Progress 01/01/16 to 12/31/19
Outputs
Target Audience:One Ph.D. student was mentored in the context of this project. Informal laboratory instruction was the key instrument. Our research team worked closely with growers in Modesto, Stockton, and Lodi CA and conducted field experiments in their fields. The PI and Ph.D. student gave several presentations related to tree modeling and mechanized harvesting. The audiences included growers, entrepreneurs, researchers, students, academic staff and the general public. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?One Ph.D. student, one M.Sc. student, two undergraduate students, and one post-doctoral researcher were mentored in the context of this project. Informal laboratory instruction was the key instrument. The post-doctoral researcher got a faculty position in Brazil. The Ph.D. student finished his Ph.D. in Fall 2019 and is continuing work on agricultural automation at UC Davis, as a post-doctoral researcher. The M.Sc. student is working at a local robotics start-up company that is developing a mushroom harvesting robot. How have the results been disseminated to communities of interest?The PI gave several presentations related to the project and to mechanical harvesting in general. The audiences were growers, entrepreneurs, researchers, students, academic staff and the general public. January 30, 2019. Cling Peach Board, Sacramento, CA. February 5, 2019. California Pear Board, Davis, CA. February 13, 2019. Morning Star Company, Davis, CA. March 19, 2019. World Bank Headquarters, Washington, D.C. April 1, 2019. Western Center for Agricultural Health and Safety, Davis, CA. April 3, 2019. California Strawberry Commission, Cal Poly at Saint Luis Obispo. August 27, 2019. Taylor Farms, Salinas, CA. What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
In the context of this project, a set of trees and their fruits were digitized and software was developed to simulate two main mechanical fruit harvesting methods: shake-catch harvesting, and robotic harvesting. Progress in mechanical harvesting is crucial for national food security, especially given the farm labor shortages we are facing. Regarding shake-catch harvesting, simulations indicated that it is possible to harvest fruits by using novel multi-level fruit catching systems. As a result, an SCRI proposal was developed and submitted to develop such systems. Regarding robotic harvesting, simulations showed that novel multi-arm robots are needed to harvest fruits efficiently. The results from this project led to a successful proposal under the NSF National Robotics Initiative - funded by NIFA - to develop a physical multi-arm robot for fruit harvesting. The project's major accomplishments are listed next. A large data set of tree geometries and attached fruits was created. Twenty V-shaped clingstone peach trees were digitized at Escalon, CA and twenty high-density Bartlett pear trees were digitized at Ukiah, CA. Matlab code was written to process, filter data and calculate fruit spatial distributions and tree branching parameters. Spatial accuracy was experimentally verified to be better than 1 cm. Also, code was written to isolate parts (branches with fruits) of the existing tree models and combine them together in order to create new instances of trees that were used to create larger virtual orchards. The data were uploaded onto an open data repository (https://datadryad.org) with permanent DOI: https://doi.org/10.25338/B8WW41. Software was developed to compute fruit motions (falling, bouncing, rolling) during their interaction with machine surfaces, and layers of parallel tubes were pushed horizontally inside ten pear and ten cling peach digitized trees. Each fruit fell vertically and the fruit catching efficiency was calculated as the percentage of dropped fruits that landed on a tube or the ground (where they are collected without damage) without colliding with a branch. Results showed that without tubes, 36% of pears and 21% of peaches hit a branch, whereas with four layers these numbers dropped to 12% and 8% respectively. So, it is possible to harvest fruits by using novel multi-level fruit catching systems. As a result, an SCRI proposal was developed and submitted to develop such systems. Also, software was written to evaluate the reachability of fruits when linear robot arms are used to pick, rather than arms with revolute joints. This is referred to as Linear Fruit Reachability (LFR) and is essentially the Fruit Picking Efficiency (FPE) metric calculated for linear arms. Simulations were performed with different arm parameters (maximum acceleration, travel speed, arm extension length, gripper size). The simulation results showed that about 81% of pears and 87% of peaches were reachable. This implies that for some trees of SNAP-type architectures fruit reachability may not require complex and expensive arms with many degrees of freedom. Also, software was written to assess the Fruit Picking throughput (FPT) for linear arms harvesting in virtual orchards. The simulator modeled the kinematics and dynamics of a harvester comprising identical, independent, non-interfering arms. Each arm moved inside a 'cell' in an array configuration and picked fruits only from the canopy volume allocated to it, in a sequence that minimized arm travel time. Overall, a ten-fold decrease in PCT (from 4.6 s to 0.46 s) was only possible with optimally sized cells and at least 14 arms. These results indicate that achieving low PCTs for cost-effective fruit tree harvesting requires many arms that must be carefully load-balanced in the presence of changing, partially known fruit distributions. Based on these results, a proposal was funded by NIFA under the NSF National Robotics Initiative to develop a physical multi-arm robot for fruit harvesting. A two-arm robot (Baxter) was programmed to detect fruit using a camera and reach fruit using its arms and suction grippers. In lab settings, picking times of 4 s per fruit were recorded, in good correspondence with simulated fruit picking using a model of the robot. Major dissemination activities included conference presentations, journal publications, presentations at grower meetings, Commodity Boards and industry, demonstrations to the public during UC Davis's picnic day, and the release of publicly open data for the digitized trees and fruits: https://doi.org/10.25338/B8WW41.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2019
Citation:
Vougioukas, S.G. (2019). Agricultural Robotics. Annual Review of Control, Robotics, and Autonomous Systems. 2:365-392.
