The University of California at Riverside Department
Contents
1. 33 i SS Lt a a Appendix A Parts List Software Microsoft Visual Basic 6 0 VisionBlox Darl Hardware Seiko D TRAN RT3200 Camera Hitachi KP M2 Personal Computer Imagenation PX610 frame grabber RS232 serial port BNC video port 15 Appendix B PX610 DLL Library Code 16 Declarations for the PXDVNT DLL library Declare Sub pxPaintDisplay Lib pxdvnt dll ByVal hDC As Long ByVal frm As Long ByVal x As Lon g ByVal y As Long ByVal dx As Long ByVal dy As Long De are Sub pxSetWindowSize Lib pxdvnt dll ByVal x As Long ByVal y As Long ByVal dx As Long al dy As Long Global fgh As Long Global frh As Long Global counter As Long Sub displayit im x As Integer y As Integer im dx As Integer dy As Integer 640 amp Controller HScrolll Value y 1008 dy 480 11 5 111 Value 100 amp 640 amp dx Livelmage HScrolll Value 1006 480 amp dy LiveImage VScrolll Value 100 amp Call pxPaintDisplay Livelmage hDC frh x y dx dy End Sub 08 08 01 7 8 440640 827 WPX5 NT BAS 69 6rz tt tg ot 0 205 04272444 00414121411332444949114240988 slobal jlobal jlobal slobal plobal jlobal lobal slobal jlobal slLobal slobal slobal slobal 541 jilobal ilobal ilobal ilobal ilobal2. Pts Jeclare Sub PutRectangle Lib wpx5 nt dll buf As Byte ByVal frh As Long ByVal x0 As Long ByV 1 yO As Long ByVal dx As Long ByVal dy Long Sub GetRow Lib 5 nt dll ByVal frh As Long buf As Byte ByVal row As Long Jeclare Sub PutRow Lib wpx5 nt dll buf As Byte ByVal frh As Long ByVal row As Long Sub GetColumn Lib wpx5 nt dll ByVal frh As Long buf As Byte ByVal row As Long leclare Sub PutColumn Lib wpx5 nt dll buf As Byte ByVal frh As Long ByVal row As Long eclare Function ReadBin Lib wpx5 nt dll Alias ReadBinA ByVal frh As Long ByVal fname As St Ping As Long 1 Function WriteBin Lib wpx5 nt dll Alias WriteBinA ByVal frh As Long ByVal fname As string ByVal overwrite As Long As Long Jeclare Function ReadBMP Lib wpx5 nt dll Alias ReadBMPA ByVal frh As Long ByVal fname As St As Long clare Function WriteBMP Lib wpx5 nt dll Alias WriteBMPA ByVal frh As Long ByVal fname As tring ByVal overwrite As Long As Long Digital I O functions strobes and syncs leclare Function FireStrobe Lib wpx5 411 ByVal fgh As Long ByVal command As Long As Long 1 Function GetStrobeState Lib 5 nt dll ByVal fgh As Long As Long Function GetSyncType Lib wpx5 nt dll ByVal fgh As Long As Long y are Function SetDrivePolarity Lib wpx5 nt dll ByVal fgh As Long ByVal vdrive As Long ByV drive As Long As
3. Composition Camera head with IR cut filter C mount cap operation manual Seiko Robot RS 232 serial cable pin outs PC Seiko Controller 9 Pin F 25 Pin M 2 3 3 2 7 1 Camera mount design 5500 gt 12 cm 4 5cm References 1 DARLfour Programming Guide Seiko Instruments Torrance CA 1987 2 DARL Talk Version 2 02A Seiko Instruments Torrance CA 1995 3 Imaging Components VisionBlox Imaging Software Vision 1 http www vision com software vbloxdesc shtml current June 7 2001 4 KPM 2 Camera Premier Electronics http www premierelect co uk kpm2 html current June 7 2001 5 Seiko RT3200 Installation and Operation Manual ed Seiko Instruments Torrance CA 1987 6 VisionBlox Training Manual Version 3 0 Integral Vision Inc Farmington Hills MI 1998 7 VisionBlox Reference Manual Version 3 0 Integral Vision Inc Farmington Hills MI 1998 8 G Perry Teach Yourself Visual Basic 5 in 24 Hours 1 ed Indianapolis Indiana Sams Publishing 1997 9 J J Craig Introduction to Robotics 2 ed Reading Massachusetts Addison Wesley Publishing Company Inc 1989 10 H M Deitel P J Deitel and T R Nieto Visual Basic 6 How to Program New Jersey Prentice Hall 1999 11 Bertell Nguyen and Wassink Final Report of Senior Design Project Winter Spring
4. 2001 2001 36 Acknowledgement We would like to thank Dan Giles for the all the help and guidance he has assisted us with all throughout We also like to thank Dr Gerardo Beni for keeping us on track with our report by making us send him weekly reports Jojo thanks for helping us prepare for the presentation it really helped us from getting flustered when we presented in front of everyone Steve thanks for lending us the Visual Basic book Also thanks to Joseph for providing the us with the video camera so we can tape our demonstration and being our cameraman too 37
5. CommPort 1 Settings 9600 N 8 1 Use 1 MSComml CommPort CommPort 9600 baud no parity 8 data and 1 stop bit MSComml Settings Settings Handshaking MSCommil Handshaking comRTS ommands 4 Open Port If MSComml PortOpen False Then MSComm1 PortOpen True nd If rivate Sub DisplayResultInfo Displays the results of when using the Use button to the screen If mintResultNumber 0 Then Commands XValue Text Commands YValue Text Commands ResultLabel Enabled False Commands FrameCoordinates Enabled False Else Commands ResultLabel Enabled True Commands FrameCoordinates Enabled True Commands FeatureFindTool ResultX mintResultNumber Commands FeatureFindTool ResultY mintResultNumber Commands XValue Text Commands YValue Text Point at the origin of the area where the feature was found when the Use method was last invoked Commands Textl Commands FeatureFindTool ResultPoint mintResultNumber If Sub rivate Sub cal As Integer Output calibration command HOME Calibrates the robot HOME vbCrLf MSComm1 Output MSComm1 Output Call delay 2 rivate Sub Exit Click Index As Integer Close Port If MSComml PortOpen lt gt False Then MSComml PortOpen False End If Exit Program End sub 1 As Double y As Double z As Double R As Double Dim buffer As String Dim destination As String manuall
6. EE counter x 0 Else keep increment until find a object counter x counter x 10 MSComm1 Output DO OUTPUT OG7 amp vbCrLf Call delay 40 Commands Textl counter x move x 300 move x2 move x counter x DO T10s amp move x2 amp 0 0 0 MSComml Output B MSComml Output vbCrLf Call delay 5 MSComm1 Output Call delay 5 DO MOVE T10 amp vbCrLf End If Call delay 15 End Sub lic Sub GetImage Not used Dim tmp As Integer tmp Grab fgh frh O Call WriteBMP frh c ee175AB temp bmp 1 ommands End Sub rivate Sub RunProg_Click Index As Integer Automation of the vision system utilizes the pick function Dim Do temp As Integer loop until object is found setting live image to abc bmp Call Controller play_Click Livelmage Timerl Interval 0 temp Grab fgh frh 0 Call WriteBMP frh h SeniorProject LastYear abc bmp 1 Commands EdibleImage Picture h SeniorProject LastYear abc bmp saves abc bmp info to abc ini file Commands FeatureFindTool FeatureFileName h SeniorProject LastYear abc Dat Commands edshpTrain save h SeniorProject LastYear abc INI TrainingShape Commands edshpInspection save h SeniorProject LastYear abc INI InspectionShape Commands FeatureFindTool save h SeniorProject LastYear abc INI FeaturefindTool load to be displayed and use Commands edshpTrain Load h SeniorProject LastYear a
