An Autonomous Fruit and Vegetable Harvester with a Low-Cost Gripper Using a 3D Sensor.
ABSTRACT: Reliable and robust systems to detect and harvest fruits and vegetables in unstructured environments are crucial for harvesting robots. In this paper, we propose an autonomous system that harvests most types of crops with peduncles. A geometric approach is first applied to obtain the cutting points of the peduncle based on the fruit bounding box, for which we have adapted the model of the state-of-the-art object detector named Mask Region-based Convolutional Neural Network (Mask R-CNN). We designed a novel gripper that simultaneously clamps and cuts the peduncles of crops without contacting the flesh. We have conducted experiments with a robotic manipulator to evaluate the effectiveness of the proposed harvesting system in being able to efficiently harvest most crops in real laboratory environments.
Project description:This paper proposes a modular gripping mechanism for the manipulation of multiple objects. The proposed micro gripper combines traditional machining techniques with MEMS technologies to produce a modular mechanism consisting of a sturdy, compliant aluminium base and replaceable end-effectors. This creates an easily-customisable solution for micro manipulation with an array of different micro tips for different applications. We have provided the kinematic analysis for the gripper to predict the output and have also optimised design parameters based on FEA (finite element analysis) simulation and the effects of altering flexure beam lengths. The gripper is operated by a piezo actuator capable of 18 μ m displacement at 150 V of applied DC voltage. This is then amplified by a factor of 8.1 to a maximum tip displacement of 154 μ m. This is achieved by incorporating bridge and lever amplifying techniques into the design. An initial experimental analysis of the micro gripper is provided to investigate the performance of the micro gripper and to gauge the accuracy of the theory and simulation through comparison with experimental results.
Project description:In this study, we propose a soft pneumatic gripper that uses a tendon-driven soft origami pump. The gripper consists of three pneumatic soft actuators that are controlled by a tendon-driven origami pump. An external air compressor that supplies air to the pneumatic actuator is replaced by an origami pump. The soft actuator is composed of silicone (Ecoflex 00-30) with a chamber-based structure, which is fabricated using a mold, and the origami pump is fabricated by folding a Kresling patterned polypropylene film. In addition, we conduct a series of experiments to evaluate the performance of the pneumatic actuator with a tendon-driven origami pump. Specifically, movement characteristics, frequency response, blocking force, and the relation between bending angle and pressure are analyzed from the results of the experiments. Furthermore, we understand the entire operation mechanism from the deformation of the origami pump to bending through pressure. Finally, we demonstrate the grasping of objects with diverse shapes and materials, and indicate the feasibility of the pneumatic gripper as an independent module without an external compressor.
Project description:In this study, the aim is to develop a population model based approach to optimize fruit harvesting strategies with regard to fruit quality and its derived economic value. This approach was applied to the case of tomato fruit harvesting under Vietnamese conditions. Fruit growth and development of tomato (cv. "Savior") was monitored in terms of fruit size and color during both the Vietnamese winter and summer growing seasons. A kinetic tomato fruit growth model was applied to quantify biological fruit-to-fruit variation in terms of their physiological maturation. This model was successfully calibrated. Finally, the model was extended to translate the fruit-to-fruit variation at harvest into the economic value of the harvested crop. It can be concluded that a model based approach to the optimization of harvest date and harvest frequency with regard to economic value of the crop as such is feasible. This approach allows growers to optimize their harvesting strategy by harvesting the crop at more uniform maturity stages meeting the stringent retail demands for homogeneous high quality product. The total farm profit would still depend on the impact a change in harvesting strategy might have on related expenditures. This model based harvest optimisation approach can be easily transferred to other fruit and vegetable crops improving homogeneity of the postharvest product streams.
