University of Malta
 

2008 Projects
UOM Main Page
 
 
 
Apply - Admissions 2016
Newspoint
Campus Map button
Facebook
Twitter

Final Year Projects Academic Year 2007/2008

INTERFACING OF FLIGHT COMPUTERS FOR AN AIRBORNE ELECTRONICS PLATFORM

Student: Warren Azzopardi
Supervisor: Prof. Ing. David Zammit Mangion

Image2008_A

Background
Embedded communication networks are playing a very important role in embedded and safety critical systems. Networking is being used to give greater system design flexibility, improve diagnosability, and most importantly when some critical data needs to be transferred safely from one point to another. In today’s modern aircrafts, “Fly-by-wire” control systems are being implemented in which critical functions are performed entirely by networked computers. As this shift forward in avionics using digital technology is taking place, the importance of designing inexpensive control networks for dependable real-time operation is increasing dramatically.

Project Objectives
The main objective of this project was to implement the Ethernet, CAN-Bus and I2C protocols for embedded communication. These protocols are to be used in the development of an airborne avionics suite for unmanned aircraft applications. The aim of the project was to implement these protocols on a selected ARM processor and to test the written drivers to make sure they are working appropriately and according to the published standards.

Project Methodology
• Phase 1: The Ethernet, CAN-Bus and I2C communication protocols and their message formats were reviewed.
• Phase 2: An appropriate ARM processor capable of supporting the required buses and communication protocols was selected. An integrated development environment (IDE), development tool chain and debugging tools to help in writing the software drivers were also selected.
• Phase 3: An assembly coded startup file and a linker script for the selected processor were developed.
• Phase 4: The Ethernet, CAN-Bus and I2C protocols were developed using the Keil IDE and the GNU compiler on the NXP LPC2378 processor as part
• Phase 5: Test programs to test the communication protocols on the development boards were finally written.

Results and Achievements
The drivers for Ethernet, CAN-Bus and I2C protocols to be run on the ARM processor for embedded communication have been implemented successfully and test programs have been written which transfer data between two embedded nodes without any problems. Arbitration procedures and other features of each protocol implemented have also been tested to verify that the functionality obtained is the one intended.


AN AUTOPILOT FUNCTION FOR LANDING IN FIXED WING AIRCRAFT

Student: Gilbert Cassar
Supervisor: Prof. Ing. David Zammit Mangion

Image_2008_B

Background
Due to the advances in Flight Dynamics and Performance of modern aircrafts,flight has become possible over a very wide flight envelope. Due to these advancements however flight became possible at many different Flight Conditions (changes in Mach, dynamic pressure, pull of gravity…) and this corresponded to large variations in coefficients that describe the Flight Dynamics of an Aircraft. For a given Aircraft, a dynamic mode that was stable and damped at one particular Flight Condition could become unstable in a different Flight Condition. Even if a dynamic mode is stable, it may result in lightly damped oscillations which cause discomfort to the passengers and make it difficult for the human pilot to maintain trajectory well. Autopilots are Automatic Flight Control Systems that can provide relief to the human pilot and also carry out special manoeuvres such as in the case of an Automatic Landing.

Project Objectives
This project aims to design a set of controllers which are used together to provide an automated landing for a Boeing B-747 Aircraft. The Controllers must make the Aircraft hold the desired Glide Slope Trajectory down to the runway without any deviations in the Longitudinal or Lateral Axis. When the Aircraft reaches an altitude of 15m above the runway the Control will switch to a Flare Controller which will reduce the rate of descent of the Aircraft while holding the front of the aircraft pitched up so that a smooth landing on the rear landing gear is ensured.

Project Methodology
This project deals with the most popular technique used in Autopilots: Gain Scheduling. The concept is to obtain Small Perturbation Models for different Altitudes and Flight Manoeuvres so that the Feedback Gains Applied will be a function of the Altitude, Airspeed and Steady State Flight Condition. In the case of the Landing Autopilot, a Small Perturbation Model will be obtained at different Altitudes for a Steady State Descent and the Controllers will be designed based on these Small Perturbation Models using Optimal Control Techniques.

Results and Achievements

A set of continuous-time Glide Slope Couplers and Lateral Stabilizers have been designed for the desired scheduling points. A prototype Flare Controller was also achieved to guide the Aircraft towards Touch Down. The Controllers were tested on a Non-Linear Model of the B-747 which the Controllers were originally designed for and the desired response as expected from the Linear Simulations was obtained.


