機械手外文翻譯--機械手的遠程控制系統(tǒng)

1、<p>　　本科畢業(yè)設(shè)計(論文)</p><p>　　外文翻譯（附外文原文）</p><p>　　學(xué) 院： 機械與控制工程學(xué)院 </p><p>　　課題名稱： 搬運機械手的結(jié)構(gòu)和液壓系統(tǒng)設(shè)計 </p><p>　　專業(yè)(方向)： 機械設(shè)計制造及其自動化（機械裝備）</p>

2、<p>　　班 級： </p><p>　　學(xué) 生： </p><p>　　指導(dǎo)教師： </p><p>　　日 期： 2015年3月10日 &

3、lt;/p><p>　　Proceedings of the 33rd Chinese Control Conference</p><p>　　July 28-30, 2014, Nanjing, China</p><p>　　The Remote Control System of the Manipulator</p><p>　　SUN

4、 Hua, ZHANG Yan, XUE Jingjing , WU Zongkai</p><p>　　College of Automation, Harbin Engineering University, Harbin 15000</p><p>　　E-mail: sunhuas@hrbeu.edu.cn</p><p>　　Abstract: A rem

5、ote control system of the 5 degree of freedom manipulator was designed. This manipulator was installed into our mobile robot to constitute a remote rescue robot. The Denavit-Hartenberg method was used to establish the ki

6、nematic models and the path planning of the manipulator was researched. The operator could remote control the manipulator by the interactive interface of PC which could display moving picture and various data of the mani

7、pulator. The servos of the manipulator were c</p><p>　　Key Words: The manipulator; Remote control; Denavit-Hartenberg; FPGA; Human-computer interaction</p><p>　　1 Introduction</p><p&g

8、t;　　With the development of the microelectronic technique and the computer technology, the manipulator has become essential equipment in the manufacturing industry. As we all known, the manipulator is usually applied to

9、accomplish dull, onerous and repeated physical work, especially used to substitute the manual operation under the dangerous and the hazardous environment such as the corrosion and the high temperature.</p><p&g

10、t;　　In this paper, the manipulator was installed our mobile robot. The tele-operation system of this manipulator was designed. The whole system is onstituted by PC and slave FPGA. The operator can remote control the mani

11、pulator by PC. The wireless communication was used for transmitting data between PC and FPGA. FPGA is controller of the the manipulator in the mobile robot. FPGA has the abundant internal resource and IP cores. And a cen

12、tral control option was built via an embedded Nios II program an</p><p>　　MATLAB software was adopted to build the kinematic models of manipulator. And using D-H (the acronym of Denavit-Hartenberg) method to

13、 solve the forward and inverse kinematic equations of the manipulator, to analyze the motivation, to plan and track the motion’s path. </p><p>　　In addition, a good interface of human-computer interaction wa

14、s enhanced in the remote control system of the manipulator in PC. Moreover, the manipulator simulation technology was built by using the mixed programming of VC and MATLAB. Thus, the motion choreographs was got quickly a

15、nd easily, also greatly saved time and cut the cost.</p><p>　　2 Manipulator Model and Path Planning</p><p>　　At first, the motion model of the manipulator was built. Then, the kinematic simulati

16、on and its path planning were researched. These works provided the foundation for the design of the remote control system of the manipulator.</p><p>　　2.1 Motion Model of the Manipulator</p><p>

17、　　The manipulator was regarded as an open loop kinematic chain. It was constituted by five rotary joints. And its one end was fixed on a base while the other end was used to achieve the ability of grabbing. Therefore, it

18、 is better to establish a chain coordinate frame as shown in Fig.1. The terminal position and attitude was determined via using forward kinematic equation after knowing the rotating angle of every joint. The D-H paramete

19、r table shown as Table 1 was established by using the frames i</p><p>　　Fig.1 Coordinate frames of mechanical arm</p><p>　　Table 1 D-H Parameters of the Robot Arm</p><p>　　Due to D-

20、H method:</p><p>　　Where , , , S . The transformation matrix of every joint was given by equation (2).</p><p><b>　　(2)</b></p><p>　　Where unit vector in equation (2)

