Real-Time Robotic Search using Hierarchical Spatial Point Processes

TL;DR

This paper introduces a hierarchical probabilistic model for aerial robot search and rescue that optimizes search efficiency by incorporating spatial priors and real-time inference, significantly improving victim detection rates.

Contribution

The paper presents a novel hierarchical spatial point process model combined with INLA and Monte Carlo tree search for real-time SAR planning, outperforming traditional methods.

Findings

Outperforms traditional search strategies in simulated experiments.

Finds up to ten times more injured victims in initial hours.

Incorporates geographic and traffic priors for improved search efficiency.

Abstract

Aerial robots hold great potential for aiding Search and Rescue (SAR) efforts over large areas. Traditional approaches typically searches an area exhaustively, thereby ignoring that the density of victims varies based on predictable factors, such as the terrain, population density and the type of disaster. We present a probabilistic model to automate SAR planning, with explicit minimization of the expected time to discovery. The proposed model is a hierarchical spatial point process with three interacting spatial fields for i) the point patterns of persons in the area, ii) the probability of detecting persons and iii) the probability of injury. This structure allows inclusion of informative priors from e.g. geographic or cell phone traffic data, while falling back to latent Gaussian processes when priors are missing or inaccurate. To solve this problem in real-time, we propose a combination of fast approximate inference using Integrated Nested Laplace Approximation (INLA), and a novel Monte Carlo tree search tailored to the problem. Experiments using data simulated from real world GIS maps show that the framework outperforms traditional search strategies, and finds up to ten times more injured in the crucial first hours.