Verifying training datasets in vision-based vehicle safety applications is crucial to understanding the potential limitations of detection capabilities that may result in a higher safety risk. Vision-based pedestrian safety applications with crash avoidance technologies rely on prompt detection to avoid a crash. This research aims to develop a verification process for vulnerable road user safety applications with vision-based detection functionalities. It consists of reviewing the application’s safety requirements, identifying the target objects of detection in the operational design domain and pre-crash scenarios, and evaluating the safety risks qualitatively by examining the training dataset based on the results of pre-crash scenarios classification. As a demonstration, the process is implemented using open-source pedestrian tracking software, and the pre-crash scenarios are classified based on the trajectories of pedestrians in an example training dataset used in a pedestrian

Hsu, Chung-Jen

Lightweight Neural Network Model and Algorithm for Pedestrian Detection

12-08-03-0027

12/18/2024

Traditional pedestrian detection methods have poor robustness. Deep learning-based methods have shown high performance in recent years but rely on substantial computational resources. Developing a lightweight, deep learning-based pedestrian detection algorithm is essential for applying deep learning-based algorithms in resource-limited scenarios, such as driverless and advanced driver assistance systems. In this article, an improved model based on YOLOv3 called “YOLOPD” (You Only Look Once—Pedestrian Detection), is proposed. It is obtained by constructing a self-attentive module, introducing a CIOU (Complete Intersection over Union) loss function and a depth separated convolutional layer. Experimental results show that on the INRIA (National Institute for Research in Computer Science and Automation), Caltech, and CityPerson pedestrian dataset, the MR (miss rate) of the model YOLOPD is better than that of the original YOLOv3 model, and the number of parameters is reduced by about 1/3

Li, Shanglin, Wang, Qi Feng, Li, Ren Fa, Xiao, Juan

MoVE: A Mobility Virtual Environment for Testing Multi-Vehicle Scenarios

2024-01-3802

11/15/2024

ABSTRACT This paper presents a Mobility Virtual Environment (MoVE) for testing multi-vehicle autonomy scenarios with real and simulated vehicles and pedestrians. MoVE is a network-centric framework designed to represent N real and M virtual vehicles interacting and possibly communicating with each other in the same coordinate frame with a common timestamp. The goal is to provide a spectrum of test options from simulation-only to semi-virtual, to all real vehicles and pedestrians. A multi-vehicle test fidelity metric is defined that captures scenario realism more accurately than traditional hardware-in-the-loop style terminology. MoVE’s simple built-in vehicle models are described that provide positions in both latitude and longitude and Cartesian UTM XYZ coordinates. Live GPS inputs from real people or vehicles allow both virtual and real vehicles to interact through the virtual environment. Test results are presented from three experiments with real and virtual vehicles and

Compere, Marc, Adkins, Kevin, Legon, Otto, Currier, Patrick

AUTONOMY MODELING AND VALIDATION IN A HIGHLY UNCERTAIN ENVIRONMENT

2024-01-3732

11/15/2024

ABSTRACT This paper describes a simulation model for autonomous vehicles operating in highly uncertain environments. Two elements of uncertainty are studied – rain and pedestrian interaction – and their effects on autonomous mobility. The model itself consists of all the essential elements of an autonomous vehicle: Scene -roads, buildings, etc., Environment - sunlight, rain, snow, etc., Sensors - gps, camera, radar, lidar, etc., Algorithms - lane detection, pedestrian detection, etc., Control - lane keeping, obstacle avoidance, etc., Vehicle Dynamics – mass, drivetrain, tires, etc., and Actuation - throttle, braking, steering, etc. Using this model, the paper presents results that assess the autonomous mobility of a Polaris GEM E6 type of vehicle in varying amounts of rain, and when the vehicle encounters multiple pedestrians crossing in front. Rain has been chosen as it impacts both situational awareness and trafficability conditions. Mobility is measured by the average speed of the

Alghodhaifi, Hesham, Lakshmanan, Sridhar, Baek, Stanley, Richardson, Paul

Skeletal Surrogate Metastructural Anisotropy Mimicry and the Influence of Fabrication Material

09-12-02-0009

11/12/2024

Human body models have been used for decades to inform efforts in promoting automobile occupant and pedestrian safety. However, many of these models fail to capture the intricacies of individual variability. Cadaveric subjects typically exceed representative age ranges and hence mechanics. Animal subjects typically require specific setups that stray from that which is representative of human crash scenarios. Computational models can only consider so many practical real-world variables. Artificial surrogates, dummies being popular among them, are very popular for reusability and robust data collection. However, even the biomechanically accurate skeletal surrogates available commercially are limited in that they do not consider human variability and skeletal microstructure local variability. The objective of the work herein is to assess computational methods of metastructural variability mimicry by fabrication material. We implement mimicry approaches focusing on bulk isotropic

Hezrony, Benjamin S., C. F. Lopes, Pedro, Brown, Philip J.

