International Journal of Network Dynamics and Intelligence

Emotion Contagion Model for Dynamical Crowd Path Planning

2024-09-24T17:12:40+08:00

Article

Emotion Contagion Model for Dynamical Crowd Path Planning

Yunpeng Wu ¹, Xiangming Huang ¹, Zhen Tian ¹, Xiaoqiang Yan ^1,*, and Hui Yu ^2,*

¹ School of Computer and Artificial Intelligence, Zhengzhou University, Zhengzhou 450001, China

² cSCAN, University of Glasgow, G12 8QB, United Kingdom

* Correspondence: iexqyan@zzu.edu.cn; hui.yu@glasgow.ac.uk

Received: 17 December 2023

Accepted: 23 April 2024

Published: 24 September 2024

Abstract: Crowd path planning aims to find the optimal path between the source and destination for multiple agents in crowd scenes. The advent of parallel theory and digital twin technologies provides a novel platform for simulating crowd path planning, which has become increasingly popular in various applications, such as pedestrian evacuation, intelligent transportation, and civil planning. The widely used strategy for crowd path planning emphasizes the objective factors, such as user-specific guidance, shortest path and crowd density. However, this strategy ignores the subjective emotion of agents, which can have significant impact on the diverse path choices of each agent. To tackle this challenge, we present a novel Emotion Contagion Model (ECM) to dynamically conduct path planning in crowded environments by incorporating the emotion of each agent. The proposed method provides a solution to the long-standing high-level affective issue for virtual agents during path search. Firstly, to bridge the gap between emotion states and path choices, the emotion preference is defined based on personality traits of multiple agents. Secondly, an emotion contagion algorithm is proposed to recognize the collective patterns of these agents, which can reveal the dynamical variation of emotion preference under crowded complex environments. Finally, to solve the emotion-to-path mapping, we propose a leastexpected-time objective function to find the optimal path choice for each agent according to the navigation graph in the given scenario. Experimental results on various scenarios, including the subway station, railway station square, fire evacuation and indoor environment, verify the effectiveness of the ECM compared with the state-of-the-art methods.

Guaranteed Cost Intermittent Control for Discrete-Time System: A Data-Driven Method

2024-09-24T18:54:18+08:00

Article

Guaranteed Cost Intermittent Control for Discrete-Time System: A Data-Driven Method

Yi Zou and Engang Tian *

School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

* Correspondence: tianengang@163.com

Received: 10 September 2023

Accepted: 27 November 2023

Published: 24 September 2024

Abstract: This paper explores a data-driven method to investigate the stabilization of intermittent controlled discrete-time systems (ICDTSs) with unknown parameter matrices. First, the pre-collected inputstate data is used to supersede the accurate prior system model. Then, in order to obtain the data-dependent stabilization conditions of ICDTSs, a novel relationship is designed among the control width, rest width, and convergence rate. Unlike existing studies on the stabilization of ICDTSs, this paper only needs the collected input-state data. Thus, the time-consuming process of model identification is avoided. In addition, to ensure an acceptable performance level, the data-based guaranteed cost control is also considered, and a new cost function for ICDTSs is correspondingly built. Finally, two simulations are presented to demonstrate the effectiveness of the theoretical analysis.

Multi-Dimensional Adaptive Learning Rate Gradient Descent Optimization Algorithm for Network Training in Magneto-Optical Defect Detection

2024-09-25T14:15:41+08:00

Article

Multi-Dimensional Adaptive Learning Rate Gradient Descent Optimization Algorithm for Network Training in Magneto-Optical Defect Detection

Yiping Liang, Lulu Tian *, Xu Zhang, Xiao Zhang, and Libing Bai

School of Automation Engineering, University of Electronic Science and Technology of China, Sichuan 611731, China

^* Correspondence: lulutian@uestc.edu.cn

Received: 24 September 2023

Accepted: 8 November 2023

Published: 25 September 2024

Abstract: As an optimization technique, the gradient descent method is widely adopted in the training process of deep learning. In traditional gradient descent methods, the gradient of each dimension has the same weight wtih the updating direction, which results in poor performances when there are multiple small gradient dimensions (e.g. near the saddle point). To improve the accuracy and convergence speed of the neural network training, we propose a novel multi-dimensional adaptive learning rate gradient descent optimization algorithm (M-AdaGrad) in this paper. Specifically, in the M-AdaGrad, the learning rate will be updated according to a newly designed weight function related to the current gradient. Experiments on a set of sigmoid-based functions verify that, compared with traditional gradient descent methods such as AdaGrad and Adam, the M-AdaGrad gives more confidence to the larger gradient direction and has a larger probability to reach a more optimal position with a faster speed. Due to its excellent performance in network training, the M-AdaGrad is successfully applied to the magneto-optical nondestructive test of crack detection based on the generative adversarial network.

