本研究利用灰色理論建立一套智慧型撞球機器人之防守策略機制，將母球及目標子球撞至對手不易進球之相對位置，即撞球術語中之「安全球」，使撞球機器人能反「守」為「攻」，因而贏得球局。 首先以CCD擷取球檯影像，經由影像辨識，搜尋並計算球檯上各球及球袋中心之位置，再以影像處理技術與所設計的VB介面程式，預測出母球經不同撞擊方式碰撞目標球後的幾何關係，並計算母球與撞擊點距離、目標球與球袋中心距離、擊球角度等幾何資料，作為灰色理論之決策條件，而且加入障礙球之偵測，特別針對具有障礙球的情況以不同之防守策略來增加對手進球難度，然後決定最不利於對手進球之預測位置，再以此位置反向推論撞球機器人之打擊方式，驅動撞球機器人打擊，完成防守動作。經實驗結果得知，撞擊後母球與目標球確實能達到不利對手進球之相對位置。且當有障礙球時，也能利用障礙球來增加對手進球的難度，成功驗證灰色理論在防守決策之應用，提高撞球機器人於球局判斷之能力，朝向贏球之目標。 The objective of this thesis is to develop a defensive strategy of the intelligentbilliard robot by using the grey system theory. The main purpose of this defensivestrategy is to make a “safety” play during the pool game. The safety play is thedefensive positioning of the balls so as to minimize the opponent’s chance to score orto make a foul stroke. In this case, it will be easier for us to win the ball gamewhenever opponents have to face to a safety play.In the research, the centers of balls are captured and caculated by a CCD and theimage processing technique in a PC. A developed software (VB) is applied to predictthe geornetrical positions of the cue ball and object balls after collision. Then, theshooting angle, the distance between the cue ball and the designate object ball, and thedistance between the designate object ball and corresponding pocket are alsodetermined by the developed software.These three parameters and the information of block ball are the lets fourfactors of this defensive strategy. Then, the billiard robot will execute the hittingcommand to make a safety play based on this algorithm. Finally, the experimentalresults show that the developed defensive strategy makes the billiard robot meake asafety play and be easier to win the ball game successfully.
