Active Neural Localization

TL;DR

Active Neural Localizer is a differentiable neural network that learns to efficiently and accurately localize an agent in various environments by combining traditional filtering ideas with reinforcement learning, capable of generalizing across environments.

Contribution

It introduces a fully differentiable neural localization model that integrates belief propagation with learned policies, trained end-to-end with reinforcement learning, for improved efficiency and accuracy.

Findings

Effective in 2D maze environments.

Capable of joint policy and perceptual learning from raw pixels.

Generalizes from simulated to photo-realistic environments.

Abstract

Localization is the problem of estimating the location of an autonomous agent from an observation and a map of the environment. Traditional methods of localization, which filter the belief based on the observations, are sub-optimal in the number of steps required, as they do not decide the actions taken by the agent. We propose "Active Neural Localizer", a fully differentiable neural network that learns to localize accurately and efficiently. The proposed model incorporates ideas of traditional filtering-based localization methods, by using a structured belief of the state with multiplicative interactions to propagate belief, and combines it with a policy model to localize accurately while minimizing the number of steps required for localization. Active Neural Localizer is trained end-to-end with reinforcement learning. We use a variety of simulation environments for our experiments which include random 2D mazes, random mazes in the Doom game engine and a photo-realistic environment in the Unreal game engine. The results on the 2D environments show the effectiveness of the learned policy in an idealistic setting while results on the 3D environments demonstrate the model's capability of learning the policy and perceptual model jointly from raw-pixel based RGB observations. We also show that a model trained on random textures in the Doom environment generalizes well to a photo-realistic office space environment in the Unreal engine.