用於路面異常偵測之智慧型手機探偵車系統

Abstract

道路品質的優劣與民眾乘載交通工具的舒適、安全與速度有著重要的影響，對於騎乘自行車與機車的民眾而言，更是影響甚多。然而對於道路品質優劣依據卻沒有一個考量到科學性、便利性、以及經濟性的方法。因此本論文提出以智慧型手機探偵車系統用於偵測路面異常點，並針對路面異常點進行分級，以期望能以較客觀的方式鑑定道路品質的優劣。一般車輛行經異常顛簸路段，例如：人孔蓋、低窪坑洞、減速丘、伸縮縫時，車輛本身會受到垂直方向的震動，因此車輛本身所受到的震動影響可作為判斷車輛是否行經顛簸路段。此外，我們考慮到影響震動幅度的因素，例如裝置置放於不同車架、不同車輛的懸吊系統、不同感測器的誤差和不同的車速等，因此單純震動的幅度並不足夠做為路面品質的客觀指標。我們提出一個異常偵測演算法，利用車輛行駛於路面異常點時震動標準差除以行駛一般正常道路時震動標準差的參考值，作為異常點大小分級。結果顯示，我們的演算法能夠消除影響震動幅度的因素，並針對異常點大小分級。

Road quality is an important index for modern traffic networks. It affects not only traffic flows but also safety and comfortableness of passengers. However, there are no scientific, convenient and economical methods to evaluate the road quality. In this work, we propose a smartphone-based probe car system that utilizes mobile sensing to pervasively detect road abnormality such as potholes, speed bumps, expansion joints, manhole covers, etc. Since the vibration caused by an abnormality mainly affects in the vertical direction of a vehicle, in our system, the road abnormalities are detected by vibration measured in embedded inertial measurement units (IMUs) of onboard smartphones. However, there are many factors such as orientation of smartphones, phone racks, sensor chips of smartphones, types of vehicles, driving speed, etc., which affect the vibration sensed by the smartphones. Therefore, several mechanisms are proposed to overcome these challenges including a vertical component extraction algorithm and an abnormality detection algorithm by the vibration characteristic (the standard deviation of vertical vibration) and the grade of the abnormality is obtained. The result indicates that our algorithm can eliminate these factors and grade the road abnormality.