Browse Topic: Accident reconstruction
Aiming at the problem of insufficient modeling of spatio-temporal heterogeneity in road traffic accident prediction, a dual task machine learning framework integrating geographical environment, location attributes and time periodicity is proposed. The dataset used in this study was derived from traffic accident records of Nanchang during 2019–2023. Firstly, geographical identifiers are generated by rounding and aggregating latitude and longitude coordinates. At the same time, the location type is processed by a one-hot encoding, so as to carry out spatial clustering analysis of accident hotspots. Compared with the North-South pattern, the contribution of geographical features shows a strong East-West trend. The kernel density heatmap identified Zone A and zone B as dual core high-risk areas. Secondly, the sinusoidal/cosine function is used to encode the time feature circularly, which effectively captures the daily change of the accident. The quantitative analysis of random forest regression model showed that time characteristics accounted for 89.2% of the variance of accident frequency interpretation, significantly exceeding the contribution of geographical factors (10.2%) and location attributes (0.6%). After hyperparameter optimization, the accuracy of XGBoost classifier in predicting serious accidents is 75.97%, and the AUC value is 0.8412, which has strong robustness, and provides reliable support for dynamic risk assessment of traffic management system.
Letter from the Guest Editor
The timing of video recordings, along with the spatial positioning of objects, is a fundamental parameter for calculating the speed time history. If the task involves determining the average speed of an object moving at approximately constant speed, it may be acceptable to average the speed over several to a dozen frames, using the fps (frames per second) parameter as the basic time unit.. However, if the objective is to compute speed from individual frames, the reliability of the timing becomes crucial. Without access to DVR hardware documentation, proprietary algorithms, or software – and considering the frequent hardware modifications and software updates - the most effective way to solve the problem is through a reverse-engineering approach. This study discusses several aspects of timing analysis, including: (1) making a test recording of a calibrated LED lightboard; (2) analyzing the relationship between the lightboard time and the presentation time stamp (pts) extracted from the file metadata; (3) investigating frame skipping and frame timing errors due to frame rate changes; (4) modeling the composite motion of the rolling shutter and the lightboard LEDs; (5) identifying the DVR’s actual frame capture rate; and (6) compensating the timing of the evidentiary recording. Establishing the timing scheme of the test recording enables reliable speed analysis based on two or three adjacent frames of the evidentiary recording, as well as the determination of the velocity time history over a short segment of the recording.
Avoiding and mitigating any potential collision is dependent on (1) road user ability to avoid entering into a conflict (conflict avoidance effect) and (2) road user response should a conflict be entered (collision avoidance effect). This study examined the collision avoidance effect of the Waymo Driver, a currently deployed SAE level 4 automated driving system (ADS), using a human behavior reference model, designed to be representative of a human driver that is non-impaired, with eyes on the conflict (NIEON). Reliable performance benchmarking methodologies for assessing ADS performance are an essential component of determining system readiness. This consistently performing, always-attentive driver does not exist in the human population. Counterfactual simulations were run on responder collision scenarios based on reconstructions from a 10-year period of human fatal crashes from the Operational Design Domain of the Waymo ADS in Chandler, Arizona. Of 16 simulated conflicts entered, 12 (75%) were prevented by the Waymo Driver, and 10 (62.5%) were prevented by the NIEON model. The NIEON Model mitigated an additional 5 collisions and did not mitigate 1 collision. In these 16 conflicts entered, 93% of serious injury risk was reduced by the Waymo Driver, whereas 84% of serious injury risk was reduced by the NIEON model. Further, in a case-by-case evaluation, the Waymo Driver’s collision avoidance led to reduced serious injury risk when compared to the NIEON model in every simulated event. The results of this paper demonstrate that a reference model like NIEON can be used to benchmark ADS responder performance in response to high-risk initiating behaviors performed by the current driving population.
