- 學位論文
整合適應性巡航與主動超車之主動安全控制策略研究
The Research for Active Safety Control Strategy Integrating Adaptive Cruise Control and Auto-Overtaking
摘要
本研究目標建立一套障礙應對策略，並應用於自主駕駛車。利用規則法，我們建立一套完整的情境判斷、決策流程及控制方法，以避免與其他障礙車輛碰撞。本研究以一組既有的主動超車方法為基礎，加入適應性巡航方法。本策略能利用感測資訊，判斷合適超車時機。透過模糊邏輯產生行車目標速度及安全距離，並透過PI線性控制車輛電門與煞車。本巡航方法可協助駕駛在超車時機出現前，與前車保持安全距離並穩定跟隨。巡航方法可協助車輛最高從100 kph穩定降速，直到與前方車輛等速。本研究能滿足乘客速度需求並保障乘客安全。符合ISO加速度規範，滿足乘車舒適性。 本研究利用MATLAB/Simulink編寫控制策略，並將各功能模組化。結合車輛與道路模型進行模型迴路測試(Model in the Loop, MiL)，初步驗證方法演算法並調整控制參數。我們利用實驗數據，確保模型正確性與動態穩定。MiL測試後，進一步將策略燒錄至可程式控制器dSPACE Micro-AutoBox(MABX)。結合實時模擬工具，進行控制器迴路測試(Controller in the Loop, CiL)，驗證策略在實時環境的可行性並調整程式。此外我們透過實車驗證適應性巡航方法。實時環境下本適應性巡航方法可協助駕駛穩定跟隨前方車輛，且兩車速度誤差在1 kph。此外等待超車時機的過程，可增加駕駛等待超車時機的時間。開啟巡航方法後，最多可增加21秒的反應時間，避免駕駛與來車擦撞。
並列摘要
The purpose of this research is to establish a collision avoidance strategy for a self-driving car. We build a strategy including situation judgement, decision-making logics and control methods by Rule-Based method. This strategy is based on a developed constant-speed autonomous overtaking strategy. In this research, we integrate it with Adaptive Cruise Control(ACC)method. This active-safety control strategy can utilize sensor information, to decide the timing to overtake. During driving car cruise, ACC method will calculate drivig target speed and safety distance by fuzzy logic machine. Then control throttle and brake by PI controller, to make sure that driving stably follows front vehicle from a safety distance at the same speed. Besides, the ACC method can deaccelerate driving car highest from 100 kph. This strategy can satisfy passengers’ need for speed and protect passengers. Furthermore, we follow the ISO standards to ensure passengers comfort. The control strategy is estabilished by MATLAB/Simulink, and modularize all the functions. And combine the program with the dynamic vehicles and road models. Then execute the Model-in-the-Loop simulation to verify the algorithm, and adjust parameter. We’ve verified the accuracy and dynamic characteristic of the vehicle model through experiment. After MiL, we compile the program into programmable controller, dSPACE “Micro-AutoBox”. Then we execute Controller-in-the-Loop simulation with real-time(RT)simulation tool. To ensure the strategy workable in real-time environment. Futhermore, we’ve execute the ACC strategy on an autonomous electric van. In RT environment, the ACC strategy can ensure the error of speed between two vehicles about 1 kph. When ACC strategy is active, this active safety control stratey can extend the reaction time most to 21 seconds before overtaking, and avoid collision with coming vehecles.
參考文獻
延伸閱讀
- 盧玟翰（2015）。整合避障與車輛穩定之主動安全控制策略研究〔碩士論文，國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU.2015.01100
- 劉書廷（2021）。基於參數識別模型之適應性巡航控制與車身動態穩定輔助系統開發實務驗證〔碩士論文，國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU202102128
- Lin, T. W. (2009). 大客車適應性巡航控制與防撞警示系統設計及駕駛者模型建立 [doctoral dissertation, National Tsing Hua University]. Airiti Library. https://doi.org/10.6843/NTHU.2009.00014
- 沈家屏（2020）。應用類主動式懸吊系統控制策略於行車舒適度之研究〔碩士論文，逢甲大學〕。華藝線上圖書館。https://doi.org/10.6341/fcu.M0707303
- 王晉元、盧宗成、張惠汶、孟億里、錢建中（2003）。The Research of the Development Strategy of Advanced Safety Vehicle。運輸學刊，15(1)，79-98。https://www.airitilibrary.com/Article/Detail?DocID=10272275-200303-202208230016-202208230016-79-98