A neural controller for lane/wall following system

In recent years neural network has been adopted to design a great variety of control applications. It provides a powerful method to design non-linear controllers using parallel connections links. First, the neural network algorithm was trained on Matlab/Simulink. Then the neural network controller (NNC) is described in Very high speed integrated circuits hardware description language (VHDL), synthesized, optimized, placed and routed, and downloading on the FPGA board. The NNC controller and a further fuzzy logic controller are compared in term of their hardware resources and frequency. Both simulation results and real-time experiments will be presented for predefined trajectories.