(4)INESC TEC
A digital twin consists of a digital representation of a real-world object or infrastructure. Allied to real-time data, a digital twin can be used for remote monitoring and to simulate or predict a set of actions. This concept is characterised by three components: a physical entity, a virtual entity and the data that connects these two parts. This technology allows the visualisation and extraction of information in real time about an entity, eliminating the need for physical contact, and can be applied to an organisation, supporting its management and monitoring.
Digital twin technology has been applied in healthcare, automotive, aeronautics and aerospace, but can also help solving many challenges in other scopes. For example, in agriculture, it helps managing problems such as resource management, food security, weather concerns and monitoring of soil and land. More recently, there have also been efforts in creating digital twins of forest areas, as they could help in tasks such as forest management planning, inventory and harvesting plans, assess carbon calculations, understanding and monitoring the effects of drought and disease on trees.
When it comes to farming, digital twins can aid farmers by monitoring and controlling operations remotely, based on real-time digital information, thereby reducing the time and effort spent on manual tasks on-site. Farmers can be automatically informed if there is a problem and can simulate and evaluate corrective and preventive actions on the digital representation. This technology also helps farmers to minimise risks from factors such as weather, as well as to increase profitability.
There are many operations involved in a farming digital twin. Some of its possible uses are to acquire answers concerning questions that are often not easy to answer without some context, such as the amounts of fertiliser needed to a certain region of the land, land preparation times, meteorological values, historical values, among others.
The physical world requires measurement technologies and sensors to collect and receive data from the physical object. These measurements may be acquired through various technologies such as weather stations to monitor the environment, overall air quality and predict the weather status. Light sensors can be used to measure sunlight exposure on growing plants. When it comes to monitoring the soil, optical and electro-chemical sensors can be used to determine fertility and measure organic matter and moisture contents of the soil, along with mechanical sensors to measure compaction, deformation, and resistance. These measurements help to identify where and how the resources are stressed, whether by invasive plants and animals, soil quality, pollution, or other factors.
The use of drones can provide substantial support in monitoring activities and other tasks. They can be mounted with various sensors and actuators to perform agricultural tasks, from gathering data to sweeping terrains and sectors with fertilisers and seeds. In farming, drones also aid in gathering data for precision agriculture by producing map projections based on multispectral imaging. These can help to determine land boundaries that can grow crops. In forest management, drones can help to monitor large sections for environmental and ecological changes. They also acquire observation data such as the topography of the terrain, how and where trees are placed, thus providing support to build a virtual model.
The current advancements in technology, and the rise of methods such LiDAR, made it possible to create full 3D scenes from observations and measurements, more easily and manageably, which enable building virtual worlds that represent reality accurately a much more feasible process [2].
Autonomous farm equipment can provide a fundamental support in saving costs and time involved in harvesting and yielding compared to manual farming machinery [3]. They can provide several advantages such as precise, fast, and repetitive operations, regardless of weather conditions. Due to these advantages, extensive use of unmanned vehicles such as tractors, as well as combine harvesters, has been increasing [4].
Nowadays, there are several real-world examples of tools that use Digital Twin technology as their basis of operation, due to the already enumerated advantages of its use:
For example, Intelligent Growth Solutions (IGS) has developed a solution focused on indoor controlled farming [5], which is able to tweak parameters such as light, water, nutrients, humidity, and temperature,and see how they affect plants. This solution scans the crops using cameras that capture two-dimensional and three-dimensional images, and it also uses sensors to measure water and nitrates.
Connecterra has created an intelligent cow-monitoring system based on artificial intelligence (AI) that allows the monitoring of health and well-being of a herd [6]. By attaching a sensor to a cow's collar, as well as other data sources, this system can generate alerts regarding an animal's health, heat, and operational changes, helping a farmer locate an injured cow.