Safe Reinforcement Learning on Vehicle Avoidance Moving Obstacles benchmark leaderboard · SOTA2 Research

Safe Reinforcement Learning on Vehicle Avoidance Moving Obstacles

100Verified Success Rate (50th Percentile)VSRL

Evaluation Results

Method	Links
VSRL2024.05		100	50	100	100	4,014
VSRL2024.05		100	50	100	100	3,644
MBPPO2024.05		82.6	6	94.2	94.2	37,510
RESPO2024.05		74.5	9	89.6	89.6	39,120
CAP2024.05		73.3	13	89.5	89.5	39,335
CBF-RL2024.05		73	6	89.3	89.3	30,115
PPO-Lag2024.05		72.8	6	87.8	87.8	30,312
PPO-PID2024.05		72	6	89.4	89.4	28,722
CAP2024.05		57.1	8	99.2	99.2	3,623
PPO-Lag2024.05		0	3	99.7	99.7	3,717
PPO-PID2024.05		0	2	99.7	99.7	3,715
MBPPO2024.05		0	4	99.3	99.3	3,746
CBF-RL2024.05		0	4	99.3	99.3	3,696
RESPO2024.05		0	6	99.1	99.1	3,736