Mobile Robotics Lab for In-Situ Sampling and Measurement
Precision agriculture is the process of integrating sensing devices and analysis tools to improve crop yields while maintaining sustainability. Generally, precision agriculture can be subdivided into categories pertaining to stress detection, prescription plans and yield maps. Stress detection meaning the nutrients available to the crops and from there prescription maps and yield maps can be created to adjust the nutrients needed to improve crop health. The first step to use the services provided by precision agriculture is to create an in-depth analysis of the land and its sources of nutrients such as the soil. This step is the primary focus of the Agrobot Project. The project’s goal is to develop a networked fleet of sensors and robots to gather real time data on crop health and thereby improve the sustainability of the farmland. The Agrobot Project looks to develop a systemic approach to grid type analysis of the soil properties of the land. The soil’s properties are measured using Acclima SDI-12 Sensor Readers. The soil sensor is posted in the field collecting data in preparation to transmit to the ROSbot, which is a terrestrial drone capable of acting autonomously.
Kevin’s main focus in the project is finding a solution to the complexities of the connection between the sensor and the drone. The soil sensor does not have a way to directly connect to the ROSbot, nor is there a way for the ROSbot to quickly gather the information provided by the sensor as it passes through the field. Based on the assumption that there is no way to physically connect the drone to the sensor the only logical way to proceed is to provide a wireless connection between them to transfer the data. This idea led to the overall plan needed to transmit the data, the soil sensor already has an SDI-12 bus that can be configured to connect to an Arduino device per an Arduino-SDI-12 library. This microcontroller will then store the data it has collected from the sensor on an external SD-Card found on a break-out board. This is once again available to use due to an Arduino-SD-Card library. This breakout allows for the microcontroller to continuously write to the SD card storing the data from the soil sensor. The microcontroller then sends the data via bluetooth to the ROSbot when it is pinged, in this way th ROSbot can inform the microcontroller that it is in close enough range to receive the data. Then the sensor data is saved as a .csv file within the on board computer of the ROSbot and finally a publisher node uses this messages file to make the sensor data available to the rest of the drone’s functions.
The work that Kevin has done dealing with the sensor communication to the ROSbot will help the project reach its overall goal of creating more sustainable farm practices with the use of technology because it brings farms closer to the tenets of precision agriculture. The most important of them being creating yield maps and prescription plans. The work that he does with the sensor communication allows for the farm handlers to understand the specific needs of certain sections of the land. The soil information allows for the farm management team to decide which areas of the land require fertilization or other types of attention and which do not. The sensor communication research contributes to the development of precision agriculture practices.