Safe Reinforcement Learning on AntCircle (test)
0Violation RateCPPOADRC
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CPPOADRCBase Algorithm=CPPO, Lagrangian Method Type=ADRC2026.01 | 0 | 0 | 3.23 | |
| TD3ADRCBase Algorithm=TD3, Lagrangian Method Type=ADRC2026.01 | 0.02 | 0 | 2.49 | |
| DDPGPIDBase Algorithm=DDPG, Lagrangian Method Type=PID2026.01 | 0.04 | 0 | 1.59 | |
| DDPGLagBase Algorithm=DDPG, Lagrangian Method Type=Lag2026.01 | 0.07 | 0 | 1.93 | |
| TRPOADRCBase Algorithm=TRPO, Lagrangian Method Type=ADRC2026.01 | 0.14 | 0 | 5.74 | |
| DDPGADRCBase Algorithm=DDPG, Lagrangian Method Type=ADRC2026.01 | 0.15 | 0 | 2.15 | |
| CPPOPIDBase Algorithm=CPPO, Lagrangian Method Type=PID2026.01 | 10.82 | 0.62 | 12.95 | |
| CPPOLagBase Algorithm=CPPO, Lagrangian Method Type=Lag2026.01 | 13.98 | 6.59 | 17.3 | |
| TRPOPIDBase Algorithm=TRPO, Lagrangian Method Type=PID2026.01 | 17.73 | 1.06 | 16.14 | |
| TRPOLagBase Algorithm=TRPO, Lagrangian Method Type=Lag2026.01 | 20.8 | 1.64 | 16.58 | |
| TD3PIDBase Algorithm=TD3, Lagrangian Method Type=PID2026.01 | 21.32 | 3.61 | 13.74 | |
| TD3LagBase Algorithm=TD3, Lagrangian Method Type=Lag2026.01 | 31.26 | 1.76 | 15.02 |