国产日韩欧美一区二区三区三州_亚洲少妇熟女av_久久久久亚洲av国产精品_波多野结衣网站一区二区_亚洲欧美色片在线91_国产亚洲精品精品国产优播av_日本一区二区三区波多野结衣 _久久国产av不卡

?

基于DF2204無級變速拖拉機(jī)的農(nóng)機(jī)無人駕駛系統(tǒng)研制

2022-02-04 12:07陳智博楊衛(wèi)中楊麗麗吳才聰
關(guān)鍵詞:線控拖拉機(jī)無人駕駛

陳智博,文 龍,楊衛(wèi)中,楊麗麗,趙 欣,吳才聰

·農(nóng)業(yè)裝備工程與機(jī)械化·

基于DF2204無級變速拖拉機(jī)的農(nóng)機(jī)無人駕駛系統(tǒng)研制

陳智博,文 龍,楊衛(wèi)中,楊麗麗,趙 欣,吳才聰※

(1. 中國農(nóng)業(yè)大學(xué)信息與電氣工程學(xué)院,北京 100083;2. 農(nóng)業(yè)農(nóng)村部農(nóng)機(jī)作業(yè)監(jiān)測與大數(shù)據(jù)應(yīng)用重點(diǎn)實(shí)驗(yàn)室,北京 100083)

針對農(nóng)機(jī)無人化作業(yè)需求,該研究基于DF2204無級變速拖拉機(jī)和機(jī)器人操作系統(tǒng)(Robot Operating System,ROS),研發(fā)了一種適于田間作業(yè)的農(nóng)機(jī)無人駕駛自主作業(yè)系統(tǒng)。系統(tǒng)由控制、規(guī)劃、安全和總線通信等模塊組成。對DF2204無級變速拖拉機(jī)進(jìn)行硬件改造與集成,設(shè)計(jì)滿足農(nóng)機(jī)無人駕駛要求的控制器局域網(wǎng)(Controller Area Network,CAN)總線協(xié)議和ROS與CAN總線通信的消息結(jié)構(gòu),包括5類控制幀和2類狀態(tài)幀;設(shè)計(jì)了基于比例-積分-微分(Proportion Integration Differentiation,PID)控制器的橫向跟蹤與縱向速度控制算法。在北京密云試驗(yàn)田開展田間小麥播種實(shí)際作業(yè)試驗(yàn)。試驗(yàn)結(jié)果表明,消息結(jié)構(gòu)滿足50 Hz通信負(fù)載,橫向跟蹤平均絕對誤差為2.96 cm,縱向速度平均絕對誤差0.68 m/s。研究結(jié)果可為無級變速拖拉機(jī)的無人化升級改造提供參考,提高農(nóng)機(jī)智能化水平和作業(yè)效率。

拖拉機(jī);無級變速;無人駕駛;自主作業(yè);線控底盤

0 引 言

國內(nèi)農(nóng)業(yè)勞動(dòng)力日益減少,用工成本不斷上漲[1],發(fā)展無人化農(nóng)機(jī)自主作業(yè)技術(shù)勢在必行[2-4]。

在可預(yù)見的未來,國內(nèi)農(nóng)機(jī)動(dòng)力仍將以內(nèi)燃機(jī)為主,并將從手動(dòng)換擋升級至動(dòng)力換擋或無級變速(Continuously Variable Transmission,CVT),以提升農(nóng)機(jī)的動(dòng)力性和經(jīng)濟(jì)性。其中,無級變速技術(shù)具有傳動(dòng)結(jié)構(gòu)簡單、變速范圍大、油耗低等優(yōu)點(diǎn)[5-7],研發(fā)基于無級變速底盤的農(nóng)機(jī)無人駕駛作業(yè)技術(shù)具有重要意義[8]。

2008年約翰迪爾基于其動(dòng)力換擋和無級變速拖拉機(jī)研發(fā)了ITEC? Pro[9](Intelligent Total Equipment Control,ITEC)導(dǎo)航儀,能夠?qū)崿F(xiàn)自動(dòng)掉頭等功能,但仍需人員在車內(nèi)監(jiān)控;2022年又研發(fā)了基于8R 410的無人駕駛拖拉機(jī)[10],采用電動(dòng)無級變速和全線控技術(shù),通過6對雙目相機(jī)實(shí)現(xiàn)環(huán)境感知。2016年凱斯紐荷蘭研發(fā)了無人駕駛概念拖拉機(jī)[9],采用Magnum系列線控底盤和無級變速技術(shù),通過多傳感器實(shí)現(xiàn)環(huán)境感知。中國一拖于2018年展示了LF1104-C無人駕駛拖拉機(jī)[11],該機(jī)采用動(dòng)力換向技術(shù),能夠?qū)崿F(xiàn)前進(jìn)與后退等部分線控。2021年濰柴雷沃重工發(fā)布P7000無級變速拖拉機(jī)[12],能夠?qū)崿F(xiàn)除制動(dòng)以外的整車線控。吳才聰?shù)萚13]基于SF2104動(dòng)力換向拖拉機(jī)研發(fā)了無人駕駛與自主作業(yè)技術(shù)。

