MAVEN: A Meta-Reinforcement Learning Framework for Varying-Dynamics Expertise in Agile Quadrotor Maneuvers

TL;DR

MAVEN introduces a meta-reinforcement learning framework with a predictive context encoder that enables quadrotor policies to adapt rapidly to significant dynamic variations, achieving robust agile navigation in simulation and real-world scenarios.

Contribution

The paper presents MAVEN, a novel meta-RL framework with a predictive context encoder for fast adaptation across diverse quadrotor dynamics, reducing training time to less than an hour.

Findings

Achieves zero-shot sim-to-real transfer for high-speed maneuvers.

Handles mass variations up to 66.7% and thrust losses of 70%.

Converges in less than an hour using GPU-vectorized simulation.

Abstract

Reinforcement learning (RL) has emerged as a powerful paradigm for achieving online agile navigation with quadrotors. Despite this success, policies trained via standard RL typically fail to generalize across significant dynamic variations, exhibiting a critical lack of adaptability. This work introduces MAVEN, a meta-RL framework that enables a single policy to achieve robust end-to-end navigation across a wide range of quadrotor dynamics. Our approach features a novel predictive context encoder, which learns to infer a latent representation of the system dynamics from interaction history. We demonstrate our method in agile waypoint traversal tasks under two challenging scenarios: large variations in quadrotor mass and severe single-rotor thrust loss. We leverage a GPU-vectorized simulator to distribute tasks across thousands of parallel environments, overcoming the long training times of meta-RL to converge in less than an hour. Through extensive experiments in both simulation and the real world, we validate that MAVEN achieves superior adaptation and agility. The policy successfully executes zero-shot sim-to-real transfer, demonstrating robust online adaptation by performing high-speed maneuvers despite mass variations of up to 66.7% and single-rotor thrust losses as severe as 70%.