A Review of Intelligent Driving Style Analysis Systems and Related Artificial Intelligence Algorithms

Analysis of harsh braking and harsh acceleration occurrence via explainable imbalanced machine learning using high-resolution smartphone telematics and traffic data

Harsh driving events such as harsh brakings (HBs) and harsh accelerations (HAs) are promising Surrogate Safety Measures, already extensively utilised in road safety research. However, their occurrence relative to normal driving conditions has not been the explicit target of research, as they are typically used as inputs for crash prediction. The present study addresses this research gap by investigating factors influencing HB and HA occurrence using real-time naturalistic driving telematics data recorded from smartphones, traffic data and road geometry & network characteristics data. These multisource data were matched in order to capture the specific circumstances under which HBs and HAs occur. The utilized telematics dataset included trips from 314 anonymous drivers in an urban arterial of Athens, Greece. Subsequently, Synthetic Minority Oversampling TEchnique (SMOTE) was applied due to class imbalance and then binary classification was conducted to detect factors leading to HB and HA occurrence. Imbalanced Machine Learning (ML) XGBoost algorithms predicted over 75% of HBs and over 84% of HAs for the test dataset, indicating suitability for real-time monitoring. The algorithms were also augmented with SHapley Additive exPlanation (SHAP) values, aiming to increase outcome explainability. Results reveal strong nonlinear effects on harsh event occurrence, with individual speed and traffic flow parameters showing the highest influence, followed by exposure parameters such as segment length and pass count. Network characteristics such as number of lanes, and speed limit had limited influence on HA and HB occurrence, as did behaviors such as mobile phone engagement and speeding.

Monitoring Distracted Driving Behaviours with Smartphones: An Extended Systematic Literature Review

Driver behaviour monitoring is a broad area of research, with a variety of methods and approaches. Distraction from the use of electronic devices, such as smartphones for texting or talking on the phone, is one of the leading causes of vehicle accidents. With the increasing number of sensors available in vehicles, there is an abundance of data available to monitor driver behaviour, but it has only been available to vehicle manufacturers and, to a limited extent, through proprietary solutions. Recently, research and practice have shifted the paradigm to the use of smartphones for driver monitoring and have fuelled efforts to support driving safety. This systematic review paper extends a preliminary, previously carried out author-centric literature review on smartphone-based driver monitoring approaches using snowballing search methods to illustrate the opportunities in using smartphones for driver distraction detection. Specifically, the paper reviews smartphone-based approaches to distracted driving behaviour detection, the smartphone sensors and detection methods applied, and the results obtained.

An Explainable Student Fatigue Monitoring Module with Joint Facial Representation

Online fatigue estimation is, inevitably, in demand as fatigue can impair the health of college students and lower the quality of higher education. Therefore, it is essential to monitor college students' fatigue to diminish its adverse effects on the health and academic performance of college students. However, former studies on student fatigue monitoring are mainly survey-based with offline analysis, instead of using constant fatigue monitoring. Hence, we proposed an explainable student fatigue estimation model based on joint facial representation. This model includes two modules: a spacial-temporal symptom classification module and a data-experience joint status inferring module. The first module tracks a student's face and generates spatial-temporal features using a deep convolutional neural network (CNN) for the relevant drivers of abnormal symptom classification; the second module infers a student's status with symptom classification results with maximum a posteriori (MAP) under the data-experience joint constraints. The model was trained on the benchmark NTHU Driver Drowsiness Detection (NTHU-DDD) dataset and tested on an Online Student Fatigue Monitoring (OSFM) dataset. Our method outperformed the other methods with an accuracy rate of 94.47% under the same training-testing setting. The results were significant for real-time monitoring of students' fatigue states during online classes and could also provide practical strategies for in-person education.

Driving behavior classification at signalized intersections using vehicle kinematics: Application of unsupervised machine learning

Driving behavior is considered as a unique driving habit of each driver and has a significant impact on road safety. This study proposed a novel data-driven Machine Learning framework that can classify driving behavior at signalized intersections considering two different signal conditions. To the best of our knowledge, this is the first study that investigates driving behavior at signalized intersections with two different conditions that are mostly used in practice, i.e., the control setting with the signal order of green-yellow-red and a flashing green setting with the signal order of green-flashing green-yellow-red. A driving simulator dataset collected from participants at Qatar University's Qatar Transportation and Traffic Safety Center, driving through multiple signalized intersections, was used. The proposed framework extracts volatility measures from vehicle kinematic parameters including longitudinal speed and acceleration. K-means clustering algorithm with elbow method was used as an unsupervised machine learning to cluster driving behavior into three classes (i.e., conservative, normal, and aggressive) and investigate the impact of signal conditions. The framework confirmed that in general driving behavior at a signalized intersection reflects drivers' habits and personality rather than the signal condition, still, it manifests the intersection nature that usually requires drivers to be more vigilant and cautious. Nonetheless, the results suggested that flashing green condition could make drivers more conservative, which could be due to the limited capabilities of human to estimate the remaining distance and the prolonged duration of the additional flashing green interval. The proposed framework and findings of the study were promising that can be used for clustering drivers into different styles for different conditions and might be beneficial for policymakers, researchers, and engineers.

