Application of AI and machine learning in modern agronomy: A comprehensive review

Introduction

The world population is expanding, resources are running out and climate change is making farming harder and harder. Agronomists are using more modern, data-driven methods to get the most out of land, water and nutrients while still meeting the world's need for food [1] . AI (artificial intelligence) and ML (machine learning) are great techniques to automatically look at difficult agricultural data and come to conclusions that other approaches can't. Artificial intelligence (AI) and machine learning (ML) are making agronomy, the science of cultivating crops and taking care of soil, more accurate, useful and long-lasting. AI algorithms may make planting schedules, watering, fertilizing, insect control and harvesting better by looking at satellite photos, sensor data, weather forecasts and data from past harvests [2] . AI systems have made a lot of progress in several domains. Studies have indicated that crop yields can go up by 20 to 30%, the use of pesticides can go down and resources can be used more efficiently. The cost ratios look excellent because of these benefits. AI-powered precision agriculture has helped major farms make more than 100-150% on their investments (ROI). This review talks a lot about how AI and ML are employed in farming these days [3] . It begins by going back at the past and talking about the first expert systems and the scientists who used computers to help them farm. Then it talks about the primary AI/ML methods and technologies, such drones, sensors and robots that have revolutionized how we gather information [4] . The main premise of the review is how farms use things like figuring out how much crops will grow, keeping the soil healthy, robotic harvesting and controlling greenhouses. There are real-world examples of how each kind of AI model and application has worked. We also discuss about the finest platforms and firms that provide AI-based farming solutions [5] . People think about topics like higher yields, cheaper costs, return on investment (ROI) and environmental benefits (such using less water) when they think about how AI will change the economy. When it makes sense, we summaries the numbers from case studies to explain how AI could be helpful. Finally, we speak about concerns (such worries about data quality and adoption) and the future (including AI advisers that make things happen and farms that run themselves) [6] . This review seeks to function as a thorough guide on the revolutionary influence of AI and ML on agriculture in the 21st century by analyzing their historical, contemporary and future advancements.

Historical Background and Pioneers 1. For a long time, farming has been using computers and other tools more and more. In the 1960s and 1970s, people first sought to utilize basic statistical models and rule-based systems to grow crops. H. built the COMAX expert system was a huge step forward for AI in farming. Lemmon used "if-then" algorithms and sensor data to determine out the optimal way to water and fertilize cotton crops in 1986 [7] . 2. In field tests, COMAX proved that it might cut production costs by automatically adjusting the amount of water and pesticides used to match what an expert agronomist would suggest. A lot of people think that this system was one of the first AI technologies that could help farmers. In other words, computers can give you advice that is just as good as that of specialists. At the same time, agricultural scientists like Norman Borlaug and others (but not AI researchers) underlined the need for innovative ideas to boost yields. This made it possible for technology to be used in the future [8] . 3. Agronomy began to use more statistical and machine learning technologies in the 1980s and 1990s. Researchers utilized regression models and decision trees to guess how much crops would yield in the early stages, learn about the types of soil they were working with and figure out how harmful something was. At Iowa State, Cornell and USDA labs, scientists including Gerald Anderson and Pete Thomison worked on algorithms for mapping soil [9] . 4. Some people tried to use neural networks to figure out how much food will be made. In the late 1990s, GPS technology and variable rate devices made "precision agriculture" possible. This entails taking care of fields that are less than an acre in size. Jack Jones and other early adopters proved that site-specific management, which entails adjusting things like fertilizer in certain regions, may make work go more smoothly. They also got the first massive sets of yield maps, which were later used to teach machine learning algorithms [10] . 5. In the 2000s, it was easier to collect information thanks to satellite photos, UAV (drone) images and sensors that were connected to the internet. Two agronomists, David Lobell and Senthold Asseng, employed machine learning and remote sensing to see how stressed crops were from space.

In the 2000s, a lot of people tried to use decision support systems (DSS) to manage nutrients and keep pests under control [11] . 6. Many governments and academic institutions started building automated systems that used weather forecasts and agronomic models to benefit people. Two big steps forward that made it possible to keep an eye on plant health in a fair way were the IRRI crop models for rice and NASA's work on the NDVI (Normalised Difference Vegetation Index) [12] . 7. AI has developed a lot in the previous ten years. Highperformance computation and new methods, especially deep learning, have made it possible to see photographs and complicated data streams in real time. Researchers in the field of ag-tech today employ deep convolution neural networks to discover diseases on leaves [13] . 8. Generative models help farmers grow better crops, while reinforcement learning helps them develop better irrigation schedules. A group of the best AI researchers, most of whom are computer scientists, have worked with agronomists. For example, John Deere eventually bought Blue River Technology, which used computer vision technology in farming. The guys who made Blue River and other companies that used to work for Google or Facebook employed the most recent machine learning to develop robots that could weed [14] . 9. Many professionals from various professions work in agricultural science these days. Computer scientists and agronomists from the past set the stage for all the smart tractors, drones and sensor networks we use today. It's easy to understand why scientists love H. has been connected to each other throughout history. Lemmon and many other AI and agronomy experts have worked together to make precision farming with AI possible today [15] .

