31 January 2024 in IEEE Robotics and Automation Letters
Realistic physics engines play a crucial role for learning to manipulate deformable objects such as garments in simulation. By doing so, researchers can circumvent challenges such as sensing the deformation of the object in the realworld. In spite of the extensive use of simulations for this task, few works have evaluated the reality gap between deformable object simulators and real-world data. We present a benchmark dataset to evaluate the sim-to-real gap in cloth manipulation. The dataset is collected by performing a dynamic as well as a quasi-static cloth manipulation task involving contact with a rigid table. We use the dataset to evaluate the reality gap, computational time, and simulation stability of four popular deformable object simulators: MuJoCo, Bullet, Flex, and SOFA. Additionally, we discuss the benefits and drawbacks of each simulator. The benchmark dataset is open-source. Supplementary material, videos, and code, can be found at https://sites.google.com/view/cloth-sim2real-benchmark .
10 March 2022 in IEEE Robotics and Automation Letters
Manipulation of granular materials such as sand or rice remains an unsolved problem due to challenges such as the difficulty of defining their configuration or modeling the materials and their particles interactions. Current approaches tend to simplify the material dynamics and omit the interactions between the particles. In this letter, we propose to use a graph-based representation to model the interaction dynamics of the material and rigid bodies manipulating it. This allows the planning of manipulation trajectories to reach a desired configuration of the material. We use a graph neural network (GNN) to model the particle interactions via message-passing. To plan manipulation trajectories, we propose to minimise the Wasserstein distance between a predicted distribution of granular particles and their desired configuration. We demonstrate that the proposed method is able to pour granular materials into the desired configuration both in simulated and real scenarios.
31 December 2023 in Advanced Intelligent Systems
For fiber technology, new approaches such as biomimetic materials, such as silks, are being intensively explored, providing new solutions for a variety of industries, including textiles, composites, and biomedical engineering. New approaches for spinning or these materials are needed. Despite recent advances in enhancing fiber tensile properties, achieving significant improvement in tensile properties remains a tedious and challenging task, suffering from little to no controlled extrusion process and difficult optimization in high‐dimensional parameter spaces. Herein, a novel robotic biomimetic pulling method that can rapidly enhance fiber tensile properties surpassing current methods in both speed and resulting fiber properties is shown. Using a controlled fiber‐pulling device with in situ tensile measurements and adaptive optimization based on Bayesian Optimization, fiber strength exceeding 300% of the traditional full factorial design method within just a few experimental iterations is reached. The rapid experimental method presents a potential avenue for enhancing the performance of artificial fibers across diverse industries and applications.