Navigation

Using AI to make assistance systems more intelligent for real-life situations

Portrait Annette Hagengruber
Annette Hagengruber is an external doctoral candidate at the Machine Learning and Data Analytics Lab at FAU. (Image: DLR)

FAU doctoral candidate Annette Hagengruber nominated as AI newcomer

FAU doctoral candidate Annette Hagengruber has set herself the goal of designing solutions based on artificial intelligence to help people with restricted mobility in their everyday lives. The Gesellschaft für Informatik e.V. (German Informatics Society) has nominated her as an AI newcomer of the year in recognition of her achievements. In our interview, she describes her research and gives us an insight into how she manages to juggle work commitments with writing her doctoral thesis.

Ms. Hagengruber, you are working at the German Aerospace Center (DLR) to design a mobile robot assistant which you hope will be able to help people with restricted mobility in their everyday lives. What exactly are you researching?

There are two core aspects to our wheelchair assistant EDAN (EMG-controlled Daily AssistaNT). Firstly, we are working to develop and implement partial autonomy, with EDAN providing support for everyday tasks such as eating and drinking. A central aspect of this is that users retain the freedom to take decisions at all times. Secondly, we are developing an interface which is aimed at allowing people with extremely restricted mobility to regain the ability to use a robot assistant. My personal research focus is muscle-based interfaces. A joystick would be the conventional choice for steering a robot in three dimensions. However, a joystick is not an option for people with severely restricted mobility caused for example by neurological disorders, who have lost control of their extremities. Alternative input devices have to be found for these people. I am working on a human-robot interface which measures remaining muscle signals using electromyographic (EMG) sensors and decodes these into a continuous control signal for the robot using methods derived from machine learning. AI-assisted partial autonomy serves to simplify complex movements. For example, if the user would like to pour water into a cup all they have to do is move towards the cup with the bottle in the robot’s hand. The system then recognises the user’s intention and carries out the rotation required for pouring the water into the cup without any further commands. The user can determine how much liquid should be poured out. Using the robot allows users to complete tasks like this independently again.

What brought you to this area of research?

I studied medical engineering, and have always been interested in how technology can be used to help people in the medical sector. Working on robot assistants at DLR gave me the possibility of combining both these aspects. Robot assistants are a relatively new and exciting field of research which offers a wide scope for new ideas and research, as well as the opportunity to get involved in putting the systems into practice.

You are also completing a doctoral degree on the same topic at FAU – how did that come about?

I met Prof. Bjoern Eskofier at a conference in 2018. We discovered that my research objectives were well suited to the topics covered by his department. After a few meetings in Erlangen, we agreed that I would fit in well as an external doctoral candidate at the Machine Learning and Data Analytics Lab at FAU.

How do you manage to cope with working and writing a doctoral thesis at the same time?

There are several areas where my project work overlaps with my research for my thesis. I make the most of these, of course. However, I do also have to plan well and be careful with my time management to ensure that I don’t neglect my doctoral thesis. It can be rather stressful at times, but if the project or paper is successful, then it was all worth it in the end.

What do you find so fascinating about AI, what makes you an AI newcomer?

AI is opening up a whole new world of opportunities which we can use to make our assistance systems more intelligent, more intuitive and more robust for real environments. However, it is important that AI is used correctly. There is a big difference between use in the laboratory and in the ‘real world’. I am working at combining AI and robots for use in real applications. I enjoy challenges like these and look forward to designing technology which can be put to good use in practice.

Voting for AI newcomer of the year

The project ‘#KI50: Künstliche Intelligenz in Deutschland – gestern, heute, morgen’ (#AI50: Artificial Intelligence in Germany – yesterday, today, tomorrow’), run by Gesellschaft für Informatik e.V. (German Informatics Society), is turning the spotlight on young, committed AI researchers from all disciplines. Ten outstanding AI newcomers will then be recognised within the context of the Federal Ministry of Education and Research’s ‘Science Year 2019 – Artificial Intelligence’. Votes can be cast until 17 November. Further information is available on the website.

You can vote directly for Annette Hagengruber here.

Another FAU doctoral candidate has also been nominated as an AI newcomer: Elisabeth Hoppe. An interview is available here, and you can vote for her via this link.

Addition information