「交通事故」目前已躍居國人十大死因之ㄧ,其中十字路口的肇事率又佔有相當高的比率。究其原因,不外乎是路口車輛的交通動線複雜,或為了搶黃燈而高速行駛,或是駕駛人及行人的注意力渙散。有鑑於此,欲發展一套安全機制,依據人-車之相對距離及相對速度制定其碰撞風險的範圍,給予十字路口的車輛與行人風險警示,藉此提高其注意力,以期能防止交通事故的發生。 本篇論文中,主要使用的設備為具有GPS功能之PDA手持裝置,考慮在未來高科技生活環境中,人手一機且一機多用的情形將會是常態,在使用手機通話的同時,以無線網路傳遞自我定位資訊到伺服端,配合最小平方法(Least-Square Method)以及地圖比對(Map Matching)修正行人行走的軌跡;同時,裝設有車上單元(On Board Unit, OBU)的車輛也由無線網路回傳定位資訊到伺服端,再於伺服端利用模糊決策(Fuzzy Decision)計算出人-車碰撞風險的程度,以此形成一套安全警示系統,針對交通事故易肇發地點-十字路口進行碰撞風險評估,依不同的風險程度將警示顯示於手機上,給予行人不同程度的警告,以達到降低事故發生率的目的。
Now traffic accident has been ranked the top ten causes of death, of which the accident rate at the crossroad is quite high. It is because of the complicated traffic circulation at the crossroad including racing through yellow lights or inattentiveness of drivers and pedestrians. Therefore, a safety mechanism is developed to warn vehicles and pedestrians at the crossroad and avoid traffic accidents by means of formulating the distance of collision risk based on the relative distance and relative speed between the pedestrian and the vehicle. In this thesis, the aim of this thesis is to develop the system for contributing to the prevention of pedestrian-to-vehicle accidents. With the widespread use of cellular phones and car navigation systems, a pedestrian-to-vehicle communication system is presented by using a cellular phone and a car navigation system equipped with GPS and wireless communication function. The location information of pedestrians and vehicles can be delivered to the traffic monitoring center. By employing Least-square method and map matching technique, the errors from poor satellite availability can mitigated. The collision risk for pedestrian is calculated based on fuzzy inference scheme and will be given to the cellular phone in the case of high collision risks.