Generating Goal-Directed Visuomotor Plans Based on Learning Using a Predictive Coding-type Deep Visuomotor Recurrent Neural Network Model

TL;DR

This paper demonstrates how a predictive coding deep recurrent neural network can generate goal-directed visuomotor plans from prior experience, enabling a robot to plan for unlearned goals through learned predictive models.

Contribution

It introduces a predictive coding-based deep recurrent neural network that learns to generate visuomotor plans and generalizes to new goals in a robot arm task.

Findings

The model successfully generates goal-directed plans for unlearned goals.

Error minimization enables inference of optimal intention states.

Large training datasets improve generalization performance.

Abstract

The current paper presents how a predictive coding type deep recurrent neural networks can generate vision-based goal-directed plans based on prior learning experience by examining experiment results using a real arm robot. The proposed deep recurrent neural network learns to predict visuo-proprioceptive sequences by extracting an adequate predictive model from various visuomotor experiences related to object-directed behaviors. The predictive model was developed in terms of mapping from intention state space to expected visuo-proprioceptive sequences space through iterative learning. Our arm robot experiments adopted with three different tasks with different levels of difficulty showed that the error minimization principle in the predictive coding framework applied to inference of the optimal intention states for given goal states can generate goal-directed plans even for unlearned goal states with generalization. It was, however, shown that sufficient generalization requires relatively large number of learning trajectories. The paper discusses possible countermeasure to overcome this problem.