|
|
Subject
While multitarget tracking (MTT) is challenging in a single radar, a high performance is achievable by a network of multiple-input-multiple-output (MIMO) radars. Such multi-radar systems have become pervasive in self-driving applications for object detection and tracking purposes.
IMEC has successfully developed multi-radar MTT algorithms based on the state-of-the-art Probability Hypothesis Density?(PHD) filter [1] and the consensus-based track fusion [2]. The PHD filter is the state-of-the-art tracking algorithm which is based on the theory of random-finite-sets (RFS).
The goal of this master thesis is to further develop the MTT and fusion algorithms based on the cardinalized PHD (CPHD) filter [3]. In parallel, a data-driven deep-learning tracking model is developed [4,5]. For the end-to-end approach, the idea is to get insights from state-of-the-art deep learning models for camera- and lidar-based tracking and detection, and adapt those to fit in radar data-based detection and tracking tasks and compare these to the CPHD-based results.
Kind of work
Implementing the CPHD filter for MIMO radar
Track fusion to obtain 360o perception in the vehicle
Implementing an end-to-end deep learning approach for MIMOM radar detection and tracking.
Master Thesis internship @ IMEC (6 months)
Preceded by optional summer internship (max 3 months) @ IMEC
Framework of the Thesis
In collaboration with
Dr. Seyed Hamed Javadi (IMEC) hamed.javadi@imec.be
References:
1. B.-N. Vo and W.-K. Ma, “The Gaussian Mixture Probability Hypothesis Density Filter,” IEEE Transactions on Signal Processing, vol. 54, no. 11, pp. 4091–4104, 2006.
2. G. Battistelli, L. Chisci, C. Fantacci, A. Farina and A. Graziano, "Consensus CPHD Filter for Distributed Multitarget Tracking," in IEEE Journal of Selected Topics in Signal Processing, vol. 7, no. 3, pp. 508-520, June 2013.
3. B.-T. Vo, B.-N. Vo and A. Cantoni, “Analytic implementations of the cardinalized probability hypothesis density filter,” IEEE Trans. on Signal Processing, vol. 55, no. 7, pp. 3553-3567, 2007.
4. 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.
5. 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.
Expected Student Profile
Following an MSc in a field related to one or more of the following: Electrical engineering, Computer Science, or Applied Computer Science.
Experience with signal processing, and machine learning. Some knowledge of radar concepts is a plus.
Strong programming skills (Python).
Interest in developing state-of-the-art Machine Learning methods and conduct experiments.
Ability to write scientific reports and communicate research results at conferences in English.
|
|