Emergency Information Communication Structure by Using Multimodel Fusion and Artificial Intelligence Algorithm

With the development of The Times, social events are increasing, and emergency management has gradually become the main helper to solve the crisis in the public domain. By observing the current situation of many countries and regions, we can find that various types of public crises often occur in many countries and regions in the world, which have severely affected people's daily life, lives, and property. Through long-term research and analysis, it can be known that the emergency management mechanism currently established in China has certain shortcomings. The communication problem of emergency information is likely to cause the emergency work to not proceed smoothly. In addition, problems in the communication channels of emergency information are likely to cause problems in the cooperation of various departments when people carry out emergency management work, and the efficiency of the government in dealing with problems will also be reduced in real scenarios. In order to improve the efficiency of emergency information management, this paper aims at the various problems existing and facing in the construction of emergency management system. On this basis, the integration of various relevant emergency information management plan models is analyzed and sorted out, and based on the research and integration of the development of artificial intelligence algorithms. The main research results of emergency information management at home and abroad are comprehensively studied and evaluated. Finally, a QG algorithm based on more model fusion is developed. In the process of analysis, this article uses artificial intelligence algorithms to build a prediction model of multiple modes and collects the data needed to build the model by random extraction. Through the analysis of different data sets, it is used as the basic training data for prediction. Through comprehensive analysis, the model constructed in this paper can promote the sharing of emergency information among departments to a certain extent.

A Fuzzy-Logic Approach Based on Driver Decision-Making Behavior Modeling and Simulation

The present study proposes a decision-making model based on different models of driver behavior, aiming to ensure integration between road safety and crash reduction based on an examination of speed limitations under weather conditions. The present study investigated differences in road safety attitude, driver behavior, and weather conditions I-69 in Flint, Genesee County, Michigan, using the fuzzy logic approach. A questionnaire-based survey was conducted among a sample of Singaporean (n = 100) professional drivers. Safety level was assessed in relation to speed limits to determine whether the proposed speed limit contributed to a risky or safe situation. The experimental results show that the speed limits investigated on different roads/in different weather were based on the participants’ responses. The participants could increase or keep their current speed limit or reduce their speed limit a little or significantly. The study results were used to determine the speed limits needed on different roads/in different weather to reduce the number of crashes and to implement safe driving conditions based on the weather. Changing the speed limit from 80 mph to 70 mph reduced the number of crashes occurring under wet road conditions. According to the results of the fuzzy logic study algorithm, a driver’s emotions can predict outputs. For this study, the fuzzy logic algorithm evaluated drivers’ emotions according to the relation between the weather/road condition and the speed limit. The fuzzy logic would contribute to assessing a powerful feature of human control. The fuzzy logic algorithm can explain smooth relationships between the input and output. The input–output relationship estimated by fuzzy logic was used to understand differences in drivers’ feelings in varying road/weather conditions at different speed limits.

A hybrid deep learning approach for driver anomalous lane changing identification

Reliable knowledge of driving states is of great importance to ensure road safety. Anomaly detection in driving behavior means recognizing anomalous driving states as a direct result of either environmental or psychological factors. This paper provides an efficient anomaly recognition approach to identify anomalous lane-changing events in a personalized manner. The proposed framework includes three unsupervised algorithms. First, a Recurrent-Convolutional Autoencoder extracts the spatio-temporal characteristics from a high-dimensional naturalistic driving dataset. Second, in order to recognize anomalous lane-changing events of individual drivers, the extracted latent feature space is analyzed using Pauta criterion-based reconstruction loss analysis, as well as one-class Support Vector Machine. Last, t-Distributed Stochastic Neighbor Embedding is employed to visualize the latent space for better understanding and interpretability. Temporal anomalies of lane-changing events were analyzed by a personalized grey relational coefficient analysis, to represent robust similarities for individual drivers. Validation and calibration were performed with a natural driving study dataset collected from 50 drivers with 59,372 lane change events. The results showed heterogeneity in the pattern of abnormal lane changing behavior across the sample. At the same time, each driver exhibited heterogeneous anomalous behaviors in both temporal and spatial sequences. Without prior labels, the proposed model effectively captures personalized driving patterns and abnormal lane-changing events from high-dimensional time-series data. This unsupervised hybrid approach is a novel attempt to complete personalized anomalous lane-changing behaviors identification based on naturalistic driving data involving various traffic environments. Our approach enables the extraction of natural individual lane-changing behavior patterns and provides insights for the improvement of personalized driving behavior monitoring systems.

