車輛輪胎與路面摩擦力之即時估測

Abstract

隨著車用電子的技術蓬勃發展，人們對車輛的要求也日益增高，不論是在駕駛的舒適性上，或是在目前所關注的環保節能上，這些條件已經成為一般民眾購車的先決條件。在車輛行駛中往往會遇到各種不同的路面狀況，而路面狀況會直接影響車輛輪胎與地面間的受力，所以可以藉由輪胎與地面間受力情況的不同，判斷目前車輛行駛路面的狀況，也就是說欲得知路面狀況的不同，可以藉由估測車輛輪胎與地面間的摩擦力不同來做判斷，所以本研究發展了一套只需使用容易測量到的車輛動態再搭配車輛的動態模型來估測摩擦力的演算法，實驗中也建構了一模擬車輛輪胎與路面間的實驗平台來驗證摩擦力估測的正確性。 本論文建立了一套摩擦力估測演算法，是利用輪胎的力矩平衡來求得縱向的摩擦力，而側向的摩擦力是使用簡化車輛模型與腳踏車模型互相切換，並引入摩擦力圓的概念來做估測，另外也是用了期望值最大化演算法來估測車輛的不確定參數，然後模擬各種駕駛狀況與路面下摩擦力估測與不確定參數估測的結果；本實驗中也架設了實驗平台，在利用此實驗平台進行的縱向力估測與結果分析。

Due to the advances in modern automotive technologies, requirements for driving comfort and being friendly to the environment are ever demanding. Vehicles run on roads with different surface conditions which make tire-road forces, and in turn the vehicle dynamics vary significantly. This study develops an algorithm that uses measurable signals and simple vehicle dynamic models to estimate tire-road friction forces. Then an experimental platform is established to simulate a tire rolling on a road surface such that the validity of the proposed algorithm can be verified experimentally. This thesis establishes a friction force estimation algorithm which uses torque balance equations to obtain longitude friction forces. Lateral friction forces are estimated by introducing the concept of friction circles and switching between a simplified four-wheel model and the bicycle model. The uncertain parameters of the vehicle are also estimated on-line by applying the expectation-maximization algorithm. Finally, simulations are conducted under various driving and road conditions to evaluate the performance of parameter and tire-road force estimation. This study also sets up a quarter-car experimental platform to estimate longitudinal forces. Experimental results exhibit a linear relation between the longitudinal tire force and the slip ratio whenever the latter is low.