Comparison of Model-Free and Model-Based Learning-Informed Planning for PointGoal Navigation

TL;DR

This paper compares model-free deep reinforcement learning and model-based POMDP approaches for point goal navigation, showing the model-based method's robustness and data efficiency with competitive performance.

Contribution

It adapts a POMDP sub-goal framework with semantic mapping, demonstrating its robustness and data efficiency over deep RL methods in indoor navigation.

Findings

Model-based POMDP approach is robust and efficient.

Compared to deep RL, it requires less data.

Achieves comparable performance to state-of-the-art deep RL methods.

Abstract

In recent years several learning approaches to point goal navigation in previously unseen environments have been proposed. They vary in the representations of the environments, problem decomposition, and experimental evaluation. In this work, we compare the state-of-the-art Deep Reinforcement Learning based approaches with Partially Observable Markov Decision Process (POMDP) formulation of the point goal navigation problem. We adapt the (POMDP) sub-goal framework proposed by [1] and modify the component that estimates frontier properties by using partial semantic maps of indoor scenes built from images' semantic segmentation. In addition to the well-known completeness of the model-based approach, we demonstrate that it is robust and efficient in that it leverages informative, learned properties of the frontiers compared to an optimistic frontier-based planner. We also demonstrate its data efficiency compared to the end-to-end deep reinforcement learning approaches. We compare our results against an optimistic planner, ANS and DD-PPO on Matterport3D dataset using the Habitat Simulator. We show comparable, though slightly worse performance than the SOTA DD-PPO approach, yet with far fewer data.