Indoor/Outdoor Pedestrian Dead Reckoning Method Using IMU

• Main Authors: Yan-Ju Chen, 陳彥茹
• Other Authors: Sheng-Wen Shih
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/20126446592967503704

碩士 === 國立暨南國際大學 === 資訊工程學系 === 102 === In this thesis, we use an inertial measurement unit (IMU) fabricated with the micro electro mechanical (MEMS) technique to study the pedestrian dead reckoning (PDR) problem. The IMU is composed of a 3-axis accelerometer, a 3-axis rate-gyro, and a 3-axis magnetometer, and is lightweight, miniature, and low cost due to the MEMS technique. In this work, the IMU is attached on the surface of the shoe tongue for trajectory tracking. The proposed method uses the particle filter, 2-D indoor/outdoor maps, and Google satellite images for indoor/outdoor PDR. In each indoor floor map, we use different colors to mark the walls, pillars, passageway, stairs, entrances and exits in order to eliminate unreasonable trajectories such as walking through a wall or a railing. Likewise, in the outdoor maps, roads, pathways, walls and the entrances of buildings are all marked clearly to facilitate accurate PDR computation. Also, we incorporate the global positioning system measurement to alleviate the trajectory drifting effectively. The color marks in the indoor/outdoor maps can also help us to determine the correct timing for changing a map to keep tracking a pedestrian walking in a mixed indoor/outdoor path. Experimental results show that the proposed method can provide accurate trajectory estimates in mixed indoor/outdoor paths.