A robot control application with neural networks

Artificial Neural Network's inherent nonlinearity gives some advantages to use them on different kinds of problems including control engineering area, Learning of inverse dynamics with using neural networks is an example of the robot control applications [1], [2]. The dynamics of nonlinear systems vary with their parameters and in some cases, determining a single global model of the plant dynamics can be a very difficult problem, Designing piecewise control laws are useful methods to overcome this problems [3], In robotics, increasing the degree of freedom and working range of each link, directly creates more complex dynamics, The structure of multilayer perceptron is depend on the controlled plant and for more complex systems we need large networks and this increases the real time calculations of the robot control, For the proposed scheme, to decrease;he real time calculations, the working range of the robot is divided into several regions and for every region, separate neural network is used, Instead of learning whole dynamics with one large network, using this kind of strategy, we divide the complexity of the dynamics to small networks, In real time control, this piecewise or regional neural network structure is used together with PD controller.