Synthetic scenario simulator based on events
A vehicle can capture data that can be converted into a synthetic scenario for use in a simulator. Objects can be identified in the data and attributes associated with the objects can be determined. The data can be used to generate a synthetic scenario of a simulated environment. The scenarios can include simulated objects that traverse the simulated environment and perform actions based on the attributes associated with the objects, the captured data, and/or interactions within the simulated environment. In some instances, the simulated objects can be filtered from the scenario based on attributes associated with the simulated objects and can be instantiated and/or destroyed based on triggers within the simulated environment. The scenarios can be used for testing and validating interactions and responses of a vehicle controller within the simulated environment.
1. A system comprising:
one or more processors; and
one or more computer-readable media storing computer-executable instructions that, when executed, cause the one or more processors to perform operations comprising:
receiving simulation instructions associated with log data that is generated by an autonomous vehicle traversing an environment;
generating, based at least in part on the simulation instructions, a simulated scenario comprising a simulated environment;
generating, based at least in part on the simulation instructions, an instantiation region that is associated with a simulated object in the simulated environment, the simulated object associated with an object that is represented in the log data;
determining, based at least in part on the log data, a termination attribute associated with the simulated object;
determining a position associated with a simulated autonomous vehicle traversing the simulated environment;
instantiating, based at least in part on determining that the position is associated with the instantiation region and independent of the log data, the simulated object in the simulated environment;
determining, based at least in part on the simulation instructions, an initial trajectory associated with the simulated object;
determining a waypoint associated with the initial trajectory;
determining a simulation cost associated with the waypoint;
controlling, based at least in part on a simulated object controller, the initial trajectory, the simulation cost, and the termination attribute, the simulated object in the simulated environment; and
controlling, based at least in part on simulation data that represents a response of an autonomous vehicle controller operating in the simulated environment, the simulated autonomous vehicle in the simulated environment.
2. The system of claim 1 , wherein controlling the simulated object comprises:
generating a candidate trajectory associated with the simulated object;
determining a candidate cost that indicates a difference between the waypoint and the candidate trajectory; and
modifying the candidate trajectory to minimize the candidate cost.
3. The system of claim 2 , wherein the position is a first position, the operations further comprising:
determining a second position associated with the simulated autonomous vehicle,
wherein generating the candidate trajectory is based at least in part on the second position.
4. The system of claim 2 , wherein the difference is a first difference, and wherein the waypoint is associated with a simulated object attribute comprising at least one of:
an object position;
an object velocity;
an object acceleration; or
an object arrival time, and
wherein determining the candidate cost comprises:
determining a second difference between the simulated object attribute and the candidate trajectory.
5. The system of claim 1 , wherein the position is a first position, the operations further comprising:
determining a second position associated with the simulated autonomous vehicle;
generating, based at least in part on the simulation instructions, a destruction region that is associated with the simulated object in the simulated environment; and
destroying, based at least in part on determining that the second position is associated with the destruction region, the simulated object.
6. A method comprising:
receiving log data that is generated by a vehicle traversing an environment;
generating, based at least in part on the log data, a simulated scenario comprising a simulated environment;
generating, based at least in part on the log data, an instantiation region that is associated with a simulated object in the simulated environment;
determining, based at least in part on the log data, a termination attribute associated with the simulated object,
instantiating, based at least in part the instantiation region and a position of a simulated vehicle and independent of the log data, the simulated object;
determining a waypoint associated with the simulated object;
determining a simulation cost associated with the waypoint;
controlling, based at least in part on a simulated object controller, the waypoint, the simulation cost, and the termination attribute, the simulated object in the simulated environment; and
controlling, based at least in part on simulation data that represents a response of an autonomous vehicle controller operating in the simulated environment, the simulated vehicle in the simulated environment.
7. The method of claim 6 , wherein instantiating the simulated object comprises:
determining the position of the simulated vehicle; and
determining that the position is within the instantiation region.
8. The method of claim 6 , wherein controlling the simulated object comprises:
generating a candidate trajectory associated with the simulated object;
determining a candidate cost that indicates a difference between the waypoint and the candidate trajectory; and
modifying the candidate trajectory to minimize the candidate cost.
9. The method of claim 8 , further comprising:
determining an attribute associated with the simulated vehicle; and
wherein generating the candidate trajectory is based at least in part on the attribute.
10. The method of claim 9 , wherein the attribute comprises at least one of:
a vehicle position;
a vehicle velocity;
a vehicle acceleration; or
a vehicle action.
11. The method of claim 8 , wherein the difference is a first difference, and wherein the waypoint is associated with a simulated object attribute comprising at least one of:
an object position;
an object velocity;
an object acceleration; or
an object arrival time, and
wherein determining the candidate cost comprises:
determining a second difference between the simulated object attribute and the candidate trajectory.
12. The method of claim 6 , wherein the position is a first position, the method further comprising:
determining a second position associated with the simulated vehicle;
generating, based at least in part on the log data, a destruction region that is associated with the simulated object; and
destroying, based at least in part on the destruction region and the second position, the simulated object.
13. A non-transitory computer-readable medium storing instructions executable by a processor, wherein the instructions, when executed, cause the processor to perform operations comprising:
receiving data that is based at least in part on sensor data captured by a vehicle traversing an environment;
generating, based at least in part on the data, a simulated scenario comprising a simulated environment;
determining, based at least in part on the data, a region that is associated with a simulated object in the simulated environment;
instantiating, based at least in part the region and a position of a simulated vehicle and independent of log data, the simulated object;
determining, based at least in part on the log data, a termination attribute associated with the simulated object;
determining a waypoint associated with the simulated object;
controlling, based at least in part on simulation data that represents a response of a simulated object controller operating in the simulated environment, the waypoint, and the termination attribute, the simulated object in the simulated environment; and
controlling, based at least in part on a vehicle controller, the simulated vehicle in the simulated environment.
14. The non-transitory computer-readable medium of claim 13 , wherein instantiating the simulated object comprises:
determining the position associated with the simulated vehicle; and
determining that the position is within the region.
15. The non-transitory computer-readable medium of claim 13 , wherein controlling the simulated object comprises:
generating a candidate trajectory associated with the simulated object;
determining a candidate cost that indicates a difference between the waypoint and the candidate trajectory; and
modifying the candidate trajectory to reduce the candidate cost.
16. The non-transitory computer-readable medium of claim 15 , the operations further comprising:
determining an attribute associated with the simulated vehicle; and
wherein generating the candidate trajectory is based at least in part on the attribute.
17. The non-transitory computer-readable medium of claim 16 , wherein the attribute comprises at least one of:
a vehicle position;
a vehicle velocity;
a vehicle acceleration; or
a vehicle action.
18. The non-transitory computer-readable medium of claim 15 , wherein:
the difference is a first difference;
the waypoint is associated with a simulated object attribute comprising at least one of:
an object position;
an object velocity;
an object acceleration; or
an object arrival time, and
determining the candidate cost comprises:
determining a second difference between the simulated object attribute and the candidate trajectory.
19. The non-transitory computer-readable medium of claim 13 , wherein the region is a first region, the operations further comprising:
determining a second region that is associated with the simulated object; and
destroying, based at least in part on the second region, the simulated object.
20. The non-transitory computer-readable medium of claim 19 , wherein the position is a first position, and wherein determining the second region comprises:
determining at least one of a:
a second position of the simulated vehicle;
a third position of the simulated object; or
a classification associated with the simulated object.