https://doi.org/10.1146/annurev-control-053018-023617
- Type:
Journal Articles
Status:
Published
Year Published:
2019
Citation:
Charlton, D., Edward Taylor, J.E., Vougioukas, S.G., Rutledge, Z. (2019). Innovations for a Shrinking Agricultural Workforce. Choices, 2nd Quarter 34(2).
http://www.choicesmagazine.org/choices-magazine/submitted-articles/estimating-value-damages-and-remedies-when-farm-data-are-misappropriated/innovations-for-a-shrinking-agricultural-workforce
- Type:
Journal Articles
Status:
Published
Year Published:
2019
Citation:
Charlton, D., Edward Taylor, J.E., Vougioukas, S.G., Rutledge, Z. (2019). Can Wages Rise Quickly Enough to Keep Workers in the Fields? Choices, 2nd Quarter 34(2).
http://www.choicesmagazine.org/choices-magazine/submitted-articles/can-wages-rise-quickly-enough-to-keep-workers-in-the-fields
|
Progress 01/01/18 to 12/31/18
Outputs
Target Audience:Students: One Ph.D. student and one post-doctoral researcher were mentored in the context of this project. Informal laboratory instruction was the key instrument. Growers: PI Vougioukas presented results from this project in grower and industry stakeholder meetings such as: Cling Peach Day on January 4, 2018; Lake County North Coast Pear Growers meeting on February 7, 2018; Sacramento River District Growers meeting, February 6, 2018; Tree Fruit Mechanization Day, Walnut Grove and Linden, CA, June 26, 2018. General public: UC Davis picnic Day, April 21, 2018; Innovation Summit, UC Davis, April 23, 2018. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?Students: One Ph.D. student and one post-doctoral researcher were mentored in the context of this project. Informal laboratory instruction was the key instrument. The post-doctoral researcher got a faculty position in Brazil. The Ph.D. student will submit his dissertation in May 2019. How have the results been disseminated to communities of interest?Growers: PI Vougioukas presented results from this project in grower and industry stakeholder meetings such as: Cling Peach Day on January 4, 2018; Lake County North Coast Pear Growers meeting on February 7, 2018; Sacramento River District Growers meeting, February 6, 2018; Tree Fruit Mechanization Day, Walnut Grove and Linden, CA, June 26, 2018. General public: A demo of a fruit grasping robot was set up during UC Davis picnic day on April 21, 2018, and hundreds of visitors - including young children - interacted with it and learned about robotic fruit harvesting research. The PI presented at the UC Davis Innovation Summit, April 23, 2018, to an audience of entrepreneurs, investors, growers, students, faculty and general public. What do you plan to do during the next reporting period to accomplish the goals?In the summer of 2018, we could not digitize trees in Lake County orchards in late August-early September because of wildfires in the area. With the no-cost extension provided to us, we'll be able to gather some more data in the next picking season. We'll make the data available in the UC Davis open-access repository. We will also finalize the harvest simulation analysis and auto-guidance experiments and complete the manuscripts that report our findings.
Impacts
What was accomplished under these goals?
