Modified extreme learning machine based motion control of robotic manipulators

Author:

Kapoor Neha,Ohri Jyoti

Abstract

Abstract Precise trajectory tracking is a difficult task due to partially known and unknown dynamics and the disturbances present in the system. For improving the tracking performance of the robotic manipulator, this work proposes a novel PSO optimized kernel based Extreme Learning Machine (PSO-KELM) learning algorithm in which PSO is used to get the optimal values of the free kernel-parameters in KELM. The simulation results represent the good generalized performance and PSO-KELM outperforms the KELM and ELM based control techniques for the manipulator trajectory tracking. Comparative analysis of the proposed control schemes have been done with NN and SVM based controllers and various ELM based variants for the trajectory tracking problem in robotic manipulator.

Publisher

IOP Publishing

Subject

General Medicine

Reference26 articles.

1. A neural network compensator for uncertainties of robotic manipulators;Ishiguro;IEEE Transactions on Industrial Electronics,1992

2. Use of a recurrent neural network in discrete sliding-mode control;Fang;IEE—Control Theory and Applications,1999

3. Neural-network hybrid control for antilock braking systems;Lin;IEEE Transactions on Neural Networks,2003

4. Universal approximation using incremental constructive feedforward networks with random hidden nodes;Huang;IEEE Transactions on Neural Networks

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3