Sensor Scheduling For Target Tracking Using Particle Swarm Optimization

This paper presents a new solution to the problem of optimal sensor scheduling for tracking a target with several noisy sensor measurements. The state of the target is modeled as a linear Gaussian model and the measurements are assumed linearly related to the state model and impaired by Gaussian noise. The state and the mean square error (MSE) of the estimated state can be calculated recursively by Kalman filtering technique. Each measurement is associated with measurement error, usage cost and physical and computational constraints. We consider the sensor scheduling problem as finding the optimal sequence of the sensors in order to minimize the measurement error and sensor usage cost for the entire time horizon subject to satisfying the constraints under consideration. In this paper we use the particle swarm optimization to find a sub-optimal sensor schedule. We study a numerical problem with tracking a vehicle with three noisy sensors and results show that the sensor scheduling obtained from the proposed method is very close to the optimal solution within a reasonable number of iterations