Multi-Robot Transfer Learning: A Dynamical System Perspective

TL;DR

This paper investigates the properties of optimal transfer maps in multi-robot transfer learning, revealing they are generally dynamic systems, and proposes an algorithm to determine their properties without detailed robot models, validated on quadrotors.

Contribution

The paper introduces a novel algorithm to identify the properties of optimal dynamic transfer maps in multi-robot transfer learning without detailed robot dynamics.

Findings

Optimal transfer maps are generally dynamic systems.

The proposed algorithm accurately determines transfer map properties.

Experimental results show 60-70% reduction in transfer error.

Abstract

Multi-robot transfer learning allows a robot to use data generated by a second, similar robot to improve its own behavior. The potential advantages are reducing the time of training and the unavoidable risks that exist during the training phase. Transfer learning algorithms aim to find an optimal transfer map between different robots. In this paper, we investigate, through a theoretical study of single-input single-output (SISO) systems, the properties of such optimal transfer maps. We first show that the optimal transfer learning map is, in general, a dynamic system. The main contribution of the paper is to provide an algorithm for determining the properties of this optimal dynamic map including its order and regressors (i.e., the variables it depends on). The proposed algorithm does not require detailed knowledge of the robots' dynamics, but relies on basic system properties easily obtainable through simple experimental tests. We validate the proposed algorithm experimentally through an example of transfer learning between two different quadrotor platforms. Experimental results show that an optimal dynamic map, with correct properties obtained from our proposed algorithm, achieves 60-70% reduction of transfer learning error compared to the cases when the data is directly transferred or transferred using an optimal static map.