以車載感性感測元件與全球定位系統量測道路高度與坡度之方法

Abstract

根據能源消耗的研究，汽車的油耗與道路坡度密切相關；對於車輛自動控制來說，道路坡度在保持車輛姿態穩定的演算法中是一個重要的參數；此外，對於車輛導航機，道路坡度和高度可以用來輔助判斷上下交流道。 在本文中，我們發展兩種方法推算道路坡度和高度資訊。第一種方法是基於全球定位系統的高度推算方法：收集車輛行經道路的三維定位資料，且對地球平面進行格子化的分割，並以單一格子的分割區塊為主體，使用數學統計的概念，平均同一格子的高度資料，消除其誤差。實驗結果顯示平均大量資料，確實可以消除定位誤差，但資料量不足的話，消除誤差的成效有限；第二種方法是基於多感測器的即時道路坡度推算方法：使用了車載慣性感測元件和全球位系統直接感測車輛的狀態，並用以推算車輛的速度、仰角變化和仰角，最後使用卡爾曼濾波器整合並推算道路的坡度和高度資訊。實驗結果顯示，推算出的高度變化趨勢與實際環境相似，推算出的坡度資料平均誤差約1.877角度。

The gasoline consumption of vehicle is closely related to the slope of roads. In automotive control, the road slope is an important factor fo to keep the vehicle stable. In addition, the road slope and elevation can help on-board navigation system to judge vehicle position on system interchange of road. In this paper, we develop two methods to estimate the road slope and elevation. The first method is a GPS-based elevation estimate method. A lot of three-dimensional positioning data of vehicles collected from GPS. The earth plane cut into grids for clustering the position data, and averaging the positioning data of the same grid . The experiment results verify that it can eliminate the positioning error by averaging large amounts of data, but the effect is limited when the amount of data is insufficient; The second method is a real-time sensor-based slope estimate method . The vehicular Inertial Measurement Unit(IMU) and GPS sensing the dynamic data of vehicle on the road and estimate the speed, variation and value of elevation angle information of vehicles. Finally, a Kalman Filter is implemented to integrate the information and estimate road slope and elevation. The estimated elevation is similar to the true elevation of road , and the average error of estimated slope about 1.877 degree.