Lessons learned from pedestrian-driver communication and yielding patterns

Exploring Factors Associated With Crossing Assertiveness of Pedestrians at Unsignalized Intersections

“…The specialists assessed pedestrian assertiveness at seven levels: extremely passive, passive, somewhat passive, neutral, somewhat assertive, assertive, and extremely assertive. Referring to Kong et al ( 9 ), seven assertiveness levels were aggregated into three levels, because some assertiveness levels in seven categories have insufficient data to mine meaningful rules from the original dataset. For example, there are only 156 records at an extremely assertive level.…”

“…Younger drivers ( 21 ) and people who drive more expensive vehicles ( 22 ) were less likely to yield the crosswalk to other road users. Besides, eye contact and facial expression could help communication between drivers and pedestrians and increase pedestrian crossing assertiveness ( 9 , 23 ).…”

“…Previous studies mainly focused on traditional statistical methods, such as logistic regression ( 14 , 15 , 18 ) or hypothesis testing ( 16 , 25 ). However, machine learning (ML) algorithms, including but not limited to the pattern recognition method ( 9 ), association rules ( 34 , 35 ), decision tree ( 36 ), XGBoost ( 37 , 38 ), and convolutional neural network ( 39 , 40 ), have achieved excellent performance in traffic safety issues, especially in pedestrian safety-related studies. Advanced association rule mining has effectively uncovered the latent trends or combinations of attributes that frequently happened together.…”

“…In contrast, successful communication could mitigate crashes, especially at USIs and uncontrolled crosswalks ( 8 ). Meanwhile, the crossing assertiveness of pedestrians significantly affects interpersonal communication and driver-yielding behavior ( 9 ). Some pedestrians cross the roadway or intersection assertively because of their nature of confidence, impulsion, or aggression ( 10 ).…”

Smiles and Angry Faces vs. Nods and Head Shakes: Facial Expressions at the Service of Autonomous Vehicles

“…It will not be long, however, before drivers of autonomous vehicles (AVs) will be unable to communicate their intention, as they will be engaged in activities unrelated to driving, such as reading, typing, or watching a movie [5,6]. This situation may prove difficult for pedestrians especially, because when deciding whether to cross the street or not, they take into consideration information provided by both vehicle kinematics, such as speed and acceleration [7][8][9], and the driver of an approaching vehicle, such as gaze direction and facial expression, according to a plethora of research conducted in different cultural contexts, such as France, China, the Czech Republic, Greece, the Netherlands, and the UK [4,[10][11][12][13][14][15][16][17][18][19]. To make matters worse, there is ample evidence for the underestimation of vehicle speed on the part of pedestrians, as well as the overestimation of the time at their disposal to attempt crossing the street safely [20][21][22].…”