Regensburg Technical University of Applied Sciences (OTH)

Unveiling Soil Biodiversity

The ELISE Project Uses Integrated Mechatronic Systems to Revolutionize Soil Assessment in Agriculture
Author: OTH Regensburg,

One of the key components for successful plant growth is soil biodiversity. The ratio of fungi to bacteria plays a particularly important role here, as it significantly influences plant growth and resilience to pests and diseases.

Until now, this ratio could only be determined through chemical analysis. The research project ELISE (Electronical Lab for Intelligent Soil Examination) has changed that. Hermann Ketterl, Professor for Measurement and Control Technology at the Faculty of Mechanical Engineering of OTH Regensburg and his team developed an innovative method to precisely measure the fungus-bacteria ratio without the use of chemical processes. The project was realized in close cooperation with Digital Workbench GmbH from Wettstetten and Agrarservice Hägler from Wernberg-Köblitz.

The ANELISE robot performing automated sampling in the field.
Base station where the collected samples are processed.

Soil Sampling Tracked with GPS Accuracy

Ketterl and his team designed a fully automated workflow to capture and analyze soil samples with remarkable precision: First, an autonomous robot drives over the study area and takes soil samples at predefined positions. These are then transferred to a base station, where the samples are automatically sieved. After that, a granule is mixed with water to form a defined suspension, applied to a microscope slide, and automatically passed on to the microscope where the sample is analysed.

  • The microscope captures the sample by gradually moving the focal plane through the depth of the specimen while recording a video with a camera.
  • Spatial representations of the sample are generated from the recorded video data using focus stacking. Thread-like fungal hyphae can be identified in these representations, which are then automatically recognized and analyzed using semantic segmentation methods.
  • In addition to analyzing the fungal hyphae, moving bacteria are also detected and counted using a blob detection algorithm. Blob detection is an image analysis method that can be used to identify particularly distinctive structures or objects in digital images. Two approaches can be used to quantify the bacterial biomass, depending on the bacterial density. At low bacterial densities, individual bacteria in a section of the field of view are identified and counted using an image recognition algorithm. The number determined is then extrapolated to the total soil mass
  • When bacteria density is high, specific regions of interest are selected in which only bacteria are present. A neuronal network that has been pre-trained with images of different bacteria density patterns then predicts the total number of bacteria in these areas.

Better Cultivation Methods Possible with ELISE

By developing a functional model as part of the project, it was possible to demonstrate that sample preparation and analysis can be carried out fully automatically with the robot. Since the robot can repeatedly visit the exact same sampling points with GPS accuracy and record a large number of sample points, statistical evaluation of the data is possible. This enables farmers to monitor the soil independently and develop cultivation methods and fertilization strategies that take soil biology into account and are adapted to the respective local and operational conditions.

Finding Your “Personal Niche” in Research

Under the guidance of Ketterl, Tobias Heinrich, a research assistant at OTH Regensburg, began his work on the project. Heinrich had first studied mechanical engineering in Regensburg, earning a bachelor's degree. It was during his Master’s in Industrial Engineering that he came into contact with Ketterl and thus with the ELISE project: Heinrich had found his "personal niche", a subject area that both inspires and moves him. "A country like Germany, which is too strongly influenced by business economics, loses its ability to produce its own goods over time," he is convinced. "We must give the soil time to regenerate and at the same time act more in harmony with nature."

I want a world where natural resources are used with humility and not for economic gain.
Tobias Heinrich, research assistant, OTH Regensburg

To ensure sustainable and regenerative agriculture, soil—which also serves as a major CO₂ reservoir—must receive greater attention and a stronger focus within the scientific community. According to Heinrich, the OTH Regensburg is making an important contribution in this area through its research activities.

In his academic work within the ELISE project, Heinrich found exactly what he believes researchers need at the beginning of their academic career: a topic that sparks personal enthusiasm. He also emphasized the excellent and supportive collaboration with Ketterl. The project leader's openness to input has greatly benefited him personally.

From ELISE to SPIN-FERT – Collaboration Continues

Following the conclusion of the ELISE project, Ketterl and Heinrich continue to work together. In 2024, the EU project SPIN-FERT (Innovative practices, tools and products to boost soil fertility and peat substitution in horticultural crops) was launched. 

In this project, the researchers are working on integrating agro-industrial waste to enhance soil fertility, once again applying innovative approaches such as AI-supported soil assessment tools.

Back to top Icon