外文資料翻譯--人工和工業(yè)機(jī)器人重復(fù)操作定位精度比較

1、<p>　　畢業(yè)設(shè)計(jì)外文資料翻譯</p><p>　　題 目 人工和工業(yè)機(jī)器人重復(fù)操作定位 </p><p>　　學(xué) 院 機(jī)械工程學(xué)院 </p><p>　　專 業(yè) 機(jī)械工程及自動(dòng)化 </p><p>　　班 級(jí)

2、 </p><p>　　學(xué) 生 </p><p>　　學(xué) 號(hào) </p><p>　　指導(dǎo)教師 </p><p&g

3、t;　　二〇一四年三月三十日</p><p>　　Computers in Biology and Medicine 28 (1998) 415-421</p><p>　　Comparison of position repeatability of a human operator and an industrial manipulating robot</p><

4、p>　　Jure Zupancic*, Tadej Bajd</p><p>　　Faculty of Electrical Engineering, University of Ljubljana, TrzÏasÏka 25, SI-1000 Ljubljana, Slovenia</p><p>　　Received 8 December 1997</p

5、><p>　　Abstract Robot performance criteria of position repeatability are studied. Weight-to-payload ratio is in manipulating robots significantly higher than in human operators. Bracing strategy improving the

6、robot performances is introduced in the paper. The strategy copies human behavior during fine motion operations. A comparison is made between the robot and the human operator performing approximately the same manipulatin

7、g task. Contactless measurements of position repeatability were accomplished </p><p>　　©1998 Elsevier Science Ltd. All rights reserved.</p><p>　　Keywords: Industrial robot; Human operator;

8、Repeatability; Standards; Measurements; Bracing strategy</p><p>　　1 Introduction</p><p>　　Modern robot manipulators replacing human operator in assembly tasks [1] are usually designed in accord

9、ance with the principles encountered in humans and their arms and hands.The main goal of developing a mechanical counterpart to human operator is achieving of improved performances such as speed, payload capacity, positi

10、on accuracy and repeatability.However, when the loads are not exceeding 3 kilograms, weight-to-payload ratio found in robots is reported to be ten times higher than the same rat</p><p>　　In a special case, i

11、t is possible to increase the robot absolute accuracy and repeatability by using appropriate bracing [4]. The method of bracing [5] is a direct copy of human behavior where the adaptation to higher accuracy and repeatabi

12、lity requirements during the fine motion operation is required. A human operator, when performing precise manipulation tasks, often finds supports for his forearm, wrist or elbow like in many working situations watch-mak

13、ers are practicing. The same simple idea</p><p>　　Apart from some estimations, there was no explicit comparison made between human operator and adequate robot performances. The aim of this investigation was

14、to make concise technical comparison of robot and human operator performance with and without bracing.</p><p><b>　　2 Method</b></p><p>　　Position repeatability tests were performed

15、in accordance with ISO 9283 standard for manipulating industrial robots [6]. The cube with maximum volume was located in the workspace of the most frequent anticipated use. Five points (P1-P5) were located on the diagona

16、ls of the selected plane as shown in Fig.1 . Contactless OPTOTRAK®/3010 motion analysis system was used for measuring of the actual positions (see Fig. 2). Measurements were taken at the poses P1, P2, P3, P4 and P5.

17、 The position repeata</p><p><b>　　(1)</b></p><p><b>　　(2)</b></p><p><b>　　(3)</b></p><p>　　Fig. 1. Definition of the measuring point

18、s (P1-P5) inside the robot workspace.</p><p>　　Fig. 2. Measuring equipment for position repeatability testing.</p><p><b>　　(4)</b></p><p>　　where xj, yj, zj represent no

19、minal positions, mean actual positions and SD is the positional standard deviation.</p><p>　　The repeatability tests were made for (1) robot, (2) braced robot, (3) human operator and (4)human operator with

20、his forearms braced. In cases (1) and (2) the nominal poses were commanded by the robot program. The same standard requirements were involved in the tests with human operator. The nominal poses were commanded by a wire f

21、rame with five ring targets which were positioned into the working space of the operator. The operator held in his hand the same measuring rigid body as the robot. App</p><p>　　A saddle shaped support body w

