Pedestrian Positioning Research Articles

Customer Context Analysis in Shopping Malls: A Method Combining Semantic Behavior and Indoor Positioning Using a Smartphone

Customer context analysis (CCA) in brick-and-mortar shopping malls can support decision makers’ marketing decisions by providing them with information about customer interest and purchases from merchants. It makes offline CCA an important topic in marketing. In order to analyze customer context, it is necessary to analyze customer behavior, as well as to obtain the customer’s location, and we propose an analysis system for customer context based on these two aspects. For customer behavior, we use a modeling approach based on the time-frequency domain, while separately identifying movement-related behaviors (MB) and semantic-related behaviors (SB), where MB are used to assist in localization and the positioning result are used to assist semantic-related behavior recognition, further realizing CCA generation. For customer locations, we use a deep-learning-based pedestrian dead reckoning (DPDR) method combined with a node map to achieve store-level pedestrian autonomous positioning, where the DPDR is assisted by simple behaviors.

Design of a Peer-to-Peer 3-D Pedestrian Positioning System Based on UWB and Sensor Fusion

Design of a Peer-to-Peer 3-D Pedestrian Positioning System Based on UWB and Sensor Fusion

Integrating Visible Light Communication and AI for Adaptive Traffic Management: A Focus on Reward Functions and Rerouting Coordination

This study combines Visible Light Communication (VLC) and Artificial Intelligence (AI) to optimize traffic signal control, reduce congestion, and enhance safety. Utilizing existing road infrastructure, VLC technology transmits real-time data on vehicle and pedestrian positions, speeds, and queues. AI agents, powered by Deep Reinforcement Learning (DRL), process these data to manage traffic flows dynamically, applying anti-bottlenecking and rerouting techniques. A global agent coordinates local agents, enabling indirect communication and a unified DRL model that adjusts traffic light phases in real time using a queue/request/response system. A key focus of this work is the design of reward functions for standard and rerouting scenarios. In standard scenarios, the reward function prioritizes wide green bands for vehicles while penalizing pedestrian rule violations, balancing efficiency and safety. In rerouting scenarios, it dynamically prevents queuing spillovers at neighboring intersections, mitigating cascading congestion and ensuring safe, timely pedestrian crossings. Simulation experiments in the SUMO urban mobility simulator and real-world trials validate the system across diverse intersection types, including four-way crossings, T-intersections, and roundabouts. Results show significant reductions in vehicle and pedestrian waiting times, particularly in rerouting scenarios, demonstrating the system’s scalability and adaptability. By integrating VLC technology and AI-driven adaptive control, this approach achieves efficient, safe, and flexible traffic management. The proposed system addresses urban mobility challenges effectively, offering a robust solution to modern traffic demands while improving the travel experience for all road users.

Integration of Visible Light Communication, Artificial Intelligence, and Rerouting Strategies for Enhanced Urban Traffic Management

This study combines Visible Light Communication (VLC) and Artificial Intelligence (AI) to enhance traffic signal control, reduce congestion, and improve safety, through real-time monitoring and dynamic traffic management. Leveraging VLC technology, the system uses existing road infrastructure to transmit live data on vehicle and pedestrian positions, speeds, and queues. AI agents, employing Deep Reinforcement Learning (DRL), process this data to manage traffic flows dynamically, applying anti-bottleneck and rerouting techniques to balance pedestrian and vehicle waiting times. A centralized global agent coordinates the local agents controlling each intersection, enabling indirect communication and data sharing to train a unified DRL model. This model makes real-time adjustments to traffic light phases, utilizing a queue/request/response system for adaptive intersection management. Tested using simulations and real-world trials involving standard and rerouting scenarios, the approach demonstrates significantly better performance in regard to the rerouting configuration, reducing congestion and enhancing traffic flow and pedestrian safety. Scalable and adaptable to various intersection types, including four-way, T-intersections, and roundabouts, the system’s efficacy is validated using the SUMO urban mobility simulator, resulting in notable reductions to travel and waiting times for both vehicles and pedestrians.

