A Decision-theoretic Approach to Detection-based Target Search with a UAV

TL;DR

This paper introduces a POMDP-based method for UAV target search that accounts for noisy observations and motion errors, significantly improving search efficiency over heuristic approaches.

Contribution

It presents a novel POMDP formulation for UAV target search that incorporates observation and motion uncertainties, enabling more effective real-time decision-making.

Findings

POMDP policy finds targets up to 3.4 times faster than heuristics.

The approach effectively handles noisy observations and motion errors.

Offline learning of policies improves real-time UAV search performance.

Abstract

Search and rescue missions and surveillance require finding targets in a large area. These tasks often use unmanned aerial vehicles (UAVs) with cameras to detect and move towards a target. However, common UAV approaches make two simplifying assumptions. First, they assume that observations made from different heights are deterministically correct. In practice, observations are noisy, with the noise increasing as the height used for observations increases. Second, they assume that a motion command executes correctly, which may not happen due to wind and other environmental factors. To address these, we propose a sequential algorithm that determines actions in real time based on observations, using partially observable Markov decision processes (POMDPs). Our formulation handles both observations and motion uncertainty and errors. We run offline simulations and learn a policy. This policy is run on a UAV to find the target efficiently. We employ a novel compact formulation to represent the coordinates of the drone relative to the target coordinates. Our POMDP policy finds the target up to 3.4 times faster when compared to a heuristic policy.