FIRA2020 — lively updated notes
Here we go! The 5th Edition of the International Forum of #AgriculturalRobotics has gone live this morning. Three days to learn, exchange and see demos about future-ready agricultural equipment.
Last year I decided to share my live notes, and it was a nice way to focus on the main ideas and outcomes of the multiple exchanges
https://agronoter.wordpress.com/2019/12/10/fira19-live-notes/
Due to the pandemic, the 2020 FIRA edition is completely virtual and it will be even much easier to share key points from convenors and exhibitors.
Main stage
The French Ministry of Agriculture addressed his speech to the International Forum participants. France is proving to be at the forefront of innovation for agricultural robotics, thanks to FIRA and RobAgri association.
In the introductory plenary, Josef Kienzle from FAO highlighted a hub of dedicated resources about sustainable agricultural mechanization. Among them, the highly relevant and recent report titled Agriculture 4.0 — Agricultural robotics and automated equipment for sustainable crop production
Adopters feedback
Labour shortage is just one side of the reasons justifying agricultural robots uptake. The flip-side is training and monitoring skills needed. This is what is being pointed out by Chuck BARESICH (Haggerty Creek Ltd.) and Joaquín REYES GONZÁLEZ (Wine Fields of Ribera del Duero) during the plenary session about vineyards. They share their feedbacks as early adopters of RoamIO (by Korachi) and other agricultural robotic solutions. They point out that the main advantage and counterparts are the wide room for customization to their specific needs and the possibility to build reciprocal trust.
Robot loving vs robot bashing round table
Robots replace tasks, especially repetitive tasks, not work. The first advantage is to free farmers’ time for agronomy and spare time for their family. Indeed, robots can help to overcome the lack of trained human power during critical agricultural periods. Even better, robots enable the farming operation to be scheduled ahead, eventually freeing farmers also from uncertainty and mental load (Green).
The environmental impact of new technologies, namely related to the use of rare earth elements, must be correctly framed. Is this impact to be compared with past technologies or with similar contemporary or upcoming ones? The real issue is not in the use of these raw materials, which is being reduced compared to classical agricultural machinery; instead, the challenge is to find processes to allow their reuse.
Cooperatives and contractors may be a good alternative for small farmers to support their access to robots and most advanced agricultural technologies (Azevedo). In addition, robots help to have technical demanding tasks accomplished at the right time, especially during harvest periods where labour shortage is most affecting farming.
Roughly 2–3 Terabytes are produced per day per field by agricultural robots. This data deluge may help to prove to the consumer the sustainability of farming practices. Farmers are _the_ data owners. Yet, decision support systems and artificial intelligence are needed to ease the decision-making. In this perspective, two consortia are promoting the coordination among actors in the French agricultural and food sectors.
Once the agricultural robots will start to deliver the expected performances, the farmers /who are curious/ will surely learn how to master them (Azevedo). Training & usability are though a clear competitive parameter for robotic manufacturers: decluttering from irrelevant tech options will be crucial (Green). Freeing farmers and agronomists from homogeneous and repetitive tasks will ensure they can focus on the interpretation of variability /what humans are best at/. So, it will be better for farmers to stick to agronomy instead of learning robotics (Green). Though, the priority in learning and acquiring electronics, robotics and similar skills will depend upon each farmer’s individual business model (Azevedo).
RobAgri Scientific Workshop
Very relevant introduction by Christian Huygue (INRAE). He highlighted the opportunity from agricultural robotics to enable agroecology transition on several aspects. First, the data augmented farming possible achieving preventive plant protection, thanks to the continuous multiple data collection, which already proved to be more effective than human capabilities to spot plant diseases and pests. Second, the possibility to manage highly spatial and temporal precision in farming practices. Third, enabling diversification of farming practices and cropping systems thanks to the modularity of implement size.
Agricultural robotics is mainly driven by the labour shortage, due to the steady reduction of the economically active population in agriculture. Nonetheless, robotics might also contribute to enhancing the attractiveness of the agricultural sector, otherwise and generally perceived as not attractive and based on painful and repetitive actions.
What might be the key topic in the training of the operators to ease the uptake and mastery of agricultural robots (or autonomous equipment)?
Operators must be trained to support and enable the co-design of robotic solutions from the very beginning. As so, they have to have skills beyond mechanics, namely in agronomy, biology as well as in sociology to properly handle societal expectations.
Farming with no tractor driver, is it possible?
Law and regulations are under analysis in the Wednesday session. Andrea Bertolini (Scuola Superiore Sant’Anna) provided a summary of legislation regulating autonomous solutions. The liability for damages is currently on the operator and, in a second stage, to the machine provider. Altogether the legislation always covers new technologies. The question is rather how related damaged should be regulated.
Insurances need a market and a defined perimeter of risk. For the moment there are not enough data for both of them. A study from 2016 in the US found that we were not yet able to affirm that automatic vehicle will be more sure than the current vehicle. For instance, hacking may add insecurity. The solution would be to identify a single source for uncertainty, such as the user as the only source of litigation. This might allow shifting cost from ex-post litigation to ex-ante payment of insurance. Nonetheless, we are still far away from one fitt-all solution.
Christophe Gossard (John Deere) addressed the ongoing proposal for relying on architecture are under development to use data to provide proof of reliability and compliance to more or less precise or clear. The revision of the machinery directive has problems following the fast pace of technology innovations. Presumption of conformity, technical files produced, and similar approaches might help to deal with the lack of precise legislation.
Demos and pitches
News from Australia about field crop robotics with SwarmFarm. In the Q&A session, Andrew Bate is clarifying that training is increasingly addressed through support materials for operators and field demos directly with farmers.
Wandering across the booths, I’m picking and updating a list of well-known recently launched agricultural robots. An early list is available in this Twitter thread https://twitter.com/pievarino/status/1336258025083052034
Save the dates
Wandering across the booths I’ve learned about the agROBOfood Industrial Challenges OpenCall Webinar for SMEs “FIRA Edition” 🗓️Thursday, Dec 10, 2020 | 11:00 AM — 12:30 PM. This is the DIRECT LINK
Originally published at https://www.linkedin.com.