Enhancing Availability of Autonomous Shuttle Services: A Conceptual Approach toward Challenges and Opportunities

12-08-03-0023

10/09/2024

Autonomous vehicle technologies have become increasingly popular over the last few years. One of their most important application is autonomous shuttle buses that could radically change public transport systems. In order to enhance the availability of shuttle service, this article outlines a series of interconnected challenges and innovative solutions to optimize the operation of autonomous shuttles based on the experience within the Shuttle Modellregion Oberfranken (SMO) project. The shuttle shall be able to work in every weather condition, including the robustness of the perception algorithm. Besides, the shuttle shall react to environmental changes, interact with other traffic participants, and ensure comfortable travel for passengers and awareness of VRUs. These challenging situations shall be solved alone or with a teleoperator’s help. Our analysis considers the basic sense–plan–act architecture for autonomous driving. Critical components like object detection, pedestrian tracking

Dehghani, Ali, Salaar, Hamza, Srinivasan, Shanmuga Priya, Zhou, Lixian, Arbeiter, Georg, Lindner, Alisa, Patino-Studencki, Lucila

Micromobility and the Next Infrastructure Wave

EPR2024022

10/04/2024

US transportation infrastructure is dominated by the automobile form factor. Alternative modalities of movement, such as bikes, golf carts, and other micromobility options, have existed but are decidedly at a lower tier of importance. Even pedestrian access ways are not overly emphasized in the US transportation system. This lack of prioritization matches the reality that the vast majority of people and commerce moves through the motor vehicle infrastructure, with micromobility sitting in the periphery. Additionally, given the current lack of commercial applications, there are limited direct fee-based funding mechanisms connected to micromobility form factors. Micromobility and the Next Infrastructure Wave discusses how recent technological innovations in electrification, e-commerce, and autonomy are enabling a new class of micromobility devices which offer palpable value to consumers and enable significant commercial applications. Unlike the past, these micromobility devices now have

Razdan, Rahul

Comprehensive Study on Challenges to Introduce Autonomous Emergency Braking (AEBS) Feature in Indian Perspective

2024-01-3050

09/08/2024

India is a diverse country in terms of road conditions, road maintenance, traffic conditions, traffic density, quality of traffic which implies presence of agricultural tractors, bullock carts, autos, motor bikes, oncoming traffic in same lane, vulnerable road users (VRU) walking in the same lanes as vehicles, VRU’s crossing roads without using zebra crossings etc. as additional traffic quality deterrents in comparison to developed countries. The braking capacity of such vivid road users may not be at par with global standards due to their maintenance, loading beyond specifications, driver behavior which includes the tendency to maintain a close gap between the preceding vehicle etc. which may lead to incidents specifically of rear collisions due to the front vehicle going through an emergency braking event. The following paper provides a comprehensive study of the special considerations or intricacies in implementation of Autonomous Emergency Braking (AEBS) feature into Indian traffic

Kartheek, Nedunuri, Khare, Rashmita, Sathyamurthy, Sainathan, Manickam, Praveenkumar, Kuchipudi, Venkata Sai Pavan

Effect of A-Pillar Blind Spots on a Driver’s Pedestrian Visibility during Vehicle Turns at an Intersection

2024-22-0002

09/05/2024

This study aims to elucidate the impact of A-pillar blind spots on drivers’ visibility of pedestrians during left and right turns at an intersection. An experiment was conducted using a sedan and a truck, with a professional test driver participating. The driver was instructed to maintain sole focus on a designated pedestrian model from the moment it was first sighted during each drive. The experimental results revealed how the blind spots caused by A-pillars occur and clarified the relationship between the pedestrian visible trajectory distance and specific vehicle windows. The results indicated that the shortest trajectory distance over which a pedestrian remained visible in the sedan was 17.6 m for a far-side pedestrian model during a right turn, where visibility was exclusively through the windshield. For the truck, this distance was 20.9 m for a near-side pedestrian model during a left turn, with visibility through the windshield of 9.5 m (45.5% of 20.9 m) and through the

Matsui, Yasuhiro, Oikawa, Shoko

Toward Trustworthy Automated Driving through Qualitative Scene Understanding and Explanations

12-08-01-0003

07/12/2024

Understanding driving scenes and communicating automated vehicle decisions are key requirements for trustworthy automated driving. In this article, we introduce the qualitative explainable graph (QXG), which is a unified symbolic and qualitative representation for scene understanding in urban mobility. The QXG enables interpreting an automated vehicle’s environment using sensor data and machine learning models. It utilizes spatiotemporal graphs and qualitative constraints to extract scene semantics from raw sensor inputs, such as LiDAR and camera data, offering an interpretable scene model. A QXG can be incrementally constructed in real-time, making it a versatile tool for in-vehicle explanations across various sensor types. Our research showcases the potential of QXG, particularly in the context of automated driving, where it can rationalize decisions by linking the graph with observed actions. These explanations can serve diverse purposes, from informing passengers and alerting

Belmecheri, Nassim, Gotlieb, Arnaud, Lazaar, Nadjib, Spieker, Helge

Harmonizing Safety Regulations and Perception: A Simulation Methodology for AVAS System Design

