中文摘要：

为解决在速度层上无穷范数最小化模型中可能出现的不连续点问题，提出一种基于双判据方法的二次型优化模型．冗余机器人运动规划与控制模型可以统一各种关节物理极限，如关节变量极限与关节速度极限．同时该模型又可以最终转化为一个标准的二次规划问题．为了实时求解该二次规划问题，提出一种基于线性变分不等式（LVI）的原对偶神经网络．该神经网络作为实时求解器具有简单的分段线性结构和较高的计算效率．计算机对PUMA560机器手臂的模拟仿真表明，该方案具有灵活性和有效性．

英文摘要：

To diminish discontinuity-points arising in the infinity-norm velocity-minimization scheme, a bicriteria velocity-minimization scheme was developed. The kinematic-control scheme of redundant manipulators could incorporate joint physical limits, such as, joint limits and joint velocity limits simultaneously. Moreover, the method could be reformulated as a quadratic programming （QP） problem. As a real-time QP solver, a LVI-based primal-dual neural network was developed with a simple piecewise-linear structure and higher computational efficiency. Computer simulations performed based on PUMA560 manipulator illustrate the validity and advantages of the proposed neural control scheme for redundant robots.