Evaluation of 1D and 2D Deep Convolutional Neural Networks for Driving Event Recognition

Driving event detection and driver behavior recognition have been widely explored for many purposes, including detecting distractions, classifying driver actions, detecting kidnappings, pricing vehicle insurance, evaluating eco-driving, and managing shared and leased vehicles. Some systems can recognize the main driving events (e.g., accelerating, braking, and turning) by using in-vehicle devices, such as inertial measurement unit (IMU) sensors. In general, feature extraction is a commonly used technique to obtain robust and meaningful information from the sensor signals to guarantee the effectiveness of the subsequent classification algorithm. However, a general assessment of deep neural networks merits further investigation, particularly regarding end-to-end models based on Convolutional Neural Networks (CNNs), which combine two components, namely feature extraction and the classification parts. This paper primarily explores supervised deep-learning models based on 1D and 2D CNNs to classify driving events from the signals of linear acceleration and angular velocity obtained with the IMU sensors of a smartphone placed in the instrument panel of the vehicle. Aggressive and non-aggressive behaviors can be recognized by monitoring driving events, such as accelerating, braking, lane changing, and turning. The experimental results obtained are promising since the best classification model achieved accuracy values of up to 82.40%, and macro- and micro-average F1 scores, respectively, equal to 75.36% and 82.40%, thus, demonstrating high performance in the classification of driving events.

Fuzzy System to Assess Dangerous Driving: A Multidisciplinary Approach

Dangerous driving can cause accidents, injuries and loss of life. An efficient assessment helps to identify the absence or degree of dangerous driving to take the appropriate decisions while driving. Previous studies assess dangerous driving through two approaches: (i) using electronic devices or sensors that provide objective variables (acceleration, turns and speed), and (ii) analyzing responses to questionnaires from behavioral science that provide subjective variables (driving thoughts, opinions and perceptions from the driver). However, we believe that a holistic and more realistic assessment requires a combination of both types of variables. Therefore, we propose a three-phase fuzzy system with a multidisciplinary (computer science and behavioral sciences) approach that draws on the strengths of sensors embedded in smartphones and questionnaires to evaluate driver behavior and social desirability. Our proposal combines objective and subjective variables while mitigating the weaknesses of the disciplines used (sensor reading errors and lack of honesty from respondents, respectively). The methods used are of proven reliability in each discipline, and their outputs feed a combined fuzzy system used to handle the vagueness of the input variables, obtaining a personalized result for each driver. The results obtained using the proposed system in a real scenario were efficient at 84.21%, and were validated with mobility experts' opinions. The presented fuzzy system can support intelligent transportation systems, driving safety, or personnel selection.

A Machine Learning Framework for Automated Accident Detection Based on Multimodal Sensors in Cars

Identifying accident patterns is one of the most vital research foci of driving analysis. Environmental or safety applications and the growing area of fleet management all benefit from accident detection contributions by minimizing the risk vehicles and drivers are subject to, improving their service and reducing overhead costs. Some solutions have been proposed in the past literature for automated accident detection that are mainly based on traffic data or external sensors. However, traffic data can be difficult to access, while external sensors can end up being difficult to set up and unreliable, depending on how they are used. Additionally, the scarcity of accident detection data has limited the type of approaches used in the past, leaving in particular, machine learning (ML) relatively unexplored. Thus, in this paper, we propose a ML framework for automated car accident detection based on mutimodal in-car sensors. Our work is a unique and innovative study on detecting real-world driving accidents by applying state-of-the-art feature extraction methods using basic sensors in cars. In total, five different feature extraction approaches, including techniques based on feature engineering and feature learning with deep learning are evaluated on the strategic highway research program (SHRP2) naturalistic driving study (NDS) crash data set. The main observations of this study are as follows: (1) CNN features with a SVM classifier obtain very promising results, outperforming all other tested approaches. (2) Feature engineering and feature learning approaches were finding different best performing features. Therefore, our fusion experiment indicates that these two feature sets can be efficiently combined. (3) Unsupervised feature extraction remarkably achieves a notable performance score.

A Comprehensive Review on the Behaviour of Motorcyclists: Motivations, Issues, Challenges, Substantial Analysis and Recommendations

With the continuous emergence of new technologies and the adaptation of smart systems in transportation, motorcyclist driving behaviour plays an important role in the transition towards intelligent transportation systems (ITS). Studying motorcyclist driving behaviour requires accurate models with accurate and complete datasets for better road safety and traffic management. As accuracy is needed in modelling, motorcyclist driving behaviour analyses can be performed using sensors that collect driving behaviour characteristics during real-time experiments. This review article systematically investigates the literature on motorcyclist driving behaviour to present many findings related to the issues, problems, challenges, and research gaps that have existed over the last 10 years (2011-2021). A number of digital databases (i.e., IEEE Xplore®, ScienceDirect, Scopus, and Web of Science) were searched and explored to collect reliable peer-reviewed articles. Out of the 2214 collected articles, only 174 articles formed the final set of articles used in the analysis of the motorcyclist research area. The filtration process consisted of two stages that were implemented on the collected articles. Inclusion criteria were the core of the first stage of the filtration process keeping articles only if they were a study or review written in English or were articles that mainly incorporated the driving style of motorcyclists. The second phase of the filtration process is based on more rules for article inclusion. The criteria of inclusion for the second phase of filtration examined the deployment of motorcyclist driver behaviour characterisation procedures using a real-time-based data acquisition system (DAS) or a questionnaire. The final number of articles was divided into three main groups: reviews (7/174), experimental studies (41/174), and social studies-based articles (126/174). This taxonomy of the literature was developed to group the literature into articles with similar types of experimental conditions. Recommendation topics are also presented to enable and enhance the pace of the development in this research area. Research gaps are presented by implementing a substantial analysis of the previously proposed methodologies. The analysis mainly identified the gaps in the development of data acquisition systems, model accuracy, and data types incorporated in the proposed models. Finally, research directions towards ITS are provided by exploring key topics necessary in the advancement of this research area.

