東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Model-based Estimation of Fruit Harv...

Arikapudi, Rajkishan .

Linked to FindBook

Google Book

Amazon

博客來

Model-based Estimation of Fruit Harvesting Performance of Arrays of Telescopic Robotic Arms.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Model-based Estimation of Fruit Harvesting Performance of Arrays of Telescopic Robotic Arms./
Author:	Arikapudi, Rajkishan .
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
Description:	165 p.
Notes:	Source: Dissertations Abstracts International, Volume: 81-11, Section: B.
Contained By:	Dissertations Abstracts International81-11B.
Subject:	Agricultural engineering. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27671259
ISBN:	9798645452513

Model-based Estimation of Fruit Harvesting Performance of Arrays of Telescopic Robotic Arms.
Arikapudi, Rajkishan .

Model-based Estimation of Fruit Harvesting Performance of Arrays of Telescopic Robotic Arms. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 165 p.

Source: Dissertations Abstracts International, Volume: 81-11, Section: B.

Thesis (Ph.D.)--University of California, Davis, 2020.

This item must not be sold to any third party vendors.

In fresh market fruit production, harvesting is one of the most labor-intensive and costly operations as most of the fresh market fruit produced in the fields are hand-harvested. So far, mass harvesting, mechanical-aided harvesting, and robotic harvesting are the three approaches used to harvest orchards, apart from manual harvesting. Mechanical harvesting for fresh-market fruits has remained an elusive target due to low picking performance. The reason for low picking performance is that mechanized harvesting productivity is dictated to a large extent by the design of the orchard and the machine. For example, the traditional open vase trees have more complex geometric structure compared to a high-density trellis or V-shaped trees. Also, the development of robotic fruit harvesters lacks appropriate modeling tools. In particular, the evaluation of the harvest capabilities of a given machine concept can only be performed in orchards using physical prototypes. The iterative process of building, testing, and modifying fruit-harvesting machines is expensive and slow and is restricted by limited harvesting windows. Therefore, model-based design tools that could enable researchers and developers to investigate the interrelationships among orchard layout, tree canopy geometry, and spatial fruit distribution, and the mechanics of tree fruit harvesting robots are needed. Such tools are anticipated to accelerate the development of novel orchard mechanized harvesting systems, thus enabling substantial gains in efficiencies of production using engineering and computing. Hence in this thesis, as a solution, a software design tool is developed that calculates the fruit picking efficiency and picking cycle time of a harvester design that utilizes an array of telescopic robot arms.To develop such a tool, one should model the tree architecture and spatial distribution of fruits in the canopy, along with the mechanics (e.g., kinematics, dynamics) and control of the tree fruit harvesting machine. However, such a model is not available. The first step in this study was to develop sensing systems that could digitize fruit trees and create geometric fruit tree models. A high throughput data acquisition system which utilized high-frequency radio signals and trilateration, as well as a highly accurate data acquisition system which utilizes an electromagnetic field for data acquisition was developed. A data set of 40 fruit trees of high-density pear and peach training systems was collected using a digitization system. All the fruits and rigid branches present on these 40 trees were digitized. A total of 6920 peaches were collected from 20 trees at an average of 346 fruits per tree. A total of 2922 pears were collected from 20 trees at an average of 146 fruits per tree. Overall, 9,842 fruit positions were recorded; with an average fruit localization rate of 9 fruits per minute. A framework to reconstruct digitized fruit tree models was developed using MATLAB software.The second part of this study is to evaluate the fruit picking efficiency (FPE) of harvesters with many cartesian/linear arms. To do this, digitized models of high-density pear trees and V-trellised cling peach trees with fruits, and harvester design specifications like gripper size, arm extension, and direction of approach were inputted into a physics simulation engine. Simulation analyses showed that high picking efficiencies could be achieved using linear arms; 93.3% and 88.9% of pears and peaches, respectively were reachable after three successive passes.The third part of this study is to evaluate the fruit picking cycle times (PCTs) of linear multi-arm harvester configurations. To do this, a MATLAB routine was developed, and all the reachable fruits were inputted into the simulation, and the harvesting algorithm was implemented under a static harvesting scenario. Results indicate that low pick cycle times for a linear multi-arm harvester (PCTM's) of < 1 sec/fruit could be achieved using harvesters carrying more than one linear arm.

