Center for Precision and Automated Agricultural Systems

Sun City Center, United States

Center for Precision and Automated Agricultural Systems

Sun City Center, United States
SEARCH FILTERS
Time filter
Source Type

Amatya S.,Center for Precision and Automated Agricultural Systems | Karkee M.,Center for Precision and Automated Agricultural Systems | Gongal A.,Center for Precision and Automated Agricultural Systems | Zhang Q.,Center for Precision and Automated Agricultural Systems | And 2 more authors.
Biosystems Engineering | Year: 2016

Fresh market sweet cherry harvesting is a labour-intensive operation that accounts for more than 50% of annual production costs. To minimise labour requirements for sweet cherry harvesting, mechanized harvesting technologies are being developed. These technologies utilise manually-placed limb actuators that apply vibrational energy to affect fruit release. Machine vision-based automated harvesting system have potential to further reduce harvest labour through improving efficiency by eliminating manual handling, positioning and operation of the harvester and/or harvesting mechanism. A machine-vision system was developed to segment and detect cherry tree branches with full foliage, when only intermittent segments of branches were visible. Firstly, an image segmentation method was developed to identify visible segments of the branches. Bayesian classifier was used to classify image pixels into four classes - branch, cherry, leaf and background. The algorithm achieved 89.6% accuracy in identifying branch pixels. The length and orientation of branch segments were then analysed to link individual sections of the same branch together and to represent the branches with an equation. Linear and logarithmic model equations were fitted to the branch segments and the equation with minimum residual was selected as the best-fit model representing the corresponding branch. Branches detected with this algorithm were compared with manual counting. The method achieved a branch detection accuracy of 89.2% in a set of 141 test images acquired during full-foliage canopy. This study shows the potential of using a machine vision system for automating shake-and-catch cherry harvesting systems. © 2015 IAgrE.


Gongal A.,Center for Precision and Automated Agricultural Systems | Silwal A.,Center for Precision and Automated Agricultural Systems | Amatya S.,Center for Precision and Automated Agricultural Systems | Karkee M.,Center for Precision and Automated Agricultural Systems | And 3 more authors.
Computers and Electronics in Agriculture | Year: 2016

Accurate crop-load estimation is important for efficient management of pre- and post-harvest operations. This information is crucial for the planning of labor and equipment requirement for harvesting and transporting fruit from the orchard to packing house. Current machine vision-based techniques for crop-load estimation have achieved only limited success mostly due to: (i) occlusion of apples by branches, leaves and/or other apples, and (ii) variable outdoor lighting conditions. In order to minimize the effect of these factors, a new sensor system was developed with an over-the-row platform integrated with a tunnel structure which acquired images from opposite sides of apple trees. The tunnel structure minimized illumination of apples with direct sunlight and reduced the variability in lighting condition. Images captured in a tall spindle orchard were processed for identifying apples, which achieved an identification accuracy of 79.8%. The location of apples in three-dimensional (3D) space was used to eliminate duplicate counting of apples that were visible to cameras from both sides of the tree canopy. The error on identifying duplicate apples was found to be 21.1%. Overall, the method achieved an accuracy of 82% on estimating crop-load on trees with dual side imaging compared to 58% with single side imaging. Over-the-row machine vision system showed promise for accurate and reliable apple crop-load estimation in the apple orchards. © 2015.


Virk S.S.,Auburn University | Mullenix D.K.,Jones Edmunds and Associates Inc. | Sharda A.,Center for Precision and Automated Agricultural Systems | Hall J.B.,Auburn University | And 4 more authors.
Applied Engineering in Agriculture | Year: 2013

Blended fertilizers are commonly applied to crop and pasture land using spinner-disc spreaders. However, the nature of blended fertilizers can make it difficult to spread uniformly due to varying physical properties of the N, P, and K raw constituents, which can lead to segregation during application. Further, variable-rate application of blended fertilizer could pose challenges in terms of accuracy and uniformity to meet target prescription rates. Therefore, the objective of this study was to evaluate the potential of fertilizer segregation during application with a spinner-disc spreader. A series of standard pan and field tests were performed to evaluate mass and nutrient distribution for a blended fertilizer (17-17-17) using a spinner-disc spreader. Field tests were performed by generating prescription maps, randomly placing pans across the field, then applying the blended fertilizer and analyzing the collected material for N, P2O5 and K2O mass fractions. Both mass and nutrient distribution patterns were generated and then compared. Field data in combination with prescription and as-applied maps were used to evaluate spreader performance and product segregation. A rating of 17-18- 17 was determined for the blended fertilizer. P2O5 and K2O mass fraction varied significantly (coefficient of variation, CV, 25.0% and 26.0%, respectively) based on the single-pass pattern measurements. Field application also produced high variability of P 2O5 and K2O mass fractions with CV's ranging between 16.8% and 37.2%, respectively. These results indicated the occurrence of fertilizer segregation. Nitrogen mass fraction, spread variability was low (CV = 4%) based on the pan tests but was contributed to N beingprovided by both ammonium nitrate and diammonium phosphate (DAP) constituents of the blend. Particle size analysis supported the notion that segregation occurred mainly due to size variability between the constituents. During field tests, actual rates were consistently higher than the prescribed rates. While the prescription and as-applied maps showed high correlation (R = 0.75 to 0.94), the as-applied maps did not reflect the actual field application (R = 0.25 to 0.67). Finally, the existence of fertilizer segregation indicated further research is required to understand segregation of blended fertilizers when applied using spinner spreaders. © 2013 American Society of Agricultural and Biological Engineers.


