21 April 2022 in IEEE Robotics and Automation Letters
A reinforcement learning (RL) control policy could fail in a new/perturbed environment that is different from the training environment, due to the presence of dynamic variations. For controlling systems with continuous state and action spaces, we propose an add-on approach to robustifying a pre-trained RL policy by augmenting it with an ${\mathcal {L}_{1}}$ adaptive controller ( ${\mathcal {L}_{1}}$ AC). Leveraging the capability of an ${\mathcal {L}_{1}}$ AC for fast estimation and active compensation of dynamic variations, the proposed approach can improve the robustness of an RL policy which is trained either in a simulator or in the real world without consideration of a broad class of dynamic variations. Numerical and real-world experiments empirically demonstrate the efficacy of the proposed approach in robustifying RL policies trained using both model-free and model-based methods.
20 February 2025 in arXiv
In trajectory optimization, Model Predictive Path Integral (MPPI) control is a sampling-based Model Predictive Control (MPC) framework that generates optimal inputs by efficiently simulating numerous trajectories. In practice, however, MPPI often struggles to guarantee safety assurance and balance efficient sampling in open spaces with the need for more extensive exploration under tight constraints. To address this challenge, we incorporate discrete barrier states (DBaS) into MPPI and propose a novel MPPI-DBaS algorithm that ensures system safety and enables adaptive exploration across diverse scenarios. We evaluate our method in simulation experiments where the vehicle navigates through closely placed obstacles. The results demonstrate that the proposed algorithm significantly outperforms standard MPPI, achieving a higher success rate and lower tracking errors.