21、was , , , . Parameters of mechanical arm were given by ,, ,. Therefore the forward kinematic</p><p>　　equation was determined by taking every parameter in equation (3).</p><p><b>　　(3)<

22、/b></p><p>　　In practical application, the manipulator was adopted to grab objects. This required that the fixed position was given from terminal to target location. That was the inverse kinematic analysi

23、s of manipulator. Inverse transformation was used to determine angle of every rotary joint toward the established coordinates. And the used method of inverse transformation was the common method to solve such problem (th

24、is method also known as algebraic method).Using inverse transformation separately to the </p><p>　　2.2 Motion Simulation of the Manipulator</p><p>　　The manipulator model was built and simulated

25、 via MATLAB toolbox. We could verify the rationality of the mathematical model. While the MATLAB model was established by table 1 and shown as Fig.2</p><p>　　Fig.2 MATLAB simulation of the manipulator</p&

26、gt;<p>　　Comparing to the Fig.1 and Fig.2, the simulation model of the manipulator was coincided to the reference frame model. That was to say, the given coordinate frame was correct. These results also could be p

27、roved by the determined inverse kinematic equations via MATLAB shown in the table (2) and table (3).</p><p>　　The target position was solved by forward kinematics. After that, the rotary angles were calculat

28、ed by inverse kinematical equation. It turned out that these rotary angles coincided to the given angles. Therefore, these results verified the correctness of forward and inverse kinematical equation.</p><p>

29、;　　Table (2) Forward Kinematics Analyze</p><p>　　Table (3) Inverse Kinematics Analyze</p><p>　　3 Path Planning of the Manipulator</p><p>　　The total displacement of joint was calcul

30、ated by inverse kinematical equation when the manipulator moved to new position. Thus, the manipulator could move to new position. Although the manipulator finally moved to the expected position in such condition, the mo

31、tion of the manipulator between these two points was unknown. Due to space limitations, motion and some certain position requirements, the manipulator was often unable to move as the above mentioned method. Therefore, th

32、e motion path was </p><p>　　In this paper, we could use these certain limitations to decide some expected points. And these expected points were used to match the planning path of the manipulator’s movement.

33、 Owing to the planning path, coordinate in every part could be calculated. The rotary angle of every joint was calculated via inverse kinetical equation and these angles realized the movement of planning path. Movement o

34、f the manipulator was shown in Fig.3 (Where‘?’ represented the points would be passed by the manipula</p><p>　　Fig.3 The path planning simulation of the manipulator</p><p>　　4 Remote Control Sy

35、stem of the Manipulator</p><p>　　The remote control system of the manipulator contains the main PC and the slave FPGA controller using DE2 Board of ALTER Company. The motors of the manipulator were controlle

36、d by multipath PWM waves. And the PWM waves were generated by IP core. The FPGA controller</p><p>　　Communicated with PC via wireless serial port. While in the PC interaction, the operator could observe the

37、 move of the manipulator in real-time and tele-control the motion of the manipulator. Also every movement of manipulator could be observed in advance via the simulation technique. The general design of the manipulator re

38、mote control system was shown in Fig.4.</p><p>　　Fig.4 The block diagram of the remote control system</p><p>　　4.1 Control Mode of the Manipulator</p><p>　　There were two control mo

39、des of the manipulator. One mode is that the inverse kinematical equations are calculated by FPGA straightly to determine angle of every rotary joint. Thus, the control of the manipulator was achieved. The advantage of t

40、his mode is more direct and independent to finish the control of the manipulator without the external devices. At the same time, this mode has large quantities of calculations, which occupy more internal storage and runn

41、ing time of FPGA. Resources of FPGA</p><p>　　The other mode accomplished the control of the manipulator by using VC and MATLAB in PC. Using VC and MATLAB finished a large number of complex calculations and d

42、etermined angle of every rotary joint. And the angle results were transmitted to FPGA in order to accomplish the control of the manipulator. This manner saved lots of internal storage and running time. In addition, FPGA

43、could finish other works under this mode. But the manipulator was not under fast control in this mode.</p><p>　　In this system, a new mode was adopted in the manipulator remote control system depending on th

44、e advantages of the two modes. Specifically, when the manipulator accomplished the specified and repeated movement the former mode was adopted under direct control by FPGA. When the manipulator wanted to achieve new moti

45、ons the latter mode was used to be commanded by orders from PC. This new mode was made good use of advantages of the two modes in the above. And this new mode lightened computational bu</p><p>　　4.2 SOPC Des

46、ign for the Remote Control System </p><p>　　Movement of the manipulator was controlled by servos. And the servos were controlled by PWM waves with the cycle of 20ms. Pulse width of these PWM waves was 0.5~2.

