Indoor Localization Using Mind Evolutionary Algorithm-Based Geomagnetic Positioning and Smartphone IMU Sensors

With the pervasiveness and ubiquitous distribution of magnetic field in indoor environments, localization using positioning (MP) has attracted considerable attention. This work concentrates on MP pedestrian dead reckoning (PDR) method, constructs a fusion system for smartphones PDR based extended Kalman filter (EKF). The mind evolutionary algorithm (MEA) is introduced to search optimal position heuristic searching strategy, which uses similartaxis dissimilation operation. In module, acceleration characteristics different walking patterns are analyzed corresponding features extracted. enhanced genetic algorithm-based extreme learning machine (EGA-ELM) adopted train these address gait recognition problem patterns. Finally, obtain lightweight high-precision MEA-based integrated with EKF. Extensive experiments conducted evaluate proposed methods. testing results showed that can location error within 2.3 m steps be recognized mean accuracy 95% when users participate testing. after reveal root-mean-square (RMSE) 1.25 1.53 respectively, outperforms MP, PDR, methods improved particle (IPF) (GPF).