Project description:Economic incentives to harvest a species usually diminish as its abundance declines, because harvest costs increase. This prevents harvesting to extinction. A known exception can occur if consumer demand causes a declining species' harvest price to rise faster than costs. This threat may affect rare and valuable species, such as large land mammals, sturgeons, and bluefin tunas. We analyze a similar but underappreciated threat, which arises when the geographic area (range) occupied by a species contracts as its abundance declines. Range contractions maintain the local densities of declining populations, which facilitates harvesting to extinction by preventing abundance declines from causing harvest costs to rise. Factors causing such range contractions include schooling, herding, or flocking behaviors-which, ironically, can be predator-avoidance adaptations; patchy environments; habitat loss; and climate change. We use a simple model to identify combinations of range contractions and price increases capable of causing extinction from profitable overharvesting, and we compare these to an empirical review. We find that some aquatic species that school or forage in patchy environments experience sufficiently severe range contractions as they decline to allow profitable harvesting to extinction even with little or no price increase; and some high-value declining aquatic species experience severe price increases. For terrestrial species, the data needed to evaluate our theory are scarce, but available evidence suggests that extinction-enabling range contractions may be common among declining mammals and birds. Thus, factors causing range contraction as abundance declines may pose unexpectedly large extinction risks to harvested species.
Project description:There have been recent developments in grippers that are based on capillary force and condensed water droplets. These are used for manipulating micro-sized objects. Recently, one-finger grippers have been produced that are able to reliably grip using the capillary force. To release objects, either the van der Waals, gravitational or inertial-forces method is used. This article presents methods for reliably gripping and releasing micro-objects using the capillary force. The moisture from the surrounding air is condensed into a thin layer of water on the contact surfaces of the objects. From the thin layer of water, a water meniscus between the micro-sized object, the gripper and the releasing surface is created. Consequently, the water meniscus between the object and the releasing surface produces a high enough capillary force to release the micro-sized object from the tip of the one-finger gripper. In this case, either polystyrene, glass beads with diameters between 5–60 µm, or irregularly shaped dust particles of similar sizes were used. 3D structures made up of micro-sized objects could be constructed using this method. This method is reliable for releasing during assembly and also for gripping, when the objects are removed from the top of the 3D structure—the so-called “disassembling gripping” process. The accuracy of the release was lower than 0.5 µm.
Project description:This article presents a methodology to recycle and upgrade a 4-DOF educational robot manipulator with a gripper. The robot is upgraded by providing it an artificial vision that allows obtaining the position and shape of objects collected by it. A low-cost and open-source hardware solution is also proposed to achieve motion control of the robot through a decentralized control scheme. The robot joints are actuated through five direct current motors coupled to optical encoders. Each encoder signal is fed to a proportional integral derivative controller with anti-windup that employs the motor velocity provided by a state observer. The motion controller works with only two open-architecture Arduino Mega boards, which carry out data acquisition of the optical encoder signals. MATLAB-Simulink is used to implement the controller as well as a friendly graphical interface, which allows the user to interact with the manipulator. The communication between the Arduino boards and MATLAB-Simulink is performed in real-time utilizing the Arduino IO Toolbox. Through the proposed controller, the robot follows a trajectory to collect a desired object, avoiding its collision with other objects. This fact is verified through a set of experiments presented in the paper.
Project description:Secure grasping of fragile fruits and other agricultural products without potential slip and damage is still a challenge due to the size and shape varying, bruise susceptible, as well as hardness changing during fruit and vegetable maturation. In the robotic grasping process, the mechanical damage mainly depends upon the aggressiveness of the gripper and the sensitivity of the product to the damage. In this study, a flexible gripper integrated with multi-sensor network is designed and tested. The network proposed includes three kinds of sensors that enable the gripper to grasp various products with the sense of touch and visual perception. Particular attention has been attached to the sensors applied between the fingers, and this makes sensing and grasping capabilities improved. To create an accurate grasping system, a grasping algorithm and the force control model are proposed for any bending state based on Cosserat theory. The boundary detection is included in the grasping algorithm, detecting the shape edge by some certain point calculation. The created grasping system guarantees mechanical compliance by evaluating and adjusting the finger status including force, angle, and direction. Multi-group tests have been done on grasping several objects of different sizes and materials in daily life. The relationship between force, bending, and surface material is also analyzed and compared under different conditions. The numerical comparisons related to the measurement error are analyzed based on their standard deviations. Experimental results indicate that this flexible manipulator with proposed system and strategy has better grasping ability for fragile fruits with its good flexibility and dexterity.
