Integrated data association and bias estimation in the presence of missed detections

This paper is concerned with performing the measurement-to-measurement association and bias estimation jointly in the presence of missed detections. An integrated mix integer programming (MINLP) model is established to determine the correspondence between local measurements and estimate sensor biases simultaneously. An alternation optimization mechanism is employed to solve the complicated MINLP model. A recursive version for bias estimation is developed that provides an access to deal with the measurement data sequentially. Monte Carlo simulation results are presented to illustrate our findings, as also demonstrating the effectiveness of the proposed approach.