SPIDER: Semi-Autonomous Precision Intelligent Detection and Elimination Robot, Phase III

Project number
25008
Organization
UA Department of Biosystems Engineering
Offering
ENGR498-F2024-S2025
Weed overgrowth on golf courses presents a significant challenge to turf management. It requires frequent maintenance that is both labor-intensive and costly. Traditional methods rely on manual labor or widespread herbicide application. Both of these methods can be inefficient and environmentally harmful. To address these challenges, the team developed SPIDER, an autonomous, data-driven agriculture robot. SPIDER can streamline turfgrass maintenance with precise weed detection and
elimination, particularly for environments like golf courses.

SPIDER’s system uses both software and hardware to enable autonomous movement, object avoidance and weed detection. Its mechanical assembly includes a main body that houses all electronic components and four legs on each side of the main body for mobility. A mechanical arm mounted on a rotational surface atop the main frame operates within a 90-degree arc between the front legs for precise weed removal. The weed detection software uses machine learning technology with PyTorch’s ResNet18 model. It is integrated with the mechanical elimination system by the repository running on an NVIDIA Jetson Nano AI platform to ensure seamless communication between software and hardware components.

Get started and sponsor a project now!

UA engineering students are ready to take your project from concept to reality.