Research Advanced in Pedestrian Re-identification based on Deep Learning

Pedestrian re-identification (re-ID) is always a research hotspot in the intelligent video surveillance community, which has been applied in various fields such as social security and criminal investigation. Though much effort has been devoted to boosting the accuracy and speed, person re-ID still faces numerous challenges and difficulties, including low-quality captured images, inconsistent camera angles, varied pedestrian postures, significant changes in background environments, and issues related to lighting and occlusion. This paper aims to summarize the recent studies of pedestrian re-ID to motivate the subsequent development. Specifically, this paper focuses on the technological development, application needs, and challenges of person re-ID from the aspects of occluded pedestrian re-ID, unsupervised person re-ID, virtual data generation, pose-aware person re-ID, and pedestrian search. It also identifies existing problems and offers a perspective on the future trends. The hope is that through the study of these methods, an understanding of the current development directions in person re-ID can be achieved and researchers can be provided with a reference.

Inertial Multi-Agent Systems Sensor Pedestrian Positioning Based on Foot Waist Vector Difference Constraint

Inertial Multi-Agent Systems Sensor Pedestrian Positioning Based on Foot Waist Vector Difference Constraint

A Triangular Structure Constraint for Pedestrian Positioning with Inertial Sensors Mounted on Foot and Shank

To suppress pedestrian positioning drift, a velocity constraint commonly known as zero-velocity update (ZUPT) is widely used. However, it cannot correct the error in the non-zero velocity interval (non-ZVI) or observe heading errors. In addition, the positioning accuracy will be further affected when a velocity error occurs in the ZVI (e.g., foot tremble). In this study, the foot, ankle, and shank were regarded as a triangular structure. Consequently, an angle constraint was established by utilizing the sum of the internal angles. Moreover, in contrast to the traditional ZUPT algorithm, a velocity constraint method combined with Coriolis theorem was constructed. Magnetometer measurements were used to correct heading. Three groups of experiments with different trajectories were carried out. The ZUPT method of the single inertial measurement unit (IMU) and the distance constraint method of dual IMUs were employed for comparisons. The experimental results showed that the proposed method had high accuracy in positioning. Furthermore, the constraints built by the lower limb structure were applied to the whole gait cycle (ZVI and non-ZVI).

Enhancing Smartphone-based Pedestrian Positioning: Using Factor Graph Optimization with Indoor/Outdoor Detection for 3DMA GNSS/Visual-Inertial State Estimation

This paper explores the pervasive challenges of pedestrian positioning using smartphones in densely populated urban environments where Global Navigation Satellite System (GNSS) signals are inaccessible, for example, in indoor areas. Existing sensor-based positioning methods, such as inertial navigation systems (INS), GNSS, and visual-inertial odometry (VIO), suffer from inherent restrictions that compromise the accuracy and reliability of the positioning performance. An approach based on machine learning is proposed to address these limitations, employing the Support Vector Machine (SVM) algorithm to accurately distinguish indoor/outdoor (IO) based on the measurement of GNSS. The proposed approach in this study seamlessly incorporates 3D mapping aided (3DMA) GNSS measurements and localized estimations derived by VIO via factor graph optimization (FGO), complemented by an IO detection switch, to achieve accurate pose estimation and effectively eliminate global drift. The system's effectiveness and robustness are rigorously assessed through comprehensive extensive real-life experiments, with an average reduction of 4 meters, leading to noteworthy and statistically significant findings. Received: 1 April 2024 | Revised: 13 September 2024 | Accepted: 26 October 2024 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The source code that supports the findings of this study is openly available in 3DMAGNSSVINS-IOFGO at https://github.com/queenie-ho/3DMAGNSSVINS-IOFGO. Author Contribution Statement Hiu-Yi Ho and Hoi-Fung Ng: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing - original draft, Writing - review & editing, Visualization. Weisong Wen: Conceptualization, Resources, Writing - review & editing, Supervision, Project administration, Funding acquisition. Yanlei Gu: Resources, Writing - review & editing, Writing - review & editing. Li-Ta Hsu: Conceptualization, Resources, Writing - review & editing, Supervision, Project administration, Funding acquisition.

Indoor Pedestrian Positioning Method Based on Ultra-Wideband with a Graph Convolutional Network and Visual Fusion.

