Task and Domain Adaptive Reinforcement Learning for Robot Control

TL;DR

This paper introduces an adaptive reinforcement learning agent that uses transfer learning to enable robots to perform multiple tasks and adapt to environmental changes, validated through a blimp control challenge.

Contribution

The paper presents a novel transfer learning-based adaptive agent for robot control that can handle multiple tasks and environmental variations in real-world scenarios.

Findings

Successful zero-shot transfer to real-world blimp control

Enhanced multitasking capabilities in simulated environments

Effective adaptation to environmental changes

Abstract

Deep reinforcement learning (DRL) has shown remarkable success in simulation domains, yet its application in designing robot controllers remains limited, due to its single-task orientation and insufficient adaptability to environmental changes. To overcome these limitations, we present a novel adaptive agent that leverages transfer learning techniques to dynamically adapt policy in response to different tasks and environmental conditions. The approach is validated through the blimp control challenge, where multitasking capabilities and environmental adaptability are essential. The agent is trained using a custom, highly parallelized simulator built on IsaacGym. We perform zero-shot transfer to fly the blimp in the real world to solve various tasks. We share our code at https://github.com/robot-perception-group/adaptive_agent.