A novel adaptation strategy for forage production under wet growing conditions - robotization and high quality forages

The main aim of the GrassRobotics project was to establish a sustainable production system for forage production using a lightweight robot with cutting, collecting and transportation tools, combined with novel seed mixtures for frequent harvests and a wetter, longer growing season. The project examined the benefits of using a lightweight robot with harvesting tools to cut and collect grass, and tested multispecies seed mixtures for frequent harvesting and increased feed quality. The project estimated animal production response and calculated the economic performance with respect to forage costs and profit by increasing the number of harvests per season. The project also studied the environmental effects of lightweight equipment and multiple harvests on soil compaction, root development and nutrient uptake. At the start of the project, two different experimental fields were established at NIBIO Fureneset. One field was used for testing yield, forage quality and soil compaction while the other was used as a test-driving field for the grass harvesting robot. Through three harvesting seasons, the researchers at NIBIO Fureneset conducted two different harvesting regimes on the main experimental field: One area was harvested three times and compacted with traditional equipment, while the other was harvested five times and compacted with the lightweight robot. Soil cores from the experimental field were sampled throughout the season and analysed for soil physical parameters. Soil moisture and soil temperature were continuously measured by sensors installed at two depths on the experimental field, while penetration resistance was measured before and after compaction. Dry matter yield was measured at each harvest, and chemical analyses were carried out in order to assess the level of nitrogen content. Laboratory silos of grass material samples were stored for three months, then analysed for nutritional content and forage quality. In order to monitor the plant development of the two different seed mixtures, plant samples were also taken for sorting the grass species. A number of field trials and demonstrations of robotised grass cutting, harvesting and transportation were carried out by the project team, both at NMBU in Ås and at NIBIO Fureneset. Two Thorvald robots were made available to the project by project partner Saga Robotics and were modified for grass harvesting purposes. One of the robots was mainly used in soil compaction trials at NIBIO Fureneset, while the other was used for testing cutting and harvesting tools. The robots were also used by master students at NMBU. The project's PhD fellow used the robots to test a safety system developed in collaboration with researchers at NMBU and the University of Lincoln in the UK. Tools for collecting and transporting the grass after cutting were developed in collaboration between researchers and engineers at NMBU and project industry partners HMR Voss and Orkel. The GrassRobotics project has reached the stage where we have shown a proof-of-concept. The next step is to make this into a fully commercial product. This is not within the scope of the project, but a preferred outcome of the project. The project has concluded that the concept is mechanically viable, and that it makes sense from an agronomic and economic point of view. The next step will be to find a commercial company to take the concept to the level where we can deliver robust and reliable robots for the entire industry.

The GrassRobotics project has shown that from a mechanical, biological, operational and agricultural point of view, it is viable to cut and collect forage up to five times in a season using autonomous robots. This has never been shown before at this scale. The project serves as a basis for taking this idea further and develop it into a fully commercial product. The main scientific contributions of the project is the development of a safety system for navigation in the vicinity of humans. We have also developed several devices for attaching cutting and collection mechanisms to the robots. Furthermore, we have shown that it is economically viable to use small autonomous robots for this task. A final impact of the work if the solution is taken to a fully commercial product is the potential reduction of imported feed. It is a stated goal to reduce this and this can have a significant impact on the environment.

A novel adaptation strategy for forage production under wet growing conditions - robotization and high quality forages (GrassRobotics) is a 4 year interdisciplinary project aiming to develop and test an equipped robot for cutting, collecting and transporting forage grass to a stationary bailing site. The project will increase the knowledge base for forage production substantially with a new approach of increasing the efficiency of forage production. Due to wet growing conditions and commonly used heavy tractors and bailing equipment, the time windows for ley harvesting are shortened during the last decades, resulting in forages of lower forage nutritive quality than necessary. Also, soil compaction and less optimal conditions for plant growth is a consequence in a situation where the dry matter yield in grassland is lower than expected from the plant nutrients supplied. A novel agricultural robot will be developed to work specifically for the application at hand. The project will during the three ley years reveal the impacts of a lightweight harvesting robot in comparison with conventional heavy tractor and bailing equipment on plant growth, soil conditions and cost efficiency. Compared to conventional equipment used for bailing, frequent robotic harvesting is expected to increase forage quality, plant growth and the root:shoot ratio. Since lightweight robots are expected to reduce soil compaction, the time window for ley harvesting will increase, and so will the dry matter and nitrogen yield. The cost benefit of improved feed quality with robotized harvesting of forage will be assessed to see how robotized harvesting reduces the production of cost of silage, the energy use in silage production and the need for concentrate in the ruminant diet. Results will be disseminated with the intention to show how this lightweight harvesting robot works in practice and with regular demonstrations and discussions with farmers and the agricultural industry.