A Registration Algorithm of Different Sensors Based on EM-EKF

Expectation-Maximization(EM) algorithm plays an important role in parameter estimation and in dealing with incomplete data.Considering the unknown parameter of the registration process,the EM algorithm is incorporated with EKF to give the simultaneous registration and track fusion.The algorithm simplifies the optimizing estimation of the maximum likelihood by making the expectation maximal.EKF and smoothing process realize the more exact estimation of the expectation.Registration is realized effectively with iterative algorithm.The simulation result shows the effectiveness of the method.