CSG模型指定碰撞面五軸加工碰撞偵測之研究

陳宥瑋(Yu-Wei Chen)

指導教授 ： 林榮信

國立中正大學/工學院/機械工程學系/碩士(2014年)

摘要

本文目的在於發展一套應用於五軸CNC工具機之即時防碰撞偵測技術，在現今講求快速與高精密度的時代，五軸CNC工具機已是目前加工技術的主流趨勢，由於全自動化的緣故，若是在加工的過程中有錯誤的指令或操作，很容易發生機台碰撞，會拖延產品的製造行程、增加維修成本以及危害到操作人員的安全。
本文使用（Constructive Solid Geometry）實體幾何模型描述方式建構機台構型，主要目的是為了改善使用網格建立機台結構模型，造成資料過多的問題以及機台透過G-Code指令移動時機台有些幾何面不需做碰撞判斷等因素，影響碰撞偵測的效率，亦可改善由網格組成所造成圓柱、球體等等擁有曲面圖形物件之不準確問題。為了業界實務上的需求，程式多增加了機台的碰撞偵測裕度，再透過分析工具機加工碼G-Code移動方向，使用分離軸定理演算法之指定碰撞面計算各軸間是否發生碰撞，最後程式配合OpenGL將畫面呈現出來做電腦動畫模擬驗證。
本文最後偵測得到五軸工具機使用CSG模型指定碰撞面碰撞平均時間(含碰撞及未碰撞)為10.59 μs，相比CSG模型未使用指定碰撞面改善53.8%，並且在碰撞偵測的平均時間遠小於CNC工具機取樣時間0.5~1 ms情況之下，可應用於機台的即時碰撞偵測。

關鍵字

碰撞偵測 ； 五軸加工

並列摘要

The main purpose of this research is the development of real-time five-axis CNC machine tools collision detection technology. Nowadays, emphasis on precision and productivity, five-axis CNC machine tools has been the mainstream processing technology trends. Due to the automated's sake, if there is an error in the instructions or operations in the course of processing, the machine is prone to collision or interference, will travel products and manufacturing delays endanger the safety of operating personnel.
In this research, the machine model is described by the way of Constructive Solid Geometry, which builds each component of five-axis CNC machine tools. Its main purpose is to improve the use of the machine to establish the structure of STL model, resulting in the problem of excessive data and the G-Code machine is moved through the back of the object do not need to determine the collision influence the efficiency of the collision detection,also improve the composition caused by the STL cylinder, the sphere, and so does not have a surface accuracy of graphic objects. To demand on industry practice, the program adds to predict the value of machine, and then through the analysis tool machining G-Code movement direction. Finally, through an entity separation axis theorem testing as a basis for collision detection and then show the simulation results with OpenGL.
Finally, under the five-axis machine tools to get use to detect the collision plane crash specified average time 10.59 μs, compared to 53.8% did not improve with the specified impact surface, and in the mean time collision detection is much smaller than the sampling time of 0.5~1ms , immediate practical machine real-time collision detection.

並列關鍵字

Collision Detection

參考文獻

【1】 Brian Mirtich, “V-Clip: Fast and Robust Polyhedral Collision Detection,” Mitsubishi Electric Research Laboratories(MERL), Inc, 1997.

【2】 Chang, Jung-Woo, Wenping Wang, and Myung-Soo Kim. "Efficient collision detection using a dual OBB-sphere bounding volume hierarchy." Computer-Aided Design 42.1 (2010): 50-57.

【4】 P. JimeHnez, F. Thomas, C. Torras, “3D collision detection: a survey”, Computers & Graphics 25 (2001) 269-285,2001.

【6】 Kuan-Chen Lai, Shih-Chung Kang, “Collision detection strategies for virtual construction simulation”, Automation in Construction 18 (2009) 724–736,2009.

【7】 Lee, R-S., and C-H. She. "Developing a postprocessor for three types of five-axis machine tools." The International Journal of Advanced Manufacturing Technology 13.9 (1997): 658-665.