張良安,唐 鍇,趙永杰,王孝義,余大壯,盧新建,王 祥
·農(nóng)業(yè)裝備工程與機(jī)械化·
四足激光除草機(jī)器人腿部結(jié)構(gòu)參數(shù)優(yōu)化
張良安1,唐 鍇1,趙永杰2※,王孝義1,余大壯1,盧新建3,王 祥1
(1. 安徽工業(yè)大學(xué)機(jī)械工程學(xué)院,馬鞍山 243000;2. 汕頭大學(xué)機(jī)械電子工程系,汕頭 515063;3. 廣東省智行機(jī)器人科技有限公司,佛山 528226)
電驅(qū)四足激光除草機(jī)器人的續(xù)航能力取決于其電池容量和自身功耗,為了提高其續(xù)航能力,降低其目標(biāo)軌跡下的驅(qū)動(dòng)力矩將具有重要意義。該文以四足激光除草機(jī)器人為研究對(duì)象,提出一種動(dòng)力學(xué)尺度綜合方法,將其腿部尺寸參數(shù)作為優(yōu)化設(shè)計(jì)目的,在給定的目標(biāo)軌跡上,針對(duì)腿部關(guān)節(jié)驅(qū)動(dòng)力矩進(jìn)行優(yōu)化,最終得到一組最優(yōu)腿部桿長(zhǎng),使其完成目標(biāo)軌跡的驅(qū)動(dòng)力矩和功耗最小。以四足機(jī)器人腿部關(guān)節(jié)驅(qū)動(dòng)力矩最大值最小作為優(yōu)化目標(biāo),腿部尺寸參數(shù)作為約束變量,利用粒子群算法和理想點(diǎn)法進(jìn)行二次優(yōu)化,將多目標(biāo)優(yōu)化問題轉(zhuǎn)化為單目標(biāo)優(yōu)化問題,得到一組最優(yōu)桿長(zhǎng)。計(jì)算結(jié)果表明:經(jīng)過尺度綜合后的四足機(jī)器人動(dòng)力學(xué)性能得到明顯改善,優(yōu)化后單腿的大腿關(guān)節(jié)驅(qū)動(dòng)力矩峰值下降5.29%,小腿關(guān)節(jié)驅(qū)動(dòng)力矩峰值下降18.05%,驗(yàn)證了該尺度綜合方法的有效性。該文提出的動(dòng)力學(xué)尺度綜合方法可為四足類機(jī)器人的設(shè)計(jì)提供參考依據(jù)。
機(jī)器人;優(yōu)化;尺度綜合;粒子群算法
激光除草不使用農(nóng)藥,對(duì)保護(hù)生態(tài)環(huán)境具有重要意義。但激光器的運(yùn)載平臺(tái)不成熟是目前限制其推廣使用的重要因素。國(guó)內(nèi)外學(xué)者對(duì)怎樣將機(jī)器人作為激光除草器的移動(dòng)搭載平臺(tái)進(jìn)行了大量的研究[1-3],而相對(duì)于目前的輪式和履帶式機(jī)器人,四足機(jī)器人由于具有離散的落足點(diǎn),能夠靈活調(diào)整行走位姿和步態(tài),因此更適合在其上搭載激光除草機(jī)構(gòu)在農(nóng)田等非結(jié)構(gòu)環(huán)境中進(jìn)行作業(yè)。
四足機(jī)器人的能源和動(dòng)力不足問題一直是限制其進(jìn)行應(yīng)用發(fā)展的一大瓶頸。因此,研究四足機(jī)器人的移動(dòng)能量消耗,分析影響其驅(qū)動(dòng)力矩和功耗的因素,對(duì)于降低其功耗、拓寬其應(yīng)用領(lǐng)域具有重要的意義。以往研究已經(jīng)證實(shí)合理的結(jié)構(gòu)參數(shù)和步態(tài)參數(shù)能夠提高機(jī)器人的能量效率[4-6]。例如,馬宗利等[7]針對(duì)四足機(jī)器人能耗進(jìn)行分析,提出了一種大腿和小腿呈一體化的柔性節(jié)能結(jié)構(gòu),并對(duì)模型進(jìn)行了動(dòng)力學(xué)分析和ADAMS仿真,驗(yàn)證了該模型能夠使得關(guān)節(jié)驅(qū)動(dòng)力矩和功耗明顯降低;雷靜桃等[8]對(duì)四足機(jī)器人的足端軌跡進(jìn)行了研究,分析了不同足端軌跡及步高、步距、關(guān)節(jié)起始角等步態(tài)參數(shù)對(duì)移動(dòng)能耗的影響。李軍等[9]針對(duì)不同步頻與步幅的耦合狀態(tài),研究了步頻與步幅的獨(dú)立變化對(duì)四足機(jī)器人關(guān)節(jié)扭矩與功耗的影響,通過仿真驗(yàn)證得出能耗對(duì)步頻參數(shù)的提升更為敏感。現(xiàn)有的四足機(jī)器人動(dòng)力學(xué)和能耗研究中,較少涉及考慮整個(gè)機(jī)身質(zhì)量分布在給定軌跡規(guī)劃下,腿部結(jié)構(gòu)參數(shù)對(duì)四足機(jī)器人動(dòng)力學(xué)性能和功耗的影響。對(duì)四足機(jī)器人而言,合適的腿部參數(shù)可明顯降低四足機(jī)器人的驅(qū)動(dòng)力矩和功耗,提高機(jī)器人的動(dòng)力學(xué)性能,拓展應(yīng)用范圍,因此,對(duì)四足機(jī)器人進(jìn)行動(dòng)力學(xué)尺度綜合尤為重要[10-12]。
綜上所述,為降低四足激光除草機(jī)器人的驅(qū)動(dòng)力矩,本文提出一種基于目標(biāo)軌跡下四足機(jī)器人動(dòng)力學(xué)尺度綜合優(yōu)化設(shè)計(jì)方法。以Swil四足機(jī)器人為研究對(duì)象,建立其整機(jī)運(yùn)動(dòng)學(xué)和剛體動(dòng)力學(xué)模型,并基于零沖擊原則確定機(jī)器人足端運(yùn)動(dòng)軌跡。在目標(biāo)軌跡下以腿部關(guān)節(jié)驅(qū)動(dòng)力矩最大值最小化作為動(dòng)力學(xué)性能優(yōu)化目標(biāo),利用粒子群算法先對(duì)多目標(biāo)函數(shù)進(jìn)行優(yōu)化,得到各關(guān)節(jié)驅(qū)動(dòng)力矩峰值的最小值。最后根據(jù)理想點(diǎn)法將多目標(biāo)問題轉(zhuǎn)化為單目標(biāo)問題進(jìn)行二次優(yōu)化,得到一組腿部尺寸參數(shù),使得四足機(jī)器人的驅(qū)動(dòng)力矩和功耗最小。
所設(shè)計(jì)的Swil四足機(jī)器人主要由軀身和四條對(duì)稱布置的腿部機(jī)構(gòu)組成。具體結(jié)構(gòu)如圖1所示。