Key AI and Machine Learning Techniques in Agronomy

1. Modern farming uses a lot of AI and ML techniques. These can be anything from old-fashioned methods to the most recent deep learning architectures. Regression, support vector machines (SVM) and decision forests are all types of supervised learning that are typically used to predict things that happen over time, like yield or soil moisture, or to sort things, like the health of a plant [16] . 2. For example, random forests and multiple linear regressions have been used for a long time to work out how much crops would yield based on things like the weather, the quality of the soil and how well they were managed in the past. Deep convolution neural networks (CNNs) are the most common way to do things with images, such figuring out what sort of crop it is or finding weeds. CNNs can detect where things are in space just by examining at pictures taken by satellites or drones [17] . 3. People typically utilize Long Short-Term Memory (LSTM) networks and other newer versions of Recurrent Neural Networks (RNNs) to guess how the weather will change over time or how things will grow over time. These networks could be able to find connections between events that happen at various periods [18] . 4. In agronomy, unsupervised learning methods like clustering algorithms (like k-means) are used to group field zones with similar soil or yield characteristics. Principal component analysis (PCA) is used to reduce the number of dimensions in high-throughput sensor data. Reinforcement learning (RL) is becoming more and more common for professions that require making judgments, such regulating irrigation and directing self-driving cars [19] . 5. By testing things out and getting feedback, an RL agent learns the best things to do, such how much to water. In this way, it can get the most out of the circumstance and not waste too much time. Bayesian networks and probabilistic models assist us deal with things we don't know and understand risks. For example, they can help us determine estimate how likely it is that a pest outbreak will happen based on the weather. Fuzzy logic systems have been used in farming to deal with uncertainty. For example, fuzzy controllers can figure out "moderate" irrigation settings using inaccurate information about the weather and the amount of moisture in the soil [20] . 6. Another useful method is transfer learning. You can use AI models that were trained on one crop or area on another crop or area with smaller datasets. This makes deployment go a lot faster. For instance, you could utilize a CNN that has previously trained a lot about plants to discover diseases in a certain crop in your area with a lot less fresh data [21] . 7. Decision support systems also usually combine both rulebased expert knowledge, such the "if-then" logic used in early expert systems and machine learning models to develop systems that work together [21] .

Table 1:

The following table summarizes some of the major AI and ML techniques and their typical agronomic uses [22,23,24] Technique These AI/ML techniques are complemented by domain-specific considerations in agronomy, such as working with unbalanced data (e.g. rare pest events) and integrating multi-source data (e.g. combining satellite images, soil sensors and farmer inputs). In practice, modern ag-tech solutions often employ ensembles of models and combine machine learning with agronomic knowledge to improve reliability [25] .

Data Sources and Technological Advances

AI can only be useful in agronomy if there is a lot of data and tools for sensing. In the last few years, the agricultural industry has gotten a lot of new sources of information. Some important technologies are:

• Remote Sensing (Satellites and Drones): Satellites like Landsat and Sentinel have cameras that can see in many colors and take pictures of farms in large areas multiple times. You can use these pictures to figure out NDVI (vegetation index), which tells you how healthy your crops are, how much biomass there is and how many plants are sick or stressed. Drones, also called unmanned aerial vehicles (UAVs), are now being used to do field surveys [26] . Drones can fly under clouds to take very clear pictures (down to the centimeter level) and can carry special sensors including RGB cameras, thermal cameras, LiDAR and multispectral cameras. AI algorithms use pictures taken by drones to do a lot of tasks, such as check on the health of plants, count fruits and make maps of areas where weeds grow [27] . • Ground Sensors and IoT: Sensor networks and Internet of Things (IoT) devices collect data on the weather and soil in real time. Some examples are leaf wetness sensors, nutrition probes, sensors that detect soil moisture and weather stations. When applied in machine learning models, this sensor data helps adjust how much fertilizer and water is used [28] . A reinforcement learning agent can acquire data from a network of soil moisture sensors that are spread out throughout a field. This agent then turns on the irrigation systems at the proper times to save water and still get amazing results. With smart irrigation controls and soil analysis kits, farmers can get data much more often than they could with manual samples taken every so often [29] . • Autonomous Machinery and Robotics: Many new tractors, combine harvesters and sprayers come with computers and sensors like GPS, cameras and LIDAR. These cars could collect data while they were driving. For instance, plants contain sensors that measure moisture, while harvesters have monitors that measure yield [30] . They can also use AI to adjust how they work in real time, as with targeted spraying and auto-steering. More and more, robotics is being used in specialist crops. For example, selfdriving weeding robots use machine vision to find their way between rows and pluck up weeds [31] . • Farm Management Software and Databases: Cloudbased platforms and apps connect information from the past about the farm, including planting dates, types of seeds used and past yields, with information from other sources, like soil maps and market prices [32] . These systems employ machine learning to discover trends and give people personalized advice. You may now train ML models using large agricultural databases, such those from the government or research groups. Data sharing projects are making agricultural data from all across the world more accessible [33] . • Genomic and Phenotypic Data: Genomic data (DNA markers, gene sequences) and high-throughput phenotyping (digital trait assessments of plants) are both growing swiftly in the field of crop breeding. AI connects genotypes to traits that are wanted, which speeds up the process of developing new types. For instance, deep learning algorithms can look at photographs of crop canopies and locate plants that are stronger or less likely to get sick [34] .

Agronomy has never been data-driven like this before, thanks to high-resolution sensors, robots, cloud computing and strong algorithms. Drones that don't cost much can fly over farms and capture images. These photographs are then relayed to CNNs to aid with automated watering valves or sprayers. These technologies all function together. We can do things now that we couldn't do ten years ago because every new sensor, connection, or AI model makes the others operate better [35] .

Key Applications in Agronomy

AI and ML work well together; therefore they can be used on a farm in a lot of different ways. We discuss about some of the most essential ways that AI is changing things below, along with some examples and what they mean.