利用CAN總線實(shí)現(xiàn)主機(jī)和機(jī)具的電控或線控(Drive by Wire)[14]是實(shí)現(xiàn)農(nóng)機(jī)無人駕駛和自主作業(yè)的重要基礎(chǔ)。當(dāng)前,CAN總線協(xié)議作為車用總線標(biāo)準(zhǔn),已廣泛應(yīng)用于農(nóng)用車輛線控及狀態(tài)反饋[15-16]。何杰等[17]基于水稻插秧機(jī)與電動(dòng)方向盤,設(shè)計(jì)了具有CAN通信接口的插植機(jī)構(gòu)控制方案。闞輝玉等[18]針對動(dòng)力換擋重型拖拉機(jī)電控單元之間實(shí)時(shí)信息共享的需求,設(shè)計(jì)了整機(jī)CAN總線通信網(wǎng)絡(luò)。劉雪珂[19]開發(fā)了一種拖拉機(jī)自動(dòng)導(dǎo)航系統(tǒng)轉(zhuǎn)向控制器,并進(jìn)行轉(zhuǎn)向控制策略設(shè)計(jì)及驗(yàn)證。Wang等[20]實(shí)現(xiàn)了一種農(nóng)用拖拉機(jī)的自主控制算法和掉頭運(yùn)動(dòng)規(guī)劃算法,試驗(yàn)農(nóng)機(jī)采用基于CAN通信的洋馬EG105。近年來,國內(nèi)有關(guān)機(jī)構(gòu)對手動(dòng)換擋和非線控農(nóng)機(jī)的無人化升級改造[21-22]盡管在一定程度上實(shí)現(xiàn)了無人駕駛與自主作業(yè)功能,但仍然不具有經(jīng)濟(jì)性和推廣性。

此外,無人農(nóng)機(jī)屬于農(nóng)業(yè)機(jī)器人范疇,采用專用的類ROS(Robot Operating System,ROS),中間件(Middleware)應(yīng)引起足夠的重視[23]。ROS不僅能夠連接下層硬件與上層功能模塊,并且具有強(qiáng)大的仿真功能[24-26]。

綜上,針對無人駕駛農(nóng)機(jī)的研發(fā)與推廣需求,本文擬基于無級變速拖拉機(jī),從系統(tǒng)整體架構(gòu)入手,設(shè)計(jì)CAN協(xié)議和研發(fā)相應(yīng)的安全、規(guī)劃與控制技術(shù),并進(jìn)行面向?qū)嶋H作業(yè)環(huán)境的田間試驗(yàn)。

1 總體架構(gòu)設(shè)計(jì)

本研究設(shè)計(jì)的無人駕駛農(nóng)機(jī)(Autonomous Agricultural Vehicle,AAV)系統(tǒng)架構(gòu)包括硬件平臺和軟件系統(tǒng)兩部分(圖1)。硬件平臺主要向軟件系統(tǒng)傳輸傳感器和農(nóng)機(jī)底盤狀態(tài)信息,軟件系統(tǒng)進(jìn)行處理后生成執(zhí)行指令下發(fā)至硬件平臺以驅(qū)動(dòng)農(nóng)機(jī)具進(jìn)行動(dòng)作執(zhí)行,兩者基于CAN總線通信。

圖1 無人駕駛農(nóng)機(jī)系統(tǒng)架構(gòu)

軟件系統(tǒng)主要包括定位、規(guī)劃、控制、安全和CAN通信5個(gè)模塊,為了實(shí)現(xiàn)應(yīng)用層模塊間通信以及與硬件平臺的信息傳輸,使用基于Linux的ROS作為中間件,以降低模塊間耦合度,提高系統(tǒng)魯棒性。

硬件平臺由車載計(jì)算單元、各型傳感器和線控農(nóng)機(jī)構(gòu)成。其主要作用分別是提供計(jì)算能力和硬件接口,獲取環(huán)境信息和上傳底盤狀態(tài)、執(zhí)行控制指令。

1.1 硬件平臺

無人駕駛農(nóng)機(jī)的整體硬件組成與布局如圖2,主要設(shè)備包括GNSS/INS(Global Navigation Satellite System/Inertial Navigation System,GNSS/INS)組合導(dǎo)航系統(tǒng)、毫米波雷達(dá)、車載計(jì)算平臺、電控方向盤等。主機(jī)為DF2204無級變速拖拉機(jī),參數(shù)見表1。毫米波雷達(dá)型號為ARS408-21,俯仰角±20°,安裝在車前方設(shè)備支架上,距地高度80 cm。GNSS/INS組合導(dǎo)航型號為CGI-610,實(shí)時(shí)動(dòng)態(tài)(Real Time Kinematic,RTK)載波相位差分模式下水平精度1 cm,垂直精度3 cm。為提高系統(tǒng)安全性,拖拉機(jī)引擎蓋右側(cè)安裝有緊急停車開關(guān)(圖2中E),按下開關(guān)可實(shí)現(xiàn)整車斷電熄火。

1.2 CAN通信協(xié)議定義

農(nóng)機(jī)的線控能力是實(shí)現(xiàn)無人駕駛的基礎(chǔ)需求,主要包括農(nóng)機(jī)的擋位線控、驅(qū)動(dòng)線控、轉(zhuǎn)向線控、制動(dòng)線控、機(jī)具線控(如動(dòng)力輸出,后液壓控制等)等。本研究對DF2204無級變速拖拉機(jī)進(jìn)行線控化改造,除D、N、R擋位實(shí)現(xiàn)線控切換,低速與高速擋切換亦實(shí)現(xiàn)線控;驅(qū)動(dòng)方式通過請求車速或請求發(fā)動(dòng)機(jī)扭矩實(shí)現(xiàn);采用基于CAN總線協(xié)議的電控方向盤實(shí)現(xiàn)線控轉(zhuǎn)向,且接入整車CAN網(wǎng)絡(luò)。同時(shí)面向農(nóng)機(jī)無人駕駛需求制定基于SAE J1939協(xié)議和ISO11783協(xié)議的拖拉機(jī)整車CAN總線通信協(xié)議(表2),主要包括5類控制幀,2類狀態(tài)幀??刂茙捎?jì)算平臺(Computing Platform,CP)發(fā)出,分別用于轉(zhuǎn)向、驅(qū)動(dòng)、液壓輸出、液壓提升與動(dòng)力輸出(Power Take Off,PTO)控制,發(fā)送周期20 ms。狀態(tài)幀由農(nóng)機(jī)的整車控制單元(Vehicle Control Unit,VCU)發(fā)出,用于反饋農(nóng)機(jī)當(dāng)前狀態(tài),包括行駛狀態(tài)與作業(yè)狀態(tài),發(fā)送周期100 ms。

