NUMERICAL STUDY AND DEVELOPMENT OF A SEMI-AUTONOMOUS ROBOT FOR FARM APPLICATIONS
Abstract
Weed infestation remains a major constraint to cassava production in Nigeria and other developing countries, leading to reduced yield, increased labor cost, and excessive reliance on manual weeding and indiscriminate herbicide application. This project presents the design, development, and preliminary evaluation of a semi-autonomous robotic cassava weeder intended to improve efficiency, safety, and precision in weed management on cassava farms.
The developed system was designed as a semi-autonomous differential drive mobile robot, powered by two rear-mounted 12 V DC wiper motors, with a passive fixed-axle front wheel assembly to enhance stability on uneven agricultural terrain. An Arduino Mega 2560 served as the main control unit, coordinating motion control, obstacle detection, and spraying operations. Obstacle avoidance was implemented using HC-SR04 ultrasonic sensors, while remote monitoring and human-in-the-loop control were achieved through an ESP32-CAM module, providing live video feedback and wireless command transmission. The weeding operation was carried out using a 12 V DC diaphragm pump and spray nozzles for targeted herbicide application between cassava ridges.
Mechanical fabrication, subsystem integration, analytical evaluations, and bench-level testing were conducted to assess system functionality and readiness. Results from these evaluations indicated reliable control response, stable locomotion, effective obstacle detection within a 30 cm threshold, and consistent spray delivery under controlled conditions. Power analysis showed that the system could operate for approximately 1.5–2 hours using a 12 V rechargeable battery, depending on load conditions.
At the time of submission of this report, extended field trials were not completed. Preliminary results however demonstrated that, the developed robotic cassava weeder is technically viable and suitable for small- to medium-scale cassava farms. Future work will focus on full-scale field testing, performance optimization, and integration of autonomous navigation and machine vision–based weed detection.
Keywords: Robotic weeder; Cassava farming; Precision agriculture; Semi-autonomous robot; Differential drive; Weed control.
Full-Text Access Notice
In accordance with the NERD Policy on promoting peer-reviewed publication, public access to the full text of a project, thesis or dissertation is restricted for three years, allowing the author and supervisors sufficient time to pursue peer-reviewed publication.
During this period, researchers with legitimate academic or research purposes may request authorisation directly from the author to enable NERD to release the indexed full texts of the work using the form below.