EPAL's IoT server engine (http://www.servers.epalab.org:9090 ) based on ThingsBoard open-source software is a centralized platform designed to collect, store, and analyze data from multiple IoT devices deployed in agricultural settings. This scalable and customizable solution can cater to the diverse needs of farmers, researchers, and other stakeholders in the agricultural sector.

Introducing EPAL's IoT server engine brings several benefits to the agricultural community and beyond. Firstly, it enables real-time monitoring and management of agricultural processes and empowers farmers to respond promptly to changing conditions while optimizing resource allocation. Secondly, by aggregating and analyzing data from multiple sources, the IoT server engine facilitates data-driven decision-making, thereby enhancing productivity and sustainability. Additionally, the open-source nature of ThingsBoard promotes transparency, collaboration, and innovation within the agricultural community.

EPAL's latest groundbreaking innovation, the MoFaRo, represents a significant leap in agricultural technology. This advanced machine specializes in autonomous chemical spraying, eliminating the need for human involvement in potentially hazardous tasks and ensuring the safety of our workforce. Equipped with cutting-edge camera technology, the MoFaRo can accurately scan and locate pests or diseases, enabling targeted and precise spraying. Furthermore, the integration of cameras and LiDAR scanning enables the machine to possess high throughput crop phenotyping capabilities. This functionality proves invaluable to crop breeders as they can assess the crop's resistance to pests and diseases without the need for labor-intensive manual scouting. The MoFaRo stands as a remarkable solution that enhances efficiency, safety, and data-driven decision-making in modern agriculture.

This work is being done by Dr Kadeghe Fue and Mr Dickson Massawe.

AgriCare is an application for surveillance and monitoring of crop diseases that uses the YEESI Dataset to accurately detect diseases and pests that affect bell peppers. With the help of artificial intelligence, the app can quickly identify crops and diagnose specific ailments by analyzing images taken on a mobile phone. In the future, AgriCare will be integrated into the Mwagilia App to provide farmers with comprehensive support. Additionally, we plan to include the ability to track the population dynamics of disease vectors for main staple crops using remote sensing technology.

This work was done by Dr Kadeghe Fue and Mr Fikiri Matatizo.

The Mwagilia App serves as an irrigation tool that provides guidance on irrigation scheduling by leveraging satellite data. Additionally, it offers advice on fertigation practices when supplied with drone imagery. By utilizing these advanced technologies, the app empowers farmers with accurate recommendations for efficient irrigation management and optimal fertilization practices.

This work was done by Dr Kadeghe Fue and funded by SG-NAPI, SUARIS and FSNET Africa.

This is an autonomous robot that can harvest cotton bolls.

A rover with centre articulation and hydrostatic propulsion, accompanied by a Cartesian manipulator, was conceptualized and put into action. This robot incorporated advanced sensing systems, including encoders, a low-cost RTK-GNSS, a potentiometer, RGB stereo cameras, and IMUs, to facilitate navigation and manipulation during the picking process. To enable precise object detection and control for harvesting cotton bolls, three controllers were integrated into the robot. The overall robotic system, responsible for tasks such as cotton boll tracking, estimating cotton boll locations, detecting cotton rows, navigation, and harvesting, was controlled and coordinated using the Robot Operating System (ROS). Autonomous localization and navigation of the robot were accomplished through the utilization of the Extended Kalman Filter (EKF) sensor fusion algorithm. For cotton harvesting, a ROS-independent finite state machine (SMACH), a modified pure pursuit algorithm, and a proportional-integral-derivative controller were employed. The robot's performance was assessed and documented, revealing its precise and efficient navigation over cotton rows and successful harvesting of cotton bolls. Additionally, the robotic design demonstrated the feasibility of adopting traditional vacuum harvesting in robotic systems for cotton boll harvesting. The successful development of this robot marks an important milestone in enhancing cotton harvesting management.

This work was done by Dr Kadeghe Fue in collaboration with Prof Glen Rains from the University of Georgia

A tool has been developed to accurately measure the downforce pressure of planters, aiming to generate a comprehensive dataset for precision planting systems. This tool takes into account various factors that can impact seed emergence, including seed depth, downforce pressure, soil texture, soil moisture, and soil structure. The project was conducted in collaboration between Dr. Kadeghe Fue and Prof. Wesley Porter from the University of Georgia, ensuring expertise from both parties.

A cutting-edge web-based tool, utilising the Water Requirement Specification Index (WRSI), has been developed to evaluate crop performance across Tanzania. This innovative tool enables the monitoring and prediction of crop performance, explicitly focusing on crops like Maize. The invaluable insights gained from this tool can be shared with practitioners and government agencies, facilitating informed decision-making in the agricultural sector.

This collaborative project involved the expertise of Prof. Siza Tumbo, Dr. Kadeghe Fue, and a team of dedicated scientists from the Sokoine University of Agriculture (SUA) in Tanzania and the University of Hohenheim, Germany. To explore the capabilities of this tool firsthand, you can visit our servers at http://climate.sua.ac.tz . 

This project was funded by the extra-mural project called Innovating Strategies to Safeguard Food Security using Technology and Knowledge Transfer (Trans-Sec)

In 2014, EPAL successfully developed and patented a groundbreaking technology in collaboration with the University of Florida. The invention, titled "Solar-Powered, Wireless Re-Programmable Precision Irrigation Controller," was the result of the collaborative efforts of Dr. Kadeghe Fue, Prof. Siza Tumbo, Prof. Arnold Schumann, and Prof. John Schueller.

This innovative technology, registered under Patent Number TZ/P/14/00383 by BRELA-Tanzania, offers a solar-powered and wireless irrigation controller that can be re-programmed for precise irrigation management. The tool garnered significant recognition and was awarded the Best Student Paper at the Pan African Conference on Science, Computing, and Telecommunications (PACT) in Arusha, Tanzania, on July 17th, 2014.

For more information about the technology and EPAL's other innovations, you can visit the SUA Technologies website at [https://www.dprtc.sua.ac.tz/sua-technologies/].

2014 Innovation