47、5ms corresponding to the rotary angle of servo with -90 degree to 90 degree. High precision of PWM waves were generated by IP core via Verilog in this system. The results were shown in Fig.5. PWM waves controlled rotary

48、angles of the servos via the servo drivers.</p><p>　　Fig.5 The PWM IP core</p><p>　　Multiple of IP cores were able to be downloaded into FPGA. And multiple PWM waves with high precision were gen

49、erated in the output. As shown in Fig.6, the pulse width of these waves could be settled by program of Nios II. The movement of the manipulator was more flexible and in higher precision in this system.</p><p&g

50、t;　　Fig.6 The IP cores generating PWM wave</p><p>　　The movement of the manipulator was accomplished by the duty ratio of PWM waves. Formula (4) inverted rotary angle to the corresponding amount of the duty

51、 ratio of PWM waves. The duty ratio of PWM waves corresponded to the Nios II output.</p><p>　　Wireless serial of 9600 baud rate was used to transmit the coordinate and the angle information from host compute

52、r to FPGA. After that, the data and orders were analyzed by FPGA Then FPGA transmitted the movement results to interactive interface of host computer via wireless transition model. This communication was realized through

53、 adding UVRT communication protocol to FPGA.</p><p>　　4.3 The Interactive Interface of the Remote Control System</p><p>　　The interactive interface of the remote control system was shown in Fig.

54、7. There were some functions in the interactive interface: video observation, the manipulator control and the simulation modeling.</p><p>　　At first, the manipulator video could be seen from camera to intera

55、ctive interface. The operator could monitor the manipulator in real-time. </p><p>　　Secondly, the angle and the coordinate could be set in control zone of the interactive interface. The angle of the manipula

56、tor could be set independently to each single joint. In addition, the angle setting could be shown in real-time in the list of interactive interface (as shown in Fig.7). In the set of coordinates, judging of coordinate s

57、etting assured that the total coordinates could achieve to the target points. Thus the manipulator could be controlled to move in the settled path depend on </p><p>　　Lastly, the MATLAB robot toolbox was emb

58、edded into this interactive interface. One interface was integrated both the control and simulation of the manipulator. MATLAB robot toolbox was directly used by interactive interface in the manipulator modeling. Each gr

59、oup of information was simulated separately in order to detect whether each movement was correct. And the general simulation could test whether movement arrangement of the manipulator was reasonable. Combining with multi

60、ple simulation metho</p><p>　　Fig.7 The interactive interface of the manipulator</p><p>　　5 Experiment and Simulation</p><p>　　In order to verify properties of the remote control s

61、ystem of the manipulator, experiments of the system were under way and were comparing to the simulation system. To be specific, manipulator modeling was built by interactive interface and a group of coordinates could be

62、designed. These coordinates were transmitted to FPGA, which controlled the servos to accomplish the movement of the manipulator. Joint angles, the terminal coordinates shot by interface video. The simulation results were

63、 shown </p><p>　　Fig.8 The experiment and the simulation</p><p>　　6 Conclusion</p><p>　　In the experiment, the 5-DOF manipulator modeling was simulated by MATLAB. In the slave FPGA

64、board, control of the manipulator was accomplished via IP core based on the Verilog language. That greatly reduced design of the peripheral circuit, cut the cost, improved the precision and made the movement smoother wit

65、hout shaking. While in the interactive interface, the mixed programming method of VC and MATLAB was embedded into the MATLAB simulation function. Thus the operability of this manipulator </p><p>　　References

66、</p><p>　　[1] Saeed B. Niku write, Sun Fuchun, Zhu Jihong, Liu Guodong</p><p>　　etc translate: Robotics Introduction, Beijing, Electronic</p><p>　　Industry Press, 2004(1):60-63,132-

