Contact planning for legged robots in extremely constrained environments is challenging. The main difficulty stems from the mixed nature of the problem, discrete search together with continuous trajectory optimization. To speed up the discrete search problem, we propose in this paper to learn the properties of transitions from one contact mode to the next. In particular, we learn a feasibility classifier and an offset network; the former predicts if a potential next contact state is feasible from the current contact state, while the latter learns to compensate for misalignment in achieving a desired contact state due to imperfections of the low-level control. We integrate these learned networks in a Monte Carlo Tree Search (MCTS) contact planner to better prune the tree and improve the heuristic. Our simulation results demonstrate that training these networks with offline data significantly speeds up the online search process and improves its accuracy.

翻译：在极度受限环境中为足式机器人进行接触规划具有挑战性。主要困难源于问题的混合性质，即离散搜索与连续轨迹优化相结合。为加速离散搜索过程，本文提出学习从一种接触模式到下一接触模式的转移特性。具体而言，我们学习一个可行性分类器和一个偏移网络：前者预测从当前接触状态出发，潜在的下一个接触状态是否可行；后者则学习补偿因底层控制不完善而导致的期望接触状态对准偏差。我们将这些学习网络集成到蒙特卡洛树搜索（MCTS）接触规划器中，以更好地剪枝搜索树并改进启发式函数。仿真结果表明，利用离线数据训练这些网络能显著加速在线搜索过程并提升其准确性。