| Summary: | 碩士 === 元智大學 === 通訊工程學系 === 106 === In recent years, the development of the Internet and smart phones has gradually matured, and brought Internet of Things and Artificial Intelligence Technology advance. Many problems can be solved through the combination of big data andmachine learning. For environments that are too complex or have security concernscan effectively be managed and prevented, and can make people’s lives moreconvenient and efficient.
In this paper, we will make a further discussion on how to use inertial measurement unit and machine learning technique to pedestrian behavior classification and tracking. The data is captured by the sensor when the user wears the inertial measurement unit sensor on the back of the right foot. We classify data as pedestrian behavior through machine learning techniques. This study will divide datas into two classes, moving and stop, and perform pedestrian behavior tracking on moving datas. In the tracking, this paper will calculate the amount of pedestrian movement, the number of steps, the length of the steps, whether there is a turn,
angle of turning, and number of turning times. On through these methods, we can achieve relative positioning.
|
|---|
