Yanzheng Zhu
Zhixiong Zhong- 1College of Electrical Engineering and Automation, Shandong University of Science and Technology, Qingdao, China
- 2College of Computer and Control Engineering, Minjiang University, Fuzhou, China
Editorial on the Research Topic
Hybrid intelligent algorithms based learning, optimization, and application to autonomic control systems, volume II
With the advent of the age of artificial intelligence, a wealth of intelligent algorithms, such as genetic algorithm (Li et al., 2022), neural networks (Zou et al., 2020), and fuzzy logics (Li et al., 2019), etc., have been widely used in various fields during the past two decades. Specifically, thanks to the natures of the issue itself and sensor/actuator restrictions, continuous control is not appropriate and cannot be fulfilled, a possible selective in such circumstances is to incorporate logic-based decisions into the control law and perform switching among a family of controllers (Lin and Antsaklis, 2022). Nowadays, endless developments have been appeared to carry out the analysis and design of dynamic systems with typical hybrid dynamic characteristics. Unfortunately, the practical applications in the different fields, such as autonomous vehicles (Zhang et al., 2021), rehabilitation robots (Allen et al., 2022), and mechatronic systems (Liu et al., 2022), are of lack by using the existing hybrid intelligent algorithms to a great degree. Besides, it is not always easy to embed these hybrid intelligent algorithms to the application research and actual development of industrial installations.
More recently, various intelligent methodology and hybrid control technology have been developed for autonomic control systems in the literature (see Na et al., 2019; Allen et al., 2022; Che et al., 2022 and reference therein), which also have been applied to different scenarios in practice, including aerial vehicles (Kumar and Michael, 2012), permanent magnet synchronous machine (Egidio et al., 2022), and active magnetic bearing (Che et al., 2022), etc. Among them, it is desired that the sophisticated hybrid intelligent algorithms can be adapted to the learning-based optimization and control design to increase the autonomous performance of equipment. In addition, the exploration on the connection between the practical applications and the theoretical methodologies and technologies is very attractive with the aid of existing results based on learning and/or switching control approaches. This Research Topic presents a further collection about the learning and optimization of dynamic systems via hybrid intelligent algorithms, as well as their applications to the autonomic control systems.
Wang et al. propose a realistic module of traffic simulating to generate authentic traffic flow in the Test scenario, and to design a Hi-Fi truck model which is evaluated to imitate the actual truck response in the real world. Then, an AI planning module is established through a learning-based decision algorithm and a multi-mode trajectory planner, simultaneously considering the truck's restrictions, the road slope variations, and the environmental traffic flow. Finally, an automatic drive truck system is realized for road transport, which is the first attempt on the design of open-sourced full automatic drive truck system for logistic operation and volume-produce.
Consider a class of functional electrical stimulation (FES)-cycling system with unknown time-varying input delays, Tong and Zhu construct an Lyapunov-Krasovskii functional to investigate the stability and robustness of the presented system. By using the switching control approach, the considered rider-tricycle system is firstly decomposed into two subsystems. To avoid the chattering and destabilizing as high frequency switching occurs between FES and motor control, a novel average dwell time constraint is introduced to ensure the input-to-state stability (ISS) of the presented systems, and then the corresponding ISS condition is obtained for the augmented system. Finally, the performance of the designed state-feedback controller is testified via the simulation example under a wide range of time-varying delays, including the robustness even the time-varying input delays reach to 250 ms.
In reality, the application of micro-robots in medicine can break through the weaknesses and limitations of numerous conventional clinical approaches. In order to make the micro-robot smarter while passing through blood vessels, Huan et al. extract the skeleton of vascular images firstly. Then, a kind of skeleton-extraction-based A*algorithm is developed to determine an optimum route for the movement of micro-robots at a safe distance from the blood vessel wall. Moreover, the well-known gradient descent algorithm is borrowed to realize the smoothing of the planning paths, which results in a safe and smooth path of the micro-robots under the blood vessel environment.
Zhang et al. exploit a Multi-Layer Convolutional Neural Network (i.e., ResNet-18) and Long Short-Term Memory (LSTM) Networks model to perform the dynamic gesture recognition. A group of velocity-range Doppler images, which are transformed from the original signal, are generated as the input of the model. In particular, ResNet-18 is employed to extract the spatial features at a deeper level and to cope with the issue of gradient extinction/explosion, and LSTM is borrowed to extract temporal features and to address the issue of long-time dependence. Finally, the dynamic gesture recognition experiment is made on the Soli Dataset to implement the proposed model, and the degree of gesture recognition accuracy arrives at 92.55%.
