University of Malta

2011 Projects
UOM Main Page
Apply - Admissions 2016
Campus Map button

Final Year Projects Academic Year 2010/2011



Student: Christian Barbara
Supervisor: Prof. Carmel Pulé


Robots are widely used nowadays, and there’s no shortage of jobs for robots to do. A significant task robots perform is to go places or do things that no human would want or is able to do. In this project, a surveillance RC car was built to perform such task. This was done by introducing a wireless router on an RC car and connecting a network camera to it.
Project Objectives
The main purpose of this project was to build a machine that can be controlled by a human operator from a distance. A camera was required so as to permit the user to drive the machine even when it is out of line of sight.
Project Methodologies
Figure [1] depicts exactly an overview of the project. The user’s laptop, connected to the router by the IEEE 802.11g standards, will send the desired directions by pressing the arrow keys. These are passed to the router as bits. The router then directs them to the serial port which is connected to the microcontroller. Consequently the microcontroller, which has the RC motor and steering servo connected to its digital pin, would translate the bits coming from the router as respective output. The project was carried out as follows:
• Literature review on wireless communications, focusing extensively on the IEEE 802.11 and its various generations.
• Studying socket programming, Unix Networks and typical server-client programs. [2]
• Familiarization with third party firmware, WRT54GL Linksys router hacks and projects. [3]
• Acquaintance on ‘cross compiling’ since the router has not sufficient memory for a compiler to be installed on it. [4]
• Modify the router for serial communication.
• Using Visual Basic for Graphical User Interface. (Figure [2]).
Results and Achievements
When applied indoor, changing the transmission power of the router, did not make any difference. This is due to severe interference caused by equipment such as cordless phones and due to propagation. However, the car’s Wi-Fi range will be tested in a football ground where interference is minimal. The range is estimated to be 500m.
[1] Shelato. “Building An Internet Controlled Security Robot”. [Online] Available:
[2] W. Richard Stevens, Bill Fenner, Andrew M. Rudo". UNIX Network Programming. Boston MA: Pearson Education, 2004.
[3] Paul Asadoorian and Larry Pesce. Linksys WRT54G Ultimate Hacking. Syngress, Burlington, MA, 2007.
[4] Eric Bishop. “Writing and Compiling A Simple Program For OpenWrt”. Internet: coding.html, Aug 23 2007 [Feb 16 2011].



Student: Rosemarie Mangion
Supervisor: Prof. Carmel Pulé


The project is about an imitation of an aircraft landing system which is very useful in bad weather conditions especially fog. Weather conditions like fog can prevent an aircraft from taking off or landing at airports. This presents a hazard, because the pilot may think that visibility is suitable for landing while in fact it wouldn’t be.
Project Objectives
The aim of this project is to create a very stable and reliable system which is mounted on wheels and has to follow light, leading the aircraft to the appropriate touchdown point. Only a simulation of the vertical contour was achieved. The modulation for the two side lobes was not tackled.
Project Methodologies
During the implementation of the project the following steps were carried out:
• Literature Review of current ILS (Instrument Landing System) technologies and also how aircraft landing was facilitated in the past.
• Construction of electronic circuit, prediction of its behaviour and tested later on. Electronic circuit basically consists of two LDR’s which are present to compare light intensities. Differences in light introduce an error which is amplified and fed to the motor.
• Control System analyzed using BBC BASIC
• System implementation and observing its performance.
Results and Achievements
When obtaining the error from the LDR’s, system was stable. To obtain a more reliable system an observer can be introduced in order to model the real system. When plotting curves on BBC BASIC the desired results were obtained since real and ideal system curves were approximately near each other. While testing the model, the platform moved along the rod following the light indicating that the circuit was working correctly. Varying the position of the torch light, it could be seen clearly that as the torch light was far away with minimal light falling on the LDR’s, the platform was moving slowly towards the light but it gained speed as the torch light was brought closer to the platform. Also, changing the direction of light instantly, the system seemed to respond efficiently as the direction in which the platform was heading changed abruptly and smoothly.



