基于遗传算法的超二次曲面间碰撞检测算法研究  

Research on the Collision Detection between Two Objects Modeled by Superquadrics Based on Genetic Algorithm

作　　者：钱东海[1] 袁斌[1] 

机构地区：[1]上海大学精机系,上海200072

出　　处：《组合机床与自动化加工技术》2010年第8期45-47,51,共4页Modular Machine Tool & Automatic Manufacturing Technique

摘　　要：文章对机器人的手臂采用超二次曲面建模,将机器人碰撞检测问题归结为两超二次曲面之间最短距离求解,该问题本质为一带约束条件的非线性优化问题。将该问题转化为非线性优化问题后,利用遗传算法对该问题进行求解。实例表明,遗传算法对该问题的求解克服了传统算法求解中的初始点难于选择、局部最优点等问题,并提高了算法的收敛速度。Superquadrics are used for modeling the Robot arm in the paper,and the robot arm＇s collision detection is reduced to calculate the shortest distance between two Superquadrics.The essence of the problem is non-linear constrained optimization.Genetic algorithm is used for the calculating,after it has been transformed to a problem of nonlinear constrained optimization.Real examples show that the algorithm proposed in this paper can dealing with the difficulty of choosing initial point and the problem of local best point in traditional algorithms,and it＇s also help to improve the convergence speed.

关 键 词：机器人手臂 超二次曲面 遗传算法 碰撞检测 

分 类 号：TP241.2[自动化与计算机技术—检测技术与自动化装置]