To address the challenges of low accuracy in indoor positioning caused by factors such as signal interference and visual distortions, this paper proposes a novel method that integrates ultra-wideband (UWB) technology with visual positioning. In the UWB positioning module, the powerful feature-extraction ability of the graph convolutional network (GCN) is used to integrate the features of adjacent positioning points and improve positioning accuracy. In the visual positioning module, the residual results learned from the bidirectional gate recurrent unit (Bi-GRU) network are compensated into the mathematical visual positioning model's solution results to improve the positioning results' continuity. Finally, the two positioning coordinates are fused based on particle filter (PF) to obtain the final positioning results and improve the accuracy. The experimental results show that the positioning accuracy of the proposed UWB positioning method based on a GCN is less than 0.72 m in a single UWB positioning, and the positioning accuracy is improved by 55% compared with the Chan-Taylor algorithm. The proposed visual positioning method based on Bi-GRU and residual fitting has a positioning accuracy of 0.42 m, 71% higher than the Zhang Zhengyou visual positioning algorithm. In the fusion experiment, 80% of the positioning accuracy is within 0.24 m, and the maximum error is 0.66 m. Compared with the single UWB and visual positioning, the positioning accuracy is improved by 56% and 52%, respectively, effectively enhancing indoor pedestrian positioning accuracy.

UWB-Assisted Bluetooth Localization Using Regression Models and Multi-Scan Processing.

Bluetooth devices have been widely used for pedestrian positioning and navigation in complex indoor scenes. Bluetooth beacons are scattered throughout the entire indoor walkable area containing stairwells, and pedestrian positioning can be obtained by the received Bluetooth packets. However, the positioning performance is sharply deteriorated by the multipath effects originating from indoor clutter and walls. In this work, an ultra-wideband (UWB)-assisted Bluetooth acquisition of signal strength value method is proposed for the construction of a Bluetooth fingerprint library, and a multi-frame fusion particle filtering approach is proposed for indoor pedestrian localization for online matching. First, a polynomial regression model is developed to fit the relationship between signal strength and location. Then, particle filtering is utilized to continuously update the hypothetical location and combine the data from multiple frames before and after to attenuate the interference generated by the multipath. Finally, the position corresponding to the maximum likelihood probability of the multi-frame signal is used to obtain a more accurate position estimation with an average error as low as 70 cm.

Evacuation from Cramped Interiors with Aisle Seats: Uncertainty Induced by the Random Choice of Initial Positions

Considering evacuation from cramped interiors with aisle seats, the initial seating positions of pedestrians represent a non-negligible source of variance in total evacuation time. By means of Pathfinder simulations, this uncertainty is quantitatively investigated for a train and lecture hall geometry with homogeneous and heterogeneous groups of agents. Furthermore, an energy-inspired description of the occupation configuration is introduced that enables a proper numerical representation of the configuration space of the initial seating positions.

Pedestrian and Vehicle Area Positioning With Multiple UWB Signals

Pedestrian and Vehicle Area Positioning With Multiple UWB Signals

Impact of Vehicle Steering Strategy on the Severity of Pedestrian Head Injury.

In response to the sudden violation of pedestrians crossing the road, intelligent vehicles take into account factors such as the road conditions in the accident zone, traffic rules, and surrounding vehicles' driving status to make emergency evasive decisions. Thus, the collision simulation models for pedestrians and three types of vehicles, i.e., sedans, Sport Utility Vehicles (SUVs), and Multi-Purpose Vehicle (MPVs), are built to investigate the impact of vehicle types, vehicle steering angles, collision speeds, collision positions, and pedestrian orientations on head injuries of pedestrians. The results indicate that the Head Injury Criterion (HIC) value of the head increases with the increase in collision speed. Regarding the steering angles, when a vehicle's steering direction aligns with a pedestrian's position, the pedestrian remains on top of the vehicle's hood for a longer period and moves together with the vehicle after the collision. This effectively reduces head injuries to pedestrians. However, when the vehicle's steering direction is opposite to the pedestrian's position, the pedestrian directly collides with the ground, resulting in higher head injuries. Among them, MPVs cause the most severe injuries, followed by SUVs, and sedans have the least impact. Overall, intelligent vehicles have great potential to reduce head injuries of pedestrians in the event of sudden pedestrian-vehicle collisions by combining with Automatic Emergency Steering (AES) measures. In the future, efforts need to be made to establish an optimized steering strategy and optimize the handling of situations where steering is ineffective or even harmful.