Student: Matthew Micallef
Supervisor: Ing. Kenneth Chirchop


Optimisation of !ight trajectories is currently a major area of research in aviation. In this dissertation optimisation of energy required by the aircraft’s engines or fuel consumed and time needed to complete a trajectory were considered. One technique how to solve multi objective optimisation problems is by using evolutionary algorithms based on the Genetic algorithm.
Project Objectives
The objective of this dissertation was to build a multiobjective optimiser which could be applied to flight trajectory optimisation. This optimiser had to be integrated with an Aircraft Performance Model to optimise the flight trajectory of a generic aircraft similar to an A320 and a B737.
Project Methodologies
A review of the existing evolutionary genetic algorithms was carried out. NSGA, SPEA, NSGA2 and SPEA2 were considered in detail. These algorithms were compared and the best one was chosen. SPEA2 and NSGA2 give almost identical results, but SPEA2 gives more diversified solutions and can be applied to more than two objective functions[1]. A modified SPEA2 to handle constraints was implemented in JAVA. To validate the optimiser, several tests were applied. Multiobjective functions with constraints found in literature were applied to the optimiser and the resulting Pareto fronts were compared to the global Pareto fronts which were obtained by Deb[2]. After obtaining satisfying results, the optimiser was used together with an Aircraft Performance Model developed by the department of Electronic Systems Engineering to optimise for the total fuel or total energy and total time required by the aircraft to complete a segment .The four segment climb and eight segment cruise trajectories were considered to find the best optimiser settings.
Results and Achievements
The optimiser succeeded in reaching the global Pareto front for all flight trajectories considered. Eight segment and four segment results were compared. Eight segment results were more accurate due to the fact that it emulates a true aircraft trajectory better. Figure 1 shows the optimal trajectories obtained for minimum energy and minimum time for a four segment climb. Figure 2 shows the Pareto front obtained for the four segment climb.
[1] Tomoyuki Hiroyasu, Seiichi Nakayama, Mitsunori Miki, ‘Comparison Study of SPEA2+,SPEA2 and NSGA2 in Diesel Engine Emissions and Fuel Economy Problem’, The 2005 IEEE Congress on Evolutionary Computation, 2005, Vol. 1, pp. 236 – 242.
[2] Kalyanmoy Deb, Amrit Pratap,Sameer Agarwal and T.Meyarivan, ‘A fast and Elitist Multiobjective Genetic Algorithm NSGA-2’, IEEE transactions on Evolutionary Computation 2002, Vol. 6, No.2, pp.182-197.



Student: Daniel Pisani
Supervisor: Prof. Ing. David Zammit Mangion

Digital platform stabilization is a technique by which an object is stabilized to keep a constant line of sight (LOS). Such systems are used in a large array of fields with the two most common fields being the aviation and military. In the aviation field, such platforms are used to stabilize cameras for filming and also for guidance purposes. These are becoming more common in unmanned aerial vehicles (UAVs). This project is a continuation of a previous project, in which an analogue based platform stabilization, was implemented using analogue electronics [1]. The main disadvantage of using an analogue system is the fact that random noise from components could make the signal loose important information and would also result in distortion of the signal. However, if these are eliminated the accuracy of the system could be improved.
Project Objectives
The aims of this project included the following:
1) Implementation of the analogue system on a microprocessor.
2) Improve the system by making use of a digital proportional-integral-derivative (PID) controller and other digital techniques. Then comparison between the systems is to be completed.
3) Implementation of a new control algorithm by placing MEMS (micro-electromechanical sensors) on the platform.
4) Study of sensor characteristics and performance.
Project Methodologies
Initially a detailed analysis of the analogue system and a literature review on several sensors and control loops was carried out. Several digital control laws were then designed by making use of MATLAB® and Simulink®. These control laws included the design of a digital PD (proportional-derivative) controller, a digital PID controller and since the transfer function of the system was known, pole-placement techniques were also carried out. The design of the system was implemented using a microprocessor. Several comparisons between different techniques were then carried out. These comparisons focused mainly on the use of different sensor con#gurations, most importantly placing the sensors on the platform rather than on the aircraft’s fuselage. The sensors used throughout the project included MEMS accelerometers and gyroscopes which are relatively low cost and accurate sensors. Sensor calibration was also carried out for improved results.
Results and Achievements
The results obtained in the project included the digital stabilization version of the algorithm used in the previous thesis. This algorithm consisted of stabilizing the platform using a potentiometer to read the aircrafts pitching angle and the motor encoder to note the platform angle. Then, stabilization was obtained by making use of di"erent control loops and sensor positioning. Stabilization of the platform was obtained by making use of an accelerometer and a gyroscope strapped down with the fuselage. Then, stabilization was obtained by using a single gyroscope placed on the platform and a digital PID controller. The final stabilization was then obtained by making use of two gyroscopes placed on the platform (using averaging) and an inclinometer to initialize the system.
[1] Grixti S, “Platform Stabilization for Airborne Applications,” B.Eng dissertation 



Student: Josef Pollacco
Supervisor: Prof. Carmel Pulé

Unmanned Aerial Vehicles (UAV) can perform tasks which would be diffcult or even hazardous for a manned vehicle. In modern days, UAVs are used for an extensive range of applications, starting from filming movies, through search-and rescue missions, up till military reconnaissance missions. An autonomous helicopter falls under such a category, having various advanced capabilities over other aircrafts, with the main capability being the ability of hovering. However, due to their advanced and complex nature, helicopters are more diffcult to control when compared to fixed wing aircrafts. In fact, modern helicopter aerodynamics is still a demanding field for researchers due to its diffculty in measurement, modelling and prediction.
Project Objectives
The main objectives of this project are to implement an autonomous helicopter, which will go through the flight dynamics of lift, hovering and landing, by using gravitational and physical sensors to perceive any disturbance affecting the over-all inclination and correcting the error caused by disturbance to achieve stable flight. Such correction computations will be quantified by an on-board microcontroller, which will control the helicopter servos accordingly.
Project Methodologies
The project was divided in three stages:
(i) Helicopter construction: choosing the appropriate equipment, such as the frame, motor, servos and speed controller was crucial in order to observe some factors that were included in the project. Such a factor is the weight limit
(ii) Inertial Measurement Unit (IMU) design and implementation: the IMU in this project is composed of physical sensors (such as accelerometers and gyroscopes) that perceive an inclination or a rotational displacement around any of the three axes. The design consisted of choosing the appropriate sensors, the power supply unit and its corresponding smoothing and biasing components and drafting the PCB on which all the components were going to be placed. After testing the functionality of the merged components, a box was fabricated so as to contain the electronics and be attached to the helicopter
(iii) Software code implementation: an Arduino Mega is the helicopter’s chosen on-board microcontroller, on which the software code is implemented. The main function of the programme will be to combine the sensor data through a sensor fusion process. In this project, such a process is achieved by utilizing the Kalman filter. This algorithm will be used so as to compute an estimate of the inclination, by utilizing the input data of the accelerometer and gyroscope. After achieving all the objectives, all the equipment was merged together as shown in Figure 1.

