Pedestrian Trajectory Data Research Articles

A deep pedestrian trajectory generator for complex indoor environments

AbstractPedestrian trajectory data, which can be used to mine pedestrian motion patterns or to model pedestrian dynamics, is crucial for indoor location‐based service studies and applications. However, researchers are faced with the challenges of data shortage and privacy restrictions when using pedestrian trajectory data. We present an Indoor Pedestrian Trajectory Generator (IPTG), which is a novel deep learning model to synthesize pedestrian trajectory data. IPTG first produces feature sequences that encode the spatial–temporal and semantic features of the walking process and then interpolates them into complete trajectories using A* and perturbation algorithms. IPTG has specially designed loss functions that preserve topological constraints and semantic characteristics. Incorporating the prior knowledge of environment constraints and pedestrian walking patterns, the IPTG model is capable of generating topologically and logically sound indoor pedestrian trajectories. We evaluated the synthesized trajectories based on multiple metrics and examined the generated trajectories qualitatively. The results show that IPTG outperforms several baselines, demonstrating its ability to generate semantically meaningful and spatiotemporally coherent trajectories.

Novel learning framework for optimal multi-object video trajectory tracking

BackgroundWith the rapid development of Web3D, virtual reality, and digital twins, virtual trajectories and decision data considerably rely on the analysis and understanding of real video data, particularly in emergency evacuation scenarios. Correctly and effectively evacuating crowds in virtual emergency scenarios are becoming increasingly urgent. One good solution is to extract pedestrian trajectories from videos of emergency situations using a multi-target tracking algorithm and use them to define evacuation procedures. MethodsTo implement this solution, a trajectory extraction and optimization framework based on multi-target tracking is developed in this study. First, a multi-target tracking algorithm is used to extract and preprocess the trajectory data of the crowd in a video. Then, the trajectory is optimized by combining the trajectory point extraction algorithm and Savitzky–Golay smoothing filtering method. Finally, related experiments are conducted, and the results show that the proposed approach can effectively and accurately extract the trajectories of multiple target objects in real time. ResultsIn addition, the proposed approach retains the real characteristics of the trajectories as much as possible while improving the trajectory smoothing index, which can provide data support for the analysis of pedestrian trajectory data and formulation of personnel evacuation schemes in emergency scenarios. ConclusionsFurther comparisons with methods used in related studies confirm the feasibility and superiority of the proposed framework.

Deep learning-based pedestrian trajectory prediction and risk assessment at signalized intersections using trajectory data captured through roadside LiDAR

In recent years, rapid advancements in the Autonomous Vehicles (AVs) industry have greatly motivated the research and development in pedestrian trajectory prediction and risk assessment. One of the critical requirements for AVs is to predict the future trajectories of pedestrians and provide collision warnings in an accurate and prompt manner. Nevertheless, accurate prediction of pedestrian trajectories remains a technical challenge, mainly caused by the heterogeneity of pedestrian crossing behavior and uncertainties in vehicle-pedestrian interactions. This paper proposes a deep learning-based method for pedestrian trajectory prediction and risk assessment, using trajectory data extracted from roadside LiDAR data and corresponding signal phasing information at MLK and Georgia Avenue in Chattanooga, TN. Meanwhile, a set of criteria referred to as the risk factor is established to quantitatively evaluate the risk of the pedestrian crossing behavior, which also serves as a learnable feature. A Long Short-Term Memory (LSTM) network is proposed, which takes the following data as the input: the pedestrian trajectory data, signal phasing data, and risk factors from the past 10 steps. Meanwhile, the network predicts the pedestrian trajectory and risk factor at the future time step. In the experimental study, the root-mean-square errors between the predicted and ground truth x and y coordinates are 0.225 meters and 0.377 meters, respectively, and the F1 score value for the risk factor is 99.6%, demonstrating the efficacy of the proposed LSTM-based methodology on pedestrian trajectory prediction and risk assessment.