7. can take our design Due to last year s project we were limited to using Visual Basic as our programming language which links the controller of the Seiko robot to the vision system We were also constrained to the VisionBlox software to perform our image processing The camera frame grabber card and the serial port were all hardware parts that we were limited to use for our project Another constraint to our project is the end effector that was provided by the Electrical Engineering Department of UCR The end effector provided is very small and have no real life applicability We will not be able get a new end effector before the end of the 2002 Spring quarter the demo for our project will be severely limited 5 3 Elements of Design The uses of robotics in the industrial field have long been an accepted and normal practice of the industry There are many benefits of using robotics for production such as safety productive reliable and relatively cheap compared to human labor For example when working with dangerous chemicals or in unsafe environments robot can be used to get the job done without endangering people s lives Robots are also very productive for it can perform specified task at a very high speed without ever having to take a break In most cases machines will outperform human labor drastically in any mass production With the advancement of robotics and technology the robots used today in industry are extremely reliable and our
8. experiences working with the Seiko RT3200 agrees with that Although purchasing an industrial robot might seem quite expensive once the robot is purchased the cost to maintain and reuse is relatively minimal So in the long run the use of robots for industrial applications will save companies tremendously in cost There will always be critics that will oppose the use of robots for the fear of losing all jobs to machines It is true that machinery and robotics have taking over many of the production that used to be done by people but by using robotics it has also made many things much more cheaper and available So until something major event occurs machinery and robots will continue to be integrated into our everyday lives and continue to grow 10 Chapter 6 Conclusions 6 1 Introduction The problem statement of this project is to upgrade the Seiko RT3200 so that it can utilize vision by incorporating a Visual Basic interface using the software VisionBlox to implement with the Seiko robot arm The robot will be able to recognize trained objects and perform a particular task with the object We have succeeded in controlling the robot by using the Visual Basic user front and the vision system can recognize a trained image we specified Using a trained image we are able to get the Seiko robot to scan a particular workspace for the trained object Then once the trained object is found we are able to allocate to the robot the location of the obj
9. for the Electrical Engineering Department at the University of California Riverside For two quarters every undergraduate student in the Electrical Engineering major at the University of California Riverside must present a design project to the Electrical Engineering Department prior to graduation The undergraduate student has the freedom to select any design project as he or she pleases but with the approval of the advising professors For our senior design project we decided to continue the Machine Vision Interface with Seiko Robot Arm project started by prior undergraduate students at University of California Riverside The main objective is to develop a machine vision interface for the Seiko RT3200 Robot Arm in one of the UCR s Electrical Engineering Department Laboratory The Seiko RT3200 Robot Arm is an industrial robot that can be used for many applications and by adding a vision system to the robot will greatly improve its capabilities and value Also once the vision system is complete it can be marketed and implemented for other Seiko RT3200 Robot Arm that needs a vision system Section 1 2 talks about the Historic Background of the robotic industry and section 1 3 contains a Glossary of Acronyms and Abbreviations important to this report 1 2 Historic Background The Seiko RT3200 Robot arm is an industrial robot and it is of the cylindrical type It is a hybrid mix that incorporates the rotary motion of the SCARA and the linear mot
10. friendly and its capabilities have increased tremendously An updated robot was developed by Seiko model D TRAN RT330 that has the vision feature which shows the importance of our project and the real life applicability of a vision system for the RT3200 robot arm Keywords robotics vision system serial port communications automation object detection Table of Contents Chapter 1 Introduction I 1 Introduction 1 2 Historic Background 1 3 A Glossary of Acronyms and Abbreviations Chapter 2 Design and Technical Formulation 2 1 Introduction 2 2 Problem Statement 2 3 Design Specification 2 4 Design Approach 2 5 Alternative Approaches Chapter 3 Design Evaluation 3 1 Introduction 3 2 Test Plan and Result 3 3 Design Comparison Design Trade Off Chapter 4 Administration 4 1 Introduction 4 2 Budget Cost Analysis 4 3 Marketability Chapter 5 Meeting Expectations 5 1 Introduction 5 2 Design Constraints 5 3 Elements of Design Chapter 6 Conclusions 6 1 Introduction 6 2 Design Expansion Improvement Suggestions 6 3 User s Manual Appendix A Parts List Appendix PX610 DLL Library Code Appendix C Visual Basic Form Codes Livelmage Controller Commands Appendix D Technical Specifications References Acknowledgement Page CRA A t9 dN bo Chapter 1 Introduction 1 1 Introduction The driving force for this design is to fulfill the senior design course requirements
11. rlobal ilobal lobal lobal lobal Tobal 1 1 lobal lobal lobal lobal lobal lobal pu 1 lobal lobal lobal lobal Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Const Flags GRAB OFF FIELDS amp H3 EITHER amp H3 GRAB ON amp H3 FIELDO amp 1 FIELD1 amp H2 IMMEDIATE amp H4 QUEUED amp H8 SINGLE_FLD amp H10 CACHE amp H20 Trigger Types RISING FALLING HIGH amp 1 LOW amp HO DEBOUNCE amp H4 amp H3 amp H2 Strobe Types STROBE 0 amp H1 STROBE_1 amp H2 STROBE_OFF amp H4 STROBING amp H8 STROBE_GAP amp H10 STROBE_STOP amp H11 STROBE_TRIG amp H12 STROBE_NORMAL amp H13 SYNC_NORMAL amp H13 SYNC_OFF amp H4 Error Numbers ERR NONE ERR VALID amp H1 ERR CORRUPT amp H2 ERR PCI amp H4 ERR BAD IRQ amp H8 ERR NO ADDRESS amp H10 ERR NO DEVICES amp H20 ERR GRAB INVALID amp H86 File I O Return Values SUCCESS 0 FILE_EXISTS 2 FILE_OPEN_ERROR BAD_WRITE 4 BAD_READ 5 WRONG_BITS 8 BAD_FILE 11 INVALID_FRAME 12 Sync Types USER_SYNC amp H1 INTERNAL SYNC WEN_SYNC amp H4 SINGLE
