明星八卦

The monitoring of the growth of healthy crops, effective resource management and soil preservation have always faced major issues in agriculture. Many methods have been developed over the years, but they have always required a lot of human work. With the progress of solid-state electronics and miniaturization of transistors, it is now possible to develop small electronic devices equipped with sensors to have a better view of the conditions of the soil and take informed actions based on the data. This will allow the optimization of resources utilisation and possibly the forecast of the soil conditions based on historical data. This project investigates the development of such a system by reviewing the current state of the art systems from literature and solutions already available on the market. Such a system is developed in collaboration with Small Robots which is a UK based start-up developing robots that can take autonomous decisions and take care of weed infestations. The soil monitoring system will eventually become part of their solution to monitor the soil and have a better view of its conditions and hence make informed decisions.

This project investigates the design of an Internet of Things Soil (IoT) Soil Monitoring System that can either be burrowed into the ground or be deployed by a mobile robot. The system allows the user to monitor the soil conditions remotely and take informed action based on soil and weather conditions. A field may have 10s or 100s of sensors deployed in it. Communications and power will be key design considerations. The nodes sense air relative humidity, temperature and soil relative moisture and temperature. The nodes send the data wirelessly to a web platform. A Virtual Sensing approach is developed using Neural Networks to predict soil temperatures. The historic data obtained from the nodes are stored and visualised on a web platform. The deployment of such a system allows robots developed by Small Robots to make autonomous decisions and take actions based on soil and weather conditions. Further developments on the Virtual Sensing approach might allow the reduction of the nodes’ sampling frequency eventually saving battery power.

The aim of the project is to develop a completely autonomous, low cost, low power and scalable up to 100s of nodes Internet of Things Soil Monitoring System. Store and visualise the historic data obtained from the nodes on a web platform. Achieve a minimum of 12 months of battery life with 1h sampling frequency. Allow the user to monitor the soil and weather conditions remotely, take informed actions and predict soil temperatures allowing the optimization of resources utilisation.