Zhipeng Shen

A general fuzzified CMAC based reinforcement learning control for ship steering using recursive least-squares algorithm

A general fuzzified cerebellar model articulation controller (GFCMAC) is proposed. The mapping of receptive field functions, the selection law of membership function and the learning algorithm are presented. Recursive least-squares temporal difference algorithm (RLS-TD) is deduced, which can use data efficiently with faster convergence and less computational burden. Using RLS-TD method a reinforcement learning structure based on GFCMAC is applied to ship steering control, as provides an efficient way for the improvement of ship steering control performance. The parameters of controller are online learned and adjusted. Simulation results show that the ship course can be properly controlled in case of the disturbances of wave and wind. It is demonstrated that the proposed algorithm is a promising alternative to conventional autopilots.

A General Fuzzified CMAC Controller with Eligibility

This paper presents an online neural network controller. Cerebellar Model Articulation Controller (CMAC) is suitable to online control due to its fast learning speed. By integrating the CMAC address scheme with fuzzy logic concept, a general fuzzified CMAC (GFAC) is proposed. Then by incorporating the concept of eligibility into the GFAC, a GFAC controller with eligibility is presented, named FACE. A learning algorithm for the FACE is derived to tune the model parameters. To achieve online control, an efficient implementation of the proposed FACE method is presented. As an example, the proposed FACE is applied to a ship steering control system. The simulation results show that the ship course can be properly controlled under the disturbances of wave, wind and current.

Ship Steering Adaptive Robust Control Based on Kalman Filtering

Based on Nomotos equation for ship steering system, a state equation of the system was constructed, in which uncertainties were concerned. For the system, a robust control algorithm was proposed according to the upper bounds of the uncertainties. Applying Lyapunov stability theory and linear matrix inequality method, we proved that the proposed approach can guarantee the global stability of the closed loop system and lead the tracking error to a small bounded neighborhood of zero. Furthermore, an adaptive robust control algorithm was put forward with biological immune system to improve the adaptability of the system. To reduce the system disturbances, Kalman filtering was adopted in the system. Simulation results demonstrate the validity of the algorithm

Fuzzy s-tuning PID steering control for ultra large container ship

Based on input and output states, a construction method of the responding ship motion nonlinear mathematical model is presented. Comparing with state space model, the computation of responding ship model is simple, and the differential equation obtained can preserve the nonlinear effect. Then the mathematical model of the 5446TEU container ship is constructed according to its relative parameters and the trial voyage datas. Combined fuzzy logic with conventional PID, a fuzzy self-tuning PID controller is introduced. Applying the proposed controller to the 5446TEU container ship steering system, the simulation results show that the controller works in effect.

Predictive Control Research Based PID Neural Network of Large Ship

Aiming to the control feature of large delay system like ships, designing a 2-rank derivative multi-step neural network predictive model and the algorithm, and presenting a fuzzy control scheme based on the model with PID neural network and fuzzy CMAC controller, we solve the problems of model online identification and controller online design in traditional adaptive control, so that the high precision output fellow-up control of systems with large delay and uncertain nonlinear features can be realized. Simulation results show that the method had perfect control effect.

Research and application on a fuzzy CMAC controller with eligibility

A fuzzy CMAC controller with eligibility (FCE) is proposed. The eligibility can forecast the controlled system, and improve the system stability. The structure of FCE system is presented, and its learning algorithm is deduced. To make the algorithm fit to on-line control, the efficient implementation of FCE method is also given. Applying the FCE controller in a ship steering control system, the simulation results show that the ship course can be properly controlled in case of the disturbances of wave, wind, current and error in measure apparatus exist. It is demonstrated that the proposed algorithm is a promising alternative to conventional autopilots.

Web-Based Marine Engineering English Intelligent Training System Design

In order to simulate the special language practicing environment of marine engineering English, a marine auxiliary boiler simulation system is designed to represent the real ship system environment. Differing from traditional simulation systems, the simulation system designed in this research not only can carry out simulation exercise, but also test learners’ marine engineering English proficiency and technical skills. Moreover, web technology is adopted, so that operation time and locations are no longer constrained, therefore, the flexibility of learners’ independent learning is significantly enhanced.

A kind of robust controller for uncertain linear system LQ tracking problem

In this paper, the problem of linear quadratic tracking with infinite time-invariant is discussed. The description of matching uncertain linear system is presented and the error equation of the system is established, which can be considered as the general error dynamic system (GEDS). Hence, the tracking problem is transformed into stabilization issue. A kind of robust linear quadratic tracking controller is designed by solving a Riccati inequation which contains the uncertain information with the LMI method. By Lyapnov function, it can be proven that the controller guarantee all signals in the closed loop system robust stable. In addition, a simulation example is provided, which illustrates that the proposed controller results in robust performances to the model perturbation. The effectiveness of the designed control law is verified. The work done in the paper also improves the controller design method for linear quadratic tracking problem.

Research on a new type of ship main engine remote control simulation system

Taking the main engine and propulsion system of the large-scale container ships as the controlled object, a type of main engine remote control simulation system is designed, using the CAN-bus and database administration technology as well as touch-screen equipments. The system is composed of ECR controls system, BCR control system, and local control system with high density of integration. It can simulate the navigation condition of the main propulsion of the ship, display the output and control curves and visualize the simulation. It provides a favorable platform for research of marine main diesel propulsion control algorithm in laboratory and crew training.

Fractional sampling polyphase filter based conversion of image resolution

In order to meet the performance of realtime system, multirate digital signal processing is discussed in frequency spectrum, then a polyphase structure based efficient implementation scheme of the fractional sampling rate converter is presented. Applying the converter to image scaling, the simulation results show that the margin of the output image is legible, demonstrating the validity of the proposed fractional polyphase converter.