cs.ROOct 9, 2026

Towards a Unified Taxonomy of Agricultural Harvesting Technologies: Linking Physical Interaction, Automation, and Selectivity

Authors: Renato Cristiano Torres, Fabiano Mariath D Oliveira, Claudio Gottschalg Duque

Organizations: Embrapa Agricultura Digital, Research Group on Scientific Computing, Information Engineering, and Automation, Campinas, Brazil · Embrapa, Information Technology Departament, Federal District, Brazill · University of Brasilia, Computational Linguistics, Federal District, Brazil

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Harvesting technologies have evolved from conventional mechanized systems based on bulk mechanical interaction toward increasingly automated and robotic systems capable of perception, decision-making, and localized actuation. This evolution has also created a classification problem, as harvesting systems are commonly described according to crop type, mechanical removal principle, sensing technology, robotic architecture, or degree of automation, making direct comparison difficult. This paper proposes a unified taxonomy for agricultural harvesting technologies based on two primary classification axes: the physical principle governing plant-machine interaction and the level of automation incorporated into the harvesting process. Selectivity is treated as an emergent functional attribute that expresses the biological and spatial scale at which perception and decision-making influence physical actuation. Conventional mechanized methods, including cutting, pulling, vibration, impact, brushing, pneumatic action, and combined systems, are analyzed together with semi-automated and robotic harvesting approaches. The analysis shows that increasing automation does not necessarily result in greater selectivity. A qualitative transition occurs when perception and decision-making directly influence product-level actuation, shifting harvesting from distributed or bulk interaction toward localized and selective manipulation. The proposed framework provides a common conceptual basis for positioning and comparing mechanized, automated, and robotic harvesting technologies. By separating the physical mechanism of product removal from the automation level and resulting selectivity, the taxonomy clarifies technological boundaries, identifies intermediate configurations, and supports the analysis and development of future harvesting systems.

