ReviewTechnology progress in mechanical harvest of fresh market apples
Graphical abstract
Introduction
Apple is one of the most consumed fruit in both fresh eating and other forms (e.g., juice and sauce), and it provides a number of health benefits, such as reducing the incidence rates of cancers and cardiovascular diseases (Boyer and Liu, 2004, STATISTA, 2016, PBH., 2015). Fresh market apples are still manually harvested using the conventional ladder-and-bucket approach (Zhang, 2015, Freivalds et al., 2006). In manual harvesting of apples, seasonal workers stand on the ground for picking the low-level fruits and need to climb a ladder to get access to high-level fruits. During the whole harvest process, workers need to wear a bucket to temporarily hold apples, and after the bucket is full, they walk to a bin to unload the fruits. This harvest method is low in efficiency, since a lot of activities performed manually by workers (e.g., moving ladders, climbing up and down, and unloading apples from buckets to bins) are not directly related to apple picking. This conventional harvest approach demands high physical strength for the carrying out of the bucket (≈19 kg when full) continuously, not to mention workers have to climb/descend ladders with a heavy bucket (Freivalds et al., 2006). In addition, the conventional harvest approach is prone to causing occupational injuries, such as ladder falls, back strains, and musculoskeletal disorders (Zhang et al., 2019a, Zhang et al., 2019b, Fathallah, 2010). Mechanical apple harvest is an approach to alleviate the disadvantages of the conventional ladder-and-bucket apple harvest method.
Researchers during the past decades worked towards the realization of mechanical apple harvest using different methods that can be categorized into shake-and-catch method, robots, and harvest-assist platform (Zhang et al., 2016a). The shake-and-catch approach is to hold the trunk or limbs and then shake to detach apples (Allshouse and Morrow, 1972). Several harvest machines based on this concept have been developed, but they are not adopted commercially for fresh market apple harvest due to extensive impacts related to bruising damages on fruit (Berlage and Langmo, 1974, Tennes et al., 1976). Harvesting robots mainly include two components – a machine vision system and an end-effector. The machine vision system detects and localizes apples, and the end-effector detaches fruit from the tree. Though researchers have developed a variety of apple harvesting robot prototypes, none of them is commercialized, due to low throughput and efficiency, unreliable performance, and high cost (Slaughter and Harrell, 1987, Slaughter and Harrell, 1989, Zhao et al., 2016). Harvesting robots, however, can have the potential to entirely replace the laborers. To immediately benefit the apple industry, researchers shifted to the development of harvest-assist platforms to replace the use of ladders (Zhang et al., 2016a, 2017a). Instead of picking by standing on a ladder, picking fruit by standing on the harvest-assist platform would increase harvest efficiency, reduce physical strength requirements, improve overall safety, and alleviate occupational injuries (Schupp et al., 2011).
Numerous research projects have been conducted with results published in the past decades on the progress of mechanical fresh market apple harvest aforementioned concepts (Zhang and Karkee, 2016, He et al., 2017, Wheat, 2019, Zhang et al., 2019d). There were some reviews published in this domain, but they only focused on specific topics: Gongal et al. (2015) reviewed the progress on the machine vision for fruit detection; Bac et al., 2014, Zhao et al., 2016 summarized the progress in robots harvest and vision-based control for robots; and Zhang et al. (2016a) focused on the shake-and-catch method. However, no reviews were published about the mechanical harvest of fresh market apples in a holistic manner in the past decade. Thus, a proper synchronization of the literature in the mechanical fresh market apple harvest has been lacking, and this article aims to fill this need of reviewing the technology progress on the methods of shake-and-catch, robots, and harvest-assist platform.
Section snippets
Shake-and-catch method
Shake-and-catch serves as a promising fresh market apple harvest technology (Markwardt et al., 1968, Fu et al., 2016). A major advantage of shake-and-catch method is its high harvest efficiency, and it was reported that the developed machines were capable of harvesting 20 to 60 trees/h (Allshouse and Morrow, 1972, Chen et al., 1982). However, the low apple detachment rates and unacceptable bruising of harvested apples hindered the machines’ commercial application (LaBelle et al., 1965, Peterson
Apple harvesting robots
Starting from the 1970s, researchers started to develop robots for fresh market fruit harvest (Sarig, 1993, Ceccarelli et al., 2000, Gongal et al., 2015, 2016). Robots for apple harvest progressed slowly, unlike their counterparts in industrial application, due to uncontrolled field environment (e.g., lighting) and complex field working conditions (e.g., water, dust, and vibration) (Sivaraman and Burks, 2006). Robots harvesting is a systematic design, with a lot of technologies involved, such
Apple harvest-assist platform progress
Since both the shake-and-catch method and robots are not ready for commercial use of fresh market apple harvest, and the pressure increases from labor cost increase and labor shortage on apple industry, agricultural engineers and industrial startups invest efforts on developing harvest-assist platforms to assist pickers for fresh market apple harvest since the 1990s (Torregrosa et al., 2009, Schupp et al., 2011). The shortcomings of conventional apple harvest method are mainly due to the use of
Progress summary of fresh market apple mechanical harvest
Table 1 provides a snapshot of the various developments and progresses made in the field of mechanical harvesting of fresh market apples using shake-and-catch, robots, and harvest-assist platforms methods, the overall merits and limitations of these methods, along with identified technology sub-categories and relevant references. This information can be used to compare the methods as well as a roadmap to comprehend the developments so far at a glance and gain detailed information through the
Conclusion and future trends
In this article, the progress in the mechanical harvest of fresh market apples has been summarized focusing on the three predominant technologies, such as shake-and-catch, robots, and harvest-assist platform methods. For the shake-and-catch technology, the recently obtained theoretical knowledge in apple detachment (location index, limb load, and hooker position) is desirable to be applied to design proper tree canopies for easy detachment of apples. Though the tree canopy structure has evolved
Disclaimer
Mention of commercial products or orchards in this paper is only for providing factual information and does not imply endorsement of them by authors over those not mentioned.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (135)
- et al.
