In agriculture, areas are increasingly cultivated in a targeted and selective manner, right up to the specific care of individual plants in the context of precision farming. To aid this, accurate information about current field conditions, such as soil and plant parameters, can be obtained from mobile field machines or stationary equipment. Here, robotics plays a key role.
Fraunhofer IPA develops and offers agricultural robotics solutions for undefined and dynamic environments, ranging from navigation software for field swarm cooperation to machine vision and new small-scale service robot concepts. The goals of each project are to achieve cost-effective final results and to transfer the developed technologies to applications.
The focus is on developing small service robots that are networked with each other and with drones. This combination creates a collective intelligence capable of making logical decisions or linking data in a meaningful way. It is hoped that such information will help all relevant actors in agriculture to make decisions, for example on the needs-based use of pesticides.
Artificial intelligence is a key technology in this regard: with the aid of machine vision and neural networks designed for fields and plants, relevant information is generated and used for later processes.
The COGNAC project is bringing about trend-setting developments: the complex and highly-diverse agricultural data is recorded automatically and processed with the aim of generating benefit. Service robots for agriculture combine information with sensor platforms and specific sensor systems, which is then analyzed for immediate use by the data infrastructure of the "Agricultural Data Space" and made available to farmers.
An ongoing, successful project is "AMU-Bot": in the context of sustainable crop protection methods, the affordable, autonomous platform mechanically removes inter- and intra-row weeds. The service robot uses optical sensor technologies to detect crop plants and remove the surrounding weeds.