GA-PSO-SVM indoor positioning algorithm based on Kalman filtering
In order to improve the accuracy of indoor positioning on smartphones,this paper established a pedestrian motion curve model and proposed an indoor positioning algorithm that combined Kalman filtering and support vector machine (SVM) with optimized parameters. Firstly,smartphones were used for collecting data from accelerometer and gyroscope. Secondly,the Kalman filtering algorithm was introduced to denoise the Gaussian white noise contained in the gyroscope data. Finally,SVM parameters were optimized using genetic algorithm (GA) and particle swarm optimization (PSO) algorithms to establish an indoor positioning model for gyroscope data prediction,which suppresses the constant drift in the gyroscope data. The indoor data collected by smartphones was used for experiments,and the results show that the positioning model proposed in this paper has significantly improved the accuracy of positioning compared to the method directly using microelectro mechanical system (MEMS) sensor data,with an average error reduction of 1.50 meters,which can meet the needs of indoor positioning services.
万方数据
维普
