Algorithm developed to map and locate objects in unknown environments

​​​Self-driving cars, unmanned airplanes, planetary rovers and other systems need to construct a map of unknown environments while keeping track of objects located within them to avoid collisions.

This is done through computational algorithms known as SLAM: Simultaneous Localization and Mapping. The resolution of these maps is directly proportional to the wavelength of the electromagnetic signals they utilize, thus, millimetre waves have become a promising spectrum for SLAM systems.

Dr. Mariette Awad​, ​Dr. Yousef Nasser​, and PhD student Ali Yassin from the department of electrical and computer engineering​ at the Maroun Semaan Faculty of Engineering and Architecture (MSFEA) investigated a 2D SLAM scenario using a wavelength between 1.0 and 1.5mm on the y-axis and a series of mirrors manually oriented on the x-axis, using a linear antenna array. Although this is an ideal situation and in practical scenarios several parameters can differ from laboratory conditions, their research proved to successfully resolve the geometry of the surroundings and the localization of the user with sub-centimetre accuracy.

This research was published by IEEE Access in November 2018.