Self Improvement

Study of the Hero-l Educational Robot

Study of the Hero-l Educational Robot
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  »a :1 .i . " .... T   P   IN EDUCATION DIVISIO N OF ASEE セ   1985 JOUR NAL COM PUTERS VOL. V NO.3 JULY - SEPTEMBER CONTENTS CoED Newsletter • e. TRANSACTIONS 'An Inexpensive Robotics Laboratory e N   . . セ   セ N   H IZ 90 Person al PR1ME 550 Com puter Workstation For The Development Of A Comp uter Course Plus Labora to  y Communications Stud y Of Th e Her o-L Educa ti onal Robot 20 . - '   t " APPLI CATION NOTE T he Simula ti on On A Gr ap h ic s Of Pl ane Ki nema tic Mecha nis ms Com pu ter : A Stu den t P ro je ct . 25 © CoED 1985 , . I . .,. J P r in t e d in U.S. A .  STUDY OF THE HERO-I EDUCATIONAL ROBOT Adel S. Elmaghraby and V. Jagannathan Department of Engineering Mathematics and Computer Science Univers lt y of Louisville Louisville, Kentuck y ABSTRACT This study analyzes Heathkit's multi-sensor educational robot - Hero-j . The robot, unlike most industrial manipulators, Is. capable of moving using a set of wheels. Discussion of the Hero-I Is presented to show programmabllit y of the robot for specific tasks. INTRODUCTION Robot technology has advanced conslderabl y in the past decade. However, the full potential of robots has yet to be realized. The Intelligent control of robots will require a departure from the current industrial method of manuall y teaching the robot for specific applications . Off -line programming of robots utilizing high level robot languages and artificial intelligence concepts Is the direct ion of current research for optimal utilization of robot and computer technology [Elmaghraby and Ja gannathan, 1983]. In a university environment, the development of an Intelligent Robotic Control Language [Elmaghraby and Jagannathan, 1983] Is current! y underwa y utiliz ing Ii microcomputer and an educational robot  Heathkit's newl y Introduced Hero-I. The foundation of Intelligent control of a robot is the stud y of its kinematic.s and dynamics. This paper focuses on such a required analysis as applied to Hero-I, an effort required pr i or to the development of a higher level langua ge . Hero-I Is a mobile microprocessor-based mult lsens or rob ot. It is programmable In a he xadecimal code using an expanded set of Motorola 6808 mach ine instruct i ons. Sensory capabilities include range, motion, sound level, and l ight level d et e cti on . I t also has a pr o gra mmable spe ech synthesis cap ab ility. Th is stud y of the Hero -l robot incl ud es the consideration of its mobility in addition to an analysis of its arm manipulation. Applications of the anal ysis to the control of specific robot tasks, future research directions and Hero-I limitations will also be discussed in the f ollowing sections. KINEMATIC REPRESENTATION Representation of the robot body, head and arm is required using some conventional frame of reference. An Initial difficult y is encountered due to the complete mobility of the robot body. Algorithms suggested by Paul (19811 and Lee (1982) were both cons idered and adapted for use with Hero-I. A s ketch of the robot Is shown In Figure 1. The procedure used to describe coordinate assignment of robot joints and links is described below and assignments depicted in Table 1 and Figure 2. Since the robot Is mobile, a fixed coordinate frame representing t he robot's In itial position and orientation will be the base reference, B. A robot coordinate frame, R, is assigned to the base of the robot body parallel to the base coordinate frame  the transformation matri x TR B represents the translation of the robot body. Its location now described, the orientation corresponding to the front wheel turn is given by another frame of reference, D, selected in such a manner to align XO with the direction of the robot face the transformation matrix TOR describes the change In the orientation w ith respect to the coordinate frame R. By separating the robot position and or ientation from its head and arm movements, one can immediatel y use the algorithm d eveloped for most robot man ipulators. The head turn, shoulder rotat ion , arm extens i on, wr ist pivoting, wr ist rotat ion, an d grip pe r pos ition are r eferenced usi ng c oord inate fra mes 1 thro u gh 6. These c oor dinate fra mes were assi gn ed follow ing the al gorithm detailed by Lee (1982). One s hould not e that d I d2, and d6 are constants based on the robot dime ns ions Tran s ction s OCoED $ 0  ( e. JOINT 8L ex:: . l, d · t, 1 B 1 -90' d 1 2 8 2 +90' d 2 3 0 -90' d 3 4 8 4 +90' 0 5 B 5 0 0 Table 1 Figure 1 Figure 2 1 21 Transa ct ion s C> CoED  I I セ   0 se 4w hile el , e 2, d 3 and 84 are the Joint variables. Wrist Pivot セ {l4 0 - Cll 4 Also a and b are variables specifying the robot To T43 { 010 S position with respect to the base frame of Extender ·ft 00 !] efere nc e. -5 0S 0 Wrist Rotate S6S ceS 0 To TS4 0 01 Wrist Pivot r !] 