Judging One's Own or Another Person's Responsibility in Interactions With Automation

OBJECTIVE

We explore users' and observers' subjective assessments of human and automation capabilities and human causal responsibility for outcomes.

BACKGROUND

In intelligent systems and advanced automation, human responsibility for outcomes becomes equivocal, as do subjective perceptions of responsibility. In particular, actors who actively work with a system may perceive responsibility differently from observers.

METHOD

In a laboratory experiment with pairs of participants, one participant (the "actor") performed a decision task, aided by an automated system, and the other (the "observer") passively observed the actor. We compared the perceptions of responsibility between the two roles when interacting with two systems with different capabilities.

RESULTS

Actors' behavior matched the theoretical predictions, and actors and observers assessed the system and human capabilities and the comparative human responsibility similarly. However, actors tended to relate adverse outcomes more to system characteristics than to their own limitations, whereas the observers insufficiently considered system capabilities when evaluating the actors' comparative responsibility.

CONCLUSION

When intelligent systems greatly exceed human capabilities, users may correctly feel they contribute little to system performance. They may interfere more than necessary, impairing the overall performance. Outside observers, such as managers, may overweigh users' contribution to outcomes, holding users responsible for adverse outcomes when they rightly trusted the system.

APPLICATION

Presenting users of intelligent systems and others with performance measures and the comparative human responsibility may help them calibrate subjective assessments of performance, reducing users' and outside observers' biases and attribution errors.

Driver Behavior Profiling and Recognition Using Deep-Learning Methods: In Accordance with Traffic Regulations and Experts Guidelines

The process of collecting driving data and using a computational model to generate a safety score for the driver is known as driver behavior profiling. Existing driver profiles attempt to categorize drivers as either safe or aggressive, which some experts say is not practical. This is due to the "safe/aggressive" categorization being a state that describes a driver's conduct at a specific point in time rather than a continuous state or a human trait. Furthermore, due to the disparity in traffic laws and regulations between countries, what is considered aggressive behavior in one place may differ from what is considered aggressive behavior in another. As a result, adopting existing profiles is not ideal. The authors provide a unique approach to driver behavior profiling based on timeframe data segmentation. The profiling procedure consists of two main parts: row labeling and segment labeling. Row labeling assigns a safety score to each second of driving data based on criteria developed with the help of Malaysian traffic safety experts. Then, rows are accumulated to form timeframe segments. In segment labeling, generated timeframe segments are assigned a safety score using a set of criteria. The score assigned to the generated timeframe segment reflects the driver's behavior during that time period. Following that, the study adopts three deep-learning-based algorithms, namely, Deep Neural Network (DNN), Recurrent Neural Network (RNN), and Convolutional Neural Network (CNN), to classify recorded driving data according to the established profiling procedure, and selects the most suitable one for a proposed recognition system. Various techniques were used to prevent the classification algorithms from overfitting. Using gathered naturalistic data, the validity of the modulated algorithms was assessed on various timeframe segments ranging from 1 to 10 s. Results showed that the CNN, which achieved an accuracy of 96.1%, outperformed the other two classification algorithms and was therefore recommended for the recognition system. In addition, recommendations were outlined on how the recognition system would assist in improving traffic safety.

A Bibliometric Analysis and Benchmark of Machine Learning and AutoML in Crash Severity Prediction: The Case Study of Three Colombian Cities

Traffic accidents are of worldwide concern, as they are one of the leading causes of death globally. One policy designed to cope with them is the design and deployment of road safety systems. These aim to predict crashes based on historical records, provided by new Internet of Things (IoT) technologies, to enhance traffic flow management and promote safer roads. Increasing data availability has helped machine learning (ML) to address the prediction of crashes and their severity. The literature reports numerous contributions regarding survey papers, experimental comparisons of various techniques, and the design of new methods at the point where crash severity prediction (CSP) and ML converge. Despite such progress, and as far as we know, there are no comprehensive research articles that theoretically and practically approach the model selection problem (MSP) in CSP. Thus, this paper introduces a bibliometric analysis and experimental benchmark of ML and automated machine learning (AutoML) as a suitable approach to automatically address the MSP in CSP. Firstly, 2318 bibliographic references were consulted to identify relevant authors, trending topics, keywords evolution, and the most common ML methods used in related-case studies, which revealed an opportunity for the use AutoML in the transportation field. Then, we compared AutoML (AutoGluon, Auto-sklearn, TPOT) and ML (CatBoost, Decision Tree, Extra Trees, Gradient Boosting, Gaussian Naive Bayes, Light Gradient Boosting Machine, Random Forest) methods in three case studies using open data portals belonging to the cities of Medellín, Bogotá, and Bucaramanga in Colombia. Our experimentation reveals that AutoGluon and CatBoost are competitive and robust ML approaches to deal with various CSP problems. In addition, we concluded that general-purpose AutoML effectively supports the MSP in CSP without developing domain-focused AutoML methods for this supervised learning problem. Finally, based on the results obtained, we introduce challenges and research opportunities that the community should explore to enhance the contributions that ML and AutoML can bring to CSP and other transportation areas.

