Center for Digital Agriculture




Overview: Crop farming plays an essential role in our society, providing us food, feed, fiber, and fuel. We heavily rely on agricultural production but at the same time, we need to reduce the footprint of agriculture production: less input of chemicals like herbicides, fertilizer, and other limited resources. Agricultural robots and other new technologies offer promising directions to address key management challenges in agricultural fields. To achieve this, autonomous field robots need the ability to perceive and model their environment, to predict possible future developments, and to make appropriate decisions in complex and changing situations. This talk will showcase recent developments towards robot-driven sustainable crop production. Dr. Stachniss will illustrate how management tasks can be automized using UAVs and UGVs and which new ways this technology can offer. 

Speaker: Dr. Cytill Stachniss, University of Bonn
Date: February 2, 2022


Overview: Precision Livestock Farming (PLF) is a tool for management of livestock by continuous automated real-time monitoring of production/reproduction, health & welfare and environmental impact. A tool means that this technology does not replace experts like farmers, veterinarians, feed experts, etc. but that PLF supports people in their decision taking by objective measurements on the animals. This contribution gives an overview of where we are today and what the proven potential is of this technology. We use results from the work we did since 1991 and the literature on this topic published in peer reviewed publications and in several conference proceedings.

When starting this research on insects and mussels, it soon became clear that animals, like humans, are so called C.I.T.D. systems: Complex, Individually different, Time-varying in their responses and Dynamic. Then we did experiments on bees, fish, mice, rats, chicken, pigs, cow, horses, dogs to from 2001 also work on humans. Results are shown in videos and graphs. The research trajectory has given principles on how to develop the technology and to implement it in products. We will discuss some potential issues and business models. The pickup in the field however goes far too slowly and that is where we must put more efforts.

Speaker: Dr. Daniel Berckmans
Date: April 6, 2022


Overview: Advancements in highly-automated agricultural machines over the last decade have generated excitement in the industry and in the general public. Major agricultural equipment companies have used national and international events to generate attention about their latest designs. U.S. and international companies tout benefits of labor savings and production efficiency. Another widely promoted idea is the ability of new highly-automated technology to improve operator experience by improving safety and ergonomics while reducing the cognitive burden of repetitive and boring tasks. In all areas where automated or autonomous machines replace older technologies (including motor vehicle design and deployment), safety is a major barrier in the pathway to adoption. Novel technology is difficult to understand among stakeholders such as insurers and regulators who focus on safety. New engineering design standards have been created in the past five years, but these tend to use older risk-assessment methods predicated on having a rich base of historical incident (accident) data. Such data does not yet exist. These standards and documented safety efforts sometimes focus on obstacle detection whereas historical data suggests that operator and bystander farm injuries often occur in situations where the hazard is not easily detected by a human operator and while equipment is being operated in suboptimal conditions or is being repaired or maintained. New engineering standards also sometimes exclude other important hazards and outcomes in their scope such as environmental damage, harm to structures and civic infrastructure, risk to livestock, and risk assessment that includes the economic impact of downtime. This presentation will explore many of these issues. Data from an industry survey of practicing engineers will show that new risk assessment methods must be developed along with datasets that might be used in training and informing software used on these machines. An analysis of 434 fatal non-automated machine events will show that even highly experienced operators are challenged by the task of “hazard detection” which has implications related to dependence on sensors, AI, and other software to handle functions of safety. A list of further research and development needs will be presented for discussion.

Speaker: Dr. John Shutske, University of Wisconsin-Madison
Date: May 4, 2022