• Precision Crop Monitoring and Yield Prediction: AI systems use data from the past and the present to forecast how well crops will do. For instance, an ML model might be able to predict how much a field will yield weeks or even months in advance by looking at weather data, soil conditions and multispectral photos [36] . Farmers can use these kinds of weather forecasts to figure out when to plant and sell their crops. You can always check on how your crops are doing at different points in time by using computer vision on photographs obtained by drones or satellites. The system shows a part of a field that isn't expanding as it should. It might not have enough nutrients or it might be sick. This kind of early warning lets you take concentrated action. Many farmers report that using AI to make better predictions about when to plant and what to use makes their crops grow 10-30% better in real life [37] . • Getting rid of pests and diseases: It's very vital to identify pests or diseases early. AI image analysis can recognise spots on leaves, blights and insect damage in photos. For example, a machine learning model that has been trained on hundreds of photographs of plant leaves can tell when a plant is sick, such when potatoes have late blight or wheat has rust. It can be just as accurate as a human specialist in a lot of circumstances [38] . AI with drones that can see may be able to cover a lot more ground. For example, neural networks can locate dirty places by looking at photographs of whole orchards or fields. Companies say that these technologies reduce the need for pesticides by just treating the regions that require it. This can cut the use of chemicals by 50% or more while keeping production losses to a minimal [39] . • Robots in the field and weed control: Robots driven by AI are revolutionizing how weeds are controlled. John Deere now owns Blue River Technology's "See & Spray" technology, which uses computer vision to identify the difference between crops and weeds as it moves through a field. At that time, only the weeds should be sprayed with herbicide [40] . Compared to spraying everything at once, this exact strategy can cut down on pesticide use by about 90%. This is better for the environment and saves a lot of money.

There are also tests going on with self-driving robots that have mechanical weeders in row crops. They may go about their business every day without any aid from people thanks to machine learning [41] . • Managing Soil and Nutrients: Machine learning algorithms look at soil data (pH, organic matter, nutrient levels) and the needs of crops to propose different amounts of fertilizer to use. Instead of putting the same amount of fertilizer all over the field, AI helps produce fertility maps that show how much to use in each location [42] . This "sitespecific management" helps crops grow better and cuts down on waste. Some studies have shown that AI-optimized fertilizer algorithms can use nutrients more than 20% better. Models can also tell you where the dirt will wash away or get hard. This can help you fix things more precisely [43] . • One of the most common uses of this technology is smart irrigation. AI systems use weather forecasts, estimates of how much water evaporates and sensors that assess soil moisture to figure out when to water. Studies demonstrate that a lot of water can be conserved. For example, an AIcontrolled irrigation system in one vineyard lowered water use by more than 30% and made the grapes better and more plentiful [44] . On a large scale, these kinds of gadgets can save millions of liters of water. As the weather changes, reinforcement learning and adaptive control algorithms keep improving the rules for watering plants better. AI also quickly discovers leaks or problems with too much water, which stops the water from going lost [45] . • Greenhouse and Controlled Environment Agriculture:

AI watches the temperature, humidity, light and fertilizer solutions in real time for greenhouse farming. Neural networks get data from sensors and then look at all the factors to identify the fastest growth rate. Some high-tech vertical farms have AI systems that run on their own. Deep learning algorithms modify the amount of fertilizer and the LED lighting in a couple of milliseconds. Cameras and weight sensors keep an eye on how big the plants are. These farms grow a lot more crops per square meter than normal farms do and they don't consume a lot of resources [46] . • After the harvest, taking care of the animals and the farm: Agronomy is largely about crops, but AI may also be able to aid with running the overall farm. AI watches on the health and behaviour of animals on farms that raise cattle, which are sometimes part of larger farms. After the harvest, computer vision automatically grades and sorts fruit or verify the quality of grain. This means that fewer people have to do this job. AI and cameras can discover problems with almonds and apples on a production line faster than people can do it by hand [47] . • Supply Chain and Market Optimization: Some AI farming systems are all about getting things where they need to go. By predicting how much crops will be worth on the market, machine learning can assist farmers decide which crops to grow and when to sell them. Supply chain software employs optimization algorithms to find the best ways to move and store goods so that less waste is created [48] .

The breadth of applications is vast

These applications are interrelated. For instance, AI-driven soil analysis informs irrigation control; yield prediction feeds into supply chain planning. Together, they constitute a precision agriculture ecosystem where data continuously flows and is acted upon by algorithms to optimize each step of the agronomic process.

Table 2: below summarizes key areas, AI approaches used and typical outcomes [49,50,51] Application

Industry and Corporate Initiatives

Agri-tech has grown so swiftly because of research done in schools, businesses and new businesses. Some of the major IT and farming organizations have built AI-based tools and platforms that will help modern agronomy move forward [52] .

Table 3: lists some well-known companies, their AI solutions and the fields they work in to help you understand what we're talking about [53,54,55] Company • These companies exemplify how business and research may operate together. When John Deere bought Blue River Technology, a firm that makes traditional farm tools, it spent a lot of money on AI. Deere has been working on prototypes for self-driving tractors for twenty years [56] . They use sensors and satellite photos on board to get the most out of their crops and use less gas. AGCO and New Holland are two companies that make farm equipment.

