论文部分内容阅读
该文将多协作冗余机械臂的运动控制问题转化成一个可完全分解的约束二次求解问题,并将该约束二次求解问题拆分成一系列独立的子问题,每个子问题对应于单个机械臂的运动控制模型,其大大简化了多协作机械臂的运动控制模型。另外,该文采用对偶回归神经网络求解该运动控制模型,每个独立的神经网络核对应于一个机械臂的控制模型。该文首次将对偶回归神经网络用于多冗余机械臂系统的控制问题中,具有很强的应用研究意义。与现有的对偶回归神经网络相比,该文提出的对偶回归神经网络结构非常特殊,其多项性能将大大提高,使得多协作机械臂的运动控制问题变成完全分布式。理论验证结果表明,该文提出的模型是全局稳定的并且可得到全局最优解。仿真实验结果表明本论文提出的方案是有效的。该文的研究结果可为多协作冗余机械臂的控制理论和技术发展提供新的思路。
In this paper, the problem of motion control of multi-cooperation redundant manipulator is transformed into a problem of quadratic solving which can be completely decomposed and the problem of quadratic solving is divided into a series of independent sub-problems. Each sub-problem corresponds to a single machine The arm motion control model greatly simplifies the motion control model of a multi-cooperative arm. In addition, this paper uses dual recurrent neural network to solve the motion control model, and each independent neural network kernel corresponds to a manipulator control model. This article for the first time uses dual dual neural network to control the multi-redundant manipulator system, which has a strong application significance. Compared with the existing dual recurrent neural networks, the dual recurrent neural network proposed in this paper is very special in structure, and its performance will be greatly improved. The motion control of multi-cooperative manipulators becomes completely distributed. The theoretical verification shows that the proposed model is globally stable and can obtain the global optimal solution. Simulation results show that the proposed scheme is effective. The results of this paper can provide new ideas for the control theory and technology development of multi-cooperative redundant manipulators.