Lyapunov-Inspired Sustained Balance on Dynamically Responsive Supports
Organizations: Department of Data and Systems Engineering, The University of Hong Kong, Hong Kong SAR, China.
Abstract
Conventional balance formulations often assume a rigid, stationary support, decoupling body stabilization from support motion. This assumption becomes inadequate on mechanically responsive supports, creating a coupled control challenge: corrective body motions intended to restore equilibrium can also excite support motion. We formulate sustained balance as a coupled regulation problem that coordinates equilibrium restoration with suppression of motion induced by the robot's own corrections. We introduce LISST (Lyapunov-Inspired Sustained Stabilization), a learning framework that accounts for how earlier actions shape the conditions for subsequent corrections. It uses recent observations to support immediate adjustments and longer histories to characterize persistent motion, with physical supervision relating the learned representation to robot-support behavior. To evaluate corrective actions, LISST uses coupled dynamics to assess both imbalance and the direction of motion. A Lyapunov-inspired criterion measures progress over finite windows, accommodating the transients required to restore balance. This physical evaluation guides joint-level policy learning through backpropagation through differentiable model rollouts. Privileged guidance focuses training on difficult states reached by the deployment policy and is transferred through distillation. The resulting policy operates using only onboard proprioception. Real-world experiments demonstrate that LISST maintains balance while limiting residual oscillations during continuous interaction with a dynamically responsive support. Video: youtu.be/iGHU7S4wPyw.
Figures & tables
Appendix figures & tables3 assets
Supplementary material from the paper’s appendix.
Appendix
| Parameter | Value |
| State and input scales | |
| , | , |
| , | , |
| Nominal dynamics and cost | |
| , | , |
| Module | Architecture |
|---|---|
| Shared leg encoder | MLP |
| Body branch | MLP |
| Ordered fusion | MLP |
| Body Observer | MLP |
| Temporal encoder | Conv1D |
| Each phase head | Linear |
| Setting | Value |
|---|---|
| Privileged / deployment environments | 3072 / 1024 |
| Privileged / deployment imagined starts | 3072 / 1024 |
| Imagined trajectories per update | 4096 |
| Rollout horizon | 32 steps (0.64 s) |
| Privileged-start replacement | 30% |
| Privileged starts: low-value / pre-failure | 15% / 15% |
Robotic Control
Robotic control studies how machines use sensing and computation to produce reliable movement and interaction in changing environments.
See all 626 papers