Interview with Mathieu Joerger, Assistant Professor at Virginia Tech

Interview:

1) Could you please tell us about yourself?

My name is Mathieu Joerger, I come from a small village between France and Germany. I went to an engineering school in France and earned a MS in Mechatronics. I then joined Illinois Tech in Chicago where I earned a MS and PhD in Aerospace Engineering, and stayed as postdoc and research professor. I was an assistant professor at the U of Arizona for three years, and I am now assistant professor at Virginia Tech.

2) Why did you want to become an aerospace engineer?

I did not. I met my advisor at Illinois Tech who inspired me to specialize in aerospace engineering. He showed me that I had a lot to learn, and that I could contribute to advancing theoretical and applied research in this field.

3) Could you give us an insight to your research in controls of dynamics systems? My research focuses on the feedback branch, on sensing, perception and navigation. I work on understanding sensor errors and faults, on advancing methods to quantify the impact on estimation of errors and faults, and on mitigating these impacts.

4) Your research expertise also covers navigation. Could you explain what a Multi-constellation global navigation satellite system is? GNSS is a generic term for the U.S. GPS and its international counterparts: the re-deployed Russian GLONASS, the emerging European Galileo, and the Chinese BDS. A GNSS is made of a space segment (the satellites), a ground segment (a network of ground stations estimating satellite orbits and monitoring satellite "health"), and a user segment (receivers). Using combinations of GNSS enables higher-accuracy continuous worldwide positioning as compared to using a single GNSS. GNSS can also be combined with other satellite constellations, such as Iridium satellites, which are intended for communication purposes but can be used for navigation. Recently, attempts have been made at leveraging low-earth-orbiting (LEO) telecommunication mega-constellations (e.g., OneWeb, Starlink, ...) for navigation.

5) Could you explain the challenges associated with creating a network like that? I am assuming that by "network" you mean a combination of GNSS. Early challenges with combining GNSS were that the nations deploying these systems, other than GPS, did not publish their minimum performance commitments. This has since changed, and we can now better quantify multi-constellation GNSS performance. Challenges with incorporating other LEO telecommunication satellites are that the owners or operators of LEO constellations have yet to appreciate the navigation accuracy, redundancy, and robustness benefits of dedicating a minute fraction of their communication capabilities to navigation. They are operating in private, with little possibility for interested and eager actors in academia to make practical contributions.

6) Your research expertise also covers autonomous multi-sensor safety systems. How does a system like that work? Man-made systems fail. For example, difficult-to-detect GPS satellite clock faults occur about once every 10 years for a given satellite, which corresponds to three faults per year for the entire constellation. If all autonomous vehicles that rely on GPS crash 3 times per year, we face a catastrophe. To be safe, we must detect these faults using redundancy from independent sources of information, including from other sensors.

7) What are some of the challenges associated with implementing a system like that?

A fundamental challenge when implementing a detector is to limit the risk of false and true alerts. False alerts, i.e., detection under nominal, fault-free conditions, are undesirable, but can also be hazardous. Imagine driving behind a car that stops suddenly, with no apparent reason -- you risk rear-ending it. In addition, and this is sometimes overlooked, the navigation system must be designed so that sensor and algorithm faults are rare to start with, otherwise true alerts (i.e., alerts due to actual faults) will interrupt operations at an unacceptably high rate.

8) What has been your toughest accomplishment so far in your career?

I won't discuss accomplishments, but I can talk about tough struggles: balancing professional life with family life has been a constant struggle for me.

9) What is a big milestone that you are looking forward to in your research? Excellent question! A few ideas come to mind. On the recommendation of Brad Parkinson, the father of GPS, we must improve GPS resiliency. GPS is part of the US critical infrastructure since its worldwide accurate timing capability enables communication and electrical power management. We must protect GPS by preserving its RF spectrum and catching and sanctioning interferers, we must toughen GPS by improving satellite, receiver and antenna technology, and we must augment GPS using other GNSS, other sensors, and other sources of navigation information. On a more specific point, and in a longer-term future, I am also looking forward to the day when we can rigorously quantify the safety of autonomous systems and of robots interacting with humans. 10) To high schoolers like me and to younger generations that want to become aerospace engineers, what would your advice be to be successful in this career? If you don't know where to go, and you believe you can do anything, then go where you will face the toughest challenges.