Time-warping for autonomous driving simulation
View Patent ↗Autonomous driving simulation using recorded driving data is disclosed. A method of simulating autonomous driving includes receiving recorded driving data from a recorded autonomous vehicle (AV), the recorded driving data includes decision-making data generated using a first decision-making algorithm, sensing data, and movement data including positions of the recorded AV; obtaining simulation data from the recorded driving data by excluding the decision-making data from the recorded driving data; and simulating, by a simulation AV, a second decision-making algorithm using the simulation data. The simulating the second decision-making algorithm includes determining a first position of the simulation AV at a first time and adjusting a playback speed of the simulation data based on a difference between the first position and a second position of the positions of the recorded AV at the first time.
1. A method of simulating autonomous driving, comprising:
receiving recorded driving data from a recorded autonomous vehicle (AV), the recorded driving data including:
decision-making data generated using a first decision-making algorithm;
sensing data; and
movement data including positions of the recorded AV;
obtaining simulation data from the recorded driving data by:
excluding the decision-making data from the recorded driving data; and
simulating, by a simulation AV, a second decision-making algorithm using the simulation data, wherein the simulating the second decision-making algorithm comprises:
determining a first position of the simulation AV at a first time; and
adjusting a playback speed of the simulation data based on a difference between the first position and a second position of the positions of the recorded AV at the first time.
2. The method of claim 1 , wherein the difference between the first position and a second of the position is measured in a longitudinal direction.
3. The method of claim 1 ,
wherein the recorded driving data further includes operator data, and
wherein obtaining the simulation data further comprises:
excluding the operator data from the recorded driving data.
4. The method of claim 1 , wherein adjusting the playback speed of the simulation data comprises:
in response to the first position being outside a sensing range of the recorded AV, adjusting the playback speed of the recorded driving data.
5. The method of claim 1 , wherein adjusting the playback speed of the simulation data comprises:
in response to a distance between the first position and the second position exceeding a threshold distance, adjusting the playback speed of the simulation data.
6. The method of claim 1 , wherein adjusting the playback speed of the simulation data comprises:
adjusting, by a Proportional-Integral-Derivative controller (a PID controller), the playback speed of the simulation data, wherein the PID controller uses a distance between the first position and the second position as an error value.
7. The method of claim 6 , wherein the PID controller adjusts the playback speed of the simulation data in a case that the distance exceeds a threshold.
8. The method of claim 6 , wherein the distance is measured between respective co-located eye-points of the recorded AV and the simulation AV.
9. The method of claim 8 , wherein the respective co-located eye-points correspond to respective centers of rear axles of the recorded AV and the simulation AV.
10. The method of claim 1 , wherein the sensing data includes representations of other road users obtained using line-of-sight sensors of the AV.
11. The method of claim 1 ,
wherein the sensing data includes a representation of another road user obtained using sensors of the recorded AV, and
wherein simulating, by the simulation AV, the second decision-making algorithm using the recorded driving data comprises:
receiving, by the second decision-making algorithm, the representation of the other road user.
12. An apparatus for simulating autonomous driving, comprising:
a memory; and
a processor configured to execute instructions of a simulation autonomous vehicle (AV) stored in the memory to:
receive recorded driving data from a recorded AV, the recorded driving data including:
decision-making data generated using a first decision-making algorithm;
sensing data; and
movement data including positions of the recorded AV;
obtain simulation data from the recorded driving data by:
excluding the decision-making data from the recorded driving data; and
simulate a second decision-making algorithm using the simulation data, wherein the simulating the second decision-making algorithm comprises to:
determine a first position of the simulation AV at a first time; and
adjust a playback speed of the simulation data based on a difference between the first position and a second position of the positions of the recorded AV at the first time.
13. The apparatus of claim 12 , wherein the difference between the first position and a second of the position is measured in a longitudinal direction.
14. The apparatus of claim 12 ,
wherein the recorded driving data further includes operator data, and
wherein obtaining the simulation data further comprises:
excluding the operator data from the recorded driving data.
15. The apparatus of claim 12 , wherein to adjust the playback speed of the simulation data comprises to:
in response to the first position being outside a sensing range of the recorded AV, adjust the playback speed of the recorded driving data such that the first position and the second position are within the sensing range of the recorded AV.
16. The apparatus of claim 12 , wherein to adjust the playback speed of the simulation data comprises to:
in response to a distance between the first position and the second position exceeding a threshold distance, adjust the playback speed of the recorded driving data such that the first position and the second position are within a sensing range of the recorded AV.
17. The apparatus of claim 12 , wherein to adjust the playback speed of the simulation data comprises to:
adjust, by a proportional-integral-derivative (PID) controller, the playback speed of the simulation data, wherein the PID controller uses a distance between the first position and the second position as an error value.
18. The apparatus of claim 17 , wherein the distance is measured between respective co-located eye-points of the recorded AV and the simulation AV corresponding to respective centers of rear axles of the recorded AV and the simulation AV.
19. The apparatus of claim 12 ,
wherein the sensing data includes a representation of another road user obtained using sensors of the recorded AV, and
wherein to simulate the second decision-making algorithm using the simulation data comprises to:
receive, by the second decision-making algorithm, the representation of the other road user.
20. A method of simulating autonomous driving, comprising:
receiving recorded driving data from a recorded autonomous vehicle (AV), the recorded AV using a first decision-making algorithm; and
simulating, by a simulation AV, a second decision-making algorithm using at least some of the recorded driving data, wherein the simulating the second decision-making algorithm comprises:
adjusting a playback speed of the at least some of the recorded driving data based on a difference between a first position of the simulation AV at a first time and a second of the position of the recorded AV at the first time.