Robots With Different Abilities Can Succeed In Mars Exploration

The exploration of Mars can best succeed when robots with different abilities work together – preferably autonomously. Important prerequisites for this are being created as part of the “VaMEx-3” project, in which three working groups from the University of Bremen are significantly involved.
The surface of Mars poses a challenge for science: the most interesting discoveries are expected to be made in those places that are most difficult to access due to the rocky and rugged terrain. In order to be able to examine them, robots with different abilities have to cooperate: walking, climbing, flying – the transport of payloads and the energy supply must also be secured. The VaMEx-3 project, which is funded by the German Space Agency at DLR with funds from the Federal Ministry for Economic Affairs and Climate Protection ( BMWK ), enables cooperation between different robot systems that are intended to land on Mars one day.

The Computer Graphics working group at the Technology Center for Computer Science and Information Technology (TZI) at the University of Bremen, headed by Professor Gabriel Zachmann, the Cognitive Neuroinformatics working group led by Professor Kerstin Schill and the High-Performance Visualization (HPV) working group led by Professor Andreas Gerndt are all centrally involved. The VaMEx-3 project consists of four parallel sub-projects with partners from all over Germany. It has a total funding volume of around five million euros.

Simulation of Mars exploration in real time
In previous projects, the TZI scientists have already virtually reproduced around 40 square kilometers of the surface of Mars based on scans from NASA in order to set up a test environment for the required technologies. The project partners then exposed digital versions of their robotic systems to the real Martian landscape in this virtual twin.
“Until now, each swarm member acted largely on their own,” reports project coordinator Dr. René Weller from the computer graphics working group at the TZI . “Now it’s a matter of bringing them together cooperatively.” The different systems are to be enabled via interfaces to interact in real time – and to a large extent autonomously.

The test environment must meet the highest standards: “The virtual twin must enable realistic statements to be made that the swarm will function in the same way on Mars in the future,” emphasizes Professor Zachmann. The test environment must also show existing weaknesses, for example if the recognition of certain objects does not yet work sufficiently with a robot. It is a challenge for IT to simulate all the vehicles quickly enough and – among other things – to reproduce the camera images and lidar scans in real time.

Navigation without GPS, Galileo and paved paths
While the Computer Graphics working group is focusing on the further development of the test field, the Cognitive Neuroinformatics working group is leading the ” Robust Ground Exploration ” sub-project. A central point is the development of a common navigation method, because on Mars the terrestrial satellite navigation does not help. The software for this is being developed at the University of Bremen. In addition, the robustness of the systems should be increased in order to withstand the difficult environmental conditions and deal with unexpected situations.

“One challenge is that the environment is partly unknown in advance,” explains Dr. Joachim Clemens, who coordinates the sub-project. “Therefore, the swarm members have to recognize obstacles independently, create a map of the environment and estimate their position on the map. The units cooperate with each other: map and position information is exchanged so that all units can benefit from it. The swarm participants then use this information to plan and coordinate further action.”

Another aspect of the sub-project is the development and integration of a mission control tool. On the one hand, this system should enable the visualization of the current status of the mission and, on the other hand, enable the scientists to communicate with the VaMEx swarm. The information transmitted to the swarm, for example about scientifically relevant target areas, is incorporated into the autonomous planning of the system and taken into account in the further course of the mission. The mission control tool is being developed by the High-Performance Visualization working group at the University of Bremen.

In search of water and signs of life
A larger demonstration campaign is planned in three to four years to test the robot swarm extensively in a Mars-like test area. The long-term goal of the VaMEx mission is to explore the Valles Marineris canyon system on Mars in order to find evidence of water resources and biological traces from climatically more habitable epochs on Mars. The “Mariner Valleys”, named after one of NASA’s first Mars space probes, form the largest network of canyons in the solar system with an extension of 4,000 kilometers and a depth of up to 10,000 meters in places.

In addition to the University of Bremen, the following partners are also involved in VaMEx-3: ANavS GmbH, German Aerospace Center (DLR), DFKI Robotics Innovation Center , DSI Aerospace Technologie GmbH, INVENT GmbH, TU Braunschweig, TU Munich, University of Erlangen- Nuremberg, University of the Federal Armed Forces Munich and University of Würzburg.