Proposal of a classification method for public space usage based on pedestrian trajectory data

AbstractHerein, we propose a machine learning method based on pedestrian trajectory data to classify public space usage states and discriminate unknown usage states. Assuming that there are several frequent patterns in the usage state of public spaces, each element of the data set representing the usage state of public spaces can be classified into several clusters. Each cluster is defined as a “type” of usage state. They were classified into usage state “types” via principal component analysis and x‐means clustering. We employed data from the detection and recording of pedestrian trajectories by six 3D laser (LiDAR) sensors, conducted by the authors during the summer of 2019 in a public space in Yokohama, Japan. In the training phase, we defined “types” of usage states based on the data obtained for 3 weeks. In the test phase, the “type” of usage state was determined for the data of other periods. Consequently, 16 types that appeared at specific times and days were identified, and 1.1% of the test data were determined to be “new usage states,” which were not found in the training data. This method helps understand long‐term and complex variations in public space utilization patterns.

Analysis of Badminton Movement Cognition Algorithm Based on Track Linear Capture

At present, more and more sports science and technology are being explored and applied in competitive sports. The birth and popularization of video tracking and capturing technology have provided more fair and just perspectives for many sports events. Track linear capture can replay the player’s behavior in real time, the flight path of the badminton can be analyzed in 3D stereoscopic analysis, and the ball’s motion trajectory can be calculated more accurately. In this paper, an objective trajectory tracking and prediction model is constructed based on the motion cognition algorithm, and the motion characteristics of the objective are extracted from the limited historical trajectory of the objective to achieve more accurate trajectory tracking. Then, the trajectory tracking model is applied to the objective tracking framework to obtain ideal objective tracking results. At the same time, in order to make use of the interaction between scene information and objective, this paper improves the trajectory tracking model. The trajectory prediction model based on neural network is constructed, which learns the pedestrian motion characteristics from the pedestrian trajectory data of the target tracking scene offline and uses its “memory” online to generate the implicit depth motion characteristics of the target according to the limited historical information of the target. It also predicts the most likely location of the future target and calculates the motion similarity between the targets. Finally, a simulation experiment platform is built to prove the effectiveness of the trajectory tracking model and objective tracking algorithm proposed in this paper. Through the research results of this paper, it can play a role in verifying the referee’s judgment on the penalty of some key balls, which is more conducive to maintaining the fairness of the game, and more helpful for athletes to optimize their exercise results according to scientific basis, and has the function of improving their performance.

PROPOSAL OF A CLASSIFICATION METHOD FOR PUBLIC SPACE USAGE BASED ON PEDESTRIAN TRAJECTORY DATA

Herein, we propose a machine learning method based on pedestrian trajectory data to classify public space usage states and discriminate unknown usage states. Aggregated feature values for each small cell were regarded as feature vectors representing the usage state. They were classified into usage state “types” via principal component analysis and x-means clustering. During validation using actual data, 16 types appearing at specific times and days were identified, and 1.1% of the test data were determined to be “new usage states” not found in the training data. This method helps understand long-term and complex variations in public space utilization patterns.

A Data-Driven Model for Pedestrian Behavior Classification and Trajectory Prediction

Pedestrian modeling remains a formidable challenge in transportation science due to the complicated nature of pedestrian behavior and the irregular movement patterns. To this extent, accurate and reliable positioning technologies and techniques play a significant role in the pedestrian simulation studies. The objective of this research is to predict pedestrian movement in various perspectives utilizing historical trajectory data. The study features considered in this research are pedestrian class, speed and position. The ensemble of these features provides a thorough description of pedestrian movement prediction, whilst contributes to the context of pedestrian modeling and Intelligent Transportation Systems. More specifically, pedestrian movement is grouped into different classes considering gender, walking pace and distraction by employing random forest algorithms. Then, position and speed prediction is computed employing suitable data-driven methods, in particular, the locally weighted regression (LOESS method), taking into account the individual pedestrian.s profile. An LSTM-based (Long Short-Term Memory) model is also applied for comparison. The methodology is applied on pedestrian trajectory data that were collected in a controlled experiment undertaken at the Campus of the National Technical University of Athens (NTUA), Greece. Prediction of pedestrian.s movement is achieved, yielding satisfactory results.

