THREAD: Trajectory Planning for Hybrid Rigid-Soft Manipulators with Environment-Aware Diffusion
This work addresses the challenging problem of planning trajectories for hybrid rigid-soft manipulators in confined spaces, which is important for applications like minimally invasive surgery or inspection, and the results show a substantial improvement over existing methods.
THREAD is the first diffusion-based trajectory planner for hybrid rigid-soft manipulators, achieving 92.4% task success with 5x fewer collisions than the strongest baseline in confined environments, and demonstrating real-world transfer through apertures as small as 1.3x the soft segment diameter.
Manipulation in confined environments, such as threading a manipulator through narrow apertures, remains a fundamental challenge, especially for conventional rigid robots. Hybrid rigid-soft manipulators offer promise but face two compounding planning challenges: backbone shapes feasible in free space become infeasible under environmental contact, and planning rigid and soft segments independently ignores their kinematic coupling. We present THREAD, the first diffusion-based trajectory planner for hybrid manipulation, learning a generative prior over physically realizable backbone trajectories conditioned on local environment geometry, with physics-inspired losses encoding curvature, smoothness, and collision constraints jointly across both segments. Trained in simulation, THREAD achieves 92.4% task success with 5x fewer collisions than the strongest baseline. We show cross-embodiment real-world transfer with minimal online updates, successfully threading through apertures as small as 1.3x the soft segment diameter.