Gait analysis beats ground radar
Student Best Paper Award for Aylin Tastan and Afief D. Pamudi
30.10.2020 von Martin Gölz, Sebastian Stamm
Darmstadt, 30.10.2020. At the 2020 IEEE Radar Conference, Aylin Tastan and Afief D. Pambudi from the Signal Processing Group of etit both successfully participated in the 2020 IEEE Radar Conference Student Contest. Aylin Tastan’s work on robust clustering for human gait signatures was awarded the Best Student Paper Award, whereas the runner-up by Afief D. Pambudi focused on robust landmine detection. Additionally, Pamudi also won the first prize in the Three-Minute Thesis competition, held for the first at an IEEE Radar Conference
Among the 143 papers submitted to the competition from across the world, the best five papers were selected as finalists through a rigorous review process by the AESS Radar Systems Panel Student Paper Competition Committee. In the final round, all participants presented their papers in a dedicated live on-line session to the jury.
Radar for human gait analysis
In Aylin Tastan’s award-winning paper, she designed a parameter-free robust clustering algorithm to cluster highly contaminated human gait radar data. The analysis of the human gait is a useful tool in medicine, because it can be used for diagnosing a gait disorder. The work of Aylin Tastan is very practical, due to the fact that her algorithm does not need any training data or a-priori information. Furthermore, it is very robust against outliers. The results of her work underline that radars are a useful tool in differentiating between different gait classes. In addition to that, her work shows that these differences can be learned by the radar in a non-parametric fashion. The experts were sure that her research has a huge impact on the smart-assisted living technology, which will help older people to continue living a self-determent life.
Increased detection of landmines
Afief D. Pambudi’s second-placed paper contributes to increase the performance and usage of ground-penetrating radar for detecting landmines. One of the reasons for deploying ground-penetrating radars instead of standard metal detectors for landmine detection is their capability of identifying very small or non-metallic landmines. In his work he integrated the dependence structures between various radar images in the decision making process by deriving a novel copula-based test statistic, which has already applied been in a different form in the financial sector. Due to his research results more landmines can be found and eliminated.
Afief also participated in the Three-Minute Thesis competition. In only 3 minutes he presented his topic to a jury and the entire audience of the conference. The audience decided that Afief´s presentation stood out from his competitors from around the world and voted for him in the final voting.
The work of Afief D. Pambudi is in close cooperation with Prof. Fauzia Ahmad from Temple University, PA, USA and the U.S. Army Research Laboratory.
Warm congratulations to Aylin and Afief!
