Leuven | More than two weeks ago
Multiple-input-multiple-output (MIMO) radars are becoming popular in indoor and outdoor environments for human target tracking and activity recognition. Complex urban/indoor environments lead to multipath reflections from static objects such as buildings, walls, static objects in the environment, etc. which complicates the task of detection and tracking. This is further complicated by the large size (compared to the radar resolution) and non-rigid nature of human bodies.
The goal of this master thesis is to develop algorithms for extended targets detection and tracking with a MIMO radar in a multipath environment. We will investigate a state-of-the-art tracking model i.e. Probability Hypothesis Density (PHD) and data-driven deep learning tracking models. The PHD filter is the state-of-the-art tracking algorithm which is based on the theory of random-finite-sets (RFS). For the end-to-end approach, the idea is to use insights from state-of-the-art deep learning models for camera- and lidar-based tracking and detection, and adapt those to be suitable for radar data-based detection and tracking tasks and compare these to PHD filter results.
The work includes:
Studying state-of-the-art radar-based detection and tracking
Implementing The PHD filter for MIMO radar
Implementing an end-to-end deep learning approach for MIMOM radar detection and tracking
Comparing and validating both approaches using existing data sets.
The successful candidate must be competent at Python coding and have some experience with relevant machine learning and AI toolkits (such as Scipy, Statsmodels, Sci-Kit-learn, PyTorch, Keras etc.). A good understanding of signal processing is also required. Prior knowledge of radar concepts is a plus.
Interested students can already get a feel for this subject from these papers:
Biruk K. Habtemariam, R. Tharmarasa, T. Kirubarajan, PHD filter based track-before-detect for MIMO radars, Signal Processing, Volume 92, Issue 3, 2012, pp:667-678, https://doi.org/10.1016/j.sigpro.2011.09.007.
Karl Granstrom, Marcus Baum, Stephan Reuter, Extended Object Tracking: Introduction, Overview and Applications, Journal of Advances in Information Fusion, Volume 12, Number 2, Pages 139-174, December 201, https://arxiv.org/abs/1604.00970
A. Palffy, J. Dong, J. F. P. Kooij and D. M. Gavrila, "CNN Based Road User Detection Using the 3D Radar Cube," in IEEE Robotics and Automation Letters, vol. 5, no. 2, pp. 1263-1270, April 2020, doi: 10.1109/LRA.2020.2967272.
J. F. Tilly et al., "Detection and Tracking on Automotive Radar Data with Deep Learning," 2020 IEEE 23rd International Conference on Information Fusion (FUSION), 2020, pp. 1-7, doi: 10.23919/FUSION45008.2020.9190261.
- Master Thesis internship (6 months)
- Preceded by optional summer internship (max 3 months)
Type of project: Combination of internship and thesis
Duration: 6 months
Required degree: Master of Science, Master of Engineering Science, Master of Engineering Technology
Required background: Electrotechnics/Electrical Engineering