Accomplishments Objective 1. Orchard modeling Data was not gathered during this reporting period because of wildfires in CA. However, the existing orchard models had a large number of trees (40) that enabled performing harvesting simulations (Objective 2). More data will be gathered in the summer of 2019. Objective 2. Harvest simulation In the previous report period, software was written to evaluate the reachability of fruits when linear arms - rather than arms with revolute joints - are used to pick fruits. This is referred to as Linear Fruit Reachability (LFR) and corresponds to the Fruit Picking Efficiency (FPE) metric calculated for linear arms. In this reporting period, our work extended the concept of LFR by introducing physical constraints for the harvesting actuators, such as maximum reach, gripper cross-section, minimum distance constraint of the machine from the canopy due to branches extending in the row, and possible directions of approach given the arm length. High density trellised pear and v-trellised cling peaches were the training systems chosen for the current research. Arm lengths of 100 to 200 cm with an increment of 25 cm for pears and 200 to 300 cm with an increment of 25 cm for peaches, and gripper sizes of 8 to 12 cm for pears and 7 to 11 cm for peaches were tested. Two gripper designs were considered. 'Adaptive' grippers, whose cross-section can increase (by opening) from a minimum size to the size of the fruit to be harvested, and fixed-size grippers (e.g., vacuum ones) whose cross-section must be larger than the largest fruits to be picked. One significant result was that adaptive grippers picked 7% more fruit compared to vacuum grippers. Another main finding was that the cumulative LFR after three passes was estimated to be 93.5% for pears and 94.2% for peaches. This means that linear arms are a practical way of harvesting these fruit trees. In this reporting period, the fruit picking cycle times (PCTs) of linear multi-arm harvester configurations were also investigated for all reachable fruits under static harvesting scenario. A methodology was developed to estimate the fruit PCT of linear multi-arm harvesters using models of digitized trees. Simulation analyses showed the dependence of single-arm PCT on maximum arm acceleration and maximum velocity. A range of accelerations and velocities were tested for pear and peach trees, and results showed that PCT dependence on maximum acceleration and maximum speed is described by a power law. The dependence of linear arm array PCT on the number of arms was investigated, and results indicated that PCT follows a decaying power law as the number of arms increases for both peaches and pears. Hence increasing the number of arms provides diminishing returns in terms of per-arm. The PCTs of linear arm arrays were estimated under two different scenarios for the size and shape of each arm's work cell: a) the work cells of all arms were equal in size (square), and b) the work cell of each arm had the same amount of fruit in it, resulting in rectangular but unequal work cells. Results indicated that harvesting pear and peach fruits using arms allocated under the same fruit load work cells is 25% to 50% faster than harvesting with arms operating in equal-sized work cells. Finally, the dependence of PCT on three different harvest scenarios were tested: harvesting both sides of a row, straddling the trees and harvesting both sides of each tree, and independent harvesting of each side of a row. PCTs achieved for the 2nd harvesting technique are higher than the PCTs achieved by 1st and 3rd techniques for the case of equal fruit load allocation to arms. The PCTs achieved for the 2nd harvesting technique are lower than the PCTs achieved by 1st and 3rd techniques. Also, the PCTs were low for 1st technique compared to the 3rd technique initially as the arms increased the 3rd technique has lower PCTs than the 1st technique for the case of equal cell size allocation to arms. Simulations results on PCT of the multi-arm harvester harvesting one side with an unequal number of arms with equal cell size allocation indicates that for a given number of arms several combinations of arms mounted on each side attain similar or better PCT's. Objective 3. Experimental evaluation During this reporting period, our efforts focused on the autonomy of the vehicle that will be used, and in particular on software for navigation in orchards using a single 3D camera, without continuous GPS signal availability. The approach developed is innovative in that it uses the 3D point cloud at places in the orchard where the presence of landmarks provides ground truth for the robot pose, to develop automatically a sensor model for the vehicle to use in a particle filtering approach. Results in simulation and orchard experiments are excellent (cm-level accuracy in lateral offset and less than 5o yaw error) and a manuscript is under preparation. Objective 4. Dissemination Our dissemination activities are described in the section "How have the results been disseminated to communities of interest?"
