Learning to plan with uncertain topological maps

TL;DR

This paper introduces a neural network-based planning method for navigation in uncertain topological maps, leveraging visual features and an inductive bias for dynamic programming to outperform classical algorithms in simulated 3D environments.

Contribution

It presents a novel data-driven neural approach for planning under uncertainty that incorporates rich node features and mimics classical shortest path algorithms.

Findings

Neural planner with visual features outperforms symbolic algorithms.

The neural model aligns with Bellman-Ford algorithm principles.

Empirical results in simulated environments validate the approach.

Abstract

We train an agent to navigate in 3D environments using a hierarchical strategy including a high-level graph based planner and a local policy. Our main contribution is a data driven learning based approach for planning under uncertainty in topological maps, requiring an estimate of shortest paths in valued graphs with a probabilistic structure. Whereas classical symbolic algorithms achieve optimal results on noise-less topologies, or optimal results in a probabilistic sense on graphs with probabilistic structure, we aim to show that machine learning can overcome missing information in the graph by taking into account rich high-dimensional node features, for instance visual information available at each location of the map. Compared to purely learned neural white box algorithms, we structure our neural model with an inductive bias for dynamic programming based shortest path algorithms, and we show that a particular parameterization of our neural model corresponds to the Bellman-Ford algorithm. By performing an empirical analysis of our method in simulated photo-realistic 3D environments, we demonstrate that the inclusion of visual features in the learned neural planner outperforms classical symbolic solutions for graph based planning.