An integrated positioning method with IMU/UWB based on geometric constraints of foot-to-foot distances

Zero velocity update aided inertial navigation system (INS) and ultra-wideband (UWB) technology are widely used in pedestrian positioning. However, INS has significant cumulative error and the application environment of UWB positioning system is limited. In this study, an integrated positioning method with IMU/UWB based on geometric constraints of foot-to-foot distances is proposed. The gait of pedestrian is divided into three phases: stance, swing, and stride. Then, stance phase correction model, swing phase ranging model and stride phase size model are established, respectively. The magnetic induction intensity, accelerations, velocities, and the foot-to-foot distances measured by UWB are used to construct the constraints. Extended Kalman filter (EKF) and error state Kalman filter (ESKF) are utilized to estimate the attitudes and positions optimally. Several experiments in various environments are carried out to verify the proposed method. The results show that the proposed method has high accuracy and good robustness to the environments.

Modeling framework of human driving behavior based on Deep Maximum Entropy Inverse Reinforcement Learning

Driving behavior modeling is extremely crucial for designing safe, intelligent, and personalized autonomous driving systems. In this paper, a modeling framework based on Markov Decision Processes (MDPs) is introduced that emulates drivers’ decision-making processes. The framework combines the Deep Maximum Entropy Inverse Reinforcement Learning (Deep MEIRL) and a reinforcement learning algorithm-proximal strategy optimization (PPO). A neural network structure is customized for Deep MEIRL, which uses the velocity of the ego vehicle, the pedestrian position, the velocity of surrounding vehicles, the lateral distance, the surrounding vehicles’ type, and the distance to the crosswalk to recover the nonlinear reward function. The dataset of drone-based video footage is collected in Xi’an (China) to train and validate the framework. The outcomes demonstrate that Deep MEIRL-PPO outperforms traditional modeling frameworks (Maximum Entropy Inverse Reinforcement Learning (MEIRL) - PPO) in modeling and predicting human driving behavior. Specifically, in predicting human driving behavior, Deep MEIRL-PPO outperforms MEIRL-PPO by 50.71% and 43.90% on the basis of the MAE and HD, respectively. Furthermore, it is discovered that Deep MEIRL-PPO accurately learns the behavior of human drivers avoiding potential conflicts when lines of sight are occluded. This research can contribute to aiding self-driving vehicles in learning human driving behavior and avoiding unforeseen risks.

Analytical Investigation of the Effects of Additional Load Mass on the Fundamental Frequency of Pedestrian Beam Bridges

The aspect of resonant vibrations due to pedestrian movement is of great significance in engineering practice. Therefore, understanding the variations in the natural frequency of bridge structures under varying positions of additional mass is of particular interest. This paper presents a procedure for the straightforward determination of the natural frequencies of a beam pedestrian bridge for various positions of pedestrians or a service vehicle based on derived analytical solutions. The calculation takes into account the inertial effects of the additional load mass, modeled as either uniformly distributed or concentrated. The importance of additional load mass effects on the fundamental frequency of a beam pedestrian bridge and its dynamic response to a moving pedestrian load is demonstrated on a bridge example. The proposed solutions are also applicable to other girder system structures with uniform mass and stiffness along their span.

Pedestrian detection using a MEMS acoustic array mounted on a moving vehicle

Automatic pedestrian detection in a vehicle is of vital importance in Advanced Driver Assistance Systems (ADAS). Sensors currently used are based on cameras, radars and lidars, whose performance is degraded in environments with reduced visibility: night, smoke, fog, etc. This paper has validated the use of an active array of 150 MEMS (Micro-Electro Mechanical Systems) microphones incorporated into a conventional car moving in real urban traffic conditions, with the system working in real time, at a rate of 8 detections per second. Together with beamforming, Constant False Alarm Rate (CFAR) detection and lane detection algorithms, a crucial algorithm has been incorporated for the proper performance of the system that discriminates the detections of objects or pedestrians generated by the system from the false alarms generated by road imperfections, such as bumps, cracks, etc. Based on 6000 captures, performed with the car moving at 30 km/h, which is the typical speed limit in urban environments, it has been possible to detect pedestrians positioned at a distance from the car varying between 5 and 20 m, with a detection probability of 0.91 and a false alarm probability of 0.01. The results obtained have validated the effectiveness of using active acoustic arrays in the field of pedestrian detection and position estimation from moving cars in urban environments. The fusion of the presented system with the systems currently used for this purpose, would significantly improve the performance of pedestrian detection systems interacting with AEB (Automatic Emergency Breaking) systems, extending their operability to environments with reduced visibility, and resulting in the reduction in the number of possible collisions with pedestrians, thus increasing their safety.

