AI and Machine Learning

Artificial Intelligence (AI) is the study of how to code computers and machines to mimic human decision-making processes and tasks with greater speed, accuracy and lack of bias. Machine learning (ML) is when computers are given access to huge sets of training data (Big Data) to automatically look for patterns and learn and improve from experience using processes involving algorithms (coded sets of rules to process and model Big Data). These algorithms can be fully supervised, partly supervised or unsupervised. ML processes involve neural networks which recognise and classify information on the basis of probability. Deep Learning (DL) adds feedback loops enabling learning via sensing or being instructed on the validity of decisions, so ML can modify approaches to constantly improve accuracy i.e., it trains itself.

AI and ML underpin the concept of ‘Smart Farming’ where agriculture is made more productive and sustainable. In combination with Big Data and ultra-fast processor computers, it is creating new opportunities for data-driven (evidence based) crop production.

Apart from driving autonomous machines, AI and ML helps analyse farm data from the thousands of data points collected from the ground each day. Real-time monitoring of local weather conditions, temperature, soil conditions, water status, pest and disease and weed status, crop health and growth are collected and processed to better inform agronomic decisions and to allow for more precisely targeted inputs. Seasonal forecasting and decision support systems are also made possible allowing for improved crop management efficiency.  Weed recognition systems can facilitate real-time spot-spraying, and remote sensing allows differential fertilisation of fields. In plant breeding, ML is used to identify and predict which genes will contribute to beneficial traits.

It allows farmers and agribusinesses to streamline operations and to make evidence-based decisions based on findings in real time, helping to make traditional agricultural activities more efficient and predictable.

https://www.mdpi.com/1424-8220/18/8/2674/pdf

https://medium.com/sciforce/machine-learning-in-agriculture-applications-and-techniques-6ab501f4d1b5