A Reinforcement Learning-Based Model for Mapping and Goal-Directed Navigation Using Multiscale Place Fields

TL;DR

This paper presents a novel reinforcement learning model for robot navigation that uses multiscale place fields, a replay-based reward system, and dynamic scale fusion to improve path efficiency and learning speed in complex environments.

Contribution

The paper introduces a new multiscale place field model with replay-based rewards and dynamic scale fusion, advancing bio-inspired navigation techniques.

Findings

Improved path efficiency over single-scale models

Faster learning in complex environments

Enhanced adaptability in navigation tasks

Abstract

Autonomous navigation in complex and partially observable environments remains a central challenge in robotics. Several bio-inspired models of mapping and navigation based on place cells in the mammalian hippocampus have been proposed. This paper introduces a new robust model that employs parallel layers of place fields at multiple spatial scales, a replay-based reward mechanism, and dynamic scale fusion. Simulations show that the model improves path efficiency and accelerates learning compared to single-scale baselines, highlighting the value of multiscale spatial representations for adaptive robot navigation.