| 摘要 |
PURPOSE:To attain a CP action at playing back in the combination with actual fine PTP actions by sampling periodically and action locus available from a manually operated robot so as to store it as NC data in a memory when the CP action is taught. CONSTITUTION:When contents in a teaching action mode buffer 1 is '1', a CPU 9 receives an interruption in an interval set to a sampling period buffer 3 by a timer interrupter 13. When the CP action (the locus of the tip of the wrist of a robot draws a straight line or an arc) is recognized, the CPU 9 reads the current value of the robot from a POS11 and sets it in a current value buffer 15. continuously contents in a current value buffer 15 are compared with those of a previous current value buffer 17. when there is a difference between them, NC data is generated by assuming that a pivot action is executed, and set in an NC data buffer 5. Upon the completion of the generation of the NC data at one sampling point, the operation is shifted to the next sampling point. When the data generation at all sampling points is terminated, the CP action is executed according to said data.
|
