Pedestrian Model Research Articles

Application of Agent-Based Modelling for Evaluating Level of Service in Airports

Airports worldwide evaluate Level of service (LOS) to assess performance of the terminal for the current scenario and to compare it to a projected scenario. The International Air Transport Association (IATA) global standard is the widely accepted framework that evaluate airport LOS based on space and time characteristics. The application of simulation tools can be used in this regard to depict the airport LOS scenarios more interactively and in detail. Moreover, this approach also enables modelling future scenarios of technological or infrastructure modifications to cater terminal capacity constraints. The use of Agent-Based Modelling (ABM) in evaluating airport LOS is discussed in this paper. The study utilized field study at the check-in hall of a large-sized airport, the Chennai International airport, India to capture the passenger flow characteristics such as passenger entry rate, walking speed, group size, passenger flow sequence and flight movement frequency. These are used as inputs to a microscopic pedestrian modelling tool, AnyLogic. Three scenarios of airport LOS are modelled to evaluate the passenger density and the dwell times. The results show that the variation in number of service points and flow rate significantly affect the passenger dwell time, queuing length, queuing time and passenger density at the check-in zone. Visual and quantitative representations of the scenarios are also presented, analyzed and compared with the existing standard.

Pedestrian Simulation with Reinforcement Learning: A Curriculum-Based Approach

Pedestrian simulation is a consolidated but still lively area of research. State of the art models mostly take an agent-based perspective, in which pedestrian decisions are made according to a manually defined model. Reinforcement learning (RL), on the other hand, is used to train an agent situated in an environment how to act so as to maximize an accumulated numerical reward signal (a feedback provided by the environment to every chosen action). We explored the possibility of applying RL to pedestrian simulation. We carefully defined a reward function combining elements related to goal orientation, basic proxemics, and basic way-finding considerations. The proposed approach employs a particular training curriculum, a set of scenarios growing in difficulty supporting an incremental acquisition of general movement competences such as orientation, walking, and pedestrian interaction. The learned pedestrian behavioral model is applicable to situations not presented to the agents in the training phase, and seems therefore reasonably general. This paper describes the basic elements of the approach, the training procedure, and an experimentation within a software framework employing Unity and ML-Agents.

Inverse augmentation: Transposing real people into pedestrian models

We introduce a scheme for immersing real human users in urban simulations, and for enabling them to transpose their embodied behavior into models. We achieve this by inverse augmentation, flipping traditional philosophies of augmented reality. Rather than beginning with real-world scenes and embellishing them with graphics, we proceed from a base of synthetic, modeled, streetscapes filled with agent characters, which we augment with real human users. Participants are then allowed to use their natural abilities to explore the simulation scenarios. We achieve this by employing mobile virtual reality to allow users to build dynamic presence in a fused geosimulation and virtual geographic environment that they can physically view and walk around in. Our central argument is that inversion of this kind allows for the detail and nuances of human behavior to be brought directly into simulation, where they would traditionally be difficult to capture and represent. We show that close matches between real physical activity on the ground and actions in the model world can be achieved, as measured by spatial analysis and encephalography of user brain activity. We demonstrate the usefulness of the approach with an application to studying pedestrian road-crossing behavior.

Developing Pedestrian Capacity Based on Movement Simulation Model: The Case of Malioboro, Yogyakarta

The activity of pedestrian movement can be predicted by a pedestrian model. Pedestrian model would assign pedestrian lane characteristic, especially for capacity. This study is intended as an effort to develop basic capacity on pedestrian pathways in Indonesia with simulation models. The function of the sidewalk was not only a connecting path between regions in the commercial area but also has a higher level of complexity due to diverse pedestrian motives. This study used a microscopic approach through model simulation that built-in PTV Viswalk as a supporting system to do it. Location of study in Jalan Malioboro in Yogyakarta, Indonesia. This study resulted that the default value coefficient about the pedestrian characteristic model needs to change and pedestrian speed was a key to giving a new value. Then, validation for simulation is given with local and global validation. It resulted that new value makes similar value with real data which could reach basic capacity. Then, basic capacity does with Viswalk simulation and reaches 365,42 pedestrian/meter/minute as a basic pedestrian lane in Yogyakarta, Indonesia. This study makes a starting point to reach the developing standard for pedestrian traffic in Indonesia.