Analyzing pedestrian individual and interaction collision avoidance dynamics in traditional scenarios

Revealing the behavioral mechanism of crowd dynamics is crucial for explaining the self-organized phenomena and making a quick emergent response for crowd management. Various approaches have been proposed to analyze self-organized phenomena of crowd dynamics, but the essential mechanisms of complicated and heterogeneous collision avoidance behaviors are still not clear. To deeply investigate the microscopic behaviors, the geometrical characteristics of Voronoi diagrams and velocity obstacles are explored and a unified spatial tessellation and velocity obstacle approach is proposed. The personal space is divided into four regions, and a velocity graph that is a subgraph of Delaunay triangulation will be established to describe the relationship between pedestrians. In a velocity graph, a link represents that a pedestrian’s speed is towards a Voronoi edge of its neighbor. According to the experimental pedestrian trajectory data, the collision avoidance probability is relevant to some parameters, such as spatial conflict time, and minimal velocity difference. Results show that the probability is related to the distance to velocity obstacle boundary and relaxation time by Sigmoid function. Besides, the behavior dynamics in each region is analyzed, and it is determined by the deviation angles between Voronoi nodes and neighbor locations. This approach can be potentially applied to pedestrian behavioral analysis and crowd management.

Knowledge Acquisition from Pedestrian Flow Analysis using Sparse Mobile Probe Data

Autonomous vehicles require high-level semantic maps, which contain the activities of pedestrians and cars, to ensure safe navigation. High-level semantics can be obtained from mobile probe sensor data. Analyzing pedestrian trajectories obtained from mobile probe data is an effective approach to avoid collisions between autonomous vehicles and pedestrians. Such analyses of pedestrian trajectories can generate new information such as pedestrian behaviors in violation of traffic regulations. However, pedestrian trajectories obtained from mobile probe data significantly sparse and noisy, making it challenging to analyze pedestrian activity. To address this issue, we propose multiple daily data and graph-based approaches to treat sparse and noisy data for estimating the flow of pedestrians based on mobile probe data. To improve the sparseness of the data, multiple daily data are fused. After that, a pedestrian graph is created to enhance the region’s coverage by connecting the sparse data indicating the flow of pedestrians. This proposed approach successfully obtained pedestrian trajectory data from the sparse and noisy data. Moreover, it was possible to identify the potential locations where pedestrians tend to cross the street by analyzing the pedestrian flow. The results indicate that 83% of well-known regions where pedestrians tend to cross the street corresponded with those extracted using the proposed approach. Furthermore, a high-level semantic map of the regions where pedestrians tend to cross the street along a 1-km road is presented. The trajectory information obtained using the proposed approach is expected to be essential for understanding different scenarios of the interactions between individuals and autonomous vehicles.

Trajectory Data Analysis on the Spatial and Temporal Influence of Pedestrian Flow on Path Planning Decision

The modeling of walking behavior and design of walk-friendly urban pathways have been of interest to many researchers over the past decades. One of the major issues in pedestrian modeling is path planning decision-making in a dynamic walking environment with different pedestrian flows. While previous studies have agreed that pedestrian flow influences path planning, only a few studies have dealt with the empirical data to show the relationship between pedestrian flow and path planning behavior. This study introduces a new methodology for analyzing pedestrian trajectory data to find the dynamic walking conditions that influence the path planning decision. The comparison of the pedestrians’ path shows that the higher proportion of opposite flows are, the greater they influence the path selection decision. In this study, we investigate the relationship between the opposite flow changes and path planning behavior and find the spatial and temporal ranges of the opposite flow that affects the path planning behavior. Lastly, we find the ratio of pedestrians that update their paths with respect to the opposite flow rate.

