RoLand
Agricultural Robotics
The aim of the RoLand project is to develop a semi-autonomous, mobile harvesting system that can pick fruit independently in open fields. Unlike other systems, which are primarily designed for use in greenhouses, the robot developed in the project is specifically designed to meet the challenges of outdoor cultivation, such as weather conditions and uneven terrain. The versatile and cost-efficient system is also suitable for smaller farms. The work output can be flexibly scaled by operating several compact units in parallel.
| Duration: | 01.10.2021 till 30.09.2025 |
| Donee: | German Research Center for Artificial Intelligence GmbH |
| Sponsor: |
Federal Ministry of Food and Agriculture (BMEL)
Bundesanstalt für Landwirtschaft und Ernährung |
| Grant number: | The project is supported by funds of the Federal Ministry of Agriculture, Food and Regional Identity (BMLEH) based on a decision of the Parliament of the Federal Republic of Germany. The Federal Office for Agriculture and Food (BLE) provides coordinating support for artificial intelligence (AI) in agriculture as funding organisation, grant number 28DK103A20. |
| Partner: |
HAW Hamburg – FTZ Smart Systems |
| Application Field: | Agricultural Robotics |
| Related Robots: |
SHIVAA
Strawberry Harvester: an Innovative Vehicle for Application in Agriculture
ARTEMIS
DLR SpaceBot Cup 2013 Rover
ASGUARD IV
Advanced Security Guard V4
|
Project details
The SHIVAA robot was developed as part of the RoLand project to specifically promote automation in sustainable agriculture. It is capable of harvesting fruit in open fields completely autonomously, thereby promoting natural cultivation that enables high-quality and ecological end products.
Another goal of the project is to improve supply security. The system can reliably compensate for short-term absences of human harvest workers, ensuring that no ripe fruit is left unused in the fields. This makes the harvest less dependent on personnel, weather conditions, and time of day. In this way, the technology contributes to securing Germany as a cultivation location in the long term, strengthens the regional cultivation of numerous types of fruit and vegetables, and improves the carbon footprint through shorter transport routes.
SHIVAA was specially designed for fully autonomous strawberry harvesting in open fields. With the help of a 3D stereo camera, the robot independently follows a row of strawberries. Additional cameras, including multispectral sensors in the invisible light spectrum, precisely detect the position and ripeness of the fruit. Image processing is based on modern machine learning methods that ensure reliable detection and selection of strawberries. A gripping mechanism gently picks the fruit directly from the plant and carefully places it in a harvest box mounted on the robot.
SHIVAA is designed to integrate seamlessly into existing agricultural operations and work closely with human staff, despite the many challenges of outdoor work. A particular advantage is its nighttime operation: cooler temperatures protect the fruit, while artificial lighting creates ideal conditions for precise image processing. This significantly improves harvesting precision and quality – and the fresh produce is ready for delivery in the morning.
SHIVAA – Technical highlights
- Size: approx. 2.45 x 1.20 x 1.00 m
- Weight: approx. 150 kg
- Drive: 4 electric wheel hub motors (BLDC), max. 6 km/h
- Power supply: 48V Li-ion batteries, 2x 29 Ah
- Sensors: 3D stereo camera, multispectral camera, GPS/IMU
- Gripping technology: Strawberries are sucked in and gently enclosed by the gripper
Videos
RoLand: Agricultural Robotics
The aim of the project is to design and develop a semi-autonomous, mobile system that is capable of harvesting fruit independent of human interaction. While robotic systems are currently used mainly in greenhouses, the target scenario of the proposed project is the open field, taking into account the typical environmental influences associated with it. Taking into account a wide range of possible applications and a low investment volume, the aim is to develop a system that can also be operated economically by smaller farms. An adaptability of the work performance to the farm size-dependent demand is then given by the number of small systems working in parallel.