1.足端 2.曲柄連桿機(jī)構(gòu) 3.大腿 4.髖關(guān)節(jié)減速器 5.側(cè)擺關(guān)節(jié)電機(jī) 6.蓄電池 7.膝關(guān)節(jié)電機(jī) 8.側(cè)擺關(guān)節(jié)減速器 9.髖關(guān)節(jié)電機(jī) 10.電動(dòng)缸 11.小腿
在四足機(jī)器人結(jié)構(gòu)設(shè)計(jì)上,Swil四足機(jī)器人的四條腿的大小尺寸相同,采用前肘后膝式的拓?fù)浣Y(jié)構(gòu),對(duì)稱布置在機(jī)身的四周,這有助于增強(qiáng)四足機(jī)器人的穩(wěn)定性。四足機(jī)器人在直線運(yùn)動(dòng)時(shí),髖關(guān)節(jié)處的電機(jī)驅(qū)動(dòng)大腿使其繞著髖關(guān)節(jié)進(jìn)行轉(zhuǎn)動(dòng),實(shí)現(xiàn)其俯仰運(yùn)動(dòng)。小腿繞膝關(guān)節(jié)的轉(zhuǎn)動(dòng)是依靠將電動(dòng)缸的直線運(yùn)動(dòng)通過曲柄滑塊結(jié)構(gòu)轉(zhuǎn)化為小腿的轉(zhuǎn)動(dòng),實(shí)現(xiàn)其腿部的抬落。通過大腿與小腿間的相互協(xié)調(diào)以及各腿部之間的連續(xù)步態(tài),來實(shí)現(xiàn)四足機(jī)器人的運(yùn)動(dòng)。
四足機(jī)器人單腿D-H模型如圖2所示,運(yùn)用D-H法建立坐標(biāo)系對(duì)四足機(jī)器人腿部結(jié)構(gòu)關(guān)系進(jìn)行描述。
注:b為機(jī)身半長(zhǎng),m;w為機(jī)身半寬,m;h為機(jī)身高度,m;zb為機(jī)身z軸坐標(biāo);yb為機(jī)身y軸坐標(biāo);xb為機(jī)身x軸坐標(biāo);Ob為機(jī)身坐標(biāo)軸原點(diǎn);x0為髖關(guān)節(jié)x軸坐標(biāo);y0為髖關(guān)節(jié)軸坐標(biāo);z0為髖關(guān)節(jié)軸坐標(biāo);θ1為側(cè)擺關(guān)節(jié)轉(zhuǎn)角,(o);L1為髖部桿長(zhǎng),m;x1為髖部軸坐標(biāo);z1為髖部軸坐標(biāo);θ2為髖關(guān)節(jié)轉(zhuǎn)角,(o);L2為大腿桿長(zhǎng),m;x2為膝關(guān)節(jié)軸坐標(biāo);z2為膝關(guān)節(jié)軸坐標(biāo);θ3為膝關(guān)節(jié)轉(zhuǎn)角,(o);L3為小腿桿長(zhǎng),m;x3為足端軸坐標(biāo);z3為足端軸坐標(biāo)。
根據(jù)四足機(jī)器人單腿結(jié)構(gòu)簡(jiǎn)圖,建立各關(guān)節(jié)處的空間坐標(biāo)系,并通過各坐標(biāo)系間的關(guān)系運(yùn)用D-H法得到D-H連桿參數(shù),如表1所示。
表1 D-H連桿參數(shù)表
注:表示關(guān)節(jié)轉(zhuǎn)角,(°);d表示關(guān)節(jié)距離,mm;a表示桿件長(zhǎng)度,mm;表示桿件扭角,(°);1表示髖部連桿,2表示大腿連桿,3表示小腿連桿。
Note:is joint angle, (°);dis linkage distance, mm ;ais linkage length, mm;is linkage twist, (°); 1 is the connecting rod of hip; 2 is the connecting rod of thigh; 3 is the connecting rod of shank.
四足機(jī)器人足端相對(duì)于側(cè)擺關(guān)節(jié)在空間中的位置=(P,P,P)可由齊次變換矩陣表示為
式中03表示足端坐標(biāo)系相對(duì)于側(cè)擺關(guān)節(jié)坐標(biāo)系的位姿,01(稱為1)表示髖關(guān)節(jié)坐標(biāo)系相對(duì)于側(cè)擺關(guān)節(jié)坐標(biāo)系的位姿,12(稱為2)表示膝關(guān)節(jié)坐標(biāo)系相對(duì)于髖關(guān)節(jié)坐標(biāo)系的位姿,23(稱為3)表示足端關(guān)節(jié)坐標(biāo)系相對(duì)于膝關(guān)節(jié)坐標(biāo)系的位姿,P表示足端在側(cè)擺關(guān)節(jié)坐標(biāo)系下的軸坐標(biāo),P表示足端在側(cè)擺關(guān)節(jié)坐標(biāo)系下的軸坐標(biāo), P表示足端在側(cè)擺關(guān)節(jié)坐標(biāo)系下的軸坐標(biāo),=(n,n,n)表示足端軸的單位方向矢量,=(o,o,o)表示足端軸的單位方向矢量,=(a,a,a)表示足端軸的單位方向矢量,1=sin(1),1=cos(1),2=sin(2),2=cos(2),3=sin(3),3=cos(3),23=sin(2+3),23=cos(2+3)
根據(jù)四足機(jī)器人D-H連桿參數(shù)表,可推導(dǎo)出其運(yùn)動(dòng)學(xué)正解為
為了對(duì)四足機(jī)器人進(jìn)行軌跡規(guī)劃和控制,我們需要對(duì)其進(jìn)行逆運(yùn)動(dòng)學(xué)分析。依據(jù)公式(1)、(2),可分別求出各關(guān)節(jié)轉(zhuǎn)動(dòng)的角度,具體如下
其中,1=sin(1),1=cos(1),2=sin(2),2=cos(2),3=sin(3),3=sin(1),23=sin(1+3),23=cos(2+3)
本文是基于給定的目標(biāo)軌跡下對(duì)四足機(jī)器人動(dòng)力學(xué)性能進(jìn)行研究,因此規(guī)劃一條合理的能夠滿足任務(wù)需求的足端軌跡顯得尤為重要[13-14]。據(jù)一些以往的研究發(fā)現(xiàn)在對(duì)四足機(jī)器人進(jìn)行軌跡規(guī)劃時(shí),復(fù)合擺線軌跡能夠減小足端與地面的沖擊,且在運(yùn)行穩(wěn)定性方面也具有良好的效果[15-16]??紤]四足機(jī)器人關(guān)節(jié)速度、加速度連續(xù)性要求,結(jié)合預(yù)設(shè)的步態(tài)參數(shù)建立四足機(jī)器人足端約束方程,并推導(dǎo)出滿足要求的復(fù)合擺線軌跡方程,再將得到的復(fù)合擺線軌跡與運(yùn)動(dòng)學(xué)方程相結(jié)合,來驗(yàn)證其軌跡規(guī)劃的有效性。
