Combining Sampling- and Gradient-based Planning for Contact-rich Manipulation

TL;DR

This paper introduces a hybrid planning method combining sampling and gradient techniques to improve contact-rich manipulation in robotics, addressing stability, speed, and safety constraints.

Contribution

A novel approach that integrates the Cross-entropy Method with gradient-based planning for more robust contact-rich manipulation tasks.

Findings

Enhanced planning stability and speed in contact-rich tasks.

Ability to handle explicit safety constraints like force limits.

Benchmark results show superiority over gradient-only MPC and CEM.

Abstract

Planning over discontinuous dynamics is needed for robotics tasks like contact-rich manipulation, which presents challenges in the numerical stability and speed of planning methods when either neural network or analytical models are used. On the one hand, sampling-based planners require higher sample complexity in high-dimensional problems and cannot describe safety constraints such as force limits. On the other hand, gradient-based solvers can suffer from local optima and convergence issues when the Hessian is poorly conditioned. We propose a planning method with both sampling- and gradient-based elements, using the Cross-entropy Method to initialize a gradient-based solver, providing better search over local minima and the ability to handle explicit constraints. We show the approach allows smooth, stable contact-rich planning for an impedance-controlled robot making contact with a stiff environment, benchmarking against gradient-only MPC and CEM.