Reinforcement Learning with Formal Performance Metrics for Quadcopter Attitude Control under Non-nominal Contexts

TL;DR

This paper applies reinforcement learning to quadcopter attitude control, introducing formal performance metrics and robustness evaluation under various non-nominal conditions, including motor failures and wind gusts.

Contribution

It presents a comprehensive methodology for designing robust RL-based controllers with formal performance metrics for quadcopters under challenging conditions.

Findings

Controllers demonstrate robustness to motor failures and wind gusts.

The approach enables reproducible RL controller design with detailed methodology.

Formal metrics effectively evaluate vehicle behavior and controller performance.

Abstract

We explore the reinforcement learning approach to designing controllers by extensively discussing the case of a quadcopter attitude controller. We provide all details allowing to reproduce our approach, starting with a model of the dynamics of a crazyflie 2.0 under various nominal and non-nominal conditions, including partial motor failures and wind gusts. We develop a robust form of a signal temporal logic to quantitatively evaluate the vehicle's behavior and measure the performance of controllers. The paper thoroughly describes the choices in training algorithms, neural net architecture, hyperparameters, observation space in view of the different performance metrics we have introduced. We discuss the robustness of the obtained controllers, both to partial loss of power for one rotor and to wind gusts and finish by drawing conclusions on practical controller design by reinforcement learning.