Study Of Cooperative Localization Algorithm In Mobile Robot And Wireless Sensor Networks

With the development of science and technology, the field of robot application is more and more extensive. Mobile robot and wireless sensor networks cooperative localization gains widely attention because of its high location accuracy. The main starting point of this paper is to improve the accuracy of mobile robot and wireless sensor networks cooperative localization, an improved marginal particle filter which is putted the newest observation information into the proposal distribution is proposed, and the algorithm is used to robot and sensor networks cooperative localization. Content can be summarized as follows:First, the background and significance of the subject is introduced, and then the development of robot location and SLAM problem is described. The purpose of the subject is confirmed.Second, some theory and algorithm needed to know about the subject is expounded.To deal with the problems that the dimension of the state space of particle filter grows with the time increases and higher variance of the importance weights, marginal particle filter is introduced. The marginal particle filter can make filtering estimation in low-dimensional state space and reduce variance of the importance weights.Third, the quality of state estimation is decided by the closeness of the importance probability density function and true posterior probability density function.To make the proposal distribution function closer to posterior distribution function, the most recent observation is put into proposal distribution based marginal particle filter. It makes the proposal distribution function closer to posterior distribution function and improve positioning accuracy.In addition, mobile robot and wireless sensor networks cooperative localization can be equivalent to SLAM problem, so we can apply the improved marginal particle filter mentioned above into solution of cooperative localization. Through the experimental comparison, we find that improved algorithm reduces the positioning error of mobile robot and wireless sensor networks, and the results verify the effectiveness and superiority of improved algorithm.