A multi-agent motion simulation method for emergency scenario deduction DOI
Jiale Wang, Zhen Liu, Tingting Liu

et al.

Cognitive Systems Research, Journal Year: 2024, Volume and Issue: 88, P. 101275 - 101275

Published: Aug. 15, 2024

Language: Английский

Multisource-Knowledge-Based Approach for Crowd Evacuation Navigation DOI
P. Zhang, Kun Zhao, Hong Liu

et al.

IEEE Transactions on Computational Social Systems, Journal Year: 2024, Volume and Issue: 11(3), P. 4524 - 4539

Published: April 30, 2024

In crowd evacuation research, the knowledge contained in is very complex and multisource. Crowd scenarios restrict pedestrians' movement decision-making, states of imply characteristics. However, existing studies on navigation approach cannot make full use multisource knowledge, which reduces effect navigation. To solve this problem, a new based proposed. First, we collect relevant data using an image sensor network establish graph to organize store data. Second, explicit scene structure movements represented graph. Then, deep-learning-based tacit model (DLTKM) designed extract different groups entities. Finally, wireless related representations plan paths for evacuees. The experiment results show that can reasonable pedestrians, improve efficiency evacuations.

Language: Английский

Citations

3
Use of Knowledge Graphs for Construction Safety Management: A Systematic Literature Review DOI Creative Commons
Fansheng Kong, Seungjun Ahn

Information, Journal Year: 2024, Volume and Issue: 15(7), P. 390 - 390

Published: July 3, 2024

Effective safety management is crucial in the construction industry. The growing interest employing Knowledge Graphs (KGs) for driven by need efficient computing-aided practices. This paper systematically reviews literature related to automating processes through knowledge base systems, focusing on creation and utilization of KGs safety. It captures current methodologies developing using management, outlining techniques each phase KG development, including scope identification, integration external data, ontological modeling, data extraction, completion. provides structured guidance building a management. Moreover, this discusses challenges limitations that hinder wider adoption leading identification goals considerations future research.

Language: Английский

Citations

3
Crowd evacuation path planning and simulation method based on deep reinforcement learning and repulsive force field DOI
Hongyue Wang, Liu Hong, Wenhao Li

et al.

Applied Intelligence, Journal Year: 2025, Volume and Issue: 55(4)

Published: Jan. 11, 2025

Language: Английский

Citations

0
QLGWYB: design of an efficient model for analyzing crowd behavior through Quad LSTM and Quad GRU fusion enhanced by Q-learning and YOLO DOI
Lokesh Heda, Parul Sahare

Iran Journal of Computer Science, Journal Year: 2025, Volume and Issue: unknown

Published: May 8, 2025

Language: Английский

Citations

0
Safety Indication Signs-Based Crowd Division and Leader Selection Approach for Evacuation Guidance DOI
Liang Li, Chen Lyu, Hong Liu

et al.

IEEE Sensors Journal, Journal Year: 2024, Volume and Issue: 24(6), P. 8937 - 8948

Published: Jan. 31, 2024

The safety indication signs (SISs) are indispensable direction environmental information indicators for crowd evacuation. However, given that pedestrians have different acceptance and understanding of information, the application effect SISs in evacuation management is still limited. This article proposed a new division leader selection algorithm based on leader–follower phenomenon to improve guiding SISs. configuration SIS similar sensing area sensors. According guide role route selection, we divide into leaders followers pedestrians' visual distance field. A modified PRM consider path planning. behavior decided by movement-driving approach combines reciprocal velocity obstacle (RVO) model social force (SFM). can adapt sensors management. Several simulations conducted verify scenarios evaluate results terms efficiency reasonableness SISs' placement. result illustrates appropriate large outdoor scenes or indoor buildings.

Language: Английский

Citations

2
KDEM: A Knowledge-Driven Exploration Model for Indoor Crowd Evacuation Simulation DOI

Yuji Shen,

Bohao Zhang, Chen Li

et al.

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 262 - 274

Published: Jan. 1, 2024

Language: Английский

Citations

1
Artificial Intelligence Methodologies for Building Evacuation Plan Modeling DOI
Rodrigo Ternero, Guillermo Fuertes, Miguel Alfaro

et al.

Journal of Building Engineering, Journal Year: 2024, Volume and Issue: 96, P. 110408 - 110408

Published: Aug. 13, 2024

Language: Английский

Citations

1
A Multi-Agent Motion Simulation Method for Emergency Scenario Deduction DOI

Jiale Wang,

Zhen Liu, Tingting Liu

et al.

Published: Jan. 1, 2024

Simulating crowd motion in emergency scenarios remains a challenge computer graphics due to heterogeneity and environmental complexity. However, existing simulation methods homogenize the agent model simplify target selection navigation of crowds. To address these problems, we propose multi-agent method for scenario deduction. First, simulate heterogeneity. This includes personality-based heterogeneous an perception that considers vision, hearing, familiarity with environment. Second, strategy based on patterns actual pedestrians. employs mathematical models our guide agents selecting appropriate targets. Finally, global algorithm combines random sampling heuristic search methods. Concurrently, use adjust agent's local planning naturally deduce states Experimental results validate can realistically reasonably scenarios.

Language: Английский

Citations

0
A multi-agent motion simulation method for emergency scenario deduction DOI
Jiale Wang, Zhen Liu, Tingting Liu

et al.

Cognitive Systems Research, Journal Year: 2024, Volume and Issue: 88, P. 101275 - 101275

Published: Aug. 15, 2024

Language: Английский

Citations

0