| Conclusions The elements of parallel robotic manipulator were designed and manufactured, including two stepper motor driven, nut-screw actuators, connected by a pin joint. Preliminary designs have been submitted in an iterative process, which included trial and error in many of its stages, until the most simple, economic lighter weigh and compatible with existing component design was obtained. |
 |
| Design & Manufacture of a Parallel Kinematic Robotic Prototype |
 |
OVERVIEW We have been designing and building a robotic manipulator prototype working as a plotter as well as a PCB driller. The mechanism will be driven by linear actuators controlled by stepper motor control system. The procedures and tasks performed during GP1 include: 1. Proposing the details of design, manufacturing and assembly of the actuator. And building an initial prototype in the college workshop. 2. Deducing the physical equations describing a variable frequency signal to be sent to the feed drives and writing a C code describing that signal. 3. Tests were conducted to verify the correction of wiring, the speed of the motor, and the computer interface. 4. Several procedures were followed to correct the motor RPM problem as going to be discussed later. 5. Direct and inverse kinematics analysis to coordinate the motion of the plotter on the plotting plane and writing a C code that does this. The procedures and tasks performed during GP2 include: 1. A test was carried to identify the range of speed at which the motor can work at. 2. A parallel robotic mechanism, that includes two actuators with their holders and a pin connection, pen mechanism and a jack mechanism were designed and manufactured. 3. The distance between the actuators centers of rotation was selected to obtain a maximum working area. 4. All the required wiring between the motors, the drive boards and the power source was made. Control circuits for a solenoid and two limit switches were built. All the circuits were compacted in a connection box and the wires were gathered in long cables. 5. A user-friendly interface was written. It consists of basic linear and circular motion functions in addition to letters typing and drilling functions. Codes were also written for controlling the solenoid and the jack. The following general objectives are fulfilled throughout working on this project: 1. Introduction to analysis, design and manufacturing of real time robotic systems. 2. Integrating the areas of design, kinematics, dynamics, control, electronics, electrical machines etc. in one whole process that aims at building a robot prototype. 3. One other objective is to gain the skill of selecting compatible system components that can fit together. |
 |
|
| Chapter 1 Introduction |
| Chapter 2 Mechanical Design |
| Chapter 3 Motor Drive & Connections |
| Chapter 4 Interfacing & Programming |
| Very Important Caution ! I do not, and will never, support neither the anti-religious nor the non-moral advertisements that would be demonstrated by the Geocities Ad Banner shown in the upper right corner of this page. The existance of this banner is a condition of my presence in GeoCities, so I can do nothing about it. You can remove it by clicking Close then Click X. Thank you . All Rights Reserved. Haytham Abdulwahab 2002 |
 |
 |
 |
 |
 |
 |
 |
 |
| Motion on one axis and in two axes for straight and circular paths were analyzed. Based on the analysis, a user interface was built to allow the user to position the end-effecter at the required point on the x-y plane, The interface allows the user to move on linear and circular segments. The parallel mechanism was tested by using it to draw linear and circular segments. An input speed of 2.5mm/s for the linear segment, and 2.5/r for the circular segment, where r is the radius of the circular segment, were used. The distance between the pen and the point connecting the two struts is 90mm. The linear motion starts from the home position of the mechanism which is point (-36.51, 340.01) and continues to the point (0, 365), which is lying on the circumference of the circular segment. A 360deg rotation was made around the point (20, 385) going back again to (0, 365). The user interface contains inverse kinematics formulas that allow us to capture the path of the motion, by recording the actual instantaneous position for the end-effecter every time step. The points where examined and the path was plotted. The program generates a linear path that reaches the point (0, 365) before rotating around the specified center and coming back to the same point again. The figures below show the path generated by the program, which is exactly the required path. This verifies the analysis of motion coordination on 2D for linear and circular paths. |
Connections Box |
