A big international buzz around space resources exploration and Luxembourg began in 2016: Luxembourg passed a legal framework for space resources utilisation and launched the spaceresources.lu initiative to attract companies in the space sector. In 2020, the European Space Resources Innovation Centre (ESRIC) was opened and the same year space robotics became part of Luxembourg’s National Research Priorities.
Luxembourg has been making an early mark in the space resources exploration endeavour, with several research projects tackling what is by many considered a kind of last frontier. Since 2016, the space sector in the Grand Duchy has been growing rapidly, with about 75 companies and research institutions active in the space sector and employing nearly 1,200 people.
Simulating the Moon surface
A strong contrast between light and shadow, the cratered surface is grainy and dusty. Rolling with movement not unlike a tank, a rover appears at the edge of the crater. What sounds like the Moon is in fact an 80 m2 lab, with 20 tonnes of basalt gravel simulating the bedrock of the Moon. Equipped with sensors throughout, students and researchers can monitor their AI algorithms and robots’ movements down to the millimetre.
The LunaLab, opened in the framework of the University’s Interdisciplinary Space Masters, is one of the few facilities across the globe that simulates lunar conditions.
“New companies are focussing on new activities where robotics are essential – such as space exploration. In the LunaLab, we focus on simulating the visual appearance of the Moon, so we can test AI algorithms for the localisation, mapping, trajectory planning and control – basically autonomous navigation of the robots”Prof Miguel Olivares-Mendez
Prof Miguel Olivares-Mendez, who heads up the LunaLab, works on developing AI algorithms on autonomous navigation in different research activities related to space with his 15-member strong research group SpaceR. This includes orbital robotics – spacecrafts or satellites that can perform certain tasks in space autonomously, semi-autonomously, or tele-operated – planetary robotics and aerial robotics.
Planetary robotics include landers, rovers, hoppers (a type of spacecraft) and other mobile systems to explore celestial bodies and to find, identify, extract and collect space resources for their utilization in Space (ISRU). Concrete examples of orbital robotics include satellite and spacecrafts equipped with robotics tools, such as robotic arms or robotic docking systems for on-orbit servicing like refuelling, maintenance, re/de-orbiting, etc, with collaborative and non-collaborative targets. Both together are defined as Space Robotics.
Together with Prof Gilbert Fridgen (PayPal-FNR PEARL Chair), Miguel recently embarked on a new space research project, where the scientists are developing space robots that follow a given governance (RegTech) and make autonomous economic decisions (FinTech) to jointly create systems where multiple robots collaborate on tasks. This kind of dynamic collaboration between space robots could lower market barriers in the space sector and transform the whole industry.
Teaching a lunar rover to navigate the Moon alone
In space exploration, time is everything. The time a lunar rover will have to fulfil its mission is limited and, ideally, no time should be wasted – a lunar rover should figure out how to get from A to B as quickly as possible. As part of his PhD, Philippe worked on creating a navigation system for a small lunar rover: teaching it to navigate around on the Moon’s surface, while creating a map, and locating itself on this map, in the process.
Philippe Ludivig, who recently completed his PhD at the University of Luxembourg’s SnT, in collaboration with Japanese company ispace.
“We needed to teach the robot some awareness of itself and its surroundings. This involves accurate localisation, which is an estimate of where the robot is relative to its starting point. Once the robot understands its position, it can then determine whether it has successfully moved from A to B.” Philippe explains.
“However, this field also investigates more complex forms of autonomy which also involve obstacle detection and avoidance as well as path planning challenges. For applications on earth, this localisation problem is largely solved through GPS systems, but for planetary robots, the difficulty lies in the lack of such external localisation systems. The work therefore focussed on internal localisation systems that can fit inside small lunar surface robots.”
In the research domain, this is known as ‘SLAM’ – ‘Simultaneous Localisation and Mapping’. It essentially allows the rover to understand its environment, which is the first step needed to take autonomous decisions within this environment.
The system Philippe worked on has the lunar rover scanning one area at a time, shown as a point cloud – a 3D map of the surface. Piece by piece, the rover creates a map. The computer scientist worked both with lunar rovers at the ispace Moon testing facility in Luxembourg, as well as testing his research on rovers in the LunaLab at the University of Luxembourg’s SnT.
In a few years, this novel navigation system could be in operation, playing a key role in ispace’s ambitious programme for polar ice exploration.
”It’s nice when you go home in the evening, you can look at the Moon and say ‘that’s where my research is going!”Philippe Ludivig
What difference has the project made for Luxembourg? The LunaLab at the University of Luxembourg and the Lunar Yard at ispace in Luxembourg are both direct products of this research. There now exist facilities in Luxembourg which focus on testing vision-based localisation systems for planetary surface robotics. Currently, there are only a handful of similar facilities worldwide!
Philippe conducted his research in the framework of a collaborative project between the University of Luxembourg’s SnT and Japanese lunar exploration company ispace. The FNR has special funding programme (Industrial Fellowships and BRIDGES) to enable researchers in Luxembourg to work with companies. Philippe’s project is one of several FNR-funded collaborations between ispace and researchers, and was one of the first space exploration projects funded by the FNR.
Understanding where there is light and where there is darkness
This is a map highlighting light hitting the Moon’s North Pole. Why is such a map of interest for scientists working on lunar rovers?
Before humanity can begin planning lunar resources exploration, good maps are an essential step. Rovers and instruments that could explore the Moon will probably rely on solar power and batteries. Maps like this one highlight which areas receive a lot of light – and which ones are dark.
Treasure in the risky darkness
Even in places near the poles that receive some sunlight, it is necessary to understand just how much sunlight they receive – a rover that is charged by the Sun should not spend too much time in the dark, as it cannot recharge its batteries there and could become stuck.
Why not just avoid dark areas? Permanently lit and permanently dark places at the lunar poles are ideal places for exploration and lunar resource development. The rovers will have to venture into the dark areas, because only these places may contain water ice – essential for any hope of life support on the Moon.