67、137.</p><p>　　[2] Brady, M.J.M.Hollerbach, T.L.Johnson, T.Lozano-Perez, and</p><p>　　M.T.Mason, editors, Robot Motion; Planning and Control,</p><p>　　MIT Press, Cambridge, Mass, 198

68、2.</p><p>　　[3] Paul Richard P., Robot Manipulators, Mathematics,</p><p>　　Programming, and Control, The MIT Press 1981.</p><p>　　[4] Li Jian, Design and Research of Multi-DOF Robot

69、, Master</p><p>　　degree theses of master of university of technology, Chinese</p><p>　　acaedemy of sciences, 2009?20-31.</p><p>　　[5] Cheng Liyan, Fei Ling, Su Zelang, The 5-DOF Ma

70、nipulator</p><p>　　Kinematics Simulation Analysis Based on MATLAB,</p><p>　　Mechanical Research & Application, 2011(06).</p><p>　　[6] Zhang Puxing Jia Qiuling, Mechanical Arm Mu

71、lti-channel</p><p>　　Servo Control Design based on FPGA, Small and special</p><p>　　electrical machine. 2011, 39(4)</p><p>　　第33屆中國控制會議論文集</p><p>　　中國，南京，2014年28-30日<

72、;/p><p>　　機械手的遠程控制系統(tǒng)</p><p>　　孫華、張媛、薛晶晶、吳宗凱</p><p>　　哈爾濱工程大學(xué)，哈爾濱15000學(xué)院，自動化專業(yè)</p><p>　　電子油箱：sunhuas@hrbeu.edu.cn</p><p>　　摘要：一種5自由度機械手的遠程控制系統(tǒng)的設(shè)計。這種機械手被安裝到我們的移

73、動機器人構(gòu)成遠程救援機器人。這種Denavit-Hartenberg方法被用于建立運動學(xué)模型和機械手操作的路徑規(guī)劃的研究。操作者可以通過個人電腦來顯示移動圖像和操縱器的各種數(shù)據(jù)的交互式界面遠程控制機械手。操縱器的伺服系統(tǒng)是由從屬的FPGA控制器控制。此外，從屬的FPGA控制器經(jīng)由所述無線通信模塊的個人計算機進行通信。由于嵌入的Nios II程序和IP（知識產(chǎn)權(quán)）核生成PWM波，該系統(tǒng)可以控制多個伺服的快速性和靈活性。為了實現(xiàn)實時操作和仿

74、真，人機交互界面，建立了VC和MATLAB的混合編程。</p><p>　　關(guān)鍵詞：機械手、遠程控制;、Denavit-Hartenberg;、FPGA、人機交互</p><p><b>　　1.緒論</b></p><p>　　隨著微電子技術(shù)的發(fā)展和計算機技術(shù)，機械手在制造業(yè)中已經(jīng)成為必不可少的設(shè)備。正如我們所熟知的，機械手常適用于完成乏味的

75、、繁重的和反復(fù)的體力勞動，特別是用于替代在危險和有害環(huán)境的手動操作，例如腐蝕和高溫。</p><p>　　在本文中，機械手安裝在我們移動機器人上。這種機器手的遠程操作系統(tǒng)的設(shè)計。整個系統(tǒng)由個人電腦和從屬的FPGA構(gòu)成。操作者可以通過個人電腦遠程控制該機器人。無線通信被用于發(fā)送個人電腦和FPGA之間的數(shù)據(jù)。 FPGA是在移動機器人上機械手的控制器。 FPGA具有豐富的內(nèi)部資源和IP內(nèi)核。和一個中央控制選項，通過內(nèi)嵌

76、的Nios II程序和IP核心FPGA.建立，Verilog語言采用設(shè)計出生成的數(shù)字PWM波來控制機械手的IP核。因此，該系統(tǒng)可以達到較高的精度，易于調(diào)試。</p><p>　　MATLAB軟件采用建立機械手的運動模型，并使用DH（Denavit-Hartenberg的縮寫）的方法來解決前進和操縱逆運動學(xué)方程，分析動機，計劃和跟蹤運動路徑。</p><p>　　此外，人機交互的良好的界面增

