Multi-Task Reinforcement Learning of Drone Aerobatics by Exploiting Geometric Symmetries

TL;DR

This paper introduces GEAR, a multi-task reinforcement learning framework that leverages geometric symmetries to improve data efficiency and robustness in drone aerobatic control, enabling successful execution of diverse maneuvers.

Contribution

The paper presents a novel equivariant RL architecture that exploits SO(2) symmetry, enhancing multi-task learning for drone aerobatics with improved efficiency and generalization.

Findings

GEAR achieves 98.85% success rate across aerobatic tasks.

It demonstrates stable execution of multiple maneuvers in real-world tests.

The framework effectively combines motion primitives for complex aerobatics.

Abstract

Flight control for autonomous micro aerial vehicles (MAVs) is evolving from steady flight near equilibrium points toward more aggressive aerobatic maneuvers, such as flips, rolls, and Power Loop. Although reinforcement learning (RL) has shown great potential in these tasks, conventional RL methods often suffer from low data efficiency and limited generalization. This challenge becomes more pronounced in multi-task scenarios where a single policy is required to master multiple maneuvers. In this paper, we propose a novel end-to-end multi-task reinforcement learning framework, called GEAR (Geometric Equivariant Aerobatics Reinforcement), which fully exploits the inherent SO(2) rotational symmetry in MAV dynamics and explicitly incorporates this property into the policy network architecture. By integrating an equivariant actor network, FiLM-based task modulation, and a multi-head critic, GEAR achieves both efficiency and flexibility in learning diverse aerobatic maneuvers, enabling a data-efficient, robust, and unified framework for aerobatic control. GEAR attains a 98.85\% success rate across various aerobatic tasks, significantly outperforming baseline methods. In real-world experiments, GEAR demonstrates stable execution of multiple maneuvers and the capability to combine basic motion primitives to complete complex aerobatics.