22、as used in case (2). In case (4) a horizontal bar was used to support the human operator's forearm. The four measuring situations are shown in Fig.3.</p><p>　　3 Testing and results</p><p>　

23、　Asea Irb 6 industrial manipulating robot was tested in our experiments. Additional segment was attached to the robot enabling bracing against the support body. For the case (1) (Fig.3a) the weight of additional segment

24、was 1.65 kg, while in the case (2) (Fig.3c) it was 2.15 kg. The difference was due to the construction details of a bracing segment. The weight of the measuring rigid body held by human operator was 0.8 kg. Three persons

25、 were tested. 5% of the total human operator's weight was ass</p><p>　　Fig.3 Four types of measurements (a) free robot, (b) human operator, (c) braced robot and (d) human operator with braced forearm.<

26、;/p><p>　　Fig.4 The results of the repeatability test in free robot and braced robot.</p><p>　　Fig.5 The results of the repeatability test in free human operator and with bracing.</p><p&

27、gt;　　In the next four histograms (Figs. 4 and 5) the results of the position repeatability for the four specified types of testing are presented.</p><p>　　The repeatability of the braced robot as compared to

28、 the free robot was improved for approximately 50%. The same improvement in the human operator was for about 25%. Note,that the weight-to-payload ratio for robot was more than 10_ higher than for human operator.</p>

29、;<p>　　4 Conclusion</p><p>　　Applying bracing strategy during robot manipulation is a copy of human behavior. Position repeatability performance criteria were studied. Repeatability measurements acco

30、rding to ISO 9283 were performed with industrial manipulating robot. For the first time, the same measurement of position repeatability under condition of ISO 9238 standard was used in the test of a human operator perfor

31、mance. The results show significant improvement of position repeatability in cases when the robot and the huma</p><p>　　5 Summary</p><p>　　The industrial manipulating robots are usually more or

32、 less accurate structural copies of human operator's arms and hands. Structural copying of the natural mechanisms alone does not always give satisfying results, hence the modern robotic manipulators with maximum rate

33、d loads below 3 kg are mechanically inferior to human operators. Bracing strategy which is used for improving particular robot performances is the combination of the structural and functional copying of natural human beh

34、avior. A h</p><p>　　The objective of our research was to get concise technical comparison between the robot and human operator's performances concerning positional repeatability. ISO 9283 standard for ma

35、nipulating industrial robot performance criteria and related test methods was the basis for the comparative study. The testing equipment was built around the contactless 3-D motion analysis system OPTOTRAK®/3010 (No

36、rthern Digital). The position repeatability tests were performed for the robot Asea Irb 6 and the huma</p><p>　　Acknowledgements</p><p>　　This work was partly sponsored by the Slovenian Ministry

37、 of Science and Technology. The authors wish to acknowledge the contribution of Dana MaurovicÏ and Ivan LoncÏar performing the described measurements.</p><p>　　References</p><p>　　[1]

38、J. ZupancÏ icÏ , Calibration of an SMT Robot Assembly Cell, Journal of Robotic Systems 11 (4) (1994) 301-310.</p><p>　　[2] E.I. Rivin, Mechanical design of robots, Mc.Graw-Hill, 1988.</p>&l

39、t;p>　　[3] G. Belforte, M. Gola, N. D'Alfio, Design and testing of carbon fiber robots, in: Proceedings of the 2nd International Conference on Robotics, Dubrovnik, 1989, pp. 361-372.</p><p>　　[4] J. Zu

40、pancÏ icÏ , Enhancing robot mechanism performances by using mechanical support: kinematic analysis, in:Proceedings of the 3rd International Workshop on Advances in Robot Kinematics, Ferrara, Italy, 1992, pp.297

41、-303.</p><p>　　[5] W. Book, S. Le, V. Sangveraphunsiri, Bracing strategy for robot operation, in: Proceedings of the Symposium on the Theory and Practice of Robots and Manipulators, Udine, 1984, pp. 179-185.