Modeling and Simulation of Crowd Pre-Evacuation Decision-Making in Complex Traffic Environments.

Human movements in complex traffic environments have been successfully simulated by various models. It is crucial to improve crowd safety and urban resilience. However, few studies focus on reproducing human behavior and predicting escape reaction time in the initial judgement stage in complex traffic environments. In this paper, a pedestrian pre-evacuation decision-making model considering pedestrian heterogeneity is proposed for complex environments. Firstly, the model takes different obvious factors into account, including cognition, information, experience, habits, stress, and decision-making ability. Then, according to the preference of the escapees, the personnel decision-making in each stage is divided into two types: stay and escape. Finally, multiple influencing factors are selected to construct the regression equation for prediction of the escape opportunity. The results show that: (1) Choices of escape opportunity are divided into several stages, which are affected by the pedestrian individual risk tolerance, risk categories strength, distance from danger, and reaction of the neighborhood crowd. (2) There are many important factors indicating the pedestrian individual risk tolerance, in which Gen, Group, Time and Mode are a positive correlation, while Age and Zone are a negative correlation. (3) The analysis of the natural response rate of different evacuation strategies shows that 19.81% of people evacuate immediately. The research in this paper can better protect public safety and promote the normal activities of the population.

Modeling Pedestrian Detour Behavior By-Passing Conflict Areas

In the process of walking, most pedestrians prefer to choose the shortest path, which requires passing through the conflict area. However, in the case of high crowd density, 5–20% of the total population will choose to follow the pre-planned route before walking or during the initial period of the trip to bypass the conflict area. Aiming at reproducing this detour behavior phenomenon, an extended social force model (SFM) is proposed according to a three-layer pedestrian simulation framework. This model not only fully considers the choice of detour mode, but also contains the avoidance and game behavior at the conflict point. At the strategic layer, a detour mode selection model based on the Logit model is established considering the pedestrian starting time and detour angle, to distinguish between the two groups of pedestrians who follow the pre-planned route and those who repeatedly adjust the route during the trip. Then, the path decision based on visual perception density at the tactical layer and the Voronoi-based SFM at the operational layer are combined to guide the specific movement of the two types of pedestrian groups. A series of evaluation indexes such as the central density, the mean local density, and detour level is selected, and Kolmogorov–Smirnov (K-S) test and dynamic time warping (DTW) method are used to evaluate and compare the scores of each index of different models. The results show that the model can improve the existing pedestrian detour simulation model to a certain extent. In sum, the travel time score, the detour level, and the mean local density score respectively increase from 0.71 to 0.81, 0.46 to 0.81, and 0.39 to 0.48, which indicates a significant improvement in walking performance.

The construction and application of event logic graph for pedestrian flow evacuation in typical scenarios

An event logic graph is a kind of knowledge mapping technology for knowledge inference and simulation analysis, which takes events as the core and portrays the hierarchical system and logical evolution pattern between events. In order to apply it to further improve the accuracy of related studies, such as pedestrian flow evacuation, simulation model optimization and risk prediction. In this paper, we use social network resources, media resources and journal database resources to build our corpus and adopt the explicit event relationship extraction method based on syntactic dependency and the implicit event relationship extraction method based on BERT+Bi-LSTM+Attention+Softmax for the characteristics of explicit event relationship and implicit event relationship, respectively. This paper constructs a pedestrian flow evacuation matter mapping for three typical scenarios and discusses its application path. It is found that once a sound knowledge base of logical reasoning and event logic graph is established, both research on optimization of pedestrian flow evacuation simulation models and research on identification and assessment of pedestrian flow evacuation safety risks will receive excellent support.

Causal Temporal–Spatial Pedestrian Trajectory Prediction With Goal Point Estimation and Contextual Interaction

