İnsansız hava aracı için bir borda bilgisayarının mimarisi ve tasarımı
Başlık çevirisi mevcut değil.
- Tez No: 46362
- Danışmanlar: Y.DOÇ.DR. T. BERAT KARYOT
- Tez Türü: Yüksek Lisans
- Konular: Astronomi ve Uzay Bilimleri, Astronomy and Space Sciences
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1995
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 77
Özet
ÖZET Bu çalışmanın amacı, insansız bir hava aracında kullanılabilecek özgün bir borda bilgisayar mimarisinin tasarımı ve basit bir otopilot yazılımının bu borda bilgisayarında koşturulmasıdır. Bu amaçla ilk önce, borda bilgisayarının merkezi işlem birimi olmak üzere, HITACHI firmasının ürettiği H8-534 mikrodenetleyicisi kullanılarak, 1 Mbyte bellek adresleyebilen, 3 tane 8 bitlik paralel, 1 tane tam çift yönlü seri giriş-çıkış kapısı, analog giriş ve çıkışları bulunan bir mikrodenetleyici kartı tasarlanmıştır. Önerilen mimari uygulamada her ne kadar borda bilgisayarının iş yükü bir Sayısal îşaret İşleyici (DSP)'ye yüklenmeli ise de, otopilot yazılımı deneysel olarak DSP üzerinde değil, giriş çıkış görevini üstlenmiş olan H8-534 tabanlı mikrodenetleyicide gerçekleştirilmiştir. Bu arada uçuşa ait temel aerodinamik bilgiler ve klasik denetleyici metotları incelenmiş ve gerekli olanlarına bu çalışma içinde yer verilmiştir. Daha sonra, üç serbestlik derecesi ile yunuslama davranışı bir otopilot tarafından denetlenen, orta büyüklükte bir nakliye uçağı model olarak ele alınmıştır. Buna ait durum denklemleri dördüncü derece Runge-Kutta yönteminin kullanıldığı bir yazılım ile çözülmüştür. Çözümleme işlemi sürekli şekilde ilk koşullar yenilenerek tekrarlanmış ve bu uçağa benzeşim yapan bir yazılım oluşturulmuştur. Ayrıca, bu benzeşim yazılımının üretiği irtifa değerini alıp, bu durum değerini referansla karşılaştırarak bir yunuslama komut işareti üreten irtifa otopilotu kişisel bilgisayarda gerçekleştirilmiştir, bu benzeşim modeli dışarıdan rastgele bozuntu girilmesine ve referans irtifanm değiştirilmesine de izin vermektedir. Son olarak bilgisayar ortamında uyumlu olarak çalışan uçak benzeşimi ve otopilot yazılımlarından, otopilot kısmı borda bilgisayarına aktarılarak kişisel bilgisayarla borda bilgisayarı arasındaki iletişim paralel olarak sağlanmıştır. Böylece kişisel bilgisayarda bir uçak benzeşimi ve borda bilgisayarında ona ait irtifa otopilotunun bulunduğu bir sistem oluşturulmuştur.
