Open AccessInverse kinematics solution and motion simulation of seven-degree-of-freedom ascending platform based on neural network
Author(s) -
Jingyuan Wu,
Xiangyi Ren
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 - 10.1088/1742-6596/1650/3/032127
Subject(s) - moore–penrose pseudoinverse , artificial neural network , singularity , inverse kinematics , control theory (sociology) , kinematics , norm (philosophy) , genetic algorithm , position (finance) , computer science , stewart platform , mathematics , inverse , robot , mathematical optimization , artificial intelligence , mathematical analysis , physics , law , geometry , classical mechanics , control (management) , finance , political science , economics
A pseudoinverse-based inverse kinetic solving scheme for 7-degree-of-freedom redundant aerial platform robotic arm, the scheme combines Minimum Velocity Norm scheme and damped least-square method, a close-loop damped minimum velocity norm scheme is set up which allows to avoid the joints’ limitation and to ensure the solution’s stability near the singularity position. Further, a genetic algorithm optimized BP neural network is built. Genetic algorithm can improve the training efficiency and avoid trapping in local optimal solution. The application of neural network can ensure the real-time of solution.