A UAV Allocation Method for Traffic Surveillance in Sparse Road Network

The method of choosing the surveillance targets was proposed, and the K-means clustering algorithm was used to divide the UAV surveillance area into multiple sub-zones to convert this problem into UAV traffic surveillance scenarios without continuous flight distance constraint.

Abstract

Unmanned aerial vehicle (UAV) technology was introduced in traffic surveillance in sparse road networks, and a UAV allocation method with/without UAV continuous flight distance constraint was proposed. First, the method of choosing the surveillance targets was proposed. The UAV traffic surveillance problem without maximum flight distance constraint was then formulated as a traveling salesman problem, and the simulated annealing algorithm was introduced to solve this problem. As for UAV traffic surveillance problem with continuous flight distance constraint, the K-means clustering algorithm was used to divide the UAV surveillance area into multiple sub-zones to convert this problem into UAV traffic surveillance scenarios without continuous flight distance constraint. Finally, taking the Korla-Kuqa expressway of Xinjiang and its road network as the example, the proposed UAV-based traffic surveillance allocation method for sparse road networks was demonstrated and validated using several field experiments. The experimental results show that UAV is an effective and useful tool for traffic surveillance in the sparse road networks of China's western regions.