We present CORD-SLS, a real-time control method for safe deformable object manipulation, focusing on ropes and cloth. At its core is a GPU-parallel differentiable simulator with contact smoothing, enabling efficient gradient-based planning through intermittent contact. To robustly satisfy constraints under model and sensing uncertainty, we develop a real-time, GPU-parallel output-feedback robust model predictive control (MPC) algorithm that plans with this simulator. We further show that the simulator accelerates model-based RL for training neural manipulation policies. To improve real-world robustness, we use conformal prediction to calibrate visual-feedback and perception-error bounds for MPC, producing reachable tubes that enable high-probability safe control. We evaluate CORD-SLS on high-dimensional, contact-rich rope and cloth manipulation tasks in simulation and hardware, including obstacle avoidance, routing, folding, and smoothing. Across settings, CORD-SLS achieves millisecond-speed planning, exceeding baselines in safety, speed, and task success.
Blogger's Review: The innovation of CORD-SLS lies in its integration of GPU-accelerated differentiable simulation with robust MPC control strategies. This approach not only enhances the safety and efficiency of operations but also provides new insights for complex deformable object manipulation, especially in uncertain environments. Its performance in real-world applications demonstrates a strong synergy between theory and practice.