A New Autoregressive Neural Network Model with Command Compensation for Imitation Learning Based on Bilateral Control

TL;DR

This paper introduces a novel autoregressive neural network model with command compensation for bilateral control-based imitation learning, enhancing stability and success rates in robot learning tasks involving human-like demonstrations.

Contribution

The study proposes the S2SM model requiring only slave states, improving stability and task success over previous models, along with a new feedback controller for better reproducibility.

Findings

S2SM model outperforms SM2SM in stability and success rates.

The feedback controller improves reproducibility of robot imitation.

The approach effectively handles environmental variations.

Abstract

In the near future, robots are expected to work with humans or operate alone and may replace human workers in various fields such as homes and factories. In a previous study, we proposed bilateral control-based imitation learning that enables robots to utilize force information and operate almost simultaneously with an expert's demonstration. In addition, we recently proposed an autoregressive neural network model (SM2SM) for bilateral control-based imitation learning to obtain long-term inferences. In the SM2SM model, both master and slave states must be input, but the master states are obtained from the previous outputs of the SM2SM model, resulting in destabilized estimation under large environmental variations. Hence, a new autoregressive neural network model (S2SM) is proposed in this study. This model requires only the slave state as input and its outputs are the next slave and master states, thereby improving the task success rates. In addition, a new feedback controller that utilizes the error between the responses and estimates of the slave is proposed, which shows better reproducibility.