Research

1. 先進駕駛輔助系統/自動駕駛決策與控制 Advanced Driver Assistance System/Decision and Control of Automated Driving

先進駕駛輔助系統利用感測器、致動器與電子控制單元，以提示、警示與控制介入方式，協助駕駛人及時應變各種駕駛狀況，並提昇行車安全性，減少因判斷錯誤或技術不足造成的交通事故發生。Level 4自駕則在受限的操作設計範圍內，可自動決策與控制車輛，不需要駕駛者介入。

Advanced driver assistance system (ADAS) utilizes sensors, actuators, and electric control units to assist the driver to deal with all the driving conditions in time via information, warning, and control intervention. Thus enhance driving safety and reduce traffic accidents due to misjudgments or insufficient driving skills. Automated Driving can make decisions and control vehicles for Level 4 automation without the driver's intervention within the constrained operational design domain.

2. 車輛動態與控制 Vehicle Dynamics and Control

車輛控制常用的車輛動態包含縱向動態、側向動態、垂直動態與最重要的輪胎動態，因為所有的車輛動態都是由輪胎力與力矩所激發產生。必須對車輛動態有一定程度的瞭解，才能設計出好的車輛控制系統。

Commonly used vehicle dynamics includes longitudinal dynamics, lateral dynamics, vertical dynamics, and the most important tire dynamics. All the vehicle dynamics are excited via tire forces and moments. It is necessary to understand the vehicle dynamics to a certain extent to design good vehicle control systems.

3. 油電複合動力/電動車能量管理控制系統 Power Management System of Hybrid Electric Powertrain/Electric Vehicle

採用內燃機與電動馬達作為動力系統的油電複合電動車(Hybrid Electric Vehicle, HEV)，為目前解決純電動車續航力不足的最佳方案，經由能量管理控制策略控制內燃機與馬達/發電機的動力分配比例，可大幅提升燃油經濟性與減少排放廢氣。

The powertrain of hybrid electric vehicles (HEV) which consists of an internal combustion engine and electric motors is the optimal solution for the insufficient cruising distance of pure electric vehicles. Power management control strategies can control the power split ratio between the internal combustion engine and motor/generator to significantly increase the fuel economy and reduce exhaust emissions.

4. 電池管理系統 Battery Management System

常見的電池管理系統(Battery Management System, BMS)是採用庫倫積分法與開路電壓校正來估測電池的電量狀態(State of Charge, SOC)，但是很容易因為電池健康狀態(State of Health, SOH)老化而使得SOC估測逐漸失準，利用雙卡爾曼濾波器配合老化電池開迴路資料估測，可大幅提升SOC與SOH估測的準確性。

A commonly used battery management systems (BMS) often employs coulomb counting method and open circuit voltage to estimate the state of charge (SOC)　of the battery. However, the estimation of SOC deviates from the actual value due to the deteriorated state of health (SOH) of an aging battery. The estimation accuracies of SOC and SOH can be significantly enhanced using a dual Kalman filter and the parameter estimation of an aging battery.