Özet (Çeviri)
A BOARD COMPUTER ARCHITECTURE AND DESIGN FOR A UNMANNED AIR VEHICLE SUMMARY Aim of this study is design of an on-board computer which can be used in an UAV and to run a simple autopilot software on this computer. For this purpose, firstly a microcontroller board that has a H8-534 microcontroller produced by HITACHI Company as its CPU and three 8-bit parallel and one full-duplex serial I/O ports and analog I/O ports. Although, in the proposed architectural application, wok load of on board computer must be carried upon by a DSP, autopilot software has been developed on a H8-534 based microcontroller board instead of DSP for experimental reasons. Meanwhile, basic aerodynamic information belonging to flight and classical controller method are analyzed and necessary information has been included in this study. After that, a mid-size transport aircraft whose pitching attitude with three degrees of freedom is controlled by an autopilot is considered as a model. State equations of the model has been solved by a software using Runge-Kutta method. Solution process is repeated by renewal of initial conditions continuously and simulation software has been developed. Besides an altitude autopilot produced a pitching command signal by comparing the altitude value produced by the simulation. software by the reference value. This simulation model allows external disturbances and change of reference altitude. Finally, autopilot software has been ported to on-board computer and parallel communication between PC and on board computer has been established. Thus a system composed of a PC running an aircraft simulation software and an on-board computer running the altitude autopilot part of the simulator has been established. Automatic Flight Control Systems (AFCS's) have become vital operational equipment of military and civil aircraft. These systems, vary from a simple roll stabilizer of a single engine private aircraft, to the sophisticated flight guidance systems that can control the flight paths of large transport aircraft from take-off to touchdown. This variety of AFCSs is due to the different flight handling characteristic of individual types of aircraft. Before dealing with the operating fundamentals of any AFCS and its application to an aircraft, it is first necessary to have enough knowledge on how an aircraft fly, its stability characteristics, and its controllability. Therefore, we firstly XIexamine the topics related to flight. ' In order an aircraft to rise from the earth's surface and fly at a constant altitude, there must be a force that opposes the gravity. This force that is called lift, is generated by the pressure difference between the upper and lower surfaces of the wing. Drag which acts on the aircraft is a force that must be opposed. This force is against the movement and is produced by the movement of a body through a fluid. Opposing to the drag is the propulsion force that is produced by a propeller or a jet engine and it causes a relative movement between the wings and surrounding air. The second subject is aircraft stability. If a system returns to a state of equilibrium after it has been displaced from a state of rest or a state of uniform motion, then this system is stable. There are two types of stability: static stability and dynamic stability. Static stability is related with immediate reaction of the aircraft and its tendency to return to equilibrium after displacement while dynamic stability refers to the subsequent long-term reaction that is of an oscillatory nature about a neutral or equilibrium position. Additionally it is important to note that, static stability is a prerequisite for dynamic stability, although the reverse is not true; it is possible to have a statically stable but dynamically unstable system. The displacements of an aircraft can take place in any one of three planes known as: pitching, yawing and rolling planes. When an aircraft has a tendency to return to a trimmed angle of attack position following a displacement, it is said to have positive static longitudinal stability. Static longitudinal stability is affected by stabilizer and distance of aerodynamic center from the center of gravity. Directional stability involves the development of yawing moments that will oppose displacement about the aircraft's vertical axis and so restore it to equilibrium. Unlike longitudinal stability, directional stability is not independent in its influence on the behavior of an aircraft. An aircraft has lateral stability if, following a displacement about the longitudinal axis, a rolling moment is produced which will oppose the displacement and return the aircraft to a wing-level condition. Dihedral angle, vertical location of wing about the fuselage, keel surface and flaps contribute the static lateral stability. The third topic is controllability of aircraft. Controllability requires aerodynamic forces and moments to be produced about the three axes of the aircraft. These forces always oppose the natural stability restoring moment and cause the aircraft to deviate from an equilibrium condition. Controllability is a different problem from stability, but stability limits controllability and vice versa. Primary control system that consists of