The Connaught Automotive Research (CAR) Group NUI Galway was established in 2005 by its directors Dr Edward Jones and Dr Martin Glavin.  The primary objective of the CAR Group is to develop technology for intelligent vehicles, through the application of state of the art techniques in signal processing, image processing, computer vision and machine learning. This is an area of increasing interest and importance to society, and ranges from advanced driver assistance systems (ADAS) that support the driver in managing the vehicle in its environment, through semi-autonomous operation where the vehicle drives itself in limited circumstances, and eventually to fully-autonomous systems. The work of the CAR Group is based on broader expertise in Electrical & Electronic Engineering on the application of signal processing and machine learning algorithms in a range of application domains includeing biomedical, speech, and environment/agriculture, that has been developed over 25 years.

Connaught Automotive Research (CAR) Group

Developing signal and image processing technology for the automotive industry.

Current Research Topics:

Applications of Wireless Communications Technologies

This research involves automotive applications of wireless communications technologies for distribution of information (control information, or entertainment) inside the vehicle.

Previous projects have investigated the potential of Bluetooth as a primary interface to the vehicle for access, control and telemetry for the purposes of diagnostics and repair.

Applications of Digital Signal Processing for In-Car Audio Systems

In this area, the group are investigating the application of DSP techniques for speech and audio processing in the automotive environment.

In particular, for the purposes of improving the quality of sound in the vehicle cabin, e.g. by reducing the perceptual effects of background noise.

Image Processing for Driver-Assist Applications

This work involves the development and prototyping of image processing algorithms suitable for so-called "driver assist" applications, particularly with safety in mind.

The basic concept is to process image streams from cameras placed at various locations on the vehicle, and to extract or emphasise useful information in these image streams. For example, important features such as other road users, signs, and road markings are detected from the image streams, and this information is then fed to the driver. The additional input from the driver assist technology will facilitate earlier detection of potential hazards.