Diffusion Predictive Control with Constraints
This addresses the limitation of diffusion policies in dynamic conditions for robotics, but it is incremental as it builds on existing diffusion methods with constraint handling.
The paper tackles the problem of diffusion policies being unable to handle unseen constraints in robotics by proposing DPCC, which incorporates model-based projections and constraint tightening into the denoising process, resulting in improved constraint satisfaction in simulations of a robot manipulator.
Diffusion models have become popular for policy learning in robotics due to their ability to capture high-dimensional and multimodal distributions. However, diffusion policies are stochastic and typically trained offline, limiting their ability to handle unseen and dynamic conditions where novel constraints not represented in the training data must be satisfied. To overcome this limitation, we propose diffusion predictive control with constraints (DPCC), an algorithm for diffusion-based control with explicit state and action constraints that can deviate from those in the training data. DPCC incorporates model-based projections into the denoising process of a trained trajectory diffusion model and uses constraint tightening to account for model mismatch. This allows us to generate constraint-satisfying, dynamically feasible, and goal-reaching trajectories for predictive control. We show through simulations of a robot manipulator that DPCC outperforms existing methods in satisfying novel test-time constraints while maintaining performance on the learned control task.