本文根據(jù)四足機(jī)器人完成任務(wù)要求,對(duì)其步態(tài)參數(shù)進(jìn)行設(shè)置:步長(zhǎng)=0.48 m,步高=0.08 m,單腿步態(tài)周期=2 s,單腿占空比=1/2,四足機(jī)器人擺動(dòng)相與支撐相在0=1 s時(shí)進(jìn)行切換。則在0~0,足端處于擺動(dòng)相;在0~,足端處于支撐相[17-18]。設(shè)水平方向?yàn)榉较?,豎直方向?yàn)榉较颍瑸閱瓮冗\(yùn)動(dòng)時(shí)間,根據(jù)速度、加速度連續(xù)性要求建立足端軌跡方程:
方向上約束方程為
方向上約束方程為
支撐相函數(shù)為
則整個(gè)復(fù)合擺線的運(yùn)動(dòng)曲線如圖3所示。
為了驗(yàn)證所推導(dǎo)的復(fù)合擺線軌跡的正確性,首先建立基于復(fù)合擺線軌跡下的運(yùn)動(dòng)學(xué)模型簡(jiǎn)圖,如圖4所示。
注:m1為髖部桿質(zhì)量,kg,m2為大腿質(zhì)量,kg;m3為小腿質(zhì)量,kg;H1為側(cè)擺關(guān)節(jié)到地面的高度,m;A點(diǎn)為足端在復(fù)合擺線軌跡上的任意一點(diǎn);h1為A點(diǎn)到地面的高度,m;B點(diǎn)為足端起始點(diǎn),且在側(cè)擺關(guān)節(jié)質(zhì)心O點(diǎn)正下方。
以髖關(guān)節(jié)質(zhì)心為原點(diǎn),水平前進(jìn)方向?yàn)檩S,豎直方向?yàn)檩S,建立坐標(biāo)系。點(diǎn)在髖關(guān)節(jié)質(zhì)心的正下方;點(diǎn)為足端在復(fù)合擺線軌跡上的任意一點(diǎn)。則可推導(dǎo)出基于復(fù)合擺線下的四足機(jī)器人單腿運(yùn)動(dòng)學(xué)數(shù)學(xué)表達(dá)式。
根據(jù)已知桿長(zhǎng)(1,2,3)和式(8)以及表1,可推導(dǎo)出四足機(jī)器人逆運(yùn)動(dòng)學(xué)數(shù)學(xué)表達(dá)式為
為了對(duì)復(fù)合擺線軌跡進(jìn)行驗(yàn)證,根據(jù)實(shí)驗(yàn)室原有的四足機(jī)器人結(jié)構(gòu)尺寸,令1=0.64 m,1=0,2=0.42 m,3=0.36 m,并將3.1節(jié)中各步態(tài)參數(shù)代入式(9)中得到腿部關(guān)節(jié)的轉(zhuǎn)角函數(shù),并對(duì)其進(jìn)行一階和二階求導(dǎo)分別得到角速度和角加速度函數(shù)。并在MATLAB中繪制出基于給定目標(biāo)軌跡下的各關(guān)節(jié)運(yùn)動(dòng)曲線,如圖5所示。
圖5 各關(guān)節(jié)運(yùn)動(dòng)曲線
由圖5可知,四足機(jī)器人在目標(biāo)軌跡上速度、加速度連續(xù),且在擺動(dòng)相與支撐相切換的瞬間及0=1 s時(shí)足端加速度=0,減小了足端與地面的沖擊,驗(yàn)證了復(fù)合擺線軌跡的可行性。
動(dòng)力學(xué)模型可描述四足機(jī)器人的動(dòng)態(tài)性能,反映各關(guān)節(jié)驅(qū)動(dòng)力矩與運(yùn)動(dòng)參數(shù)間的關(guān)系,是四足機(jī)器人驅(qū)動(dòng)力矩和移動(dòng)功耗分析的基礎(chǔ)。由于四足機(jī)器人在trot步態(tài)下,位于對(duì)角線上的兩條腿同時(shí)處于支撐相或擺動(dòng)相,所以四足機(jī)器人在運(yùn)動(dòng)過程中是變結(jié)構(gòu)的,擺動(dòng)相和支撐相時(shí)的運(yùn)動(dòng)學(xué)模型不同[19]。本文優(yōu)化的目標(biāo)是使四足機(jī)器人在目標(biāo)軌跡上各關(guān)節(jié)驅(qū)動(dòng)力矩峰值下降,而相對(duì)于擺動(dòng)相時(shí),支撐相的各關(guān)節(jié)驅(qū)動(dòng)力矩要大的多,因此本文只對(duì)支撐相時(shí)的各關(guān)節(jié)驅(qū)動(dòng)力矩進(jìn)行分析。
當(dāng)僅考慮到四足激光除草機(jī)器人的機(jī)身質(zhì)量,而忽略其腿部質(zhì)量時(shí),假定四足機(jī)器人的所有質(zhì)量都集中在機(jī)身的中心。由于四足激光除草機(jī)器人是一個(gè)復(fù)雜的多剛體系統(tǒng),其足端與地面間的接觸很難用模型來描述,因此本文將四足機(jī)器人簡(jiǎn)化為一個(gè)做平面運(yùn)動(dòng)的五桿機(jī)構(gòu),且在行走過程進(jìn)行如下假設(shè)[20]
1)足端與地面間接觸良好,且不存在相對(duì)滑動(dòng),則四足機(jī)器人與地面之間可以看作是鉸鏈連接;
2)四足激光除草機(jī)器人以trot步態(tài)勻速運(yùn)動(dòng)時(shí),地面對(duì)足端的支撐反力豎直向上,且只在豎直方向上有分量;
3)相對(duì)于四足機(jī)器人的機(jī)身質(zhì)量,其腿部質(zhì)量忽略不計(jì),且假定所有質(zhì)量都集中在機(jī)身的中心。
基于以上假說,對(duì)四足機(jī)器人支撐相時(shí)的模型進(jìn)行簡(jiǎn)化,其中1、3為當(dāng)前的支撐腿,如圖6所示。
式中為步幅長(zhǎng)度,m;為步行周期,s;?為以當(dāng)前支撐相為起點(diǎn)的步行時(shí)間,s。當(dāng)1、3腿為支撐相時(shí),取簡(jiǎn)化為五桿機(jī)構(gòu)的四足機(jī)器人整體為研究對(duì)象,可得到力與力矩的平衡方程
