DiffSRL: Learning Dynamical State Representation for Deformable Object Manipulation with Differentiable Simulator
This addresses the challenge of dynamic state representation learning for deformable object manipulation in robotics, which is incremental as it builds on existing methods by integrating differentiable simulation.
The paper tackles the problem of learning dynamic state representations for complex systems like deformable objects, which current methods scale poorly on, by proposing DiffSRL, a pipeline that uses differentiable simulation to embed dynamics models into training; it demonstrates superior performance in capturing long-term dynamics and reward prediction on the PlasticineLab benchmark.
Dynamic state representation learning is an important task in robot learning. Latent space that can capture dynamics related information has wide application in areas such as accelerating model free reinforcement learning, closing the simulation to reality gap, as well as reducing the motion planning complexity. However, current dynamic state representation learning methods scale poorly on complex dynamic systems such as deformable objects, and cannot directly embed well defined simulation function into the training pipeline. We propose DiffSRL, a dynamic state representation learning pipeline utilizing differentiable simulation that can embed complex dynamics models as part of the end-to-end training. We also integrate differentiable dynamic constraints as part of the pipeline which provide incentives for the latent state to be aware of dynamical constraints. We further establish a state representation learning benchmark on a soft-body simulation system, PlasticineLab, and our model demonstrates superior performance in terms of capturing long-term dynamics as well as reward prediction.