Open AccessTarget Locating and Grabbing Algorithm for Robot Prosthesis Based on PSO-RBF
Author(s) -
Jianbo Zhou,
Chuanjiang Wang,
Xiujuan Sun,
Kunpeng Wang,
Zhengkai Feng
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1693/1/012195
Subject(s) - workspace , computer vision , position (finance) , artificial intelligence , computer science , robot , engineering , simulation , finance , economics
Aiming at the target locating and grabbing problem for a five-degree-of-freedom intelligent robot prosthesis worn by the disabled, an algorithm is proposed. Firstly, the laser rangefinder and attitude sensor installed at the head of the wearer are used to simulate the motion of the human eye to realize the position of the target in the prosthesis workspace. Secondly, the laser ranging information is obtained by the 3D attitude sensor installed at the wearer’s shoulder to obtain the position coordinates of the target relative to the prosthesis shoulder, so as to obtain the position coordinates of the target in the prosthesis workspace. Finally, the PSO-RBF neural network is used to train the samples to obtain the nonlinear mapping relationship between the position coordinates and the joint angle, thus predicting the joint angle and grabbing the target. The simulation and experimental results show that this method can locate the target accurately and calculate the angle of each joint of the prosthesis in real-time, so as to complete the action of target grabbing.