Orchard Automation - Harvest Assist Technologies

Labor costs associated with fruit harvest are roughly 40% of an orchard enterprise annual budget, and fruit industry has identified harvest mechanization and automation as a research priority.
Orchard Automation - Harvest Assist Technologies - Articles


Mechanical harvest aids offer the potential for more efficient harvest and increased consistency in fruit handling. However, in-field bin filler technologies available at the start of our research resulted in excessive bruising of fruit. The complex fruit handling and equipment/ operator interface is a major hurdle engineers must address for successful technology transfer.

Preliminary Engineering Investigations

Penn State and Olin College engineering students worked with the Specialty Crop Innovations team to assess bin filling methods and design and simulate new concepts for gently transferring fruit to bins in the field. The most innovative and promising design was a "false floor" bin filler. The Specialty Crop Innovations team developed a cooperative agreement with USDA Appalachian Fruit Research Station Research Engineer Amy Tabb to conduct commercial trials on a dry bin filler with a similar "disappearing floor" design concept. Bruising studies were conducted to quantify the efficacy of the bin filler in a packinghouse setting. The bin filler also was tested for potential applications in assisted harvest operations. A research paper was published in the Journal of the American Society of Agricultural and Biological Engineers.

During the 2007 harvest, Penn State University and Pennsylvania growers hosted a "Specialty Crop Engineering Solutions" tour for robotics and precision agriculture engineers. The outcome of the tour and various planning sessions was the funding of a USDA Specialty Crop Research Initiative project led by Carnegie Mellon to investigate new solutions for assisted harvest and other labor intensive operations. During the initial year of this project, two passive bin filler prototypes showed promise in laboratory tests to assess potential reductions in damage to fruit during the bin filling process.

From 2010 through 2013, design and testing were performed on apple transport systems with a bin filler design to keep fruit singulated all the way to the bin. In 2010, the project team began working with a commercialization partner, DBR Conveyor Concepts, on a vacuum tube transport system and automated bin filler that can be retrofitted to existing grower equipment. For Penn State trials, the harvest system was initially adapted to the orchard platform automated by Carnegie Mellon.

A second generation vacuum assist system designed specifically for eastern orchard systems was tested in Pennsylvania and Michigan orchards in 2012. We performed efficiency and bruise analysis trials on the newest prototype DBR vacuum harvest system to assess efficiency gains over the use of traditional ladders and picking buckets. This assessment included examination of bruise volume from hand harvested and vacuum harvested apples. Testing the new prototype with plantings of Golden Delicious, York and Cameo, four harvest workers could simultaneously pick a 23-bushel bin of apples every 11 to 12 minutes. Compared to harvesting with ladders, this represented a 15% to 33% increase in harvest labor efficiency. Bruising was slightly elevated, and varied between 2.5% and 7.9%, depending on variety. Several refinements were made for reducing bruise incidence.

Table 1. Efficiency in apple orchard plots harvested with vacuum assist system and platform compared to hand harvest and ladders. (PA Trials, 2012)

CultivarHarvest SystemHarvest Time/bin
(min/1 person)z
Harvest time/acre
(hrs/4 workers)
% change
Golden DeliciousVacuum assist45.6 cy8.333
YorkVacuum assist45.5 c5.833
CameoVacuum assist58.0 b11.615
CameoHand68.0 a13.6--

z = 23 bushel plastic bin
y = completely randomized design with 4 replicates. Mean separation by Fisher's protected least significant difference at P<0.05

On-Going and Future Investigations

A low-cost and low-energy harvest-assist unit was developed in 2012-13 with funds from the Pennsylvania Department of Agriculture and the State Horticultural Association of Pennsylvania. The unit was refined and further field tested in Fall 2014, and a commercialization plan is currently being developed, with funds from the Penn State College of Ag Science's Research Applications and Innovations (RAIN) grant program.

Lab- and field-testing in Fall 2013 successfully demonstrated that the unit could work in a commercial-scale orchard and demonstrated possible market potential. Challenges with bruising of apples, in particular, drove refinements in the design. One of the key issues in the first year of field-testing was the incidence of bruising that occurred between the distributor and the bin. This has been the primary limiting factor with mechanized apple harvest units in the past. The focus of research in 2014 was in three primary areas: bruise reduction, ergonomics and efficiency. Redesign greatly reduced bruising in 2014 field-testing. Field tests showed that only 5% of the apples harvested were lowered in grade due to bruising after modifications in the device.


Eliminating ladders and picking buckets can greatly reduce potential hazards, and utilizing picking platforms with harvest-assist devices is a good way to do this. The ergonomics of ladder picking versus harvest with this device and platform were compared in different ways. The "Rapid Upper Limb Assessment," or RULA, approach was used to make ergonomic comparisons. This method evaluates postures and rates them on a scale of 1 (negligible risk) to 6 or above (high risk).

Conventional apple harvest activities were categorized using the RULA method, and awkward activities were identified. Awkward activities of ladder descending, moving a ladder and picking high apples while standing on a ladder were eliminated with the harvest-assist unit. The highest RULA scores recorded in ladder picking reached a value of 7, while the highest score using the harvest-assist device was 6, and the time in those potentially dangerous postures was greatly reduced. Overall, the time spent in awkward postures (RULA score of 5 or higher) was decreased from 65.0% of the whole harvest process in conventional apple harvest approach to 43.3% of the whole harvest process in harvest-assist unit approach.


Efficiency (apples picked per unit time) was based on two pickers on the platform and two pickers walking on the ground, compared with four pickers picking from the ground and using ladders for higher apples. Use of the harvest-assist unit increased the overall apple harvest efficiency by 28.6% for these two scenarios.

Prepared by: Paul Heinemann, Jude Liu, Judd Michael, Zhao Zhang, Jim Schupp, Tara Baugher