Forecasting pedestrian trajectories in complex dynamic environments is highly critical for the application of autonomous vehicles and robots. Accordingly, this paper proposes a novel pedestrian trajectory prediction model called CTSGI, which, utilizes self-attention mechanism to construct an interactive graph between pedestrians and their neighbors based on their spatial relationship, to model the crowd’s interaction. At the same time, it uses self-attention to extract the temporal dependence for a single pedestrian. In order to effectively model the interaction between the pedestrian and the context, such as where a pedestrian can walk or approach, the semantic segmentation of background image is utilized. CTSGI estimates the goal points of pedestrian and their neighbors to assist in predicting the future trajectory. In addition, the causal structure model is used to analyze the confounding factors existing in the encoding stage, and Do-calculus is introduced for eliminating the confounding impact to improve the prediction performance. Moreover, extensive experiments are conducted for the proposed model on ETH and UCY datasets, and clearly, the experimental results reveal that the model reported herein outperforms the comparative state-of-the-art methods by 27.59% in Average Displacement Error (ADE) and 8.33% in Final Displacement Error (FDE). Furthermore, visualization of attention indicates that our model can capture the interaction between some specific pedestrian and their neighbors better.

Pedestrian Detection Using MB-CSP Model and Boosted Identity Aware Non-Maximum Suppression

Pedestrian detection is an important task in autonomous surveillance systems. Despite the rapid progress in pedestrian detection field, detecting occluded pedestrians remains a challenging task due to the great variations in occluded pedestrians appearance and the drastic loss of pedestrian information in some severe cases. In this paper, we tackle the occlusion problem by proposing a multi-branch pedestrian detection model based on center and scale prediction framework. The proposed model employs features extracted from full pedestrian’s body as well as its upper, middle, and lower body parts using four detection branches. This multi-branch approach ensures that data representing the true pedestrian appearances, whether they are partially or completely visible, can dominate the final decision-making, minimizing the interference of non-pedestrian data in the detection. Furthermore, to implement the proposed model, the visibility of different pedestrian parts is appropriately annotated, which facilitates the training process. The final decision is made based on the four MB-CSP branches outputs, using a proposed fusing method, named Boosted Identity Aware-Non Maximum Suppression. On heavy occlusion settings, the proposed model resulted in the miss rates of 27.83%, 47.29% and 33.3% for Caltech-USA, Citypersons and EuroCity Persons datasets, respectively.

Discussions on pedestrian delay models and applications at signalized crosswalks

Discussions on pedestrian delay models and applications at signalized crosswalks

Factorial Design with Simulation for the Optimization of the Level of Service in the Platform-Train Interface of Metro Stations—A Pilot Study

Metro stations are considered complex areas of pedestrian mobility due to the increasing congestion, due to the a high level of demand of different circulation spaces. Given this situation and the limited physical spaces remaining to develop transport systems in urban areas, railways acquire greater relevance given the need to mobilize pedestrians. Within the stations, the most problematic area is the platform-train interface (PTI) due to the high number of interactions between passengers boarding and alighting. The objective of this study is to identify the PTI configuration that maximizes the level of service for passengers, safeguards the operational continuity of the system and improves user experience by reducing dissatisfaction and delay times. For this, a pedestrian microsimulation model is used in order to recreate the reality of a generic metro station and its different scenarios given the combinations of two factors: the platform configurations (topology) and the traffic control elements. Subsequently, these scenarios are analyzed through a factorial design, looking for the situation that optimizes the combination of metrics chosen in a weighted way. Finally, it is found that the PTI configuration that maximizes the level of service for users is the mixed station with signaling. It is this which includes the factors with the greatest positive effect on the chosen metrics.

Simulation-based heterogeneous pedestrian evacuation in subway stations

Pedestrian heterogeneity is one of the important factors affecting evacuation efficiency in subway stations. This paper mainly studies the impact of pedestrian heterogeneity on evacuation based on simulations. With the help of Massmotion, the Qingdao Jinggangshan Road subway station is modeled. The social force model is used as the pedestrian dynamics model and the minimum cost model is used as the decision-making mechanism of pedestrian path selection. The models are verified by comparing the field data with the corresponding simulation data. Fully considering the impact of different pedestrian attributes on evacuation efficiency, pedestrians are divided into three categories with different speed levels and three categories with different body size levels. Simulation experiments are carried out by adjusting the proportional relationship of the number of pedestrians with different attributes. The simulation results indicate that the larger the proportion of fast pedestrians under the same number of evacuees, the higher the evacuation efficiency to a certain extent. The evacuation efficiency could be reduced accordingly with the increase in the proportion of pedestrians with large body sizes. When the pedestrian density is large, the impact of pedestrian heterogeneity on evacuation cannot be clearly reflected. Moreover, the quantitative fitting relationship between evacuation time and pedestrian quantity could be obtained. This paper provides a theoretical basis for the determination of evacuation strategy for the heterogeneous crowd.

