RL-Based Coverage Path Planning for Deformable Objects on 3D Surfaces

TL;DR

This paper introduces a reinforcement learning approach for planning coverage paths to manipulate deformable objects on 3D surfaces, using simulation and tactile feedback to improve surface wiping tasks.

Contribution

It presents a novel RL-based method utilizing harmonic UV mapping and scaled grouped convolutions for efficient coverage path planning on deformable objects.

Findings

Outperforms previous methods in path length and coverage

Successfully deployed on Kinova Gen3 for wiping tasks

Validates simulation results with real-world experiments

Abstract

Currently, manipulation tasks for deformable objects often focus on activities like folding clothes, handling ropes, and manipulating bags. However, research on contact-rich tasks involving deformable objects remains relatively underdeveloped. When humans use cloth or sponges to wipe surfaces, they rely on both vision and tactile feedback. Yet, current algorithms still face challenges with issues like occlusion, while research on tactile perception for manipulation is still evolving. Tasks such as covering surfaces with deformable objects demand not only perception but also precise robotic manipulation. To address this, we propose a method that leverages efficient and accessible simulators for task execution. Specifically, we train a reinforcement learning agent in a simulator to manipulate deformable objects for surface wiping tasks. We simplify the state representation of object surfaces using harmonic UV mapping, process contact feedback from the simulator on 2D feature maps, and use scaled grouped convolutions (SGCNN) to extract features efficiently. The agent then outputs actions in a reduced-dimensional action space to generate coverage paths. Experiments demonstrate that our method outperforms previous approaches in key metrics, including total path length and coverage area. We deploy these paths on a Kinova Gen3 manipulator to perform wiping experiments on the back of a torso model, validating the feasibility of our approach.