Quadruped Locomotion on Barkour MJX simulation (test)
2.1CoTSTL-based reward
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| STL-based rewardCommanded forward velocity (vx)=0.32026.07 | 2.1 | 100 | 100 | |
| Heuristic-defaultCommanded forward velocity (vx)=0.32026.07 | 2.1 | 100 | 0 | |
| STL-based rewardCommanded forward velocity (vx)=0.52026.07 | 1.5 | 100 | 100 | |
| Heuristic-defaultCommanded forward velocity (vx)=1.02026.07 | 1.5 | 100 | 80 | |
| Heuristic-bestCommanded forward velocity (vx)=1.62026.07 | 1.4 | 100 | 100 | |
| Heuristic-bestCommanded forward velocity (vx)=1.92026.07 | 1.4 | 100 | 100 | |
| Heuristic-bestCommanded forward velocity (vx)=2.02026.07 | 1.4 | 100 | 10 | |
| Heuristic-bestCommanded forward velocity (vx)=2.12026.07 | 1.4 | 100 | 0 | |
| Heuristic-defaultCommanded forward velocity (vx)=0.52026.07 | 1.3 | 100 | 0 | |
| Heuristic-defaultCommanded forward velocity (vx)=1.32026.07 | 1.3 | 100 | 30 | |
| Heuristic-defaultCommanded forward velocity (vx)=2.02026.07 | 1.3 | 50 | 0 | |
| Heuristic-defaultCommanded forward velocity (vx)=2.12026.07 | 1.3 | 30 | 0 | |
| Heuristic-bestCommanded forward velocity (vx)=1.32026.07 | 1.3 | 100 | 100 | |
| STL-based rewardCommanded forward velocity (vx)=0.72026.07 | 1.2 | 100 | 100 | |
| STL-based rewardCommanded forward velocity (vx)=1.02026.07 | 1.2 | 100 | 100 | |
| Heuristic-defaultCommanded forward velocity (vx)=1.62026.07 | 1.2 | 100 | 30 | |
| Heuristic-defaultCommanded forward velocity (vx)=1.92026.07 | 1.2 | 50 | 0 | |
| Heuristic-bestCommanded forward velocity (vx)=0.32026.07 | 1.2 | 100 | 100 | |
| Heuristic-bestCommanded forward velocity (vx)=1.02026.07 | 1.2 | 100 | 100 | |
| STL-based rewardCommanded forward velocity (vx)=1.32026.07 | 1.1 | 100 | 100 | |
| STL-based rewardCommanded forward velocity (vx)=1.92026.07 | 1.1 | 100 | 100 | |
| STL-based rewardCommanded forward velocity (vx)=2.02026.07 | 1.1 | 100 | 100 | |
| STL-based rewardCommanded forward velocity (vx)=2.12026.07 | 1.1 | 100 | 100 | |
| Heuristic-defaultCommanded forward velocity (vx)=0.72026.07 | 1.1 | 100 | 10 | |
| STL-based rewardCommanded forward velocity (vx)=1.62026.07 | 1 | 100 | 100 | |
| Heuristic-bestCommanded forward velocity (vx)=0.52026.07 | 1 | 100 | 100 | |
| Heuristic-bestCommanded forward velocity (vx)=0.72026.07 | 1 | 100 | 100 |