注:G為機(jī)身質(zhì)量,kg;Zb為四足機(jī)器人前進(jìn)方向;M12, M32分別為前后足端對(duì)兩大腿關(guān)節(jié)的轉(zhuǎn)矩,N·m;N12, N32分別為大腿關(guān)節(jié)對(duì)機(jī)身的反作用力,N·m;M13, M33分別為前后足端對(duì)兩小腿關(guān)節(jié)的轉(zhuǎn)矩,N·m;N13, N33分別為地面對(duì)兩足端的反作用力,N·m。
對(duì)式(10)、(11)進(jìn)行連列求解可得到反作用力13和33的表達(dá)式為
現(xiàn)以四足機(jī)器人單腿為研究對(duì)象,在四足機(jī)器人直線運(yùn)動(dòng)時(shí)髖關(guān)節(jié)驅(qū)動(dòng)力矩1=0。為了平衡地面對(duì)右前腿足端的支撐反力,須給大腿和小腿關(guān)節(jié)施加驅(qū)力矩12(N·m)和13(N·m),保持力的作用效果不變。則四足機(jī)器人單腿支撐相時(shí)的各關(guān)節(jié)驅(qū)動(dòng)力矩?cái)?shù)學(xué)表達(dá)式如下。
對(duì)于有確定任務(wù)需求和目標(biāo)軌跡的四足機(jī)器人而言,考慮到關(guān)節(jié)驅(qū)動(dòng)力矩峰值過大,會(huì)增加能量損耗,產(chǎn)生較大的力矩波動(dòng),降低了機(jī)器人運(yùn)行時(shí)的平穩(wěn)性,且減小電機(jī)的疲勞壽命[21-22]。因此設(shè)法降低四足機(jī)器人在目標(biāo)軌跡上的驅(qū)動(dòng)力矩峰值對(duì)于提高四足機(jī)器人動(dòng)力學(xué)性能具有重要意義。
本文研究的目的是使四足機(jī)器人在目標(biāo)軌跡上的驅(qū)動(dòng)力矩峰值T(=1,2,3)下降,根據(jù)動(dòng)力學(xué)公式(13)可知,在給定目標(biāo)軌跡和步態(tài)參數(shù)下,影響各關(guān)節(jié)驅(qū)動(dòng)力矩的因素為2、3,因此其設(shè)計(jì)變量為[2,3]。
本文屬于多目標(biāo)優(yōu)化問題,通過改變腿部的尺寸參數(shù)可以得到四足機(jī)器人在整個(gè)運(yùn)動(dòng)周期上各關(guān)節(jié)的驅(qū)動(dòng)力矩峰值。由于大腿和小腿關(guān)節(jié)受力不同,所以在實(shí)際過程中使四足機(jī)器人大腿和小腿關(guān)節(jié)同時(shí)滿足驅(qū)動(dòng)力矩峰值最大值最小是不可現(xiàn)實(shí)的。因此需要通過理想點(diǎn)法將多目標(biāo)問題轉(zhuǎn)化為單目標(biāo)問題加以優(yōu)化。取基于目標(biāo)軌跡優(yōu)化后的大腿與小腿驅(qū)動(dòng)力矩最大值Max(2)和Max(3)中的最小值min(Max2)和min(Max3)作為理想點(diǎn),利用最短距離理想點(diǎn)法將原多目標(biāo)規(guī)劃問題轉(zhuǎn)化成如下單目標(biāo)優(yōu)化問題[23-25],得到最終的目標(biāo)函數(shù)為
由于四足機(jī)器人是基于給定的目標(biāo)軌跡進(jìn)行運(yùn)動(dòng),考慮到腿部尺寸過長(zhǎng)和過短都會(huì)對(duì)四足機(jī)器人的穩(wěn)定性、靈活性以及工作空間范圍產(chǎn)生影響[26-28],故需要先對(duì)腿部尺寸施加合理的范圍約束。本文前期通過枚舉法選取大量不同組合的桿長(zhǎng)對(duì)結(jié)果進(jìn)行初步對(duì)比驗(yàn)證,試驗(yàn)數(shù)據(jù)表明,選取2,3=[0.2,0.5](m)范圍內(nèi),會(huì)使得最終的輸出力矩相對(duì)合理有效。
由于本文使用D-H法進(jìn)行運(yùn)動(dòng)學(xué)建模,所以在粒子群算法搜索的過程中會(huì)因不同桿長(zhǎng)而出現(xiàn)奇異點(diǎn),導(dǎo)致粒子群算法不會(huì)收斂到最優(yōu)解。為保證數(shù)據(jù)的合理性、正確性,對(duì)目標(biāo)函數(shù)施加以下約束條件:
式中min(Max2)、min(Max3)和min()均屬于實(shí)數(shù)域,從而能夠搜索到最優(yōu)解。
本文分2次進(jìn)行優(yōu)化,第一次初步優(yōu)化是將腿部桿長(zhǎng)L(=1,2,3)作為其設(shè)計(jì)變量,腿部各關(guān)節(jié)驅(qū)動(dòng)力矩T(=1,2,3)作為適應(yīng)度函數(shù),并添加一定的約束條件,通過粒子群算法[29-30]進(jìn)行優(yōu)化,其參數(shù)設(shè)置如下:粒子數(shù)目=40,學(xué)習(xí)因子為1=2=2,最大權(quán)重max=0.9,最小權(quán)重min=0.4,迭代步數(shù)取1 000次,其適應(yīng)度函數(shù)為=min(MaxT)(=1,2,3),分別得到腿部驅(qū)動(dòng)力矩T(=1,2,3)最大值最小的值,并以此作為理想點(diǎn),各理想點(diǎn)分別對(duì)應(yīng)2組桿長(zhǎng)(2,3)和(2,3)。
第二次優(yōu)化是將第一次優(yōu)化得到的理想點(diǎn)運(yùn)用理想點(diǎn)法將多目標(biāo)函數(shù)轉(zhuǎn)化為單目標(biāo)函數(shù)[31]:
通過將第一次優(yōu)化得到的2組桿長(zhǎng)(2,2)和(3,3)作為粒子群算法搜索的約束條件,將構(gòu)造的評(píng)價(jià)函數(shù)作為算法的適應(yīng)度函數(shù),再次運(yùn)用粒子群算法,其參數(shù)設(shè)置同上,最終得到使得四足激光除草機(jī)器人腿部各關(guān)節(jié)驅(qū)動(dòng)力矩最大值最小的一組桿長(zhǎng)。
通過第一次運(yùn)用粒子群算法得到支撐相時(shí)大腿關(guān)節(jié)驅(qū)動(dòng)力矩最大值最小的理想點(diǎn)為min(Max(2))= 23.253 1 N·m,對(duì)應(yīng)桿長(zhǎng)為2=0.339 2 m,3=0.340 3m;小腿關(guān)節(jié)驅(qū)動(dòng)力矩最大值最小的理想點(diǎn)為min(Max(3))= 75.708 6 N·m,對(duì)應(yīng)桿長(zhǎng)為2=0.340 1 m,3=0.339 6 m,其基于粒子群算法適應(yīng)度與迭代次數(shù)的關(guān)系如圖7所示。
