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

2026-03-03

AI Summary

This work addresses the challenge of contact-intensive task planning for manipulating deformable objects—such as cloths—over complex 3D surfaces, where occlusions and limited tactile perception hinder effective coverage. To overcome this, the authors propose a reinforcement learning–based autonomous path planning method that leverages harmonic UV mapping to project intricate 3D geometries into a simplified 2D state representation. Tactile feedback is encoded via 2D feature maps, and a novel scaled grouped convolutional neural network (SGCNN) is introduced to efficiently extract spatial features and generate coverage paths within a low-dimensional action space. Experimental results demonstrate that the proposed approach outperforms existing methods in key metrics including total path length and covered area. Its practical efficacy is further validated through successful execution of a back-wiping task on a human torso model using a Kinova Gen3 robotic arm.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

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.

Problem

• deformable objects
• coverage path planning
• 3D surfaces
• tactile perception
• contact-rich manipulation

Innovation

• Reinforcement Learning
• Deformable Object Manipulation
• Harmonic UV Mapping
• Scaled Grouped Convolution
• Coverage Path Planning