Figures & tables

ReferencePhysical interactionPerception/roboticsAutomationSelectivityUnified cross-domain taxonomy
Droukas et al. [ 7 ]PartialStrongStrongPartial–
Kootstra et al. [ 24 ]PartialStrongStrongStrong–
Pu [ 34 ]StrongPartialPartialPartial–
Rajendran et al. [ 35 ]PartialStrongStrongStrong–
Zhang [ 43 ]PartialStrongStrongStrong–
Present frameworkStrongStrongStrongExplicit attribute✓
Table 1: Primary emphasis of representative harvesting reviews and the integrative scope of the present framework. “Partial” indicates that the dimension is discussed but is not the principal organizing axis of the review.
RequirementRepresentative range or condition reported in the literatureReferences
Terrain slopeLower than approximately 12–15%, supporting machine stability and operational safety[ 1 , 8 , 22 , 29 , 30 , 32 , 42 ]
Row spacingCrop-dependent spacing allowing adequate machine traffic and harvesting operation[ 4 , 9 , 22 ]
Plant spacingApproximately 0.5–0.7 m in reported applications requiring compatibility between plant size and mechanized systems; not a general limitation for grain and seed harvesting[ 12 , 22 , 39 ]
Obstacles between rowsRemoval or management of stones, stumps, crop residues, and other obstacles that may interfere with machine displacement[ 13 , 22 , 23 , 39 , 40 ]
Planting row lengthLong rows are generally advantageous because they reduce turning maneuvers and improve operational efficiency[ 22 , 23 , 39 ]
Headland widthApproximately 6.0 m in reported cases to allow machine maneuvering[ 9 , 22 , 23 , 39 ]
Table 2: Representative production-system requirements reported for mechanized harvesting.
Figure 1: Conceptual transition from conventional bulk harvesting to selective product-level harvesting. Increasing automation alone does not guarantee selectivity; the decisive transition occurs when perception and decision-making govern localized actuation. Source: authors’ conceptual synthesis.
MethodPhysical principleTypical cropsReferences
CuttingProduct separation by mechanical shearing of the stem or entire plantGrains and seeds (wheat, rice), sugarcane, forage crops[ 22 , 37 , 39 ]
PullingProduct removal by tensile force between plant and soilRoots and tubers (potato, carrot, beetroot)[ 39 ]
VibrationMechanical excitation used to detach the product from the plantWoody crops (coffee), fruit crops (citrus, olive)[ 8 , 10 , 34 ]
ImpactTransfer of kinetic energy through repeated impactsWoody and industrial fruit crops (coffee, olive)[ 10 , 22 , 34 ]
BrushingContinuous mechanical contact using flexible elementsFruit and woody crops under specific production systems[ 8 , 10 , 15 ]
PneumaticPressure differentials and airflow used for capture or transportSeeds, lightweight fruits, auxiliary harvesting systems[ 22 , 34 , 39 ]
Table 3: Mechanized harvesting methods according to physical interaction principle and typical crops.
MethodPhysical principle and automation functionTypical cropsReferences
Selective graspingControlled mechanical force applied by robotic manipulators based on perception of the target productApple, orange, strawberry, tomato, bell pepper[ 4 , 19 , 43 ]
Selective cuttingLocalized cutting of the peduncle using automated blades or actuatorsLettuce, broccoli, high-value fruit crops[ 34 , 43 ]
Assisted suctionPressure differentials used for capture and transport, combined with visual perceptionStrawberry, seeds, seedlings, light products[ 34 , 43 ]
Semi-automated vibrationMechanical excitation whose parameters are adjusted automatically using sensors and control algorithmsCitrus, olive, woody crops[ 8 , 34 ]
Human–robot collaborationPhysical and cognitive interaction with task sharing between human operators and automated systemsHorticultural and fruit crops[ 4 , 15 , 36 ]
Automated harvesting platformsAutomated navigation and operator assistance integrated with harvesting operationsHorticultural and fruit crops[ 4 , 20 , 34 , 44 ]
Table 4: Automated and robotic harvesting approaches according to physical interaction principle.
ClassDescriptionTypical selectivityReferences
Mechanization (no automation)Harvesting execution based on mechanical energy; perception, decision-making, and control performed by the operatorLow[ 12 , 22 , 39 ]
Automated navigationAutomation limited mainly to machine guidance and displacement while the underlying harvesting method remains mechanizedLow[ 9 , 22 , 33 ]
Automated controlAutomatic adjustment of harvesting parameters such as height, speed, or vibration without individual product selectionLow to medium[ 9 , 11 , 39 ]
Semi-automated harvestingPartial product selection based on automated perception with human intervention in decision-making or executionMedium[ 4 , 36 , 34 ]
Autonomous robotic harvestingIntegration of perception, decision-making, planning, and actuation in an automated harvesting systemHigh[ 4 , 16 , 26 , 38 , 41 , 43 ]
Human–robot collaborative harvestingDynamic task sharing between human operators and automated or robotic systemsVariable[ 4 , 15 , 36 ]
Table 5: Automation classes in agricultural harvesting and their typical selectivity.
Physical interactionMechanizedAuto. nav./controlSemi-auto.AutonomousTypical selectivityRepresentative refs.
Cutting✓\checkmark✓\checkmark✓\checkmark✓\checkmarkLow–high[ 22 , 34 , 43 ]
Pulling✓\checkmark✓\checkmark––Low–medium[ 39 ]
Vibration✓\checkmark✓\checkmark✓\checkmark–Low–medium[ 8 , 10 , 34 ]
Impact✓\checkmark✓\checkmark––Low[ 10 , 22 , 34 ]
Brushing✓\checkmark✓\checkmark––Low–medium[ 8 , 10 , 15 ]
Pneumatic/suction✓\checkmark✓\checkmark✓\checkmark✓\checkmarkLow–high[ 22 , 34 , 43 ]
Table 6: Conceptual matrix combining physical interaction and automation level. Check marks indicate combinations represented or directly supported by the harvesting approaches discussed in the reviewed literature; a dash indicates that the combination is not established by the present review.
More in this topic

Robotic Harvesting

See all 22 papers

Explore similar work

CardsList
Sep 16, 2026

CLASP: A Cluster-Level Autonomous Selective Picking Robot with a Soft Rolling-Band Gripper for Fresh-Market Blueberry Harvesting

