Fast Neural-Network Approximation of Active Target Search Under Uncertainty

TL;DR

This paper introduces a neural network-based approach to approximate active target search decisions, significantly reducing online computation while maintaining high detection accuracy.

Contribution

The authors propose a convolutional neural network that approximates existing search planners, enabling faster decision-making in target search tasks under uncertainty.

Findings

The neural network achieves detection rates comparable to traditional planners.

The approach reduces online computation by orders of magnitude.

Simulations demonstrate effectiveness across different target distributions.

Abstract

We address the problem of searching for an unknown number of stationary targets at unknown positions with a mobile agent. A probability hypothesis density filter is used to estimate the expected number of targets under measurement uncertainty. Existing planners, such as Active Search (AS) and its Intermittent variant (ASI), achieve accurate detection but require costly online optimization. To reduce online computation, we propose to use a convolutional neural network to approximate AS or ASI decisions through direct inference. The network is trained on AS/ASI data using a multi-channel grid that encodes target beliefs, the agent position, visitation history, and boundary information. Simulations with uniform and clustered target distributions show that the network achieves detection rates comparable to AS or ASI while reducing computation by orders of magnitude.