Automated visual positioning and precision placement of a workpiece using deep learning

Li, Chih-Hung G.; Chang, Yang-Lang

doi:10.1007/s00170-019-04293-x

Cited by 37 publications

(12 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Segmenting the workpiece area from the nonworkpiece area in the point cloud data helps to increase the number of detected workpieces and estimate the correct object poses [17]. Using the position image of the workpieces to train the CNN to push the position deviation of the mechanical part for accurately placing the workpieces on the fixture [18]. Thanks to the powerful fitting capabilities of various deep learning network frameworks, although workpiece detection tasks are more diverse and complex, they can do tasks that traditional algorithms cannot.…”

Section: Related Workmentioning

confidence: 99%

Multiple Object Detection of Workpieces Based on Fusion of Deep Learning and Image Processing

Lei

Yao

Chen

et al. 2020

2020 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

A workpiece detection method based on fusion of deep learning and image processing is proposed. Firstly, the workpiece bounding boxes are located in the workpiece images by YOLOv3, whose parameters are compressed by an improved convolutional neural network residual structure pruning strategy. Then, the workpiece images are cropped based on the bounding boxes with cropping biases. Finally, the contours and suitable gripping points of the workpieces are obtained through image processing. The experimental results show that mean Average Precision (mAP) is 98.60% for YOLOv3, and 99.38% for that one by pruning 50.89% of its parameters, and the inference time is shortened by 31.13%. Image processing effectively corrects the bounding boxes obtained by deep learning, and obtains workpiece contour and gripping point information.

show abstract

Section: Related Workmentioning

confidence: 99%

Multiple Object Detection of Workpieces Based on Fusion of Deep Learning and Image Processing

Lei

Yao

Chen

et al. 2020

2020 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

show abstract

“…The performance evaluation results of the proposed algorithm have shown accurate predictions in detecting tool wear under various cutting conditions with a high-speed response rate. Li and Chang [4] proposed an automated visual positioning system for precision placement of a workpiece on the xture. The experimental evidence of workpiece placement con rms that the low-resolution (640× 480 pixels) camera can obtain a translational precision of ± 0.2 mm; by using the binocular system can control the rotational error within ± 0.1°.…”

Section: Introductionmentioning

confidence: 99%

Robotic Obstacle Avoidance: A Virtual Modeling And Reinforcement Learning

Chen¹,

Tsai²,

Hsiao³

2021

Preprint

View full text Add to dashboard Cite

This study developed a robotic arm self-learning system based on virtual modeling and reinforcement learning. Using the model of a robotic arm, information concerning obstacles in the environment, initial coordinates of the robotic arm, and the target position, this system automatically generated a set of rotational angles to enable a robotic arm to be positioned such that it can avoid all obstacles and reach a target. The developed program was divided into three parts. The first part involves robotic arm simulation and collision detection; specifically, images of a six-axis robotic arm and obstacles were input to the Visualization ToolKit library to visualize the movements and surrounding environment of the robotic arm. Subsequently, an oriented bounding box algorithm was used to determine whether collisions had occurred. The second part concerned machine-learning–based route planning. The TensorFlow was used to establish a deep deterministic policy gradient model, and reinforcement learning was employed for the response to environmental variables. Different reward functions were designed for tests and discussions, and the program’s practicality was verified through actual machine operations. Finally, the application of reinforcement learning in route planning for a robotic arm was proved feasible by the experiment. Such an application facilitated automatic route planning and achieved an error of less than 10 mm from the target position.

show abstract

“…2 (b). In those tasks, the 3D vision sensors can be utilized to capture the point cloud data (PCD) and estimate the pose of parts through the iterative closest point (ICP) algorithm [14] or using neural networks [11,13,28].…”

Section: Introductionmentioning

confidence: 99%

Fast 6D Object Pose Estimation of Shell Parts for Robotic Assembly

Yang

et al. 2021

Preprint

View full text Add to dashboard Cite

Shell parts which have similar and close inner and outer surfaces are common in industrial manufacturing applications. In view of the 6D pose error compensation of parts in high-precision robotic assembly tasks, this work proposes a fast 6D pose estimation approach tailored for shell parts. With a binocular structured light camera, the proposed approach consists of two phases, namely initial pose estimation phase and local pose estimation phase. In the former one, an initial pose correction and translation offset methods serve to solve the local optimal estimation problem of the iterative closest point (ICP) algorithm. This problem is caused by the poorly assigned initial pose and the similar inner and outer surfaces of shell parts. In the latter one, the voxel sampling and the weighted point-to-plane ICP algorithms are applied to boost the efficiency of the pose estimation approach. With two typical shell parts, a simulation and an experiment of pose estimation are conducted to verify the effectiveness of the proposed approach. Experiment results prove that the accuracy of the pose estimation approach is 0:27mm/0:38°, and the runtime is 680ms.

show abstract

Automated visual positioning and precision placement of a workpiece using deep learning

Cited by 37 publications

References 19 publications

Multiple Object Detection of Workpieces Based on Fusion of Deep Learning and Image Processing

Multiple Object Detection of Workpieces Based on Fusion of Deep Learning and Image Processing

Robotic Obstacle Avoidance: A Virtual Modeling And Reinforcement Learning

Fast 6D Object Pose Estimation of Shell Parts for Robotic Assembly

Contact Info

Product

Resources

About