Time-optimal Flight in Cluttered Environments via Safe Reinforcement Learning

TL;DR

This paper introduces a safe reinforcement learning method enabling quadrotors to navigate cluttered environments quickly and safely, outperforming existing algorithms in time efficiency and obstacle avoidance.

Contribution

The paper proposes a novel reinforcement learning approach with safety and terminal rewards for time-optimal, collision-free drone navigation in complex environments.

Findings

Outperforms state-of-the-art algorithms in flight time and safety.

Achieves 66.7% success rate in unseen challenging environments.

Demonstrates effective obstacle avoidance and time minimization.

Abstract

This paper addresses the problem of guiding a quadrotor through a predefined sequence of waypoints in cluttered environments, aiming to minimize the flight time while avoiding collisions. Previous approaches either suffer from prolonged computational time caused by solving complex non-convex optimization problems or are limited by the inherent smoothness of polynomial trajectory representations, thereby restricting the flexibility of movement. In this work, we present a safe reinforcement learning approach for autonomous drone racing with time-optimal flight in cluttered environments. The reinforcement learning policy, trained using safety and terminal rewards specifically designed to enforce near time-optimal and collision-free flight, outperforms current state-of-the-art algorithms. Additionally, experimental results demonstrate the efficacy of the proposed approach in achieving both minimum flight time and obstacle avoidance objectives in complex environments, with a commendable $66.7\%$ success rate in unseen, challenging settings.