Coordinated Road Network Search for Multiple UAVs Using Dubins Path

Abstract

This paper proposes a coordinated road network search algorithm for multiple heterogeneous unmanned aerial vehicles (UAVs). The road network search problem can be interpreted as the problem to seek minimum-weight postman tours with a graphic representation of the road network. Therefore, the conventional Chinese Postman Problem (CPP) is first presented. We, then, consider physical constraints of UAVs into the search problem, since they cannot be addressed in the typical CPP. This modified search problem is formulated as Multi-choice Multidimensional Knapsack Problem (MMKP), which is to find an optimal solution minimising flight time. The Dubins path planning produces the shortest and flyable paths in consideration of physical constraints, so that the Dubins path is used to calculate the cost function of the modified search problem. The performance of the proposed multiple UAVs road network search algorithm is verified via numerical simulation for a given map.