用神经网络求解机器人动态调度中的近似指派问题

本文利用HOPFIELD神经网络，对机器人动态调度中的近似指派问题提出了合理的神经网络表示方法，给出了网络的能量函数表示法及神经元状态方程，从而得出了机器人动态调度中近似指派问题的快速求解策略，满足了动态调度的实时性要求。本文从理论上论证了所提算法的收敛性。软件仿真结果表明，本文提出的近似指派问题网络求解方法是有效的，计算结果是满意的。

An Assigning Solution in Robots Dynamic Scheduling with Hopfield Neural Network

On the basis of Hopfield neural net work, this paper proposes a reasonable method to represent approximate assigning problem in robots dynamic scheduling with Hopfield neural network, and sets up the Energy Function and neuron state equation. A rapid solution to the approximate assigning problem in robots dynamic scheduling is resulted. It can meet the request of speed in dynamic scheduling. The convergence of this method has been theoretically proved. The satisfactory results of software simulation show that this method in this paper is efficient.