3D asynchronous multisensor target tracking performance with bias compensation

Sensor registration deals with the correction of sensor biases, and the registration error is vital in multiple sensor systems in order to carry out data fusion. This paper provides a new real time algorithm for three dimensional (3D) asynchronous sensor bias compensation based on decentralized fusion architecture. Firstly, the bias model and the bias estimation method for 3D asynchronous sensors with sensor location uncertainty are presented, and then the bias compensation is done considering the effect of the variance of bias estimates that is generally ignored in current literatures. Finally, four bias compensation algorithms, including compensation with the variance of bias estimates, compensation without it, perfect compensation and no compensation, are compared by Monte Carlo simulation. The simulation results show that tracking performance is significantly improved with the bias compensation using the variance of bias estimates compared with the bias compensation omitting it, when there are more uncertainties in bias estimation.