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

2026-02-11

AI Summary

This work addresses the challenges of low data efficiency and poor generalization in traditional reinforcement learning for multi-task drone acrobatics. The authors propose GEAR, a novel framework that, for the first time, integrates SO(2) geometric equivariance into an end-to-end multi-task policy network. By combining FiLM-based task modulation with a multi-head critic architecture, GEAR embeds dynamical symmetry as an inductive bias, yielding a highly sample-efficient control policy. The method achieves a 98.85% success rate across diverse acrobatic maneuvers, substantially outperforming existing baselines. Furthermore, it demonstrates robust generalization and practical applicability by stably executing complex composite maneuvers on a real-world drone platform.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

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.

Problem

• multi-task reinforcement learning
• drone aerobatics
• micro aerial vehicles
• data efficiency
• policy generalization

Innovation

• Geometric Equivariance
• Multi-Task Reinforcement Learning
• SO(2) Symmetry
• Aerobatic Control
• Equivariant Policy Network