Hardware Implementation of a Real-Time Neural Network Controller With a DSP and an FPGA for Nonlinear Systems

Abstract

In this paper, we implement the intelligent neural network controller hardware with a field programmable gate array (FPGA)-based general purpose chip and a digital signal processing (DSP) board to solve nonlinear system control problems. The designed intelligent control hardware can perform real-time control of the backpropagation learning algorithm of a neural network. The basic proportional-integral-derivative (PID) control algorithms are implemented in an FPGA chip and a neural network controller is implemented in a DSP board. By using a high capacity of an FPGA chip, the additional hardware such as an encoder counter and a pulsewidth modulation (PWM) generator is implemented in a single FPGA chip. As a result, the controller becomes cost effective. It was tested for controlling nonlinear systems such as a robot finger and an inverted pendulum on a moving cart to show performance of the controller.Index Terms-Digital signal processing (DSP), field programmable gate array (FPGA), inverted pendulum, neural network controller, proportional-integral-derivative (PID) controller, robot finger.

Hardware Design for Machine Learning

“…This is accomplished by means of VHDL code that can be easily translated into an FPGA implementation, using suitable electronic-design-automation software. The work in [67] describes the hardware implementation of a real-time neural network controller with a DSP and an FPGA for nonlinear systems.…”

Experimental Studies of Swing Up and Balancing Control of an Inverted Pendulum System Using Intelligent Algorithms Aimed at Advanced Control Education

“…Here, neural network control was added to the PID controlled system, as shown in Figure 6, to give robustness to the system. The scheme known as the reference compensation technique is shown in Figure 6 [14,17]. Neural network control modifies the desired trajectories to minimize the output errors.…”

Experimental Studies of Neural Compensation Technique for a Fuzzy Controlled Inverted Pendulum System

“…A neural network controller has been known as a powerful nonlinear controller so it can be used as a nonlinear controller by itself [11][12][13][14]. Combing the neural network with the fuzzy controller is expected to yield the better performance since merits of two intelligent tools are used.…”

“…We add a separate neural network to change fuzzy rules by modifying reference input values [10]. This forms the reference compensation technique control method as one of neural network control methods that has been used for controlling the inverted pendulum system [11][12][13].…”