Trajectory Generation, Control, and Safety with Denoising Diffusion Probabilistic Models

TL;DR

This paper introduces a novel control framework combining denoising diffusion probabilistic models with control barrier functions to generate safe, optimal trajectories for physical systems, integrating safety constraints with task-specific rewards.

Contribution

It develops a new offline, model-based reinforcement learning approach that leverages DDPMs and CBFs for safe, optimal control in safety-critical systems.

Findings

Successfully integrates CBFs with DDPMs for trajectory planning.

Ensures safety constraints while optimizing task rewards.

Resembles model-predictive control with receding horizon.

Abstract

We present a framework for safety-critical optimal control of physical systems based on denoising diffusion probabilistic models (DDPMs). The technology of control barrier functions (CBFs), encoding desired safety constraints, is used in combination with DDPMs to plan actions by iteratively denoising trajectories through a CBF-based guided sampling procedure. At the same time, the generated trajectories are also guided to maximize a future cumulative reward representing a specific task to be optimally executed. The proposed scheme can be seen as an offline and model-based reinforcement learning algorithm resembling in its functionalities a model-predictive control optimization scheme with receding horizon in which the selected actions lead to optimal and safe trajectories.