Journal
2021 4TH INTERNATIONAL CONFERENCE ON INTELLIGENT AUTONOMOUS SYSTEMS (ICOIAS 2021)
Volume -, Issue -, Pages 97-102Publisher
IEEE
DOI: 10.1109/ICoIAS53694.2021.00025
Keywords
charging system; force control; F/T sensor; peg-in-hole; 3D camera
Funding
- National Key R&D Program of China [2017YFB1300400]
- National Natural Science Foundation of China [62073249]
- Major Project of Hubei Province Technology Innovation [2019AAA071]
- Scientific Research Program Foundation for Talents from Department of Education of Hubei Province [Q20191108]
- Wuhan Science and Technology Planning Project [2018010401011275]
Ask authors/readers for more resources
In this paper, a new energy vehicle automatic charging system is developed using 3D vision and F/T sensor on a cooperative robot. A position and attitude adjustment method based on force and torque information classification is proposed, achieving an accuracy of 92.59% for the classifier and a 94% success rate for insertion. These experiments verify the feasibility of the system in terms of precision rate, recall rate, F-Measure, and actual insertion success rate of the classifier.
In this paper, a new energy vehicle automatic charging system is built on the cooperative robot by using 3D vision and F/T sensor. In particular, aiming at the problem that the attitude error is not taken into account in the peg-in-hole problem of the existing system, this paper proposes a position and attitude adjustment method based on the classification of force and torque information, which divides the contact force and torque obtained by the F/T sensor into nine categories. According to different classification results, different compensation strategies are adopted to adjust the motion control, and finally the charging plug is inserted accurately and flexibly into the charging port. Finally, the random contact data classification experiment and the automatic charging actual insertion experiment are designed. the experimental results show that the accuracy of the classifier is 92.59%, and the success rate of insertion is 94%. With these two experiments, the feasibility of our system is verified from the aspects of precision rate, recall rate, F-Measure and actual insertion success rate of the classifier.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available