[1]曾旭翔,孔 颖.求解时变二次规划的自适应参数归零神经网络[J].浙江科技大学学报,2024,(05):384-393.[doi:10.3969/j.issn.2097-5236.2024.05.004]
 ZENG Xuxiang,KONG Ying.Adaptive parameter zeroing neural network for solving time-varying quadratic programming[J].,2024,(05):384-393.[doi:10.3969/j.issn.2097-5236.2024.05.004]
点击复制

求解时变二次规划的自适应参数归零神经网络(/HTML)
分享到:

《浙江科技大学学报》[ISSN:2097-5236/CN:33-1431/Z]

卷:
期数:
2024年05期
页码:
384-393
栏目:
出版日期:
2024-10-28

文章信息/Info

Title:
Adaptive parameter zeroing neural network for solving time-varying quadratic programming
文章编号:
2097-5236(2024)05-0384-10
作者:
曾旭翔孔 颖
(浙江科技大学 信息与电子工程学院,杭州 310023)
Author(s):
ZENG Xuxiang KONG Ying
(School of Information and Electronic Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, Zhejiang, China)
关键词:
归零神经网络 时变二次规划 自适应参数 预设时间
分类号:
TP183
DOI:
10.3969/j.issn.2097-5236.2024.05.004
文献标志码:
A
摘要:
【目的】针对时变二次规划(time-varying quadratic programming,TVQP)中的时变参数求解问题,提出了一种自适应参数归零神经网络(adaptive parameter zeroing neural network,APZNN)模型。【方法】首先,在归零神经网络(zeroing neural network,ZNN)模型的基础上引入一种基于误差的自适应参数及增强型双幂(enhanced sign-bi-power,ESBP)激活函数,从而提出了APZNN模型; 然后,利用李雅普诺夫定理分析了APZNN模型的稳定性,预设时间收敛性和鲁棒性; 最后,通过仿真试验以验证APZNN模型的有效性。【结果】在求解时变二次规划问题时,APZNN模型相比ZNN模型和时变参数归零神经网络(time-varying parameters zeroing neural network,TVPZNN)模型,具有更快的收敛速度和更强的鲁棒性,其误差函数能在0.2 s内收敛到0; 得益于自适应参数的引入,APZNN模型在仿真试验中的计算时间较TVPZNN模型减少了16.6 s,节省了计算资源。此外,将APZNN模型应用于UR5机械臂轨迹跟踪试验中,机械臂的末端执行器可以很好地跟踪期望的路径,末端执行器的位置误差被限制在-1.5×10-4 m和1.5×10-4 m之间,这进一步说明模型的可行性。【结论】本研究提出的APZNN模型能够有效地求解时变二次规划问题,可为神经网络模型设计提供参考。

参考文献/References:

[1] HU Z, XIAO L, DAI J H, et al. A unified predefined-time convergent and robust ZNN model for constrained quadratic programming[J]. IEEE Transactions on Industrial Informatics,2021,17(3):1998.
[2] 孔颖,吴佳佳,雷景生,等.基于有限时间对偶神经网络的冗余机械臂重复运动规划[J].控制理论与应用,2023,40(1):139.
[3] 金龙,张凡,刘佰阳,等.基于数据驱动的冗余机器人末端执行器位姿控制方案[J].自动化学报,2024,50(3):518.
[4] 陈思远.基于神经动力学的冗余度机器人混合优化策略[D].广州:华南理工大学,2022.
[5] MATTINGLEY J, BOYD S. Real-time convex optimization in signal processing[J]. IEEE Signal Processing Magazine,2010,27(3):50.
[6] WANG P, SHEN C H, HENGEL A V D, et al. Large-scale binary quadratic optimization using semidefinite relaxation and applications[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(3):470.
[7] XU H K. An iterative approach to quadratic optimization[J]. Journal of Optimization Theory and Applications,2003,116(3):659.
[8] GUO D S, ZHANG Y N. Neural dynamics and Newton-Raphson iteration for nonlinear optimization[J]. Journal of Computational and Nonlinear Dynamics,2014,9(2):021016.
[9] FENG J Q, QIN S, SHI F L, et al. A recurrent neural network with finite-time convergence for convex quadratic bilevel programming problems[J]. Neural Computing and Applications,2018,30(11):3399.
[10] GUO D S, YI C F, ZHANG Y N. Zhang neural network versus gradient based neural network for time-varying linear matrix equation solving[J]. Neurocomputing,2011,74(17):3708.
[11] LIAO B L, ZHANG Y N. Different complex ZFs leading to different complex ZNN models for time-varying complex generalized inverse matrices[J]. IEEE Transactions on Neural Networks and Learning Systems,2014,25(9):1621.
[12] ZHANG Z J, ZHENG L N, QIU T R, et al. Varying-parameter convergent-differential neural solution to time-varying overdetermined system of linear equations[J]. IEEE Transactions on Automatic Control,2020,65(2):874.
[13] JIN J, ZHAO L, LI M, et al. Improved zeroing neural networks for finite time solving nonlinear equations[J]. Neural Computing and Applications,2020,32:4151.
[14] SUN M X, LI X, ZHONG G M. Performance enhancing ZNN models for time-variant equality-constraint convex optimization solving:a transition-state based attracting system approach[J]. Expert Systems with Applications,2024,237:121559.
[15] LI S, CHEN S F, LIU B. Accelerating a recurrent neural network to finite-time convergence for solving time-varying Sylvester equation by using a sign-bi-power activation function[J]. Neural Processing Letters,2013,37(2):189.
[16] XIAO L, HE Y J, LIAO B L. A parameter-changing zeroing neural network for solving linear equations with superior fixed-time convergence[J]. Expert Systems with Applications,2022,208:118086.
[17] JIN J, CHEN W J, CHEN C Y, et al. A predefined fixed-time convergence ZNN and its applications to time-varying quadratic programming solving and dual-arm manipulator cooperative trajectory tracking[J]. IEEE Transactions on Industrial Informatics,2023,19(8):8691.
[18] ZHENG L N, ZHANG Z J. Time-varying quadratic-programming-based error redefinition neural network control and its application to mobile redundant manipulators[J]. IEEE Transactions on Automatic Control,2022,67(11):6151.

备注/Memo

备注/Memo:
收稿日期:2024-04-25
基金项目:浙江省自然科学基金项目(LZY22E050002)
通信作者:孔 颖(1980— ),女,浙江省杭州人,教授,博士,主要从事神经网络与机械臂轨迹规划研究。E-mail:kongying@zust.edu.cn。
更新日期/Last Update: 2024-10-28