Neural Network Memory Architectures for Autonomous Robot Navigation

TL;DR

This paper investigates how different neural network memory architectures affect autonomous robot navigation, emphasizing the importance of memory for better generalization and proposing new evaluation tools for these models.

Contribution

It provides a comprehensive analysis of memory structures in neural networks for robot navigation and introduces a novel VC-dimension based evaluation method for generalization.

Findings

Memory structures improve navigation performance in unseen scenarios.

VC-dimension estimates correlate with actual test performance.

Feedforward, LSTM, and DNC networks show varying generalization abilities.

Abstract

This paper highlights the significance of including memory structures in neural networks when the latter are used to learn perception-action loops for autonomous robot navigation. Traditional navigation approaches rely on global maps of the environment to overcome cul-de-sacs and plan feasible motions. Yet, maintaining an accurate global map may be challenging in real-world settings. A possible way to mitigate this limitation is to use learning techniques that forgo hand-engineered map representations and infer appropriate control responses directly from sensed information. An important but unexplored aspect of such approaches is the effect of memory on their performance. This work is a first thorough study of memory structures for deep-neural-network-based robot navigation, and offers novel tools to train such networks from supervision and quantify their ability to generalize to unseen scenarios. We analyze the separation and generalization abilities of feedforward, long short-term memory, and differentiable neural computer networks. We introduce a new method to evaluate the generalization ability by estimating the VC-dimension of networks with a final linear readout layer. We validate that the VC estimates are good predictors of actual test performance. The reported method can be applied to deep learning problems beyond robotics.