Integrating Low Cost IMU with Building Heading In Indoor Pedestrian Navigation

Abstract

This paper proposes an integration of ‘building heading’ information with ZUPT in a Kalman filter, using a shoe mounted IMU approach. This is done to reduce heading drift error, which remains a major problem in a standalone shoe mounted pedestrian navigation system. The standalone system used in this paper consists of only single low cost MEMS IMU that contains 3-axis accelerometers and gyros. Several trials represented by regular and irregular walking trials were undertaken inside typical public buildings. The results were then compared with HSGPS solution and IMU+ZUPT only solution. Based on these trials, an average return position error of below 5 m was consistently achieved for an average time of 24 minutes – at times as long as 40 minutes - using only a low cost MEMS IMU.