2024-01-2915

06/12/2024

The development of an effective Acoustic Vehicle Alerting System (AVAS) is not solely about adhering to safety regulations; it also involves crafting an auditory experience that aligns with the expectations of vulnerable road users. To achieve this, a deep understanding of the acoustic transfer function is essential, as it defines the relationship between the sound emitter (the speaker inside the vehicle) and the receiver (the vulnerable road user). Maintaining the constancy of this acoustic transfer function is paramount, as it ensures that the sound emitted by the vehicle aligns with the intended safety cues and brand identity that is defined by the car manufacturer. In this research paper, three distinct methodologies for calculating the acoustic transfer function are presented: the classical Boundary Element method, the H-Matrix BEM accelerated method, and the Ray Tracing method. Furthermore, the paper encompasses an assessment of the correlation between these methods and their

Calloni, Massimiliano, Hadjit, Rabah, Salvekar, Pinak, Musser, Chad

Lighting and Visual Information for Vulnerable Road User (VRU) Safety: An Introductory Review

VRUSC-002-202405/03/2024

Injuries and fatalities among pedestrians, cyclists, scooterists, highway road workers, and safety and emergency personnel—often referred to as vulnerable road users (VRUs)—continue to rise at alarming rates worldwide. Emphasizing the urgent need for enhanced safety measures, this study, commissioned by the Vulnerable Road User Safety Consortium™ (VRUSC) and conducted by the Light and Health Research Center at the Icahn School of Medicine at Mount Sinai, evaluates the potential effectiveness of lighting and visual information systems in improving VRU safety. The white paper presents a synthesis of published research on lighting and markings from the perspective of both human drivers and machine vision systems. It identifies potential preliminary guidelines for the intensity, color, temporal, and spatial characteristics of lighting and visual information that can help prevent crashes involving VRUs. The study also highlights knowledge gaps that require further research, including the

Vulnerable Road Users Safety Consortium

Evaluation of the Full-Frontal Crash Regulation for the M1 Category of Vehicles from an Indian Perspective

2024-01-2750

04/09/2024

Background: The Indian automobile industry, including the auto component industry, is a significant part of the country’s economy and has experienced growth over the years. India is now the world’s 3rd largest passenger car market and the world’s second-largest two-wheeler market. Along with the boon, the bane of road accident fatalities is also a reality that needs urgent attention, as per a study titled ‘Estimation of Socio-Economic Loss due to Road Traffic Accidents in India’, the socio-economic loss due to road accidents is estimated to be around 0.55% to 1.35% of India’s GDP [27] Ministry of road transport and highways (MoRTH) accident data shows that the total number of fatalities on the road are the highest (in number terms) in the world. Though passenger car occupant fatalities have decreased over the years, the fatalities of vulnerable road users are showing an increasing trend. India has committed to reduce road fatalities by 50% by 2030. In this context, the automotive

Mehta, Pooja, Prasad, Avinash, Srivastava, Aakash, Arora, Pankaj, Howlader, Ashim

Evaluating Safety Metrics for Vulnerable Road Users at Urban Traffic Intersections Using High-Density Infrastructure LiDAR System

2024-01-2641

04/09/2024

Ensuring the safety of vulnerable road users (VRUs) such as pedestrians, users of micro-mobility vehicles, and cyclists is imperative for the commercialization of automated vehicles (AVs) in urban traffic scenarios. City traffic intersections are of particular concern due to the precarious situations VRUs often encounter when navigating these locations, primarily because of the unpredictable nature of urban traffic. Earlier work from the Institute of Automated Vehicles (IAM) has developed and evaluated Driving Assessment (DA) metrics for analyzing car following scenarios. In this work, we extend those evaluations to an urban traffic intersection testbed located in downtown Tempe, Arizona. A multimodal infrastructure sensor setup, comprising a high-density, 128-channel LiDAR and a 720p RGB camera, was employed to collect data during the dusk period, with the objective of capturing data during the transition from daylight to night. In this study, we present and empirically assess the

Rath, Prabin Kumar, Harrison, Blake, Lu, Duo, Yang, Yezhou, Wishart, Jeffrey, Yu, Hongbin

Representative Cyclist Collision Injury Risk Distributions for a Dense-Urban US ODD Using Naturalistic Dash Camera Data

2024-01-2645

04/09/2024

Automated driving systems (ADS) are designed toward safely navigating the roadway environment, which also includes consideration of potential conflict with other road users. Of particular concern is understanding the cumulative risk associated with vulnerable road users (VRUs) conflicts and collisions. VRUs represent a population of road users that have limited protection compared to vehicle occupants. These severity distributions are particularly useful in evaluating ADS real-world performance with respect to the existing fleet of vehicles. The objective of this study was to present event severity distributions associated with vehicle-cyclist collisions within an urban naturalistic driving environment by leveraging data from third-party vehicles instrumented with forward-facing cameras and a sensor suite (accelerometer sampling at 20 Hz and GPS [variable sampling frequency]). From over 66 million miles of driving, 30 collision events were identified. A global optimization routine was

Campolettano, Eamon T., Scanlon, John M., Kusano, Kristofer D.