12. system to be used by the Seiko robot Section 2 2 will contains the Problem Statement and section 2 3 is the Design Specifications In sections 2 4 and 2 5 the focus is on the Design Approach and the Alternative Approaches to the project 2 2 Problem Statement The goal of this project is to upgrade the Seiko RT3200 so that it utilizes a third party vision system The problem includes creating vision software that can incorporate the already existing VisionBlox software with the Seiko robot arm In addition the vision system needs to be able to recognize some trained objects and perform particular task with the object Moreover the vision system needs to be able demonstrate some automation procedure with the robot arm 2 3 Design Specifications The specification for our design requires the vision system to be able to capture and analyze an image in real time The robot also needs to be able to be controlled through a personal computer Using the PC a vision system must be incorporated into the Seiko RT3200 robot The vision system that needs to be created have to be able to recognize an object get the coordinates of the found object and then send the coordinates of the object to the robot so it can interact with it Automation between the vision system and the Seiko robot arm is desirable feature of the design project Camera Darl Driver Visual Basic Seiko Robot VisionBlox Fig 2 2 Technical Design 2 4 Design Appro
13. Final Report of Senior Design Project Winter Spring 2002 The University of California at Riverside Department of Electrical Engineering Machine Vision Interface With Seiko Robot Arm II Prepared by Zhi Xu and Ghazi Ting Technical Faculty Advisors Mr Dan Giles Principal Electronics Technician Dr Gerardo Beni Electrical Engineering Faculty of the University of California at Riverside This project is support by the Electrical Engineering Department of the University of California at Riverside Submitted on June 7 2002 at 5 00 PM Executive Summary The objective of this project is to create a third party vision system to be utilized by the Seiko D TRAN RT3200 robot arm that belongs to the Electrical Engineering Department of UCR Working with the Visual Basic 6 0 platform started by last year s senior design group the industrial image processing software VisionBlox is used to design a vision system capable of live image A user friendly interface has been developed to integrate the Seiko robot with the vision system so the two components can operate together In addition with the interface users can control the robot and use the vision system without any prior knowledge of DARL commands or VisionBlox With the aid of the vision system the arm is able to perform regular industrial type task such as performance check object detection and objection recognition automatically The developed vision system has made the robot much more user
14. Long 1 Function SetDriveType Lib wpx5 nt dll ByVal fgh As Long ByVal mode As Long As Long eclare Function SetStrobePeriods Lib wpx5 nt dll ByVal fgh As Long ByVal ti As Long ByVal t As Long ByVal t3 As Long As Long 1 Function SetStrobeType Lib wpx5 nt dll ByVal fgh As Long ByVal mode As Long As Long Appendix C Visual Basic Form Codes ivelmage 1 ivate Sub Form Load Dim x As Integer Dim y As Integer Fixed Window Size This requires moving the sciollbabs Livelmage ScaleMode 3 VScrolli Top O VScrolll Left 640 VScrolll Height 512 HScrolll Top 512 HScrolll Left 0 HScrolll Width 640 Call pxSetWindowSize 0 0 640 512 We calculate the new window size in twips because Screen coordinates are in twips Livelmage ScaleMode 1 VScrolll Left VScrolll Width y HScrolll Top HScrolll Height account for the window border size X X Livelmage Width Livelmage ScaleWidth LiveImage Livelmage ScaleHeight Livelmage Width LiveImage Height ge Livelmage Timeri1 Interval E 10 Sub Private Sub Form Paint Call displayit md Sub ate Sub 5 111 Change Call displayit Sub rivate Sub Timerl Timer counter counter 1 tmp Grab fgh frh 0 If tmp 0 Then Debug Print Grab failed End If Call displayit ind Sub rivate Sub 5 111 Change Call displayit md Sub ontroller 1 rivate Sub Co
15. Sample INI Inspections lape Y Commands FeatureFindTool Refresh Sub rivate Sub cmdMOVE Click wCommands FeatureFindTool Load h SeniorProject LastYear FeatureFindSample INI Featurefind Moves robot to one location right now for testing purposes Need to be incorporated later allow user a list of coordinate choices or allow user to input location MSComml Output DO SPEED 50 vbCrLf Call Move1 380 170 120 O nd Sub rivate Sub cmdPICK Click Call pick True nd Sub rivate Sub cmdSave Click Saves the current trained image to a file Filename imagel Dat is used at the default name Commands CommonDialog2 DialogTitle File Save Commands CommonDialog2 Filter Dat Commands CommonDialog2 FileName imagel Dat Commands CommonDialog2 Action 2 If Commands CommonDialog2 FileName lt gt Then screen MousePointer 11 ommands If WriteBMP frh Commands CommonDialog2 FileName 1 Then Screen MousePointer 0 MsgBox Could Not Write File 48 Error End If Screen MousePointer 0 If saves image and feature find information Commands FeatureFindTool FeatureFileName Commands CommonDialog2 FileName Commands edshpTrain save h SeniorProject LastYear FeatureFindSample INI TrainingShape Commands edshpInspection save h SeniorProject LastYear FeatureFindSample INI Inspections Commands FeatureFindTool save h SeniorProject LastYear Fea
16. Sync system Internal or external automatic switching External sync input HV VD 2 to 6 volts p p input impedance Ikohm frequency deviation 1 negative polarity Resolution 570 H 485 V TV lines Standard sensitivity 400 lux f4 3200K Minimum sensitivity 0 5 lux at f1 4 with AGC on gamma 0 45 without IR cut filter Electronic Shutter off 1 100th 1 125th 1 250th 1 500th 1 1 000th 1 2 000th 1 4 000th 1 10 000th selectable by external switch Gamma correction 0 45 or 1 0 selectable by internal switch Signal to noise ratio 56dB AGC Normal or AGC selectable by internal switch 33 Video output VS 1 0 volt p p 75 ohms unbalanced 0 7 volts p p video and 0 3 volts p p sync negative polarity Integration mode Field or frame integration selectable by external switch Field on demand function Switchable on off by internal switch Restart reset operation On Off selectable by component change Lens mount C e Flange focal distance 7 526 mm Ambient temperature and humidity Operating 10 C to 60 C RH 90 or less Storage 20 C to 60 C RH 70 or less Resistance to vibration 9G max 10 to 60Hz amplitude constant at 0 98mm 7G constant 60 to 150Hz amplitude variable 10 to 150Hz sweeping for one minute each of X Y and Z directions for 30 minutes Dimensions 44 W x 29 H x 72 D mm Weight 120 grams approx Supply voltage 12 volts dc 1 volt Current consumption 180 mA approx