77、強在PC上的機械手的遠程控制系統(tǒng)。而且，機器人仿真技術(shù)是通過使用VC和MATLAB的混合編程建成。因此，運動編排變得快速，方便，也大大節(jié)省了時間，降低了成本。</p><p>　　2 機械手的模型和路徑的規(guī)劃</p><p>　　首先，建立機械手的運動模型。然后，研究它的運動學(xué)仿真和路徑規(guī)劃。這些工作被用于基礎(chǔ)的操縱器的遙控系統(tǒng)的設(shè)計。</p><p>　　2.1

78、機械手的運動模型</p><p>　　機械手就被視為開環(huán)運動鏈。它由五個旋轉(zhuǎn)接頭構(gòu)成。其一端被固定在一個基座上，同時另一端被用來實現(xiàn)斂的能力。因此，最好是建立一個鏈坐標幀，如圖1所示。該終端的位置和姿態(tài)是通過使用正向運動學(xué)方程知道每個關(guān)節(jié)的旋轉(zhuǎn)角度后確定。如1所示的DH參數(shù)表，建立了使用框架如圖1所示。</p><p><b>　　圖1坐標機械手的幀</b></

79、p><p>　　表1機器人手的D-h參數(shù)值</p><p><b>　　由D-H方法得：</b></p><p>　　當 , , , S，每個關(guān)節(jié)的變換矩陣是由方程（2）給出。</p><p><b>　　(2)</b></p><p>　　當單位矢量在方程（2）為, , ,

80、 ，機械手臂參數(shù)被賦予，，，</p><p>　　因此，正向運動方程通過取在公式（3）每一個參數(shù)來確定。</p><p><b>　　(3)</b></p><p>　　在實際應(yīng)用中，機械手要求能抓住物體。這要求不斷該修正位置給出終端給目標位置。這是機械手的逆運動學(xué)分析。逆變換被用來確定朝著既定的坐標的每個旋轉(zhuǎn)接頭的角，并且逆變換的使用的方法是

81、為了解決這樣的問題（該方法也被稱為代數(shù)方法）的常用方法。使用逆變換分別向左側(cè)乘以，每個旋轉(zhuǎn)接頭的角被確定，通過這些結(jié)果，旋轉(zhuǎn)角度在機械手的末端位置被完全的目標位置決定，角θ4被用于改變機械手的終端的姿態(tài)，并改變由已知的法線向量。然而，傾斜的θ5，由目標對象的大小來決定。</p><p>　　2.2 機械手的運動仿真</p><p>　　建立機械手模型，并通過MATLAB仿真工具箱。我們可以

82、驗證的數(shù)學(xué)模型的合理性。同時MATLAB模型建立了表1和如圖2所示</p><p>　　圖2 MATLAB仿真機械手</p><p>　　比較于圖1和2所示，機械手的仿真模型與參照幀的模型相吻合。這是說，在給定的坐標系是正確的。這些結(jié)果還可以通過經(jīng)由MATLAB所確定的逆運動學(xué)方程表中的（2）和表（3）所示來證明。</p><p>　　表（2）正向運動學(xué)分析<

83、/p><p>　　表（3）反向運動學(xué)分析</p><p>　　目標位置由正向運動學(xué)解決。在此之后，在旋轉(zhuǎn)角度的計算采用逆運動學(xué)方程。事實證明，這些旋轉(zhuǎn)角度正好給定的角度。因此，這些結(jié)果驗證了正向和反向運動的正確性。</p><p>　　3. 機械手的路徑規(guī)劃</p><p>　　聯(lián)合的總量進行了計算和反向運動學(xué)方程當機器人移動到新的位置。因此，機

84、械手可以移動到新的位置，但是機械手最后移動到預(yù)期位置在這樣的條件下，這兩個點之間的機械手的運動是未知的。由于篇幅的限制，運動和一定的位置要求，機械手往往無法移動的上述方法。因此，該運動路徑被設(shè)計為在有限的條件下重合。</p><p>　　在本文中，我們可以利用這些一定的局限性來決定了一些預(yù)期的點。這些預(yù)期的點被用于匹配操作者的運動的規(guī)劃路徑。由于規(guī)劃路徑，在每一個部分的坐標可以被計算出來。每個關(guān)節(jié)的旋轉(zhuǎn)角度是通過

