Margarita V. Sotnikova

Dynamic Positioning Based on Nonlinear MPC

Abstract This paper is devoted to nonlinear model predictive control (MPC) application for dynamic positioning (DP) problem. The MPC approach has prominent features among other nonlinear control schemes. They include the explicit way for using nonlinear models and taking into account constraints imposed on controlled and manipulated variables while provide optimal performance with respect to a given cost functional. Dynamic positioning is a nonlinear control process with three-dimensional input and output, constraints and external disturbances that allows to state that MPC strategy is a quite suitable approach to be used here. In this paper MPC based control algorithm for DP problem is proposed. This algorithm includes two stages: the nonlinear stage, where nonlinear predictive model is used, and the linear stage, which is switch on in the vicinity of desired position. The algorithm provides offset-free performance. The effectiveness of the MPC approach is demonstrated by the simulation study examples.

Control Systems Analysis and Design Labs with Educational Plants

Abstract This article is devoted to using of the special laboratory plants in control education. The complex of control system analysis and design labs is presented, where each lab deals with one of the experimental control plants. The brief description of the plants is given and their main features are analyzed from the control point of view. The substantial formulations of the labs, which are discussed in the paper, include the problems of system identification, digital control systems design, signal processing and real-time implementation. The advantages of using MATLAB environment for the problem solution are pointed out. Finally, the examples of the labs are presented.

Modal Parametric Optimization of Control Laws with Special Structure

The paper is devoted to particular cases of optimization problems for dynamical systems with scalar control and external disturbance. The aim is to find optimal parameters of an output feedback controller having initially given structure. An admissible set of controllers is determined by the requirement to place the poles of linear closed-loop connection inside desirable regions on the complex left-half plane. Numerical algorithm is proposed to solve the posed problem using special multi-purposes structure of the controller. To show practical applicability and effectiveness of the proposed approach and the benefits of developed methods, the illustrative example of marine autopilot design is presented.

Transoceanic routes optimization using dynamic properties of ship and weather conditions

This paper is devoted to the problem of optimal weather routing for transoceanic ship voyages. The goal of optimization is to reduce travel time and fuel consumptions. The optimal route is searched on the admissible set, which comprises routes satisfying all of the imposed static and dynamic constraints. The static constraints are represented mostly by coastal lines and shallow waters. The dynamic constraints are time-varying alarm zones restricted for ship motion due to dangerous weather conditions. The proposed route optimization algorithm uses long-term weather forecast for a sailing region and a mathematical model of ship motion. The key feature of algorithm is the three-dimensional representation of the route, which allows transforming initial finite-dimensional optimization problem to the problem of path finding on special graph.

Object recognition and distance evaluation using single camera

The template images of object and camera motion parameters are used as input data. The proposed approach is based on interest points detection and tracking through the sequence of images. The practical significance is demonstrated using Matlab examples and real-time experiments with a mobile robot.

Ship Dynamics Control using Predictive Models

Abstract This article is devoted to using of model predictive control (MPC) approach for ship motion control problems. Currently, MPC method is one of the modern approaches in control theory. MPC algorithms have prominent features in comparison with other control schemes. The main advantage is the explicit way for using nonlinear models and taking into account constraints involved on the controlled and manipulated variables. In this paper two different cases of the MPC-strategy application for ship dynamics control are considered. First case is dedicated to ship motion control in the turning circle process where additional constraint on the dynamic roll angle is imposed. The second case is concerned with the control of the velocity for high-speed craft fitted with the interceptors. Both these processes are essentially nonlinear and imply input and output constraints. It is shown that the MPC scheme is superior to other control algorithms in the considered situations.