Integrating Machine Learning in Pedestrian Forensics: A Comprehensive Tool for Analysing Pedestrian Collisions

2024-01-2468

04/09/2024

Analysis of pedestrian-to-vehicle collisions can be complex due to the nature of the interaction and the physics involved. The scarcity of evidence like video evidence (from CCTV or dashcams), data from the vehicle's ECU, witness accounts, and physical evidence such as tyre marks, complicates the analysis of these incidents. In cases with limited evidence, current forensic methods often rely on prolonged inquiry processes or computationally intensive simulations. Without adequate data, accurately estimating pedestrian kinematics and addressing hit-and-run scenarios becomes challenging. This research provides an alternative approach to enhancing pedestrian forensic analysis based on machine learning (ML) algorithms trained on over 3000 multi-body computer simulations with a diverse set of vehicle profiles and pedestrian anthropometries. Leveraging information such as vehicle profile, damage, and pedestrian attributes like height and weight, the ML algorithm estimates essential

Shrinivas, Vadhiraj, Bastien, Christophe, Davies, Huw, Daneshkhah, Alireza, Hardwicke, Joseph, Neal-Sturgess, Clive, Lamaj, Albi

Vehicle-in-Virtual-Environment Method for ADAS and Connected and Automated Driving Function Development, Demonstration and Evaluation

2024-01-1967

04/09/2024

The current approach for new Advanced Driver Assistance System (ADAS) and Connected and Automated Driving (CAD) function development involves a significant amount of public road testing which is inefficient due to the number miles that need to be driven for rare and extreme events to take place, thereby being very costly also, and unsafe as the rest of the road users become involuntary test subjects. A new development, evaluation and demonstration method for safe, efficient, and repeatable development, demonstration and evaluation of ADAS and CAD functions called Vehicle-in-Virtual –Environment (VVE) was recently introduced as a solution to this problem. The vehicle is operated in a large, empty, and flat area during VVE while its localization and perception sensor data is fed from the virtual environment with other traffic and rare and extreme events being generated as needed. The virtual environment can be easily configured and modified to construct different testing scenarios on

Cao, Xincheng, Chen, Haochong, Gelbal, Sukru Yaren, Aksun Guvenc, Bilin, Guvenc, Levent

Evaluating Vehicle Response Through Non-Traditional Pedestrian Automatic Emergency Braking Scenarios

2024-01-1975

04/09/2024

Pedestrian Automatic Emergency Braking (P-AEB) is a technology designed to avoid or reduce the severity of vehicle to pedestrian collisions. This technology is currently assessed and evaluated via EuroNCAP and similar procedures in which a pedestrian test target is crossing the road, walking alongside the road, or stationary in the forward vehicle travel path. While these assessment methods serve the purpose of providing cross-comparison of technology performance in a standardized set of scenarios, there are many scenarios which could occur which are not considered or studied. By identifying and performing non-EuroNCAP, non-standardized scenarios using similar methodology, the robustness of P-AEB systems can be analyzed. These scenarios help identify areas of further development and consideration for future testing programs. Three scenarios were considered as a part of this work: straight line approach, curved path approach, and parking lot testing. Exemplar tests were performed for

Bartholomew, Meredith, Helber, Nicholas, Heydinger, Gary, Zagorski, Scott

Development of Predictive Model for Flexible Pedestrian Legform Impactor Injury

2024-01-2511

04/09/2024

Road accidents are a major concern worldwide and vulnerable road users make up more than half of the victims of road accident deaths. In order to combat this issue, several countries worldwide have mandated pedestrian safety test regulations viz., AIS100 & UN-R127. One of the requirements of the regulations is when Flexible Pedestrian Legform Impactor (Flex-PLI) is impacted onto the frontal structure of the vehicle at a speed of 40kmph, the Bending moment (BM) of tibia bone of Flex-PLI shall not exceed the regulatory limit of 340Nm. In this paper, we have built a statistical model for predicting the BM of tibia in Flex-PLI using regression analysis. 13 vehicles have been selected from all applicable vehicle categories viz., Sedan, hatchback, Coupe & SUV/MUV for this undertaking. An exhaustive analysis of the vehicle frontal structures and Flex-PLI test videos have been done to identify & measure the design parameters to be used as predictor variables. The vehicles have then been

Barbhuiya, Junaid Hassan, Jain, Subhav

Sensor-Fused Low Light Pedestrian Detection System with Transfer Learning

2024-01-2043

04/09/2024

Objection detection using a camera sensor is essential for developing Advanced Driver Assistance Systems (ADAS) and Autonomous Driving (AD) vehicles. Due to the recent advancement in deep Convolution Neural Networks (CNNs), object detection based on CNNs has achieved state-of-the-art performance during daytime. However, using an RGB camera alone in object detection under poor lighting conditions, such as sun flare, snow, and foggy nights, causes the system's performance to drop and increases the likelihood of a crash. In addition, the object detection system based on an RGB camera performs poorly during nighttime because the camera sensors are susceptible to lighting conditions. This paper explores different pedestrian detection systems at low-lighting conditions and proposes a sensor-fused pedestrian detection system under low-lighting conditions, including nighttime. The proposed system fuses RGB and infrared (IR) thermal camera information. IR thermal cameras are used as they are

