Scaling Whole-body Multi-contact Manipulation with Contact Optimization

2025-08-18

AI Summary

Traditional discrete-sampling planning methods suffer from poor scalability in whole-body multi-contact manipulation for humanoid robots due to the continuity of contact points. To address this, we propose an autonomous planning framework based on continuous contact surface modeling. Our approach introduces three key contributions: (1) a robot–object geometric representation enabling closed-form computation of proximity points; (2) an operation-oriented cost function designed for whole-body coordination; and (3) gradient-based optimization for efficient, differentiable contact policy synthesis. Experiments demonstrate a 77% reduction in planning time, successful resolution of previously infeasible complex manipulation tasks, and real-world validation on a physical humanoid robot performing whole-body box manipulation. The framework significantly enhances both the autonomy and applicability of multi-contact manipulation.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Daily tasks require us to use our whole body to manipulate objects, for instance when our hands are unavailable. We consider the issue of providing humanoid robots with the ability to autonomously perform similar whole-body manipulation tasks. In this context, the infinite possibilities for where and how contact can occur on the robot and object surfaces hinder the scalability of existing planning methods, which predominantly rely on discrete sampling. Given the continuous nature of contact surfaces, gradient-based optimization offers a more suitable approach for finding solutions. However, a key remaining challenge is the lack of an efficient representation of robot surfaces. In this work, we propose (i) a representation of robot and object surfaces that enables closed-form computation of proximity points, and (ii) a cost design that effectively guides whole-body manipulation planning. Our experiments demonstrate that the proposed framework can solve problems unaddressed by existing methods, and achieves a 77% improvement in planning time over the state of the art. We also validate the suitability of our approach on real hardware through the whole-body manipulation of boxes by a humanoid robot.

Problem

• Enabling humanoid robots to autonomously perform whole-body manipulation tasks
• Addressing scalability issues in contact planning for continuous surfaces
• Developing efficient robot surface representation for gradient-based optimization

Innovation

• Closed-form proximity points computation for surfaces
• Gradient-based optimization for continuous contact planning
• Cost design guiding whole-body manipulation efficiently