Identifying the Indoor Space Characteristics of an Urban Railway Station Based on Pedestrian Trajectory Data

Stations are being converted into various living spaces that can be used for public transportation, work, commerce, and leisure. To satisfy the various requirements and expectations for functional extension, it is necessary to investigate and understand the phenomena caused by users. A methodology to cluster the characteristics of pedestrian space of a railway station through the pedestrian trajectory data collected from an actual operating station is proposed in this paper. Then the spatial usability of the movement and stay of pedestrians were defined through the results of the clustering. The procedure to cluster the indoor space characteristics of an urban railway station in this study consists of four steps: data collection, feature vector extraction, K-means clustering, and cluster characteristics analysis. A case study was conducted for the Samseong station. The results of the proposed spatial clustering analysis showed that there are several types of spaces depending on the space occupancy characteristics of pedestrians. The proposed methodology could be applied to indoor space diagnosis from the perspective of station monitoring and management. In addition, the station operator could respond flexibly to unexpected events by monitoring the indoor spaces according to whether the flow is normal or suggestive of an emergency.

A method for the treatment of pedestrian trajectory data noise

This paper provides an improved algorithm for eliminating noise of pedestrian trajectory data. Data have been collected from the field through video recordings. A semi-automatic process extracts pedestrian trajectories that include noise. The proposed algorithm relies on the Kalman filter framework. In particular, the Unscented Kalman Filter is employed for relaxing standard Kalman filter assumptions. An innovation of this paper is the incorporation of moving average in the Unscented Kalman Filter that provides more accurate pedestrian trajectory estimations. In addition, a procedure for evaluating Kalman filter noise covariance matrices is suggested. Algorithm results from real pedestrian trajectory data indicate high efficacy level in reducing data noise, thus improving their usefulness for calibrating and validating pedestrian simulation models.

Pedestrian dynamics at transit stations: an integrated pedestrian flow modeling approach

ABSTRACTThis paper presents an integrated modelling framework to capture pedestrian walking behaviour in congested and uncongested conditions. The framework is built using a combination of concepts from the Social Force model (basic one-to-one interaction), behavioural heuristics (physiological and cognitive constraints), and materials science (multi-body potential concept). The approach is ultimately designed to capture pedestrian interactions in transit stations. Due to the lack of available trajectory data of pedestrians within transit stations, the model is calibrated using pedestrian trajectory data from narrow bottleneck and bidirectional flow experiments provided by the Delft University of Technology. These two scenarios were chosen due to the frequency with which they occur in transit stations. The integrated modelling framework reproduced similar trajectory patterns observed in the experiments which encouraged a transit station simulation in an environment similar to that at the Foggy-Bottom METRO station in Washington, D.C.

Macroscopic Fundamental Diagram for pedestrian networks: Theory and applications

The Network or Macroscopic Fundamental diagram (MFD) has been a topic receiving a lot of attention in the past decade. Both from a theoretical angle and from a more application-oriented perspective, the MFD has proven to be a powerful concept in understanding and managing vehicular network dynamics.In particular, the application in traffic management has inspired the research presented in this contribution, where we explore the existence and the characteristics of the pedestrian Macroscopic Fundamental Diagram (p-MFD). This is first of all done from a theoretical perspective, which results in the main contribution of this research showing how we can derive the p-MFD from assumed local fundamental diagrams (FDs). In doing so, we show that we can relate the average (out-)flow from a pedestrian network as a function of the average spatial density ρ¯ and the density spatial variation σ2. We show that the latter is essential to provide a reasonable description of the overall network conditions. For simple relations between density and speed (i.e. Greenshields and Underwood fundamental diagrams), we derive analytical results; for more commonly used FDs in pedestrian flow theory, such as the triangular FD of Newell or the FD of Weidmann, we show the resulting relation by proposing a straightforward simulation approach.As a secondary contribution of the paper, we show how the p-MFD can be constructed from pedestrian trajectory data stemming from either microsimulation or from experimental studies. We argue that the results found are in line with the theoretical results, providing further evidence for the validity of the p-MFD concept. We furthermore discuss concepts of hysteresis, also observed in vehicular network dynamics, due to the differences in the queue build up and recuperation phases.We finally present some applications of the presented concepts in crowd management, network level-of-service determination, and coarse-scale modelling.