They have also integrated telematics and navigation systems that use machine learning to their products [57] . • Businesses that develop software and store data are also incredibly essential. Bayer and Monsanto were the first to create FieldView, a platform that uses maps and prediction algorithms to aggregate field data including yields, weather and inputs. People who utilize FieldView suggest that farmers may minimize their nitrogen use by 40% to 60% and still achieve the same or greater yields by only adding fertilizer where it is needed [58] . CropX and other new companies offer full solutions. Their soil moisture sensors feed information to a cloud service that utilizes AI to tell them when to water the plants. Prospera's AI technology helps enormous tomato and lettuce farms all around the world discover problems before they get worse. People think this will keep millions of dollars from being lost [59] . • Companies that don't ordinarily work in farming are starting to do so. Microsoft and other big IT businesses are working with pesticide companies and farming cooperatives like Land O'Lakes to deploy cloud-AI technologies. For instance, Microsoft's FarmBeats project employs drones, sensors and Azure AI to deliver smart farming based on the Internet of Things (IoT) to places that are hard to get to [60] .

Agrostar and CropIn both have smartphone apps that help small farmers plant by combining data from the area with satellite photos. IBM is a big corporation that has an Environmental Intelligence Suite. This suite uses AI to combine hyper-local weather data from The Meteorological firm with other data to tell farmers when to plant and warn them when the weather is bad [61] . • Most of the time, these company ideas contain real-life examples of people who have made them work. For example, a large grain cooperative reported that utilizing AI-guided variable-rate technology saved them more than $25 per acre every season by cutting the costs of seeds and fertilizers that weren't needed [62] . The Australian project COALA employed AI to make irrigation 20% more effective. This helped one farm save tens of thousands of dollars. Even on a small basis, individual farmers have seen big shifts. One American farmer reported that precision farming technology helped him cut his nitrogen use from 1.2 pounds per bushel of maize to 0.5 pounds, a 60% drop, while still receiving exceptional yields [63] . • Along with well-known enterprises, a lot of new and interesting businesses are getting their names out there. A lot more new businesses are concentrating on things like robotic fruit pickers, predictive breeding algorithms, blockchain-based supply systems and more. Table 3 only shows a few of them. The essential point is that AI and ML are being utilized more and more in farming and the results demonstrate that they work better and are better for the planet [64] . These changes also show that the ecology is getting bigger. For example, original equipment manufacturers (OEMs) of farm tools, software businesses, data analytics groups and research institutes are all working together on pilot projects and joint ventures. These companies compete with each other and share ideas, which helps the overall sector move forward more swiftly [65] .

Economic and Environmental Impact

There are many economic benefits of using AI in farming. Many studies and case studies show that the benefits of technology usually outweigh the drawbacks. The return on investment (ROI) for AI investments varies by circumstance; however, for large corporations, it typically ranges from 120% to 150%. A study found that large commercial farms that used AI-powered systems had an average return on investment (ROI) of about 150%. But smaller farms still made roughly 120% on what they put in. These data indicate how AI might help with things like getting more crops; spending less on inputs and helping things run more smoothly [66] .

People in the field discuss about some significant ways to quantify impact

• More Yields: AI management usually leads to more crop yields. Many reports claim that average yields for all crops have gone up by roughly 25% because of better inputs and faster reactions. Some farmers report that employing adaptive fertilization and careful monitoring has helped them by more than 30% [67] .

• Getting rid of pests and diseases: AI may assist discover diseases early, which can cut losses from them by almost half. AI-powered scouting and targeted treatment can cut pest damage by more than half [68] .

• Input Usage Efficiency: AI makes it easier to use chemicals, water and fertilizer the right way. Tests have shown that smart irrigation systems can save up to 30% of water. People say that the precision of using fertilizer has gone up by 80-85%. This means that plants get the nutrients they need right where they need them, which cuts down on waste. AI-optimized scheduling made one situation 80% better at using water [69] .

• Lower Costs: By modifying how they do things in each zone, farmers can save money on inputs. A well-known study found that AI-guided precision farming saved an average of $25 per acre every season by lowering the expenses of seeds, herbicides and fertilizers. Spraying with caution is another example. If you only spray weeds, you can cut down on the amount of pesticides you use by up to 90%, which saves you money straight away [70] . • Work and operations that are more efficient: Automation means that fewer people need to work. People who have worked in these fields claim that for things like gardening and keeping an eye on things, they might only have to do half as much work. The exact amounts could be different. Drones and self-driving harvesters also help you get things done on time, like picking at the appropriate moment, which makes the fruit better and worth more on the market [71] . • Effects on the Environment: AI is excellent for the environment since it uses less energy and makes less pollution. Some people think that AI-powered precision farming might cut down on greenhouse gas emissions from farming by a lot by using less fertilizer and making machines operate better. One study found that applying AI to manage crops might lower their carbon footprint by 10% to 20% for each unit [72] .

Table 4: below synthesizes some of these impacts and cost metrics from various projects and studies [73,74,75,76] Impact such when weather stations don't cover a lot of ground. It's still a work in progress to make sure that data can work together and that there are standard formats [77] . To employ AI technology, you also need the correct tech tools. In some isolated areas, you can't always count on having consistent power and internet.

Edge computing and devices that can run without an internet connection can help, but they might also cost more. The digital divide is a different challenge. Big farms can buy and deploy costly AI systems, while small farmers in developing nations might not be able to. Some projects are trying to fix this by linking AI consultancy services with mobile networks. Some people are working on making AI tools that are inexpensive and easy for everyone to get. It's also very crucial to have trust and training [78] . Farmers need to be able to believe what AI says them and understand it. Putting people first is really important and agronomists can help you develop interfaces that are simple to use. Money is also a problem. It costs a lot of money to make robots and sensors that are better. It could take small farmers a long time to get their money back. Some tools that are incredibly accurate cost a lot of money and take a long time to make back what you spent on them. But as more success stories come in and business models evolve (for example, businesses might lease equipment instead of buying it outright or sign service contracts), the economics get better. Governments, NGOs and corporations can also help farmers use new technologies more safely by giving them loans or sharing resources [79] .

