Two more labs at TU/e join ranks of Dutch AI network

January 26, 2021

The labs focus on making cars and industrial robots understand the world around them.

The worlds of mobility and manufacturing are more and more driven by artificial intelligence. Now two leading TU/e labs in these fields, EAISI Mobility Lab and EAISI FAST Lab, are joining the ranks of Dutch ICAI labs. ICAI is a network of Dutch research programs that is designed to bring together researchers in the field of AI. The network helps TU/e to further strengthen its already close ties with Dutch industry, and exchange AI expertise and talent with other universities and knowledge institutes. TU/e now has a total of three ICAI labs.

EAISI Mobility Lab

EAISI Mobility Lab is a collaboration between TU/e and NXP Semiconductors. It aims to use the ongoing digitization of vehicles and transport services to bring about accident-free mobility. At present, too many people die or get hurt on our roads. The researchers at the Mobility Lab believe that technology and AI in particular can make vehicles inherently safe by giving drivers real-time support in critical situations. In order to achieve this goal, they use technologies like machine learning, world modelling, optimized human-machine cooperation, new vision and radar systems, and ultrafast actuators.

Foto: Shutterstock

The Mobility Lab aims to design intelligent mobility systems that sense their environment, learn and understand it, and, in interaction with humans, reason about which action to take to achieve specific goals. Examples include new types of cruise control systems, pro-actively adjusting speed based on surrounding vehicles, traffic jam assistance and platooning.

PERCEPTION

One of the focus areas of the EAISI Mobility Lab is perception technology. At present, intelligent mobile systems are increasingly able to sense the world around them, but in order to operate in an open, dynamic and often unpredictable environment, they also need to understand it. By understanding the outside world, (autonomous) vehicles can anticipate possible future events and plan safe and effective actions in response.

Program manager Margriet van Schijndel-de Nooij: “I firmly believe that artificial intelligence can boost safety and energy efficiency of vehicles, but also of our overall mobility system. To establish this, close collaboration with industrial partners like NXP is crucial. Being part of the ICAI network helps us to maximize the impact of our innovations.”

Program

The current research program of the EAISI Mobility Lab has been running since September 2019 and comprises the expertise of two research groups within TU/e (Mobile Perception Systems, and Control Systems Technology). It is carried out by five PhD researchers, plus a large number of students who are actively participating in the research. The team is currently working on extending the program for another two years.

For more information about the EAISI Mobility Lab, please contact Margriet van Schijndel - de Nooij at m.v.schijndel@tue.nl.

EAISI FAST Lab

Making machines understand the world around them, is also the focus of EAISI FAST Lab, the third TU/e lab to become part of ICAI labs. FAST stands for new Frontiers in Autonomous Systems Technologies, and is an industry funded research collaboration between TU/e and five companies: Vanderlande, ExRobotics, Rademaker, Diversey and Lely Industries.

Dealing with the unexpected (read: non-programmed events) is a routine task for TU/e’s prize-winning soccer and care robots, as you never know what an opponent or user will do next. Industrial robots, too, can benefit from increased independence and autonomy.

EAISI FAST Lab aims to develop mobile manufacturing robots with increased flexibility, which enables them to operate in changing environments. This has many advantages. For instance, the environment does not have to be fenced off, signaled or demarcated and specific zones or reference points will not be required.

World models

In order to be more mobile and flexible, the robots need so-called ‘semantic world models’, so that they can give meaning to observations, and recognize situations and objects. Being able to distinguish between, for example, a pallet with goods, a group of people and a forklift truck means there are wide range of different situations the robot can respond to. Robots also need to proactively gather information, so that they are able to carry out their task better.

By adding this new information to the robot’s world view, the robot can actually learn from previously solved tasks. This way, the next time a specific situation occurs, it will be able to deal with it more easily. As people will increasingly interact and engage with robots – think of workers in a factory, interaction with humans is also a key component within the semantic world model.

Collaboration

Program manager Jesse Scholtes is happy that EAISI FAST Lab has become part of the ICAI family: “Joining the ICAI LAB community provides us with a deeper connection to the AI network in the Netherlands, not only to exchange technical knowledge but also to effectively work together with industry to maximize the impact of the collaboration.”

The EAISI FAST Lab research program has been running since January 2018 and comprises the expertise of three different research groups within TU/e (Control Systems Technology, Mobile Perception Systems and Human Technology Interaction) and is carried out by four PhD researchers, strengthened even more with a large amount of involved students. The Lab is currently working on extending the program for another four years.

Are you interested in the work of the EAISI FAST Lab? Check out their webpage, or send an email to Shane O’Seasnáin (s.m.oseasnain@tue.nl).

THREE ICAI LABS

With the EAISI Mobility Lab and FAST, TU/e now has three ICAI labs. Earlier, EAISI AI-enabled Manufacturing and Maintenance (AIMM) Lab became part of the national ICAI network.

Carlo van de Weijer, general manager of the Eindhoven AI Systems Institute (EAISI): 
“EAISI started a year and a half ago with one of its main objectives being to accelerate AI research in collaboration with industry. We are therefore very happy to have already established three AI labs around the core themes of our research. Business is increasingly in need of AI solutions, we are developing more and more AI knowledge, and together we are bursting with AI ideas.”

Media contact

Henk van Appeven
(Communications Adviser)