A Hierarchical Learning Control Framework for an Aerial Manipulation System

4th International Conference on Mechanics and Mechatronics Research

机械制造;无线电电子学

Ma, Le^1 ; Chi, Yanxun^1 ; Li, Jiapeng^1 ; Li, Zhongsheng^1 ; Ding, Yalei^1 ; Liu, Lixing^1

Robotics Institution, School of Automation Engineering, Northeast Electric Power University, China^1

关键词: Flight platforms;    Hierarchical learning;    Manipulation system;    Mechanical design;    Newton Euler equation;    RBF(radial basis function);   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/224/1/012030/pdf
DOI  :  10.1088/1757-899X/224/1/012030

来源: IOP

【 摘 要 】

A hierarchical learning control framework for an aerial manipulation system is proposed. Firstly, the mechanical design of aerial manipulation system is introduced and analyzed, and the kinematics and the dynamics based on Newton-Euler equation are modeled. Secondly, the framework of hierarchical learning for this system is presented, in which flight platform and manipulator are controlled by different controller respectively. The RBF (Radial Basis Function) neural networks are employed to estimate parameters and control. The Simulation and experiment demonstrate that the methods proposed effective and advanced.