Lane changing is one of the main contributors to car crashes in the U.S. The complexity of the decision-making process associated with lane changing makes such maneuvers prone to driving errors, and hence, increases the possibility of car crashes. Thus, researchers have been investigating ways to model and predict lane changing maneuvers for optimally designed crash avoidance systems. Such systems rely on the accuracy of detecting the onset of lane-change maneuvers, which requires comprehensive vehicle trajectory data. Connected Vehicles (CV) data provide opportunities for accurate modeling of lane changing maneuvers, especially with the variety of advanced tools available nowadays. The review of the literature indicates that most of the implemented modeling tools do not achieve reliable accuracy for such critical safety application of lane-change prediction. Recently, eXtreme Gradient Boosting (XGB) became a well-recognized algorithm among the computer science community in solving classification problems due to its accuracy, scalability, and speed. This study implements the XGB in predicting the onset of lane changing maneuvers using CV trajectory data. The performance of XGB is compared to three other tree-based algorithms namely, decision trees, gradient boosting, and random forests. The Next Generation SIMulation trajectory data are used to represent the high-resolution CV data. The results indicate that XGB is superior to the other algorithms with a high accuracy value of 99.7%. This outstanding accuracy is achieved when considering vehicle trajectory data two seconds prior to a potential lane change maneuver. The findings of this study are promising for detection of lane change maneuvers in CV environments.
This study introduces a machine learning model for near-crash prediction from observed vehicle kinematics data. The main hypothesis is that vehicles tend to experience discernible turbulence in their kinematics shortly before involvement in near-crashes. To test this hypothesis, the SHRP2 NDS vehicle kinematics data (speed, longitudinal acceleration, lateral acceleration, yaw rate, and pedal position) are utilized. Several machine learning algorithms are trained and comparatively analyzed including K nearest neighbor (KNN), random forest, support vector machine (SVM), decision trees, Gaussian naïve Bayes (Gaussian NB), and adaptive boost (AdaBoost). Sensitivity analysis is performed to determine the optimal prediction horizon length (the time period before the occurrence of a near-crash) and the turbulence horizon length (the time period during which near-crash related changes in vehicle kinematics take place). The results indicate that optimal prediction performance can be achieved at a 1 s prediction horizons and a 3 s turbulence horizon. At these values, the AdaBoost model outperforms all other models in relation to its recall (100%), precision (98%), and F1-score (99%). These values imply that the near-crash prediction model is highly efficient in predicting most instances of near-crashes with minimal false near-crash predictions. This promising prediction performance offers a viable tool for supporting crash avoidance systems in the emerging connected/autonomous vehicle technology.
Despite the research efforts for reducing traffic accidents, the number of global annual vehicle accidents is still on the rise. This continues to motivate researchers to examine the factors contributing to crash and near-crash events (CNC). Recently, many studies attempted to identify the associated crash factors using naturalistic driving study (SHRP2-NDS) data. Despite the many classifiers developed in the literature, the high dimensionality and multicollinearity within the SHRP2-NDS data limit the accuracy and reliability of the developed models. This study develops an extreme gradient boosting (XGB) classifier, robust to multicollinearity, using the SHRP2-NDS dataset for identifying the factors contributing to CNC events. The performance of the XGB classifier is evaluated against three other advanced machine-learning algorithms. Results indicate that the XGB model outperformed the other models with a detection accuracy of 85% and identified the “driver behavior” and “intersection influence” as the most contributing factors to CNC detection.
Distracted driving behavior is a perennial safety concern that affects not only the vehicle’s occupants but other road users as well. Distraction is typically caused by engagement in secondary tasks and activities such as manipulating objects and passenger interaction, among many others. This study provides an in-depth analysis of the increased crash/near-crash risk associated with different secondary tasks using the largest real-world naturalistic driving dataset (SHRP2 Naturalistic Driving Study). Several statistical and data-mining techniques were developed to analyze the distracted driving and crash risk. First, a bivariate probit model was constructed to investigate the relationship between engagement in a secondary task and the safety-critical events likelihood. Subsequently, two different techniques were implemented to quantify the increased crash/near-crash risk because of involvement in a particular secondary task. The first technique used the baseline-category logits model to estimate the increased crash risk in terms of conditional odds ratios. The second technique used the a priori association rule mining algorithm to reveal the risk associated with each secondary task in terms of support, confidence, and lift indexes. The results indicate that reaching for objects, manipulating objects, reading, and cell phone texting are the highest crash risk factors among various secondary tasks. Recognizing the effect of different secondary tasks on traffic safety in a real-world environment helps legislators enact laws that reduce crashes resulting from distracted driving, as well as enabling government officials to make informed decisions about the allocation of available resources to reduce roadway crashes and improve traffic safety.
Lane changing is a complex decision-making process that is affected by factors such as vehicle features, driver characteristics, network attributes, and traffic conditions. Understanding the changes in driver behavior and vehicle trajectory before the lane change initiation process is essential to the design of a safe and reliable crash avoidance system. The recently introduced connected vehicle (CV) technology provides opportunities for real-time, high-resolution data exchange capability between vehicles. This study explored the high-resolution vehicle trajectory data attainable in CV environments for detecting the onset of lane change maneuvers. The observed change in behavior before the initiation of such a maneuver was examined to identify the associated driving pattern. This pattern was used to develop two lane change detection models: an artificial neural network (ANN) model and a multiple logistic regression (MLR) model. The two models were trained and tested with Next Generation Simulation data collected from a weaving freeway segment in Arlington, Virginia. The results show 80% detection accuracy for the ANN model, compared with 72% for the MLR model. The developed models identified the vehicle speed, acceleration, and speed relative to the lead vehicle as the most significant attributes for lane change detection. Drivers’ intentions could be detected early and potential crashes could be prevented by training these models to capture similar driving behavior patterns.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.