17. _FIELD_SYNC AUTOMATIC SYNC amp H2 amp H10 80 Configuration Values PXC_CACHE amp H1 PXC_NONINTERLACE amp H2 31081 Const PXC_BUS Slobal Const amp HO 3lobal Const PXC 104 PLUS amp H40 31obal Const PXC COMPACT PCI amp H80 3 1 Const VIDEO DRIVE amp H100 l Const PXC H CROP amp H100 Const PXC V SCALE amp H100 3lobal Const PXC V CROP amp H800 1 Const PXC SCALE amp H800 Jlobal Const PXC STROBES amp H200 3lobal Const PXC WEN SYNC amp H400 3lobal Const PXC CUSTOM HW 000 Setup Functions Jeclare Function InitLibrary Lib wpx5 nt dll As Long Declare Sub ExitLibrary Lib wpx5_nt dll Frame Data Functions Jeclare Function AllocateBuffer Lib wpx5 nt dll ByVal dx As Long ByVal dy As Long ByVal bits As Long As Long Jeclare Function AllocateAddress Lib wpx5 nt dll ByVal address As Long ByVal dx As Long L dy As Long ByVal bits As Long As Long Jeclare Sub FreeFrame Lib wpx5 nt dll ByVal frh As Long Jeclare Function FrameWidth Lib wpx5 411 ByVal frh As Long As Long Jeclare Function FrameHeight Lib wpx5 nt dll ByVal frh As Long As Long Jeclare Function FrameBits Lib wpx5 nt dll ByVal frh Long As Long Jeclare Function FrameBuffer Lib 5 nt dll ByVal frh As Long As Long FrameAddress NOT supported on NT ByVa Frame Grabber Allocation Functions 1 F
18. ach The approach taken to fulfill the specifications of our senior design project is to continue with the development of last year s senior design group In the preceding year a group of seniors in the University of California Riverside Electrical Program started the implementation of vision for the Seiko robot arm They were able to utilize Visual Basic 6 0 to incorporate the vision software VisionBlox and the Seiko robot language Darl to work together as one entity With parts of the machine vision interface already developed we needed only to continue the development by upgrading the capabilities of the vision system creating a more user friendly interface and making the Seiko D TRAN RT3200 Robot Arm have a totally automated vision system 2 5 Alternative Approaches Another possible approach could have been to start the project fresh and not use the codes provided by last year s group The advantage would be that we will have total control of the decisions of the project and any errors or issues that arises we would be able to answer The drawback to this approach is that we would be reproducing many of the same information that is already available and the difference from continuing from last year s end point versus starting anew might be minimal Other possible alternative approaches can be found in section 3 3 of the report Chapter 3 Design Evaluation 3 1 Introduction The most important measure of the effectiveness or how go
19. ame Grabber Info Functions Jeclare Function ReadRevision Lib wpx5 nt dll ByVal fgh As Long As Long Jeclare Function ReadProtection Lib wpx5_nt dll ByVal fgh As Long As Long Jeclare Function HaveCache Lib wpx5 nt dll ByVal fgh As Long As Long declare Function VideoType Lib wpx5 nt dll ByVal fgh As Long As Long Jeclare Function CheckGreater Lib wpx5 nt dll ByVal fgh As Long As Long clare Function CheckEqual Lib wpx5 nt dll ByVal fgh As Long As Long Jeclare Function GetTriggerType Lib wpx5 nt dll ByVal fgh As Long As Long Jeclare Function GetFineGain Lib wpx5 nt dll ByVal fgh As Long As Long Jeclare Function GetGainRange Lib wpx5 nt dll ByVal fgh As Long As Long Jeclare Function GetOffset Lib wpx5 nt dll ByVal fgh As Long As Long Jeclare Function GetCamera Lib wpx5 nt dll ByVal fgh As Long As Long Declare Function GetLUT Lib wpx5 nt dll ByVal fgh As Long ByVal first address As Long ByVal Length As Long lutarray As Long As Long re Function CheckError Lib wpx5 nt dll ByVal fgh As Long As Long are Function GetFieldCount Lib wpx5_nt dll ByVal fgh As Long As Long Jeclare Function GetFieldLength Lib wpx5 nt dll ByVal fgh As Long As Long Jeclare Function ReadConfiguration Lib wpx5 nt dll ByVal fgh As Long As Long Data Manipulation Functions eclare Sub GetRectangle Lib wpx5 nt dll ByVal frh As Long buf As Byte ByVal x0 As Long ByV l yO As Long ByVal dx As Long ByVal dy As Long
20. bec INI TrainingShape Commands edshpInspection Load h SeniorProject LastYear abc INI InspectionShape Commands FeatureFindTool Load h SeniorProject LastYear abc INI FeaturefindTool Commands FeatureFindTool Refresh Commands EdibleImage Refresh Controller Refresh Livelmage Refresh check if trained image is found by calling the use function Call commandUse Click Call pick Commands FeatureFindTool Found Call delay 20 setting the edible image to abc bmp loop counter flag 1 Commands WindowState 1 loop counter loop counter 1 Commands CalYValue Text loop counter Loop While Commands FeatureFindTool Found False And loop counter 10 reseting variable after loop counter x 0 loop counter 0 Commands WindowState O Appendix D Technical Specifications KP M2U Specification EIA Sensor pick up device 1 2 KP M2 Interline transfer CCD 1 3 KP M3 interline transfer CCD Sensing scanning area 6 45 H x 4 84 V mm KP M2 4 88 H x 3 09 V mm KP M3 Total number of pixels 811 H x 508 V Number of effective pixels 768 H x 494 V Pixel pitch 8 4 H 9 8 V um KP M2 6 35 H x 7 4 V um KP M3 Spectral sensitivity Scanning system 2 1 interlaced 525 lines 60 fields per second 30 frames per second or non interlaced only in external sync mode or with modification Scanning frequency 15 734 kHz H 59 94Hz V Signal standard EIA Monochrome
21. ect and have it pick it up and ali of this can automated Therefore we have accomplished all the goals that we set out to accomplish and the rest will be left to the future seniors that will be working on the improvement of this project 6 2 Design Expansion Improvement Suggestions There are many suggestions that we have for the improvement and expansion of this project in the future The first is to mount the camera in a higher position so that the pictures captured by the camera will be in a better focus As of now the height of the object from the end effector when the image is in focus is far too low and when the end effector tries to pick up a focused object it cannot reach it The alternative to this approach may be making a longer end effector so that the end effector can reach to a lower position and therefore will not need to change the position of the camera The second approach may be better because the calculation of the pixel to world coordinate system would stay relatively the same Important note if the camera and the distance between the platform is changed then the area of the camera relative to the spatial area will also change causing the pixel to world coordinate system calculation to change as well Another suggestion we have for the improvement of this project is to change the end effector on the Seiko RT3200 robot arm to something much more useful The end effector we have now is able to pick up 2 4 mm wide objects o
