From Sensing to Understanding with Smart Radars – Technology Innovations and Applications

Wednesday 17 November, 13h00-17h00
@ Flanders Make, Heverlee

TOPIC OF THIS HYBRID WORKSHOP

In this workshop that is co-organised by imec and Flanders Make, we will explain and demonstrate the potential of today’s smart radars for advanced RF sensing applications.

Due to the shift to higher frequencies (60 GHz and up) and the integration of multiple antennas, current radars are very small and offer high-resolution sensing for range, angular position and speed.

Powerful AI algorithms enable unprecedented capabilities for detection and classification of (human) movement and behavior, and for vital sign monitoring.
Moreover, radar sensors do this in a non-contact and privacy preserving fashion.

Experts from academia will present the state-of-the-art in radar sensing technology and discuss the challenges and opportunities ahead.
Speakers from industry will zoom in on the possibilities and challenges that these innovative radars offer for their applications, such as industrial automation, automotive, healthcare, etc.
Several innovative technologies and solutions will be presented during a guided demo tour.

We will end the workshop with a plenary Q&A session.

Participation is open to all technology enthusiasts who want to understand the technology behind radar sensing and its potential applications.

PROGRAMME

13h00Introduction
Kris Hermus, Coordinator Wireless Community & Innovation Program Manager Flanders, imec
Wouter Lammens, Business Development Manager, Flanders Make
Kris Van de Voorde, Business Program Manager, imec
slides (members only, login required)
13h10Past, present and future of technologies for smart radar sensing
Ilja Ocket, Program Manager RF Sensing, imec
slides (members only, login required)
  • working principles of radarthe shift to higher frequenciesfrom sensing to understanding with AI algorithmssize, power, cost,performance,… of off-the-shelf radars
13h40Radar-based hand gesture recognition using spiking neural networks
Inton Tsang, Senior Research Scientist, imec-IDLab-UAntwerp
slides (members only, login required)

We present a radar-based hand gesture recognition system using spiking neural networks and tested on publicly available reference datasets based on FMCW millimeter-wave radar. More specifically, we used liquid state machines of less than 1000 neurons and different classifiers, including logistic regression, random forest, and support vector machine (SVM) as readout maps. We achieved better than state-of-the-art results, reaching over 98% accuracy on 10-fold cross-validation for both data sets.
13h50Activity recognition for patient monitoring
Lorin Werthen-Brabants, PhD Student, imec-IDLab-UGent
slides (members only, login required)

Radar systems can be used to perform human activity recognition in a privacy preserving manner. This can be achieved by using Deep Neural Networks, which are able to effectively process the complex radar data. Often these networks are large and do not scale well when processing a large amount of radar streams at once, for example when monitoring multiple rooms in a hospital. We present a work that enables fast notifications to staff if troublesome activities occur (such as falling) by an on-premise device, while an off-premise device captures activities missed or misclassified by the on-premise device.
14h05Joint communication and radar sensing
Sofie Pollin, Professor, KU Leuven and Principal Scientist, imec
slides (members only, login required)

We introduce the topic of joint communication and radar sensing. Then, we show how an in-band full duplex communication radio can easily be extended with radar signal processing, almost for free. After a short video of the system, we conclude with future work and challenges.
14h20On the Performance of Radar vs LIDAR vs Vision in Challenging Environments
Niels Van Damme, Co-Founder & Tech Lead, OTIV
slides (members only, login required)

We will compare the performance of radar, lidar and camera sensor data in complex urban and industrial rail environments.
14h35COFFEE BREAK
14h55GUIDED DEMO TOUR
  • (imec) Detection, localisation, tracking and counting of people with an off-the-shelf mm-wave radar
  • (imec) High-resolution radar sensing at 140 GHz
  • (imec) Vital sign monitoring with radar sensors
  • (Flanders Make) Collwork AI
  • (imec) Radar-based hand recognition with spiking neural networks
  • (Flanders Make) Heavy and bulky load transportation with mobile robots
  • (SICK) Radar vs LIDAR based people detection

15h40Challenges and opportunities of radar sensing in manufacturing
Risang Yudanto, Research Engineer, Flanders Make
slides (members only, login required)

The future of manufacturing revolves around smarter productions agents, able to fail better and recover better. This requires an
enhanced understanding and modeling of the world in which the production agents are operating. Traditionally world modeling is
done by vision, proximity and light sensors. The recent breakthroughs in radar technologies provide new avenues for such world
modeling and therefore intelligent actions of productions agents like robots, AGVs and drones. In this presentation Flanders Make
provides an insight into these possibilities of the future.
16h00Radar sensing solutions for industrial applications
Johan DeCoster, Product Manager Identification & Measuring, SICK Belgium
slides (members only, login required)

SICK Sensor Intelligence will introduce its new RMS1000 Radar.
Important product features and characteristics will be discussed during this presentation.
Some typical applications will indicate where this radar can be used.
16h15Monitoring passenger flows in train and metro
Pascal Marly, Project Manager, Siemens
slides (members only, login required)
16h30Promising use cases of radar in railway and healthcare
Steven Lauwereins, Research lead, Televic Rail
slides (members only, login required)

Televic will present promising use cases of radar in railway and healthcare. These include unintrusive patient monitoring, real-time imaging of rail defects and catenary wires. The presentation will focus on the opportunities, requirements and current technological limitations.
16h45Detection and localization challenges in the orchestration of automation agents
Jurgen Dekeyser, Owner, Flagstone
slides (members only, login required)

Flagstone will discuss the current state of practice around orchestration and deployment of mobile robots, tools and vehicles,
highlighting the detection and localization challenges they face for which radar can present a possible solution.
17h00Radar: from the living room to the factory floor
Eva Maria Buchkremer, Head of Product Management, InnoSenT
slides (members only, login required)

As product manager for radar products at InnoSenT, Eva Maria will guide us through the various applications of radar.
17h15Plenary Q&A session
17h30Networking reception
18h00End of the workshop

REGISTRATION

Due to COVID19 measures, the number of seats for on site presence was very limited, and we have limited the number of participants per company/organisation.

Registration is still open, but for online attendance only.

Please complete the registration form below.

Registrations can be cancelled until November 10, 2021. After this date the complete registration fee will be charged. Please send an email to wireless-community@imec.be to cancel your registration.

No Fields Found.
Skip to toolbar