Fresh-market blueberries require selective, gentle picking, which is labor-intensive and expensive. Over-the-row machine harvesters are fast but non-selective, bruising mixed-ripeness fruit and limiting yield to the processing market. Selective robotic harvesters typically target individual fruits rather than fruit clusters, which limits harvesting efficiency for small, densely clustered blueberries. This paper presents CLASP, a Cluster-Level Autonomous Selective Picking robot with a Soft Active Rolling-Band Gripper (SARB-Gripper). Two compliant bands envelop the cluster and roll against the fruit, drawing mature berries off in sequence, while closed-loop regulation of the pulling force keeps the applied load below the immature detachment threshold. A global-to-local perception pipeline pairs an eye-to-hand camera for global cluster detection and target selection with an eye-in-hand camera for local localization and cluster orientation estimation. Field measurements confirm a clear detachment-force separation between mature and immature fruit, and the SARB-Gripper reproduces a commanded pulling force to within \SI{3.7}{\percent}, enabling selective harvesting at the cluster level. In end-to-end field trials, CLASP autonomously grasped 23 of 25 presented clusters (\SI{92}{\percent}). With the component cost of approximately $3326 per unit, CLASP offers a scalable approach to selective cluster-level harvesting for fresh-market blueberries.

Yixuan Xia, Yilin Cai, Natalia Belen Espinoza +4

George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332 USA · Department of Crop & Soil Sciences, University of Georgia, Athens, GA 30605, USA · Bio-Sensing, Automation, and Intelligence Laboratory, Department of Agricultural and Biological Engineering, University of Florida, Gainesville, Florida 32611, USA +3

Robotic GraspingForce-Controlled Robotic ManipulationAgricultural RoboticsSoft RoboticsRobotic Harvesting
AskPDF
Mar 24, 2026

Learning What Can Be Picked: Active Reachability Estimation for Efficient Robotic Fruit Harvesting

Agriculture remains a cornerstone of global health and economic sustainability, yet labor-intensive tasks such as harvesting high-value crops continue to face growing workforce shortages. Robotic harvesting systems offer a promising solution; however, their deployment in unstructured orchard environments is constrained by inefficient perception-to-action pipelines. In particular, existing approaches often rely on exhaustive inverse kinematics or motion planning to determine whether a target fruit is reachable, leading to unnecessary computation and delayed decision-making. Our approach combines RGB-D perception with active learning to directly learn reachability as a binary decision problem. We then leverage active learning to selectively query the most informative samples for reachability labeling, significantly reducing annotation effort while maintaining high predictive accuracy. Extensive experiments demonstrate that the proposed framework achieves accurate reachability prediction with substantially fewer labeled samples, yielding approximately 6--8% higher accuracy than random sampling and enabling label-efficient adaptation to new orchard configurations. Among the evaluated strategies, entropy- and margin-based sampling outperform Query-by-Committee and standard uncertainty sampling in low-label regimes, while all strategies converge to comparable performance as the labeled set grows. These results highlight the effectiveness of active learning for task-level perception in agricultural robotics and position our approach as a scalable alternative to computation-heavy kinematic reachability analysis. Our code is available through https://github.com/wsu-cyber-security-lab-ai/active-learning.

Nur Afsa Syeda, Mohamed Elmahallawy, Luis Fernando de la Torre +1

Washington State University, Richland, WA 99354, USA

Agricultural RoboticsRobotic HarvestingRobotic PerceptionActive Learning
AskPDF
Feb 22, 2026

FruitTouch: A Perceptive Gripper for Gentle and Scalable Fruit Harvesting

The automation of fruit harvesting has gained increasing significance in response to rising labor shortages. A sensorized gripper is a key component of this process, which must be compact enough for confined spaces, able to stably grasp diverse fruits, and provide reliable feedback on fruit conditions for efficient harvesting. To address this need, we propose FruitTouch, a compact gripper that integrates high-resolution, vision-based tactile sensing through an optimized optical design. This configuration accommodates a wide range of fruit sizes while maintaining low cost and mechanical simplicity. Tactile images captured by an embedded camera provide rich information for real-time force estimation, slip detection, and softness prediction. We validate the gripper in real-world fruit harvesting experiments, demonstrating robust grasp stability and effective damage prevention.

Ruohan Zhang, Mohammad Amin Mirzaee, Wenzhen Yuan

University of Illinois at Urbana-Champaign, Champaign, IL, USA

Robotic GraspingRobotic Hand DesignAgricultural RoboticsVision-Based Tactile Sensing
AskPDF