22. espect to the editable image 12 Loop Counter Use to display the number of scans when in the live mode Found Object Returns true or false that the object is in the editable image Minscore Use to control how similar an object has to be with respect to the train image before it is considered the same object Before performing any motion of the robot for the first time the robot must be calibrated and clicking on the Calibrate button on the user interface does this Calibrating will place the robot to the HOME position 13 ae Lt 5 5 MT mage m in E 025 oem eee Se pe _ _ Pixel Loc i um a Lh ur M cl ee ce a ee iin 8 88 1 rs 1 io T oe a dee A 53 4 ore gt Pi j II rk spe LIE as TT ET JuriptRn i 2 147 E LEE I m nmm E gd c i E T p a 1 4 LI M EEUU PUN E ITE 4 e Wie L3 d 111111111 did Oe 111111 aL 2 4 4 T GNE T E amm
23. ffset Lib wpx5 nt dll ByVal fgh As Long ByVal offset As Long As Long As Long eclare Function SetTriggerType Lib 5 nt dll ByVal fgh As Long ByVal mode As Long As Long Function SetLUT Lib wpx5 nt dll ByVal fgh As Long ByVal first address As Long ByVal ength As Long lutarray As Long As Long ByVal gain As Long ByVal flag ByVal flag Declare Function SetCamera Lib wpx5 nt dll ByVal fgh As Long ByVal camera As Long ByVal flag 5 As Long As Long Declare Sub ResetFG Lib wpx5 nt dll ByVal fgh As Long Declare Function SetImageSize Lib wpx5 nt dll ByVal fgh As Long ByVal resx As Long ByVal res y Long ByVal x0 As Long ByVal y0 As Long ByVal dx As Long ByVal dy As Long ByVal bits As L As Long Declare Function SetCompare Lib wpx5 nt dll ByVal fgh As Long ByVal n As Long As Long Declare Function SetFieldCount Lib wpx5 nt dll ByVal fgh As Long ByVal x As Long As Long SetCurrentWindow NOT supported on NT Declare Function SetStrobePolarity Lib wpx5 nt dll ByVal fgh As Long ByVal s As Long ByVal p As Long As Long declare Function SetFieldSize Lib 5 nt dll ByVal fgh As Long ByVal resx As Long ByVal res As Long ByVal x0 As Long ByVal yO As Long ByVal dx As Long ByVal dy As Long ByVal bits As Long As Long Jeclare Function SetVideoFormat Lib wpx5 nt dll ByVal fgh As Long ByVal field length As Long ByVal blank length As Long ByVal flags As Long As Long Fr
24. gl FileName CMDialogl Action 2 If CMDialogl FileName Then Screen MousePointer 11 If WriteBMP frh CMDialogl FileName 1 Then Screen MousePointer 0 MsgBox Could Not Write File 48 Error End If Screen MousePointer 0 End If Call WriteBMP frh c eel75AB test bmp 1 Sub rivate Sub stop click gt Sie 0 Sub rivate Sub cmdQuit_Click Livelmage Timerl Interval 0 Unload LiveImage Unload Controller Load Commands Commands Visible True Commands WindowState 0 FreeFrame frh FreeFG fgh ExitLibrary nd Sub ption Explicit fining global variable of the form rivate mintResultNumber As Integer rivate counter x As Integer jim loop counter As Integer rivate flag As Integer rivate Sub Commandl Click Call GetImage nd Sub rivate Sub cmdLoad Click Loads a saved bitmap image and sets to the editable image screen When clicked a file must be selected or an error occurs A if statement can be used but we consider this not the most important issue right now so disregard for now Commands CommonDialogl DialogTitle File Open Commands CommonDialogl Filter bmp Commands CommonDialog1 FileName bmp Commands CommonDialog1 ShowOpen load the image information from the FeatureFindSample INI file Commands edshpTrain Load h SeniorProject LastYear FeatureFindSample INI TrainingShape Commands edshpInspection Load h SeniorProject LastYear FeatureFind
25. ined region so if the same object is placed on a different background from the trained image FeatureFindTool Found will likely to return a False Setting the minimum on the form will change the FeatureFindTool MinScore and this can make some difference Test shows that 75 works well for objects with the same object type The test we used is two washers Testing for the MinScore Value Comparing two washers Washer A is the washer in the trained image Original image Washer A True False 0 100 none Washer B True False 0 89 90 100 Image 2 Washer A True False 0 87 88 100 Washer B True False 0 88 89 100 Image 3 Washer A True False 0 84 85 100 Washer C True False smaller washer 0 34 35 100 Training an image The edshpTrain Load and edshpTrain Save is used to load and save an image in the editable shape region of the editable image To use the FeatureFindTool Found function the image must be first trained The edshpInspection save and FeatureFindTool save should always be used whenever the edshpTrain Save command is used The same for the edshpTrain Load command It appears that the CaliperTool commands works the same way as the FeatureFindTool but there were errors when we try to incorporate them into our program mainly when training the CaliperTool Saving and Loading an image Mainly to store and load the image in the editable image and the editable
26. ion of a CARTESIAN There are four axis of movement for the robot x y z a The x axis and the y axis take care of the flat surface of the robot s movement while z axis maps the height of the robot The a axis is rotational movement of the end effector on the robot The RT3200 has a standard cycle time of 0 8 seconds that the robot can Work in a very high speed compare to any other kind of robot DARL is the original image developed by Seiko for the RT3200 model robot which is very similar to the programming language BASIC 1 3 A Glossary of Acronyms and Abbreviations UCR University of California Riverside DARL D Tran Assembly Robot Language Visual Basic Computer programming language that runs in Microsoft Window End effectors The tool at the end of the Seiko RT3200 robot used to perform specific tasks Seiko RT3200 Industrial robot used in the production industry to perform assembly type jobs Editable Image This is the display of image entire image on the user interface Train Image The small region within the editable image use to select a specific object out of the whole image Chapter 2 Design and Technical Formulation 2 1 Introduction The design approaches selected for this project make use of the work completed from last year s senior design group and improve on their work The main part of the design deals with controlling the Seiko D TRAN robot in Visual Basic and incorporating a third party vision
