ass日本风韵熟妇pics男人扒开女人屁屁桶到爽|扒开胸露出奶头亲吻视频|邻居少妇的诱惑|人人妻在线播放|日日摸夜夜摸狠狠摸婷婷|制服 丝袜 人妻|激情熟妇中文字幕|看黄色欧美特一级|日本av人妻系列|高潮对白av,丰满岳妇乱熟妇之荡,日本丰满熟妇乱又伦,日韩欧美一区二区三区在线

基于臨場(chǎng)感增強(qiáng)的果園機(jī)器人遙操作可視化系統(tǒng)研究
CSTR:
作者:
作者單位:

作者簡(jiǎn)介:

通訊作者:

中圖分類號(hào):

基金項(xiàng)目:

江蘇省重點(diǎn)研發(fā)計(jì)劃項(xiàng)目(BE2017370)


Teleoperation Visualization System of Orchard Robot Based on Enhancing Telepresence
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 圖/表
  • |
  • 訪問(wèn)統(tǒng)計(jì)
  • |
  • 參考文獻(xiàn)
  • |
  • 相似文獻(xiàn)
  • |
  • 引證文獻(xiàn)
  • |
  • 資源附件
  • |
  • 文章評(píng)論
    摘要:

    針對(duì)果園作業(yè)機(jī)器人使用單目相機(jī)進(jìn)行遙操作時(shí),,僅用二維視頻獲取環(huán)境信息缺乏臨場(chǎng)感的問(wèn)題,設(shè)計(jì)了一套基于臨場(chǎng)感增強(qiáng)的果園環(huán)境信息可視化系統(tǒng),,用于果園機(jī)器人遙操作,。系統(tǒng)由計(jì)算服務(wù)器、云服務(wù)器,、網(wǎng)絡(luò)攝像頭,、激光雷達(dá),、嵌入式開(kāi)發(fā)平臺(tái)等組成。計(jì)算服務(wù)器采用T7920工作站,,并在其上部署Tensorflow計(jì)算框架和Open3D點(diǎn)云算法庫(kù),,計(jì)算服務(wù)器在接收到云服務(wù)器轉(zhuǎn)發(fā)來(lái)的環(huán)境圖像和點(diǎn)云數(shù)據(jù)后,分別對(duì)圖像進(jìn)行導(dǎo)航信息增強(qiáng),,對(duì)點(diǎn)云進(jìn)行曲面重建,;嵌入式開(kāi)發(fā)平臺(tái)可以收集來(lái)自于網(wǎng)絡(luò)攝像頭和激光雷達(dá)的原始數(shù)據(jù),并上傳至云服務(wù)器,;在云服務(wù)器部署了以ZeroMQ為基礎(chǔ)的消息中轉(zhuǎn)程序和HTML5后臺(tái)服務(wù),,提供跨互聯(lián)網(wǎng)的消息通信服務(wù)和可移動(dòng)的遙操作環(huán)境信息可視化服務(wù)。測(cè)試結(jié)果表明,,部署在計(jì)算服務(wù)器的導(dǎo)航信息提取模型平均提取導(dǎo)航線時(shí)間86ms,,提取導(dǎo)航線平均精度16°,均優(yōu)于對(duì)比模型結(jié)果,。點(diǎn)云重建算法可以有效建立場(chǎng)景輪廓,平均精度4.9cm,,平均重建時(shí)間24ms,。壓縮圖像傳輸及增強(qiáng)處理時(shí)延不超過(guò)230ms,點(diǎn)云的傳輸時(shí)延不超過(guò)400ms,。各項(xiàng)參數(shù)可以滿足遙操作機(jī)器人進(jìn)行果園作業(yè)的基本要求,,相比僅有單目相機(jī)的遙操作,臨場(chǎng)感明顯增強(qiáng),,可為果園機(jī)器人遙操作提供參考,。

    Abstract:

    Aiming at the problem of lack of presence when the orchard robot used a monocular camera for remote operation, only using two-dimensional video to obtain environmental information lacked presence, a set of orchard environment information visualization system based on the enhanced sense of presence was designed. The system consisted of computing server, cloud server, network camera, LiDAR, embedded development platform, etc. The computing server adopted the T7920 workstation, and deployed the Tensorflow computing framework and the Open3D algorithm library of point cloud on it. After receiving the environmental image and point cloud data forwarded by the cloud server, the computing server enhanced the navigation information of the image, and surface reconstructed the point cloud. The embedded development platform could collect raw data from webcam and LiDAR, and uploaded them to cloud servers. A ZeroMQ-based message transfer program and HTML5 background service were deployed on the cloud server, providing cross-Internet message communication services and mobile teleoperation environment information visualization services. The test results showed that the average extraction time of the extraction model for navigation information deployed on the computing server was 86ms, and the average precision of the navigation line extraction was 16°, which were better than the results of the comparison model. The algorithm of point cloud reconstruction can effectively establish scene contours with an average accuracy of 4.9cm and an average reconstruction time of 24ms. The delay of compressed image transmission and enhancement processing did not exceed 230ms, and the transmission delay of point cloud did not exceed 400ms. The parameters could meet the basic requirements of the remote operation robot for orchard operation. The system significantly enhanced telepresence compared with that only with monocular cameras, which provided an effective reference for the remote operation of the orchard robot.

    參考文獻(xiàn)
    相似文獻(xiàn)
    引證文獻(xiàn)
引用本文

王運(yùn)東,周俊,孫經(jīng)緯,王凱,江自真,張震.基于臨場(chǎng)感增強(qiáng)的果園機(jī)器人遙操作可視化系統(tǒng)研究[J].農(nóng)業(yè)機(jī)械學(xué)報(bào),2023,54(3):22-31,78. WANG Yundong, ZHOU Jun, SUN Jingwei, WANG Kai, JIANG Zizhen, ZHANG Zhen. Teleoperation Visualization System of Orchard Robot Based on Enhancing Telepresence[J]. Transactions of the Chinese Society for Agricultural Machinery,2023,54(3):22-31,,78.

復(fù)制
分享
文章指標(biāo)
  • 點(diǎn)擊次數(shù):
  • 下載次數(shù):
  • HTML閱讀次數(shù):
  • 引用次數(shù):
歷史
  • 收稿日期:2022-06-17
  • 最后修改日期:
  • 錄用日期:
  • 在線發(fā)布日期: 2023-03-10
  • 出版日期:
文章二維碼