圖7 各適應(yīng)度函數(shù)與迭代次數(shù)關(guān)系圖
將第一次優(yōu)化得到的理想點(diǎn)帶入評(píng)價(jià)函數(shù)min()中,再次運(yùn)用粒子群算法在新的搜索區(qū)間中得到最優(yōu)解min()=9.165 6,對(duì)應(yīng)的最優(yōu)桿長(zhǎng)為2=0.340 1 m,3=0.340 0 m。其目標(biāo)函數(shù)適應(yīng)度值與迭代次數(shù)的關(guān)系如圖7c所示。
四足機(jī)器人優(yōu)化前后的參數(shù)如表2所示。當(dāng)選取最優(yōu)桿長(zhǎng)(2,3)=(0.340 1, 0.340 0)時(shí),根據(jù)式(7)、(9)、(13)通過MATLAB進(jìn)行仿真計(jì)算,得到在整個(gè)支撐相上腿部各關(guān)節(jié)驅(qū)動(dòng)力矩優(yōu)化前后變化曲線,如圖8所示。由圖8可知大腿關(guān)節(jié)驅(qū)動(dòng)力矩峰值由優(yōu)化前的24.552 6 N·m降至23.253 1 N·m,降幅為5.29%;小腿關(guān)節(jié)驅(qū)動(dòng)力矩峰值由優(yōu)化前的92.385 7 N·m降至75.708 6 N·m,降幅為18.05%。
表2 四足機(jī)器人優(yōu)化前后結(jié)構(gòu)參數(shù)
圖8 優(yōu)化前后腿部各關(guān)節(jié)驅(qū)動(dòng)力矩對(duì)比
在不考慮能量損失等其他因素時(shí),單腿支撐相在一個(gè)周期內(nèi)所消耗的能量可表示為[31]:
式中P1為大腿關(guān)節(jié)驅(qū)動(dòng)功率,N·m/s;P2為小腿關(guān)節(jié)驅(qū)動(dòng)功率,N·m/s;為大腿關(guān)節(jié)角速度,rad/s;為小腿關(guān)節(jié)角速度,rad/s。再根據(jù)式(7)、(9)、(17)得到整個(gè)支撐相上腿部關(guān)節(jié)驅(qū)動(dòng)力矩優(yōu)化前后的功率對(duì)比,如圖9a和圖9b所示。由圖10可知,優(yōu)化后四足機(jī)器人的驅(qū)動(dòng)功率下降,則電機(jī)功耗也隨之下降。不考慮其他影響,優(yōu)化后四足激光除草機(jī)器人的動(dòng)力學(xué)性能和功耗得到明顯改善。
本文以得到的腿部最優(yōu)桿長(zhǎng)尺寸對(duì)其進(jìn)行四足激光除草機(jī)器人的結(jié)構(gòu)設(shè)計(jì),并制造實(shí)物樣機(jī)對(duì)結(jié)果進(jìn)行驗(yàn)證。圖10為設(shè)計(jì)的實(shí)物樣機(jī)步態(tài)連續(xù)行走過程,其中圖 10a為四足機(jī)器人初始位姿,此時(shí)左前腿和右后腿組成對(duì)角線,右前腿和左后腿組成另一組對(duì)角線;圖10b為左后腿和右前腿作為支撐相,左前腿和右后腿作為擺動(dòng)相同時(shí)向前邁進(jìn);圖10c為左前腿和右后腿作為支撐相,右前腿和左后腿作為擺動(dòng)相同時(shí)向前邁進(jìn),落地后回到初始位姿,此時(shí)四足機(jī)器人移動(dòng)步長(zhǎng);圖10d為四足機(jī)器人第二個(gè)周期時(shí)的步態(tài),兩組對(duì)角腿循環(huán)向前邁進(jìn)實(shí)現(xiàn)了四足機(jī)器人對(duì)角小跑步態(tài)向前運(yùn)動(dòng)。實(shí)驗(yàn)測(cè)試結(jié)果如表3所示,則驗(yàn)證結(jié)果表明與優(yōu)化前的實(shí)物樣機(jī)相比,其運(yùn)動(dòng)性能得到明顯的改善。
圖10 實(shí)物樣機(jī)trot步態(tài)下的行走過程
表3 四足機(jī)器人參數(shù)優(yōu)化前后性能對(duì)比
1)本文以四足激光除草機(jī)器人腿部支撐相時(shí),關(guān)節(jié)驅(qū)動(dòng)力矩最大值最小化作為優(yōu)化目標(biāo),以腿部尺寸參數(shù)作為其設(shè)計(jì)變量,提出了一種基于目標(biāo)軌跡下的四足激光除草機(jī)器人動(dòng)力學(xué)尺度綜合方法。
2)在滿足任務(wù)要求的情況下,規(guī)劃了一條零沖擊、且速度和加速度連續(xù)的復(fù)合擺線軌跡,并分析了在機(jī)身質(zhì)量作用下的四足激光除草機(jī)器人在支撐相時(shí)的輸出力矩變化情況。
3)運(yùn)用理想點(diǎn)法將四足激光除草機(jī)器人腿部結(jié)構(gòu)參數(shù)優(yōu)化的尺度變量轉(zhuǎn)化為粒子群算法的維度決策變量,得到了一組最優(yōu)桿長(zhǎng)(2,3)=(0.340 1, 0.340 0),并對(duì)基于這組最優(yōu)桿長(zhǎng)下的四足機(jī)器人驅(qū)動(dòng)力矩進(jìn)行分析和驗(yàn)證。實(shí)驗(yàn)結(jié)果表明單腿的大腿關(guān)節(jié)驅(qū)動(dòng)力矩峰值下降5.29%,小腿關(guān)節(jié)驅(qū)動(dòng)力矩峰值下降18.05%,大幅降低了四足激光除草機(jī)器人的各關(guān)節(jié)驅(qū)動(dòng)力矩峰值和能耗,驗(yàn)證了本文給定目標(biāo)軌跡下的四足激光除草機(jī)器人動(dòng)力學(xué)尺度綜合方法的正確性和優(yōu)越性,可為四足類的機(jī)器人動(dòng)力學(xué)尺度綜合提供參考。
[1]周福君,王文明,李小利,等. 搖桿式擺動(dòng)型玉米株間除草裝置設(shè)計(jì)與實(shí)驗(yàn)[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2018,49(1):77-85.
