Safe Reinforcement Learning on CarGoal
29.12Violation RateTRPOADRC
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TRPOADRCBackbone=TRPO, Controller=ADRC2026.01 | 29.12 | 3.44 | 20.48 | |
| CPPOADRCBackbone=PPO, Controller=ADRC2026.01 | 34.97 | 4.72 | 21.99 | |
| TD3ADRCBackbone=TD3, Controller=ADRC2026.01 | 40.62 | 2.85 | 20.65 | |
| TRPOPIDBackbone=TRPO, Controller=PID2026.01 | 44.79 | 7.34 | 25.84 | |
| DDPGADRCBackbone=DDPG, Controller=ADRC2026.01 | 47.36 | 2.88 | 21.55 | |
| DDPGLagBackbone=DDPG, Controller=Lagrangian2026.01 | 52.43 | 7.25 | 28.35 | |
| TRPOLagBackbone=TRPO, Controller=Lagrangian2026.01 | 54.86 | 10.46 | 30.97 | |
| CPPOPIDBackbone=PPO, Controller=PID2026.01 | 62.04 | 12.95 | 34.41 | |
| CPPOLagBackbone=PPO, Controller=Lagrangian2026.01 | 65.2 | 18.54 | 40.05 | |
| DDPGPIDBackbone=DDPG, Controller=PID2026.01 | 65.52 | 12.53 | 35.03 | |
| TD3LagBackbone=TD3, Controller=Lagrangian2026.01 | 70.47 | 17 | 38.9 | |
| TD3PIDBackbone=TD3, Controller=PID2026.01 | 80.94 | 20.68 | 43.43 |