Parameter Optimization of Cartesian Serial-Parallel Manipulator with Redundant Actuation Based on BP Neural Networks

Abstract:

Design the redundant driving structure based on a new type of universal Cartesian serial-parallel manipulator, structural parameters are optimized through objective function, relative workspace and accuracy of motion included as the main index, four BP neural networks are constructed with reiterative exercise carried out, transferred function and input parameters through the constructed and exercised neural networks, analysis and approach computation completed with pretty optimal data suggested.

• Periodical: Applied Mechanics and Materials (Volumes 602-605)
• Pages: 908-911
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.602-605.908
• Online since: August 2014
• Authors: Jian Guo Luo*, Mao Yan He
• Keywords: Neural Network (NN), Parameter, Redundant Driving, Serial-Parallel

References

[1] Wang Yuxin and so on. Spatial layout of complex mechanical systems [J]. Chinese Journal of mechanical engineering . 2005, 41(2)6-13.

[2] Kourosh E and so on. Instantaneous Kinematics and Design of a Novel Redundant Parallel Manipulator IEEE, (1994).

[3] Xie Jin and so on. Based on BP neural network for kinematic synthesis of planar linkage [J]. Chinese Journal of mechanical engineering . 2005, 41(2)24-27.

[4] Wang Jinsong. VAMITIY virtual axis machine tool [J]. Manufacturing technology and machine tool. 1998, (2)42-43.

[5] Huang Zhen. Spatial mechanism [M]. Beijing: Mechanical Industry Press. (1991).

[6] Yang Yuhu and so on. The position error of the mechanism analysis of transfer matrix method[J]. Chinese Journal of mechanical engineering . 2005, 41(2)20-23.