85、逆運動學(xué)方程計算和這些角度實現(xiàn)規(guī)劃路徑的運動。機械手的運動是如圖3所示（其中'o'代表的點由機械手傳遞;'*'代表每一段的預(yù)期點;'-'代表機械手的路徑規(guī)劃）。在圖3中，我們可以看到，手的動作通過每一個規(guī)劃點和移動路徑相吻合的規(guī)劃路徑。</p><p>　　圖3路徑規(guī)劃仿真機械手</p><p>　　4 機械手的遠程控制系統(tǒng)</p&g

86、t;<p>　　機械手的遠程控制系統(tǒng)包含有主機PC和使用ALTER公司董事會DE2從屬的FPGA控制器。操縱器的馬達是由多路徑的PWM波控制。并且PWM波是由IP內(nèi)核生成的。該FPGA控制器通過無線串口通訊PC。而在PC的交互中，操作者可觀察到的操縱器的實時和移動遠程控制的操縱器的運動。同時機械手的每一個動作可以事先通過模擬技術(shù)進行觀察。機械手遠程控制系統(tǒng)的總體設(shè)計</p><p>　　圖4的遠程控

87、制系統(tǒng)的框圖</p><p>　　4.1 機械手的控制模式</p><p>　　有兩種控制模式的控制器。一種模式是，逆運動學(xué)方程計算由FPGA直線，以確定每一個旋轉(zhuǎn)接頭的角度。因此，機械手的控制能實現(xiàn)。這種模式的優(yōu)點是更直接和獨立完成的操縱器的控制，而無需外部設(shè)備。與此同時，這種模式有大量的計算，其占據(jù)更多的內(nèi)部存儲和FPGA的運行時間。 FPGA的資源在此模式下浪費。</p>

88、<p>　　另一種模式通過PC使用的VC和MATLAB實現(xiàn)了機械手的控制。用VC和MATLAB完成了大量復(fù)雜的計算，每個旋轉(zhuǎn)接頭的決心角度。角度的結(jié)果傳送到FPGA中以實現(xiàn)機械手的控制。這種方式節(jié)約了大量的內(nèi)部存儲和運行時間。此外，F(xiàn)PGA可以在此模式下完成其他工作。但機械手不是在該模式下快速控制。</p><p>　　在這個系統(tǒng)中，在機械手遠程控制系統(tǒng)中，一種依賴于兩種模式的優(yōu)點的新模式被采用。具

89、體地，當機械手完成的指定和反復(fù)運動，前者模式通過從屬的FPGA控制器實現(xiàn)控制。當機械手想要實現(xiàn)新的提案 ，后一種模式被訂單者從個人PC指揮。這種新模式是取得了良好的使用在以上兩種模式的優(yōu)點。這種新的模式減輕了計算負擔(dān)，提高了機械手的工作效率。</p><p>　　4.2 SOPC設(shè)計的遠程控制系統(tǒng)</p><p>　　機械手的運動是通過伺服控制的，而伺服機構(gòu)是由PWM波以20毫秒的周期來控

90、制。這些PWM波的脈沖寬度是0.5?2.5ms，對應(yīng)伺服的旋轉(zhuǎn)角度-90度到90度。由通過的Verilog IP核在本系統(tǒng)中生成PWM波的高精度。該結(jié)果示于圖5。 PWM波通過伺服驅(qū)動器控制伺服電機的旋轉(zhuǎn)角度。</p><p>　　圖5的PWM IP核</p><p>　　IP核的多個能夠被下載到FPGA。并以高精度多路PWM波在輸出生成。如圖6所示，這些波的脈沖寬度將由Nios II的方

91、案來解決。操縱器的運動是更靈活的，并且在該系統(tǒng)中更高的精度。</p><p>　　圖6的IP內(nèi)核生成PWM波</p><p>　　機械手的運動是由PWM波的占空比來實現(xiàn)。式（4）倒轉(zhuǎn)角至PWM波的占空比的相應(yīng)量。這種PWM波的占空比對應(yīng)于Nios II的輸出。</p><p>　　9600波特率無線串行被用來傳送坐標，并從主計算機的角度信息的FPGA。在這之后，數(shù)據(jù)

