FLARE: Agile Flights for Quadrotor Cable-Suspended Payload System via Reinforcement Learning

TL;DR

FLARE introduces a reinforcement learning framework enabling quadrotors with cable-suspended payloads to perform agile, real-time navigation maneuvers with superior speed and safety, validated through simulation and real-world experiments.

Contribution

The paper presents FLARE, a novel RL-based method for agile quadrotor payload navigation that outperforms traditional optimization techniques in speed and transferability.

Findings

3x faster gate traversal compared to optimization-based methods

Successful zero-shot sim-to-real transfer demonstrating real-time agility

Effective in complex, challenging flight scenarios

Abstract

Agile flight for the quadrotor cable-suspended payload system is a formidable challenge due to its underactuated, highly nonlinear, and hybrid dynamics. Traditional optimization-based methods often struggle with high computational costs and the complexities of cable mode transitions, limiting their real-time applicability and maneuverability exploitation. In this letter, we present FLARE, a reinforcement learning (RL) framework that directly learns agile navigation policy from high-fidelity simulation. Our method is validated across three designed challenging scenarios, notably outperforming a state-of-the-art optimization-based approach by a 3x speedup during gate traversal maneuvers. Furthermore, the learned policies achieve successful zero-shot sim-to-real transfer, demonstrating remarkable agility and safety in real-world experiments, running in real time on an onboard computer.