00 Gr ipper o o To 1 o o j o o atrix transformations for the complete robot are Wrist T6S 1 listed below. o o 1 o TR o 1 Robot To oo Base j セ   KINEMATIC SOLUTION Wheel c eo -5 eo o Hero-l Is a microprocessor based programmable Orientation 5 0 ceo o robot. Atypical problem Is: Given a set of robot To TCR [ oo 1 motion instructions one needs to determine the Robot o o o position of the robot's end effector - the gripper. This problem, known as the d irect kinematic solution, is obtained through the mult iplication of the transformation matrices to refer the end effector with respect to the base coordinate frame. The resultant position and orientation are descr ibed by: o -581 Head o C81 To Tl0 -1 o TB t n . .! £. J oo heels lB nx sx ax ークセ   ny sy ay py C 82 o sa Shoulder nz sz az pz 2 o -032 [To o 001 T21 2 o Head [ r   o where E Is the vector describing the center of the gripper with respect to the base coord inate frame, a Is the approach vector and s Is the sliding vector o o and n Is the vector norma I to a and s , These vectorsExtender o 1 are Shown In Figure 3. TB canbe derived through the To セ 32 -1 o d3 following matrl x multiplications: Shoulder oo 1 TS TRS*TOR*Tl0*T21*T32*T43*TS4*T65. Tran s ctions OCoED $ 2  GRIPPER Figure 3 A complete symbolic solution has been obtained using Reduce on a DEC-10 system. Elements of the matrix TB corresponding to e セ   セ   and.!! are shown In Appendix A. . The inverse problem Is that of determining the appropriate motion instructions. Given a target position and orientation for the gripper with respect to the base. one needs a set of values for the joint variables - namely 8 1. 8 2. d3,' 8 4. 85. A structured approach Is presented by Paul (1981). Equating matrix TB to a desired gripper position and orientation direct! y yields a sol ution for 8 5 and (6 2 + 84) . A series of algebraic manipulations will be required for the solution of all joint variables. SENSOR BASED MOTIONS Solutions derived from the kinematic equations are of theoretical Interest. However, In controlling robots for Impreclsel y defined tasks with imprecise robot movements requires adaptive control. This can be achieved by incorporating sensor y feedback and heuristic algorithms -a direction of current and future research. A typical robot application Is the pick and pl ace pr obl em. Using Hero-1 this was possible by I セ N   t , using the light sensor or by usin g the range finder. Coordination and interaction with a human controller was also po ss ible by u sing synthesized sp eech and sound dete ction. Resulting motion was more flexible than those resulting through use of th e teaching pendant of Hero-}, y 6 Z 6 CONCLUSIONS AND FUTURE WORK The Hero-1 robot Is an inexpensive education al tool to study sensor-based motion and other ro bo t i cs applications. Hero-l's major limitation is its lac k of tactile sensor's and the absence of a hig h level language Interface. Current resea r ch pr ojects us ing Hero-1 include use of artificial intelligence a nd simulation techniques in the design and developm e nt of a high level language and interfaces wit h a microcomputer and a low cost camera. ACKNOWLEDGEMENT This research was partia lly supported by Universlt y of Louisville. Graduate School Initiation grant. REFERENCES 1. A. S. Elmaghraby and V. Jagannathan, "Intelligent Robot Control." Initiation Grant Proposal. Graduate School. Unlversit y of Louisville. 1983. 2. C. S. G. Lee. "Robot Arm Kinematics, Dynam ic s. and Control." Computer. December, pp , 62- 80. 1982. 3. T. Lozano-Perez. "Robot Programming." Prcc, of the IEEE, Vol. 71. No.7. July. pp 821-841. 1983. 4. R. P. Paul. "Robot Manipulators, Mathemat ic s. Programming and Control." The MIT Press, Cambrid ge. Mass •• 1981. Tran s cti on s Cl CoED $ 23
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks