Deformable Linear Object
Deformable linear object (DLO) manipulation focuses on developing robotic systems capable of accurately controlling the shape and movement of flexible objects like ropes and cables. Current research heavily emphasizes model-based control approaches, often incorporating physics-based models like the Cosserat model or discrete elastic rods, enhanced by closed-loop control with visual and/or tactile feedback and data-driven methods such as neural networks and graph neural networks to improve accuracy and robustness. This field is significant for its potential impact on various industries, including automotive manufacturing, surgery, and construction, where precise manipulation of DLOs is crucial for automation and efficiency.
Papers
Deep Reinforcement Learning Based on Local GNN for Goal-conditioned Deformable Object Rearranging
Yuhong Deng, Chongkun Xia, Xueqian Wang, Lipeng Chen
Graph-Transporter: A Graph-based Learning Method for Goal-Conditioned Deformable Object Rearranging Task
Yuhong Deng, Chongkun Xia, Xueqian Wang, Lipeng Chen