Resource-Constrained Task Assignment for Event-Driven Sensing Using Public Bicycles in Smart Cities

Abstract

Many cities have provided public bicycle services to reduce traffic congestion and air pollution. Mobility of public bicycles makes them very suitable for event-driven sensing for smart cities, i.e., collecting data relevant to special events such as car accidents or street parades. The problem is how to assign a set of bicycles to best fulfil the sensing mission, considering the constrained storage, battery energy, and communication capability of bicycles. The problem is referred to as the resource-constrained task assignment for event-driven sensing (ReConTAES). The goal is to minimise the number of bicycles used, while balancing the energy consumption of the selected bicycles. We first formulate the problem as a mixed integer programming and then propose a set of greedy heuristics to solve the problem. We evaluate the proposed algorithms by using real trajectories to show their feasibility.