De Kleine M.E.,Center for Precision and Automated Agricultural Systems | Karkee M.,Center for Precision and Automated Agricultural Systems | Lewis K.,Center for Precision and Automated Agricultural Systems | Lewis K.,Washington State University | Zhang Q.,Center for Precision and Automated Agricultural Systems
American Society of Agricultural and Biological Engineers Annual International Meeting 2014, ASABE 2014 | Year: 2014

An end-effector concept for fresh-market fruit harvesting was evaluated in term of removing apples from branches. The rotational end-effector was developed as a simple device to spin, or rotate, an apple from a branch. This mechanism used two bi-directional electric motors to rotate rubber wheels. Tests were conducted to determine the optimal performance for the end-effector by varying individual wheel speed, direction of rotation, and variety. The performance of the fruit removal process was characterized by the fruit removal condition: stem-intact (preferred for the fresh-market), stempull (stem was removed), spurpull (fruit spur was removed), and removal rate (number of fruit per unit time). Variety and direction of rotation were significant factors affecting the fruit removal condition whereas wheel speed was not. The direction of rotation that resulted in the highest fresh-market percentage was equal, i.e. both wheels rotating in the clockwise direction. The resulting stem-intact removal conditions using the rotational end-effector were: 69%, 71%, 72% and 88%, for 'Gala', 'Fuji'. 'Golden Delicious', and 'WA-17' variety apples, respectively. Removal rates ranged 29-43 apples minute'. This study indicates that a rotational end-effector has potential to remove apples from branches while maintaining a reasonable level of fruit with the stem intact. Future research on expanding the rotational prototype for higher removal rate and developing appropriate catch mechanisms for both techniques could lead to the development and comparison of complete harvesting systems.


Silwal A.,Center for Precision and Automated Agricultural Systems | Karkee M.,Center for Precision and Automated Agricultural Systems | Mo C.,United Materials and Systems | Zhang Q.,Washington State University
Transactions of the ASABE | Year: 2016

This article evaluates hand-picking methods as candidate grasping techniques for implementation in a robotic system designed to harvest apples. The standard method of hand-picking apples is highly selective to the apple's orientation and stem location. However, sensory detection of the fruit's orientation and stem while the apple is on the tree is a challenging problem requiring significant computation time. In this study, four picking techniques that do not require knowledge of fruit orientation were applied to five apple varieties growing in several different cultivation systems. The sensors used during hand-picking included force sensors and an inertial measurement unit. Experimental results were obtained for normal contact forces during a three-fingered power grasp as well as the angle of rotation around the axis of the forearm. Field data and controlled laboratory experiments show that fruit separation can be clearly detected. Accelerometer measurements were also used to calculate the average distance to fruit separation, which varied from 3 to 7 cm. The optimum picking method relative to stem attachment was identified for each apple variety. © 2016 American Society of Agricultural and Biological Engineers.


Karkee M.,Center for Precision and Automated Agricultural Systems | McNaull R.P.,Iowa State University | Birrell S.J.,Iowa State University | Steward B.L.,Iowa State University
Transactions of the ASABE | Year: 2012

As the demand for biomass feedstocks grows, it is likely that agricultural residue will be removed in a way that compromises soil sustainability due to increased soil erosion, depletion of organic matter, and deterioration of soil physical characteristics. Since soil erosion from agricultural fields depends on several factors including soil type, field terrain, and cropping practices, the amount of biomass that can be removed while maintaining soil tilth varies substantially over space and time. The RUSLE2 soil erosion model, which takes into account these spatio-temporal variations, was used to estimate tolerable agricultural biomass removal rates at field scales for a single-pass crop grain and biomass harvesting system. Soil type, field topography, climate data, management practices, and conservation practices were stored in individual databases on a state or county basis. Geographic position of the field was used as a spatial key to access the databases to select site-specific information such as soil, topography, and management related parameters. These parameters along with actual grain yield were provided as inputs to the RUSLE2 model to calculate yearly soil loss per unit area of the field. An iterative technique was then used to determine site-specific tolerable biomass removal rates that keep the soil loss below the soil loss thresholds (T) of the field. The tolerable removal rates varied substantially with field terrain, crop management practices, and soil type. At a location in a field in Winnebago county, Iowa, with ~1% slope and conventional tillage practices, up to 98% of the 11 Mg ha -1 total above-ground biomass was available for collection with negligible soil loss. There was no biomass available to remove with conventional tillage practices on steep slopes, as in a field in Crawford county, Iowa, with a 12.6% slope. If no-till crop practices were adopted, up to 70% of the total above-ground biomass could be collected at the same location with 12.6% slope. In the case of a soybean-corn rotation with no-till practices, about 98% of total biomass was available for removal at the locations in the Winnebago field with low slopes, whereas 77% of total biomass was available at a location in the Crawford field with a 7.5% slope. Tolerable removal rates varied substantially over an agricultural field, which showed the importance of site-specific removal rate estimation. These removal rates can be useful in developing recommended rates for producers to use during a single-pass crop grain and biomass harvesting operation. However, this study only considered the soil erosion tolerance level in estimating biomass removal rates. Before providing the final recommendation to end users, further investigations will be necessary to study the potential effects of continuous biomass removal on organic matter content and other biophysical properties of the soil. © 2012 American Society of Agricultural and Biological Engineers.

Loading Center for Precision and Automated Agricultural Systems collaborators
Loading Center for Precision and Automated Agricultural Systems collaborators