스마트 온실 방제 로봇의 요구조건을 고려한 평가 방법 연구

김경철(Kyoung-Chul Kim),유범상(Beom-Sahng Ryuh),이시영(Siyoung Lee),김국환(Gookhwan Kim),이명훈(Meonghun Lee),홍영기(Young-ki Hong),김현종(Hyunjong Kim),유병기(Byeong-Kee Yu) 한국산학기술학회 2019 한국산학기술학회논문지 Vol.20 No.10

최근, 스마트 농업에 대한 관심이 증가함에 따라 농업 자동화 및 로봇에 대한 연구개발이 증가하고 있다. 스마트 온실에서 사용되는 로봇은 작업 특성과 생육 환경이 고려되어야 한다. 이를 위해 스마트 온실의 환경 분석을 통해 개발되어진 방제 로봇들을 대상으로 하였다. 본 논문은 스마트 온실에 적용 가능한 방제 로봇의 요구조건을 고려한 평가방법에 대한 연구를 수행하였다. 로봇의 요구조건을 통하여 성능 및 품질 평가 기준을 수립하고 시험을 실시하였다. 로봇관련표준을 참고하여 방제 로봇의 요구 기능과 목표를 도출하였다. 로봇의 성능을 위해 주행과 작업 능력 시험을 실시하였다. 주행시험은 로봇의 주행성능에 대한 시험을 실시하였고 작업능력은 방제성능에 대한 시험을 실시하였다. 로봇의 품질을 위해 내구성 시험을 실시하였다. 시험결과를 통하여 스마트 온실 로봇에게 요구되는 지표들을 도출할 수 있었다. 이를 통해 스마트 온실에 적용하기 위한 다양한 로봇들의 평가 기준이 될 것으로 판단된다. Recently, research and development on agricultural robots have been on the rise as the interest in smart farming has increased. Robots used in smart greenhouses should be taken into account with the working characteristics and growing environment. This study examined cleaning robots developed through the environmental analysis of smart greenhouses. This study assessed the evaluation method considering the requirements of the pest control robot applicable to the smart greenhouse. The performance and quality assessment criteria were established to conduct tests through the requirements of robots. The required functions and goals of the pest control robot were derived by referring to the robot-related standards. A driving and working ability test was conducted to assess the performance of the robot. The driving test was conducted on the driving performance of the robot and the work capability was tested on the pest control performance. In addition, a durability test was conducted to assess the quality of the robot. The required factors for smart greenhouse robots were derived from the test results. The study results are expected be a standard for an evaluation of a variety of robots for applications to smart greenhouses.

Autonomous Harvesting Robotic Design for ASABE Robotics Competition, Team SNU-Bot

( Chulwhan Yoon ),( Yong-hyun Kim ),( Gyujin Jang ),( Seongje Moon ),( Changho Yun ),( Hak-jin Kim ) 한국농업기계학회 2023 한국농업기계학회 학술발표논문집 Vol.28 No.2

Agricultural robots have been rapidly improving labor and agricultural productivity. Harvesting robots, one of the main applications of agricultural robots, are mobile devices that detect crops using cameras and harvest crops with robot arms. A lot of harvesting robots for strawberries, tomatoes, and apples have been developed, and cotton industry is also showing an interest in harvesting cottons using robots. However, one of the main challenges is in that cotton deforms easily and has varying shapes and orientations. The ultimate goal of this study was to develop an autonomous cotton harvesting robot. This study reports on the development of cotton harvesting algorithms required for the 2023 ASABE Robotics Competition. A four wheeled mobile robot equipped with a robotic arm and a depth camera was used. Experiments were conducted in a 2.5 m x 2.5 m indoor field consisting of simulated cotton samples and ping-pong balls representing the unripe cotton. For recognition, the YOLOv5 model was used to detect the cotton and compute the center of each cotton ball in real-time. The gripper was then programmed to be tilted for picking individual cotton balls. The detection model showed 94.8% in accuracy, and the harvesting algorithm using the tilted gripper showed an improved success rate of more than 10% as compared to those obtained without tilt. It was shown that the use of the algorithms developed in the study was successful in harvesting cotton balls in a laboratory scale-based field.

