Abstract

Agricultural robotics is a rapidly growing industry with the possibility to revolutionize food production. Agricultural robots often require custom sensing solutions to perform specific tasks. Developing the appropriate sensing capabilities for an application is therefore an important step in the automation process. In this thesis I present two case studies in developing sensing capabilities for agricultural robotics applications. First, I introduce a sensor that can provide simultaneous measurements of force and position. The sensor was developed to enable high-throughput measurement of maize stalk stiffness on a robot platform. It uses multiple sets of strain gauges on a cantilever beam to obtain information about the magnitude of an applied force and its position along the beam. Three different prototypes for the sensor concept were designed, tested, and validated. The results indicate that the force-position sensor prototypes are highly linear and accurate, with the fit of measured positions, forces and angles to true applied values having an R2 of over 0.98. I address the parameters that influence performance, design considerations, potential applications, and limitations of the force-position sensor concept. Preliminary results from applying the sensor to stalk stiffness measurement are also presented. Second, I demonstrate the use of a deep neural network object detection model as a solution for in-field detection of saffron flowers and evaluation of flower harvestability. The model used the YOLOv10b architecture and was trained on a dataset consisting of 35,472 images of saffron. This dataset is 10x larger than any other saffron image database collected to date, and the only one to include saffron throughout its emergence, growth, and flowering stages. The trained model achieves an Average Precision of 93.1% in identifying and distinguishing harvestable saffron from immature saffron. This model is an important step towards automated saffron harvesting using agricultural robots.

Degree

College and Department

Ira A. Fulton College of Engineering; Mechanical Engineering

Rights

https://lib.byu.edu/about/copyright/