Thota, Bharath kumar, Somashekar, Karthik, Park, Jungme

Research on Design of Electric Vehicle Sound Synthesis Based on Frequency Shift Algorithm

2024-01-2335

04/09/2024

The active sound generation systems (ASGS) for electric vehicles (EVs) play an important role in improving sound perception and transmission in the car, and can meet the needs of different user groups for driving and riding experiences. The active sound synthesis algorithm is the core part of ASGS. This paper uses an efficient variable-range fast linear interpolation method to design a frequency-shifted and pitch-modified sound synthesis algorithm. By obtaining the operating parameters of EVs, such as vehicle speed, motor speed, pedal opening, etc., the original sound signal is interpolated to varying degrees to change the frequency of the sound signal, and then the amplitude of the sound signal is determined according to different driving states. This simulates an effect similar to the sound of a traditional car engine. Then, a dynamic superposition strategy is proposed based on the Hann window function. Through windowing and superposition processing of each sound signal segment

Yu, Shangbo, Xie, Liping, Lu, Chihua, Qian, Yushu, Liu, Zhien, Songze, Du

Assessing the Impact of Rubberized Asphalt on Reducing Hip Fracture Risk in Elderly Populations Using Human Body Models

09-12-01-0007

04/08/2024

Compared to other age groups, older adults are at more significant risk of hip fracture when they fall. In addition to the higher risk of falls for the elderly, fear of falls can reduce this population’s outdoor activity. Various preventive solutions have been proposed to reduce the risk of hip fractures ranging from wearable hip protectors to indoor flooring systems. A previously developed rubberized asphalt mixture demonstrated the potential to reduce the risk of head injury. In the current study, the capability of the rubberized asphalt sample was evaluated for the risk of hip fracture for an average elderly male and an average elderly female. A previously developed human body model was positioned in a fall configuration that would give the highest impact forces toward regular asphalt. Three different rubber contents with 14, 28, 33 weight percent (% wt.) were implemented as the ground alongside one regular non-rubberized (0%) asphalt mixture, one baseline, and one extra-compliant

Sahandifar, Pooya, Wallqvist, Viveca, Kleiven, Svein

A Simulation Methodology to Design the AVAS System to Meet Safety Regulations and Create the Expected Perception for the Vulnerable Road User

2024-26-0230

01/16/2024

Designing an effective AVAS system, not only to meet safety regulations, but also to create the expected perception for the vulnerable road user, relies on knowledge of the acoustic transfer function between the sound actuator and the receiver. It is preferable that the acoustic transfer function be as constant as possible to allow transferring the sound designed by the car OEM to ensure the safety of vulnerable road users while conveying the proper brand image. In this paper three different methodologies for the acoustic transfer function calculations are presented and compared in terms of accuracy and calculation time: classic Boundary Element method, H-Matrix BEM accelerated method and Ray tracing method. An example of binaural listening experience at different certification positions in the modeled simulated space is also presented

Calloni, Massimiliano, Hadjit, Rabah, Salvekar, Pinak, Musser, Chadwyck

Comprehensive Assessment of Driver Monitoring System for Commercial Vehicle Applications Using Innovative Lab Testing Approach

2024-26-0027

01/16/2024

The commercial vehicle sector (especially trucks) has major role in economic growth of a nation. With improving infrastructure, increasing number of commercial vehicles and growing amount of Vulnerable Road Users (VRUs) on roads, accidents are also increasing. As per RASSI (Road Accident Sampling System India) FY2016-21 database, commercial vehicles are involved in 43% of total accidents on Indian roads. One of the major causes of these accidents is Driver Drowsiness and Inattention (DDI) (approx. 10% contribution in total accidents). This paper describes novel driver-in-loop performance assessment methodology for comprehensive verification of Driver Monitoring System (DMS) for commercial vehicle application. Novelty lies in specification of test subjects, driving styles and variety of road traffic scenarios for verification of DMS system. Test setup is made modular to cater to different platform environments (Heavy, Intermediate, Light) with minor modifications. The test setup

Sudarshan, BV, Jarhad, Manoj, Dey, Susanta Kumar, Joshi, Kedar Shrikant, Gadekar, Ganesh

Modelling and Evaluation of Pedestrian Active Safety System Based on Accidents in India

