关键词: 干涉检测, 时间间隔放大因子, 模糊控制

Abstract: We analyze the research state of collision detection and discuss the relationship of reliability, efficiency, and cost of algorithms. Referring to time slicing method, we present an algorithm using fuzzy theory to control the time-interval-enlargement factor. In order to control the time-interval-enlargement factor reasonably, we analyze the complexity of both geometry and moving-ways, and consider the possibility and risk of collision. To get more efficient, we use suitable approximate object in different situations. It has been proved to be an intelligent, efficient, fast and accurate algorithm for collision detection between general models in dynamic environments. At the same time, it has been implemented as part of NC-instruction-driven simulation systems for NC tube bender. In practice, it can accurately detect all the collision and meet the requirement of real-time.

Key words: collision detection, time-interval-enlargement factor, fuzzy control