Study of typical electric two‐wheelers pre-crash scenarios using K-medoids clustering methodology based on video recordings in China

Pan, Di; Han, Yong; Jin, Qianqian; Wu, He; Huang, Hongwu

doi:10.1016/j.aap.2021.106320

Cited by 33 publications

(14 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To validate the collision probability model, three car-to-cyclist accident cases were selected from the VRU Traffic Accident Database with Video ( 30 , 31 ) to investigate the collision probability. The database aims to obtain more detailed and objective precrash, crash, and postcrash accident information by collecting in-depth accident data with videos as well as public internet videos.…”

Section: Methodsmentioning

confidence: 99%

Probabilistic Prediction of Collisions between Cyclists and Vehicles Based on Uncertainty of Cyclists’ Movements

Pan

Han

Jin

et al. 2022

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

The uncertainty of cyclists’ movements has a significant impact on predicting the risk of collisions between cyclists and vehicles. The purpose of this study was to provide a method for assessing collision risk using probability, taking into account the uncertainty of cyclists’ movements. A cyclist model was first developed using a first-order Markov model. Then, based on Monte Carlo sampling, the distribution characteristics of the minimum distance and the time-to-collision (TTC) between the vehicle and the cyclist were extracted. By fitting these features, the probability density functions of the collision distance and TTC were estimated to derive the collision probabilities. The effectiveness of the collision probability prediction model was benchmarked against a deterministic crash risk prediction model (autonomous emergency braking [AEB] system) applied to three real-world cases previously reconstructed in an in-depth crash database. The results show that the collision probability prediction model can effectively predict the risk of collisions between cyclists and vehicles with better accuracy than AEB systems using a fixed trigger threshold. This study is a valuable reference for the development of advanced vehicle collision avoidance systems to protect cyclists and other vulnerable road users.

show abstract

Section: Methodsmentioning

confidence: 99%

Probabilistic Prediction of Collisions between Cyclists and Vehicles Based on Uncertainty of Cyclists’ Movements

Pan

Han

Jin

et al. 2022

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

show abstract

“…database 44 with detailed head injury information regarding common head injury types such as SF, SDH, SAH, Contusion, and DAI, as shown in Table 1. The occurrence of SDH injury is often accompanied by SAH injury 36 .…”

Section: The Developed and Validated Wicmentioning

confidence: 99%

Establishment of a weighted injury criterion for evaluation of multiple types of head severe injuries for vulnerable road users

Han

Yang

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

In traffic accidents, multiple head injury types often occur simultaneously and cause severe brain injury for vulnerable road users (VRUs). In this study, a head-weighted injury criterion was developed to assess the risk of severe brain injury considering various injury types. Firstly, 50 in-depth accidents were reconstructed using a high-precision reconstruction method to reconstruct the overall kinematic response and head injury severity of VRUs and analyze the correlation between various head injury criteria and severe brain injury. Then, four injury criteria were selected that correlated well with severe brain injury, namely HIC15, angular acceleration, coup pressure, and maximum principal strain (MPS). Finally, weighted head injury criteria (WIC4) of severe injuries were established based on the four selected injury criteria, and the correlation between WIC4 and severe brain injury was validated based on the area under of receiver operating characteristic curve (AUROC) and the reconstructed results of another 10 selected accidents. The results showed that WIC4 had a good predictive capability for both severe and non-severe brain injury cases, and the AUROC was 0.983, which was significantly higher than that of the single head injury criterion. This study further improved the correlation between head injury criteria and severe brain injury.

show abstract

“…The AsPeCSS (Assessment methodologies for forward looking integrated pedestrian and further extension to cyclist safety systems) project applied a similar approach as CATS and generated four scenarios for testing ( 35 ). Clustering methods such as K-means, K-medoids and hierarchical clustering ( 29 , 36 , 37 ) have also been widely used to develop car-to-TW test scenarios. Despite different clustering methods, the essential variables that play decisive roles are consistent: motion of the car, motion of the TW, and the relative motion direction between the car and TW.…”

Section: Literature Reviewmentioning

confidence: 99%

Evaluation of Automatic Emergency Braking Systems in Two-Wheeler Crash Scenarios

Zhou

Wang

2023

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

The two-wheeler (TW) is a popular means of transportation in China, but TWs often suffer from serious traffic crashes because of their highly flexible trajectory and low detectability. Therefore, they present a challenge for the sensing and decision-making systems in autonomous vehicles (AVs). Collision avoidance systems, such as automatic emergency braking (AEB), have provided an effective way for AVs to avoid collisions with different objects, including TWs. The effectiveness of the AEB system is highly dependent on its parameter configurations, however, which vary among TW crash scenarios. This study, therefore, evaluates the AEB parameters, including time to collision (TTC), deceleration, and detection area (including detection range, field of vision, and trigger width) in two AEB systems: one-stage AEB and three-stage AEB. A total of 243 crashes extracted from the China In-Depth Accident Study database were simulated in Matlab’s Simulink. Results show: (i) one-stage AEB crash avoidance rates range from 15.2% to 81.5%, while three-stage AEB has crash avoidance rates as high as 87.2%; (ii) deceleration, TTC, and detection area all have significant main effects on crash rate, but detection area has less influence in longitudinal than in crossing scenarios; (iii) higher crash avoidance rates resulted in lower traffic efficiency for both AEB systems, but resulted in greater speed reduction only for one-stage AEB; and (iv) collisions are less likely to be avoided in scenarios with high initial speed of TW and AV. This study demonstrates the performance of AEB algorithms in multiple actual crash scenarios and provides a reliable basis for the development of AEB systems.

show abstract

Study of typical electric two‐wheelers pre-crash scenarios using K-medoids clustering methodology based on video recordings in China

Cited by 33 publications

References 26 publications

Probabilistic Prediction of Collisions between Cyclists and Vehicles Based on Uncertainty of Cyclists’ Movements

Probabilistic Prediction of Collisions between Cyclists and Vehicles Based on Uncertainty of Cyclists’ Movements

Establishment of a weighted injury criterion for evaluation of multiple types of head severe injuries for vulnerable road users

Evaluation of Automatic Emergency Braking Systems in Two-Wheeler Crash Scenarios

Contact Info

Product

Resources

About