Volumetric mass flow sensor for citrus mechanical harvesting machines

• Volumetric mass flow system was developed to monitor citrus yield on harvester conveyor.
• Algorithms were developed to estimate total fruit mass (yield) based upon scanned volume.
• Average fruit mass estimation error was 7% in laboratory tests and 10% in field tests.

A volume-based mass flow sensor system was developed to estimate the total mass of fruit travelling on the inclined and horizontal conveyors of two citrus mechanical harvesting systems. A LIDAR (light detection and ranging) sensor was used to scan the conveyor cross-sectional area and use the distance related-information to estimate the bulk volume and total mass of fruit on the conveyor by integrating the data over time. A custom algorithm was developed to analyze the data and calculate the fruit volume. Fruit volume was used to estimate the total fruit mass.Sensor system was tested in the laboratory on two different conveyor systems: an inclined conveyor used in a mechanical harvester (conveyor flap height = 8 cm), and a horizontal conveyor used in a debris removal system (conveyor flap height ⩽3 cm). The tested conveyor speeds ranged from 0.59 m/s to 1.71 m/s. The system performance for the tested speed ranges was an average error of 7% (standard deviation [SD] ± 7%) and 7% (SD ± 5%) for the inclined and horizontal conveyors, respectively. The average total fruit mass estimation error during mechanical harvesting was 10% (SD ± 6%) for 11 field trails that involved scanning 512–1356 kg of the fruit on a horizontal conveyor of a debris removal system. Overall, the developed sensor system can be used for volumetric yield monitoring of citrus. The sensor system performance will likely be improved using instant conveyor speed during the estimation of total fruit mass.