Curriculum-based Sample Efficient Reinforcement Learning for Robust Stabilization of a Quadrotor

TL;DR

This paper presents a curriculum learning approach for reinforcement learning that efficiently trains a quadrotor to stabilize from random initial conditions, reducing training time and computational resources.

Contribution

The authors introduce a three-stage curriculum learning method that improves sample efficiency and robustness in training quadrotor stabilization policies.

Findings

The curriculum approach outperforms conventional one-stage training in performance.

It significantly reduces training samples and convergence time.

The trained policy demonstrates robustness in simulation and pose-tracking scenarios.

Abstract

This article introduces a novel sample-efficient curriculum learning (CL) approach for training an end-to-end reinforcement learning (RL) policy for robust stabilization of a Quadrotor. The learning objective is to simultaneously stabilize position and yaw-orientation from random initial conditions through direct control over motor RPMs (end-to-end), while adhering to pre-specified transient and steady-state specifications. This objective, relevant in aerial inspection applications, is challenging for conventional one-stage end-to-end RL, which requires substantial computational resources and lengthy training times. To address this challenge, this article draws inspiration from human-inspired curriculum learning and decomposes the learning objective into a three-stage curriculum that incrementally increases task complexity, while transferring knowledge from one stage to the next. In the proposed curriculum, the policy sequentially learns hovering, the coupling between translational and rotational degrees of freedom, and robustness to random non-zero initial velocities, utilizing a custom reward function and episode truncation conditions. The results demonstrate that the proposed CL approach achieves superior performance compared to a policy trained conventionally in one stage, with the same reward function and hyperparameters, while significantly reducing computational resource needs (samples) and convergence time. The CL-trained policy's performance and robustness are thoroughly validated in a simulation engine (Gym-PyBullet-Drones), under random initial conditions, and in an inspection pose-tracking scenario. A video presenting our results is available at https://youtu.be/9wv6T4eezAU.