27. mage to a temporary image and compare it with the stored trained image If a trained image is found then the process stops and the Pick function is called At first I thought the loop I created was causing the computer to freeze but after placing a variable to count the loops and setting a limit on how much the Private Sub RunProg loops the freeze seems to come from the infinite loop of the image always returning False 8 0 650 pixel 512 pixel y 658 512 Fig 3 1 Pixel Coordinate Testing movement limitation of the robot arm Home position is 300 0 0 0 X is from 300 to 600 Y is from 277 615 to 277 615 Z is from 120 to 0 A is from 100degree to 100 for the end effector For all movement of the robot we first programmed and tested it in DARL first and then after it works the code is incorporated into the Visual Basic code Testing for delay on the robot First we set the robot at speed 50 while teaching the robot to learn a new position a small delay is needed and we set it to 5 When I try to move the robot arm from one position to another position another delay is needed to allow the robot time to get to the new position The delay was first set to 50 which is too short and the robot sometimes will skip commands So the delay has been set to 100 now it works fine Then we try to set the robot at full speed 300 between teaching the robot to learn a new position the delay was set to 5 When we try
28. mmandi Click Call Commands GetImage nd Sub rivate Sub Form Load If InitLibrary 0 Then Debug Print init fail End End If fgh AllocateFG 1 If fgh 0 Then ExitLibrary Debug Print frame grabber fail End End If frh AllocateBuffer 640 480 8 If frh 0 Then FreeFG fgh ExitLibrary Debug Print buffer fail End End If tmp SetImageSize fgh 640 256 0 4 640 480 8 If tmp 0 Then Debug Print SetImageSize failed End If Livelmage Show Sub ate Sub Form_Unload Cancel As Integer LiveImage Timeri Interval 0 Unload LiveImage FreeFrame frh FreeFG fgh ExitLibrary 1d Sub rivate Sub grange_Click Index As Integer tmp SetGainRange fgh Index 0 nd Sub rivate Sub gslide Change tmp SetFineGain fgh gslide Value 0 nd Sub rivate Sub HScrolli Change Call displayit nd Sub rivate Sub Label4 Click nd Sub rivate Sub offslide Change _ tmp SetOffset fgh offslide Value 6 Sub ivate Sub Open Click stop click CMDialogl FileName CMDialogi Action 1 If CMDialogl FileName lt gt Then ontroller 2 Screen MousePointer 11 If ReadBMP frh CMDialogi FileName Then Screen MousePointer O0 MsgBox Could Not Read File 48 Error End If Screen MousePointer 0 End If LiveImage Refresh nd Sub ublic Sub play Click es Livelmage Timerl Interval 10 counter 0 tl Timer Private Sub save Click stop click CMDialo
29. nality to the existing vision system The addition of edge detection caliper and blob tools of VisionBlox would be great addition to the existing vision system 6 3 User s Manual To start controlling the robot from the computer first you have to type three commands from the teach pendant DO COMM 1 78 DO OUNIT 1 DO IUNIT 1 These inputs instructs the controller to read command lines through the serial port instead of the teach pendant So all commands forth is assumed to be typed using the Visual Basic interface created Buttons Calibrate go to HOME position Load Image Loads a bitmap image onto the editable image screen Save TrainImage Saves the image currently on the edshpTrain of the FeatureFindTool Pick Calls the function that moves the robot to pick up an object Live Mode Performs the scanning and picking up the trained object Display LiveImage Goes to the live image screen and controller display Move Arm Moves the robot to a specified location TrainImage Trains a specific region of an image and the image is used for recognition Use TrainImage This checks if the train image is currently on the editable image display and if so the coordinates of the object is displayed Exit Quits the program Textbox World Loc Displays the location of the object with respect to the robot these are coordinates that the robot uses to pick up the object Pixel Loc The location of the object with r
30. nly We suggest buying a new end effector that can actually be used to do something much more useful so a more impressive demonstration can use to showcase the usefulness of the project Other suggestions that future groups can also consider are working in C instead of Visual Basic incorporation the vision system for the Adaptec robot storing the commands to read code through the serial port and adding more features to the existing vision system The reason for working in the computer language C is because the PX610 driver seems to be more compatible with C and all its sample codes are in C therefore it will be much easier to implement new features in the future Also vision system created for the Seiko robot should be easily incorporated into the Adaptec robot Of course the commands to control the robot and camera location will be different the general structure should be the same One of the additions that we were meaning to add but did not get to is getting rid of the input commands of DO COMM 1 78 DO OUNIT 1 DO IUNIT 1 When going from the control pendant to the Visual Basic program these three commands must be first typed on the control pendant to get the controller to respond to commands sent through the serial port This is a redundant procedure and getting rid of it will please future users who want to send commands to the robot via the serial port Finally this project can be improved also just by adding more features and functio
31. od the design of this project is deals with the compatibility and accuracy between the robot and the vision system Therefore most of the test were focused on the vision system being able to identify and differentiate objects in an image and determine whether or not the trained object is in the image Next how accurate is the robot in going to location of the desired object specified by the vision system From the results of our tests we conclude that are design has achieved the satisfactory level that we desire Section 3 2 is the Test Plan and Result and section 3 3 is the Design Comparison Design Trade Off 3 2 Test Plan and Result e Delay needed when moving robot with VB Sending commands using VB through serial port e Capturing and storing of live image Vision Parameters for comparing trained with live image Pixel to world coordinate system calculation Accuracy of the vision system in real time The repeatability of the robot The capability of the vision system A compatible speed between robot and vision Ability to perform multiple task simultaneously Using the FeatureFindTool in VisionBlox to compare an object Utilized the save command from the Controller form to store an image and then wrote a load function in the Commands form to load the saved images saved Next the FeatureFindTool Found function was used to return the comparison result and displayed in a text box The FeatureFindTool compares everything within the tra
