It is widely acknowledged that navigation is a significant source of between sites.The Global Positioning System(GPS)has numerous navi-gational advancements,and hence it is used widely.GPS navigation can be compromise...It is widely acknowledged that navigation is a significant source of between sites.The Global Positioning System(GPS)has numerous navi-gational advancements,and hence it is used widely.GPS navigation can be compromised at any level between position,location,and estimation,to the detriment of the user.Consequently,a navigation system requires the precise location and underpinning tracking of an object without signal loss.The objective of a hybrid environment prediction system is to foresee the location of the user and their territory by employing a variety of sensors for position estimation and monitoring navigation.This article presents a state estimation of the relative position for indoor and outdoor activity solved with a state estimation algorithm utilizing Kalman filter.Also,a comparative study of variants of the Kalman filter,where linearizing current mean and covariance with nonlinear state estimation as an approach of Extended Kalman Filter(EFK)is applied to the collected data.The third comparative aspect uses probability distribution for the selected points with a Sigma Point Kalman Filter(SPKF)for evaluating an accelerometer,gyroscope,and GPS data in hybrid environments for various activities for different data collection scenar-ios from users.The findings of the presented model demonstrate the robust performance of all forms of the Kalman filter algorithm for diverse user-performed activities in totally contaminated indoor and outdoor environ-ments.Experimental findings with various patterns and data,conducted by different subjects using multiple modes of navigation,show that the approach can indeed lead to the intelligent development of sensor-based navigation and monitoring.State estimation and prediction is extraordinarily beneficial for mining applications,autonomous vehicle localization/tracking,and location-based services.This research work demonstrates both EKF-based and SPKF-based sensor fusion to provide an appropriate estimation.展开更多
文摘It is widely acknowledged that navigation is a significant source of between sites.The Global Positioning System(GPS)has numerous navi-gational advancements,and hence it is used widely.GPS navigation can be compromised at any level between position,location,and estimation,to the detriment of the user.Consequently,a navigation system requires the precise location and underpinning tracking of an object without signal loss.The objective of a hybrid environment prediction system is to foresee the location of the user and their territory by employing a variety of sensors for position estimation and monitoring navigation.This article presents a state estimation of the relative position for indoor and outdoor activity solved with a state estimation algorithm utilizing Kalman filter.Also,a comparative study of variants of the Kalman filter,where linearizing current mean and covariance with nonlinear state estimation as an approach of Extended Kalman Filter(EFK)is applied to the collected data.The third comparative aspect uses probability distribution for the selected points with a Sigma Point Kalman Filter(SPKF)for evaluating an accelerometer,gyroscope,and GPS data in hybrid environments for various activities for different data collection scenar-ios from users.The findings of the presented model demonstrate the robust performance of all forms of the Kalman filter algorithm for diverse user-performed activities in totally contaminated indoor and outdoor environ-ments.Experimental findings with various patterns and data,conducted by different subjects using multiple modes of navigation,show that the approach can indeed lead to the intelligent development of sensor-based navigation and monitoring.State estimation and prediction is extraordinarily beneficial for mining applications,autonomous vehicle localization/tracking,and location-based services.This research work demonstrates both EKF-based and SPKF-based sensor fusion to provide an appropriate estimation.