A.毫米波雷達(dá) B.傳感器掛載支架 C.角度傳感器 D.拖拉機(jī) E.緊急停車開關(guān) F.電控方向盤 G.GNSS天線 H.車載計(jì)算平臺 I.鋰電池與分線箱 J.GNSS/INS組合導(dǎo)航 K.線控液壓輸出 L.線控動(dòng)力輸出 M.線控液壓提升

表1 DF2204拖拉機(jī)主要參數(shù)

表2 整車CAN總線協(xié)議

注:CP為計(jì)算平臺;VCU為整車控制單元。

Note: CP is the computing platform; VCU is vehicle control unit.

1.3 軟件系統(tǒng)

1.3.1 整體結(jié)構(gòu)

本研究采用ROS作為系統(tǒng)中間件及應(yīng)用層軟件運(yùn)行環(huán)境(圖3),橢圓形表示軟件的節(jié)點(diǎn)序號與名稱,主要包括數(shù)據(jù)獲?。ü?jié)點(diǎn)1、3、6)、規(guī)劃(節(jié)點(diǎn)5)與控制(節(jié)點(diǎn)4),矩形表示節(jié)點(diǎn)間通信的話題(Topic)與消息(Message)名稱。

圖3 軟件系統(tǒng)整體架構(gòu)

數(shù)據(jù)輸入層分別為2個(gè)CAN接口與1個(gè)串口,其中CAN0與車輛底盤進(jìn)行通信,CAN1與毫米波雷達(dá)進(jìn)行通信,串口COM0接收GNSS/INS組合導(dǎo)航發(fā)送的衛(wèi)星定位數(shù)據(jù)和姿態(tài)數(shù)據(jù)。

各節(jié)點(diǎn)的主要功能如表3,為了解決多進(jìn)程無法訪問同一CAN設(shè)備的局限并提高程序可移植性,采用socketCAN技術(shù)將CAN接口抽象為套接字接口,從而實(shí)現(xiàn)節(jié)點(diǎn)1與節(jié)點(diǎn)6同時(shí)與CAN0通信。

表3 節(jié)點(diǎn)功能

由于CAN總線協(xié)議只定義了物理層以及鏈路層傳輸方式,讀取到的原始數(shù)據(jù)需按照CAN協(xié)議進(jìn)行解碼。預(yù)先設(shè)計(jì)的DBC(DataBase CAN)文件定義了CAN報(bào)文的編碼方式,包括數(shù)據(jù)幀ID、縮放以及偏移量。通過將DBC文件轉(zhuǎn)換為結(jié)構(gòu)體載入節(jié)點(diǎn)2將16進(jìn)制CAN數(shù)據(jù)轉(zhuǎn)為明文數(shù)據(jù),并發(fā)布至ROS消息隊(duì)列供其他節(jié)點(diǎn)使用。節(jié)點(diǎn)3還負(fù)責(zé)將自車坐標(biāo)從WGS 84(World Geodetic System 1984, WGS 84)坐標(biāo)系轉(zhuǎn)換為UTM(Universal Transverse Mercato,UTM)坐標(biāo)系,從而方便后續(xù)規(guī)劃與控制模塊的使用。

節(jié)點(diǎn)5生成的信息如表4,包括作業(yè)期望軌跡信息與機(jī)具控制信息,通過話題訂閱方式傳輸至控制節(jié)點(diǎn)。作業(yè)方式采用效率較高的分塊套圈法,采用翟衛(wèi)欣等[27]方法,通過讀取預(yù)先采集的農(nóng)田地理信息、作業(yè)方向、幅寬等參數(shù),生成路由順序再進(jìn)行運(yùn)動(dòng)規(guī)劃,生成作業(yè)軌跡和控制指令。

表4 規(guī)劃節(jié)點(diǎn)5的輸出內(nèi)容

由于農(nóng)田土壤松軟,阻力較大。經(jīng)測試,在不采取制動(dòng)措施情況下,農(nóng)機(jī)從10 km/h到靜止的滑行距離僅為2 m。因此節(jié)點(diǎn)6采用駕駛員預(yù)瞄安全距離模型[28],安全距離設(shè)置為10 m。配置雷達(dá)以O(shè)bject模式作為目標(biāo)輸出模式,該模式集成了目標(biāo)識別、跟蹤等算法。為了降低誤報(bào),將輸出目標(biāo)物體數(shù)據(jù)中的橫坐標(biāo)設(shè)置在?2~2 m內(nèi),大于3 m作業(yè)幅寬,滿足作業(yè)需求。當(dāng)檢測到跟蹤目標(biāo)距離小于安全距離時(shí),節(jié)點(diǎn)6直接向CAN0發(fā)送目標(biāo)速度為0的指令,防止農(nóng)機(jī)與障礙物碰撞。

