Cooperative survival principles for underground flooding: Vitae System based multi-agent simulation

Higo, Eishiro; Okada, Norio; Hipel, Keith W.; Fang, Liping

doi:10.1016/j.eswa.2017.04.034

Cited by 27 publications

(5 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another interesting study was conducted by Li et al (2019), which combined a cellular automata model and a ABM system to simulate crowd evacuations in flood disasters, and the model was evaluated by real-participant experiments based on virtual reality (VR) environments. Similar studies on ABM applications in flood evacuation are also found in Higo et al (2017); Lim (2016, 2018); Yamamoto and Takizawa (2019); Nakanishi et al (2019); and Eivazy and Malek (2019).…”

Section: Real-time Flood Emergency Managementsupporting

confidence: 67%

Agent-based modelling and flood risk management: A compendious literature review

Zhuo

Han

2020

Journal of Hydrology

View full text Add to dashboard Cite

The use of agent-based modelling (ABM) to tackle flood-related risk challenges is becoming increasingly popular in recent years. This paper reviews the literature at the interface of ABM and flood-related studies in view of understanding the technique's advantages and limitations to flood risk management, based on a set of 61 representative articles. In particular, to understand how this process-based technique can help to link human (also institutional) decisions and behaviour with flood risks through the whole human-flood systems. Overall, the temporal and spatial distributions demonstrate a growing interest in this research area around the world, especially since 2017. Three topic areas are identified, addressing different research challenges in the field: real-time flood emergency management, long-term flood adaptation planning, and flood hydrological modelling. The review has shown that the potential contribution of ABM to future flood risk management lays in its practical application to decision-making in adaptation policy and strategy planning. The review also critically reveals the limitation of ad hoc implementations of decision-making and behaviour in the ABM models that could make the application less realistic in the field. It is recommended that the future development should be guided/influenced by the continuing development and refinement of ABM modelling framework and theoretical foundations, and enhancement of model testing and documenting capabilities. More importantly, active collaborations between disciplines and sectors such as to involve more social and psychological sciences in ABM decision-making 2 modelling should be encouraged; and knowledge sharing will encourage more effective uses of ABM by wider audiences.

show abstract

Section: Real-time Flood Emergency Managementsupporting

confidence: 67%

Agent-based modelling and flood risk management: A compendious literature review

Zhuo

Han

2020

Journal of Hydrology

View full text Add to dashboard Cite

show abstract

“…There were three entrances in the exits with a relatively large amount of flood water stagnant, namely Exit 4, Exit D, and Exit F. Setting up specific monitoring points 1 m before the entrance midpoint of all emergency safety passages and the three exits, we can monitor the relationship between the water depth changes within 10 min at the population of each evacuation passage. Based on Higo's research on pedestrian movement in floods [46], the pedestrian movement speed v g is defined as follows:…”

Section: Analysis Of Evacuation Passagesmentioning

confidence: 99%

Numerical Simulation of Flood Intrusion Process under Malfunction of Flood Retaining Facilities in Complex Subway Stations

Lin

Zhou

et al. 2022

Buildings

View full text Add to dashboard Cite

In recent years, heavy rain and waterlogging accidents in subway stations have occurred many times around the world. With the comprehensive development trend of underground space, the accidents caused by flood flow intruding complex subway stations and other underground complexes in extreme precipitation disasters will be lead to more serious casualties and property damage. Therefore, it is necessary to conduct numerical simulation of flood intrusion process under malfunction of flood retaining facilities in complex subway stations. In order to prevent floods from intruding subway stations and explore coping strategies, in this study, the simulation method was used to study the entire process of flood intrusion into complex subway stations when the flood retaining facilities fail in extreme rain and flood disasters that occur once-in-a-century. The three-dimensional numerical simulation model was constructed by taking a subway interchange station with a property development floor in Nanning as a prototype. Based on the Volume of Fluid (VOF) model method, the inundated area in the subway station during the process of flood intrusion from the beginning to the basic stability was simulated, and it was found that the property development floor has serious large-scale water accumulation under extreme rainfall conditions. Through the dynamic monitoring of the flood water level depth at important positions such as the entrances of the evacuation passages, and the analysis of the influence of the design structure and location distribution of different passages on the personnel evacuation plan, it was found that the deep water accumulation at the entrances of the narrow, long, and multi-run emergency safety passages are not conducive to the evacuation of personnel. Finally, the flow of flood water into the subway tunnel through the subway station was calculated. The research results provide certain reference and guidance for the safety design of subway stations under extreme rainfall climatic conditions.

show abstract

“…Because of a limited perception of the entire environment of each agent in the system, agents need to cooperate with one another, follow specific social rules, and make joint decisions that aim to simulate the cognitive process of human beings. MAS captures the dynamics of natural and human activities in a cooperative way, which makes it a suitable solution to solve the problem of crowd evacuation [15]- [17].…”

Section: Related Work a Multi-agent Reinforcement Learningmentioning

confidence: 99%

Improved Multi-Agent Deep Deterministic Policy Gradient for Path Planning-Based Crowd Simulation

Zheng¹,

Liu²

2019

IEEE Access

View full text Add to dashboard Cite

Deep reinforcement learning (DRL) has been proved to be more suitable than reinforcement learning for path planning in large-scale scenarios. In order to more effectively complete the DRL-based collaborative path planning in crowd evacuation, it is necessary to consider the space expansion problem brought by the increase of the number of agents. In addition, it is often faced with complicated circumstances, such as exit selection and congestion in crowd evacuation. However, few existing works have integrated these two aspects jointly. To solve this problem, we propose a planning approach for crowd evacuation based on the improved DRL algorithm, which will improve evacuation efficiency for large-scale crowd path planning. First, we propose a framework of congestion detection-based multi-agent reinforcement learning, the framework divides the crowd into leaders and followers and simulates leaders with a multi-agent system, it considers the congestion detection area is set up to evaluate the degree of congestion at each exit. Next, under the specification of this framework, we propose the improved Multi-Agent Deep Deterministic Policy Gradient (IMADDPG) algorithm, which adds the mean field network to maximize the returns of other agents, enables all agents to maximize the performance of a collaborative planning task in our training period. Then, we implement the hierarchical path planning method, which upper layer is based on the IMADDPG algorithm to solve the global path, and lower layer uses the reciprocal velocity obstacles method to avoid collisions in crowds. Finally, we simulate the proposed method with the crowd simulation system. The experimental results show the effectiveness of our method.INDEX TERMS Deep reinforcement learning, multi-agent reinforcement learning, path planning, crowd simulation for evacuation, improved multi-agent deep deterministic policy gradient algorithm.

show abstract

Cooperative survival principles for underground flooding: Vitae System based multi-agent simulation

Cited by 27 publications

References 66 publications

Agent-based modelling and flood risk management: A compendious literature review

Agent-based modelling and flood risk management: A compendious literature review

Numerical Simulation of Flood Intrusion Process under Malfunction of Flood Retaining Facilities in Complex Subway Stations

Improved Multi-Agent Deep Deterministic Policy Gradient for Path Planning-Based Crowd Simulation

Contact Info

Product

Resources

About