42、</p><p>　　[6] Manipulating industrial robots: performance criteria and related test methods, ISO 9283, International Organization for Standardization, 1988.</p><p>　　[7] J. ZupancÏ icÏ

43、 , A. Kralj, Modeling of a braced robot with four-bar mechanism, in: J. LenarcÏ icÏ , V. Parenti-Castelli(Eds.), Recent Advances in Robot Kinematics, Kluwer Academic Publishers, Dordrecht, 1996, pp. 307-316.<

44、;/p><p>　　計(jì)算機(jī)在生物學(xué)和醫(yī)學(xué)的應(yīng)用28（1998）415-421</p><p>　　人工和工業(yè)機(jī)器人重復(fù)操作定位精度比較</p><p>　　Jure Zupancic*, Tadej Bajd</p><p>　　盧布爾雅那大學(xué)電氣工程學(xué)院，TrzÏasÏka 25, SI-1000 盧布爾雅那，斯洛文尼亞&l

45、t;/p><p>　　1997年12月8日</p><p>　　摘 要 機(jī)器人重復(fù)定位精度性能所受的影響。操作機(jī)器人的重量-有效載荷的比例要明顯的高于人工操作。支撐的使用提高了機(jī)器人的上述性能。這個(gè)策略模仿了人工操作時(shí)的動(dòng)作。將機(jī)器人和操作人員在相同的條件下操作做一次比較。通過OPTOTRAK®運(yùn)動(dòng)分析系統(tǒng)完成了非接觸式重復(fù)定位精度測(cè)量的測(cè)試。實(shí)驗(yàn)結(jié)果表明機(jī)器人和操作人員的定位精

46、度在使用支撐后得到了相當(dāng)大的改善。</p><p>　　©1998年，艾斯維爾科技有限公司保留所有權(quán)利。</p><p>　　關(guān)鍵詞 工業(yè)機(jī)器人，人為操作的重復(fù)性，標(biāo)準(zhǔn)，測(cè)量，支撐戰(zhàn)略</p><p><b>　　1 介紹</b></p><p>　　現(xiàn)代機(jī)器人中代替操作人員執(zhí)行裝配任務(wù)的機(jī)械手通常是按照人

47、類的胳膊和手來設(shè)計(jì)的。機(jī)械對(duì)應(yīng)人工操作發(fā)展的主要目標(biāo)是實(shí)現(xiàn)功能改善，如提高速度，增加有效載荷能力，提高定位精度和可重復(fù)性。然而，當(dāng)負(fù)載不超過3公斤時(shí)，經(jīng)評(píng)估發(fā)現(xiàn)機(jī)器人的重量—有效載荷比是人類在相同操作情況下的10倍以上。從技術(shù)和經(jīng)濟(jì)的角度來看這個(gè)比例的減少與機(jī)器人的效率密切相關(guān)。提高這個(gè)比率的傳統(tǒng)原則是引入更輕的材料，創(chuàng)建一個(gè)新的結(jié)構(gòu)，設(shè)計(jì)新的執(zhí)行器。</p><p>　　在特殊情況下，使用適當(dāng)?shù)闹误w能夠增加機(jī)

48、器人的絕對(duì)精度和可重復(fù)性。支撐的方法是模仿人類在精細(xì)運(yùn)動(dòng)操作時(shí)的行為，以便能適應(yīng)更高的精度和可重復(fù)性要求的需要。操作員在進(jìn)行精確的操作任務(wù)時(shí)往往會(huì)像工作時(shí)的鐘表匠一樣為他的前臂，手腕處，肘關(guān)節(jié)等找一支撐體。同樣道理也可以用在機(jī)器人身上。</p><p>　　除了一些估計(jì)，在人工和機(jī)器人的表現(xiàn)之間并沒有明確的比較。這次研究的目的是為了是機(jī)器人和人工操作者在有支撐和沒有支撐的條件下作一個(gè)操作技術(shù)的比較。</p&

49、gt;<p><b>　　2 方法</b></p><p>　　工業(yè)操作機(jī)器人的重復(fù)定位精度測(cè)試按照ISO 9283的標(biāo)準(zhǔn)進(jìn)行測(cè)試。多維數(shù)據(jù)集與最大音量是在工作區(qū)中最常見的預(yù)期用途。五個(gè)點(diǎn)（P1—P5），如圖所示，分別位于圖中選定平面的對(duì)角線上。接觸式的OPTOTRAK®/3010運(yùn)動(dòng)分析系統(tǒng)用于測(cè)量實(shí)際位置（如圖2所示）。測(cè)量P1，P2，P3，P4和P5的位置。經(jīng)