Many-Objective Simulation Optimization for Camp Location Problems in Humanitarian Logistics

2024-09-26T15:17:09+08:00

Article

Many-Objective Simulation Optimization for Camp Location Problems in Humanitarian Logistics

Yani Xue ^1,*, Miqing Li ², Hamid Arabnejad ¹, Diana Suleimenova ¹, Alireza Jahani ¹, Bernhard C. Geiger ³, Freek Boesjes ⁴, Anastasia Anagnostou ¹, Simon J.E. Taylor ¹, Xiaohui Liu ¹, and Derek Groen ^1,*

¹ Department of Computer Science, Brunel University London, Uxbridge, United Kingdom

² School of Computer Science, University of Birmingham, Birmingham, United Kingdom

³ Know-Center GmbH, Graz, Austria

⁴ Faculty of Geosciences, Utrecht University, Utrecht, The Netherlands

^* Correspondence: Yani Xue (Yani.Xue3@brunel.ac.uk); Derek Groen (Derek.Groen@brunel.ac.uk)

Received: 10 March 2024

Accepted: 19 August 2024

Published: 26 September 2024

Abstract: Humanitarian organizations face a rising number of people fleeing violence or persecution, people who need their protection and support. When this support is given in the right locations, it can be timely, effective and cost-efficient. Successful refugee settlement planning not only considers the support needs of displaced people, but also local environmental conditions and available resources for ensuring survival and health. It is indeed very challenging to find optimal locations for establishing a new refugee camp that satisfy all these objectives. In this paper, we present a novel formulation of the facility location problem with a simulation-based evolutionary many-objective optimization approach to address this problem. We show how this approach, applied to migration simulations, can inform camp selection decisions by demonstrating it for a recent conflict in South Sudan. Our approach may be applicable to diverse humanitarian contexts, and the experimental results have shown it is capable of providing a set of solutions that effectively balance up to five objectives.

Adaptive Fixed-time Control for Multiple Switched Coupled Neural Networks

2024-09-26T20:39:44+08:00

Article

Adaptive Fixed-time Control for Multiple Switched Coupled Neural Networks

Linhao Zhao, and Boqian Li *

Australian AI Institute, Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, NSW 2007, Australia

^* Correspondence: BoqianLi@student.uts.edu.au (Boqian Li)

Received: 15 July 2024

Accepted: 24 August 2024

Published: 26 September 2024

Abstract: Adaptive control is an effective approach for mitigating undesirable deviations in prescribed closed-loop plant behavior. However, conventional adaptive control methods often exhibit slow responses in various control tasks. This paper introduces a novel adaptive control method to achieve fixed-time synchronization in a class of coupled neural networks. We present coupled neural networks with multiple switching topologies and design a fixed-time adaptive control strategy for this system. Furthermore, we establish a criterion to ensure the fixed-time stability of the closed-loop system. Two numerical examples are provided to demonstrate the effectiveness and accuracy of the theoretical results.

CCE-Net: Causal Convolution Embedding Network for Streaming Automatic Speech Recognition

2024-09-27T14:30:18+08:00

Article

CCE-Net: Causal Convolution Embedding Network for Streaming Automatic Speech Recognition

Feiteng Deng, Yue Ming *, and Boyang Lyu

Beijing Key Laboratory of Work Safety and Intelligent Monitoring, School of Electronic Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China

^* Correspondence: yming@bupt.edu.cn

Received: 11 March 2024

Accepted: 15 August 2024

Published: 27 September 2024

Abstract: Streaming Automatic Speech Recognition (ASR) has gained significant attention across various application scenarios, including video conferencing, live sports events, and intelligent terminals. However, chunk division for current streaming speech recognition results in insufficient contextual information, thus weakening the ability of attention modeling and leading to a decrease in recognition accuracy. For Mandarin speech recognition, there is also a risk of splitting Chinese character phonemes into different chunks, which may lead to incorrect recognition of Chinese characters at chunk boundaries due to incomplete phonemes. To alleviate these problems, we propose a novel front-end network - Causal Convolution Embedding Network (CCE-Net). The network introduces a causal convolution embedding module to obtain richer historical context information, while capturing Chinese character phoneme information at chunk boundaries and feeding it to the current chunk. We conducted experiments on Aishell-1 and Aidatatang. The results showed that our method achieves a character error rate (CER) of 5.07% and 4.90%, respectively, without introducing any additional latency, showing competitive performances.