TY - GEN
T1 - Hole detection algorithm for square peg-in-hole using force-based shape recognition
AU - Kim, Young Loul
AU - Kim, Byeong Sang
AU - Song, Jae Bok
PY - 2012
Y1 - 2012
N2 - It is very difficult to assemble precision parts using only position control. To deal with these problems, force control approaches, which provide a proper motion response against the contact force, were investigated. Moreover, hole detection is a crucial step to eliminate the uncertainty in robotic assembly. Without a proper hole detection algorithm, assembly time increases along with the position difference between the peg and the hole. In this study, we propose a shape recognition algorithm based on a 6 axis F/T sensor and the hole detection algorithm. The proposed hole detection algorithm can find the direction of the hole regardless of the size of the position error between a peg and a hole. The same algorithm can be implemented not only to a circular peg, but also to a polygonal convex peg. A series of experimental results show that the proposed algorithms can estimate the shape and location of a peg reasonably well.
AB - It is very difficult to assemble precision parts using only position control. To deal with these problems, force control approaches, which provide a proper motion response against the contact force, were investigated. Moreover, hole detection is a crucial step to eliminate the uncertainty in robotic assembly. Without a proper hole detection algorithm, assembly time increases along with the position difference between the peg and the hole. In this study, we propose a shape recognition algorithm based on a 6 axis F/T sensor and the hole detection algorithm. The proposed hole detection algorithm can find the direction of the hole regardless of the size of the position error between a peg and a hole. The same algorithm can be implemented not only to a circular peg, but also to a polygonal convex peg. A series of experimental results show that the proposed algorithms can estimate the shape and location of a peg reasonably well.
UR - http://www.scopus.com/inward/record.url?scp=84872551123&partnerID=8YFLogxK
U2 - 10.1109/CoASE.2012.6386340
DO - 10.1109/CoASE.2012.6386340
M3 - Conference contribution
AN - SCOPUS:84872551123
SN - 9781467304283
T3 - IEEE International Conference on Automation Science and Engineering
SP - 1074
EP - 1079
BT - 2012 IEEE International Conference on Automation Science and Engineering
T2 - 2012 IEEE International Conference on Automation Science and Engineering: Green Automation Toward a Sustainable Society, CASE 2012
Y2 - 20 August 2012 through 24 August 2012
ER -