基于SOA-Newton迭代的六自由度平台正解算法

A Hybrid Forward Kinematic Solution of 6-DOF Platform Based on SOA-Newton Iterative Algorithm

运动学正解是实现六自由度平台高精度控制和三维可视化仿真的基础,但是传统方法存在着求解难度大、精度低的缺陷。针对Newton-Raphson迭代法求解时依赖迭代初值的问题进行了研究,提出了一种基于SOA-Newton迭代的六自由度平台正解混合算法。该混合算法充分发挥了SOA算法的群体搜索性和拟Newton法的局部细致搜索性,同时也克服了粒子群算法后期搜索效率降低和Newton法对初始点敏感的缺陷。以研制的车载位姿平台为例,在Simulink软件中建立了混合求解算法的仿真模型,给出了计算实例,并与单独的Newton-Raphson迭代法和SOA算法进行了对比。对比结果表明,SOA-Newton混合算法具有极好的稳定性和较高的收敛速度及精度,更能满足工程实际应用。

The positive kinematics solution is the basis for realizing high-precision control and 3 D visual simulation of 6-DOF platform,but the traditional method has the defects of difficulty in solving and low precision.Aiming at the problem of high sensitivity to iteration initial guess when using Newton-Raphson iterative method to compute the forward kinematic solution,a hybrid forward kinematic solution algorithm based on SOA-Newton iteration is proposed.The hybrid algorithm has sufficiently displayed the characteristics of SOA group searching and Newton method local strong searching.At the same time,it effectively overcomes the problem of high sensitivity to initial point of Newton method and shortcoming of PSO which reduces the searching efficiency in later period.Based on the developed vehicle attitude platform,a simulation model of hybrid solution algorithm is established in Simulink software.The calculation examples are given and compared with the Newton-Raphson iterative method and SOA algorithm.The comparison results show that the SOA-Newton hybrid algorithm has extremely stability,high convergence rate and precision,and can better meet the practical engineering application.