1.3.2 控制模塊

無人駕駛農(nóng)機(jī)整體控制流程如圖4,控制節(jié)點(diǎn)通過訂閱規(guī)劃模塊發(fā)送的數(shù)據(jù)獲取跟蹤軌跡點(diǎn)、目標(biāo)速度、機(jī)具目標(biāo)動(dòng)作等信息,節(jié)點(diǎn)中的橫向控制器負(fù)責(zé)路徑的橫向跟蹤,從而將自車與規(guī)劃路徑的橫向誤差控制在一定范圍??刂乒?jié)點(diǎn)通過CAN總線傳輸控制指令至VCU。VCU內(nèi)包含2個(gè)控制器,分別是轉(zhuǎn)角控制器和縱向速度控制器,前者通過獲取前輪轉(zhuǎn)角傳感器數(shù)據(jù)與控制電控方向盤消除目標(biāo)轉(zhuǎn)角與當(dāng)前轉(zhuǎn)角誤差;后者通過發(fā)送請求扭矩到發(fā)動(dòng)機(jī)電子控制單元(Engine Control Unit,ECU),變速箱控制單元(Transmission Control Unit,TCU)根據(jù)工況調(diào)整發(fā)動(dòng)機(jī)傳動(dòng)比調(diào)節(jié)扭矩輸出,從而消除目標(biāo)速度與實(shí)際速度誤差。VCU通過轉(zhuǎn)發(fā)上層控制節(jié)點(diǎn)生成的機(jī)具控制命令到PTO控制單元與后液壓控制單元實(shí)現(xiàn)機(jī)具控制,以上控制單元由車廠提供。

整車控制主要包含橫向控制模塊與縱向速度控制模塊。橫向控制流程如圖5a,其中電控方向盤通過控制自車轉(zhuǎn)向軸實(shí)現(xiàn)轉(zhuǎn)向控制??v向速度控制如圖5b,通過專家PID算法消除自車輪速v與目標(biāo)速度v的誤差ε,(km/h),計(jì)算扭矩輸出量τ后發(fā)送至ECU實(shí)現(xiàn)縱向速度的控制。

注:φd為前輪目標(biāo)轉(zhuǎn)角,rad;vd為目標(biāo)速度,km×h-1;Mreq為農(nóng)機(jī)具目標(biāo)動(dòng)作。

注:δt、δe、ψ分別為橫向控制器設(shè)定值、誤差值與測量值,rad;φf、φd分別為當(dāng)前前輪轉(zhuǎn)角與目標(biāo)前輪轉(zhuǎn)角,rad;εφ為轉(zhuǎn)角控制器誤差值,rad;vd、vc、εv分別為目標(biāo)速度、當(dāng)前速度與速度誤差,km·h-1;τd為發(fā)動(dòng)機(jī)請求扭矩百分比,%。

δ代表自車與規(guī)劃路徑的橫向誤差:

注:A,B分別為車輛前后軸中點(diǎn),B=(X,Y)為農(nóng)機(jī)質(zhì)心;L為軸距,cm;V為車輛AB方向線速度,km·h-1;ψ、ψd分別為當(dāng)前偏航角與目標(biāo)偏航角,rad;Ti=(xi,yi)為當(dāng)前預(yù)瞄點(diǎn),,N為預(yù)瞄點(diǎn)數(shù)量;δlat代表橫向誤差,cm;E為B在上的投影;δe為與的夾角,rad;δt為與X軸正方向夾角,rad。

其中,

由式(2)知,當(dāng)δ足夠小時(shí)δ逐漸接近0,可以認(rèn)為農(nóng)機(jī)正按照預(yù)定路徑行駛。為了減少穩(wěn)態(tài)誤差,采用帶積分的PID控制器,為了提高控制器穩(wěn)定性,防止積分過飽和,橫向控制采用位置式遇限削弱積分PID控制方法:

式中()為目標(biāo)前輪轉(zhuǎn)角,rad,正值左轉(zhuǎn),負(fù)值右轉(zhuǎn),δ()為橫向控制器當(dāng)前誤差量,rad,K、K分別代表比例和微分系數(shù),u()代表積分環(huán)節(jié),由下式給出:

K為積分系數(shù),即當(dāng)積分環(huán)節(jié)大于π/6,且誤差與輸出同號時(shí)不進(jìn)行積分。PID控制器參數(shù)調(diào)整主要依靠經(jīng)驗(yàn)和田間測試,通過在田間測試時(shí)設(shè)置不同的參數(shù)組合記錄軌跡點(diǎn)并定量分析最終確定各系數(shù)如下:K=1.1,K=0.1,K=0.5。

2 田間試驗(yàn)

2.1 試驗(yàn)材料與方法

為了驗(yàn)證算法和系統(tǒng)整體可靠性,在北京市密云區(qū)某農(nóng)機(jī)合作社(40.347851°N,116.858439°E)進(jìn)行田間試驗(yàn)。為了便于分析作業(yè)效率與誤差,選擇矩形地塊作為無人駕駛試驗(yàn)區(qū)域(圖7),區(qū)域長333 m,寬72 m,總作業(yè)面積2.4 hm2。

田間試驗(yàn)于2021年10月開展,試驗(yàn)控制頻率50 Hz,數(shù)據(jù)記錄頻率10 Hz。通過型號為EMUC-B202的CAN記錄儀記錄CAN總線數(shù)據(jù)。位姿數(shù)據(jù)由厘米級GNSS/INS組合導(dǎo)航獲取,數(shù)據(jù)獲取頻率20 Hz。

注:A、B、C、D為作業(yè)區(qū)域4個(gè)頂點(diǎn),BC為導(dǎo)航參考線。

作業(yè)任務(wù)為冬小麥播種,作業(yè)幅寬3 m。農(nóng)機(jī)在進(jìn)行U形掉頭時(shí),考慮安全和作業(yè)效率,目標(biāo)速度為3 km/h。動(dòng)力驅(qū)動(dòng)耙與播種機(jī)所需總動(dòng)力約125~185 kW,試驗(yàn)農(nóng)機(jī)動(dòng)力輸出軸輸出功率134 kW,經(jīng)過測試發(fā)現(xiàn)目標(biāo)速度設(shè)定為7 km/h時(shí),發(fā)動(dòng)機(jī)負(fù)載保持在90%左右,因此直線作業(yè)時(shí)目標(biāo)速度設(shè)定為7 km/h。

