Integration of Multiple Low-Cost Wearable Inertial/Magnetic Sensors and Kinematics of Lower Limbs for Improving Pedestrian Navigation Systems
Date
2019-07-25
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis presents a work in pedestrian navigation that utilizes multiple low-cost wearable inertial/magnetic sensors and kinematics of lower limbs for improving the positioning performance. A multi-wearable sensor system is developed for this research in order to record the walking motion using multiple MPU-6050 and MPU-9250 sensors. The focus of this research is to investigate the feasibility of using multiple low-cost sensors distributed on lower limb segments as a pedestrian navigation system. The proposed method uses seven wearable sensor modules distributed on pelvis, thighs and shanks. A skeletal model consists of five limb segments is used to model the forward kinematics of lower limbs. Different attitude estimator algorithms are tested and compared. The step size and heading are computed using the forward kinematics. To evaluate the proposed navigation method, two experiments are conducted. The first experiment is a treadmill walk to evaluate the accuracy of the estimated segment orientation angles and step sizes. The second experiment involves turning motion where a test subject walks around a rectangular path. The results show that the use of a wearable multi-sensor system can provide a pedestrian navigation solution with error comparable to the solution computed using a single higher-cost sensor.
Description
Keywords
inertial navigation, pedestrian navigation system, pedestrian dead-reckoning, wearable sensor, wearable multi-sensor system, low-cost sensor, step length, stride length, orientation estimation, inertial/magnetic sensor, kinematics of lower limbs, forward kinematics, step detection, gait event detection
Citation
Tjhai, C. (2019). Integration of Multiple Low-Cost Wearable Inertial/Magnetic Sensors and Kinematics of Lower Limbs for Improving Pedestrian Navigation Systems (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca.