以電腦視覺搭配改良式A*搜尋法路徑搜尋做室外自動車導航之研究

A Study on Outdoor Guidance of Autonomous Land Vehicles by Computer Vision Based on Improved A* Search

陳永德(Yung-Der Chen)

指導教授 ： 駱榮欽

國立臺北科技大學/電資學院/電腦與通訊研究所/碩士(2007年)

https://doi.org/10.6841/NTUT.2007.00340

摘要

本論文中，我們發展一套利用雙眼立體電腦視覺搭配人工智慧策略的室外導航車系統，可讓自動車導航於室外的複雜環境中。這系統利用兩個攝影機所拍攝的影像來計算場景的3D資訊，藉由得到的3D資訊可以分類出車子前方可行走的區域與障礙物的區域，配合人工智慧策略，使得車子可以在室外的環境中導航。
在導航方面，結合蟻群演算法與 搜尋法規劃出最佳路徑。首先，從攝影機獲得車子前方所要導航的地圖，經由蟻群演算法將可從地圖上得到一種新的信息(費洛蒙)，利用所得到的信息來改進 搜尋法的參數，得到新的節點評估值，依據在搜尋路徑上節點的評估值，得到一條由起點到終點的最佳路徑。
得到道路資訊後，我們利用電子羅盤搭配改良式 搜尋法規劃路徑，使自動車可以安全繞過障礙物並在規劃的路徑下前進。透過實驗測試此系統可以實際在室外道路上行走，證明所提方法之可行性。

關鍵字

立體視覺 ； 自動車 ； 人工智慧 ； 蟻群演算法 ； 路徑規劃

並列摘要

In this paper, we develop the system of autonomous land vehicles (ALV) by computer vision based on artificial intelligent (AI) policy. It makes ALV adapted to the complex environment. This system uses two cameras to capture two images from the front of the vehicle and calculate the 3D information of environment. Though the obtained 3D information, we can recognize region of road and region of obstacle in front of the vehicle. Using navigation policy based on AI to make vehicle navigate in outdoor environment.
In path planning, combining ant colony optimization (ACO) and search program the best path. First, we obtain the navigation map of the front of the vehicle from charge-coupled device (CCD). After path planning of ACO, we can get a kind of novel trail information (pheromone). Using trail information improves the parameters of search to get novel estimation of node. According to estimation of nodes on searching path, we obtain a path from origin to target.
After obtaining road information, the path of navigation is searched by E-compass and improved search. We make ALV avoid the obstacle safely and run toward the goal in an appropriate path. The ALV system has been performed in the outdoor to demonstrate the effectiveness of the presented method.

並列關鍵字

Stereo Vision ； Autonomous Land Vehicle ； Artificial Intelligence ； Ant Colony Optimization ； Path Planning

參考文獻

[3] 葉哲昌，利用鬆弛相關性改進3D對應點匹配與重建應用於室外自動導航車，碩士論文，國立臺北科技大學自動化科技研究所，台北，2006。

[5] 陳冠州，以型態學搭配分水嶺為基礎偵測室外路面及障礙物作自動車導航，碩士論文，國立台北科技大學機電整合研究所，台北，2005。

[10] Chris Harris and Mike Stephens, “A Combed Corner And Edge Detector”, The Plessey Computer plc, 1988, pp 147-151.

[11] Zhengyou Zhang, “A Robust Technique for Matching Two Uncalibrated Images Through the Recovery of the Unknown Epipolar Geometry”, Institut National De Recherche En Informatique Et En Automatique, 1994, pp 87-119.

[12] Beatriz A. Garro, Humberto Sossa and Roberto A. Vázquez., “Path Planning Optimization Using Bio-Inspirited Algorithms”, IEEE Proceedings of the Fifth Mexican International Conference on Artificial Intelligence (MICAI'06), 2006.