Estimating Tall Fescue and Alfalfa Forage Biomass Using an Unmanned Ground Vehicle

Improving the efficiency and utilization of forage crops is important for enhancing productivity. Precise assessments of biomass levels play a vital role in assisting producers of hay, silage, and grazers in determining the optimal harvest timing and achieving an efficient stocking rate. This study aimed to develop a pre-harvest biomass estimation method by utilizing crop height measurements obtained through a ski-shaped plate mounted on an unmanned ground vehicle (UGV). The research focused on evaluating the performance of the proposed method on tall fescue (Schedonorus phoenix) and alfalfa (Medicago sativa) trial plots. Measurements on the plots were made and then harvested at approximately 10, 20, and 30-day intervals. The ski-shaped plate, referred to as the “compression ski,” was constructed and attached to a ground vehicle. Continuous height measurements of tall fescue and alfalfa were captured using an ultrasonic sensor and a microcontroller. Geotagged measurement locations were acquired through an RTK-GPS receiver mounted on the UGV. Plot boundaries were determined using aerial images processed in Agisoft Metashape and ArcGIS Pro. The compressed height indicators obtained were correlated with wet yield measurements, leading to the development of estimation models. Dry matter fractions (DMFs) of the crops were determined using oven drying method. The wet yield estimation models were combined with DMFs to create a prediction model for dry matter yield (kg DM/ha). The best-performing compression ski-based model achieved a standard error of 291 kg DM/ha for tall fescue and 491 kg DM/ha for alfalfa. The proposed pre-harvest biomass estimation method utilizing compressed height measurements with a UGV and a ski-shaped plate offers a promising approach to optimize efficiency and utilization in grassland management. This method holds potential for enhancing productivity and resource utilization in the aforementioned areas of forage production systems.