Learning Agile Flights through Narrow Gaps with Varying Angles using Onboard Sensing

TL;DR

This paper presents a novel onboard sensing and deep reinforcement learning approach enabling quadrotors to successfully navigate through unknown, tilted narrow gaps in real-world environments without prior environmental knowledge.

Contribution

It introduces a learning framework that combines onboard sensing with neural network control for variable-tilted gap traversal, achieving high success rates in real-world tests.

Findings

Success rate of 84.51% in real-world gap traversal

Effective transfer of simulation-trained policies to real-world scenarios

First real-world implementation of variable-tilted narrow gap traversal using learning

Abstract

This paper addresses the problem of traversing through unknown, tilted, and narrow gaps for quadrotors using Deep Reinforcement Learning (DRL). Previous learning-based methods relied on accurate knowledge of the environment, including the gap's pose and size. In contrast, we integrate onboard sensing and detect the gap from a single onboard camera. The training problem is challenging for two reasons: a precise and robust whole-body planning and control policy is required for variable-tilted and narrow gaps, and an effective Sim2Real method is needed to successfully conduct real-world experiments. To this end, we propose a learning framework for agile gap traversal flight, which successfully trains the vehicle to traverse through the center of the gap at an approximate attitude to the gap with aggressive tilted angles. The policy trained only in a simulation environment can be transferred into different domains with fine-tuning while maintaining the success rate. Our proposed framework, which integrates onboard sensing and a neural network controller, achieves a success rate of 84.51% in real-world experiments, with gap orientations up to 60deg. To the best of our knowledge, this is the first paper that performs the learning-based variable-tilted narrow gap traversal flight in the real world, without prior knowledge of the environment.