Fast SLAM Algorithm Based on Distributed Target Bayesian Detection in Compound Gaussian Noise

Abstract

Fast detecting the distributed targets is appreciated in the SLAM algorithm. In this paper, Bayesian Rao detection and Wald detection methods are proposed for the distributed targets detecting in compound Gaussian noise, which can be applied to the rapid generation and iteration of SLAM (simultaneous localization and mapping) algorithm. Firstly, it is assumed that the covariance matrix of compound Gaussian noise obeys the inverse Wishart prior distribution, on this basis, the maximum posterior estimate of the covariance matrix is used to propose Bayesian Rao detection and Wald detection, then the affects by the prior distribution degree of freedom and the number of distance units spanned on the performance of the proposed detector are analyzed, Finally, Simulation and experiment results verify the effectiveness of the proposed method. Applying the detection method to the SLAM algorithm, there will be a significant improvement in detecting the distributed target.