Peg-in-hole assembly based on hybrid vision/force guidance and dual-arm coordination

In this paper, a peg-in-hole assembly strategy based on vision/force guidance and dual-arm coordination is proposed. Vision guidance using corner detection algorithm is applied for rough adjustment and ensures a suitable dual-arm operating space. To realize precise adjustment, position adjustment and orientation adjustment are realized respectively according to the forces and torques information detected by a 6-axis F/T sensor. Besides, master-slave arms are distinguished in dual-arm coordinated operation. The proposed strategy can be applied not only for round assembly parts, but also for triangle and square parts. Experimental results show that the proposed assembly strategy perform well for different-shaped parts and 0.5mm maximum clearance between peg and hole can be assembled successfully.