Lightweight model estimates full shapes of flexible objects fast
Deformable Object Manipulation under Partial Observability via Real-Time Full-Shape Estimation
Robotics
Summary
Manipulating objects that bend and stretch, like ropes and fabrics, is hard because you can't see or measure their entire shape easily. The authors created a fast and simple computer model called cRVAE that guesses the whole shape of these objects using only limited information from their corners. This model works much faster than older methods and is accurate enough to control robots in real time. They tested it with simulations and a robot, showing it can handle complex shapes quickly without needing to know physical details upfront.
What this means in practice
- •For robotic system builders: Control robots to manipulate flexible objects using fast full-shape estimation from limited corner data.
- •For industrial automation engineers: Automate handling of deformable materials like fabric or cables with precise, real-time shape tracking for obstacle avoidance.
Authors
Kosar Behnia, Ville Kyrki, Gokhan Alcan
Abstract
Manipulating deformable objects (DOs) is challenging due to their high-dimensional state space, underactuated dynamics, and partial observability. In this paper, we propose cRVAE, a lightweight conditional recurrent variational autoencoder that estimates the full DO state from only partial corner-node observations during inference. The resulting model is used as the forward model in a receding-horizon optimal control framework for obstacle-aware collaborative DO manipulation. In simulation on rope and fabric, cRVAE estimates the full DO state from the available corner-node measurements alone, matching the accuracy of a parameter-identified XPBD model. At inference it uses no physical parameters as model inputs and performs no online parameter identification. It also runs approximately 350 times faster on the rope and over 1500 times faster on the fabric per forward pass, keeping horizon-based planning within the 100 ms control budget where XPBD exceeds it already at short horizons. Full-shape estimation from corner sensing at in-loop speed is what makes the model deployable on hardware, which we demonstrate on a Unitree Go2 robot.