由于試驗(yàn)前農(nóng)田已進(jìn)行旋耕作業(yè),因此試驗(yàn)區(qū)域土塊較大。為了提高作業(yè)質(zhì)量與作業(yè)效率,作業(yè)農(nóng)機(jī)具選擇當(dāng)康牌直播一體機(jī),該一體機(jī)由動(dòng)力驅(qū)動(dòng)耙與播種機(jī)組成,能夠同時(shí)完成耙地與播種作業(yè),其中播種機(jī)型號為2BF-20,驅(qū)動(dòng)耙型號為1BQ-3.0,具體參數(shù)如表5。

2.2 結(jié)果與分析

如圖8a,農(nóng)機(jī)按照規(guī)劃的路徑順序進(jìn)行作業(yè),整體路徑跟蹤效果與規(guī)劃路徑基本重合,共有22條直線作業(yè)路徑,將其按順序編號,分別命名為路徑1,2,…,22。如圖8b,試驗(yàn)區(qū)高程信息由組合導(dǎo)航獲取的高程信息通過克里金插值生成,試驗(yàn)區(qū)整體地形呈東南高,西北低,最高處與最低處相差2.6 m。

表5 動(dòng)力耙播種一體機(jī)參數(shù)

根據(jù)農(nóng)機(jī)播種作業(yè)實(shí)際需求,分別對農(nóng)機(jī)作業(yè)時(shí)的前輪轉(zhuǎn)向角誤差、偏航誤差、速度誤差與橫向誤差進(jìn)行定量分析。轉(zhuǎn)向角誤差φ(°)代表當(dāng)前轉(zhuǎn)向角與期望轉(zhuǎn)向角之差;偏航角誤差ψ(°)代表當(dāng)前偏航角與期望偏航角之差;速度誤差(km/h)為當(dāng)前速度與期望速度之差,橫向誤差由公式(2)給出。

量化指標(biāo)主要有均方根值、平均值、平均絕對誤差、標(biāo)準(zhǔn)差與導(dǎo)航誤差,導(dǎo)航誤差參考GB/T 37164-2018標(biāo)準(zhǔn),由下式計(jì)算:

其中NE為導(dǎo)航誤差,cm;δ,h為第個(gè)采樣點(diǎn)的橫向偏差,cm,為采樣總數(shù)。

如表6,共統(tǒng)計(jì)22條直線作業(yè)路徑,總作業(yè)時(shí)長5 943 s(去除加種時(shí)間),直線作業(yè)階段總時(shí)間4 037 s,掉頭時(shí)長1 906 s,掉頭用時(shí)占總時(shí)長32%,作業(yè)效率1.33 hm2/h。

表6 田間試驗(yàn)誤差分析

注:RMS為均方根值;Mean為平均值;MAE為平均絕對誤差;SD為標(biāo)準(zhǔn)差;NE為導(dǎo)航誤差。

Note: RMS is the root mean square; Mean is the average value; MAE is the mean absolute error; SD is the standard deviation; NE is the navigation error.

選取橫向誤差較大的路徑7為例進(jìn)行轉(zhuǎn)向角跟蹤分析,如圖9,農(nóng)機(jī)掉頭后進(jìn)入直線作業(yè)階段,完成直線作業(yè)后再次進(jìn)行掉頭,掉頭時(shí)誤差較大,直線作業(yè)時(shí)誤差穩(wěn)定在?5°~5°之間,直行階段轉(zhuǎn)向角均方根誤差2.76°,平均絕對誤差1.84°。

圖9 路徑7轉(zhuǎn)向角誤差變化

試驗(yàn)期間發(fā)動(dòng)機(jī)扭矩百分比平均值為80%,圖10為直線作業(yè)路徑16的速度誤差與發(fā)動(dòng)機(jī)扭矩百分比示意圖,掉頭時(shí),目標(biāo)速度為3 km/h,驅(qū)動(dòng)耙停止工作,播種機(jī)提升,發(fā)動(dòng)機(jī)扭矩百分比在10%~30%波動(dòng),負(fù)載較低。進(jìn)行直線作業(yè)時(shí),目標(biāo)速度為7 km/h,驅(qū)動(dòng)耙與播種機(jī)同時(shí)作業(yè),80 s后負(fù)載持續(xù)增加,扭矩輸出百分比保持在80%以上,最大值94%。為了滿足瞬時(shí)扭矩需求(如阻力增大、爬坡等),發(fā)動(dòng)機(jī)會進(jìn)行扭矩儲備,當(dāng)轉(zhuǎn)速超過扭矩峰值時(shí),負(fù)載增加將導(dǎo)致發(fā)動(dòng)機(jī)過載,轉(zhuǎn)速降低[31],作業(yè)速度降低,從而輸出更大扭矩。

圖10 路徑16速度誤差變化

如圖11,橫坐標(biāo)為所有作業(yè)路徑按時(shí)間順序的采樣點(diǎn),左側(cè)縱坐標(biāo)為農(nóng)機(jī)直線作業(yè)時(shí)的橫向誤差的絕對值,右側(cè)縱坐標(biāo)為農(nóng)機(jī)直線作業(yè)時(shí)的偏航角誤差絕對值,兩者呈現(xiàn)較強(qiáng)的相關(guān)性,Pearson相關(guān)系數(shù)為0.61。偏航角均方根誤差1.19°,平均絕對誤差0.91°,橫向均方根誤差為9.22 cm,平均值?2.91 cm,平均絕對誤差7.17 cm,導(dǎo)航誤差11.69 cm,其中97.3%的作業(yè)時(shí)間內(nèi)橫向誤差小于20 cm,說明拖拉機(jī)控制方案滿足實(shí)際作業(yè)需求。