Development of a machine-learning-based decision support mechanism for predicting chemical tanker cleaning activity

Purpose

This paper aims to propose a machine learning-based automatic labeling methodology for chemical tanker activities that can be applied to any port with any number of active tankers and the identification of important predictors. The methodology can be applied to any type of activity tracking that is based on automatically generated geospatial data.

Design/methodology/approach

The proposed methodology uses three machine learning algorithms (artificial neural networks, support vector machines (SVMs) and random forest) along with information fusion (IF)-based sensitivity analysis to classify chemical tanker activities. The data set is split into training and test data based on vessels, with two vessels in the training data and one in the test data set. Important predictors were identified using a receiver operating characteristic comparative approach, and overall variable importance was calculated using IF from the top models.

Findings

Results show that an SVM model has the best balance between sensitivity and specificity, at 93.5% and 91.4%, respectively. Speed, acceleration and change in the course on the ground for the vessels are identified as the most important predictors for classifying vessel activity.

Research limitations/implications

The study evaluates the vessel movements waiting between different terminals in the same port, but not their movements between different ports for their tank-cleaning activities.

Practical implications

The findings in this study can be used by port authorities, shipping companies, vessel operators and other stakeholders for decision support, performance tracking, as well as for automated alerts.

Originality/value

This analysis makes original contributions to the existing literature by defining and demonstrating a methodology that can automatically label vehicle activity based on location data and identify certain characteristics of the activity by finding important location-based predictors that effectively classify the activity status.

Driving style recognition and comparisons among driving tasks based on driver behavior in the online car-hailing industry

As a product of the shared economy, online car-hailing platforms can be used effectively to help maximize resources and alleviate traffic congestion. The driver's behavior is characterized by his or her driving style and plays an important role in traffic safety. This paper proposes a novel framework to classify driving styles (defined as aggressive, normal, and cautious) based on online car-hailing data to investigate the distinct characteristics of drivers when performing various driving tasks (defined as cruising, ride requests, and drop-off) and undergoing certain maneuvers (defined as turning, acceleration, and deceleration). The proposed model is constructed based on the detection and classification of driving maneuvers using a threshold-based endpoint detection approach, principal component analysis, and k-means clustering. The driving styles that the driver exhibits for the different driving tasks are compared and analyzed based on the classified maneuvers. The empirical results for Nanjing, China demonstrate that the proposed framework can detect driving maneuvers and classify driving styles accurately. Moreover, according to this framework, driving tasks lead to variations in driving style, and the variations in driving style during the different driving tasks differ significantly for turning, acceleration, and deceleration maneuvers.

Beside and Behind the Wheel: Factors that Influence Driving Stress and Driving Behavior

A large percentage of traffic accidents are due to human errors. Driving behavior and driving stress influence the probability of making these mistakes. Both are influenced by multiple factors, among which might be elements such as age, gender, sleeping hours, or working hours. The objective of this paper is to study, in a real scenario and without forcing the driver’s state, the relationship between driving behavior, driving stress, and these elements. Furthermore, we aim to provide guidelines to improve driving assistants. In this study, we used 1050 driving samples obtained from 35 volunteers. The driving samples correspond to regular commutes from home to the workplace. ANOVA and ANCOVA tests were carried out to check if there are significant differences in the four factors analyzed. Although the results show that driving behavior and driving stress are affected by gender, age, and sleeping hours, the most critical variable is working hours. Drivers with long working days suffer significantly more driving stress compared to other drivers, with the corresponding effect on their driving style. These drivers were the worst at maintaining the safety distance.

Theoretical, Measured, and Subjective Responsibility in Aided Decision Making

When humans interact with intelligent systems, their causal responsibility for outcomes becomes equivocal. We analyze the descriptive abilities of a newly developed responsibility quantification model (ResQu) to predict actual human responsibility and perceptions of responsibility in the interaction with intelligent systems. In two laboratory experiments, participants performed a classification task. They were aided by classification systems with different capabilities. We compared the predicted theoretical responsibility values to the actual measured responsibility participants took on and to their subjective rankings of responsibility. The model predictions were strongly correlated with both measured and subjective responsibility. Participants’ behavior with each system was influenced by the system and human capabilities, but also by the subjective perceptions of these capabilities and the perception of the participant's own contribution. A bias existed only when participants with poor classification capabilities relied less than optimally on a system that had superior classification capabilities and assumed higher-than-optimal responsibility. The study implies that when humans interact with advanced intelligent systems, with capabilities that greatly exceed their own, their comparative causal responsibility will be small, even if formally the human is assigned major roles. Simply putting a human into the loop does not ensure that the human will meaningfully contribute to the outcomes. The results demonstrate the descriptive value of the ResQu model to predict behavior and perceptions of responsibility by considering the characteristics of the human, the intelligent system, the environment, and some systematic behavioral biases. The ResQu model is a new quantitative method that can be used in system design and can guide policy and legal decisions regarding human responsibility in events involving intelligent systems.