On the existence and uniqueness properties of the Hoogendoorn–Bovy pedestrian flow model

This study addresses the existence and uniqueness of solutions to the Hoogendoorn–Bovy (HB) pedestrian flow model, which describes the dynamic user-optimal pedestrian flow assignment problem in continuous space and time. The HB model consists of a forward conservation law (CL) equation that governs density and a backward Hamilton–Jacobi–Bellman (HJB) equation that contains a maximum admissible speed constraint (MASC), in which the flow direction is determined by the path-choice strategy. The existence and uniqueness of solutions are significantly more difficult to determine when the HJB equation contains an MASC; however, we prove that the HB model can be formulated as a forward CL equation and backward Hamilton–Jacobi (HJ) equation in which the MASC is non-binding if suitable model parameters are selected. This model is formulated as a fixed-point problem upon the simultaneous satisfaction of both equations. To verify the existence and uniqueness results, we first demonstrate the existence and uniqueness of the solutions to the CL and HJ equations, and then show that the coupled HB model is well-posed and has a unique solution. A numerical example is presented to illustrate the properties of the HB model.

Assuring Quality of Pedestrian Behavior Results Derived From Simulation Models Applied to Airport Terminals

Computer simulation is an important tool nowadays in the design and assessment of airport terminals and other transport hubs, reducing costs, speeding up decision-making, and bringing new perspectives on scenarios targeted on either legacy or disruptive technologies. Concepts such as agent-based modelling, floor field cellular automata and social force models support a range of software technologies available in easy-to-use packages, allowing complex pedestrian behavior to be modelled in a variety of situations. On the other hand, old challenges concerning model verification and validation can potentially hamper the benefits associated with pedestrian microsimulation, leading to untrustworthy results depicted through appealing screen animations. This research aims to deliver a systematization of verification and validation procedures in pedestrian simulation models applied to the airport terminal. Acknowledging that this research is not exhaustive, categories of tests are proposed, along with measures for their assessment. Regarding pedestrian traffic diagrams, solutions for data treatment are proposed and tested, including a novel approach to the delineation of the steady-state regime. The proposed methods were tested with experimental data and agent-based modelling simulation data, demonstrating the feasibility of the procedures, along with certain cautions on their usage.

Review on Microscopic Model of Continuous Pedestrian Flow

Continuous pedestrian flow micro model is an important method to study pedestrian flow at present. Compared with macro model and discrete model, continuous pedestrian flow micro model can better simulate pedestrian flow phenomenon. This paper summarizes the research significance and achievements of the continuous pedestrian flow micro model. The research contents and corresponding modeling methods of relevant models are mainly introduced, and the future development of pedestrian flow micro models is prospected.

Obtaining Pedestrian Interest Regions Using Experimental Methods

In pedestrian simulation, using the region of interest of pedestrians can solve the problem of data redundancy and complexity and lack of representation in the pedestrian saccade range, and establish an efficient pedestrian collision avoidance behavior model. The real-time eye movement data of unidirectional and bidirectional pedestrian flow were collected by eye tracker, and the mathematical models of pedestrian flow density, radius and angle of each region of interest were established. Analyzing the experimental data, it is found that the range of pedestrian interest region is affected by pedestrian density. The angle of the pedestrian interest region is generally not more than 90°, and the radius of the high-interest region and the low-interest region are negatively correlated with the pedestrian density. When the pedestrian density reaches a certain degree, the radius of the high-interest region and the low-interest region approach the fixed value respectively.

Geographic Information System-Based Model of Outdoor Thermal Comfort: Case Study for Zurich

The importance of walking, the most basic form of transportation, is growing. Climate change and the associated warmer temperatures could reduce comfortable walking distances drastically. There is a clear need to better understand how outdoor thermal comfort (OTC) and walking interact. In this work, thermoregulation of the human body is modeled with the two-node model to determine the influence of the microclimate on pedestrians’ OTC. First, the impact of the current microclimate in Zurich on the route choice of pedestrians is analyzed. No significant correlation is found between simulated OTC of walking a particular route and route choices for all trips, but results for longer trips indicate a possible influence of OTC. It is pointed out that the same assessment could be done for other regions, and the results could contribute to more accurate pedestrian modeling. Second, a tool is developed that can estimate OTC-corrected walking distances from any location. The tool is applied to the current climate and future climate scenarios. The results show that, in the future, OTC of pedestrians in Zurich will be severely decreased. Further, the tool can detect where there is potential for and, through its accessibility approach, quantify improvements to the built environment citywide. Future work should focus on enhancing physiological input parameters to the model. This work provides a novel use of the two-node model for walking subjects in a citywide assessment.

