Hardware in loop testing and generation of latency profiles for use in simulation
Systems, methods, tangible non-transitory computer-readable media, and devices associated with testing, simulation, or operation of an autonomous device including an autonomous vehicle are provided. For example, a service entity computing system can perform operations including obtaining operating software data associated with operating software of the autonomous vehicle. Log data associated with one or more real-world scenarios can also be obtained. One or more first simulations of the operating software can be performed based on the one or more real-world scenarios. A latency distribution profile associated with the operating software can be generated based on the one or more first simulations. One or more second simulations of the operating software can be performed based on the latency distribution profile and one or more artificially generated scenarios. Furthermore, a real-world behavior of the autonomous vehicle can be predicted based on the one or more second simulations.
1. A computer-implemented method for predicting a behavior of an autonomous vehicle, the method comprising:
obtaining, by a computing system comprising one or more computing devices, operating software data associated with operating software of the autonomous vehicle;
obtaining, by the computing system, log data associated with one or more real-world scenarios;
performing, by the computing system, one or more first simulations of the operating software based at least in part on the one or more real-world scenarios;
generating, by the computing system, a latency distribution profile associated with the operating software based at least in part on the one or more first simulations;
performing, by the computing system, one or more second simulations of the operating software based at least in part on the latency distribution profile and one or more artificially generated scenarios; and
determining, by the computing system, a predicted behavior of the autonomous vehicle based at least in part on the one or more second simulations.
2. The computer-implemented method of claim 1 , wherein the performing the one or more first simulations of the operating software comprises:
initializing, by the computing system, a vehicle computing system corresponding to the autonomous vehicle to an initial state based at least in part on the operating software; and
performing, by the computing system, the one or more first simulations using the initialized vehicle computing system.
3. The computer-implemented method of claim 2 , wherein the latency distribution profile comprises a distribution of one or more latency values associated with one or more components of the vehicle computing system when performing one or more functions in response to the one or more real-world scenarios.
4. The computer-implemented method of claim 3 , wherein the performing the one or more second simulations of the operating software based at least in part on the latency distribution profile and the one or more artificially generated scenarios comprises:
selecting, by the computing system, a latency value associated with a component from the distribution of one or more latency values associated with the component; and
performing, by the computing system, the one or more second simulations of the operating software based at least in part on the selected latency value for the component.
5. The computer-implemented method of claim 4 , further comprising:
performing, by the computing system, the one or more second simulations of the operating software for each of a plurality of selected values for the component.
6. The computer-implemented method of claim 1 , wherein the one or more second simulations of the operating software are performed in a virtual environment on emulated vehicle hardware.
7. The computer-implemented method of claim 1 , further comprising:
obtaining, by the computing system, bench test data associated with one or more latency values associated with one or more components that are not included in a vehicle computing system; and
generating, by the computing system, the latency distribution profile based at least in part on the bench test data.
8. The computer-implemented method of claim 1 , further comprising:
storing, by the computing system, prediction data associated with the predicted behavior of the autonomous vehicle in a structured data table.
9. The computer-implemented method of claim 1 , wherein the operating software data comprises a difference between the operating software and a prior version of the operating software.
10. A computing system comprising:
one or more processors; and
one or more tangible, non-transitory, computer readable media that collectively store instructions that when executed by the one or more processors cause the computing system to perform operations, the operations comprising:
obtaining operating software data associated with operating software of an autonomous vehicle;
obtaining log data associated with one or more real-world scenarios;
performing one or more first simulations of the operating software based at least in part on the one or more real-world scenarios;
generating a latency distribution profile associated with the operating software based at least in part on the one or more first simulations;
performing one or more second simulations of the operating software based at least in part on the latency distribution profile and one or more artificially generated scenarios; and
predicting a real-world behavior of the autonomous vehicle based at least in part on the one or more second simulations.
11. The computing system of claim 10 , wherein the performing the one or more first simulations of the operating software comprises:
initializing a vehicle computing system corresponding to the autonomous vehicle to an initial state based at least in part on the operating software; and
performing the one or more first simulations using the initialized vehicle computing system.
12. The computing system of claim 11 , wherein the performing the one or more second simulations of the operating software based at least in part on the latency distribution profile and the one or more artificially generated scenarios comprises:
selecting a latency value associated with a component of the vehicle computing system when performing one or more functions in response to the one or more real-world scenarios, the latency value being selected from a distribution of latency values associated with the component in the latency distribution profile; and
performing the one or more second simulations of the operating software based at least in part on the selected latency value for the component.
13. The computing system of claim 12 , wherein the operations further comprise:
performing the one or more second simulations of the operating software for each of a plurality of selected values for the component.
14. The computing system of claim 10 , wherein the operations further comprise:
obtaining bench test data associated with one or more latency values associated with one or more components that are not included in a vehicle computing system; and
generating the latency distribution profile based at least in part on the bench test data.
15. The computing system of claim 10 , wherein the operations further comprise:
storing prediction data associated with the predicted real-world behavior of the autonomous vehicle in a structured data table.
16. A tangible, non-transitory, computer readable media that stores instructions that when executed by one or more processors cause the one or more processors to perform operations, the operations comprising:
obtaining operating software data associated with operating software of an autonomous vehicle;
obtaining log data associated with one or more real-world scenarios;
performing one or more first simulations of the operating software based at least in part on the one or more real-world scenarios;
generating a latency distribution profile associated with the operating software based at least in part on the one or more first simulations;
performing one or more second simulations of the operating software based at least in part on the latency distribution profile and one or more artificially generated scenarios; and
predicting a real-world behavior of the autonomous vehicle based at least in part on the one or more second simulations.
17. The computer readable media of claim 16 , wherein the performing the one or more first simulations of the operating software comprises:
initializing a vehicle computing system corresponding to the autonomous vehicle to an initial state based at least in part on the operating software; and
performing the one or more first simulations using the initialized vehicle computing system.
18. The computer readable media of claim 17 , wherein the performing the one or more second simulations of the operating software based at least in part on the latency distribution profile and the one or more artificially generated scenarios comprises:
selecting a latency value associated with a component of the vehicle computing system when performing one or more functions in response to the one or more real-world scenarios, the latency value being selected from a distribution of latency values associated with the component in the latency distribution profile; and
performing the one or more second simulations of the operating software based at least in part on the selected latency value for the component.
19. The computer readable media of claim 18 , wherein the operations further comprise:
performing the one or more second simulations of the operating software for each of a plurality of selected values for the component.
20. The computer readable media of claim 16 , wherein the operations further comprise:
obtaining bench test data associated with one or more latency values associated with one or more components that are not included in a vehicle computing system; and
generating the latency distribution profile based at least in part on the bench test data.