A Novel Semantic Matching Method for Indoor Trajectory Tracking

The rapid development of smartphone sensors has provided rich indoor pedestrian trajectory data for indoor location-based applications. To improve the quality of these collected trajectory data, map matching methods are widely used to correct trajectories. However, these existing matching methods usually cannot achieve satisfactory accuracy and efficiency and have difficulty in exploiting the rich information contained in the obtained trajectory data. In this study, we proposed a novel semantic matching method for indoor pedestrian trajectory tracking. Similar to our previous work, pedestrian dead reckoning (PDR) and human activity recognition (HAR) are used to obtain the raw user trajectory data and the corresponding semantic information involved in the trajectory, respectively. To improve the accuracy and efficiency for user trajectory tracking, a semantic-rich indoor link-node model is then constructed based on the input floor plan, in which navigation-related semantics are extracted and formalized for the following trajectory matching. PDR and HAR are further utilized to segment the trajectory and infer the semantics (e.g., “Turn left”, “Turn right”, and “Go straight”). Finally, the inferred semantic information is matched with the semantic-rich indoor link-node model to derive the correct user trajectory. To accelerate the matching process, the semantics inferred from the trajectory are also assigned weights according to their relative importance. The experiments confirm that the proposed method achieves accurate trajectory tracking results while guaranteeing a high matching efficiency. In addition, the resulting semantic information has great application potential in further indoor location-based services.

Model-based Pedestrian Trajectory Prediction using Environmental Sensor for Mobile Robots Navigation

Safety is the most important to the mobile robots that coexist with human. There are many studies that investigate obstacle detection and collision avoidance by predicting obstacles’ trajectories several seconds into the future using mounted sensors such as cameras and laser range finder (LRF) for the safe behavior control of robots. In environments such as crossing roads where blind areas occur because of visual barriers like walls, obstacle detection might be delayed and collisions might be difficult to avoid. Using environmental sensors to detect obstacles is effective in such environments. When crossing roads, there are several passages pedestrian might move and it is difficult to depict going each passage in the same movement model. Therefore, we hypothesize that a more effective way to predict pedestrian movement is by predicting passages pedestrian might move and estimating the trajectories to the passages. We acquire pedestrian trajectory data using an environmental LRF with an extended Kalman filter (EKF) and construct pedestrian movement models using vector auto regressive (VAR) models, which pedestrian state is consisting of the position, speed and direction. Then, we test the validity of the constructed pedestrian movement models using experimental data. We narrow down the selection of a pedestrian movement model by comparing the prediction error for each path between the estimated pedestrian state using an EKF, and the predicted state using each movement model. We predict the trajectory using the selected movement model. Finally, we confirm that an appropriate path model that a pedestrian can actually move through is selected before the crossing area and that only the appropriate model is selected near the crossing area.

Macroscopic Fundamental Diagram for pedestrian networks: theory and applications

The Macroscopic Fundamental diagram (MFD) has proven to be a powerful concept in understanding and managing vehicular network dynamics, both from a theoretical angle and from a more application-oriented perspective. In this contribution, we explore the existence and the characteristics of the pedestrian Macroscopic Fundamental Diagram (p-MFD). From a theoretical perspective, the main contribution of this research shows how we can derive the p-MFD from assumed local fundamental diagrams (FDs). We show that we can relate the average (out-) flow from a pedestrian network as a function of the average spatial density ρ and the density spatial variation σ2. We show that the latter is essential to provide a reasonable description of the overall network conditions. For simple linear relations between density and speed, we derive analytical results; for more commonly used FDs in pedestrian flow theory we show the resulting relation using a straightforward simulation approach. As a secondary contribution of the paper, we show how the p-MFD can be constructed from pedestrian trajectory data stemming from either microsimulation or from experimental studies. The results found are in line with the theoretical result, providing further evidence for the validity of the p-MFD concept. We furthermore discuss concepts of hysteresis, due to the differences in the queue build up and recuperation phases. We end with applications of the presented concepts, e.g. in crowd management.