일본의 농작업 로봇 개발 동향 분석

이충근,홍영기 한국국제농업개발학회 2017 韓國國際農業開發學會誌 Vol.29 No.1

In order to distribute agricultural robots to the fields, based on the review of Japanese trends, three preliminary tasks and implications for the development of agricultural robots in Korea are derived as followings. The first result is that farmers are small-scale, agricultural robot is necessary to standardize through the public-private partnerships and reduce the supply price through the utilization of existing technologies and components. Secondly, the car-based robots are GPS-based system, and it is necessary to construct a nation-wide regional base stations that can trensmit the GPS correction signals free of charge in real time. Third, additional research is needed to cope with the unexpected situation in the fields, such as obstacle avoidance technology, robot safety evaluation technology, and safety standard development. In addition, although agricultural robots are being developed in Korea, a step-by-step approach is necessary for solving the above-mentioned pre-requisites for field distribution. In the first stage, it is necessary to supply an autonomous straight guidance system that assists the worker. The second stage is the spread of cooperation system between human and robot. The third stage is a complete robot system, in which the robot works on its own. Therefore, the supply of agricultural robots needs to be approached sequentially.

농작환경에 유연하게 적응하는 다기능 농업 로봇개발에 관한 연구

김민수,박홍석 한국컨설팅학회 2023 컨설팅융합연구 Vol.3 No.2

이 연구는 농업 분야에서의 생산성을 향상시키기 위해 다기능 로봇을 개발하는데 초점을 맞추고 있다. 기존의 농업 로봇은 제한된 기능과 단일 작업 수행 능력으로 인해 한정된 역할을 수행할 수밖에 없었다. 이러한 제약을 극복하기 위해, 본 연구에서는 다양한 작업을 수행할 수 있는 로봇을 설계하였다. 본 연구에서는 농업에 필수적으로 요구되는 작업인 급수, 휴림 피복, 쟁기질, 토양 습도 측정 등 다양한 작업을 수행할 수 있는 로봇을 개발하는 데 중점을 두고 있다. 설계된 로봇은 4개의 바 연결 메커니즘 을 사용하여 높이를 조정하고 습도측정 센서를 통해 토양의 건조 상태를 확인할 수 있다. 이에 대한 사 용모듈을 알고리즘화 하여 시스템을 구성함으로써 다양한 농업 환경에 적응할 수 있다. 이러한 농업용 다기능 로봇의 개발은 농업 생산성 향상과 농업 로봇 기술 분야의 발전에 크게 기여할 것으로 기대된다. Conventional agriculture robots have been mostly limited in their functions and are only capable of performing a single task. To overcome this problem, this research focuses on developing a robot that is able to perform various tasks required in agriculture such as water supply, mulching, plowing, and soil humidity measurement. The designed robot utilizes a 4-bar linkage mechanism for height adjustment and incorporates a humidity sensing sensor to monitor the soil's moisture level. By algorithmizing the utilization modules for these functions, the system is configured to adapt to various agricultural environments. The development of such a multi-functional robot for agriculture is expected to contribute significantly to improving agricultural productivity and advancing the field of robot technology in agriculture.

Non-Holonomic Agricultural Robot with Neural Network On-line Learning Controller

이현동,노대현 한국정밀공학회 2014 International Journal of Precision Engineering and Vol. No.

The present study represents agricultural applications of non-holonomic mobile robots. Agricultural robots are significantly affected by various disturbances such as a loading capacity and modeling errors of robots. Abundant studies of mobile robots using online learning have investigated real-time elimination of control errors that are associated to inaccurate modeling of robots and controls of disturbance effects by online learning using neural network. However, a certain problem from errors of online learning may occur in case of the robot that is located on out of tracking path. The above problem may be resulted in the difference by control errors occurring between desired values and current values. Therefore, stepwise optimization of robot control with desired values should be necessary. Online learning for agricultural mobile robots is possibly performed with accurate calculation of control errors using the stepwise optimization of desired values as a standard. In summary, the present study demonstrates a reference robot is used to calculate accurate control errors for non-holomic mobile robot that is driven by pulse. The control error of the non-holonomic mobile robot through online feedback-error learning is almost 1.3% in agricultural application.

농업 로봇의 품질을 위한 가속 수명평가 방법 및 장치 개발

김경철,홍영기,김국환,권경도,양창주,이기범 한국산학기술학회 2023 한국산학기술학회논문지 Vol.24 No.11

The food self-sufficiency rate is decreasing due to the decline in the agricultural production population. Advances in robot technology are increasing the number of robot applications in agriculture, but agricultural robots applied to agriculture are still inadequate. Quality improvement is required for robots to be used in agriculture, so evaluation technology is needed to reduce the frequency of failures. This study examined the durability of three types of greenhouse robots. A life evaluation test method was designed for the drive motor, a major component of the greenhouse robot, and the accelerated life test method using an accelerated life test device was applied for a life evaluation. Life evaluation tests were conducted on three types of greenhouse robots using the designed test method and test equipment. As a result of the evaluation, it was possible to obtain results that satisfied the minimum expected lifespan of five years required for agricultural machinery. The study result can be used as an evaluation index for commercializing greenhouse robots.

