An Algorithm For Satellite Tracking Antenna Based On Servo Mechanism

ABSTRACT

There are increasing needs to develop control systems that can be used to automatically point Direct Current (DC) servomotor driven parabolic antennas to moving targets, notably in satellite tracking, to maintain the desired line of sight for quality communication. Although many researchers nowadays focus on artificial intelligence (AI) techniques, this work reckons that a well designed optimal linear controller can still give acceptable results at reduced system cost and complexity.

In this paper the design and control of DC servomotor-based antenna positioning system has been addressed and simulated in MATLAB/SIMULINK software. The response of the system is analyzed and results are first obtained by using well tuned Proportional-Integral-Derivative (PID) controller. The PID controller is then replaced by Linear Quadratic Regulator (LQR) which apart from optimizing the system response increases the accuracy of the state variables by estimating the states. It has been shown that the LQR results are much better than the results obtained by PID controller in terms of reduced steady state error, settling time and overshoot. Moreover, since low speed rotations are required and assuming linear conditions, the results obtained with LQR method can challenge those obtained by AI based techniques.

TABLE OF CONTENTS

TITLE PAGE

APPROVAL PAGE

DEDICATION

ACKNOWLEDGEMENT

ABSTRACT

TABLE OF CONTENT

CHAPTER ONE

1.0      INTRODUCTION

1.2      OBJECTIVE OF THE PROJECT

1.3      SCOPE OF THE PROJECT

1.4      PROBLEM  STATEMENT

1.5     JUSTIFICATION OF THE STUDY

1.6      AIM OF THE RESEARCH

1.7     CONTRIBUTION OF THE THESIS

1.7      LIMITATION OF THE PROJECT

1.8      PROJECT ORGANISATION

CHAPTER TWO

2.0     LITERATURE REVIEW

2.1      OVERVIEW OF SERVOMECHANISM

2.2       REVIEW OF THE RELATED STUDY

2.3      PREFERENCE OF SERVO MOTORS TO STEPPER MOTORS

2.4      DC SERVO MOTOR DESCRIPTION

CHAPTER THREE

3.0     METHODOLOGY

3.1  BLOCK DIAGRAM OF ANTENNA CONTROL MECHANISM

3.2      SYSTEM DESCRIPTION

3.3      ANTENNA POSITIONING BASICS

3.3.1  AZIMUTH

3.3.2  ELEVATION

3.4      ANTENNA POLARIZATION

3.5      TRACKING ANTENNA SYSTEMS

3.6      MATHEMATICAL MODELING OF DC (SERVO) MOTOR SYSTEM

3.7      PROPORTIONAL-INTEGRAL-DERIVATIVE (PID) CONTROLLER

3.8 LINEAR QUADRATIC REGULATOR (LQR) CONTROLLER & PID-LQR

 DESIGN

3.9      SIMULINK MODEL

3.10   SYSTEM FLOW CHART

3.11   TESTING CONTROL ALGORITHM AND SOFTWARE DESIGN IMPLEMENTATION

CHAPTER FOUR

RESULT ANALYSIS

4.1      RESULTS AND DISCUSSIONS

CHAPTER FIVE

5.1      CONCLUSION AND FUTURE WORK

5.2      REFERENCES

CHAPTER ONE

1.0                                               INTRODUCTION

Parabolic antennas mounted at earth stations which are commonly used in satellite tracking applications, are susceptible to suffer from environmental disturbances [1]. For many years, DC servomotor-based controllers have been in use to automatically position the satellite dishes [2]. Several controller models have been developed over time to solve the problem of antenna pointing in satellite and movable targets tracking using servomechanism [3], [4] and [5]. The case of overseas satellite telecommunication is considered in [3] where the control system directs on-board motorized antenna towards a selected satellite. Fault Tolerant Control (FTC) system is designed using the ship simulator facility to maintain the tracking functionality. However, the fault estimation has proved to be an extremely challenging task. An overview of most common servomotor-based linear antenna pointing mechanisms: Proportional-Integral (PI), Proportional-Integral-Derivative (PID), Linear Quadratic Gaussian (LQG) modeled and implemented for varied space applications is provided in [5]. PI controllers are easy to implement but take much time to reach setpoint and have degraded performance under system nonlinearities. LQG controllers are not only optimal but also have the ability to estimate non-measurable states by using observers to reconstruct them and provide better performance in case of wind gusts noise. However, the performances of these methods depend on the accuracy of system models and parameters. Generally, an accurate non-linear model of actual DC motor is difficult to find. Recently, new intelligent control techniques such as Neural Networks, Genetic Algorithms, Fuzzy Logic and hybrid AI methods are under research consideration as a viable solution to the problem [5]. The DC servomotor suitable for automatic dish antenna positioning in the case system ought to possess a minimum torque of 5.0Kg.cm at low speed and a maximum torque of 7.9Kg.cm at full speed. Within those ranges, the motor can support the following categories of dish antenna sizes: 0.30m, 0.45m, 0.55m and 0.60m nominal diameter, weighing from 4Kg to 7.0Kg for typical material and design.