50、過30次的重復(fù)測(cè)量后，重復(fù)定位精度達(dá)到了ISO 9283標(biāo)準(zhǔn)的要求?？芍貜?fù)性由以下的公式計(jì)算：</p><p><b>　　(1)</b></p><p><b>　　(2)</b></p><p><b>　　(3)</b></p><p>　　圖1 在機(jī)器人工作區(qū)內(nèi)定義測(cè)量

51、點(diǎn)（P1-P5）。</p><p>　　圖2 重復(fù)定位精度測(cè)量設(shè)備。</p><p><b>　　(4)</b></p><p>　　其中xj, yj, zj表示標(biāo)稱位置，表示實(shí)際位置和SD的位置的偏差。重復(fù)性試驗(yàn)測(cè)試的是機(jī)器人（1），有支撐的機(jī)器人（2），操作員（3），前臂支撐的操作員（4）。其中在（1）和（2）中機(jī)器人執(zhí)行的是指定的程序。參

52、加測(cè)試的操作人員要按照相同的標(biāo)準(zhǔn)執(zhí)行相同的命令。標(biāo)準(zhǔn)的操作是將有五個(gè)環(huán)的目標(biāo)定位到操作員工作空間線框。以相同的標(biāo)準(zhǔn)測(cè)量操作員的手和機(jī)器人的操作臂。為操作員和機(jī)器人準(zhǔn)備相同大小的操作空間。在（2）中使用相同的鞍形支撐體。在（4）中用一個(gè)水平棒支撐操作員的前臂。四個(gè)測(cè)量情形如下圖3所示。</p><p><b>　　3 測(cè)試和結(jié)果</b></p><p>　　我們?cè)趯?shí)驗(yàn)

53、室里用工業(yè)操作機(jī)器人Asea Irb 6來進(jìn)行測(cè)試。附加的部分連接到機(jī)器人上從而能夠?qū)ζ渥龀鲇欣?。?duì)于（1）（圖3a）的情況下附加部分的重量是1.65公斤，而在（2）中（圖3c）附加部分的重量是2.15公斤。差異是由支撐部分引起的。人工操作員的持有測(cè)量剛體的重量是0.8公斤。3個(gè)操作員參加了測(cè)試。假定人體體重的5%為手臂的重量。在人手上附加0.9公斤的重量來模擬機(jī)器人手上的相同重量的工具。重量-有效載荷比大約為（a）無支撐機(jī)器人82

54、，（b）無支撐操作人員2.4，（c）支撐機(jī)器人62，（d）有支撐的操作員5.</p><p>　　圖3 四個(gè)類型的測(cè)量（a）自由移動(dòng)機(jī)器人，（b）操作人員，（c）支撐機(jī)器人，（d）支撐前臂的操作員</p><p>　　圖4 基于自由機(jī)器人和支撐機(jī)器人的可重復(fù)性測(cè)試結(jié)果</p><p>　　圖5 基于自由操作人員和支撐前臂的操作人員的可重復(fù)性測(cè)試結(jié)果</p>

55、;<p>　　在以上四個(gè)直方圖中（圖4和圖5）展示了指定的四個(gè)類型的位置重復(fù)性測(cè)量結(jié)果。</p><p>　　支撐機(jī)器人的可重復(fù)性比自由機(jī)器人改善了大約50%。有支撐的操作人員同比增長(zhǎng)了大約25%。請(qǐng)注意，該機(jī)器人的重量-有效載荷比高出了操作人員的10倍多。</p><p><b>　　4 結(jié)論</b></p><p>　　對(duì)機(jī)

56、器人應(yīng)用支撐是對(duì)人類行為的模仿。對(duì)重復(fù)定位精度的性能標(biāo)準(zhǔn)進(jìn)行了研究。按照ISO 9283的標(biāo)準(zhǔn)對(duì)工業(yè)機(jī)器人進(jìn)行了重復(fù)性測(cè)量。這是第一次在ISO 9283的標(biāo)準(zhǔn)下對(duì)操作人員的操作技術(shù)進(jìn)行的位置重復(fù)性測(cè)量。結(jié)果表明機(jī)器人和操作人員在有支撐體支撐的條件下位置重復(fù)性得到了顯著的改善。這項(xiàng)研究鼓勵(lì)了研究人員對(duì)于支撐方案進(jìn)行進(jìn)一步的研究。操作人員位置重復(fù)性的研究同樣也可用于人體工程學(xué)的研究。</p><p><b>

