Research On Indoor Positioning Technology Of Unmanned Vehicle Based On Multi-sensor Fusion

With the change of life style brought by the progress of science and technology,people’s demand for indoor location information is increasing day by day.At present,many indoor navigation and positioning technologies have been proposed,among which ultra-wideband(UWB)positioning technology has shown its importance in indoor positioning by virtue of its high time resolution and good ranging performance.In addition,inertial navigation system(INS)is widely used in indoor positioning due to its advantages of not easy to be affected by the environment and high positioning accuracy.The combination of UWB and INS is a combination technology that has been widely studied and concerned recently and has become the mainstream combination scheme to solve indoor positioning.Therefore,the combined localization of UWB and INS in this paper has important research value and application prospect.This paper first consulted the domestic and foreign references,summed up the research status and development trend of the field,and then for the indoor positioning of the scheme of each part of the design.The basic elements and principle of INS are analyzed,the update method of strapdown inertial navigation is deduced,the error source is analyzed and the error equation is deduced.For UWB technology,the basic principle and four location methods based on time of arrival,time difference of arrival,angle of arrival and received signal strength indication are analyzed,and the error sources of UWB signal are discussed.Then the error based on the time of flight is described,and the TOA method is selected as the distance measurement method.Then,two localization algorithms based on TOA--Chan algorithm and Taylor algorithm are studied,and an improved chan-Taylor fusion localization algorithm based on simulated annealing algorithm(SA)is presented by combining their advantages and disadvantages with simulated annealing algorithm.At the same time,the accuracy of several algorithms is evaluated,and the error evaluation criteria of mean absolute error,root mean square error,cumulative distribution function and cramero lower bound are given to judge the improvement degree of the improved algorithm.Aiming at the problems of multi-sensor fusion algorithm,this paper studies the improvement of particle filter fusion algorithm based on improved bat algorithm.In this paper,the bat algorithm(BA)of swarm intelligence algorithm is introduced,but at the same time,BA has some problems,such as lack of variation,easy to fall into local minimum.To solve these problems,the iterative updating mechanism of BA is improved,and a scheme based on improved bat algorithm improved particle filter(IBA-PF)is proposed.IBA-PF and its basic algorithm are analyzed by using ideal nonlinear model and ideal carrier motion model,and the fusion accuracy is further improved.Finally,the static and dynamic positioning tests of SA-Chan-Taylor algorithm and the basic algorithm are carried out with Mecanum wheel trolley platform,and the accuracy of the improved algorithm and the basic algorithm are compared according to the accuracy evaluation of the algorithm.At the same time,the influence of geometric precision factor on positioning accuracy is considered.In the combined positioning experiment,the UWB positioning results of SA-Chan-Taylor algorithm and the semiphysical INS are combined with the IBA-PF algorithm.The experimental results show that the UWB/INS combination positioning results can effectively improve the accuracy of the vehicle positioning algorithm,and improve the positioning effect under the condition of non-line-of-sight.