LiDAR 센서 활용 객체 인식기술이 적용된 농업용 자율주행 이송 로봇 개발

김종실(Jong-Sil Kim),주영태(Yeong-Tae Ju),김응곤(Eung-Kon Kim) 한국전자통신학회 2021 한국전자통신학회 논문지 Vol.16 No.3

우리나라 농업은 고령화로 인해 인력이 부족해 생산성이 감소하고 있다. 이를 극복하기 위해 농업용 로봇시장이 빠르게 성장하고 있으며 농업용 로봇의 원격제어와 자율주행에 관한 연구가 진행되고 있다. 본 연구는 농업용 로봇의 외형 및 구조를 설계하고 구동을 위한 장치 및 제어시스템을 구현하였다. LiDAR 센서를 활용, 최적화해 자율주행을 위한 필수 기능인 객체 인식기술을 적용하였다. 이를 통해 농업에서 노동력이 가장 많이 필요한 운송작업의 인건비 절감과 더불어 생산성을 향상 시킬 수 있다. Agriculture in South Korea is losing productivity due to the lack of manpower as aging population increases. To overcome this, the agricultural robot market is growing rapidly, and research is being conducted on remote control and autonomous driving of agricultural robots. This work designs the appearance and structure of agricultural robots and implements the devices and control systems for driving. By utilizing and optimizing LiDAR sensors, we applied object recognition technology, which is an essential function for autonomous driving. This can reduce labor costs and improve productivity of transportation tasks that require the most labor in agriculture.

A Review: Technology approach for autonomous driving in agricultural robot industry

( Seok Ho Kang ),( Seung Min Woo ),( Yu Shin Ha ) 한국농업기계학회 2023 한국농업기계학회 학술발표논문집 Vol.28 No.2

As autonomous technology takes place of human power, the robotic system becomes a compelling convergence method in most of the industry, especially in agriculture. High performance of plowing the land, harvesting, and real-time monitoring are such advantages of using it. Despite the constant performance in the protected system, the out door agricultural robot still faces challenges due to changing environmental reasons. The well-known problems are the path learning, limitation of sensors, battery management and etc. Moreover, it seems that the software used for data acquisition, has different performance with the driving mechanisms, 4-wheels and terrain tracked. In this study, a general summary of the challenges that autonomous technology in agriculture was served and solutions were reviewed. Then the applicable solutions were suggested for preventing the expected problem while performing the rotary operation. In order to perform the targeting operation, the path of field should be learned regarding soil and GPS precision resist, then the clarified technology required for clear identification in vision system. The reliability and durability of robot is required due to high performance of rotary operation and data filter is required for not to give a significant influence to data acquisition in open-field.

시설 농업 무인 관리를 위한 식물 생산 로봇 개발

김경철(Kyoung-Chul Kim),유범상(Beom-Sahng Ryuh) 제어로봇시스템학회 2011 제어·로봇·시스템학회 논문지 Vol.17 No.5

Environmental change, labor shortage, and international trade politics make agricultural automation ever more important. The automation demands the highest technology due to the nature of agriculture. In this paper, autonomous pesticide spray robot system has been developed for rose farming in the glass house. We developed drive platform, navigation/localization system, atomization spray system, autonomous, remote, and manual operation system, and monitoring system. The robot will be a great contribution to automation of hazardous labor-demanding chore of pesticide control in glass houses.

Soil sensing survey robots based on electronic nose

Theerapat Pobkrut,Teerakiat Kerdcharoen 제어로봇시스템학회 2014 제어로봇시스템학회 국제학술대회 논문집 Vol.2014 No.10

Nowadays autonomous robots are used in many filed of agriculture for increasing efficiency, and especially reducing the cost of the scarce human labor. Mobile ground robot and flying drone can be used to survey the farmland collecting necessary information such as ambient and crop conditions, soil fertility, pest and disease, etc. In this work, we have implemented an electronics nose (e-nose) based on metal oxide gas sensors installed on a mobile robot for detection of organic volatile compounds contained in the soil surface to examine the fertility of soil. Farmers can use such information to manage crops fertilizing or watering. To demonstrate this idea, we have mapped the soil volatiles at different places and having different types of soil, using laboratory-made portable e-nose and mobile e-nose robot. Therefore, the experiment was separated into 2 parts. The first part was to collect soils from different places and to measure the soil volatiles using portable e-nose that has eight metal-oxide gas sensors, in order to classify and confirm the difference between each sample. Secondly, the array of six gas sensors was then installed on a semi-autonomous six-wheeled robot chassis in order to measure real situations of soil surface located at four different places. It was found that both portable and robotized e-noses can identify and confirm the difference of soil volatiles when using principal component analysis (PCA) as pattern recognition.