Exploring and Learning Structure: Active Inference Approach in Navigational Agents

TL;DR

This paper presents a biologically inspired Active Inference framework for navigation that enables agents to rapidly learn environmental structures with minimal prior knowledge, mimicking animal navigation strategies.

Contribution

The novel integration of topological mapping with Active Inference allows for quick, robust environment structure learning without prior environment information.

Findings

Rapid structure learning in a single episode

Effective navigation in ambiguous environments

Minimal overlap during exploration

Abstract

Drawing inspiration from animal navigation strategies, we introduce a novel computational model for navigation and mapping, rooted in biologically inspired principles. Animals exhibit remarkable navigation abilities by efficiently using memory, imagination, and strategic decision-making to navigate complex and aliased environments. Building on these insights, we integrate traditional cognitive mapping approaches with an Active Inference Framework (AIF) to learn an environment structure in a few steps. Through the incorporation of topological mapping for long-term memory and AIF for navigation planning and structure learning, our model can dynamically apprehend environmental structures and expand its internal map with predicted beliefs during exploration. Comparative experiments with the Clone-Structured Graph (CSCG) model highlight our model's ability to rapidly learn environmental structures in a single episode, with minimal navigation overlap. this is achieved without prior knowledge of the dimensions of the environment or the type of observations, showcasing its robustness and effectiveness in navigating ambiguous environments.