Peg-In-Hole Operation Using a Cobot Without Using External Sensors (original) (raw)

Abstract

sparkles

AI

This research explores the automation of peg-in-hole operations using a collaborative robot (cobot) without external force or torque sensors. By leveraging internal motor current consumption data from a UR5 cobot, the study successfully implements monitoring mechanisms that ensure precise part insertion through a two-phase process. The results reveal a high success rate of 96.7% in inserting parts, with implications for reducing costs and complexity in robotic automation for smaller enterprises.

Key takeaways

sparkles

AI

  1. Achieved a 96.7% success rate in peg-in-hole operations using a UR5 cobot.
  2. The algorithm utilizes internal motor current for force and torque feedback without external sensors.
  3. First phase centers the part using a force threshold of 60N; second phase monitors torque variation.
  4. Over 1000 operations performed, with 90% successful on the first trial.
  5. Mean operation time is 5.5 seconds, excluding part picking.

Loading...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

References (2)

  1. Y. Kim, B. Kim, J. Song, "Hole detection algorithm for square peg-in-hole using force- based shape recognition", 8th IEEE Int. Conf. on Automation Science and Engineering, pp. 1074-1079, August 2012.
  2. Kramberger et al., "Transfer of contact skills to new environmental conditions," 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids), Cancun, 2016, pp. 668-675.

FAQs

sparkles

AI

What was the success rate of peg-in-hole operations using the cobot?add

The research demonstrated a 96.7% success rate over 1000 peg-in-hole operations, with 90% achieved on the first trial.

How does the cobot determine successful part centering during operations?add

The cobot utilizes an internal force vector threshold of 60N to monitor part centering during the initial phase.

Which technology was employed for the peg-in-hole operation?add

A UR5 collaborative robot from Universal Robots was utilized, with a standard RG2 gripper and a 3D printed adaptor.

What methodology was used for the peg-in-hole descent phase?add

The robot performed a helicoidal descent, decreasing the vertical z value by 1mm with a 2° rotation at each step.

How can torque variation indicate successful insertion in the operation?add

A 50% threshold variation in shoulder torque during the second phase signals a successful part insertion.