Zhou Fuzun, Wang Wenming, Li Xiaoli, et al. Design and experiment of cam rocker swing intra-row weeding device for Maize[J]. Transactions of The Chinese Society for Agricultural Machinery, 2018, 49(1): 77-85. (in Chinese with English abstract)
[2]賈洪雷,李森森,王剛,等. 中耕期玉米田間避苗除草裝置設(shè)計(jì)與實(shí)驗(yàn)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2018,34(7):15-22.
Jia Honglei, Li Sensen, Wang Gang, et al. Design and experiment of seedling avoidable weeding control device for intertillage maize[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(7): 15-22. (in Chinese with English abstract)
[3]徐麗明,于暢暢,劉文,等. 籬架式栽培葡萄株間除草機(jī)自動(dòng)避障機(jī)構(gòu)優(yōu)化設(shè)計(jì)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2018,34(7):23-30. Xu Liming, Yu Changchang, Liu Wen, et al. Optimal design on auto obstacle avoidance mechanism of intra-row weeder for trellis cultivated grape[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(7): 23-30. (in Chinese with English abstract)
[4]柏龍,龍樟,陳曉紅,等. 連續(xù)電驅(qū)動(dòng)四足機(jī)器人腿部機(jī)構(gòu)設(shè)計(jì)與分析[J]. 機(jī)器人,2018,40(2):136-144.
Bai Long, Long Zhang, Chen Xiaohong, et al. Design and analysis of a leg mechanism for a continuous electrically-driven quadruped robot[J]. Robot, 2018, 40(2): 136-144. (in Chinese with English abstract)
[5]高峰,雷靜桃,徐國(guó)艷. 四足步行機(jī)的對(duì)角小跑步態(tài)及能耗仿真分析[J]. 北京航空航天大學(xué)學(xué)報(bào),2007,33(6):719-722.
Gao Feng, Lei Jingtao, Xu Guoyan. Trot gait and energy consumption simulation of a quadruped robot[J]. Journal of Beijing University of Aeronautics, 2007, 33(6): 719-722. (in Chinese with English abstract)
[6]鄭楚婷,宋光明,喬貴方,等. 具有主動(dòng)腰關(guān)節(jié)的四足機(jī)器人在間歇對(duì)角步態(tài)下的姿態(tài)平衡控制[J]. 機(jī)器人,2007,38(6):670-677.
Zheng Chuting, Song Guangming, Qiao Guifang, et al. Posture balance control of the quadruped robot with an waist joint during intermittent trot locomotion[J]. Robot, 2007, 38(6): 670-677. (in Chinese with English abstract)
[7]馬宗利,朱彥防,劉永超,等. 四足機(jī)器人新型節(jié)能腿的設(shè)計(jì)與分析[J]. 東北大學(xué)學(xué)報(bào),2016,37(4):543-547.
Ma Zongli, Zhu Yanfang, Liu Yongchao, et al. Design and analysis of new energy-efficient legs for quadruped robots[J]. Journal of Northeastern University, 2016, 37(4): 543-547. (in Chinese with English abstract)
[8]雷靜桃,王峰,俞煌穎. 四足機(jī)器人軌跡規(guī)劃及移動(dòng)能耗分析[J]. 機(jī)械設(shè)計(jì)與研究,2014,30(1):29-34.
Lei Jingtao, Wang Feng, Yu Huangying. Analysis on trajectory planning and energy consumption of quadruped robot[J]. Machine Design & Research, 2014, 30(1): 29-34. (in Chinese with English abstract)
[9]李軍,苗新聰,張曉宇. 四足機(jī)器人步幅、步頻與扭矩和能耗關(guān)系研究[J]. 組合機(jī)床與自動(dòng)化加工技術(shù),2018,12(12):10-14.
Li Jun, Miao Xincong, Zhang Xiaoyu. The research on effect of stride frequency and stride length on joint torque and energy cost in trotting of quadruped robot[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2018, 12(12): 10-14. (in Chinese with English abstract)
[10]鄭坤明,張秋菊. 一種高速輕型化并聯(lián)機(jī)器人尺度綜合的研究[J]. 機(jī)械設(shè)計(jì),2017,34(4):29-38.
Zheng Kunming, Zhang Qiuju. Research on dimensional synthesis of high speed light-duty parallel robot[J]. Journal of Machine Design, 2017, 34(4): 29-38. (in Chinese with English abstract)
[11]鹿玲,張東勝,許允斗,等. 五自由度混聯(lián)機(jī)器人尺度與結(jié)構(gòu)優(yōu)化設(shè)計(jì)[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2018,49(4):412-419.