圖11 偏航角誤差與橫向誤差

3 結(jié) 論

本文以東風(fēng)無級變速拖拉機(jī)為試驗(yàn)平臺,針對無人駕駛農(nóng)機(jī)的作業(yè)需求,制訂了滿足整車控制需求的CAN總線協(xié)議,以ROS為中間件開發(fā)了滿足作業(yè)需求的軟件平臺,包括定位、規(guī)劃、控制、CAN通信、安全模塊,設(shè)計(jì)了橫向控制模塊與縱向速度控制模塊。開展了實(shí)際作業(yè)的小麥播種田間試驗(yàn),其中橫向誤差平均值2.96 cm,導(dǎo)航誤差11.69 cm,速度均方根誤差0.98 km/h,平均絕對誤差0.68 km/h,轉(zhuǎn)向角均方根誤差1.91°,平均絕對誤差1.47°。試驗(yàn)結(jié)果表明,基于線控技術(shù)與無級變速技術(shù)的拖拉機(jī)滿足無人駕駛農(nóng)機(jī)研制需求,控制與規(guī)劃軟件能夠滿足無人駕駛農(nóng)機(jī)進(jìn)行播種作業(yè)的需求。下一步將基于當(dāng)前研究基礎(chǔ)進(jìn)一步開展農(nóng)機(jī)田內(nèi)決策規(guī)劃與環(huán)境感知等農(nóng)機(jī)無人駕駛相關(guān)技術(shù)研究。

[1] 龐春霖. 智能農(nóng)機(jī)技術(shù)路線圖[M]. 成都:電子科技大學(xué)出版社,2022.

[2] Roshanianfard A, Noguchi N, Okamoto H, et al. A review of autonomous agricultural vehicles[J]. Journal of Terramechanics, 2020, 91: 155-183.

[3] Yin X, Wang Y, Chen Y, et al. Development of autonomous navigation controller for agricultural vehicles[J]. International Journal of Agricultural and Biological Engineering, 2020, 13(4): 70-76.

[4] 胡靜濤,高雷,白曉平,等. 農(nóng)業(yè)機(jī)械自動(dòng)導(dǎo)航技術(shù)研究進(jìn)展[J]. 農(nóng)業(yè)工程學(xué)報(bào),2015,31(10):1-10.

Hu Jingtao, Gao Lei, Bai Xiaoping, et al. Review of research on automatic guidance of agricultural vehicles[J]. Transactions of The Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2015, 31(10): 1-10. (in Chinese with English abstract)

[5] 薛麗君,趙業(yè)慧,宋悅,等. 無級變速和電傳動(dòng)農(nóng)業(yè)作業(yè)機(jī)械現(xiàn)狀研究[J]. 中國農(nóng)機(jī)化學(xué)報(bào),2022,43(7):81-89.

Xue Lijun, Zhao Yehui, Song Yue, et al. Research on the current situation of continuously variable transmission and electric drive technology[J]. Jorunal of Chinese Agricultural Mechanization, 2022, 43(7): 81-89. (in Chinese with English abstract)

[6] 王韋韋,陳黎卿,楊洋,等. 農(nóng)業(yè)機(jī)械底盤技術(shù)研究現(xiàn)狀與展望[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2021,52(8):1-15.

Wang Weiwei, Chen Liqing, Yang Yang, et al. Development and prospect of agricultural machinery chassis technology[J]. Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(8): 1-15. (in Chinese with English abstract)

[7] Coffman B, Kocher M, Adamchuk V, et al. Testing fuel efficiency of a tractor with a continuously variable transmission[J]. Applied Engineering in Agriculture, 2010, 26(1): 31-36.

[8] Xiao M, Zhao J, Wang Y, et al. Fuel economy of multiple conditions self-adaptive tractors with hydro-mechanical CVT[J]. International Journal of Agricultural and Biological Engineering, 2018, 11(3): 102-109.

[9] Baillie C, Lobsey C, Antille D, et al. A review of the state of the art in agricultural automation. Part III: Agricultural machinery navigation systems[C]//Detroit: Proceedings of the 2018 ASABE Annual International Meeting. ASABE, 2018.

[10] Allison M. The autonomous future of industrial vehicles: Interview with John Deere business manager Jesse Haecker[EB/OL]. Robotics and Automation News, 2022-01-12[2022-07-15]. https://roboticsandautomationnews.com/2022/01/12/the-autonomous-future-of-large-industrial-vehicles-interview-with-john-deere-business-manager-jesse-haecker/48245/.

[11] 許予永. 東方紅無人駕駛拖拉機(jī)閃耀央視舞臺[N]. 農(nóng)機(jī)質(zhì)量與監(jiān)督, 2018-09-25.

[12] 姚彬. 濰柴雷沃CVT智能拖拉機(jī)榮耀發(fā)布[N]. 農(nóng)機(jī)市場,2021-11-18.

[13] 吳才聰,王東旭,陳智博. SF2104 拖拉機(jī)自主行駛與作業(yè)控制方法[J]. 農(nóng)業(yè)工程學(xué)報(bào),2020,36(18):42-48.

Wu Caicong, Wang Dongxu, Chen Zhibo. Autonomous driving and operation control method for SF2104 tractors[J]. Transactions of The Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(18): 42-48. (in Chinese with English abstract)

[14] Han S, He Y, Fang H. Recent development in automatic guidance and autonomous vehicle for agriculture: A Review[J]. Journal of Zhejiang University (Agriculture & Life Sciences), 2018, 44(4): 381-391.

