Bachelor and Master Theses

Motivation

Hand gesture recognition (HGR) is a versatile tool for human-computer interaction in various real-life interactions like Virtual or Augmented Reality (VR/AR). Most often, (HGR) is performed on camera-based systems which are susceptible to light conditions and can cause privacy issues. This thesis aims to investigate hand gesture recognition based on Frequency Modulated Continuous Wave (FMCW) radar signals. In FMCW radar, a continuous electromagnetic microwave is generated whose frequency varies over time. FMCW radar is very sensitive to movement, penetrates solid objects, and from the reflected signal even the velocity of a target can be determined.

For this thesis, an FMCW evaluation kit is available for data acquisition. A machine learning algorithm will then be applied to train a neural network for gesture recognition based on the received radar signals.

Tasks

In this thesis, the student will:

-       Get familiar with Frequency Modulated Continuous Wave (FMCW) radar technology and the AWR1443BOOST radar development board from Texas Instruments

-       Implement a machine-learning algorithm for gesture recognition from the acquired radar signals

-       Evaluate the capabilities and limitations of the implemented system

Literature

J.-W. Choi, C.-W. Park, and J.-H. Kim, “FMCW Radar-Based Real-Time Hand Gesture Recognition System Capable of Out-of-Distribution Detection,” IEEE Access, vol. 10, pp. 87425–87434, 2022, doi: 10.1109/ACCESS.2022.3200757.

Basic programming skills. Experience with setting up hardware systems is beneficial but not a requirement.