with the first driving behavior is higher than a rate of changing acceleration of the at least one simulated vehicle with the second driving behavior.
15. The method of claim 9 including generating a plurality of waypoints for each of the plurality of simulated vehicles, the waypoints representing movement and behavior of each simulated vehicle of the plurality of simulated vehicles in a simulation environment, wherein each of the waypoints further comprises the target position of the respective simulated vehicle.
16. The method of claim 9 further comprising using the plurality of trajectories and acceleration profiles for each of the plurality of simulated vehicles to create a virtual simulation environment.
17. A non-transitory machine-useable storage medium embodying instructions which, when executed by a machine, cause the machine to:
obtain configuration instructions and data for each of a plurality of simulated vehicles, a specific driving behavior for each of the plurality of simulated vehicles corresponding to perception data obtained from perception data sensors; and
generate a plurality of trajectories and acceleration profiles to transition each of the plurality of simulated vehicles from a current position and speed to a corresponding target position
and target speed, the target position and the target speed corresponding to the specific driving behavior for each of the plurality of simulated vehicles.
18. The non-transitory machine-useable storage medium of claim 17 wherein the perception data sensors being of a type selected from the group consisting of: image generating devices, light amplification by stimulated emission of radiation (laser) devices, light detection and ranging (LIDAR) devices, global positioning system (GPS) devices, sound navigation and ranging (sonar) devices, radio detection and ranging (radar) devices, and distance measuring systems.
19. The non-transitory machine-useable storage medium of claim 17 wherein the perception data comprises positions of obstacles and locations of available roadways.
20. The non-transitory machine-useable storage medium of claim 17 wherein the configuration instructions and data represent at least one simulated vehicle of the plurality of simulated vehicles with an aggressive driving behavior and at least one simulated vehicle of the plurality of simulated vehicles with a conservative driving behavior.