| Errors are generated by the positioning mechanism. For example, the datum point is read on the computer as (-39.57, 281.14) while it was measured to be (-44.0, 283.4). The difference is due to the two following reasons: 1. Inaccuracy in inputting the dimensional parameters to the interface. 2. Inaccuracy in measuring the pen position on the x-y plane. The difference can be reduced by eliminating the error sources. Errors in the positioning process will result from the previously mentioned reasons in addition to the accumulation of error. The segments moved are not as smooth as required because of the following manufacturing errors: 1. The screws are not perfectly straight and are badly manufactured resulting in vibrations when the nut moves on the axis of the screw. 2. The vibration is also resulting from the motion of nuts of the two struts at the same time. More straightened segments results from the motion of one nut only. 3. One other source of vibration is the nature of the cantilevered shape end-effecter. Vibration is amplified by the lever action of the pen mechanism. 4. The axis of rotation of the actuator b is not perpendicular to the working plain due to bad manufacturing of its holder. A static end-effecter that eliminates the lever action and has a shorter distance between the center of the pen and the joint was used draw linear and circular segments. The generated segments were much smoother which verifies some of the sources of error mentioned above. The new end-effecter is shown in the figure: |
Verification of straight path generation |
Verification of circular path generation |
| The mechanism can be used in drawing. A pen mechanism end-effecter can be installed on the pin joint. It is driven by a solenoid control circuit embedded in the connection box and is controlled by the computer through the parallel port. A letter typing system was built using sequences of basic linear and circular motion functions and solenoid control functions. |
| The mechanism can be also used in a PCB drilling system. A drill is installed as an end-effecter; while a stepper motor driven jack mechanism that holds the card is used to perform the motion in the z direction. A sequence of basic linear function was also used in the control code. We have to mention that the drill end-effecter was not built because the budget was not enough to purchase a light weight small drill that is compatible to our design. ] |
| A return to datum system was also made to allow returning to a known position. Two limit switches that can send signals to the computer through the parallel port were used to accomplish the mission. Their control circuits are embedded in the connection box. |
Recommendations Basically, all tasks that were carried from Graduation Project I were fulfilled first in Graduation Project II, including correcting the problem of the stepper motor speed. The tasks of Project II were also all attempted. There are however a number of tasks that could have been tried and can be attempted in similar future projects: 1. Verification of mechanical designs before manufacturing by building them using cardboard and small sticks. The DOF of the jack mechanism could have been examined by doing this instead of the long trial and error process that was undertaken. 2. A new pen mechanism that have not lever action and that have the pen as near as possible to the joint should be built. 3. A ball-screw mechanism should replace the lead-screw mechanism that we manufactured in the college machine shop. This will reduce the friction and will eliminate the unnecessary loss of power by the motors. 4. Better motors should be employed to achieve higher feed-rates. 5. Interrupt driven real time system should yield better control of the system. |
Pen Mechanism Splenoid Driven End Effecter |
| The correct wiring diagram as well as the motor performance were verified. A connection box that mainly contains the motors connections and feed drives was made. We have encountered a problem in the power supply. The computer ATX-PIV 400W power supply is unable to drive the motors. This is most probably due to low current supply. A PS 5010 programmable power supply was to feed the motors and the rest of the electronics instead. |
Computer Mouse Limit switch used to bulid a return to datum system |
Mechanical Setup of the robot |
| Simple End Effector to eliminate manufacturing errors |
| United Arab Emirates University College of Engineering Graduation Project I & II Acadamic Year: 2002/2003 Work of the students: Haytham Mahmoud Abdulwahab Abdulraheem Amin Al-Zarooni Khalid Jumaa Al-Housani Advisor: Dr. Khalifa Harib |
| Project Documentaton Chapters: |
Cost Esitmation of the robot The total components cost was 2434.5 Dhms. The labor cost is 20 Dhms per hour and is equal to 500 Dhms for 25 work hours. The total cost of the machine is 2934.5 Dhms which is equivalent to $815. |
 |
| Other Projects: |
| Bonus Projects! |
| Control Code: Download here(closed) |
| Mechainical Drawings: (closed) drawing 1 drawing 2 drawing 3 drawing 4 drawing 5 drawing 6 drawing 7 drawing 8 drawing 9 drawing 10 drawing 11 drawing 12 drawing 13 drawing 14 drawing 15 |