2018 ASHS Annual Conference

Fresh market tree fruit harvesting is a difficult task that relies entirely on manual labor, but there exists a clear need for the technology in today’s economy. Extensive research has been done on the development of mechanical harvesting techniques. Several selective harvesting robots have been developed for research studies, but there are no commercially available robotic systems.

This talk summarizes recent advances in mechanical tree fruit harvesting technologies. Because the fresh market apple industry is an agricultural sector with significant potential for the incorporation of robotic harvesting technologies, factors and considerations that are particularly applicable to apple picking are highlighted. In addition, it presents two studies of the design and field evaluation of robotic apple harvesters that have been conducted recently. In order to assess required functionality in modern orchard systems with ideal fruit distributions, an undersensed, low-cost system was developed. Based on local growers’ willingness to modify the tree to optimize fruit distribution for robotic harvesting, apples adjacent to trellis wires and trunks were removed prior to field studies. The robotic system integrated a global camera set-up, seven DOF manipulator, and grasping end-effector of 3D printed tendon driven fingers to execute fruit picking with open-loop control.

The design and development of a novel 3D printed soft robotic end-effector to facilitate apple separation was also presented. In this study, the field test was conducted in an unmodified field environment. Additionally, a secondary robot was implemented to catch the harvested fruit to facilitate fruit collection as an approach to potentially reduce the overall cycle time of robotic tree fruit harvesting. The soft robotic end-effector performed very well in grasping apples and facilitating apple separation. The compliant actuators were unharmed in collisions with the tree canopy and trellis wires offering a significant improvement in grasping speed over the tendon driven fingers.

Results from field studies show that horticultural practices play a critical role in the selection of functionality requirements. Improved harvesting efficiency will require enhanced robustness, especially obstacle detection with increased visual sensing and force sensing on the end-effector for feedback on grasp status.