Developmental Reinforcement Learning of Control Policy of a Quadcopter UAV with Thrust Vectoring Rotors

TL;DR

This paper introduces a reinforcement learning-based control policy for a tilt-rotor quadcopter, leveraging transfer learning from a simpler UAV to improve learning speed, robustness, and fault tolerance in simulation tasks.

Contribution

It presents a novel developmental reinforcement learning approach that transfers policies from a simple quadcopter to a more complex tilt-rotor UAV, enhancing learning efficiency and robustness.

Findings

Faster learning of control policies compared to learning from scratch.

Demonstrated robustness in recovering from non-static initial conditions.

Superior fault tolerance of transferred policies over scratch-learned policies.

Abstract

In this paper, we present a novel developmental reinforcement learning-based controller for a quadcopter with thrust vectoring capabilities. This multirotor UAV design has tilt-enabled rotors. It utilizes the rotor force magnitude and direction to achieve the desired state during flight. The control policy of this robot is learned using the policy transfer from the learned controller of the quadcopter (comparatively simple UAV design without thrust vectoring). This approach allows learning a control policy for systems with multiple inputs and multiple outputs. The performance of the learned policy is evaluated by physics-based simulations for the tasks of hovering and way-point navigation. The flight simulations utilize a flight controller based on reinforcement learning without any additional PID components. The results show faster learning with the presented approach as opposed to learning the control policy from scratch for this new UAV design created by modifications in a conventional quadcopter, i.e., the addition of more degrees of freedom (4-actuators in conventional quadcopter to 8-actuators in tilt-rotor quadcopter). We demonstrate the robustness of our learned policy by showing the recovery of the tilt-rotor platform in the simulation from various non-static initial conditions in order to reach a desired state. The developmental policy for the tilt-rotor UAV also showed superior fault tolerance when compared with the policy learned from the scratch. The results show the ability of the presented approach to bootstrap the learned behavior from a simpler system (lower-dimensional action-space) to a more complex robot (comparatively higher-dimensional action-space) and reach better performance faster.