32. ra is rotated by some inconsistent degree which would be difficult to calculate the relationship between the robot movements with respect to the angle change of the camera By not having the robot being able to move sideways limits the workspace in which our robot can work with in our automation mode it simplified our calculation of translating pixel locations into world locations that the Seiko robot can use Chapter 4 Administration 4 1 Introduction Our project is a continuation of an existing project therefore we did not spend any money for this project The Budget Cost Analysis of section 4 2 contains prices of what it would have cost had we had to buy all the parts ourselves Section 4 3 talks about the Marketability of the design we help develop 4 2 Budget Cost Analysis Seiko D TRAN RT3200 10 000 Vision1 Mono kit Hitachi KP M2 Camera Imagenation PX610 PCI Frame grabber Card e VisionBlox software e 12 volt DC Power Supply amp Cables 1 975 Computar C Mount 25mm f1 3 lens 128 Visual Basic 6 0 99 Development cost e 200 hours x 100 per hour 20 000 Total 32 202 Purchasing newer model Seiko RT 3300 60 000 4 3 Marketability The newer model RT3300 which is basically the same robot as the RT3200 model but with a vision system is priced at 60 000 Evaluating our budget cost analysis it shows that we will be able to provide the vision system we developed and offer it at a much lower price Being tha
33. riginal design plan use of the CaliperTool feature in VisionBlox and the workspace used for our automation system Originally before fully understanding the capabilities and functionality of the FeatureFindTool of VisionBlox we were considering adding the CaliperTool component to our vision system We assumed that without some sort of way to separate the foreground from the background there will be no way to detect an object However since we can control the background we forced the background of all images we work with to be always white or black With the ability to control the background then just by using the FeatureFindTool we are able to utilize the features of an object alone to distinguish one object from another Another major change we made to our original design was the workspace for the robot when we automate our vision system Originally we were planning to have a box shaped workspace for our vision system and robot would scan for objects within that workspace To do this would require some extremely difficult algorithms to compute the translation from the pixel coordinate system to the world coordinate system however so we simplified the process by having the robot just scan forward To have the workspace of a box would require the robot and camera to move sideways which can throw off the camera orientation with respect to the robot since the camera is mounted on the robot For when the robot moves in the y direction the whole came
34. t is displayed Commands FeatureFindTool NumberOfResults 0 Then mintResultNumber 1 DisplayResultInfo Else mintResultNumber 0 DisplayResultInfo End If Sub rivate Sub Form Load Loads the form initializes and sets parameter before anything are done Dim CommPort As String Handshaking As String Settings As String Dim counter x As Integer intializes these global variables counter_x 0 counter 0 flag 0 need to be specified for featurefindtool to work Commands FeatureFindTool hInspectionlImage EdibleImage Commands FeatureFindTool hInspectionShape edshpInspection Commands FeatureFindTool hTrainingImage Commands FeatureFindTool hlnspectionImage Commands FeatureFindTool hTrainingShape edshpTrain n image shown when program start at beginning Commands Ediblelmage Picture h SeniorProject LastYear abc bmp Commands edshpTrain Load h SeniorProject LastYear abc INI TrainingShape Commands edshpInspection Load h SeniorProject LastYear abc INI InspectionShape Commands FeatureFindTool Load h SeniorProject LastYear abc INI FeaturefindTool Commands FeatureFindTool Refresh to show when program is running Commands edshpTrain Visible True Commands edshpInspection Visible True Commands commandUse Enabled True Commands commandTrain Enabled True initial value used can be changed when program is running Commands txtMinScore 75 serial port settings very important
35. t the development of the vision system with the Seiko D TRAN RT3200 is already complete it is possible to sell the vision system at a much lower cost than the 32 202 shown Seiko has sold many RT3200 robots which means that there are many potential RT3200 owners that would need vision system update With our ability to offer our vision system at a much lower price than purchasing a RT3300 for that includes companies can save a great deal of money Evaluating the existing options of current RT3200 model owners it appears that there is a possibility of a very profitable opportunity if our vision system is marketed properly However due to limited research and knowledge about the number of owners of the Seiko RT3200 robot it is impossible to predict the size of the market for our product Chapter 5 Meeting Expectations 5 1 Introduction Our project has met all the specifications in the problem statement along with all of our primary objectives The robot can now be controlled through Visual Basic and with the addition of the vision system the robot can also scan for objects within the vision s workspace and perform specified task The vision system has increased the capabilities and value to the Seiko robot arm The Design Constraints are shown in section 6 2 and the Elements of Design are in section 6 3 5 2 Design Constraints Since our project is a continuation of last years design we are limited to the approaches and direction in which we
36. to move the robot arm from one position to another the delay is set to 40 and it works fine Shorter delay will work for shorter distance traveled by the robot but delay 40 is a safe and reasonable delay time Delay function used is actually a for loop that just loops multiple times Delay of 1 is equivalent to 1 000 000 loops Converting pixel coordinate system to world coordinate system When camera to ground height is 450mm the screen is 110mm X 90m and pixel to mm conversion is 1 5 8mm When we try to use the conversion to get the robot to the converted location at this height it is always off We found out that the camera is not mounted at the center of the robot and is the cause of our conversion to be off Later we found at the 450mm height the image is out of focus So the new height that the image is in focus and the end effector can still pick up object is at the height of 340mm At this height the screen is 95mm X 67mm and the pixel to mm conversion is 1 6 737 At each height to test for screen size and unit conversion four corners are drawn on a paper at the size of the screen Displayed on the screen and using image size on screen and the size of the four corners the new unit conversion is determined To test object is set at different location and the robot will try going to the object Many trials are taken to test for accuracy 3 3 Design Comparison Design Trade Off There are two major changes that we made to our o