elevators, ailerons and rudder, is operated by pilot for controlling the aircraft. Ailerons provide lateral control or rolling displacements, about the longitudinal axis, the rolling moment produced being opposed by aerodynamic damping in roll. They are always connected so that in response to instinctive movements of the control wheel by the pilot, they move opposite direction thereby assisting each other in producing a roll displacement. Elevators provide longitudinal or pitching control about the lateral axis, and they also assist the horizontal stabilizer to maintain longitudinal stability. They are connected to xitthe control column, that can be moved backwards and forwards. Pitch displacements are opposed by aerodynamic damping in pitch and by the longitudinal stability. As the response to elevator deflections is a steady change of attitude, elevators are essentially displacement control device. Rudder provides yawing moments or directional control about the normal axis of aircraft, such control being opposed by damping in yaw, and by the directional stability. The rudder is operated in response to instinctive movements by the pilot of a pair of rudder pedals. The response to rudder deflections is a steady state of change at angle of attack on the keel surfaces, so like the elevators, the rudder is a displacement control device. The displacements resulting from the various movements of the flight control surface are those intentionally set up by the pilot to maneuver his aircraft into required flight attitudes. Such attitudes are: straight and level, climbing, descending, rolling, turning and a combination of these. There are four principal forces affecting an aircraft in flight. First is lift that acts from the center of pressure. Second is weight that acts vertically downward through the center of gravity. Third is thrust which is forward propulsive force produced by turbine engine or propeller. Last is drag that is produced by movement. To be in equilibrium in straight and level flight attitude at constant speed, lift must be equal to weight and thrust must be equal to drag, and it is arranged that these forces act from points that are not coincident thereby producing couples that cause pitching moments. Climbing and descending are pitch attitude maneuvers which are set up by upward and downward movements respectively of the elevators. When the elevators are deflected a pitching moment is produced to rotate the aircraft about the center of gravity causing the lift vector to be inclined, and thereby constitute an accelerating force at right angles to the direction of flight causing the aircraft to initially follow a curvilinear path. Rolling of aircraft take place about the longitudinal axis, and is initiated by deflecting the ailerons in the required direction. When the effective angle of attack of each wing is changed, the down-going wing produces a greater lift than the up-going wing and thus rolling moment is established. Turning can be achieved by operating only aileron system in aircraft that has a well-designed aileron system. However, during this operation rudder can be used to assist. AFCS fulfills attitude changes intended by the pilot to overcome the aircraft stability. To achieve this operation, AFCS needs to follow-up the attitude of aircraft by means of sensors. The follow-up action is called as feedback. The roles of AFCS are as follows:.Overcome a stability and control deficiency..Improve the handling or ride quality..Carry out a maneuver that the pilot is unable to perform either due to tiie accuracy required the time over which it is necessary to carry out the task, or the lack of visual cues. The response of an AFCS is much more rapid than that of the human pilot, and prevents disturbances reaching sizeable proportion. AFCSs can be sub-divided into three distinct groups. xin.Sensors. These measure the relevant parameters and transmit the information in the computation group. These include gyroscope, rate gyroscope, compass, accelerometer, radar altimeter, and position sensors, etc..Computers. These convert the information from the sensors into the signals that are fed to the system output device. Computers perform fiinctions such as amplification, integration, differentiation, limiting, shaping and programming..Output Devices. These convert the computed signals into a form that will result in the necessary aircraft control surface movements. Servomechanisms are defined as a closed-loop control system in which a small power input controls a much larger power output in a strictly proportionate manner. In order to apply such a mechanism to the automatic control of an aircraft, system must be capable of continuous operation and have the ability to detect the difference between an input and an output, amplify error signal and control the servo- loop. There are two main classes of servomechanism: position control servomechanism and speed control servomechanism. AFCSs are designed with several control methods. These methods can classify conventional and, advanced or optimized methods. The conventional methods are appropriate to time-invariant, linear single input single output (si. so.) systems. They belong to the class of frequency domain methods, which are essentially