Advanced technologies for monitoring and growing crops

The growing consumption of food worldwide requires increasing the amount of agricultural productions, and the development of new production technologies that must meet the increased needs, while reducing anthropogenic pressure on the environment. Precision farming is an innovative concept that included a satellite positioning systems, sensor technologies for data collection, and geographic information systems. It integrates the effects of various spatial variables such as soil properties, topography, plant population, fertilization, irrigation, and pest attacks. Modern technologies for monitoring the condition of cultivated plants include the application of a number of sensor technologies through aerial remote sensing, or ground systems. Most of the tools for remote analysis of vegetation are based on colour images and spectral sensors. The reflection of electromagnetic radiation in the visible and near infrared region of plants depends on many factors — the surface of the leaves, their structure, chemical composition, water content, pigments, the presence of abiotic and biotic stress, etc. Thermal sensors determine the temperature of the plants, which correlates with the water status of the plants. Fluorescent sensors are used to determine the chlorophyll content in plants. The new multi and hyperspectral sensors allow the acquisition of both spatial and spectral information from any point of the studied plants. Spectral sensors and computer vision can be used not only for the analysis of plants, but also on the composition and quality of production. Various methods for image processing in the visible range and mathematical methods for quantitative analysis and classification based on the obtained spectral characteristics and hyperspectral images of plants have been developed.
Sensors for the plant and the soil analysis can be located on satellites, aircrafts or unmanned aerial vehicles (UAV). Satellites provide a global overview and regular information, but the resolution in most cases, especially free data, is a low and not applicable to small areas. UAVs provide unique opportunities for a low altitude flight, and hence to obtain high resolution images, low’s operating costs, less dependence on weather conditions, a multiple repetition of observations. For the use of UAVs in plant protection measures, the accuracy and the reduction of the positioning error must be improved. This can be done by applying an RTK GNSS system to reduce an error to 1-2 cm.
Another approach to improving the efficiency of agriculture is related to the use of robots in agriculture. The current understanding is to create enough intelligent machines to be able to make independent decisions and to work in a natural environment without taking away human control functions. These robots can work continuously, perform functions in bad weather conditions or in reduced visibility. The use of such autonomous machines at a sowing, the processing of various crops, a pest and a weed control, an irrigation, and a treatment of the plants, a remote monitoring and an evaluation of the damage to the crops are very important. Agricultural robots need to have specific equipment for the specific tasks to be able to perform such a wide range of activities. Agrorobots also need to have a wide range of sensors to collect the necessary information and make adequate decisions to act in specific conditions.
The research teams that will work in this work package will do research related to:
• Creation of an autonomous control system for an agricultural robot and an unmanned aerial vehicle and specific equipment for them to perform various tasks.
• Development and research of alternative technologies through robot and UAV for: monitoring of soil conditions and tillage; for monitoring and destruction of weeds; monitoring and implementation of integrated pest management; for monitoring and removal by physical and mechanical method of pests in the cultivation of agricultural crops; monitoring and establishing the effect of topical application of pesticides, foliar fertilizers and anti-stress products when using UAVs.
• Application of remote and non-destructive methods (spectral characteristics and digital and hyperspectral images) for: a creation of models for a classification of a tomato plants for the assessment of a water stress and a nitrogen deficiency in plants; a creation of models for a classification of tomatoes, depending on the degree of maturity; an early diagnosis of a fungal diseases on crops grown in an aquaponic system; management of an irrigation regimes of a vegetable crops and soils; a development of a method for correlation of a data from sensors and remote measurements in order to establish relationship between plants’ temperature and a soil moisture at a different depths; determining the most informative wavelengths and vegetative indices for remote diagnosis of diseases in the main orchards; a creating a database of images of Sentinel-2A and Sentinel-2B from areas with a major field crops (Winter common wheat, Sunflower, Maize and peas), generating a data including vegetation indices, quantitative and qualitative characteristics of crops and establishing a risk factors of an abiotic and a biotic stress, identifying the spread of diseases and pests in these areas.
• Investigation of the MOHID-Land model to simulate the dynamics of soil moisture.
• Formation of time series of data on the state of the constellation of the GNSS GALILEO; calculation of accuracy criteria (DOP) based on the collected data; analysis and evaluation of the expected accuracy in determining the location of the UAV in the three-dimensional space.
Targeted scientific and applied research related to the application of remote sensing, artificial intelligence and robotic technologies in agriculture will reduce farmers’ costs, increase productivity, limit the use of plant protection products and reduce greenhouse gas emissions.