Pose Estimation of Circular Workpieces With Occlusion Based on GAN-Support Ellipse Detection in Manufacturing

Bie, Haodong; Xu, Wenjun; Yao, Bitao; Cui, Jia; Hu, Yang

doi:10.1109/cscwd54268.2022.9776130

Cited by 1 publication

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The pose matching is based on the estimated pose and the standard pose of the parts. In Reference [18], we proposed a pose estimation method of circular parts based on GAN-support ellipse detection in manufacturing, which can handle severe occlusion, and does not spend any other time for slight occlusion or with occlusion. The parts in this paper are with circular features, so we use the pose estimation method proposed in [18].…”

Section: Pose Estimationmentioning

confidence: 99%

Assembly Status Detection Based on Step-by-step Strategy for Robotic Assembly

Chen

Bie

Yao

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Assembly status detection is of great significance in the automated assembly of industrial robots. However, in addition to the traditional manual inspection, the existing methods for assembly status detection are mainly through methods such as torque sensors, etc., which are difficult to implement in practical assembly and lack the detection of the entire assembly cycle. To solve this problem, this paper proposes a method for assembly statute detection based on a step-by-step strategy. The first step is to use fast bounding box collision detection to determine whether the robot enters the correct assembly neighborhood. The second step is to use accurate pose information to determine whether the product is assembled correctly. The experiments show that the proposed method can accurately detect the assembly status for the entire assembly cycle with high assembly efficiency.

show abstract

Section: Pose Estimationmentioning

confidence: 99%