graphical. These methods are pole-placement, model-following, root locus and frequency response. The performance of a linear, time-invariant, S.I.S.O. System is usually assessed by considering the nature of the roots of its characteristic equation. In the pole-placement method, designer can place the system's roots to desired locations by using comments. This operation can be achieved in five different ways. The simplest pole placement method involves using the specified values of the poles to form a characteristic polynomial of closed loop system. Method of eigenvalues assignment tries to make the eigenvalues of basic system as close to the desired values as possible. Coefficients of characteristic polynomial of closed loop system can be equated to coefficients of desired polynomial. Additionally, methods of modal matrix and output feedback can be handled for pole placement operation. In the model following method, firstly an aircraft which has all the dynamic characteristics necessary to achieve the flying qualities, are chosen as a model. Then a linear control law is found to produce a perfect match between the output variables of the closed loop controlled aircraft and those of the chosen model aircraft. In the classical analysis of S.I.S.O. linear system^ the description is predominantly by means of its transfer function. Design of AFCSs using the root locus technique involves selecting an appropriate value for the gain of the closed loop system, such that locations of closed loop roots in the s-plane correspond to acceptable flying qualities or introducing additional zeros and poles, thereby altering XIVthe dynamics of the closed loop system in a manner such that particular pole locations are achieved. If a pole is added to system then in its root locus representation, locus branch is driven away from that pole, and ifa zero added, it tends to attract the locus branch towards it. The introduction of a zero can improve the stability of a closed loop system because it can attract the locus branch away from the imaginary axis, or even from the right-half of s-plane itself to the far left of the s-plane. Although the frequency response function of a system is a measure of the steady state response of a system to sinusoidal inputs, a designer who uses this method can assess the stability of closed loop system from a knowledge of the open loop response. A number of easily determined frequency response parameters have been established as providing excellent guides to the resultant response in the time domain of the closed loop system. The frequency response diagrams that are most commonly used are:.The Nyquist diagram on which G(jw)H(jw) is plotted over some desired range of w..The Bode diagram, which shows a graph of 20 logT |G(jw)H(jw)| versus w plotted on a logarithmic scale, together with a separate graph of the phase angle, arg{G(jw)H(jw)}, plotted against w on the same logarithmic scale..The Nichols diagram is a plot of 20 logio |G(jw)H(jw)| versus phase angle, arg{G(jw)H(jw)}, with the particular points on the graph annotated with the frequency to which it relates. For AFCSs, the basic rules, using frequency response as a design method, can be summarized in the following way. At those frequencies where good maneuvering and stabilization are required, use high loop gain, and at those frequencies, where noise rejection and insensitivity to changes in the aircraft dynamics are important, use low loop gains. Settling the gain is the first step in adjusting the system for satisfactory performance. As is frequently the case, increasing the gain value will improve the steady-state behavior but will result in poor stability or even instability. It is then necessary to redesign the system in order to alter the overall behavior so that the system will behave as desired. An additional device inserted into the system for such purpose is called a compansator. Lead compensation essentially yields an appreciable improvement in transient response and a small improvement in steady-state accuracy. Lag compensation yields an appreciable improvement in steady-state accuracy at the expense of increasing the transient-response time. Lag-lead compensation combines the characteristics of both lead compensation and lag compensation. Those conventional control techniques all depend upon an interpretation of the system dynamic response, in terms of such parameters as settling time, frequency of oscillation of the transient, value of the peak overshoot, time-to-half amplitude, gain margin and so on. The design which results from using such methods is obtained as a consequence of some compromise, and it may not be unique. By using the modern theory of optimal control, a specified performance criterion is met exactly and the corresponding control design is unique. xvControllability is a property relating the effect of the control inputs to the changes in the state variables of some mathematical representation of the aircraft dynamics. The model of aircraft dynamics is said to completely controllable if any and only if all state variables, namely Xi(0), can be transferred in finite time, by the application of some control function, u(t), to any final state x(T). On the other hand, observability means that it is possible to determine the state vector x(0), at time t=0, from the output variables which occur in the future. The couple are very important for the optimal control. Microcomputer is essential part of the AFCS. In developing the AFCS microcomputer can be either designed or bought. In the case of design procedure the choice of microcontroller or microprocessor is of importance. In this study, designing the microcomputer is preferred instead of buying a ready one. First step of the design procedure is the determination of requirements. Following conditions must be considered while the requirements are determined. 