ISBN: 9798645452513Subjects--Topical Terms:

3168406
Agricultural engineering.
Subjects--Index Terms:

Harvesting

Model-based Estimation of Fruit Harvesting Performance of Arrays of Telescopic Robotic Arms.
LDR:05208nmm a2200361 4500 001 2269241
005 20200910100205.5
008 220629s2020 ||||||||||||||||| ||eng d
020 $a 9798645452513
035 $a (MiAaPQ)AAI27671259
035 $a AAI27671259
040 $a MiAaPQ $c MiAaPQ
100 1 $a Arikapudi, Rajkishan . $3 3546548
245 1 0 $a Model-based Estimation of Fruit Harvesting Performance of Arrays of Telescopic Robotic Arms.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 165 p.
500 $a Source: Dissertations Abstracts International, Volume: 81-11, Section: B.
500 $a Advisor: Vougioukas, Stavros G.
502 $a Thesis (Ph.D.)--University of California, Davis, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a In fresh market fruit production, harvesting is one of the most labor-intensive and costly operations as most of the fresh market fruit produced in the fields are hand-harvested. So far, mass harvesting, mechanical-aided harvesting, and robotic harvesting are the three approaches used to harvest orchards, apart from manual harvesting. Mechanical harvesting for fresh-market fruits has remained an elusive target due to low picking performance. The reason for low picking performance is that mechanized harvesting productivity is dictated to a large extent by the design of the orchard and the machine. For example, the traditional open vase trees have more complex geometric structure compared to a high-density trellis or V-shaped trees. Also, the development of robotic fruit harvesters lacks appropriate modeling tools. In particular, the evaluation of the harvest capabilities of a given machine concept can only be performed in orchards using physical prototypes. The iterative process of building, testing, and modifying fruit-harvesting machines is expensive and slow and is restricted by limited harvesting windows. Therefore, model-based design tools that could enable researchers and developers to investigate the interrelationships among orchard layout, tree canopy geometry, and spatial fruit distribution, and the mechanics of tree fruit harvesting robots are needed. Such tools are anticipated to accelerate the development of novel orchard mechanized harvesting systems, thus enabling substantial gains in efficiencies of production using engineering and computing. Hence in this thesis, as a solution, a software design tool is developed that calculates the fruit picking efficiency and picking cycle time of a harvester design that utilizes an array of telescopic robot arms.To develop such a tool, one should model the tree architecture and spatial distribution of fruits in the canopy, along with the mechanics (e.g., kinematics, dynamics) and control of the tree fruit harvesting machine. However, such a model is not available. The first step in this study was to develop sensing systems that could digitize fruit trees and create geometric fruit tree models. A high throughput data acquisition system which utilized high-frequency radio signals and trilateration, as well as a highly accurate data acquisition system which utilizes an electromagnetic field for data acquisition was developed. A data set of 40 fruit trees of high-density pear and peach training systems was collected using a digitization system. All the fruits and rigid branches present on these 40 trees were digitized. A total of 6920 peaches were collected from 20 trees at an average of 346 fruits per tree. A total of 2922 pears were collected from 20 trees at an average of 146 fruits per tree. Overall, 9,842 fruit positions were recorded; with an average fruit localization rate of 9 fruits per minute. A framework to reconstruct digitized fruit tree models was developed using MATLAB software.The second part of this study is to evaluate the fruit picking efficiency (FPE) of harvesters with many cartesian/linear arms. To do this, digitized models of high-density pear trees and V-trellised cling peach trees with fruits, and harvester design specifications like gripper size, arm extension, and direction of approach were inputted into a physics simulation engine. Simulation analyses showed that high picking efficiencies could be achieved using linear arms; 93.3% and 88.9% of pears and peaches, respectively were reachable after three successive passes.The third part of this study is to evaluate the fruit picking cycle times (PCTs) of linear multi-arm harvester configurations. To do this, a MATLAB routine was developed, and all the reachable fruits were inputted into the simulation, and the harvesting algorithm was implemented under a static harvesting scenario. Results indicate that low pick cycle times for a linear multi-arm harvester (PCTM's) of < 1 sec/fruit could be achieved using harvesters carrying more than one linear arm.
590 $a School code: 0029.
650 4 $a Agricultural engineering. $3 3168406
653 $a Harvesting
653 $a Linear arms
653 $a Multi arm harvester
653 $a Pick cycle times
653 $a Robotics
653 $a Simulator
690 $a 0539
710 2 $a University of California, Davis. $b Biological Systems Engineering. $3 1684951
773 0 $t Dissertations Abstracts International $g 81-11B.
790 $a 0029
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27671259