Lu Ling, Zhang Dongsheng, Xu Yundou, et al. Dimension and structure optimization design of 5-DOF hybird manipulator[J]. Transactions of The Chinese Society of Agricultural Machinery, 2018, 49(4): 412-419. (in Chinese with English abstract)
[12]王杰,管聲啟,夏齊霄. 手指康復(fù)外骨骼機(jī)器人的結(jié)構(gòu)設(shè)計(jì)優(yōu)化[J]. 中國(guó)機(jī)械工程,2018,29(2):224-228.
Wang Jie, Guan Shengqi, Xia Qixiao. Structural design of finger rehabilitation exoskeleton robots[J]. China Mechanical Engineering, 2018, 29(2): 224-228. (in Chinese with English abstract)
[13]雷靜桃,俞煌穎,王峰. 四足機(jī)器人對(duì)角小跑步態(tài)動(dòng)態(tài)穩(wěn)定步行足端非連續(xù)約束及動(dòng)力學(xué)建模[J]. 中國(guó)機(jī)械工程,2015,26(5):592-597.
Lei Jingtao, Yu Huangying, Wang Feng. Analysis on non-continuous constraints and dynamics modeling of quadruped robot dynamically stable walking with trot gait[J]. China Mechanical Engineering, 2015, 26(5): 592-597. (in Chinese with English abstract)
[14]郝仁劍,王軍政,史大威,等. 基于速度矢量的四足機(jī)器人間歇步態(tài)規(guī)劃方法[J]. 機(jī)器人,2016,38(5):540-549.
Hao Renjian, Wang Junzheng, Shi Dawei, et al. Intermittent gait planning methon of quadruped robot based on velocity vector[J]. Robot, 2016, 38(5): 540-549. (in Chinese with English abstract)
[15]高炳微,王思凱,高元峰. 液壓四足機(jī)器人單腿豎直跳躍步態(tài)規(guī)劃[J]. 儀表儀器學(xué)報(bào),2017,38(5):1086-1092.
Gao Bingwei, Wang Sikai, Gao Yuanfeng. Single leg vertical hopping gait planning for hydraulic quadruped robot[J]. Chinese Journal of Scientific Instrument, 2017, 38(5): 1086-1092. (in Chinese with English abstract)
[16]韓寶玲,汪清強(qiáng),賈燕,等. 提升四足機(jī)器人行走穩(wěn)定性的對(duì)角步態(tài)規(guī)劃方法[J]. 北京理工大學(xué)學(xué)報(bào),2018,38(9):917-920.
Han Baoling, Wang Qingqiang, Jia Yan, et al. Trot gait planning method for improving the stability of quadruped robot[J]. Transaction of Beijing Institute of Technology, 2018, 38(9): 917-920. (in Chinese with English abstract)
[17]何冬青,馬培蓀,曹曦,等. 四足機(jī)器人對(duì)角小跑起步姿態(tài)對(duì)穩(wěn)定步行的影響[J]. 機(jī)器人,2004,26(6):925-932.
He Dongqing, Ma Peisun, Cao Xi, et al. Impact of Initial stance of quadruped trotting on walking stability[J]. Robot, 2004, 26(6): 925-932. (in Chinese with English abstract)
[18]張志宇. 基于ADAMS的四足機(jī)器人虛擬樣機(jī)仿真及剛?cè)狁詈戏治鯷D]. 哈爾濱:哈爾濱工業(yè)大學(xué),2016.
Zhang Zhiyu. Virtual Prototype Simulation and Rigid-flexible Coupling Analysis of Quadruped Robot Based on ADAMS[D]. Harbin: Harbin Institute of Technology, 2016. (in Chinese with English abstract)
[19]Lei J T, Yu H Y. Dynamics analysis of bionic flexible body driven by pneumatic artificial muscle for quadruped robot[J]. Journal of shanghai Jiao Tong University, 2014, 48(12): 1688-1693: 1699.
[20]Rong X W, Song E, Li B. Joint driving forces calculation for a quadruped for a quadruped robot in sagittal plane motions[J]. Journal of Information and Computational Science, 2012, 9(12): 3413-3419.
[21]任東一,邵俊鵬,孫桂濤,等. 液壓四足機(jī)器人機(jī)身擾動(dòng)抑制及實(shí)驗(yàn)研究[J]. 機(jī)器人,2019,40(4):163-171.
Ren Dongyi, Shao Junpeng, Sun Guitao, et al. Torso disturbance inhibition and experiment research of hydraulic quadruped robot[J]. Robot, 2019,40(4): 163-171. (in Chinese with English abstract)
[22]陳騰,李貽斌,榮學(xué)文. 四足機(jī)器人動(dòng)步態(tài)下實(shí)時(shí)足底力優(yōu)化方法的設(shè)計(jì)與驗(yàn)證[J]. 機(jī)器人,2019,41(3):307-316.
Chen Teng, Li Yibin, Rong Xuewen. Design and verification of real-time plantar force optimization quadruped robots in dynamic gait[J]. Robot, 2019, 41(3): 307-316. (in Chinese with English abstract)
[23]蔣榮超,劉大維,王登峰. 基于熵權(quán)TOPSIS方法的整車動(dòng)力學(xué)性能多目標(biāo)優(yōu)化[J]. 機(jī)械工程學(xué)報(bào),2018,54(2):150-158.
Jiang Rongchao, Liu Dawei, Wang Dengfeng. Multi-objective optimization of vehicle dynamics performance based on entropy weighted TOPSIS method[J]. Journal of Mechanical Engineering, 2018 54(2): 150-158. (in Chinese with English abstract)
[24]張良安,馬演東,單家正,等. 4自由度含局部閉鏈?zhǔn)酱a垛機(jī)器人動(dòng)力學(xué)優(yōu)化設(shè)計(jì)[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2013,44(11):336-341.
Zhang Liang’an, Ma Yandong, Shan Jiazheng, et al. Optimal dynamic design of 4-DOF palletizing robot with closed-chain[J]. Transactions of The Chinese Society of Agricultural Machinery, 2013, 4(11): 336-341. (in Chinese with English abstract)
[25]趙海鳴,蔣彬彬,李密. 基于拓?fù)鋬?yōu)化與多目標(biāo)的機(jī)床底座結(jié)構(gòu)設(shè)計(jì)[J]. 機(jī)械設(shè)計(jì)與研究,2018,34(4):100-105.