2024-26-0013

01/16/2024

Many Vulnerable Road Users (VRUs) experience injuries and fatalities every year, making road safety a challenge in the World. According to the Ministry of Road Transport and Highway (MoRTH) during the year 2021, a total number of 4,12,432 road accidents have been reported in India, claiming 1,53,972 lives and causing injuries to 3,84,448 persons. In terms of road-user categories, the total fatality of the Pedestrian road-users was 18.9 per cent of persons killed in road accidents. One of the ways to handle the situation is to protect pedestrians utilizing active safety measures in the vehicle. In addition, active safety research heavily relies on perceptions of pre-crash scenarios. The objective of the study is to examine passenger car-to-pedestrian scenarios and model active safety system in a car to prevent or mitigate collisions. Road Accident Sampling System – India (RASSI) database is used for this research, which provided a comprehensive set of accident data describing the

Doshi, Urvilkumar HItendrabhai, Agrawal, Saurav, Muthanandam, Muthukumar, Yarramsetti, Karthik, Penumaka, Avinash, Kalakala, Vijaya Prakash

Research on Vulnerable Road User Detection Algorithm based on Improved Deep Learning

2023-01-7050

12/20/2023

This paper proposes a detection algorithm based on deep learning for Vulnerable Road Users such as pedestrians and cyclists, which is improved on the basis of YOLOv5 network model. (1) Aiming at the problems of low resolution and insufficient information for small targets, a multi-scale feature fusion method is adopted to integrate shallow features with deep features. In this way, the effective information of small target is enriched, and the accuracy of target detection is improved. (2) In view of the interference of image noise, background and other factors, the channel attention method is introduced to strengthen key features and suppress the interference of noise, which can improve the model's attention to small targets and enable the network to better identify blocked targets; (3) Aiming at the problem that the computing speed of the model is difficult to achieve real-time performance, a deep separable convolution optimization method is proposed to reduce the amount of computation

Yi, Zhenxing, Zhan, Zhenfei, Zhou, Guilin, Lv, Fengyao, Mao, Qing, Sun, Bowen

Initial Research Summation Overview: Vehicle-VRU Safety Landscape

VRUSC-001-202312/18/2023

This white paper is an examination of the increasing rate of injuries and deaths worldwide of a class of road users, often referred to as “vulnerable road users” (VRUs), with a focus on pedestrians, cyclists, scooterists, highway road workers, safety, and emergency personnel. The potential to leverage pragmatic, evidence-based technology countermeasures to reduce these collisions, and the severity of those that do occur, is also examined

Vulnerable Road Users Safety Consortium

Pedestrian Intention Prediction and Style Recognition in Bird’s-Eye View

12-07-01-0008

11/16/2023

In this article, pedestrian crossing intention and pedestrian crossing style are studied by means of statistical theory and artificial neural network. Feature parameters such as the average speed of pedestrians, pedestrian attention to vehicles, and vehicle arrival speed are extracted before and during the time pedestrians cross the street from a bird’s-eye view. Based on these parameters, an artificial neural network is used to predict the pedestrian crossing intention. K-means statistical method was used to cluster the pedestrian crossing styles, and the results showed that clustering the crossing styles into three categories, conservative, cautious, and adventurous, has a better classification effect, and the crossing behaviors of different types of pedestrians were analyzed. A random forest-based model is used to identify pedestrian crossing styles, the prediction accuracy reaches 91.83% and the recognition accuracy reaches 93.3%. The research content of this article can provide

Li, Wenli, Wang, Mengxin, Gong, Xiaohao, Tang, Yuanhang

Letter from the Special Issue Editors

09-11-02-0009

09/20/2023

Letter from the Special Issue Editors

Mueller, Becky, Bautsch, Brian, Mansfield, Julie

Cooperative Driving Automation (CDA) Feature: Perception Status Sharing for Occluded Pedestrian Collision Avoidance

J3251_202308 (Current)

08/22/2023

This SAE Information Report develops a concept of operations (ConOps) to evaluate a cooperative driving automation (CDA) Feature for occluded pedestrian collision avoidance using perception status sharing. It provides a test procedure to evaluate this CDA Feature, which is suitable for proof-of-concept testing in both virtual and test track settings

Cooperative Driving Automation(CDA) Committee

Effect of Tissue Erosion Modeling Techniques on Pedestrian Impact Kinematics

2022-22-0007

06/27/2023

The pedestrian is one of the most vulnerable road users and has experienced increased numbers of injuries and deaths caused by car-to-pedestrian collisions over the last decade. To curb this trend, finite element models of pedestrians have been developed to investigate pedestrian protection in vehicle impact simulations. While useful, modeling practices vary across research groups, especially when applying knee/ankle ligament and bone failure. To help better standardize modeling practices this study explored the effect of knee ligament and bone element elimination on pedestrian impact outcomes. A male 50th percentile model was impacted by three European generic vehicles at 30, 40, and 50 km/h. The pedestrian model was set to three element elimination settings: the “Off-model” didn’t allow any element erosion, the “Lig-model” allowed lower-extremity ligament erosion, and the “All-model” allowed lower-extremity ligament and bone erosion. Failure toggling had a significant effect on

Grindle, Daniel, Untaroiu, Costin

Ligaments Laxity and Elongation at Injury in Flexed knees during Lateral Impact Conditions

