TrajectoTree: Trajectory Optimization Meets Tree Search for Planning Multi-contact Dexterous Manipulation

TL;DR

This paper introduces TrajectoTree, a novel planning method combining high-level contact sequence planning with contact-implicit trajectory optimization to improve efficiency and accuracy in multi-contact dexterous manipulation tasks.

Contribution

The paper presents a new integrated approach that significantly speeds up trajectory planning and enhances tracking accuracy in dexterous manipulation involving contact switching.

Findings

Trajectory planning is approximately 7 times faster than baseline methods.

Planned trajectories enable closer tracking of object poses in simulations.

The method effectively handles contact switching in dexterous manipulation.

Abstract

Dexterous manipulation tasks often require contact switching, where fingers make and break contact with the object. We propose a method that plans trajectories for dexterous manipulation tasks involving contact switching using contact-implicit trajectory optimization (CITO) augmented with a high-level discrete contact sequence planner. We first use the high-level planner to find a sequence of finger contact switches given a desired object trajectory. With this contact sequence plan, we impose additional constraints in the CITO problem. We show that our method finds trajectories approximately 7 times faster than a general CITO baseline for a four-finger planar manipulation scenario. Furthermore, when executing the planned trajectories in a full dynamics simulator, we are able to more closely track the object pose trajectories planned by our method than those planned by the baselines.