Zhao Haiming, Jiang Binbin, Li Mi. Structure design of machine tool based on topology optimization and multi-objective optimization[J]. Machine Design and Research, 2018, 34(4): 100-105. (in Chinese with English abstract)
[26]沈惠平,李菊,王振,等. 基于結(jié)構(gòu)降耦和運(yùn)動(dòng)解耦的并聯(lián)機(jī)構(gòu)拓?fù)浣Y(jié)構(gòu)優(yōu)化及性能改善[J]. 機(jī)械工程學(xué)報(bào),2017,53(19):100-105.
Shen Huiping, Li Ju, Wang Zhen, et al. Topology structure optimization and performance improvement for parallel mechanisms based on structure coupling-reducing and motion decoupling[J]. Journal of Mechanical Engineering, 2017, 53(19): 176-186. (in Chinese with English abstract)
[27]丁良宏. BigDog四足機(jī)器人關(guān)鍵技術(shù)分析[J]. 機(jī)械工程學(xué)報(bào),2015,51(7):1-22.
Ding Lianghong. Key Technology analysis of BigDog quadruped robot[J]. Journal of Mechanical Engineering, 2015, 51(7): 1-22. (in Chinese with English abstract)
[28]孫龍飛,房立金. 機(jī)械手臂結(jié)構(gòu)設(shè)計(jì)與性能分析[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2017,48(9):402-410.
Sun Longfei, Fang Lijin. Design and performance analysis of novel robotic arm[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(9): 402-410. (in Chinese with English abstract)
[29]張良安,萬(wàn)俊,譚玉良. Ahut-Delta 并聯(lián)機(jī)構(gòu)改進(jìn)混沌粒子群算法尺度綜合[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2015,46(8):344-351.
Zhang Liang’an, Wan Jun, Tan Yuliang. Dimensional synthesis of ahut-delta parallel mechanism based on improved chaotic particle swarm algorithm[J]. Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(8): 344-351. (in Chinese with English abstract)
[30]王強(qiáng),張培林,玉懷光,等. 基于多目標(biāo)粒子群算法的稀疏分解參數(shù)優(yōu)化[J]. 振動(dòng)與沖擊,2017,36(23):45-50.
Wang Qiang, Zhang Peilin, Yu Huaiguang, et al. Parametric optimization of sparse decomposision based on multi-objective particle swarm optimization algorithm[J]. Journal of Vibration and Shock, 2017, 36(23): 45-50. (in Chinese with English abstract)
[31]劉建,葛世榮,朱華,等. 基于多目標(biāo)優(yōu)化的礦用救援機(jī)器人動(dòng)力匹配[J]. 機(jī)械工程學(xué)報(bào),2015,51(3):18-28.
Liu Jian, Ge Shirong, Zhu Hua, et al. Mine rescue robot power matching based on multi-objective particle swarm optimization[J]. Journal of Mechanical Engineering, 2015, 51(3): 18-28. (in Chinese with English abstract)
Optimization of leg structure parameter of quadruped laser weeding robot
Zhang Liang’an1, Tang Kai1, Zhao Yongjie2※, Wang Xiaoyi1,Yu Dazhuang1, Lu Xinjian3, Wang Xiang1
(1.243000;2.515063;3.,528226,)
Robot has been increasingly used in weeding. Taking the quadruped laser weeding robot as an example, this paper presents a method to optimize its four legs using the dynamic scale synthesis. The objective of the optimization was the lengths of the legs, in which, prior to optimizing the driving torque of the joint in each leg, we first optimized the driving torque of the leg based on the load it was required to take. The laser weeding robot used compound cycloid trajectory to plan its trajectory, which is ready for optimizing the driving torque of the joint in the rear leg. The optimization focused on the driving force moment on both the thigh joint and the calf joint, for which obtained a set of optimal thigh bar lengths and calf bar lengths. We first calculated the foot trajectory and the gait parameters of the robot, and then used the thigh joint force moment, the calf joint driving force moment and different leg rod lengths to maximize the dynamic target; the dimensional parameters of the legs were also taken as design parameter in the multi-objective constraint optimization. After the thigh joint driving force moment and the calf joint driving moment were firstly optimized using the particle group algorithm, the maximum torque of the leg joints within the each group of the rods was calculated under a given target trajectory by combining the leg sizes. Based on the smallest set of driving force moments, we found the maximum and minimum driving torque of each joint in the target trajectory, which was further optimized using the ideal dot method and the particle group algorithm. This transformed the multi-objective optimization to a single-objective optimization. These two-step optimizations allowed us to obtain the leg length in the group of optimal four-legged laser weeding robots. We verified the optimal results against experiments. The results showed that the optimized four-legged laser weeding robot significantly improved its dynamic performance compared to that without optimization. It was also found that the optimization reduced the maximum thigh joint driving force by 5.29% and the maximum driving force in the calf joint by 18.05%. The comprehensive methods presented in this paper cannot only help developing four-legged laser weeding robot prototype, but also provide references for studying driving force moment and energy consumption of four-legged robot.
quadruped robot; optimization; dimensional synthesis; particle swarm optimization
張良安,唐 鍇,趙永杰,王孝義,余大壯,盧新建,王 祥. 四足激光除草機(jī)器人腿部結(jié)構(gòu)參數(shù)優(yōu)化[J]. 農(nóng)業(yè)工程學(xué)報(bào),2020,36(2):7-15.doi:10.11975/j.issn.1002-6819.2020.02.002 http://www.tcsae.org
Zhang Liang’an, Tang Kai, Zhao Yongjie , Wang Xiaoyi, Yu Dazhuang, Lu Xinjian, Wang Xiang. Optimization of leg structure parameter of quadruped laser weeding robot[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(2): 7-15. (in Chinese with English abstract) doi:10.11975/j.issn.1002-6819.2020.02.002 http://www.tcsae.org
2019-10-24
2019-11-04
安徽省重點(diǎn)研究與開發(fā)計(jì)劃項(xiàng)目(201904b11020030);廣東省普通高校省級(jí)重大科研項(xiàng)目—基礎(chǔ)研究重大項(xiàng)目及應(yīng)用研究重大項(xiàng)目(自然科學(xué)類)(2017KZDXM036);佛山市科技創(chuàng)新團(tuán)隊(duì)專項(xiàng)項(xiàng)目資助(2018IT100052)
張良安,博士,教授,主要從事機(jī)器人技術(shù)及應(yīng)用研究。Email:robotlab@ahut.edu.cn
趙永杰,博士,教授,主要從事機(jī)器人技術(shù)與機(jī)器設(shè)計(jì)研究。Email:meyizhao@stu.edu.cn
10.11975/j.issn.1002-6819.2020.02.002
TP242.6
A
1002-6819(2020)-02-0007-09