2022-22-0003

06/27/2023

The knee is one of the regions of interest for pedestrian safety assessment. Past testing to study knee ligament injuries for pedestrian impact only included knees in full extension and mostly focused on global responses. As the knee flexion angle and the initial ligament laxity may affect the elongation at which ligaments fail, the objectives of this study were (1) to design an experimental protocol to assess the laxity of knee ligaments before measuring their elongation at failure, (2) to apply it in paired knee tests at two flexion angles (10 and 45 degrees). The laxity tests combined strain gauges to measure bone strains near insertions that would result from ligament forces and a custom machine to exercise the knee in all directions. Failure was assessed using a four-point bending setup with additional degrees of freedom on the axial rotation and displacement of the femur. A template was designed to ensure that the two setups used the exact same starting position. The protocol was

Benadi, Sahar, Trosseille, Xavier, Petit, Philippe, Uriot, Jérôme, Lafon, Yoann, Beillas, Philippe

Recognition Assistance Interface for Human-Automation Cooperation in Pedestrian Risk Prediction

12-06-03-0023

06/06/2023

Autonomous driving systems (ADS) have been widely tested in real-world environments with operators who must monitor and intervene due to remaining technical challenges. However, intervention methods that require operators to take over control of the vehicle involve many drawbacks related to human performance. ADS consist of recognition, decision, and control modules. The latter two phases are dependent on the recognition phase, which still struggles with tasks involving the prediction of human behavior, such as pedestrian risk prediction. As an alternative to full automation of the recognition task, cooperative recognition approaches utilize the human operator to assist the automated system in performing challenging recognition tasks, using a recognition assistance interface to realize human-machine cooperation. In this study, we propose a recognition assistance interface for cooperative recognition in order to achieve safer and more efficient driving through improved human-automation

Kuribayashi, Atsushi, Takeuchi, Eijiro, Carballo, Alexander, Ishiguro, Yoshio, Takeda, Kazuya

Tonal Component Separation of e-Vehicles Using the High-Resolution Spectral Analysis (HSA

2023-01-1141

05/08/2023

E-vehicles can generate strong tonal components that may disturb people inside the vehicle. However, such components, deliberately generated, may be necessary to meet audibility standards that ensure the safety of pedestrians outside the vehicle. A tradeoff must be made between pedestrian audibility and internal sound quality, but any iteration that requires additional measurements is costly. One solution to this problem is to modify the recorded signals to find the variant with the best sound quality that complies with regulations. This is only possible if there is a good separation of the tonal components of the signal. In this work, a method is proposed that uses the High-resolution Spectral Analysis (HSA) to extract the tonal components of the signal, which can then be recombined to optimize any sound quality metric, such as the tonality using the Sottek Hearing Model (standardized in ECMA 418-2

Sottek, Roland, Gomes Lobato, Thiago Henrique

A Process to Characterize the Sound Directivity Pattern of AVAS Speaker

2023-01-1095

05/08/2023

Speaker performance in Acoustic Vehicle Alerting System (AVAS) plays a crucial role for pedestrian safety. Sound radiation from AVAS speaker has obvious directivity pattern. Considering this feature is critical for accurately simulating the exterior sound field of electrical vehicles. This paper proposes a new process to characterize the sound directivity pattern of AVAS speaker. The first step of the process is to perform an acoustic testing to measure the sound pressure radiated from the speaker at a certain number of microphone locations in a free field environment. Based on the geometry of a virtual speaker, the locations of each microphone and measured sound pressure data, an inverse method, namely the inverse pellicular analysis, is adopted to recover a set of vibration pattern of the virtual speaker surface. The recovered surface vibration pattern can then be incorporated in the full vehicle numerical model as an excitation for simulating the exterior sound field. In this study

Yang, Wenlong, Wang, Chong, Zhang, Qijun

Hierarchical Neural Network-Based Prediction Model of Pedestrian Crossing Behavior at Unsignalized Crosswalks

2023-01-0865

04/11/2023

To enable smooth and low-risk autonomous driving in the presence of other road users, such as cyclists and pedestrians, appropriate predictive safe speed control strategies relying on accurate and robust prediction models should be employed. However, difficulties related to driving scene understanding and a wide variety of features influencing decisions of other road users significantly complexifies prediction tasks and related controls. This paper proposes a hierarchical neural network (NN)-based prediction model of pedestrian crossing behavior, which is aimed to be applied within an autonomous vehicle (AV) safe speed control strategy. Additionally, different single-level prediction models are presented and analyzed as well, to serve as baseline approaches. The hierarchical NN model is designed to predict the probability of pedestrian crossing the crosswalk prior to the vehicle at the high level, and parameters of Gaussian probability distribution of pedestrian entry time to the

Ćorić, Mate, Skugor, Branimir, Deur, Josko, Ivanovic, Vladimir, Tseng, H. Eric

Automotive Hood Design Based on Machine Learning and Structural Design Optimization

