Safe Reinforcement Learning on RacecarButton (test)
72.49Violation RateTD3Lag
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TD3LagBase Algorithm=TD3, Lagrangian Controller=Lag2026.01 | 72.49 | 16.31 | 39.65 | |
| DDPGLagBase Algorithm=DDPG, Lagrangian Controller=Lag2026.01 | 75.88 | 15.77 | 39.47 | |
| TRPOADRCBase Algorithm=TRPO, Lagrangian Controller=ADRC2026.01 | 80.06 | 21.38 | 45 | |
| CPPOADRCBase Algorithm=CPPO, Lagrangian Controller=ADRC2026.01 | 81.18 | 35.76 | 59.11 | |
| TD3ADRCBase Algorithm=TD3, Lagrangian Controller=ADRC2026.01 | 84.48 | 18.37 | 42.65 | |
| DDPGADRCBase Algorithm=DDPG, Lagrangian Controller=ADRC2026.01 | 85.18 | 18.49 | 42.87 | |
| TD3PIDBase Algorithm=TD3, Lagrangian Controller=PID2026.01 | 85.99 | 21.46 | 45.75 | |
| TRPOPIDBase Algorithm=TRPO, Lagrangian Controller=PID2026.01 | 87.04 | 32.94 | 56.86 | |
| TRPOLagBase Algorithm=TRPO, Lagrangian Controller=Lag2026.01 | 87.31 | 33.07 | 57.34 | |
| DDPGPIDBase Algorithm=DDPG, Lagrangian Controller=PID2026.01 | 88.26 | 20.12 | 44.62 | |
| CPPOPIDBase Algorithm=CPPO, Lagrangian Controller=PID2026.01 | 97.37 | 78.44 | 103.21 | |
| CPPOLagBase Algorithm=CPPO, Lagrangian Controller=Lag2026.01 | 97.38 | 65.75 | 90.53 |