Machine vision system for automatic quality grading of fruit
Biosyst. Eng.
(2003) - et al.
Image fusion of visible and thermal images for fruit detection
Biosyst. Eng.
(2009) - et al.
The West Virginia University tree fruit harvester
J. Agric. Eng. Res.
(1982) - et al.
Mechanical harvesting of apricots
Biosyst. Eng.
(2003) Musculoskeletal disorders in labor-intensive agriculture
Appl. Ergonomics
(2010)- et al.
Effect of belt/bucket interface in apple harvesting
Int. J. Ind. Ergon.
(2006) - et al.
A novel image processing algorithm to separate linearly clustered kiwifruits
Biosys. Eng.
(2019) - et al.
Multi-modal deep learning for Fuji apple detection using RGB-D cameras and their radiometric capabilities
Comput. Electron. Agric.
(2019) - et al.
Sensors and systems for fruit detection and localization: A review
Comput. Electron. Agric.
(2015) - et al.
Apple crop-load estimation with over-the-row machine vision system
Comput. Electron. Agric.
(2016)
Effect of fruit location on apple detachment with mechanical shaking
Biosyst. Eng.
Deep learning–Method overview and review of use for fruit detection and yield estimation
Comput. Electron. Agric.
Green citrus detection using ‘eigenfruit’, color and circular Gabor texture features under natural outdoor conditions
Comput. Electron. Agric.
The design of a prototype apple harvester
J. Agric. Eng. Res.
Characterizing apple picking patterns for robotic harvesting
Comput. Electron. Agric.
In-field citrus detection and localisation based on RGB-D image analysis
Biosyst. Eng.
Determination of the number of green apples in RGB images recorded in orchards
Comput. Electron. Agric.
Air suspension-based catching mechanism for mechanical harvesting of apples
IFAC-PapersOnLine
An image segmentation method for apple sorting and grading using support vector machine and Otsu’s method
Comput. Electron. Agric.
Computer vision-based apple grading for golden delicious apples based on surface features
Inf. Process. Agric.
Shock absorbing surfaces for collecting fruit during the mechanical harvesting of citrus
Biosyst. Eng.
Advances in mechanical harvesting of fresh market quality apples
J. Agric. Eng. Res.
Detecting fruits in natural scenes by using spatial-frequency based texture analysis and multiview geometry
Comput. Electron. Agric.
Robotics of fruit harvesting: A state-of-the-art review
J. Agric. Eng. Res.
Location of apples in trees using stereoscopic vision
Comput. Electron. Agric.
Design of an automatic apple sorting system using machine vision
Comput. Electron. Agric.
Estimation of number and diameter of apple fruits in an orchard during the growing season by thermal imaging
Comput. Electron. Agric.
Over-the-row harvester for dwarf fruit trees
Trans. ASAE
Harvesting robots for high-value crops: State-of-the-art review and challenges ahead
J. Field Rob.
Harvesting apples with straddle-frame trunk shaker
Trans. ASAE
Apple phytochemicals and their health benefits
Nutr. J.
A multispectral imaging analysis for enhancing citrus fruit detection
Environ. Control. in Biol.
Designing a robotic gripper for harvesting horticulture products
Robotica
Harvesting Valencia oranges with flexible curved fingers
Trans. ASAE
Advances in machine vision applications for automatic inspection and quality evaluation of fruits and vegetables
Food Bioprocess Technol.
Hand-picking dynamic analysis for undersensed robotic apple harvesting
Trans. ASABE
A semi-automated harvesting prototype for shaking fruit tree limbs
Trans. ASABE
Vibration characteristics of trellis-trained apple trees with reference to fruit detachment
Trans. ASAE
The highly adaptive SDM hand: Design and performance evaluation
Int. J. Robotics Res.
Semantic mapping for orchard environments by merging two-sides reconstructions of tree rows
J. Field Rob.
Cited by (75)
Seeding detection and distribution evaluation using the developed automatic maize seeding machine
2024, Computers and Electronics in AgricultureInfield corn kernel detection using image processing, machine learning, and deep learning methodologies under natural lighting
2024, Expert Systems with ApplicationsComprehensive wheat lodging detection after initial lodging using UAV RGB images
2024, Expert Systems with ApplicationsEffect of cold plasma pretreatment on drying kinetics and quality attributes of apple slices in Refractance window drying
2024, Innovative Food Science and Emerging TechnologiesBruising damage in apple-to-apple collision via a sliding method
2023, Biosystems EngineeringApple rapid recognition and processing method based on an improved version of YOLOv5
2023, Ecological Informatics