Food is an elementary necessity for human existence that is fulfilled through agriculture. Agriculture is an indispensable segment for economy of any country (Meshram et al., 2021). Inopportunely, its market is unpredictable as several factors affect agriculture. For example: drought or flood can easily influence future commodity prices and have grave consequences on all food prices. The only approach to come across the growing food demand is to keep track of the environment, crops and the market. This is the time when machine learning plays role in agriculture. Agriculture decision making is empowered by machine learning applications through the analysis of real-time sensor data and past trends. This supports the manufacturers for improving crop yields, better predicting demand and reducing the food production costs. Thus, ML applications in agriculture can optimize the food production and revolutionize the economy of any country.
Smart Farming and Precision Agriculture
Last seven decades have witnessed the use of modern techniques and tools in agriculture and new concepts like smart farming and precision agriculture have emerged. Though, a large number of farmers are far far away from these, but the future is calling to all such farmers in order to improve quality and quantity in agriculture. Smart farming and precision agriculture are very often used interchangeably, but these have slight difference in sense. In Smart farming, data and information techniques are used for optimizing complicated farming systems. Emphasis is on the smart way of applications of collected information instead of on access to data and its application. In precision agriculture, for monitoring and optimizing the production processes, recent farming management ideas together with digital techniques are employed. Here, optimization plays vital role. Precision agriculture consents the determination of actions on the basis of per plant or per square meter. For example, if one has to use fertilizers in his field, the precision agriculture suggests him not to use equal amount of fertilizers over whole field, but to measure the soil variations within the field and use the fertilizers accordingly.
In smart farming, the technologies like deep learning, machine learning, artificial intelligence, sensors, robots, internet of things (IoT), etc. (Tripathi et al., 2022, Goel et al., 2022) are used extensively. Farmers have great opportunity to upsurge both the quantity and quality of crops (Tyagi, 2016).
Figure 1.
Smart farming in action (Thales/Nesta)
Some technologies being used in smart farming are:
- •
Location systems: This includes satellites, GPS, etc.
- •
Sensors: These are used for monitoring temperature, moisture, light soil, water, etc.
- •
Robots: These are used for seeding, spaying the herbicides, pesticides, etc.
- •
Drone cameras: These are used to monitor the crops status.
- •
Precision irrigation and plant nutrition
- •
Different software platforms
- •
Platforms for analysis and optimization.
Figure 2.
Use of drones in smart farming
A UN report says that by 2050, the world population will be 9.7 billions that is 1.73 billions more people will require food in comparison to September 2022. As per the estimation of United Nations, world population in September 2022 is 7.97 billions. Unless and until, more and more farmers use the smart farming, it will not be possible to feed all in 2050.
Use of Smart farming is very beneficial for farmers as well as society. Through smart farming production can be increased, water can be saved, quality can be improved, costs can be reduced, environment impact can be minimized, etc.