2023-01-0744

04/11/2023

Nowadays, the automobile industry is booming and the number of vehicles is proliferating while the road traffic environment is also deteriorating. Therefore, attention should be paid to the protection of vulnerable road users in traffic accidents, such as pedestrians. In order to reduce the pedestrians’ head injury in collision accidents, in this study, the vehicle engine hood which responds significantly to head injuries was taken as the design object, so as to put forward a new optimization design process. The parameters of the hood’s main components, manufacturing materials and structural scheme were considered to carry out simultaneous optimization from various aspects such as pedestrian protection and hood stiffness. Meanwhile, the approximate model approach was adopted to design the main parameters to improve the efficiency, and based on Bayesian inference, the approximate model bias correction method was proposed which solved the related problems of low accuracy of the

Zhan, Zhenfei, Fengyao, LV, Xin, Ran, Zhou, Guilin, Zhao, Shuen, He, Xin, Wang, Ju, Li, Jie

Introduction of the Small Test Robot for Individuals in Dangerous Environments (STRIDE) Platform for Use in ADAS Testing

2023-01-0795

04/11/2023

The use of platforms to carry vulnerable road user (VRU) targets has become increasingly necessary with the rise of automated driver assistance systems (ADAS) on vehicles. These ADAS features must be tested in a wide variety of collision-imminent scenarios which necessitates the use of strikable targets carried by an overrun-able platform. To enable the testing of ADAS sensors such as lidar, radar, and vision systems, S-E-A, a longtime supplier of vehicle testing equipment, has created the STRIDE robotic platform (Small Test Robot for Individuals in Dangerous Environments). This platform contains many of the key ingredients of other platforms on the market, such as a hot-swappable battery, E-stop, and mounting points for targets. However, the STRIDE platform additionally provides features which can enable non-routine testing such as: turning in place, driving with an app on a mobile phone, user-scripting, and steep grade climbing capability. The combination of these elements makes

Bartholomew, Meredith, Helber, Nicholas, Nguyen, An, Zagorski, Scott, Heydinger, Gary

Full-Scale Validation of Modified Pedestrian Dummy

2023-01-0786

04/11/2023

Injury assessment by using a whole-body pedestrian dummy is one of the ways to investigate pedestrian safety performance of vehicles. The authors’ group has improved the biofidelity of the lower limb and the pelvis of the mid-sized male pedestrian dummy (POLAR III) by modifying those components. This study aims to evaluate the biofidelity of the whole-body response of the modified dummy in full-scale impact tests. The pelvis, the thigh and the leg of POLAR III have been modified in a past study by optimizing their compliance by means of the installation of plastic and rubber parts, which were used for the tests. The generic buck developed for the assessment of pedestrian dummy whole-body impact response and specified in SAE J3093 was used for this study. The buck representing the geometry of a small family car is comprised of six parts: lower bumper, bumper, grille, hood edge, hood and windshield. Tests were performed by conforming to SAE J2782 that specifies test conditions to

Asanuma, Hiroyuki, Bae, Hyejin, Nakamura, Hidetoshi, Gunji, Yasuaki, Nagashima, Akiko, Mori, Fumie

Prescan Extension Testing of an ADAS Camera

2023-01-0831

04/11/2023

Testing vision-based advanced driver assistance systems (ADAS) in a Camera-in-the-Loop (CiL) bench setup, where external visual inputs are used to stimulate the system, provides an opportunity to experiment with a wide variety of test scenarios, different types of vehicle actors, vulnerable road users, and weather conditions that may be difficult to replicate in the real world. In addition, once the CiL bench is setup and operating, experiments can be performed in less time when compared to track testing alternatives. In order to better quantify normal operating zones, track testing results were used to identify behavior corridors via a statistical methodology. After determining normal operational variability via track testing of baseline stationary surrogate vehicle and pedestrian scenarios, these operating zones were applied to screen-based testing in a CiL test setup to determine particularly challenging scenarios which might benefit from replication in a track testing environment

Bartholomew, Meredith Carol, Guenther, Dennis, Midlam-Mohler, Shawn, Heydinger, Gary, Forkenbrock, Garrick, Elsasser, Devin, Rao, Sughosh

A Data-Driven Framework of Crash Scenario Typology Development for Child Vulnerable Road Users in the U.S

2023-01-0787

04/11/2023

Motor vehicle crashes involving child Vulnerable Road Users (VRUs) remain a critical public health concern in the United States. While previous studies successfully utilized the crash scenario typology to examine traffic crashes, these studies focus on all types of motor vehicle crashes thus the method might not apply to VRU crashes. Therefore, to better understand the context and causes of child VRU crashes on the U.S. road, this paper proposes a multi-step framework to define crash scenario typology based on the Fatality Analysis Reporting System (FARS) and the Crash Report Sampling System (CRSS). A comprehensive examination of the data elements in FARS and CRSS was first conducted to determine elements that could facilitate crash scenario identification from a systematic perspective. A follow-up context description depicts the typical behavioral, environmental, and vehicular conditions associated with an identified crash scenario. In addition, hypothesis tests are used to reveal

Guo, Huizhong, Wang, Zifei, Sherony, Rini, Bao, Shan

Items per page:

1 – 50 of 273