[15] Mattetti M, Maraldi M, Lenzini N, et al. Outlining the mission profile of agricultural tractors through CAN-BUS data analytics[J]. Computers and Electronics in Agriculture, 2021, 184: 106078.

[16] 劉陽. 拖拉機(jī)自動(dòng)導(dǎo)航系統(tǒng)CAN總線的設(shè)計(jì)與實(shí)現(xiàn)[D]. 石河子:石河子大學(xué),2012.

Liu Yang. Design and Implement CAN-Bus of Automatic Navigation System on the Tractor[D]. Shihezi: Shihezi University, 2012. (in Chinese with English abstract)

[17] 何杰,朱金光,羅錫文,等. 電動(dòng)方向盤插秧機(jī)轉(zhuǎn)向控制系統(tǒng)設(shè)計(jì)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2019,35(6):10-17.

He Jie, Zhu Jinguang, Luo Xiwen, et al. Design of steering control system for rice transplanter equipped with steering wheel-like motor[J]. Transactions of The Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(6): 10-17. (in Chinese with English abstract)

[18] 闞輝玉,李軍偉,李德芳,等. 重型拖拉機(jī)CAN通信網(wǎng)絡(luò)設(shè)計(jì)[J]. 山東理工大學(xué)學(xué)報(bào)(自然科學(xué)版),2021,35(3):30-36.

Kan Huiyu, Li Junwei, Li Defang, et al. Design of CAN communication network for heavy duty tractor[J]. Journal of Shandong University of Technology (Natural Science Edition), 2021, 35(3): 30-36. (in Chinese with English abstract)

[19] 劉雪珂. 基于CAN總線的拖拉機(jī)自動(dòng)導(dǎo)航系統(tǒng)轉(zhuǎn)向控制策略研究[D]. 重慶:重慶理工大學(xué),2017.

Liu Xueke. Research on Auto-steering Control Strategies of Tractor Automatic Navigation System Based on CAN Bus[D]. Chongqing: Chongqing University of Technology, 2017. (in Chinese with English abstract)

[20] Wang H, Noguchi N. Autonomous maneuvers of a robotic tractor for farming[C]//Shiroishi: Proceedings of the 2016 IEEE/SICE International Symposium on System Integration (SII). IEEE, 2016: 592-597.

[21] 劉進(jìn)一. 基于速度自適應(yīng)的拖拉機(jī)自動(dòng)導(dǎo)航控制系統(tǒng)研究[D]. 北京:中國農(nóng)業(yè)大學(xué),2017.

Liu Jinyi. Research on Automatic Navigation Control System of Tractor Based on Speed Adaptation[D]. Beijing: China Agricultural University, 2017. (in Chinese with English abstract)

[22] 賈全. 拖拉機(jī)自動(dòng)導(dǎo)航系統(tǒng)自適應(yīng)控制方法研究[D]. 北京:中國農(nóng)業(yè)機(jī)械化科學(xué)研究院,2018.

Jia Quan. Adaptive Control Method for Tractor Auto-Guidance System[D]. Beijing: Chinese Academy of Agricultural Mechanization Sciences, 2018. (in Chinese with English abstract)

[23] Christiansen M, Laursen M, Jorgensen R, et al. Designing and testing a UAV mapping system for agricultural field surveying[J]. Sensors, 2017, 17(12): 2703.

[24] Liu Z, Lü Z, Zheng W, et al. Design of obstacle avoidance controller for agricultural tractor based on ROS[J]. International Journal of Agricultural and Biological Engineering, 2019, 12(6): 58-65.

[25] Iqbal J, Xu R, Halloran H, et al. Development of a multi-purpose autonomous differential drive mobile robot for plant phenotyping and soil sensing[J]. Electronics, 2020, 9(9): 1550.

[26] Barth R, Hemming J, Van Henten E. Design of an eye-in-hand sensing and servo control framework for harvesting robotics in dense vegetation[J]. Biosystems Engineering, 2016, 146: 71-84.

[27] 翟衛(wèi)欣,王東旭,陳智博,等. 無人駕駛農(nóng)機(jī)自主作業(yè)路徑規(guī)劃方法[J]. 農(nóng)業(yè)工程學(xué)報(bào),2021,375(16):1-7.

Zhai Weixin, Wang Dongxu, Chen Zhibo, et al. Autonomous operation path planning method for unmanned agricultural machinery[J]. Transactions of The Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 375(16): 1-7. (in Chinese with English abstract)

[28] 丁亞強(qiáng). 基于毫米波雷達(dá)的車輛縱向防碰撞系統(tǒng)研究[D]. 鎮(zhèn)江:江蘇大學(xué),2018.

Ding Yaqiang. Research on Vehicle Longitudinal Collision Avoidance System Based on Millimeter Wave Radar[D]. Zhenjiang: Jiangsu University, 2018. (in Chinese with English abstract)

[29] Wang D, Qi F. Trajectory planning for a four-wheel-steering vehicle[J]. Proceedings 2001 ICRA IEEE International Conference on Robotics and Automation (Cat No01CH37164), 2001, 4: 3320-3325.

[30] Rajamani R. Vehicle Dynamics and Control[M]. New York: Springer, 2012.

[31] 趙小輝,張明柱,白東洋,等. 柴油發(fā)動(dòng)機(jī)特性曲線分析與建模[J]. 中國農(nóng)機(jī)化學(xué)報(bào),2016,37(7):112-115.