92、和命令由FPGA進行分析。然后FPGA發(fā)送的運動的結(jié)果到主計算機的交互式接口經(jīng)由無線過渡模式。這種通信是通過增加UART通信協(xié)議FPGA實現(xiàn)。</p><p>　　4.3遠程控制系統(tǒng)的交互界面</p><p>　　遠程控制系統(tǒng)的人機交互界面示于圖7。還有，在人機交互界面的一些功能：視頻觀察，操縱控制與仿真建模。</p><p>　　首先，機械手的視頻可以從相機傳到人

93、機交互界面看到。操作人員可以監(jiān)測實時操縱。</p><p>　　其次，角度和坐標可以在交互界面的控制區(qū)進行設(shè)置。機械手的角度可以被獨立地設(shè)置到每個單關(guān)節(jié)。另外，該角度設(shè)定可顯示在實時交互界面的列表（如圖7所示）。在該組坐標，判斷坐標設(shè)定保證，總坐標能夠達到目標點。因此，機械手可以被控制以在結(jié)算路徑移動取決于所述角度信息。</p><p>　　最后，MATLAB機器人工具箱被嵌入到這個互動界

94、面。一個接口被集成在控制和模擬操盤。 MATLAB機器人工具箱是直接使用在機器人造型的人機交互界面。每個組的信息是為了檢測各個動作是否是正確的單獨的模擬。和一般的模擬可以測試機器人的運動安排是否合理。結(jié)合多種模擬方法所做的運動安排更加靈活，在操縱簡單，界面交互的操作更加完美。</p><p>　　圖7操縱的人機交互界面</p><p><b>　　5．實驗與仿真</b>

95、;</p><p>　　為了驗證操作者的遠程控制系統(tǒng)的性能，該系統(tǒng)的實驗正在進行，并進行比較，以模擬系統(tǒng)。具體地，機械手的建模由交互式接口內(nèi)置和一組坐標可以設(shè)計。這些坐標被傳輸?shù)紽PGA，從而控制舵機來完成機器人的運動。關(guān)節(jié)角度，末端坐標通過接口的視頻拍攝。</p><p><b>　　圖8實驗和仿真</b></p><p><b>

96、　　6．總結(jié)</b></p><p>　　在實驗中，將5自由度機械手模型利用MATLAB模擬。利用MATLAB。在從屬FPGA板，機械手的控制是基于Verilog語言經(jīng)由IP核來完成的。這大大降低了外圍電路的設(shè)計，降低成本，提高了精確度和取得的運動更平滑無晃動。而在人機交互界面，VC和MATLAB的混合編程方法嵌入到MATLAB仿真功能</p><p>　　因此，這個操縱器的可

97、靠性得到加強。該系統(tǒng)具有人機交互界面的良好能力。整個系統(tǒng)進行了驗證，并取得所預(yù)期的效果。在這個系統(tǒng)中的一個新的東西體現(xiàn)在人機交互界面的MATLAB機器人工具箱上，對于D-H模型，路徑規(guī)劃和遠程操作等方面進行了直接使用這個交互界面完成。相比于其他的開發(fā)工具，這個互動接口的便攜性，良好的兼容性，開發(fā)周期短，操作簡單。</p><p><b>　　參考文獻</b></p><p

98、>　　[1] SaeeB ，Niku.機器人介紹[M].孫富春,朱繼宏,劉成國棟翻譯.北京:電子工業(yè)出版社,2004(1):60 - 63132 - 137。</p><p>　　[2] Brady,M.J.M.Hollerbach,T.L Johnson.,T.Lozano-Perez和M.T.Mason M.T.機器人運動、計劃和控制[M].美國：麻省理工學(xué)院(劍橋)出版,1982年。</p>

99、;<p>　　[3] Paul Richard P.機器人機械手、數(shù)學(xué)、編程和控制[M].美國：麻省理工學(xué)院出版社，1981年。</p><p>　　[4]李健.多自由度機器人的設(shè)計和研究[D].合肥：中國科學(xué)技術(shù)大學(xué), 2009.6。</p><p>　　[5]程立艷，費凌，蘇澤郎.基于MATLAB五自由度機械手運動學(xué)仿真分析[D].成都：西華大學(xué)機械工程及自動化學(xué)院,20