冗余仿人臂避关节物理约束的一种逆运动学问题求解方法

A Joint Physical Constraints Avoidance Method for Inverse Kinematics Problem of Redundant Humanoid Manipulator

冗余仿人臂有无穷多组逆运动学解满足末端位姿要求,但由于其连杆机构与形状的设计仿人臂每个自由度都有关节位置物理约束。为使所求取的逆运动学解最大化地远离仿人臂关节限位,提出冗余仿人臂(8-DOF)一种几何-数值迭代相结合求解逆运动学问题方法,该方法在计算出仿人臂逆运动学解几何表达式基础上,以仿人臂的'远离限位度'指标构建适应度函数并以此为寻优目标,通过粒子群优化方法(Particle swarm optimization,PSO)搜索冗余关节角最优组合,并获得满足仿人臂末端位姿要求的最优逆运动学解。该方法不仅解决了冗余仿人臂逆运动学多解问题,而且所求关节角不存在理论误差、求解速度快及所求关节角远离关节位置限位裕量大的优点,仿真试验验证了该方法有效性。

There are innumerable inverse kinematics solutions that can meet the demand of end-effector position and orientation for the redundant humanoid manipulator. Also each manipulator joint has its physical constraints for the connecting rod structure and figuration design. In order to make the inverse kinematics solution away from the joint position limitation maximization, an algorithm that combines geometry method and numerical iterative method is proposed to solve the inverse kinematics problem of redundant humanoid manipulator（8-DOF）. Based on the derived geometric formula of the inverse kinematics problem, the ＂Away Limitation Level＂ of the humanoid manipulator is adopted as fitness function and optimization goal, and then particle swarm optimization（PSO） is used to search for the optimal combination of the redundant joint, and the optimal inverse kinematics solution can be obtained in this method. The proposed method can not only solve the multiple inverse kinematics solutions problem of the redundant humanoid manipulator, but also have some advantages, i.e., the solution no theoretical error, fast solving speed and the large margin away from the joint position limitation. Numerical simulation results demonstrate the effectiveness of this algorithm.