Automated design of soft-rigid hybrid robots for dynamic locomotion

2026-05-28

AI Summary

This work addresses the limitations of purely rigid robots in environmental adaptability and fully soft robots in load-bearing capacity and scalability by proposing an automated co-design framework for hybrid soft-rigid robots. For the first time in this domain, differentiable simulation coupled with gradient-based optimization is employed to jointly optimize free-form soft structures, rigid truss layouts, and multi-channel actuation configurations. The developed differentiable simulator integrates the Material Point Method (MPM) with Extended Position-Based Dynamics (XPBD), enabling end-to-end gradient optimization of coupled soft-rigid systems and automatically generating truss skeletons that efficiently transmit actuation forces and elicit effective gaits. Physical prototypes successfully reproduce the optimized locomotion patterns, and modal analysis confirms alignment between structural deformation modes and actuation frequencies, significantly enhancing locomotion performance.

Abstract

Rigid-bodied robots often lack compliance needed to adapt to unstructured environments, while fully soft robots, though highly adaptable, struggle with scalability and load capacity. In nature, musculoskeletal systems balance strength and flexibility by integrating hard and soft tissues. Inspired by this principle, we present an automated design method for soft-rigid hybrids that optimizes a freeform soft-body shape, a stiff truss layout, and multi-channel actuation. Our differentiable simulator couples the material point method (MPM) for deformable bodies with extended position-based dynamics (XPBD) for truss elements, enabling gradient-based search. The optimization generates truss skeletons that transmit actuation forces to the soft body. We fabricate the optimized design and evaluate it on a walking task. Experiments reproduce the walking mode predicted by the optimization, which does not emerge without the skeleton. Modal analysis further suggests that the skeleton enables deformation modes near the actuation frequency that promote effective stride generation.

Problem

• soft-rigid hybrid robots
• dynamic locomotion
• compliance
• scalability
• load capacity

Innovation

• soft-rigid hybrid robots
• differentiable simulation
• material point method (MPM)
• truss optimization
• gradient-based design