Publications
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2018
Citation:
Arikapudi, R., Vougioukas, S. (2018). A Study on Pick-Cycle-Times of Robotic Multi-Arm Tree Fruit Harvesters. Intl. Conference on Agricultural Engineering (AgEng 2018)
- Type:
Journal Articles
Status:
Accepted
Year Published:
2019
Citation:
Vougioukas, S.G. (2019). Agricultural Robotics. Annual Review of Control, Robotics, and Autonomous Systems. (2):xx-xx. https://doi.org/10.1146/annurev-control-053018-023617
- Type:
Journal Articles
Status:
Published
Year Published:
2018
Citation:
Durand-Petiteville, A., Le Flecher, E., Cadenat, V., Sentenac, T., Vougioukas, S.G. (2018). Tree detection with low-cost 3D sensors for autonomous navigation in orchards. IEEE Robotics and Automation Letters. 3(4): 3876-3883. https://doi.org/10.1109/LRA.2018.2857005
|
Progress 01/01/17 to 12/31/17
Outputs
Target Audience:Students: One Ph.D. student and two undergraduate students were mentored in the context of this project. Informal laboratory instruction was the key instrument. The Ph.D student also presented at the 2017 ASABE Inlt. meeting at Spokane, WA. Growers: Our team visited one grower in Escalon, CA and another one in Ukiah, CA and informed them about the project and also performed a lot of digitization work in their orchards. Growers were also reached at annual grower meetings (see section on dissemination). General public: Precision Ag conference series, organized by the World Food Center at UC Davis, April 27, 2017; CITRIS Agricultural Technology Fair, UC Merced, March 8, 2017; UC Davis picknic Day, April 22, 2017. Researchers: The international researcher community was reached vat the ASABE 2017 international meeting at Spokane, WA. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?One Ph.D., two undergraduate students and one staff member were mentored and trained in tree and fruit digitization and geometric modeling. Training was performed via collaboration, mentoring and supervision in the lab and in field experiments. The Ph.D. student, Mr. Arikapudi, attended and presented his work at the 2017 ASABE Intl. meeting, at Spokane, WA. How have the results been disseminated to communities of interest?The PI presented results from this project in grower and industry stakeholder meetings such as: sweet cherries grower meeting on Jan. 5, 2017 at Cabral Agricultural Center, Stockton, CA; Washington Tree Fruit Commission, at UC Davis, March 1, 2017; Almond Board of California, Harvest Technology Roundtable at UC Davis, March 15, 2017; Silicon Valley Forum at UC Davis, 4 April 4, 2017; Interpera conference, June 15, 2017; A demo of a fruit grasping robot was set up during UC Davis picknic day on April 22, 2017 and hundreds of visitors - including young children - interacted with it and learned about robotic fruit harvesting research. Presentations were also made to wide audiences (growers, students, academics, industry) at the Precision Ag conference series, organized by the World Food Center at UC Davis, April 27, 2017; at the CITRIS Agricultural Technology Fair, UC Merced, March 8, 2017. What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
Objective 1. Orchard modeling A new, large data set of tree geometries and attached fruits was created in summer 2017. Twenty V-shaped clingstone peach trees were digitized at Escalon, CA and twenty high-density Bartlett pear trees were digitized at Ukiah, CA. Matlab code was written to process, filter data and calculate fruit spatial distributions and tree branching parameters. Also, code was written to isolate parts (branches with fruits) of the existing tree models and combine them together in order to create new instances of trees that were used to create larger virtual orchards. Objective 2. Harvest simulation Software was written to evaluate the reachability of fruits when linear arms are used to pick, rather than arms with revolute joints. This is referred to as Linear Fruit Reachability (LFR) and is essentially the Fruit Picking Efficiency (FPE) metric calculated for linear arms. Simulations were performed with different arm parameters (maximum acceleration, travel speed, arm extension length, gripper size). The simulation results based on the digitized geometric tree models and fruit locations showed that a highest percentile of the fruit i.e., about 81% were reachable using an arm extension of 150 cm and the gripper size being 13 cm for pears. For peaches 87% fruits were reachable using an arm extension of 250 cm and the gripper size of 12 cm. This implies that for some trees of SNAP-type architectures fruit reachability may not require complex and expensive arms with many degrees of freedom. Of course, thin, flexible branches were not included in this study and their effects would need to be evaluated. The LFR's for 1st pass was 78%, 2nd pass was 2% and 1% for the 3rd pass respectively considering the gripper size of 13 cm and an arm extension of 150 