In some important areas, the future of AI in farming will be different

• Generative AI and Virtual Agronomists: New generative models like ChatGPT make it look like virtual agronomists AI bots that can fully analyze agricultural data and talk to farmers in simple terms might be possible. Using more than one model, these kinds of systems would be able to answer queries such, "What should I do to change my watering next week based on the weather report?" [80] .

• Combining genomics and phenotyping: AI will connect plant DNA to how well plants develop in the field more and more. For example, neural networks may find the best conditions for a new crop variety to grow or run millions of tests to evaluate what might happen if two plants were cross-bred. Robotics will speed up phenotyping (measuring traits), which will offer these models additional data to work with [81] .

• Edge AI with 5G: Decisions may be made in real time with less lag time thanks to better edge computing (running AI on devices in the field) and 5G connectivity. More and more tractors and drones will have AI processors built into them. This will let people look at pictures straight away without needing to connect to the internet. When you mix blockchain with the Internet of Things (IoT), it's even more important to keep data safe and easy to find. Some agri-tech projects use both IoT sensors and blockchain to maintain track of farming activities and conditions that can't be modified. AI can then check these records to make sure they are correct and meet quality standards, such making sure that organic standards were met [82] .

• Autonomous Farms: The idea is to develop farms that don't need people to work on them in the long run. Most of the work is done by AI and robots. There are already test farms with greenhouses that don't need any aid, driverless tractors and robotic pickers. This vision will move forward faster when general AI and robotics get better, such as by having higher perception and dexterity [83] .

• Farming that is good for the environment and can be used again: AI will help us attain our goals for sustainability. Machine learning may help you figure out the best cover crops, crop rotations and techniques to use less tillage, all of which can improve soil health and store carbon. AI tools can also help farmers switch to regenerative agriculture without losing too much productivity by showing how different parts of the ecosystem work together [84] .

• Ethical and Policy Dimensions: As AI gets smarter, there will be more issues about what is right and wrong. Policies may be needed to protect data ownership (farmers' data should help them) and to deal with problems like algorithmic bias (making sure that AI doesn't unfairly favour certain places or crops). International partnerships to make food safer may set rules for how AI can be used in farming [85] .

Conclusion

Machine learning and artificial intelligence are changing the way we do agronomy right now. This research has looked at how rule-based systems have grown over time to become the powerful deep learning and robotic solutions we have now. H. Lemmon started it all with the first expert systems. Since then, each new generation of researchers has improved on that base by using better models and data sources. AI is used in all parts of farming today, from checking the soil and watching the crops to using machines that function on their own and predicting market trends. AI has a big effect on farming in the real world. Farmers and businesses say they get greater yields (typically 25% or more), use a lot less input (for example, 50-90% less insecticide) and save a lot of money. Most of the time, using new technology is worth it. For example, AI-driven management usually pays for itself in just a few seasons, which makes the benefit/cost ratios very appealing. The environmental benefits, such less fertilizer and water runoff and less greenhouse gas emissions, show that the goals for the economy and the environment are often the same. But we need to fix problems with data, infrastructure and getting people to use them if we want to get these benefits. It's important to spend money on things like connecting rural areas, making data more consistent and teaching farmers. To make AI technology available to all kinds of farms, we need legislation that supports them and cooperation between the public and private sectors. In short, AI and ML give agronomists tools that could change the way things are done. They take intricate data from the field and turn it into useful information that helps farmers make better decisions, lower their risks and use resources in a way that doesn't harm the environment. As technology gets better, AI will play a bigger role in farming, making food systems smarter and better prepared to deal with challenges. If people and machines work together, we could have a future where farms work as well as labs, feeding the world and taking care of the planet at the same time.