A comparative study of state-of-the-art driving strategies for autonomous vehicles

The autonomous vehicle is regarded as a promising technology with the potential to reshape mobility and solve many traffic issues, such as accessibility, efficiency, convenience, and especially safety. Many previous studies on driving strategies mainly focused on the low-level detailed driving behaviors or specific traffic scenarios but lacked the high-level driving strategy studies. Though researchers showed increasing interest in driving strategies, there still has no comprehensive answer on how to proactively implement safe driving. After analyzing several representative driving strategies, we propose three characteristic dimensions that are important to measure driving strategies: preferred objective, risk appetite, and collaborative manner. According to these three characteristic dimensions, we categorize existing driving strategies of autonomous vehicles into four kinds: defensive driving strategies, competitive driving strategies, negotiated driving strategies, and cooperative driving strategies. This paper provides a timely comparative review of these four strategies and highlights the possible directions for improving the high-level driving strategy design.

Driver Fatigue Detection Systems Using Multi-Sensors, Smartphone, and Cloud-Based Computing Platforms: A Comparative Analysis

Internet of things (IoT) cloud-based applications deliver advanced solutions for smart cities to decrease traffic accidents caused by driver fatigue while driving on the road. Environmental conditions or driver behavior can ultimately lead to serious roadside accidents. In recent years, the authors have developed many low-cost, computerized, driver fatigue detection systems (DFDs) to help drivers, by using multi-sensors, and mobile and cloud-based computing architecture. To promote safe driving, these are the most current emerging platforms that were introduced in the past. In this paper, we reviewed state-of-the-art approaches for predicting unsafe driving styles using three common IoT-based architectures. The novelty of this article is to show major differences among multi-sensors, smartphone-based, and cloud-based architectures in multimodal feature processing. We discussed all of the problems that machine learning techniques faced in recent years, particularly the deep learning (DL) model, to predict driver hypovigilance, especially in terms of these three IoT-based architectures. Moreover, we performed state-of-the-art comparisons by using driving simulators to incorporate multimodal features of the driver. We also mention online data sources in this article to test and train network architecture in the field of DFDs on public available multimodal datasets. These comparisons assist other authors to continue future research in this domain. To evaluate the performance, we mention the major problems in these three architectures to help researchers use the best IoT-based architecture for detecting DFDs in a real-time environment. Moreover, the important factors of Multi-Access Edge Computing (MEC) and 5th generation (5G) networks are analyzed in the context of deep learning architecture to improve the response time of DFD systems. Lastly, it is concluded that there is a research gap when it comes to implementing the DFD systems on MEC and 5G technologies by using multimodal features and DL architecture.

An Intelligent System-on-a-Chip for a Real-Time Assessment of Fuel Consumption to Promote Eco-Driving

Pollution that originates from automobiles is a concern in the current world, not only because of global warming, but also due to the harmful effects on people’s health and lives. Despite regulations on exhaust gas emissions being applied, minimizing unsuitable driving habits that cause elevated fuel consumption and emissions would achieve further reductions. For that reason, this work proposes a self-organized map (SOM)-based intelligent system in order to provide drivers with eco-driving-intended driving style (DS) recommendations. The development of the DS advisor uses driving data from the Uyanik instrumented car. The system classifies drivers regarding the underlying causes of non-optimal DSs from the eco-driving viewpoint. When compared with other solutions, the main advantage of this approach is the personalization of the recommendations that are provided to motorists, comprising the handling of the pedals and the gearbox, with potential improvements in both fuel consumption and emissions ranging from the 9.5% to the 31.5%, or even higher for drivers that are strongly engaged with the system. It was successfully implemented using a field-programmable gate array (FPGA) device of the Xilinx ZynQ programmable system-on-a-chip (PSoC) family. This SOM-based system allows for real-time implementation, state-of-the-art timing performances, and low power consumption, which are suitable for developing advanced driving assistance systems (ADASs).

The Effects of the Driver's Mental State and Passenger Compartment Conditions on Driving Performance and Driving Stress

Globalization has increased the number of road trips and vehicles. The result has been an intensification of traffic accidents, which are becoming one of the most important causes of death worldwide. Traffic accidents are often due to human error, the probability of which increases when the cognitive ability of the driver decreases. Cognitive capacity is closely related to the driver’s mental state, as well as other external factors such as the CO₂ concentration inside the vehicle. The objective of this work is to analyze how these elements affect driving. We have conducted an experiment with 50 drivers who have driven for 25 min using a driving simulator. These drivers completed a survey at the start and end of the experiment to obtain information about their mental state. In addition, during the test, their stress level was monitored using biometric sensors and the state of the environment (temperature, humidity and CO₂ level) was recorded. The results of the experiment show that the initial level of stress and tiredness of the driver can have a strong impact on stress, driving behavior and fatigue produced by the driving test. Other elements such as sadness and the conditions of the interior of the vehicle also cause impaired driving and affect compliance with traffic regulations.

Artificial intelligence method to classify ophthalmic emergency severity based on symptoms: a validation study

OBJECTIVES

We investigated the usefulness of machine learning artificial intelligence (AI) in classifying the severity of ophthalmic emergency for timely hospital visits.

