Simultaneous Contact Sequence and Patch Planning for Dynamic Locomotion

2025-08-18

AI Summary

Dynamic motion planning for legged robots in highly constrained environments requires simultaneous optimization of discrete contact sequences and continuous contact patch selection—a challenging combinatorial-continuous co-design problem. Method: We propose a unified planning framework that synergistically integrates Monte Carlo Tree Search (MCTS) with whole-body dynamic trajectory optimization. MCTS guides discrete decision-making over multi-contact sequences, while a full-body dynamics-based optimizer refines continuous contact locations and joint trajectories. Contribution/Results: This is the first approach to employ MCTS for end-to-end synthesis of non-periodic, multimodal contact behaviors—including stepping, single-leg stance, and hand-foot coordination—without pre-defined templates. In simulation, it efficiently generates dynamically feasible, terrain-adaptive locomotion plans; these are successfully transferred to a real quadrupedal robot. Moreover, the framework generalizes to complex aperiodic humanoid motions such as crawling and攀附 (climbing-assisted locomotion), demonstrating robustness across morphologies and tasks.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Mixed Discrete/Continuous Planning
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Legged robots have the potential to traverse highly constrained environments with agile maneuvers. However, planning such motions requires solving a highly challenging optimization problem with a mixture of continuous and discrete decision variables. In this paper, we present a full pipeline based on Monte-Carlo tree search (MCTS) and whole-body trajectory optimization (TO) to perform simultaneous contact sequence and patch selection on highly challenging environments. Through extensive simulation experiments, we show that our framework can quickly find a diverse set of dynamically consistent plans. We experimentally show that these plans are transferable to a real quadruped robot. We further show that the same framework can find highly complex acyclic humanoid maneuvers. To the best of our knowledge, this is the first demonstration of simultaneous contact sequence and patch selection for acyclic multi-contact locomotion using the whole-body dynamics of a quadruped.

Problem

• Planning dynamic locomotion for legged robots in constrained environments
• Solving mixed continuous-discrete optimization for contact-patch selection
• Ensuring dynamically consistent plans transferable to real quadrupeds

Innovation

• Monte-Carlo tree search for contact planning
• Whole-body trajectory optimization integration
• Simultaneous sequence and patch selection