Comprehensive Survey of Autonomous Driving Systems and Advanced Technologies: Literature Review, Gap Analysis and Future Directions

Abstract: This article presents a literature review of numerous research papers focusing around the topic ‘Development of Autonomous Driving Vehicle System’, and related works. The advent of autonomous driving technology has revolutionized the automotive industry, promising enhanced safety, efficiency, and convenience on the road. A critical aspect of autonomous vehicle development is the assessment of their performance ina controlled and risk-free environment. Virtual environments, enabled by cutting-edge technologies, offer a dynamic plat- form for rigorous testing and training of autonomous vehicles. This paper aims to study and analyze different approaches and methodologies that are implemented for the development of Autonomous Driving Systems(ADS). Some of the primary aspects that are found common in most papers are Virtual Environment, pedestrians’ safety, privacy, security, real-time datapresentation, precise torque control, Reinforcement Learning, Deep Learning, Hardware-in-the-loop (HIL)- simulation, Model- in-the-loop (MIL)-simulation, etc. The sole idea of extracting valuable information from a virtual environment ensures a sense of safety as there are no humans involuntarily involved to take part in the development. However, the data extracted from the virtual environment must be highly accurate and reliable, as it will be trained and tested in real environments post deployment. Collision scenarios need to be carefully studied, for which relative positioning of vehicle and pedestrians should be taken into account, so as to examine their velocities, time to collision, and appropriately taking actions. Above all, it should be noted that the safety of human lives holds the highest priority, if a method suggests a high risk factor for a human life, then it should be either discarded, or improved

An advanced adaptive algorithm driven by online blind noise level estimation for pedestrian positioning

An effective and universal positioning algorithm is essential for a highly reliable and accurate foot-mounted inertial pedestrian navigation system. Algorithm performance is significantly influenced by the accuracy of noise level estimation. This study introduces a novel blind noise estimation algorithm to reduce the impact of uncertain and variable system noise characteristics on positioning accuracy. The algorithm combines a quasi-population growth wavelet transform threshold with an adaptive window length adjustment strategy. Firstly, a quasi-population growth wavelet threshold function is employed to separate motion signals from noise and assess noise levels. Secondly, an adaptive sliding window strategy based on motion-noise correlation optimizing the balance between estimation precision and computational efficiency. These methods enable robust noise level estimation without prior knowledge and achieve real-time adjustments to the system noise matrix in Extended Kalman Filter of the inertial navigation system, thereby significantly enhancing navigation error suppression. The effectiveness of above methodologies has been experimentally validated.

Effect of Impact Speed and Position of Pedestrian on Lower Extremity Injuries Caused by a Pickup Truck in Traffic Accidents

Pedestrians often suffer severe injuries in road traffic accidents. Therefore, pedestrian safety and, more precisely, analysis of injuries of the knee joint and lower limb bones were the focus of this research. A Vietnamese-sized human body model (HBM), named V-THUMS, scaled from the Total Human Model for Safety (THUMS) representing a walking pedestrian, was utilized in a numerical simulation. A pickup truck model was used to simulate pedestrian collisions with impact speeds from 20 km/h to 50 km/h. To reduce the computational time, the pickup truck model was simplified. In addition, the collision angle was varied from 0o, 30o, 60o, 90o, -30o, -60o to -90o to explore the effect of angle position on the injury mechanisms of the leg. The results illustrate that in some cases at low speeds there is a relationship between bone fracture and ligament rupture that does not exist at high speeds. The femur is very difficult to fracture because the bone is very hard and thick. It was found that at 20 km/h impact speed, the lower limbs are unharmed.