Year: 2025
Type: conference-paper
Abstract: In recent years, with the development of artificial intelligence, about the robot stabilization grasping has also made great progress. Especially about the force closure grasping, also made greater development, but the existing force closure grasping still exists the problem of poor real-time and low success rate. In this paper, a fast three-finger force closure grasping stabilization algorithm is proposed. By optimizing the contact map of the target object, the time of calculating the stable grasping point is reduced. The contact map is meshed for the model of the target object to reduce the number of contact points to improve the real-time performance of the algorithm. The YBC dataset is used as a sample for training and testing. Finally, the effectiveness of this algorithm is verified by simulation experiments with other three-finger closed grasping algorithms. Compared with other three-finger force closure grasping algorithms, this algorithm and the traditional force closure algorithm are consistent in terms of stability, while the real-time and success rates of this algorithm are significantly improved, and our algorithm balances the real-time and success rates of the algorithm compared with other algorithms.
Authors: Wenhao Bai, Liucun Zhu, Peitao Hong, Liang Li, Jiyuan Liu, Zhenyu Zhang
Institution: Beibu Gulf University
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4414288822