SWITCHED HIGH FREQUENCY PULSED WELDING SET

Student: Stephen Dalli
;Supervisor: Prof. Carmel Pule'

Image_2008_C

Background
Some materials as aluminium and magnesium form an oxide layer very fast making them difficult to weld. The removal of oxide layers by etching acid is now replaced by the control of the rise time and polarity of the welding current and voltage waveforms used in the electric source feeding the power to the arc.Since aluminium is a very good heat conductor, heat is dissipated very quickly during the welding process. The rise time of a pulsed current will concentrate the heat to the welding zone instead of allowing it to be dissipated, resulting in less power needed to obtain a good weld. The high frequency switching also allows the transformer size to be reduced and to minimize the current zero crossing time which allows the arc to be extinguished.

Project Objectives
The project consists in researching in the electrical characteristics required toweld various materials. Analyzing different inverter topologies and also constructing a full bridge inverter with the adequate gate driving circuit and snubbing circuits in order to obtain a high frequency switched welding power source with constant current and pulsed current control. The intention of the project is also to research high frequency welding sources which would operate at 100KHz with a suitable current waveform to clean oxide layers on the welding surface.

Project Methodology
The final project consisted of 4 sections:
• DC voltage supply consisting of a rectifying voltage doubler producing a DC link voltage of 650V peak.
• The MOSFET full-bridge inverter with the adequate snubbing circuit
• The gate driving circuitry
• The control system
Once the delays, rise and fall time of the MOSFETs were obtained by testing them one by one, the dead-time needed was calculated and the half bridge topology followed by the full bridge supplying a ferrite transformer were tested.The circuitry was first operated at low voltage to ensure that no damage was done during the fine tuning of the circuit. The current and voltage waveforms where observed when loading the secondary side of the transformer. The control system provided the dead-time necessary and using a current transducer a closed loop system was constructed to obtain an adequate constant and pulsed current control.

Results and Achievements
Using the constructed control system a pulsed current output was obtained at enough power to maintain an arc between the work piece and the electrode. The transformer produced an output of approximately 60V when open-circuited to be able to start the arc. The voltage is then dropped down by the control system when short circuited to maintain an appropriate welding current.


RESCUE PERSONNEL ENVIRONMENT MONITORING SYSTEM

Student: Mark Alan McKeon
Supervisor: Prof. Carmel Pule'

Image_2008_D
Background
Modern-day fire-rescue and fire-fighting personnel enter hazard zones with a standard set of equipment consisting of a Self–Contained Breathing Apparatus as well as an Automatic Distress Signal Unit. The latter sounds a high decibel alarm when sufficient motion is not detected, and the former is a breathable air supply system that gives notice of remaining cylinder pressure via a mechanical pressure gauge. Units with digital readouts are commercially available but give no more information than their mechanical counterparts. Knowing the pressure of an air cylinder allows mental calculations of ‘time-to-whistle’, which is a low pressure level (indicated, once reached, by a mechanical whistle) when breathing can no longer be adequately supported and evacuation of the personnel is the top priority. Monitoring of the user’s gear is an area where enhancement is possible. Monitoring of the environment in which personnel are working is an area that, so far, relies solely on observations made by the personnel themselves.

Project Objectives
The goal is to design an onboard package that is able to provide personnel with information about Temperature, Cylinder pressure and Time-to-Whistle. In addition, an attempt to design an intelligent unit capable of indicating the possibility and probability of flashover (spontaneous ignition of hot combustible gas and vapour) will be made. The main feature will be a unit capable of keeping track of the user’s movements, and hence position, using accelerometers in 3-dimensions.

Project Methodology
Each separate system was developed using LabVIEW software for BlackFin Digital Signal Processors, with real or simulated hardware sensors providing the necessary input data. The motion tracking system was tackled using principles based on double integration. An acceleration signal was integrated, resulting in a velocity signal. Another integration step gives a displacement signal. This signal can be plotted to give a visual indication of position vs. time. Using three accelerometers a 3D map of movement can be drawn. If no motion is detected for a pre-set time, the
Automatic Distress Signal is triggered. Time-to-whistle estimation was tackled by sensing air-pressure in the cylinder
and air-flow from breathing. Based on the volume of air remaining in the cylinder and the air-consumption per-minute, time-to-whistle can be calculated. A formula for relating the probability of flashover to air-temperature and the oxygen percentage in air was developed empirically through hands-on-research. The formula was applied in software, with inputs from temperature and oxygen sensors.

Results and Achievements
Each system was developed to a degree where their use in a real-life situation would be possible, though very fine calibration would be necessary to achieve results of high accuracy.

Calendar
 
 
Last Updated: 22 October 2012

Log In back to UoM Homepage