G-ROBOT: An Intelligent Greenhouse Seedling Height Inspection Robot. (13th December 2022)
- Record Type:
- Journal Article
- Title:
- G-ROBOT: An Intelligent Greenhouse Seedling Height Inspection Robot. (13th December 2022)
- Main Title:
- G-ROBOT: An Intelligent Greenhouse Seedling Height Inspection Robot
- Authors:
- Wang, Baoqi
Ding, Yang
Wang, Chunyan
Li, Deyu
Wang, Hongchang
Bie, Zhilong
Huang, Yuan
Xu, Shengyong - Other Names:
- Yang Shuofei Academic Editor.
- Abstract:
- Abstract : An intelligent and modular greenhouse seedling height inspection robot was designed to meet the demand for high-throughput, low-cost, and nondestructive inspection during the growth of greenhouse seedlings. The robot structure mainly consists of a multiterrain replacement chassis, an electronic control lift image acquisition support, and a quick disassembly mechanism. SolidWorks was used to design the robot and Adams was used for motion simulations. Based on STM32 and Raspberry Pi as the core, the robot is equipped with various sensors to build a reliable control system for intelligent navigation for inspection tasks as well as acquisition of high-quality images and environmental information data of seedling crops. The developed growth point detection algorithm based on the EfficientNet deep learning network can efficiently measure the heights of seedlings and the application of the host software and cloud server makes it easy to monitor and control the robot and store and manage various data. The results of the greenhouse experiment showed that the robot has an average battery life of 5.2 h after being fully charged, with satisfactory motion stability and environmental adaptability; the environmental information data collected were valid, and errors were within the acceptable range; the captured seedling crop images were of high quality, and the seedling height data obtained through algorithm analysis were valid and reliable. The robot is expected to be anAbstract : An intelligent and modular greenhouse seedling height inspection robot was designed to meet the demand for high-throughput, low-cost, and nondestructive inspection during the growth of greenhouse seedlings. The robot structure mainly consists of a multiterrain replacement chassis, an electronic control lift image acquisition support, and a quick disassembly mechanism. SolidWorks was used to design the robot and Adams was used for motion simulations. Based on STM32 and Raspberry Pi as the core, the robot is equipped with various sensors to build a reliable control system for intelligent navigation for inspection tasks as well as acquisition of high-quality images and environmental information data of seedling crops. The developed growth point detection algorithm based on the EfficientNet deep learning network can efficiently measure the heights of seedlings and the application of the host software and cloud server makes it easy to monitor and control the robot and store and manage various data. The results of the greenhouse experiment showed that the robot has an average battery life of 5.2 h after being fully charged, with satisfactory motion stability and environmental adaptability; the environmental information data collected were valid, and errors were within the acceptable range; the captured seedling crop images were of high quality, and the seedling height data obtained through algorithm analysis were valid and reliable. The robot is expected to be an intelligent assistant for seedling research and production. … (more)
- Is Part Of:
- Journal of robotics. Volume 2022(2022)
- Journal:
- Journal of robotics
- Issue:
- Volume 2022(2022)
- Issue Display:
- Volume 2022, Issue 2022 (2022)
- Year:
- 2022
- Volume:
- 2022
- Issue:
- 2022
- Issue Sort Value:
- 2022-2022-2022-0000
- Page Start:
- Page End:
- Publication Date:
- 2022-12-13
- Subjects:
- Robotics -- Periodicals
Robotics
Periodicals
629.892 - Journal URLs:
- https://www.hindawi.com/journals/jr/ ↗
- DOI:
- 10.1155/2022/9355234 ↗
- Languages:
- English
- ISSNs:
- 1687-9600
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library HMNTS - ELD Digital store
- Ingest File:
- 24828.xml