Zhao Xiaohui, Zhang Mingzhu, Bai Dongyang, et al. Modeling and analysis on diesel engine characteristic curve[J]. Journal of Chinese Agricultural Mechanization, 2016, 37(7): 112-115. (in Chinese with English abstract)

Development of the unmanned driving system for agricultural machinery based on DF2204 continuously variable transmission tractor

Chen Zhibo, Wen Long, Yang Weizhong, Yang Lili, Zhao Xin, Wu Caicong※

(1.100083; 2.100083)

Autonomous agricultural vehicle (AAV) is believed to play a significant role in smart agriculture. In future periods, AAV will still be dominated by fuel engines, and the transmission technology will be upgraded from manual shifting to power shifting or CVT (Continuously Variable Transmission) simultaneously for which to improve the power and economy. In response to the demand of AAV, this study devotes to developing an autonomous driving and operation system based on ROS (Robot Operating System) and CVT tractor. The proposed system includes safety, planning, control and CAN bus communication modules. To do that, we integrated and deployed hardware on CVT tractor, designed a CAN bus protocol and implemented data structure for communication between ROS and CAN bus. A close-range anti-collision capability of the tractor is realized based on radar, and a lateral controller based on PID algorithm is adopted. To further validation, a corn sowing experiment was carried out in Miyun District, Beijing, with a total operating area of 2.4 hectares. This research takes DF2204 CVT tractor as the test platform according to the working requirements of autonomous tractors. We also designed the hardware platform of AAV, which was divided into computing layer, sensing layer and actuation layer. Based on the idea of modularization and hierarchy, a software architecture that meets the job requirements was developed with ROS as the middleware, which includes localization, planning, control, CAN communication, and safety modules. And a CAN bus protocol to meet the needs of vehicle control was developed. According to the control characteristics and operation requirements of the CVT tractor, a lateral control module and a longitudinal velocity control module were designed. We counted 22 straight working paths with a total working time of 5 943 s (excluding supplemental seeds and fertilizers), of which the total working time of the straight working stage was 4 037 s and the time of the U-turn stage was 1 906 s, approximately 32% of the total working time. The efficiency of our system was 1.33 hm2/h, and the experimental result shows that the communication node could meet the communication requirements of 50 Hz.The target speed was set to 3 km/h when the tractor turns around, while the harrow was stopped and the seeder was lifted.The engine load was lower when the tractor makes a U-turn, and the torque percentage fluctuates between 10%-30%. After several seconds the engine load continued to increase, and the torque output percentage remained above 80%, with a maximum value of 94%. The engine and transmission will perform a torque reserve in order to meet the instantaneous torque demand (such as increased resistance, climbing, etc.). When the engine revs is higher than the revs corresponding to the torque peak, the increased load will cause the engine to overload, thereby reducing the revs and ego velocity, so as to output more torque. The average lateral error is 2.96 cm and the navigation error was 11.69 cm. The velocity RMSE is 0.98 km/h and MAE was 0.68 km/h. The steering angle RMSE was 1.91°, and MAE was 1.47°. This research shows that the tractor based on wire control technology and CVT tractor can fit the needs of autonomous agricultural vehicles, and the control and planning modules can meet sowing operations. This research could provide a reference for the unmanned upgrading of CVT tractors, and improve the intelligent level and operation efficiency of agricultural machinery.

tractor; continuously variable transmission; autonomous driving; autonomous operation; wire control chassis

10.11975/j.issn.1002-6819.2022.19.001

S24

A

1002-6819(2022)-19-0001-09

陳智博,文龍,楊衛(wèi)中,等. 基于DF2204無級變速拖拉機(jī)的農(nóng)機(jī)無人駕駛系統(tǒng)研制[J]. 農(nóng)業(yè)工程學(xué)報(bào),2022,38(19):1-9.doi:10.11975/j.issn.1002-6819.2022.19.001 http://www.tcsae.org

Chen Zhibo, Wen Long, Yang Weizhong, et al. Development of the unmanned driving system for agricultural machinery based on DF2204 continuously variable transmission tractor[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2022, 38(19): 1-9. (in Chinese with English abstract) doi:10.11975/j.issn.1002-6819.2022.19.001 http://www.tcsae.org

2022-08-05

2022-09-27

北京市科技計(jì)劃項(xiàng)目(Z201100008020008)

陳智博,博士生,研究方向?yàn)檗r(nóng)機(jī)自動(dòng)駕駛與機(jī)群協(xié)同作業(yè)。Email: chenzb@cau.edu.cn

吳才聰,博士,教授,博士生導(dǎo)師,研究方向?yàn)檗r(nóng)機(jī)自動(dòng)駕駛與農(nóng)機(jī)大數(shù)據(jù)。Email:wucc@cau.edu.cn

猜你喜歡
線控拖拉機(jī)無人駕駛
我們村的無人駕駛公交
拖拉機(jī)內(nèi)燃電傳動(dòng)系統(tǒng)
飛上天的“拖拉機(jī)”
2023年實(shí)現(xiàn)量產(chǎn),長城汽車智慧線控底盤技術(shù)全揭秘
無人駕駛車輛
無人駕駛公園
紅色拖拉機(jī)
基于磁流變液的線控轉(zhuǎn)向力反饋裝置探究
基于線控轉(zhuǎn)向系統(tǒng)的無人駕駛技術(shù)發(fā)展
紅色拖拉機(jī)
宁国市| 桂平市| 章丘市| 梅州市| 大余县| 张家口市| 英吉沙县| 延津县| 静宁县| 商都县| 加查县| 湾仔区| 白城市| 临武县| 特克斯县| 长海县| 连云港市| 高州市| 呈贡县| 宜君县| 瓮安县| 琼海市| 清远市| 漳平市| 东山县| 临安市| 宣武区| 达尔| 合作市| 全州县| 淳安县| 肥乡县| 大方县| 巢湖市| 子洲县| 连江县| 阜康市| 齐齐哈尔市| 利川市| 青川县| 衡东县|