Surrogate Safety Analysis of Pedestrian-Vehicle Conflict at Intersections Using Unmanned Aerial Vehicle Videos

Conflict analysis using surrogate safety measures (SSMs) has become an efficient approach to investigate safety issues. The state-of-the-art studies largely resort to video images taken from high buildings. However, it suffers from heavy labor work, high cost of maintenance, and even security restrictions. Data collection and processing remains a common challenge to traffic conflict analysis. Unmanned Aerial Systems (UASs) or Unmanned Aerial Vehicles (UAVs), known for easy maneuvering, outstanding flexibility, and low costs, are considered to be a novel aerial sensor. By taking full advantage of the bird’s eye view offered by UAV, this study, as a pioneer work, applied UAV videos for surrogate safety analysis of pedestrian-vehicle conflicts at one urban intersection in Beijing, China. Aerial video sequences for a period of one hour were analyzed. The detection and tracking systems for vehicle and pedestrian trajectory data extraction were developed, respectively. Two SSMs, that is, Postencroachment Time (PET) and Relative Time to Collision (RTTC), were employed to represent how spatially and temporally close the pedestrian-vehicle conflict is to a collision. The results of analysis showed a high exposure of pedestrians to traffic conflict both inside and outside the crosswalk and relatively risking behavior of right-turn vehicles around the corner. The findings demonstrate that UAV can support intersection safety analysis in an accurate and cost-effective way.

Trajectory Data Analyses for Pedestrian Space-time Activity Study

It is well recognized that human movement in the spatial and temporal dimensions has direct influence on disease transmission(1-3). An infectious disease typically spreads via contact between infected and susceptible individuals in their overlapped activity spaces. Therefore, daily mobility-activity information can be used as an indicator to measure exposures to risk factors of infection. However, a major difficulty and thus the reason for paucity of studies of infectious disease transmission at the micro scale arise from the lack of detailed individual mobility data. Previously in transportation and tourism research detailed space-time activity data often relied on the time-space diary technique, which requires subjects to actively record their activities in time and space. This is highly demanding for the participants and collaboration from the participants greatly affects the quality of data(4). Modern technologies such as GPS and mobile communications have made possible the automatic collection of trajectory data. The data collected, however, is not ideal for modeling human space-time activities, limited by the accuracies of existing devices. There is also no readily available tool for efficient processing of the data for human behavior study. We present here a suite of methods and an integrated ArcGIS desktop-based visual interface for the pre-processing and spatiotemporal analyses of trajectory data. We provide examples of how such processing may be used to model human space-time activities, especially with error-rich pedestrian trajectory data, that could be useful in public health studies such as infectious disease transmission modeling. The procedure presented includes pre-processing, trajectory segmentation, activity space characterization, density estimation and visualization, and a few other exploratory analysis methods. Pre-processing is the cleaning of noisy raw trajectory data. We introduce an interactive visual pre-processing interface as well as an automatic module. Trajectory segmentation(5) involves the identification of indoor and outdoor parts from pre-processed space-time tracks. Again, both interactive visual segmentation and automatic segmentation are supported. Segmented space-time tracks are then analyzed to derive characteristics of one's activity space such as activity radius etc. Density estimation and visualization are used to examine large amount of trajectory data to model hot spots and interactions. We demonstrate both density surface mapping(6) and density volume rendering(7). We also include a couple of other exploratory data analyses (EDA) and visualizations tools, such as Google Earth animation support and connection analysis. The suite of analytical as well as visual methods presented in this paper may be applied to any trajectory data for space-time activity studies.

Microscopic Parameter Identification of Pedestrian Models and Implications for Pedestrian Flow Modeling

This paper focuses on the calibration of microscopic pedestrian models using pedestrian trajectory data. A generic approach that enables parameter identification for microscopic models in general and for walker models in particular is proposed. The application results provide insight into the behavior of individual pedestrians and interpedestrian differences as well as the resulting pedestrian flow characteristics. Different models of increasing complexity were compared to investigate which model amendments are significant from a statistical point of view. Besides anisotropy, finite reaction times play an important role in correctly describing microscopic walking behavior. The implications of these findings in the microscopic description of pedestrian flows are considered by studying the predicted flow operations at a narrow bottleneck. The finite reaction times have a significant effect on the pedestrian flow operations and should be included in pedestrian models to ensure realistic predictions.