Research on Route Optimization Based on Multiagent and Genetic Algorithm for Community Patrol

To solve the problem of route planning for security patrol in smart communities, a simulation framework comprising of multi-agent based model and genetic algorithm (GA) has been proposed for community patrol. The GA is used to determine and evolve the route collection and find the optimal results, while multi-agent simulation model can be used to set constraints and get the objective values of routes. First of all, in view of the traditional directed graph model is insufficient to describe the route information, GIS map is used as the environment of the community patrol inspection, which can efficiently reflect the environment features and facilitate the expansion of traffic and road information. Secondly, the task nodes are visited by the movement of the person agent. The implementation of the simulation system is based on Anylogic, which is beneficial for interacting GA program code. Simulation results show that not only the designed multi-agent system can obtain the optimal results, but also the route planning process is intuitive and visible, which meets the requirements of dynamic route planning.