Safe Reinforcement Learning on RacecarPush
26.06Violation RateTRPO
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TRPOSafety Mechanism=ADRC2026.01 | 26.06 | 5.41 | 18.28 | |
| TD3Safety Mechanism=ADRC2026.01 | 36.21 | 4.25 | 20.7 | |
| TD3Safety Mechanism=Lagrangian2026.01 | 36.76 | 5.25 | 21.64 | |
| DDPGSafety Mechanism=ADRC2026.01 | 38.08 | 5.1 | 21.99 | |
| TD3Safety Mechanism=PID2026.01 | 38.87 | 6.76 | 23.32 | |
| DDPGSafety Mechanism=PID2026.01 | 39.31 | 6.89 | 23.46 | |
| DDPGSafety Mechanism=Lagrangian2026.01 | 39.58 | 5 | 22.52 | |
| TRPOSafety Mechanism=PID2026.01 | 40.06 | 7.98 | 24.46 | |
| CPPOSafety Mechanism=ADRC2026.01 | 47.28 | 12.5 | 29.35 | |
| TRPOSafety Mechanism=Lagrangian2026.01 | 51.18 | 12.82 | 31.19 | |
| CPPOSafety Mechanism=Lagrangian2026.01 | 57.91 | 15.45 | 35.54 | |
| CPPOSafety Mechanism=PID2026.01 | 70.84 | 28.16 | 49.67 |