Dynamic Target Detection and Tracking in Water for Mobile Robot Based on Deep Learning

Yang, Chen; Xu, He; Li, Xin; Huang, Jian

doi:10.1007/978-3-030-63403-2_48

Cited by 1 publication

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The method is used to eliminate the requirement of hand-eye calibration and camera fixation relative to the robot during system reconstruction, increasing the flexibility of system performance. A real-time dynamic image processing algorithm for robots based on deep learning was proposed [16]. Images were collected through different water depths and different motion directions; a target detection strategy was designed according to the YOLO-V3 algorithm, and the accuracy rate of the tested model reached 94.82%.…”

Section: Introductionmentioning

confidence: 99%

Driving Force Estimation of 3-PRS Parallel Robot Based on Deep Learning

2024

IJOMAM

View full text Add to dashboard Cite

Aiming at the problem of the difficulty in solving the dynamics driving force of the parallel robot, a deep learning model is proposed to estimate the slider driving force of the 3-PRS parallel robot. The 3-PRS parallel robot is presented, and its dynamics model is established using the Newton-Euler method. The dynamics of the 3-PRS parallel robot is analysed by using the theory of the "black box model". The parameters of the designed 3-PRS parallel robot are used as input, and the driving force can be output directly through the deep learning algorithm. The regression convolutional neural network (CNN) algorithm based on deep learning is established to estimate the slider driving force. An overall framework is proposed according to the characteristics of the three sliders. The parameters of the parallel robot are collected, and the collected data are converted into a grayscale image to establish the data sets for driving force estimation. Two cases with friction and without friction are considered respectively in the experiment. The experimental results show that measuring the driving force using the regression CNN can improve calculation accuracy and efficiency, which lays a foundation for studying the dynamics of the parallel robot.

show abstract

Section: Introductionmentioning

confidence: 99%

Driving Force Estimation of 3-PRS Parallel Robot Based on Deep Learning

2024

IJOMAM

View full text Add to dashboard Cite

show abstract

Dynamic Target Detection and Tracking in Water for Mobile Robot Based on Deep Learning

Cited by 1 publication

References 9 publications

Driving Force Estimation of 3-PRS Parallel Robot Based on Deep Learning

Driving Force Estimation of 3-PRS Parallel Robot Based on Deep Learning

Contact Info

Product

Resources

About