57、;　　5 總結(jié)</b></p><p>　　工業(yè)操控機(jī)器人的準(zhǔn)確結(jié)構(gòu)或多或少的模仿了人類的手臂。完全復(fù)制人類的結(jié)構(gòu)通常并不能得到讓人得到滿意的結(jié)果，因此最大額定負(fù)載低于3千克的現(xiàn)代機(jī)械手要遜色于人工操作。用于改善機(jī)器人操作性能的支撐方法是對(duì)自然人行為的結(jié)構(gòu)和功能的模仿。操作人員在進(jìn)行精準(zhǔn)的操作任務(wù)時(shí)經(jīng)常會(huì)為他的前臂、手腕處或肘關(guān)節(jié)做一些支撐，就像是鐘表匠工作時(shí)那樣。同樣的道理可用在機(jī)器人上，用來改善

58、精度、可重復(fù)性、剛度、有效載荷能力和機(jī)械振動(dòng)等機(jī)械特性。</p><p>　　我們研究的目的是通過比較機(jī)器人和操作人員的表現(xiàn)來得到關(guān)于位置重復(fù)性的簡(jiǎn)明技術(shù)。ISO 9283關(guān)于操控工業(yè)機(jī)器人性能標(biāo)準(zhǔn)及相關(guān)測(cè)試方法的標(biāo)準(zhǔn)是對(duì)其進(jìn)行比較研究的基礎(chǔ)。檢測(cè)設(shè)備是圍繞著非接觸式3-D運(yùn)動(dòng)分析系統(tǒng)OPTOTRAK®/3010 (Northern Digital)建立。位置重復(fù)性實(shí)驗(yàn)是由機(jī)器人Asea Irb 6和操

59、作人員在相同的實(shí)驗(yàn)條件下完成的相同的試驗(yàn)任務(wù)。機(jī)器人和操作人員在自由和有支撐的條件下進(jìn)行了測(cè)試。有支撐的機(jī)器人的重復(fù)定位精度比自由機(jī)器人高出了約50%。有支撐的操作人員則同比增長(zhǎng)了25%。機(jī)器人的重量-有效載荷比超過了操作人員的10倍。這研究結(jié)果鼓勵(lì)了研究人員對(duì)支撐機(jī)器人的運(yùn)動(dòng)學(xué)作進(jìn)一步的研究。本次研究結(jié)果可以作為人體工程學(xué)研究的參考。</p><p><b>　　致謝</b></p&

60、gt;<p>　　斯洛文尼亞科學(xué)和技術(shù)部對(duì)本次研究工作提供了部分贊助。感謝Dana MaurovicÏ和Ivan LoncÏar對(duì)上述測(cè)量作出的貢獻(xiàn)。</p><p><b>　　參考文獻(xiàn)</b></p><p>　　[1] J. ZupancÏ icÏ , Calibration of an SMT Robot

61、Assembly Cell, Journal of Robotic Systems 11 (4) (1994) 301-310.</p><p>　　[2] E.I. Rivin, Mechanical design of robots, Mc.Graw-Hill, 1988.</p><p>　　[3] G. Belforte, M. Gola, N. D'Alfio, Desi

62、gn and testing of carbon fiber robots, in: Proceedings of the 2nd International Conference on Robotics, Dubrovnik, 1989, pp. 361-372.</p><p>　　[4] J. ZupancÏ icÏ , Enhancing robot mechanism perform

63、ances by using mechanical support: kinematic analysis, in:Proceedings of the 3rd International Workshop on Advances in Robot Kinematics, Ferrara, Italy, 1992, pp.297-303.</p><p>　　[5] W. Book, S. Le, V. Sang

64、veraphunsiri, Bracing strategy for robot operation, in: Proceedings of the Symposium on the Theory and Practice of Robots and Manipulators, Udine, 1984, pp. 179-185.</p><p>　　[6] Manipulating industrial robo

65、ts: performance criteria and related test methods, ISO 9283, International Organization for Standardization, 1988.</p><p>　　[7] J. ZupancÏ icÏ , A. Kralj, Modeling of a braced robot with four-bar m