37. train shape region It is possible to load an editable image and a trained image not in the editable image Getting coordinates when a trained image is found in the editable image The commands FeatureFindTool ResultX and FeatureFindTool ResultY is used to get the center position of the trained image with respect to the editable image when FeatureFindTool Found returns a True The top left corner of the editable image is 0 0 and the bottom right corner is 640 480 These are pixel coordinates and a calculation will be needed to convert the coordinates into spatial coordinates so the Seiko Robot can utilize it More details on the conversion from pixel coordinates to world coordinate in the Robotic test section Displaying live images on the editable image The image captured by the frame grabber must be stored onto a temporary variable and then using the temporary variable to have it display on the editable image Look at the Private Sub RunProg function to see how this is done I tried utilize the OCX file for the frame grabber to set as the video card and reference the image from the frame grabber to the editable image directly without first storing it in a temporary image However I was unable to reference frame grabber card in my code and had use the alternate option Automating the process of displaying the live images Robotic To automate the process I created a Do While a loop to continue reference the frame grabber i
38. tring 2 DO Tl amp amp amp darly amp 100 0 Commands cmdA Text set robot movement speed MSComml Output DO SPEED 300 amp vbCrLf Call delay 1 Commands cmdA Text A 6 Call delay 1 MSComml Output A MSComml1 Output vbCrLf Call delay 5 open end effector MSComm1 Output DO OUTPUT OG6 amp vbCrLf Call delay 40 move to found object MSComml Output DO MOVE 1 amp vbCrLf Call delay 40 end effector go down MSComml Output DO OUTPUT OG7 amp vbCrLf Call delay 40 close end effector 57 MSComml Output DO OUTPUT OG6 amp vbCrLf Call delay 40 end effector go up MSComm Output DO OUTPUT OG7 amp vbCrLf Call delay 40 set the destination that want to drop the object MSComml Output DO T3 17 473 120 0 vbCrLf Call delay 5 go to the destination that want to drop the object MSComml Output DO MOVE amp vbCrLf Call delay 40 drop object MSComml Output DO OUTPUT OG6 amp vbCrLf Call delay 40 example for last demo MSComml Output DO T4 369 34 91 0 amp vbCrLf Call delay 5 MSCommil Output DO MOVE 4 amp vbCrLf Call delay 40 MSComml Output DO 5 369 34 96 0 amp vbCrLf Call delay 5 uu MSComml Output DO MOVE T5 amp vbCrLf Call delay 40 AN MSComml Output DO OUTPUT OG6 amp vbCrLf Call delay 40
39. tureFindSample INI Featurefind ool ivate Sub commandUse Click Compares the features of the trained image with the image being displayed on the editable i Commands FeatureFindTool Use Commands FeatureFindTool MinimumScore Commands txtMinScore Commands CalXValue Text Commands FeatureFindTool Found rivate Sub commandTrain Click w Trains the image in the train shape box Rectangular blue shaped on the editable image Cli nd hold to drag shape Commands FeatureFindTool Train Commands commandUse Enabled True Commands edshpInspection Visible True Commands FeatureFindTool Refresh Sub rivate Sub disp Click Index As Integer Displays the live image on the Liveimage window Commands WindowState 1 Controller Show Livelmage Show nd Sub rivate Sub Ediblelmage Paint Controls what to display on the edible image Commands Ediblelmage Picture h SeniorProject LastYear abc bmp If Commands CommonDialog1 FileName Then Commands EdibleImage Picture h SeniorProject LastYear abc bmp ElseIf flag 1 Then Commands EdibleImage Picture h SeniorProject LastYear abc bmp Commands EdibleImage Refresh flag 0 commands 3 Else Commands EdibleImage Picture Commands EdibleImage Picture End If h SeniorProject LastYear abc bmp Commands CommonDialog1l FileName Sub rivate Sub FeatureFindTool Use When features desired are found on the edible image this controls wha
40. unction AllocateFG Lib wpx5 nt dll ByVal n As Long As Long Jeclare Sub FreeFG Lib wpx5_nt dll ByVal fgh As Long Grab Functions leclare Function GrabContinuous Lib 5 nt dll ByVal fgh As Long ByVal frh As Long ByVal ac Live As Long As Long 1 Function Grab Lib wpx5 nt dll ByVal fgh As Long ByVal frh As Long ByVal flags As Lon J As Long leclare Function GrabTriggered Lib wpx5 nt dll ByVal fgh As Long ByVal frh As Long ByVal del ime As Long ByVal flags As Long As Long eclare Function ReadCache Lib wpx5 nt dll ByVal fgh As Long ByVal frh As Long ByVal flags A Long As Long 1 Function GrabToCache Lib wpx5 nt dl1l ByVal fgh As Long ByVal flags As Long As Long eclare Function CacheTriggered Lib wpx5 nt dll ByVal fgh As Long ByVal deltime As Long ByVa flags As Long As Long leclare Function Wait Lib wpx5 nt dll ByVal fgh As Long ByVal flags As Long As Long 1 Function WaitVB Lib wpx5 nt dll ByVal fgh As Long As Long leclare Function IsFinished Lib wpx5 nt dll ByVal fgh As Long ByVal handle As Long As Long 1 Sub KillQueue Lib wpx5 nt dll ByVal fgh As Long Frame Grabber State Functions leclare Function SetFineGain Lib wpx5 nt dll ByVal fgh As Long N Long As Long Function SetGainRange Lib wpx5 nt dll ByVal fgh As Long ByVal range As Long ByVal fl gs As Long As Long leclare Function SetO
41. y move arm for given destination Dim T 1 To 4 As String T 1 CStr x T 2 CStr y T 3 CStr z 4 CStr R Define point m destination T 1 T 2 T 3 T 4 buffer DO T1 destination MSComml Output buffer vbCrLf Call delay 7 MSComml Output DO MOVE T1 vbCrLf Sub ub delay c As Integer Delay for specified times Used quite often to make sure robot and vision can be worked toge For counter 1 1000000 Next counter 474 nd Sub Sub pick y As Boolean To pick object if found or increment to next area if not found Loop is under the RunProg function to scan workspace of robot Dim screen x As Double Dim screen y As Double im darl x As Integer Jim darl y As Integer Dim A As String Dim move x As Integer Dim B As String Dim move x2 As Integer Dim ndarl x As Integer clear robot memory bank MSComml Output DO CLEAR amp vbCrLf Call delay 5 y Commands FeatureFindTool Found 1f y True Then get pixel location screen Commands FeatureFindTool ResultX mintResultNumber screen y Commands FeatureFindTool ResultY mintResultNumber darl y CInt 64 516 screen x 6 124 convert pixel location to real cooridate darl x CInt screen y 6 124 387 convert pixel location to real cooridate ndarl x x counter x Hog Commands txtWorldX Text Commands txtWorldY Text ndarl x darl y set darl command into s
Download Pdf Manuals
Related Search