Breadcrumbs

Software Engineering

The relevant projects are listed below. Alternatively, you can return to the main list of subject areas.

Please note: There are many overlapping aspects in the Chemical and Biological, Materials and Process, and Mechanical Engineering programmes.  Students should contact the academics of the various projects they are interested in. Note that it is possible to have design teams made up of students from different engineering programmes.

These pages are updated as new projects become available.  Please check back regularly.


SPICE model of a BC547

Project type:
Research and Design

Updated:
30/01/2018

Project description:
The BC547 has been around a long time. A detailed SPICE model has been available for over 40 years, but the values do not seem to be consistent over time. This is likely to be a result of changes in the silicon fab technology. This project aims to obtain values on a single transistor for the majority of the SPICE parameters. A point of interest will be comparing them with values obtained in the 1970s.

Supervisor:
Jonathan Scott


Kickstarter Pitch

Project type:
Research and Design

Updated:
30/01/2018

Project description:
This project would suit a small team (2 to 5) students, at least one of whom should be an EE. From a previous, very successful, project, we have the plans and code for a plug-in upgrade to Scalextric slotcars that installs the equivalent of anti-skid stability systems on real cars. The team will build a number of these then put together an offer on Kickstarter, including manuals, promotional video, etc.

Supervisor:
Jonathan Scott


Hardware Accelerated Analysis of Video Sequences for Direction and Depth

Project type:
Research and Design

Updated:
30/01/2018

Project description:
Humans and primates have sophisticated visual systems that enable them to navigate complex scenes by vision alone. Matlab and AVX2 implementations of the numerous filters that model the human visual system exist but take at least a few seconds to process one frame of video data. This project is to implement image processing filters on accelerated computer hardware to achieve real-time processing of video sequences. The hardware is the Intel Phi---a multiprocessor and vectorised processing unit, and is available in the Faculty of Computing and Mathematical Sciences. Video test sequences can be constructed with equipment and laboratories in the School of Psychology and the School of Engineering. The experimental apparatus can provide accurately known camera motion to which the output of the Intel Phi implementation can be compared. This project will require a student who is comfortable working in the Linux environment and programming in C or C++. This RESEARCH and DESIGN project would suit either a software engineering student or an electronic engineering student.

Supervisor:
Michael Cree (Engineering) and Assoc Prof John Perrone (Psychology)


An Experimental Visual Odometry Mobile Robotic Platform

Project type:
Research and Design

Updated:
30/01/2018

Project description:
Visual odometry is the process of getting odometry (velocity and heading information) on mobile robots by analysis of monocular video. This project is to develop a mobile robotic platform with GPU accelerated video processing and logging of video and telemetry data. What we have: A number of quadcopters with opensource/hardware flight controllers. A Jetson TK1 GPU/ARM embedded device. A Go-Pro camera and gimble. Other bits and pieces. What we would like to have: A camera mounted on a mobile platform, either a quadcopter or a small land-based vehicle, with the camera connected to the Jetson so that video data can be streamed into the GPU and logged. The Jetson connected to inertial sensors and/or the flight controller so that we can log inertial and telemetry data along with the video data. Stretch/challenge (particularly if have more than one student): Implementation of basic visual odometry so that we can get heading and speed information from the video/GPU (e.g. using standard OpenCV algorithms). Compare output of implemented visual odometry against inertial/telemetry data in a test. Particularly the camera interfacing to GPU part of the project would require students who are comfortable with working in the Linux environment and programming in C++/python. This research and design project would suit electronic engineering and/or software engineering students.

Supervisor:
Michael Cree, John Perrone


Modeling Disease Outbreaks in Urban Social Networks

Project type:
Research and Design

Updated:
30/01/2018

Project description:
Communication paradigm can be employed as a powerful canonical model for analysis of complex contagion processes. The project is to use communication models to characterize the spread of disease in realistic urban social networks for the control process.

Supervisor:
Yifan Chen


Molecular Communication Model of Targeted Drug Delivery

Project type:
Research and Design

Updated:
30/01/2018

Project description:
The project is to use molecular communication model, which is an emerging beyond-electromagnetism communication technique, to characterize the targeted drug delivery (TDD) processes for optimal targeted therapies. The student will also develop a tabletop system to verify the proposed techniques.

Supervisor:
Yifan Chen


Touchable Computation Model of Cancer Detection

Project type:
Research and Design

Updated:
30/01/2018

Project description:
Computation paradigm can be employed as a powerful canonical model for design and analysis of complex physical, chemical, and biological processes. The project is to use computation/optimization models to characterize the detection of cancers by using contrast-agent-loaded nanorobots. The student will also develop a tabletop system to verify the proposed techniques.

Supervisor:
Yifan Chen


NIARC 2017 - various research projects related to autonomous robots

Project type:
Research and Design

Updated:
30/01/2018

Project description:
Projects related to Kalman filter algorithm, SLAM and sensor fusion.

Supervisor:
CK Au