Safe Reinforcement Learning on RacecarCircle (test)
0.2364Violation RateTRPOADRC
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| TRPOADRCBase Algorithm=TRPO, Lagrangian Controller=ADRC2026.01 | 0.2364 | 3.54 | 17.3 | |
| TRPOPIDBase Algorithm=TRPO, Lagrangian Controller=PID2026.01 | 0.3799 | 7.92 | 23.45 | |
| CPPOADRCBase Algorithm=CPPO, Lagrangian Controller=ADRC2026.01 | 0.3994 | 21.78 | 35.72 | |
| TD3ADRCBase Algorithm=TD3, Lagrangian Controller=ADRC2026.01 | 0.4208 | 4.56 | 24.21 | |
| DDPGPIDBase Algorithm=DDPG, Lagrangian Controller=PID2026.01 | 0.4617 | 7.94 | 27.09 | |
| TD3PIDBase Algorithm=TD3, Lagrangian Controller=PID2026.01 | 0.4872 | 8.57 | 27.76 | |
| TD3LagBase Algorithm=TD3, Lagrangian Controller=Lag2026.01 | 0.4942 | 8.22 | 24.95 | |
| DDPGLagBase Algorithm=DDPG, Lagrangian Controller=Lag2026.01 | 0.5052 | 7.87 | 25.03 | |
| DDPGADRCBase Algorithm=DDPG, Lagrangian Controller=ADRC2026.01 | 0.5092 | 5 | 24.51 | |
| TRPOLagBase Algorithm=TRPO, Lagrangian Controller=Lag2026.01 | 0.5139 | 16.1 | 34.48 | |
| CPPOPIDBase Algorithm=CPPO, Lagrangian Controller=PID2026.01 | 0.5682 | 21.77 | 40.95 | |
| CPPOLagBase Algorithm=CPPO, Lagrangian Controller=Lag2026.01 | 0.5832 | 30.72 | 50.41 |