Project description:Grippers are widely used for the gripping, manipulation, and assembly of objects with a wide range of scales, shapes, and quantities in research, industry, and our daily lives. A simple yet universal solution is very challenging. Here, we manage to address this challenge utilizing a simple shape memory polymer (SMP) block. The embedding of objects into the SMP enables the gripping while the shape recovery upon stimulation facilitates the releasing. Systematic studies show that friction, suction, and interlocking effects dominate the grip force individually or collectively. This universal SMP gripper design provides a versatile solution to grip and manipulate multiscaled (from centimeter scale down to 10-?m scale) 3D objects with arbitrary shapes, in individual, deterministic, or massive, selective ways. These extraordinary capabilities are demonstrated by the gripping and manipulation of macroscaled objects, mesoscaled steel sphere arrays and microparticles, and the selective and patterned transfer printing of micro light-emitting diodes.
Project description:In the pineapple sector of Benin, poor fruit quality prevents pineapple producers to enter the European market. We investigated effects of common cultural practices, flowering and maturity synchronisation, (1) to quantify the trade-offs of flowering and maturity synchronisation for pineapple quality and the proportion of fruits exportable to European markets, and (2) to determine the effect of harvesting practice on quality attributes. Four on-farm experiments were conducted during three years using cultivars Sugarloaf and Smooth Cayenne. A split-split plot design was used in each experiment, with flowering induction practice as main factor (artificial or natural flowering induction), maturity induction practice as split factor (artificial or natural maturity induction) and harvesting practice as the split-split factor (farmers' harvest practice or individual fruit harvesting at optimum maturity). Artificial flowering induction gave fruits with lower infructescence weight, higher ratio crown: infructescence length, and a lower proportion of fruits exportable to European markets than natural flowering induction. The costs of the improvements by natural flowering induction were huge: the longer durations from planting to flowering induction and harvesting, the higher number of harvestings of the fruits increasing the labour cost and the lower proportion of plants producing fruits compared with crops from artificially flowering-induced plants. Artificial maturity induction decreased the total soluble solids concentration in the fruits compared with natural maturity induction thus decreasing the proportion of fruits exportable to European markets, at a benefit of only a slightly shorter time from flowering induction to harvesting. Harvesting individual fruits at optimum maturity gave fruits with higher total soluble solids in naturally maturity induced fruits compared with the farmers' harvest practice. Given the huge costs of natural flowering induction, options to use artificial flowering induction effectively for obtaining high fruit quality are discussed.
Project description:A delayed harvest of maize and soybean crops is associated with yield or revenue losses, whereas a premature harvest requires additional costs for artificial grain drying. Accurately predicting the ideal harvest date can increase profitability of US Midwest farms, but today's predictive capacity is low. To fill this gap, we collected and analyzed time-series grain moisture datasets from field experiments in Iowa, Minnesota and North Dakota, US with various maize (n = 102) and soybean (n = 36) genotype-by-environment treatments. Our goal was to examine factors driving the post-maturity grain drying process, and develop scalable algorithms for decision-making. The algorithms evaluated are driven by changes in the grain equilibrium moisture content (function of air relative humidity and temperature) and require three input parameters: moisture content at physiological maturity, a drying coefficient and a power constant. Across independent genotypes and environments, the calibrated algorithms accurately predicted grain dry-down of maize (r2 = 0.79; root mean square error, RMSE = 1.8% grain moisture) and soybean field crops (r2 = 0.72; RMSE = 6.7% grain moisture). Evaluation of variance components and treatment effects revealed that genotypes, weather-years, and planting dates had little influence on the post-maturity drying coefficient, but significantly influenced grain moisture content at physiological maturity. Therefore, accurate implementation of the algorithms across environments would require estimating the initial grain moisture content, via modeling approaches or in-field measurements. Our work contributes new insights to understand the post-maturity grain dry-down and provides a robust and scalable predictive algorithm to forecast grain dry-down and ideal harvest dates across environments in the US Corn Belt.