References

1. Liakos KG, Busato P, Moshou D, Pearson S, Bochtis D. Machine learning in agriculture: A review. Sensors. 2018;18(8):2674. doi:10.3390/s18082674
2. Osisanwo FY, Akinsola JET, Awodele O, Hinmikaiye JO, Olakanmi O, Akinjobi J. Supervised machine learning algorithms: Classification and comparison. Int J Comput Trends Technol. 2017;48(3):128-138.
3. Veeragandham S, Santhi H. A review on the role of machine learning in agriculture. Scalable Comput Pract Exp. 2020;21(4):583-589. doi:10.12694/scpe.v21i4.1699
4. Kotsiantis SB, Zaharakis I, Pintelas P. Supervised machine learning: A review of classification techniques. Emerg Artif Intell Appl Comput Eng. 2007;160(1):3-24.
5. Durgesh KS, Lekha B. Data classification using support vector machine. J Theor Appl Inf Technol. 2010;12(1):1-7.
6. Akar Ö, Güngör O. Classification of multispectral images using random forest algorithm. J Geod Geoinf. 2012;1(2):105-112.
7. Priya R, Ramesh D, Khosla E. Crop prediction on the region belts of India: A Naïve Bayes MapReduce precision agricultural model. In: Proc 2018 Int Conf Adv Comput Commun Inform; 2018 Sep; India. p. 99-104. doi:10.1109/ICACCI.2018.8554948
8. Yaseen ZM, Al-Juboori AM, Beyaztas U, Al-Ansari N, Chau KW, Qi C, Shahid S. Prediction of evaporation in arid and semi-arid regions: A comparative study using different machine learning models. Eng Appl Comput Fluid Mech. 2020;14(1):70-89. doi:10.1080/19942060.2019.1680576
9. Feng Y, Peng Y, Cui N, Gong D, Zhang K. Modeling reference evapotranspiration using extreme learning machine and generalized regression neural network only with temperature data. Comput Electron Agric. 2017;136:71-78. doi:10.1016/j.compag.2017.01.027
10. Feng Y, Cui N, Hao W, Gao L, Gong D. Estimation of soil temperature from meteorological data using different machine learning models. Geoderma. 2019;338:67-77. doi:10.1016/j.geoderma.2018.11.044
11. Coopersmith EJ, Minsker BS, Wenzel CE, Gilmore BJ. Machine learning assessments of soil drying for agricultural planning. Comput Electron Agric. 2014;104:93-104. doi:10.1016/j.compag.2014.04.004
12. Heuvelink GB, Angelini ME, Poggio L, Bai Z, Batjes NH, van den Bosch R, et al. Machine learning in space and time for modelling soil organic carbon change. Eur J Soil Sci. 2020. doi:10.1111/ejss.12998
13. Crane-Droesch A. Machine learning methods for crop yield prediction and climate change impact assessment in agriculture. Environ Res Lett. 2018;13(11):114003. doi:10.1088/1748-9326/aae159
14. Abbas F, Afzaal H, Farooque AA, Tang S. Crop yield prediction through proximal sensing and machine learning algorithms. Agronomy. 2020;10(7):1046. doi:10.3390/agronomy10071046
15. Pantazi XE, Moshou D, Alexandridis T, Whetton RL, Mouazen AM. Wheat yield prediction using machine learning and advanced sensing techniques. Comput Electron Agric. 2016;121:57-65. doi:10.1016/j.compag.2015.11.018
16. Pantazi XE, Tamouridou AA, Alexandridis TK, Lagopodi AL, Kashefi J, Moshou D. Evaluation of hierarchical selforganising maps for weed mapping using UAS multispectral imagery. Comput Electron Agric. 2017;139:224-230. doi:10.1016/j.compag.2017.05.026
17. Kaur S, Pandey S, Goel S. Plant disease identification and classification through leaf images: A survey. Arch Comput Methods Eng. 2019;26(2):507-530.
18. Badage A. Crop disease detection using machine learning: Indian agriculture. Int Res J Eng Technol. 2018;5(9).
19. Arbuckle T, Schröder S, Steinhage V, Wittmann D. Biodiversity informatics in action: Identification and monitoring of bee species using ABIS. In: Proc 15th Int Symp Informatics for Environmental Protection; 2001; Zurich. p. 425-430.
20. Buschbacher K, Ahrens D, Espeland M, Steinhage V. Image-based species identification of wild bees using convolutional neural networks. Ecol Inform. 2020;55:101017.
21. Sharma A, Jain A, Gupta P, Chowdary V. Machine learning applications for precision agriculture: A comprehensive review. IEEE Access. 2020.
22. Dutta R, Smith D, Rawnsley R, Bishop-Hurley G, Hills J, Timms G, Henry D. Dynamic cattle behavioural classification using supervised ensemble classifiers. Comput Electron Agric. 2015;111:18-28. doi:10.1016/j.compag.2014.12.002
23. Jacobs JA, Siegford JM. The impact of automatic milking systems on dairy cow management, behavior, health and welfare. J Dairy Sci. 2012;95(5):2227-2247. doi:10.3168/jds.2011-4943
24. Goraya MS, Kaur H. Cloud computing in agriculture. HCTL Open Int J Technol Innov Res. 2015;16:2321-1814.
25. Rahman SAZ, Mitra KC, Islam SM. Soil classification using machine learning methods and crop suggestion based on soil series. In: Proc 21st Int Conf Comput Inf Technol; 2018 Dec; Bangladesh. p. 1-4.
26. Kaur T, Kaur H. Machine learning: An internal review. J Emerg Technol Innov Res. 2018;5:1-6.
27. Kaur T, Chana I. Energy efficiency techniques in cloud computing: A survey and taxonomy. ACM Comput Surv. 2015;48(2):1-46. doi:10.1145/2742488
28. Kaur T, Chana I. Energy aware scheduling of deadlineconstrained tasks in cloud computing. Cluster Comput. 2016;19(2):679-698. doi:10.1007/s10586-016-0566-9
29. Kaur T, Chana I. GreenSched: An intelligent energy aware scheduling for deadline-and-budget constrained cloud tasks. Simul Model Pract Theory. 2018;82:55-83. doi:10.1016/j.simpat.2017.11.008
