Leo Rover conquers the Moon once more – the final round of the Space Resources Challenge

See how the Leo Rover robots of the LUVMI-XR team tackled the finals of the Space Resources Challenge.

May 16, 2023


Aleksandra Szczepaniak

Space Resources Challege arena

Let’s be honest – competition or not, space exploration always gives us the chills, doesn’t it? And it surely did give the chills to the contestants of the Space Resources Challenge finals. 5 teams competed on an artificial lunar surface in Luxembourg showcasing their technology and its possibilities. Let’s see how the Leo Rover robotic platform fit into that :).

What is the Space Resources Challenge?

The Space Resources Challenge is an international space and robotics competition that aims to give the contestants the opportunity to propose and test a robotic system for resource reconnaissance on the Moon in order to enable future exploitation of extraterrestrial materials.

The first round

Back in November, 2021, 12 teams from Europe and Canada competed with each other on an artificial, specially prepared surface of the Moon during the first stage of the Space Resources Challenge that was held in the Netherlands. The competitors had to face various demanding tasks including navigating their rovers around obstacles, analyzing rocks, searching for water and other valuable resources. Of the 12 contestants, the 5 following teams proved to be the best and moved on to the next stage of the event:

  • Mission Control (Canada)
  • Glimpse from ETH Zürich & University of Zürich (Switzerland)
  • FZI Forschungszentrum Informatik (Germany)
  • Łukasiewicz–PIAP (Poland)
  • LUVMI-XR (Belgium/Luxembourg/UK/France)

Read more about the first round of the challenge here.

The second round

The second, final round of the Space Resources Challenge took place from 5 to 9 September 2022 in Luxembourg. Although similar to the previous stage in terms of the tasks, the finals set the bar higher.  

The whole arena of the challenge was covered with very fine basalt sand with some areas of materials hidden in the surface for the contestants to identify. On the competition site, there were also craters and boulders. The competing teams’ task was to analyze and identify the material of the lunar rocks, as well as establish their positions on the map.

Here's an official video of the Space Resources Challenge final round:

During the final stage, ESA encouraged the competitors to take a multi-robot approach. The LUVMI-XR team used three Leo Rovers and the LUVMI-X robot from Space Application Services (SAS).

The LUVMI-XR team with their robot fleet
source: www.esric.lu

Leo Rovers’ setup for the finals

Following the first round of the contest, the LUVMI-XR team upgraded their Leo Rovers to a newer version of LeoOS – 1.0 – that’s based on Ubuntu 20.04. and ROS Noetic. “We needed to make sure that our previous modifications were still working and they needed to be rewritten partially. Once this was set up, we could focus on the Nvidia Jetson computer that we had mounted on top of the rover.” – the team says. For the first challenge, they used two Leo Rovers – one equipped with Nvidia Jetson Xavier NX and the other with NVIDIA Jetson Nano. However, the team felt that the latter was insufficient for their objectives, so they replaced it with another Jetson Xavier NX, which they also mounted on a third Leo Rover they’d prepared. Ultimately, for the Space Resources Challenge finals, LUVMI-XR ended up with three Leo Rover mobile robots, each with LeoOS 1.0 running on a Raspberry Pi and with Nvidia Jetson Xavier NX on top.

Because the team encountered some difficulties with multi-robot interaction in ROS 1 during the first stage of the competition, they went for ROS 2 in the finals, although none of them had had experience with it before. As the team claims: “It was a risk that we were willing to take in order to leverage the advantages of ROS 2. Its main advantage is the simplicity to add several robots in the same robotics network without having to deal with multiple ROS masters. This makes the entire setup much easier and scalable.”

In order to perform visual SLAM in the first round of the Space Resources Challenge, LUVMI-XR used RTAB-map with an RGB-D sensor and the IMU data. They decided to stick to the RTAB-map for the finals, but with extra sensor data. The team had made custom mounting plates that they equipped two of their Leo Rovers with and added 2D LiDAR sensors to them as well. Although the sensor is not exactly cut out for space use as its moving parts could be easily damaged in a rocket launch, by using it, LUVMI-XR intended to augment the reliability of their SLAM system with LiDAR data.

visual inputs
Visual inputs via Intel RealSense used for SLAM (left) and via internal camera connected to the Raspberry Pi with low bandwidth visual feedback intended for manual navigation (right)
credit: the University of Luxembourg

During the first round of the Space Resources Challenge, the Leo Rovers’ standard internal batteries capacity was enough to power the robots along with the additional hardware mounted on them for the whole 2.5 hours of the competition. However, knowing that the final round was going to last twice as long, LUVMI-XR decided to boost the robots’ battery capacity to make sure that their Leo Rovers would be able to run for the entire contest. For this purpose, they used Makita 18V batteries that served as the power source for the Jetson, the router, antenna and camera, increasing the internal battery’s capacity in the process as well, which translated to the robots being operational for throughout the challenge. This proved to be a wise decision, especially given the fact that some of the competitors’ rovers turned out incapable to run this long.

Leo Rover setup for the Space Resources Challenge
One of the Leo Rovers’ set up with LiDAR and external battery
credit: the University of Luxembourg

To control their Leo Rovers in the first round of the competition,LUVMI-XR used two laptops connected separately to each robot, and with RViz as the main operation tool. For the finals, the team switched to Foxglove – an open-source tool that provides a wide array of features to visualize and debug data. Based on it, the team created a custom panel, thanks to which they were able to operate each of their three Leo Rovers. Given the tool is modular, if needed, LUVMI-XR could easily switch back to RViz. What’s more, the new UI enabled them to control their Leo Rover fleet from a single station. And thanks to ROS 2, operating multiple robots in the same network eased things a lot.

Foxglove custom UI panel
UI based on Foxglove developed by the Luxembourg team
credit: the University of Luxembourg

What’s more, as the team shared with us, “instead of using the MikroTik antenna to connect to an access point, we used it to create our own ad-hoc network, a mesh network, that allowed us to have all robots to be connected directly to each other in a decentralized manner. Additionally, each rover could then act as a relay to extend the range of the network if necessary.” While such a solution isn’t much of an advantage for terrestrial applications given there’s a good network infrastructure on Earth, the team is working on a space scenario, one that more closely reflects the way a multi-robot system would work on the Moon, where no network infrastructure exists.

The winner of the Space Resources Challenge 2022

The ultimate winner of the Space Resources Challenge has yet to be announced, so stay tuned!

The winning team will be awarded with funds as development contracts from ESRIC and ESA and will have the possibility to embark their technology on the Moon for the upcoming missions.

Discover possibilities of Leo Rovers

We’re incredibly proud and happy that the Leo Rover robots could commit to this excellent event that is the Space Resources Challenge! But this robotic platform has many other various applications under its belt, not only “extraterrestrial” like this one. 

Here, you can see how other of our customers have been using their Leo Rovers. And maybe you can find a use for this robot that no one has thought of yet ;).

Want to see more posts like that?

Subscribe to stay informed.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

See more blog posts:

<- get back to the Blog