Safe Reinforcement Learning on CarCircle (test)
17.71Violation RateTRPOADRC
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TRPOADRCBase Algorithm=TRPO, Controller=ADRC2026.01 | 17.71 | 1.92 | 16.22 | |
| CPPOADRCBase Algorithm=PPO, Controller=ADRC2026.01 | 21.35 | 7.74 | 18.85 | |
| TD3PIDBase Algorithm=TD3, Controller=PID2026.01 | 39.2 | 12.52 | 30.87 | |
| DDPGPIDBase Algorithm=DDPG, Controller=PID2026.01 | 39.77 | 13.56 | 30.89 | |
| TRPOLagBase Algorithm=TRPO, Controller=Lagrangian2026.01 | 40.04 | 6.45 | 25.52 | |
| TRPOPIDBase Algorithm=TRPO, Controller=PID2026.01 | 40.67 | 6.58 | 24.24 | |
| CPPOLagBase Algorithm=PPO, Controller=Lagrangian2026.01 | 46.73 | 15.85 | 33.54 | |
| TD3ADRCBase Algorithm=TD3, Controller=ADRC2026.01 | 48.41 | 7.49 | 24.26 | |
| DDPGADRCBase Algorithm=DDPG, Controller=ADRC2026.01 | 51.5 | 7.73 | 23.82 | |
| DDPGLagBase Algorithm=DDPG, Controller=Lagrangian2026.01 | 51.85 | 13.55 | 33.49 | |
| CPPOPIDBase Algorithm=PPO, Controller=PID2026.01 | 52.77 | 12.29 | 31.24 | |
| TD3LagBase Algorithm=TD3, Controller=Lagrangian2026.01 | 53 | 17.18 | 38.06 |