30. Radadiya BL. Cloud computing and agriculture. Int J Agric Sci. 2016.
31. Mekala MS, Viswanathan P. Smart agriculture IoT with cloud computing. In: Proc Int Conf Microelectron Devices Circuits Syst; 2017 Aug; India. p. 1-7.
32. Singh S, Chana I, Buyya R. Agri-info: Cloud based autonomic system for delivering agriculture as a service. Internet Things. 2020;9:100131.
33. Singh V, Sankhwar S, Pandey D. The role of information communication technology in agriculture. Glob J Multidiscip Stud. 2015;3(4):2-10.
34. Cox S. Information technology: The global key to precision agriculture and sustainability. Comput Electron Agric. 2002;36(2-3):93-111.
35. Singh S, Ahlawat S, Sanwal S. Role of ICT in agriculture: Policy implications. Orient J Comput Sci Technol. 2017;10(3):691-697. doi:10.13005/ojcst/10.03.20
36. Malavade VN, Akulwar PK. Role of IoT in agriculture. IOSR J Comput Eng. 2016;13(1):56-57.
37. Kulkarni AA, Dhanush P, Chetan BS, Gowda CT, Shrivastava PK. Applications of automation and robotics in agriculture industries: A review. IOP Conf Ser Mater Sci Eng. 2020;748(1):012002. doi:10.1088/1757-899X/748/1/012002
38. Billingsley J, Visala A, Dunn M. Robotics in agriculture and forestry. 2008.
39. Kaundal R, Kapoor AS, Raghava GP. Machine learning techniques in disease forecasting: A case study on rice blast prediction. BMC Bioinformatics. 2006;7(1):1-16.
40. Sonnino A. Towards sustainable food and agriculture systems. Rend Accad Naz Lincei. 2018;42(1):103-114.
41. Naresh RK, Chandra MS, Vivek S, Charankumar GR, Chaitanya J, et al. The prospect of artificial intelligence in precision agriculture for farming systems productivity in sub-tropical India: A review. Curr J Appl Sci Technol. 2020;39(48):96-110. doi:10.9734/cjast/2020/v39i4831205
42. Nagendraswamy C, Salis A. A review article on artificial intelligence. Ann Biomed Sci Eng. 2021;5:13-14. doi:10.29328/journal.abse.1001012
43. Fadziso T. How artificial intelligence improves agricultural productivity and sustainability: A global thematic analysis. Asia Pac J Energy Environ. 2019;6:91-100. doi:10.18034/apjee.v6i2.542
44. Erh-Chun, Shan L, Chan P. How artificial intelligence is transforming agriculture. Rev Bus Res. 2023;23(1):59-70. doi:10.18374/RBR-23-1.6
45. Javaid M, Haleem A, Khan IH, Suman R. Understanding the potential applications of artificial intelligence in agriculture sector. Adv Agrochem. 2023;2(1):15-30. doi:10.1016/j.aac.2022.10.001
46. Kolikipogu R, Darak V, Yennapu R, Reddy S, Sureddi RMK, Kuchipudi R. Agriculture recommender system for precision farming using machine learning. In: Proc 3rd Int Conf Innov Mech Ind Appl; 2023 Dec 21; India. IEEE; p. 921-927. doi:10.1109/ICIMIA60377.2023.10426510
47. Begum M. Impact of climate change on agriculture and its allied sectors: An overview. Emerg Trends Clim Change. 2022;1(1):19-28. doi:10.18782/2583-4770.103
48. Jose Mekha, Parthasarathy V. Automated pest identification and classification in crops using artificial intelligence: A state-of-the-art review. Autom Control Comput Sci. 2022;56(3):283-290. doi:10.3103/S0146411622030038
49. Prabha R, Kennedy JS, Vanitha G, Sathiah N, Priya MB. Artificial intelligence-powered expert system model identifying fall armyworm infestation in maize (Zea mays L.). J Appl Nat Sci. 2021;13(4):1339-1349. doi:10.31018/jans.v13i4.3040
50. Susheel KS, Rajkumar R. A comprehensive review on intelligent techniques in crop pests and diseases. Int J Recent Innov Trends Comput Commun. 2023;11(9):137149. doi:10.17762/ijritcc.v11i9.8328
51. Shoaib M, Shah B, El-Sappagh S, Ali A, Ullah A, Alenezi F, et al. Advanced deep learning models-based plant disease detection: A review of recent research. Front Plant Sci. 2023;14:1158933. doi:10.3389/fpls.2023.1282443
52. Storey G, Meng Q, Li B. Leaf disease segmentation and detection in apple orchards for precise smart spraying in sustainable agriculture. Sustain. 2022;14(3):1458. doi:10.3390/su14031458
53. Karar ME, Alsunaydi F, Albusaymi S, Alotaibi S. Mobile application for agricultural pest recognition using deep learning in cloud computing systems. Alex Eng J. 2021;60(5):4423-4432. doi:10.1016/j.aej.2021.03.009
54. Tannous M, Stefanini C, Romano D. Deep-learning-based detection approach for identification of insect species of economic importance. Insects. 2023;14(2):148. doi:10.3390/insects14020148
55. Adikari KE, Shrestha S, Ratnayake DT, Budhathoki A, Mohanasundaram S, Dailey MN. Evaluation of artificial intelligence models for flood and drought forecasting in arid and tropical regions. Environ Model Softw. 2021;144:105136. doi:10.1016/j.envsoft.2021.105136
56. Awais M, Naqvi SMZA, Zhang H, Li L, Zhang W, Awwad FA, et al. AI and machine learning for soil analysis: Assessment of sustainable agricultural practices. Bioresour Bioprocess. 2023;10(1):90. doi:10.1186/s40643-023-00710y
57. Wei H, Xu W, Kang B, Eisner R, Muleke A, Rodriguez D, et al. Irrigation with artificial intelligence: Problems, premises, promises. Hum-Centric Intell Syst. 2024;4(2):187-205. doi:10.1007/s44230-024-00072-4
58. Raouhi EM, Zouizza M, Lachgar M, Zouani Y, Hrimech H, Kartit A. AIDSII: An AI-based digital system for intelligent irrigation. Softw Impacts. 2023;17:100574. doi:10.1016/j.simpa.2023.100574