Results and Achievements
The IMU and Kalman filter combination proved to be an exceptional choice, successfully producing accurate and consistent output estimates. Up till now the helicopter has managed to achieve momentary flights although it tends to sway to a random direction. However this inaccuracy can be concluded to be a minor calibration issue in the software and by reviewing the code, this flaw should be solved.  



Student: Jeremy John Scicluna
Supervisor: Prof. Ing. David Zammit Mangion

A Standard Instrument Departure (SID) is a departure route designed to meet Air Traffic Control (ATC) requests and is normally used in busy terminal areas. The main purpose of an SID is to reduce the workload for both ATC and the human pilot while still observing the minimum obstacle clearance requirements. The benefits achieved by the construction of such a departure are: effectively controlling the !ow of traffic with the least communication being required between ATC and the pilot, increasing the traffic capacity within the terminal area and reducing the environmental impact by making use of the noise abatement procedures.
Project Objectives
The aim of this project is to simulate a departure. The system will need to simulate the basic autopilot and auto throttle systems with similarities to that of a Boeing 747. Both systems will need to be as accurate as possible with no instability, so as to be able to simulate the departure with the highest level of precision.
Project Methodologies
The project was carried out as follows:
• An extensive literature review covering various subjects of the aviation theory.
• Understand the Boeing 747 model.
• Design of an auto throttle system that will manage the required airspeed.
• Design of an autopilot system that will climb or descend so as to capture and hold the required height.
• Design of an autopilot system that will roll the aircraft in order to manage the required track.
• Simulate an actual SID.
• Interface with the Flight Gear Simulator.
Results and Achievements
The results obtained show that the design and implementation of both the autopilot and the autothrottle systems worked successfully, while the simulations show that the aircraft performed the desired departure successfully. Finally, the simulations were interfaced with the Flight Gear Simulator to obtain a visual result of the SID.



Supervisor: Prof. Carmel Pulé

Helicopters play an important role both in military and in humanitarian operations. This is attributed to their agility and their ability to maintain hover over a specific target. The challenge in maintaining hover on stormy days is due to the diffculty in controlling the aircraft against external disturbances, such as wind, while a hoisting procedure is being executed. The development of Unmanned Aerial Vehicles (UAVs) is essential so as to reduce human resources and to increase the safety factor significantly.

Project Objectives
The objectives of this project are:
• To design an electronic control system to control the rotors of a helicopter platform.
• To implement the system and stabilise it.
• To demonstrate the e"ectiveness of the technique and the achieved performance.
 Project Methodologies
• To solve the problem and tackle the objectives, a small-scale quadrotor helicopter was designed, constructed and tested. The design is shown in Figure [1]. The design allowed for easier construction and increased rigidity when compared to the traditional quadrotor helicopter design.
• Research was also carried out to select the required components appropriately. Familiarisation with an Arduino microcontroller was required. The latter was programmed to use the information gathered from the inertial sensors (accelerometer and gyroscope) and to filter this information. The filter is used to combine the sensors together and to reduce the impact of any fast movements, such as vibrations, on the system.
• A Proportional Integral Derivative (PID) controller was implemented so that the system would be able to respond to external disturbances by balancing the speed and thrust of the four motors and thus maintaining a stable hover. This controller is widely used in industrial applications because of its relative simplicity yet very effective technique. It takes into account the past, present and the theoretical future of the system’s performance so as to achieve the desired control.

Results and Achievements
• The quadrotor helicopter was constructed successfully. Brushless outrunner motors are used which are very powerful for their relative size. They are controlled by Electronic Speed Controller (ESCs) and powered by a suitable Lithium-ion Polymer (Li-Po) battery.
• The Arduino microcontroller was programmed to serve as the brain of the system, incorporating the filter algorithm and the PID controller.
• The accelerometer and gyroscope were filtered and combined together to control the roll (x-axis) and pitch (y axis) of the craft. The filtering technique is shown in figure [2] (The movement is light coloured and the filtered output is superimposed for both the roll and pitch).




Last Updated: 16 November 2012

Log In back to UoM Homepage