Learning Ground Interaction Models to Increase the Autonomy of Mobile Robotic Exploration Systems

In the project, data from different robots on different soils will be recorded in order to learn soil interaction models with them. These will be integrated into a simulation environment and then used for autonomous adaptation of path planning or locomotion to increase the safety and efficiency of the robots.

Duration: 01.02.2022 till 31.01.2025
Donee: German Research Center for Artificial Intelligence GmbH
Sponsor: Federal Ministry for Economic Affairs and Climate Action
German Aerospace Center e.V.
Grant number: 50RA2122
Application Field: Space Robotics
Related Projects: RoBivaL
Robot Soil Interaction Evaluation in Agriculture (08.2021- 10.2023)
Environment Modelling and Navigation for Robotic Space-Exploration (10.2014- 12.2017)
Exploration in terrain difficult to access (e.g. Valles Marineris) using visual and proprioceptive data. (05.2015- 06.2018)
Learning Intelligent Motions for Kinematically Complex Robots for Exploration in Space (05.2012- 04.2016)
Semi-autonomous cooperative exploration of planetary surfaces including the installation of a logistic chain as well as consideration of the terrestrial applicability of individual aspects (05.2013- 12.2017)
Related Robots: CREX
Crater Explorer
Multi-legged Manipulation and Locomotion System
iStruct Demonstrator
DLR SpaceBot Cup 2013 Rover
Coyote III
Related Software: MARS
Machina Arte Robotum Simulans
Behavior Optimization and Learning for Robots
Robot Construction Kit

Project details

Artemis representing a rover with passive suspension (Photo: DFKI)
Charlie representing a four-legged waling systems (Photo: DFKI)
Coyote III representing a rover with legged wheels (Photo: DFKI)
CREX representing a small six-legged waling systems (Photo: DFKI)
Mantis representing a big six-legged waling systems (Photo: DFKI)
NoStrandAMust is a project with the goal of experimentally investigating the ground interaction of different robots and generating models for the interaction using AI methods. These models are to be used and verified in simulation and to improve the autonomy of the robots. On the one hand, the models will be used for motion planning and thus increase the efficiency and safety of the systems. On the other hand, the models will be used to analyze the current ground conditions at runtime and thus prevent the systems from maneuvering into dangerous areas. Thus, the methods, models and simulations to be developed in NoStrandAMust increase the autonomy, safety and efficiency of mobile robot systems. The methods to be developed are thus seen as an essential contribution to the success of future missions.

In order to achieve the overall goal of the joint project, the autonomy of robotic systems by ground interaction models, four essential contributions are made in NoStrandAMust four essential contributions are made:
 - Different systems here refers to different locomotion modes, propulsion systems, weight classes, and contact geometries and materials.  Data will be collected from six walking systems with different morphologies and foot designs, as well as data from different wheel-driven concepts, each of which will be tested on four different surfaces under circumstances with different locomotion behaviors.
 - state-of-the-art regression and classification methods as well as other machine learning methods are required for the generation of soil interaction models. In particular, preliminary work with artificial neural networks has shown that they are very well suited to represent models comparable in complexity to those required in this project.
 - By integrating the models into a simulation environment, they can be evaluated based on the holistic robot behavior and compared with the behavior from the data acquisition of the real systems. The goal is an accurate real-time simulation that can be used for system design as well as path planning.
 - by continuously matching robot performance with a prediction generated from a simulation running on the robot, the systems will be able to continuously determine soil conditions and detect deteriorating conditions at an early stage. The robot should react accordingly to changing soil conditions, for example by triggering new path planning or appropriate behavior adjustments.

Since the data-driven approach means that the soil interaction models have to be trained on a system-specific basis, it is not possible to generate a generally applicable model in NoStrandAMust, as is possible with analytical models. Nevertheless, in order to demonstrate the general applicability of the method, six different systems will be used in NoStrandAMust. These include three walking machines (Charlie, CREX, and Mantis), two wheel-driven systems with different concepts (SherpaTT, and Artemis), and a mixture between walking and wheel systems in the form of a so-called star wheel (Coyote III). In addition, the weight class of the systems ranges from about 20 kg to about 120 kg for both the running systems and the wheel concepts. Each of the six systems is tested on substrates with different properties. In addition to a distinction between loose sedimentary soils and hard subsoils, there will also be different variants within these two classes. It is envisaged here that the ''loose sediment'' class will include fine sand, fine gravel, coarse gravel or basalt split. For hard soils, the systems will be used to record data that have different friction coefficients, such as dust-covered surfaces or high friction as found in lava caves.


ARTEMIS: First test run with a penetrometer

The rover Artemis, developed at the DFKI Robotics Innovation Center, has been equipped with a penetrometer that measures the soil's penetration resistance to obtain precise information about soil strength. Such measurements allow for conclusions about the current soil condition, applicable in both agricultural settings and space exploration. The video showcases an initial test run with the device mounted on the robot. During this test, the robot was remotely controlled, and the maximum penetration depth was limited to 15 mm.



Ground Interaction Models for Increased Autonomy of Planetary Exploration Systems
Alexander Dettmann, Malte Langosz, Jonas Eisenmenger, Marc Otto, Sebastian Kasperski, Malte Wirkus, Nayari Lessa
In 13th EASN International Conference on Innovation in Aviation & Space for opening New Horizons (13th EASN 2023), (EASN), 5.9.-8.9.2023, Salerno, IOP Publishing Ltd, series Journal of Physics: Conference Series, volume 2716, Mar/2024.
Development of a leveling and loosening mechanism for fine sediments on a test track for planetary robots
Jonas Eisenmenger, Jonas Benz
In 13th EASN International Conference on Innovation in Aviation & Space for opening New Horizons (13th EASN 2023), (EASN-2023), 5.9.-8.9.2023, Salerno, IOP Publishing Ltd, series Journal of Physics: Conference Series, volume 2716, Mar/2024.

Back to the list of projects
last updated 04.01.2024