A cellular automata approach for modelling pedestrian-vehicle mixed traffic flow in urban city

In urban streets, the intrusion of pedestrians into the motorway will pose a considerable challenge. However, modelling mixed pedestrian-vehicle traffic scenes is difficult. Pedestrians and vehicles have different motion characteristics and spatial dimensions, so it is hard to carry out unified modelling, and there are few related studies. This paper combines the multi-grid cellular automata model to explore these scenes to connect the vehicle and pedestrian models. An Improved Kerner-Klenov-Wolf (IKKW) model and a pedestrian motion model considering Time-To-Collision (TTC) are proposed, in which the spatial motion of a vehicle and a pedestrian is uniformly updated horizontally and longitudinally. The preliminary analysis of the model shows that it has high simulation accuracy. Applying the model to the simulation of real-life situations, research results reveal the impact of pedestrian intrusion behavior on traffic and the changes in vehicles’ speed and flow rate caused by pedestrian intrusion behavior. The fundamental diagram of pedestrian-vehicle heterogeneous traffic shows significant differences, reflecting the impact of pedestrian intrusion on the state of traffic flow, and exposes six phase regions of pedestrian-vehicle mixed traffic flow. Moreover, this paper analyzes the conflicts between pedestrians and vehicles under different speed limits and sidewalk widths, showing that lower speeds and wider sidewalks can effectively reduce the frequency of pedestrian-vehicle conflicts. The peak conflicts’ frequency when the vehicles’ speed limit is 60.48 km/h exceeds three times that when the speed limit is 30.24 km/h. The cellular automata model proposed in this paper provides a good direction and idea for studying mixed traffic flow and has high scalability.

Safety and Sustainable Development of Automated Driving in Mixed-Traffic Urban Areas—Considering Vulnerable Road Users and Network Efficiency

Next to environmental aspects, establishing areas for safe and economically viable automated driving in mixed-traffic settings is one major challenge for sustainable development of Autonomous Vehicles (AVs). This work investigates safety in the interactions between AVs, human-driven vehicles, and vulnerable road users, including cyclists and pedestrians, within a simulated urban environment in the Dutch city of Rotterdam. New junction and pedestrian models are introduced, and virtual AVs with an occlusion-aware driving system are deployed to deliver cargo autonomously. The safety of applying this autonomous cargo delivery service is assessed using a large set of Surrogate Safety Indicators (SSIs). Furthermore, Macroscopic Fundamental Diagrams (MFDs) and travel time loss are incorporated to evaluate the network efficiency. By assessing the impact of various measures involving Vehicle-to-Vehicle (V2V), Vehicle-to-Infrastructure (V2I), Vehicle-to-Everything (V2X) communications, infrastructure modifications, and driving behavior, we show that traffic safety and network efficiency can be achieved in a living lab setting for the considered case. Our findings further suggest that V2X gets implemented, new buildings are not placed close to intersections, and the speed limit of non-arterial roads is lowered.

MPDNet: A Transformer-based real-time passenger detection network in metro stations

We propose a passenger flow detection method for dense areas of subway stations to address the current situation that existing pedestrian detection models cannot meet the real-time performance requirements in subway applications and lack validation in multiple subway scenarios. First, we designed the MPDNet model, which uses PVT-small to extract features and an improved feature pyramid network (FPN) for upsampling using the adaptively spatial feature fusion (ASFF) algorithm to retain more local information in the output of the FPN. Second, to better evaluate the performance of models in the metro, we collected subway surveillance video data and proposed the MetroStation dataset. Finally, we trained and evaluated the performance of the MPDNet model on the MetroStation dataset. We compare our method with several common object detection models on the MetroStation dataset, using mAP and frames per second (FPS) to verify its accuracy. The experiments on the MetroStation dataset demonstrated that the MPDNet performed well and satisfied inference speed requirements in metro passenger flow detection.