DLO-Lab: Benchmarking Deformable Linear Object Manipulations with Differentiable Physics

2026-06-02

AI Summary

This work addresses the challenge of generalizing manipulation strategies for deformable linear objects (DLOs), which is hindered by material diversity and task complexity, with existing approaches often relying on handcrafted rules or task-specific designs. The authors propose the first differentiable physics simulation platform tailored for DLO manipulation, capable of modeling both elastic and plastic materials, and introduce a benchmark encompassing a variety of representative tasks. Building upon this framework, they design a specialized reinforcement learning agent that explicitly plans grasp points and employs a task decomposition mechanism to handle topological changes and grasp sensitivity. Experiments demonstrate that the proposed method efficiently learns effective policies in simulation and achieves successful sim-to-real transfer, significantly enhancing the generalization and robustness of DLO manipulation.

Abstract

We address the challenge of enabling robots to manipulate deformable linear objects (DLOs), such as ropes, cables, and rubber bands. Prior work has primarily focused on narrow, task-specific problems, often relying on real-world demonstrations or handcrafted heuristics. Such approaches, however, struggle to scale to the wide variety of materials and tasks encountered in practice, and collecting sufficiently diverse real-world data is often impractical. Additionally, existing simulation environments offer limited support for the broad spectrum of material behaviors necessary for generalizable DLO manipulation. To overcome these limitations, we introduce a differentiable simulator explicitly designed for versatile DLO manipulation. Our simulator models a wide range of material properties-including (in)extensibility, elasticity, bending plasticity, and complex interactions with other objects-providing a robust foundation for learning and evaluating manipulation skills. Building on this simulator, we propose a benchmark suite of representative tasks that highlight the unique challenges of DLO manipulation. The successful execution of these tasks is often hindered by the topological complexity and grasp sensitivity inherent to DLOs. Therefore, we introduce a specialized DLO agent that explicitly manages these challenges by proposing strategic grasping points and decomposing long-horizon tasks to maximize control authority. Finally, we evaluate various policy-learning algorithms using our framework, alongside sim-to-real transfer experiments, demonstrating our platform's potential to advance DLO manipulation.

Problem

• Deformable Linear Objects
• Robotic Manipulation
• Topological Complexity
• Grasp Sensitivity
• Material Diversity

Innovation

• differentiable physics
• deformable linear objects
• simulation benchmark
• grasp planning
• sim-to-real transfer