Robotic Control Using Model Based Meta Adaption

TL;DR

This paper introduces a Meta Adaptation Controller (MAC) that leverages model-based meta-reinforcement learning to enable robots to quickly adapt to new tasks by replicating preferred behaviors learned from previous tasks, improving sample efficiency and behavior consistency.

Contribution

The paper proposes a novel MAC framework that applies meta-learning to model-based reinforcement learning, allowing robots to adapt behaviors across similar tasks without designing new reward functions.

Findings

MAC enables rapid adaptation to new tasks.

Agents behave appropriately without new reward functions.

Improves sample efficiency in robot control tasks.

Abstract

In machine learning, meta-learning methods aim for fast adaptability to unknown tasks using prior knowledge. Model-based meta-reinforcement learning combines reinforcement learning via world models with Meta Reinforcement Learning (MRL) for increased sample efficiency. However, adaption to unknown tasks does not always result in preferable agent behavior. This paper introduces a new Meta Adaptation Controller (MAC) that employs MRL to apply a preferred robot behavior from one task to many similar tasks. To do this, MAC aims to find actions an agent has to take in a new task to reach a similar outcome as in a learned task. As a result, the agent will adapt quickly to the change in the dynamic and behave appropriately without the need to construct a reward function that enforces the preferred behavior.