1.1                                   OBJECTIVES OF THE STUDY

1. To design a DC servo motor Controller using expert knowledge and simulate in MATLAB/SIMULINK.
2. To design a DC servo motor using Proportional-Integral (PI), Proportional-Integral-Derivative (PID), Linear Quadratic Gaussian (LQG) System Controller and simulate in MATLAB/SIMULINK.

• To implement     the     designed      controller    and    compare performance    with conventional PID

1.2                                         SCOPE OF THE STUDY

This work has covered how mathematical modeling of a practical DC servomotor based parabolic antenna positioning and tracking system can be carried out. This enabled the said system to be represented in terms of a transfer function. Focus was made on the control system design which included PID, FLC and NFSC control strategies, each algorithm developed and applied in turn to the system in MATLAB/SIMULINK simulation environment and the obtained results were analyzed. A prototype was developed using Arduino Mega 2560 microcontroller and other components. This was then used to test the NFSC performance off-line with Spring SM-S4309M DC servomotor from which experimental results were obtained. Since the dish was considered off-line the aspect of real time telecommunication signal tracking has not been covered as it is beyond the scope of this study.

1.3                                                                        PROBLEM STATEMENT

The ability to maintain communication over long distances beyond 550km in space for moving targets and satellites has always been a challenge. For instance, in mobile platform satellite communication, receiver systems are mounted on the movable device such as ship, train, car or airplane. For a seamless reception of signals, the antenna system must be steered in both azimuth and elevation angles to track a specified satellite or target [1], [2], [3] and [4]. The most common problem when aligning a dish is aiming at the correct satellite or target for the broadcasts required. The problem can be addressed by developing a dedicated control algorithm that uses the received signals from the satellite to control a DC servo motor system which will in turn point the antenna to the desired position. On the other hand, a DC servo motor control system suffers from disturbances and errors caused by nonlinear variations in load conditions, motor saturation, backlash, friction and wind pressures and gusts. For a long time, classical controllers such as PID control have been utilized mostly due to low cost and simplicity. However, the PID controller is poor at dealing with nonlinearities and disturbances and also requires an accurate mathematical model of the system which may not be practically realizable. Such weaknesses of the PID controller have made the Fuzzy Logic Controllers (FLC) a viable alternative [10] and [11]. The FLC can handle problems with nonlinearities, imprecise and incomplete data, uncertainties and vague description of the system because it uses expert knowledge based on fuzzy rules with linguistic labels. However, the main drawback of FLC is that it usually takes a lot of time to design and tune the membership functions which quantitatively define the linguistic labels. Neural Network learning techniques can automate this process and substantially reduce development time and cost while improving performance but it will not explain how decisions have been made. These limitations formed a central driving force behind the Neuro-Fuzzy System Controller under  study  in   this   research.   Consequently, the  NFSC   has   been   used  to  merge advantages of FLC with those of Artificial Neural Networks (ANN) and its performance investigated in comparison with the PID controller for bench-marking purposes.

1.4                                                                        JUSTIFICATION OF THE STUDY

In a motorized parabolic dish satellite tracking application, the most common conventional method is the PID controller owing to its low cost and ease of implementation but it can only be designed for linear conditions using precise mathematical models. However, in the case servo motor system internal nonlinearities such as saturation, backlash and friction as well as external disturbances like wind pressures cause error in achieving desired output motor position. The NFSC method adopted in this research offers the following benefits over existing strategies:

• It takes care of the operational knowledge of the dynamical servo system as it is a knowledge-based approach.
• There is no need of precise system mathematical
• It can work for the stated nonlinearities and external
• The approach combines the mapping and learning ability of ANN with the linguistic and fuzzy inference advantages of
• It is more accurate in achieving target position with not only reduced overshoots and settling times but also faster rise times as compared to the PID controller.

1.5                                                                        AIM OF THE RESEARCH

To design System Controller algorithms for controlling the position of DC servo motor to achieve desired antenna elevation and azimuth orientation in a satellite tracking system.

1.6                                                                        CONTRIBUTION OF THE THESIS

The key contribution made through this research work is the development of a NFSC with better dynamic performance and response to the nonlinearities of a DC servo motorized parabolic antenna pointing in terms of reduced rise time, settling time and percentage overshoot as compared to the most common classical PID controller. Also, the dependency of the DC servo motor Fuzzy Logic Controller performance on the knowledge level of the control human expert has been greatly reduced by fusing ideas of Proportional-Integral (PI), Proportional-Integral-Derivative (PID), Linear Quadratic Gaussian (LQG)