This paper presents a survey of neural network controllers for AUVs. A direct adaptive neural net controller incorporating integral action was designed for the heave motion of the ODIN underwater vehicle. The neural net controller is trained on-line by parallel recursive error prediction method and the critic equation. The influence of the various design parameters of the neural net controller ...

In this paper, we compare two different evolvable controller models based on their performance for a simple robotic problem, where a robot has to find a light source using two luminance sensors. The first controller is a fully meshed artificial neural network. Though neural networks are the most common type of controllers used in evolutionary robotics, validating and understanding the resulting...

A control scheme for dynamic tracking of mobile robots is presented, which integrates a velocity controller based on backstepping techniques and a torque controller based on improved RBF neural networks. The proposed torque control strategy derived from sliding modes depends on the dynamics of mobile robots. Because of the uncertainties in robot dynamics, the robustness of the system cannot be ...

– This paper presents the optimization of a photovoltaic (PV) water pumping system using maximum power point tracking technique (MPPT). The optimization is suspended to reference optimal power. This optimization technique is developed to assure the optimum chopping ratio of buck-boost converter. The presented MPPT technique is used in photovoltaic water pumping system in order to optimize its e...

This paper presents a neural network approach to controlling nonlinear system. Trial-and-error correlation learning, which is a generally useful method for optimizing parameters, is applied to training a neural controller to balance an inverted pendulum. The controller is simplified by automatically pruning the hidden neurons to only two. Computer simulation shows that the trained neural contro...

In this paper, a wavelet-based neural network is proposed for the control of nonlinear systems. Activation functions of neural network nodes are determined based on the wavelet transform. The controller can efficiently compensate for the undesired effects of hard nonlinearities such as saturation and/or dead zone of control input. Compared with standard neuro-controllers, the structure of the c...

In this paper, the design and implementation of an effective neural network model for turning process identification as well as a neural network controller to track a desired vibration level of the turning machine is as an example of using the neural network for manufacturing process control. Multi – Layer Perceptron (MLP) neural network architecture with Levenberg Marquardt (LM) algorithm has ...

In this paper, a novel PID-like neural network controller (PIDNNC) is created. It is composed of a neural network with no more than 3 neural nodes in hidden layer, and there are an activation feedback and (or) an output feedback in hidden layer, respectively. This special structure makes the network be able to be a P, PI, PD, or PID controller as needed. The proposed controller weights can be u...