STUDY DESIGN

This retrospective study analysed the patients who first visited the Armed Forces Daegu Hospital between May and December 2019. General patient information, events and symptoms were input variables. Events, symptoms, diagnoses and treatments were output variables. The output variables were classified into four classes (red, orange, yellow and green, indicating immediate to no emergency cases). About 200 cases of the class-balanced validation data set were randomly selected before all training procedures. An ensemble AI model using combinations of fully connected neural networks with the synthetic minority oversampling technique algorithm was adopted.

PARTICIPANTS

A total of 1681 patients were included.

MAJOR OUTCOMES

Model performance was evaluated using accuracy, precision, recall and F1 scores.

RESULTS

The accuracy of the model was 99.05%. The precision of each class (red, orange, yellow and green) was 100%, 98.10%, 92.73% and 100%. The recalls of each class were 100%, 100%, 98.08% and 95.33%. The F1 scores of each class were 100%, 99.04%, 95.33% and 96.00%.

CONCLUSIONS

We provided support for an AI method to classify ophthalmic emergency severity based on symptoms.

A Systematic Methodology to Evaluate Prediction Models for Driving Style Classification

Identifying driving styles using classification models with in-vehicle data can provide automated feedback to drivers on their driving behavior, particularly if they are driving safely. Although several classification models have been developed for this purpose, there is no consensus on which classifier performs better at identifying driving styles. Therefore, more research is needed to evaluate classification models by comparing performance metrics. In this paper, a data-driven machine-learning methodology for classifying driving styles is introduced. This methodology is grounded in well-established machine-learning (ML) methods and literature related to driving-styles research. The methodology is illustrated through a study involving data collected from 50 drivers from two different cities in a naturalistic setting. Five features were extracted from the raw data. Fifteen experts were involved in the data labeling to derive the ground truth of the dataset. The dataset fed five different models (Support Vector Machines (SVM), Artificial Neural Networks (ANN), fuzzy logic, k-Nearest Neighbor (kNN), and Random Forests (RF)). These models were evaluated in terms of a set of performance metrics and statistical tests. The experimental results from performance metrics showed that SVM outperformed the other four models, achieving an average accuracy of 0.96, F1-Score of 0.9595, Area Under the Curve (AUC) of 0.9730, and Kappa of 0.9375. In addition, Wilcoxon tests indicated that ANN predicts differently to the other four models. These promising results demonstrate that the proposed methodology may support researchers in making informed decisions about which ML model performs better for driving-styles classification.

An FPGA-Based Neuro-Fuzzy Sensor for Personalized Driving Assistance

Advanced driving-assistance systems (ADAS) are intended to automatize driver tasks, as well as improve driving and vehicle safety. This work proposes an intelligent neuro-fuzzy sensor for driving style (DS) recognition, suitable for ADAS enhancement. The development of the driving style intelligent sensor uses naturalistic driving data from the SHRP2 study, which includes data from a CAN bus, inertial measurement unit, and front radar. The system has been successfully implemented using a field-programmable gate array (FPGA) device of the Xilinx Zynq programmable system-on-chip (PSoC). It can mimic the typical timing parameters of a group of drivers as well as tune these typical parameters to model individual DSs. The neuro-fuzzy intelligent sensor provides high-speed real-time active ADAS implementation and is able to personalize its behavior into safe margins without driver intervention. In particular, the personalization procedure of the time headway (THW) parameter for an ACC in steady car following was developed, achieving a performance of 0.53 microseconds. This performance fulfilled the requirements of cutting-edge active ADAS specifications.

Drowsiness measures for commercial motor vehicle operations

Timely detection of drowsiness in Commercial Motor Vehicle (C MV) operations is necessary to reduce drowsiness-related CMV crashes. This is relevant for manual driving and, paradoxically, even more so with increasing levels of driving automation. Measures available for drowsiness detection vary in reliability, validity, usability, and effectiveness. Passively recorded physiologic measures such as electroencephalography (EEG) and a variety of ocular parameters tend to accurately identify states of considerable drowsiness, but are limited in their potential to detect lower levels of drowsiness. They also do not correlate well with measures of driver performance. Objective measures of vigilant attention performance capture drowsiness reliably, but they require active driver involvement in a performance task and are prone to confounds from distraction and (lack of) motivation. Embedded performance measures of actual driving, such as lane deviation, have been found to correlate with physiologic and vigilance performance measures, yet to what extent drowsiness levels can be derived from them reliably remains a topic of investigation. Transient effects from external circumstances and behaviors - such as task load, light exposure, physical activity, and caffeine intake - may mask a driver's underlying state of drowsiness. Also, drivers differ in the degree to which drowsiness affects their driving performance, based on trait vulnerability as well as age. This paper provides a broad overview of the current science pertinent to a range of drowsiness measures, with an emphasis on those that may be most relevant for CMV operations. There is a need for smart technologies that in a transparent manner combine different measurement modalities with mathematical representations of the neurobiological processes driving drowsiness, that account for various mediators and confounds, and that are appropriately adapted to the individual driver. The research for and development of such technologies requires a multi-disciplinary approach and significant resources, but is technically within reach.