To provide the real-time regolabile magnetic field distribution in its workspace through external programmable current suppliers, a novel quadrupole electromagnetic drive system is established by Ma et al. which is made up of four electromagnetic coils, each coil being energized via an independent DC power supplier. The system structure is constructed to accomplish an adjustable workspace and the parameters of the system are optimized via the parametric modeling approach and ANSYS. Moreover, a magnetic field map is created to promptly obtain the expected driving current from the needed magnetic flux density. Finally, experiments are set up for manipulation of micro-particles with the developed machinery.
Author contributions
All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.
Funding
This Research Topic has been supported in part by National Natural Science Foundation of China (62222310 and 61973131), in part by the Research Fund for the Taishan Scholar Project of Shandong Province of China, and in part by the Fujian Outstanding Youth Science Fund under Grant 2020J06022.
Acknowledgments
All of the topic editors would like to thank all authors for providing high-quality manuscripts on this Research Topic in a very timely fashion. We are also indebted to the referees for their time and valuable comments, criticism, and suggestions. Finally, we very much appreciate their support, advice, and assistance of Journal Development Specialists Hannah Mellor and Oliver Boylan, and our Frontiers in Neurorobotics team to successfully organize this Research Topic.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
Allen, B. C., Stubbs, K. J., and Dixon, W. E. (2022). Data-based and opportunistic integral concurrent learning for adaptive trajectory tracking during switched FES-induced biceps curls. IEEE Transact. Neural Syst. Rehabil. Eng. 30, 2557–2566. doi: 10.1109/TNSRE.2022.3204247
Che, J., Zhu, Y., He, X., and Zhou, D. (2022). Active fault tolerant control design using switching linear parameter varying controllers with inexact fault-effect parameters. Int. J. Robust Nonlinear Control 32, 4477–4494. doi: 10.1002/rnc.6032
Egidio, L. N., Deaecto, G. S., Hespanha, J. P., and Geromel, J. C. (2022). Trajectory tracking for a class of switched nonlinear systems: application to PMSM. Nonlinear Anal. Hybrid Syst. 44, 101164. doi: 10.1016/j.nahs.2022.101164
Kumar, V., and Michael, N. (2012). Opportunities and challenges with autonomous micro aerial vehicles. Int. J. Rob. Res. 31, 1279–1291. doi: 10.1177/0278364912455954
Li, Y., Sun, K., and Tong, S. (2019). Observer-based adaptive fuzzy fault-tolerant optimal control for SISO nonlinear systems. IEEE Trans. Cybern. 49, 649–661. doi: 10.1109/TCYB.2017.2785801
Li, Z., Yu, X., Qiu, J., and Gao, H. (2022). Cell division genetic algorithm for component allocation optimization in multi-functional placers. IEEE Transact. Ind. Informat. 18, 559–570. doi: 10.1109/TII.2021.3069459
Lin, H., and Antsaklis, P. J. (2022). Hybrid Dynamical Systems: Fundamentals and Methods. Switzerland: Springer.
Liu, Z., Lin, W., Yu, X., Rodríguez-Andina, J. J., and Gao, H. (2022). Approximation-free robust synchronization control for dual-linear-motors-driven systems with uncertainties and disturbances. IEEE Transact. Ind. Electron. 69, 10500–10509. doi: 10.1109/TIE.2021.3137619
Na, J., Li, G., Wang, B., Herrmann, G., and Zhan, S. (2019). Robust optimal control of wave energy converters based on adaptive dynamic programming. IEEE Transact. Sustain. Energy 10, 961–970. doi: 10.1109/TSTE.2018.2856802
Zhang, L., Zhang, R., Wu, T., Weng, R., Han, M., Zhao, Y., et al. (2021). Safe reinforcement learning with stability guarantee for motion planning of autonomous vehicles. IEEE Transact. Neural Netw. Learn. Syst. 32, 5435–5444. doi: 10.1109/TNNLS.2021.3084685
Keywords: autonomous control systems, intelligent algorithms, optimization, robotics, switching control
Citation: Zhu Y and Zhong Z (2023) Editorial: Hybrid intelligent algorithms based learning, optimization, and application to autonomic control systems, volume II. Front. Neurorobot. 17:1151473. doi: 10.3389/fnbot.2023.1151473
Received: 26 January 2023; Accepted: 02 February 2023;
Published: 20 February 2023.
Edited and reviewed by: Alois C. Knoll, Technical University of Munich, Germany
Copyright © 2023 Zhu and Zhong. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Yanzheng Zhu, 
yanzhengzhu@sdust.edu.cn