基于人工生命算法的并联机器人最优轨迹规划

Optimal trajectory planning for parallel robots based on artificial life algorithm

基于机器人直纹面概念和人工生命算法,提出一种并联机器人位姿轨迹最优规划方法。应用计算几何中的三维直纹面生成原理,对机器人末端执行器的位置和姿态进行统一描述。考虑到机器人姿态直纹面面积及其变化率能够反映和评价机器人的运动学和动力学性能,通过求解等效角位移矢量在空间的运动轨迹形成的三维直纹曲面面积及其变化率,并将其作为泛函的泛函极值,同时考虑运动时机器人的灵活度,建立机器人位置和姿态轨迹优化的数学模型。采用人工生命优化算法对代表并联机器人位姿轨迹的高阶参数化空间曲线的参数进行优选,通过优化轨迹直纹曲面面积及其变化率和机器人的灵巧度,使并联机器人具有良好的运动学和动力学性能。最后以一三自由度球面并联机器人轨迹规划实例,验证所提出方法的可行性。

Based on the robot pose ruled surface concept and artificial life algorithm, an optimal position and orientation trajectory planning approach for parallel robots was presented. The unified description of the parallel robots end-effector positions and orientations was based on the three-dimension ruled surfaces generating method in computational geometry. The area of robot pose ruled surface and its change ratios could indicate or justify its kinematics and dynamics performances. Thus, by determining the maximum or minimum value of the functional of the area and its change rate of the three-dimension ruled surface generated by the space motion locus of the angular displacement vector and the robot dexterity, the optimization model of pose trajectory planning was established. An enhanced artificial life algorithm was used to select the parameters of high-order parametric curves as position and orientation trajectory so that the pose ruled surface and its area as well as area change ratios and the robot dexterity could be optimized to achieve good kinematics and dynamics performances. Finally, a case of trajectory planning for 3-Degree- of-Freedom （DOF） spherical parallel robot was provided to verify the feasibility of the proposed approach.