1. The microcomputer must have analog and digital input-output port, so that the controlled systems may be analog or digital. 2.1f the controlled systems are dynamic or if more than one system are ' controlled, the speed of microcomputer is very important. 3.The microcomputer must be capable of addressing very large of memory if it is going to be used in image processing applications or etc. 4.The microcomputer must be suitable for running software that is developed using high level languages. 5. The microcomputer must have the required facilities to communicate with the other computers serially or parallelly. 6.The microcomputer must consist of DMA (Direct Memory Access) unit, if it is going to perform data acquisition. In this design H8-534 chip is chosen as a microcontroller. The H8-534 is a high performance single-chip microcontroller produced by Hitachi Corporation having 16-bit internal data bus. The on-chip supporting functions include RAM, ROM, Timer, a serial communication interface (SCI), A/D Conversion and I/O ports. An on-chip data transfer controller (DTC) can transfer data in either direction between memory and I/O independently of the CPU. For the on-chip ROM, a choice is offered between masked and programmable ROM (PROM). t The microcontroller has 20-bit external address pin,son it can address 1M Byte memory which is separated into 16 pages. According to usage of memory, five modes are available: Expanded Minimum Mode (enable on-chip ROM), Expanded Minimum Mode (disable on-chip ROM), Expanded Maximum Mode (enable on-chip ROM), Expanded Maximum Mode (disable on-chip ROM) and Single-chip Mode. CPU has seven addressing modes: 1. Register direct 2Register indirect 3. Register direct with increment 4Register indirect with pre-decrement or with post-increment 5. Immediate xvi6. Absolute 7.PC-relative CPU has eight 16-bit general registers and seven control registers (two 16-bit registers and five 8-bit registers). Microcontroller that can operate up to 10 MHz, is, optimized for efficient programming the C-language. While designing board of microcontroller, all pins which include same function, are combined at the same connectors. Pins that have the same names same, are placed ât same locations on each board. Pins which belong to each same on-chip supporting module are connected to separate connectors. All boards which are piled up can be connected easily with flat cable. Boards which consist of memory components are designed separately from microcontroller board. Pins which have more function are divided by using jumper and attached to related connector. Every board has to layers. Two types of memory boards, are produced. First is EPROM board which includes four 128 K bytes (27C010) chips and the second is RAM board which includes four 32K bytes (62256) chips. The boards both are pin compatible with respect to their connectors. All electrical scheme and PCB printout are shown in, Appendix. By means of the board explained above, it has been observed that altitude autopilot has connected to the simulation software running on the PC vja parallel port and controlled the altitude changes of the aircraft chosen as model. xvn
Benzer Tezler
- Quadrotor VTOL aracının modellenmesi ve kontrolü
Modeling and control of a quadrotor VTOL vehicle
BORA ERGİNER
Yüksek Lisans
Türkçe
2007
Elektrik ve Elektronik Mühendisliğiİstanbul Teknik ÜniversitesiMakine Mühendisliği Ana Bilim Dalı
Y.DOÇ.DR. ERDİNÇ ALTUĞ
- Tracking control methodologies for a quadrotor UAV
Dört rotorlu bir İHA için yol takibi kontrol yöntemleri
BORA BAYRAKTAROĞLU
Yüksek Lisans
İngilizce
2017
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiKontrol ve Otomasyon Mühendisliği Ana Bilim Dalı
PROF. DR. MÜJDE GÜZELKAYA
- Active slam with informative path planning for heterogeneous robot teams
Heterojen robot takımları için bilgilendirici yol planlamalı aktif EZKH
MEHMET CANER AKAY
Doktora
İngilizce
2020
Elektrik ve Elektronik Mühendisliğiİstanbul Teknik ÜniversitesiKontrol ve Otomasyon Mühendisliği Ana Bilim Dalı
PROF. DR. HAKAN TEMELTAŞ
- Mikro hava araçlarının bilinmeyen ortamlarda görüntü temelli kontrolü
Vision based control of micro air vehicles in unknown environments
CİHAT BORA YİĞİT
Yüksek Lisans
Türkçe
2012
Mekatronik Mühendisliğiİstanbul Teknik ÜniversitesiMekatronik Mühendisliği Ana Bilim Dalı
YRD. DOÇ. DR. ERDİNÇ ALTUĞ
- Experimental analysis of direct injection in wankel engine
Wankel motorunda direkt püskürtmenin deneysel analizi
MAJID JAVADZADEHKALKHORAN
Yüksek Lisans
İngilizce
2018
Makine Mühendisliğiİstanbul Teknik ÜniversitesiMakine Mühendisliği Ana Bilim Dalı
DR. ÖĞR. ÜYESİ OSMAN AKIN KUTLAR