59. Padhiary M, Saha D, Kumar R, Sethi LN, Kumar A. Enhancing precision agriculture: Review of machine learning and AI vision applications in all-terrain vehicles for farm automation. Smart Agric Technol. 2024;8:100483. doi:10.1016/j.atech.2024.100483
60. Alazzai WK, Abood BShZ, Al-Jawahry HM, Obaid MK. Precision farming: The power of AI and IoT technologies. In: Li G, Subramaniam U, Sekar M, editors. Proc Int Conf Environ Dev Using Comput Sci (ICECS'24). E3S Web Conf. 2024;491:04006. doi:10.1051/e3sconf/202449104006
61. Bao J, Xie Q. Artificial intelligence in animal farming: A systematic literature review. J Clean Prod. 2022;331:129956. doi:10.1016/j.jclepro.2021.129956
62. Boruah K, Hembram PK, Deb D, Rahman S, Ghosh N. Internet of Things and artificial intelligence in livestock farming. BioNE. 2025;28(32).
63. Neethirajan S. Artificial intelligence and sensor innovations: Enhancing livestock welfare with a humancentric approach. Hum-Centric Intell Syst. 2023;4(1):77-92. doi:10.1007/s44230-023-00050-2
64. Melak A, Aseged T, Shitaw T. Influence of artificial intelligence technology on the management of livestock farms. Int J Distrib Sens Netw. 2024;2024:1-12. doi:10.1155/2024/8929748
65. Patel H, Samad A, Hamza M, Muazzam A, Harahap MK. Role of artificial intelligence in livestock and poultry farming. Sinkron. 2022;7(4):2425-2429. doi:10.33395/sinkron.v7i4.11837
66. Cho Y, Kim J. AI-based intelligent monitoring system for estrus prediction in livestock industry. Appl Sci. 2023;13(4):2442. doi:10.3390/app13042442
67. Nagahara M, Tatemoto S, Ito T, Fujimoto O, Ono T, Taniguchi M, et al. Designing a diagnostic method to predict optimal artificial insemination timing in cows using artificial intelligence. Front Anim Sci. 2024;5:1399434. doi:10.3389/fanim.2024.1399434
68. Haleem A, Javaid M, Qadri MA, Singh RP, Suman R. Artificial intelligence applications for marketing: A literature-based study. Int J Intell Netw. 2022;3:119-132. doi:10.1016/j.ijin.2022.08.005
69. Ljepava N. AI-enabled marketing solutions in marketing decision making: Application of AI in different stages of the marketing process. TEM J. 2022;11(3):1308-1315. doi:10.18421/TEM113-40
70. Elufioye OA, Ike CU, Odeyemi O, Usman FO, Mhlongo NZ. AI-driven predictive analytics in agricultural supply chains: Benefits and challenges. Comput Sci IT Res J. 2024;5(2):473-497. doi:10.51594/csitrj.v5i2.817
71. Hongbing W, Jing G, Bohan K, Peng L, Yuxian S. Marketing strategy of agricultural products based on artificial intelligence. Math Probl Eng. 2022;2022:1-7. doi:10.1155/2022/7798640
72. Sakr GE, Elhajj IH, Mitri G, Wejinya UC. Artificial intelligence for forest fire prediction. In: Proc IEEE/ASME Int Conf Adv Intell Mechatronics; 2010; Montreal, QC. IEEE; p. 1311-1316. doi:10.1109/AIM.2010.5695809
73. Shivaprakash KN, Swami N, Mysorekar S, Arora R, Gangadharan A, Vohra K, et al. Potential for artificial intelligence and machine learning applications in biodiversity conservation and forest management in India. Sustain. 2022;14(12):7154. doi:10.3390/su14127154
74. Da Silva DQ, Dos Santos FN, Filipe V, Sousa AJ, Oliveira PM. Edge AI-based tree trunk detection for forestry monitoring robotics. Robotics. 2022;11(6):136. doi:10.3390/robotics11060136
75. Honarmand Ebrahimi S, Ossewaarde M, Need A. Smart fishery: A systematic review and research agenda for sustainable fisheries in the age of AI. Sustain. 2021;13(11):6037. doi:10.3390/su13116037
76. Mohale HP, Jawahar P, Jayakumar N, Arul Oli G, Ravikumar T. Application of deep learning in marine fisheries resource management. Trends Agri Sci. 2023;2(9):753-763.
77. Ahmed MS, Aurpa TT, Azad MdAK. Fish disease detection using image-based machine learning technique in aquaculture. J King Saud Univ Comput Inf Sci. 2022;34(8):5170-5182. doi:10.1016/j.jksuci.2021.05.003
78. Rejeb A, Rejeb K, Zailani S, Keogh JG, Appolloni A. Examining the interplay between artificial intelligence and the agri-food industry. Artif Intell Agric. 2022;6:111-128. doi:10.1016/j.aiia.2022.08.002
79. Kakani V, Nguyen VH, Kumar BP, Kim H, Pasupuleti VR. Critical review on computer vision and artificial intelligence in food industry. J Agric Food Res. 2020;2:100033. doi:10.1016/j.jafr.2020.100033
80. Misra NN, Dixit Y, Al-Mallahi A, Bhullar MS, Upadhyay R, Martynenko A. IoT, big data and artificial intelligence in agriculture and food industry. IEEE Internet Things J. 2022;9(9):6305-6324. doi:10.1109/JIOT.2020.2998584
81. Kutyauripo I, Rushambwa M, Chiwazi L. Artificial intelligence applications in agrifood sectors. J Agric Food Res. 2023;11:100502. doi:10.1016/j.jafr.2023.100502
82. Golshani T. Role of artificial intelligence in managing risk in agricultural engineering. SSRN Electron J. 2024. doi:10.2139/ssrn.4842193
83. Wakchaure M, Patle BK, Mahindrakar AK. Application of AI techniques and robotics in agriculture: A review. Artif Intell Life Sci. 2023;3:100057. doi:10.1016/j.ailsci.2023.100057
84. Subeesh A, Mehta CR. Automation and digitization of agriculture using artificial intelligence and internet of things. Artif Intell Agric. 2021;5:278-291. doi:10.1016/j.aiia.2021.11.004