Non-Gaited Legged Locomotion With Monte-Carlo Tree Search and Supervised Learning
🤖 AI Summary
Legged robots struggle to generate efficient and stable locomotion in real time on complex, uncertain terrains when no prior gait templates are available.
Method: This paper proposes a gait-agnostic real-time motion planning framework. It integrates Monte Carlo Tree Search (MCTS) with a value function trained via supervised learning to guide online optimization of foot contact sequences and timing; further, it unifies an optimization-based controller with real-time contact dynamics modeling for end-to-end closed-loop control.
Contribution/Results: The core innovation lies in eliminating reliance on predefined gait templates, enabling terrain-adaptive gait synthesis. Evaluated on a 22 kg quadrupedal robot, the method achieves significantly higher locomotion efficiency and stability than fixed-gait baselines under irregular terrain and external disturbances, while satisfying real-time computational constraints (<10 ms per control cycle).
Application Category
📝 Abstract
Legged robots are able to navigate complex terrains by continuously interacting with the environment through careful selection of contact sequences and timings. However, the combinatorial nature behind contact planning hinders the applicability of such optimization problems on hardware. In this work, we present a novel approach that optimizes gait sequences and respective timings for legged robots in the context of optimization-based controllers through the use of sampling-based methods and supervised learning techniques. We propose to bootstrap the search by learning an optimal value function in order to speed-up the gait planning procedure making it applicable in real-time. To validate our proposed method, we showcase its performance both in simulation and on hardware using a 22 kg electric quadruped robot. The method is assessed on different terrains, under external perturbations, and in comparison to a standard control approach where the gait sequence is fixed a priori.
Problem
Research questions and friction points this paper is trying to address.
Legged Robot
Real-time Gait Optimization
Complex Terrain
Innovation
Methods, ideas, or system contributions that make the work stand out.
Monte Carlo Tree Search
Learning Algorithm
Legged Robot Adaptability
🔎 Similar Papers
No similar papers found.
