Towards a Smart Farming Platform: From IoT-Based Crop Sensing to Data Analytics

Autor: Héctor Cadavid, Germán López, Alexander Pérez, Cristian Mendivelso, Carlos Ramírez, Wilmer Garzón
Přispěvatelé: Informática
Jazyk: angličtina
Rok vydání: 2018
Předmět:
Zdroj: Communications in Computer and Information Science ISBN: 9783319989976
Popis: Colombia is a country with a huge agricultural potential, thanks to its size and geography diversity. Unfortunately, it is far from using it efficiently: 65% of its farmland is either unused or underused due to political problems. Furthermore, vast of Colombian agriculture is characterized - when compared with other countries - by low levels of productivity, due to the lack of good farming practices and technologies. The new political framework created by the recently signed peace agreement in this country opens new opportunities to increase its agricultural vocation. However, a lot of work is still required in this country to improve the synergy between academia, industry, agricultural experts, and farmers towards improving productivity in this field. Advances in smart-farming technologies such as Remote Sensing (RS), Internet of Things (IoT), Big Data/Data Analytics and Geographic Information Systems (GIS), bring a great opportunity to contribute to such synergy. These technologies allow not only to collect and analyze data directly from the crops in real time, but to extract new knowledge from it. Furthermore, this new knowledge, combined with the knowledge of local experts, could become the core of future technical assistance and decision support systems tools for countries with a great variety of soils and tropical floors such as Colombia. Motivated by these issues, this paper proposes an extension to Thingsboard, a popular open-source IoT platform. This extended version aims to be the core of a cloud-based Smart Farming platform that will concentrate sensors, a decision support system, and a configuration of remotely controlled and autonomous devices (e.g. water dispensers, rovers or drones). The architecture of the platform is described in detail and then showcased in a scenario with simulated sensors. In such scenario early warnings of an important plant pathogen in Colombia are generated by data analytics, and actions on third-party devices are dispatched in consequence.Colombia is a country with a huge agricultural potential, thanks to its size and geography diversity. Unfortunately, it is far from using it efficiently: 65% of its farmland is either unused or underused due to political problems. Furthermore, vast of Colombian agriculture is characterized - when compared with other countries - by low levels of productivity, due to the lack of good farming practices and technologies. The new political framework created by the recently signed peace agreement in this country opens new opportunities to increase its agricultural vocation. However, a lot of work is still required in this country to improve the synergy between academia, industry, agricultural experts, and farmers towards improving productivity in this field. Advances in smart-farming technologies such as Remote Sensing (RS), Internet of Things (IoT), Big Data/Data Analytics and Geographic Information Systems (GIS), bring a great opportunity to contribute to such synergy. These technologies allow not only to collect and analyze data directly from the crops in real time, but to extract new knowledge from it. Furthermore, this new knowledge, combined with the knowledge of local experts, could become the core of future technical assistance and decision support systems tools for countries with a great variety of soils and tropical floors such as Colombia. Motivated by these issues, this paper proposes an extension to Thingsboard, a popular open-source IoT platform. This extended version aims to be the core of a cloud-based Smart Farming platform that will concentrate sensors, a decision support system, and a configuration of remotely controlled and autonomous devices (e.g. water dispensers, rovers or drones). The architecture of the platform is described in detail and then showcased in a scenario with simulated sensors. In such scenario early warnings of an important plant pathogen in Colombia are generated by data analytics, and actions on third-party devices are dispatched in consequence.
Databáze: OpenAIRE