Hypernetwork-Conditioned Reinforcement Learning for Robust Control of Fixed-Wing Aircraft under Actuator Failures
For drone control engineers, this work offers a method to improve robustness to actuator failures, though the improvement is incremental over existing RL-based control.
This paper develops a reinforcement learning controller for fixed-wing drones that remains robust to actuator failures by conditioning the policy on fault parameters via hypernetworks. The approach achieves better generalization to unseen time-varying failures than standard MLP policies, validated in high-fidelity simulations.
This paper presents a reinforcement learning-based path-following controller for a fixed-wing small uncrewed aircraft system (sUAS) that is robust to certain actuator failures. The controller is conditioned on a parameterization of actuator faults using hypernetwork-based adaptation. We consider parameter-efficient formulations based on Feature-wise Linear Modulation (FiLM) and Low-Rank Adaptation (LoRA), trained using proximal policy optimization. We demonstrate that hypernetwork-conditioned policies can improve robustness compared to standard multilayer perceptron policies. In particular, hypernetwork-conditioned policies generalize effectively to time-varying actuator failure modes not encountered during training. The approach is validated through high-fidelity simulations, using a realistic six-degree-of-freedom fixed-wing aircraft model.