cm. The LFR's for 1st pass was 80%, 2nd pass was 5% and 2% for the 3rd pass respectively considering the gripper size of 12 cm and an arm extension of 250 cm for peaches. This knowledge about the percentile fruit reachability for individual passes was used to determine the type of robot structure that could be used to harvest the fruit on these training systems. This helps in making economic decisions in the harvester design as to see the harvest cost per fruit for individual passes. Also, software was written to assess the Fruit Picking throughput (FPT) for linear arms harvesting in virtual orchards. The simulator modeled the kinematics and dynamics of a harvester comprising identical, independent, non-interfering arms. Each arm moved inside a 'cell' in an array configuration and picked fruits only from the canopy volume allocated to it, in a sequence that minimized arm travel time. Each tree was harvested independently, while the harvester was static in front of it. Digitized models of fruit-bearing, high-density trellised pear trees were used. Individual fruit distributions were non-uniform and varied significantly among trees. Such yield variation can introduce significant inefficiencies in multi-arm robot harvesting. Two scenarios were considered: A) all arm cells had the same size; B) cells were unequal and optimally sized to contain approximately the same number of fruits. Scenario (B) serves as an 'upper bound' on performance, since all arms harvest the same load and finish almost together. Results showed that in both scenarios Picking Cycle Time (PCT), which is the inverse of FPT, followed a power law; it did not drop linearly as the number of arms increased. Overall, the PCTs of scenario A were three times worse than the corresponding PCTs of scenario B; this is the result of non-uniform fruit distribution. A ten-fold decrease in PCT (from 4.6 s to 0.46 s) was only possible with optimally sized cells and at least 14 arms. Achieving such low PCT with uniform cells required more than 40 arms. These results indicate that achieving low PCTs for cost-effective fruit tree harvesting requires many arms that must be carefully load-balanced in the presence of changing, partially known fruit distributions. Objective 3. Experimental evaluation During this reporting period our efforts focused on the autonomy of the mobile vehicle that will be used, and in particular on software for navigation in orchards using 3D cameras, without continuous GPS signal availability. A low cost, efficient vision-based system was developed to detect accurately and robustly the trees in orchard rows so that the line to be followed can be extracted. It is made of four stereo cameras which provide a point cloud characterizing the environment. The key idea is to find the tree trunks by detecting their shadows which are materialized by concavities in the obtained point cloud. In this way, branches and leaves are not taken into account, improving the detection robustness which is an important feature to obtain an efficient navigation. A complete, real-time point cloud processing pipeline was developed and implemented using ROS and validated using data sequences taken in our orchards. A manuscript is under preparation. Objective 4. Dissemination Our dissemination activities are described in the section "How have the results been disseminated to communities of interest?"
Publications
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2017
Citation:
Arikapudi R., Vougioukas, S.G., (2017). Linear Reachability of Fruits in Orchard Trees with Actuator Constraints. ASABE Annual International Meeting. Paper Number 1701421, Spokane, Washington
|
Progress 01/01/16 to 12/31/16
Outputs
Target Audience:Students: The PI mentored one Ph.D. student, one MSc. student, one post-doctoral researcher and two undergraduate students (internships) in the context of this project. Informal laboratory instruction was the key instrument. Also, undergraduate and graduate students were reached via seminars: Mechanical and Aerospace Dept., UC Davis (4.28.16); UC Merced (4.22.16), and UC Davis Biological and Agricultural Eng. Dept. (6.9.16). Researchers: The international researcher community was reached via one journal publication in one of the best journals in our field (Biosystems Engineering) and in the context of two international conferences: AgriControl, in Seattle, USA, and CIGR-Eurageng, in Aarhus, Denmark. Growers: On 5.24.2016 the PI presented results from cling-peach digitization to growers from the Canning Peach Mechanization Research Fund. Co-PI Elkins presented at the Hood River Winter Horticulture Meeting, Oregon, USA, (2.5.16). Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?One Ph.D. student (Rajkishan Arikapudi) was mentored and trained in tree and fruit digitization and in modeling. One post-doctoral researcher (Adrien Durand Petitvile) was mentored and trained in autonomous vehicle navigation control and robot control. One MSc (Mathew Lefort) student was mentored and trained in lidar-based path-tracking control for autonomous vehicles. Two undergraduate students (Joshua Munic, Emile Le Flecher) were mentored and trained in simulation development and in robot programming. How have the results been disseminated to communities of interest?The international researcher community was reached via one journal publication in one of the best journals in our field (Biosystems Engineering) and in the context of two international conferences: AgriControl, in Seattle, USA, and CIGR-Eurageng, in Aarhus, Denmark. On 5.24.2016 the PI presented results from cling-peach digitization to growers from the Canning Peach Mechanization Research Fund. co_PI Elkins presented at the Hood River Winter Horticulture Meeting, Oregon, USA, (2.5.16). What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
The tree and fruit models generated during the first year of the project were used to explore - initialy in software - alternative ways to harvest mechanically fresh-market fruits using either novel shake-and-catch systems or novel robot designs. Preliminary results are very encouraging. Progress in harvest mechanization will offer significant relief to the labor shortage proglems that the Nation faces. Objective 1. Orchard modeling A1. Collect fruit position and size data in orchards. A2. Digitize orchard trees to extract structural-geometrical data. Activities A1 and A2 were conducted together in cling-peach (Martini Ranch, Echelon, CA) and pear (Ruddick Ranch, Ukiah) high-density orchards. Fruits positions and sizes were digitized by recording three points on each fruit. Also, the parent branch of each fruit was recorded. A3. Develop and validate fruit spatial distribution models. Matlab code was written to calculate fruit spatial distributions. Marginal one and two-dimensional distributions were calculated to characterize fruit density as functin of height, length along-row and combined. A4. Develop and validate tree structural-geometrical models. Matlab code was written to approximate tree geometries and structure. Branches were represented as sequences of frustums and tree structure was represented using an encoding scheme that captures tree branch topology. A5. Develop integrated tree and fruit models. Matlab code was developed to place the fruits on tree models developed in activity A4 and to visualize the fruits. Objective 2. Harvest simulation A6. Customize robotics simulator. The open-source Gazebo simulator was downloaded and installed and a model of the autonomous vehicle was created from scratch in Simulator Description Format (SDF) and imported to the simulator. The geometric and kinematic SDF model of the Baxter robot was also imported to Gazebo. Integration of vehicle and robot started but is under development. A7. Develop virtual orchard module. Geometric models of the digitized trees and the fruits were created in SDF format and imported into the Gazebo simulator. Objective 3. Experimental evaluation A8. Develop robotic platform. A ROS layer was developed to control the autonomous vehicle; its kinematics were identified. Autonomous navigation software is being developed using a commercial stereo camera. Software was written to communicate with the Baxter robot and perform basic control of its arms and read images from its camera. Closed-loop camera-based fruit reaching was programmed. A9. Compare predicted and achieved productivities No activity. Objective 4. Dissemination A10. Open-source digital content No activity. This will start when datasets have been completed. A11. Mechanization stakeholders Results were disseminated to growers and researchers, as described in the dissemination section (next).
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2016
Citation:
Arikapudi, R., Vougioukas, S.G., Jim�nez- Jim�nez, F., Farangis Khosro Anjom, F. (2016). Estimation of Fruit Locations in Orchard Tree Canopies Using Radio Signal Ranging and Trilateration. Computers and Electronics in Agriculture (125):160-172. http://dx.doi.org/10.1016/j.compag.2016.05.004
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2016
Citation:
Vougioukas, S., Arikapudi R., Munic, J. (2016). A Study of Fruit Reachability in Orchard Trees by Linear-Only Motion. 5th Agricontrol Conference, Seattle, USA. IFAC-PapersOnLine, 49(16), pp.277-280
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2016
Citation:
Vougioukas, S., Arikapudi R., Munic, J. (2016). Upper Bound Estimates of Fruit Reachability in Orchard Trees using Linear Motion. Proceedings of the Intl. Conference on Agricultural Engineering (AgEng CIGR 2016), Paper 529, Aarhus, Denmark.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2016
Citation:
Munic, J.P., Vougioukas, S.G., Arikapudi, R. (2016). A Study on Intercepting Falling Fruits with Canopy Penetrating Rods. 2016 ASABE Annual International Meeting. Paper Number 162456923, Orlando, Florida
|
|