Mobile Crowdsourcing Task Allocation in Armed Reconnaissance Operations

In contemporary urban warfare, human armed reconnaissance assumes an indispensable role in serving as a supplement to intelligent reconnaissance technology. How to allocate armed reconnaissance tasks efficiently and rationally plays a crucial role in the process of urban warfare and the outcome of the mission. Among numerous task allocation solutions, the mobile crowdsourcing is popular for the widespread use. In this paper, we introduce the pattern of armed reconnaissance in urban operations and summarize Few Squads-More Tasks (FSMT) as the problem of mobile crowd sourcing one-to-many task allocation. Then, we treat the problem as a Multiple Traveling Salesman Problem (MTSP) model of multiple starting points, which is solved by maximizing the number of reconnaissance tasks under the constraints of reconnaissance squad distances. To find the solutions, we propose the cluster partitioning and improved genetics based Two-Stage Heuristic Algorithm (TSHA). Finally, the optimization effect of the proposed model on the assignment of armed reconnaissance tasks in the urban battlefield is demonstrated through experimental simulation, by which the superiority of the improved algorithm in enhancing the efficiency of task allocation is verified.