Traditionally, occupant safety research has centered on passive safety systems such as seatbelts, airbags, and energy-absorbing vehicle structures, all designed under the assumption of a nominal occupant posture at the moment of impact. However, with increasing deployment of active safety technologies such as Forward Collision Warning (FCW) and Autonomous Emergency Braking (AEB), vehicle occupants are exposed to pre-crash decelerations that alter their seated position before the crash. Although AEB mitigates the crash severity, the induced occupant movement leads to out-of-position behavior (OOP), compromising the available survival space phase and effectiveness of passive restraint systems during the crash. Despite these evolving real-world conditions, global regulatory bodies and NCAP programs continue to evaluate pre-crash and crash phases independently, with limited integration. Moreover, traditional Anthropomorphic Test Devices (ATDs) such as Hybrid III dummies, although highly repeatable, lack the bio-fidelity necessary to capture human-like kinematics during pre-crash braking events involving low g. ATDs do not simulate the spinal articulation, posture adjustments and active muscle contraction that occur during emergency maneuvers or pre-crash scenarios. To overcome these limitations, researchers have increasingly turned to Human Body Models (HBMs) such as Total Human Model for Safety (THUMS) and Global Human Body Model Consortium (GHBMC). These models enable high-fidelity finite element (FE) simulations with anatomical realism, allowing for the inclusion of active musculature and posture changes. This study aims to quantify the occupant forward excursion under pre-crash phase (due to AEB) and explore the possibility of an integrated simulation framework that evaluates occupant safety across both pre-crash and crash events. For this, the approach was to carry out full vehicle braking tests (1g braking pulse) with adult male (AM50) volunteers at different speeds to measure forward head excursion during pre-crash. These scenarios were replicated in LS-Dyna using THUMS HBM, showing strong agreement with experimental data. The resulting excursed postures were then used in crash simulations with ATDs to evaluate the effect on injury outcomes. Overall, the findings demonstrate effect of forward excursion on occupant injuries and the effectiveness of HBMs in capturing occupant kinematics, during pre-crash events.
This study provides a comprehensive framework for establishing land use compatibility guidelines specific to vertiports serving electric Vertical Takeoff and Landing (eVTOL) aircraft within urban settings. Recognizing critical gaps in current regulatory standards, the research systematically integrates analyses of accident risk, noise propagation, and aerodynamic impacts—including downwash and outwash—to delineate compatibility zones around vertiports. Employing an artificial intelligence (AI) augmented system, the study conducted safety and hazard assessments, various quantitative analyses, and simulations to identify spatial constructs of operational risks and environmental impacts. Results indicated significant discrepancies between existing aviation infrastructure guidelines and the unique operational characteristics of eVTOLs, necessitating revised zoning parameters. The proposed multi-tiered safety zoning framework provides precise, evidencebased recommendations for urban planners, enhancing safety, minimizing noise impacts, and ensuring environmental sustainability. Ultimately, this research offers policymakers and stakeholders practical tools for integrating advanced air mobility infrastructure safely into complex urban landscapes.
Theory and principles of occupant protection for automobiles in rear-end collisions have experienced significant evolution over the decades. Performance of the seatback, specifically the stiffness of the structure, during such a collision has been a subject of particular interest and debate among design engineers, accident reconstruction experts, critics, etc. The majority of current seat designs rely on plastic deformation of the seatback structure to protect the occupant from the dynamics of the crash. In attempt to highlight and provide background information for understanding this subject, this work highlights significant events, research, and publications over the past five decades to illustrate how this subject, automobile design, government regulation and public opinion has evolved. It is observed that technology and design for improving rear-impact protection has received less attention than collisions of other principal directions of force. The different types of Anthropometric Test Devices (ATDs) used in research are quantified. Lessons may be learned by design engineers to ensure seatback safety moves forward with progress to improve future generations of automobile seats. It is also observed that the quantity of published works could be influenced by historical events as well as certain research topics trending in the automotive industry.
Items per page:
50
1 – 50 of 869