Controller and Observer Gain Optimization on Differentiable Simulation JAX-based (train)
0.67RMSEAdj-Adaptive
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Adj-AdaptiveGradient Strategy=Backward2026.03 | 0.67 | 0.27 | |
| DT-FixedGradient Strategy=Forward2026.03 | 0.69 | 0.57 | |
| Adj-FixedGradient Strategy=Backward2026.03 | 0.69 | 0.25 | |
| DT-Adaptive (CTG)Gradient Strategy=Forward2026.03 | 0.92 | 17.81 | |
| DT-BaseGradient Strategy=Forward2026.03 | 1.56 | 10 | |
| DT-Adaptive (History)Gradient Strategy=Forward2026.03 | 15.58 | 1.52 |