Estimation of Grasp States in Prosthetic Hands using Deep Learning

Parque, Victor; Miyashita, Tomoyuki

doi:10.1109/compsac48688.2020.00-79

Cited by 5 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, deep learning-based image processing technology has significantly promoted the advancement of grasping strategy research, thereby expanding the application prospects of top-grasping strategies. One of the mainstream research methods to improve the object detection algorithm is through the generation of rectangle grasping strategies [ 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 ]. For instance, Joseph Redmon et al [ 8 ] developed an object detection model with a mixture of convolution layer and full connection layer and realized grasp detection and object classification at the same time, with excellent performance in speed and accuracy.…”

Section: Introductionmentioning

confidence: 99%

“…For instance, Joseph Redmon et al [ 8 ] developed an object detection model with a mixture of convolution layer and full connection layer and realized grasp detection and object classification at the same time, with excellent performance in speed and accuracy. Victor Parque et al [ 11 ] proposed an RGB-based grasping attitude prediction model using the GoogLeNet framework. The model can achieve excellent prediction performance under the training of a small number of label datasets.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Research on Robot Grasping Based on Deep Learning for Real-Life Scenarios

Bai

2023

Micromachines

View full text Add to dashboard Cite

The application of robots, especially robotic arms, has been primarily focused on the industrial sector due to their relatively low level of intelligence. However, the rapid development of deep learning has provided a powerful tool for conducting research on highly intelligent robots, thereby offering tremendous potential for the application of robotic arms in daily life scenarios. This paper investigates multi-object grasping in real-life scenarios. We first analyzed and improved the structural advantages and disadvantages of convolutional neural networks and residual networks from a theoretical perspective. We then constructed a hybrid grasping strategy prediction model, combining both networks for predicting multi-object grasping strategies. Finally, we deployed the trained model in the robot control system to validate its performance. The results demonstrate that both the model prediction accuracy and the success rate of robot grasping achieved by this study are leading in terms of performance.

show abstract