Data Engineering for Affective Understanding Systems

Affective understanding is an area of affective computing which is concerned with advancing the ability of a computer to understand the affective state of its user. This area continues to receive attention in order to improve the human-computer interactions of automated systems and services. Systems within this area typically deal with big data from different sources, which require the attention of data engineers to collect, process, integrate and store. Although many studies are reported in this area, few look at the issues that should be considered when designing the data pipeline for a new system or study. By reviewing the literature of affective understanding systems one can deduct important issues to consider during this design process. This paper presents a design model that works as a guideline to assist data engineers when designing data pipelines for affective understanding systems, in order to avoid implementation faults that may increase cost and time. We illustrate the feasibility of this model by presenting its utilization to develop a stress detection application for drivers as a case study. This case study shows that failure to consider issues in the model causes major errors during implementation leading to highly expensive solutions and the wasting of resources. Some of these issues are emergent such as performance, thus implementing prototypes is recommended before finalizing the data pipeline design.

Modular design of fatigue detection in naturalistic driving environments

Research in driver mental fatigue is motivated by the fact that errors made by drivers often have life-threatening consequences. This paper proposes a new modular design approach for the early detection of driver fatigue system taking into account optimisation of system performance using particle swarm optimisation (PSO). The proposed system is designed and implemented using an existing dataset that was simultaneously collected from participants and vehicles in a naturalistic environment. Four types of data are considered as fatigue-related metrics including: vehicle acceleration, vehicle rotation pattern, driver's head position and driver's head rotation. The driver's blink rate data is used in this work as a proxy for ground truth for the classification algorithm. The collected data elements are initially fed to input modules represented by ternary neural network classifiers that estimates alertness. A Bayesian algorithm with PSO is then used to combine and optimise detection performance based on the number of existing input modules as well as their output states. Performance of the developed fatigue-detection system is assessed experimentally with a small data samples of driver trips. The obtained results are found in agreement with the state-of-the-art in terms of accuracy (90.4%), sensitivity (92.6%) and specificity (90.7%). These results are achieved with significant design flexibility and robustness against partial loss of input data source(s). However, due to small sample size of dataset (N = 3), a larger dataset need to be tested with the same system framework to generalise the findings of this work.

Real-Time Traffic Risk Detection Model Using Smart Mobile Device

Automatically recognizing dangerous situations for a vehicle and quickly sharing this information with nearby vehicles is the most essential technology for road safety. In this paper, we propose a real-time deceleration pattern-based traffic risk detection system using smart mobile devices. Our system detects a dangerous situation through machine learning on the deceleration patterns of a driver by considering the vehicle’s headway distance. In order to estimate the vehicle’s headway distance, we introduce a practical vehicle detection method that exploits the shadows on the road and the taillights of the vehicle. For deceleration pattern analysis, the proposed system leverages three machine learning models: neural network, random forest, and clustering. Based on these learning models, we propose two types of decision models to make the final decisions on dangerous situations, and suggest three types of improvements to continuously enhance the traffic risk detection model. Finally, we analyze the accuracy of the proposed model based on actual driving data collected by driving on Seoul city roadways and the Gyeongbu expressway. We also propose an optimal solution for traffic risk detection by analyzing the performance between the proposed decision models and the improvement techniques.

IoT On-Board System for Driving Style Assessment

The assessment of skills is essential and desirable in areas such as medicine, security, and other professions where mental, physical, and manual skills are crucial. However, often such assessments are performed by people called “experts” who may be subjective and are able to consider a limited number of factors and indicators. This article addresses the problem of the objective assessment of driving style independent of circumstances. The proposed objective assessment of driving style is based on eight indicators, which are associated with the vehicle’s speed, acceleration, jerk, engine rotational speed and driving time. These indicators are used to estimate three driving style criteria: safety, economy, and comfort. The presented solution is based on the embedded system designed according to the Internet of Things concept. The useful data are acquired from the car diagnostic port—OBD-II—and from an additional accelerometer sensor and GPS module. The proposed driving skills assessment method has been implemented and experimentally validated on a group of drivers. The obtained results prove the system’s ability to quantitatively distinguish different driving styles. The system was verified on long-route tests for analysis and could then improve the driver’s behavior behind the wheel. Moreover, the spider diagram approach that was used established a convenient visualization platform for multidimensional comparison of the result and comprehensive assessment in an intelligible manner.

Driving style recognition method using braking characteristics based on hidden Markov model

Since the advantage of hidden Markov model in dealing with time series data and for the sake of identifying driving style, three driving style (aggressive, moderate and mild) are modeled reasonably through hidden Markov model based on driver braking characteristics to achieve efficient driving style. Firstly, braking impulse and the maximum braking unit area of vacuum booster within a certain time are collected from braking operation, and then general braking and emergency braking characteristics are extracted to code the braking characteristics. Secondly, the braking behavior observation sequence is used to describe the initial parameters of hidden Markov model, and the generation of the hidden Markov model for differentiating and an observation sequence which is trained and judged by the driving style is introduced. Thirdly, the maximum likelihood logarithm could be implied from the observable parameters. The recognition accuracy of algorithm is verified through experiments and two common pattern recognition algorithms. The results showed that the driving style discrimination based on hidden Markov model algorithm could realize effective discriminant of driving style.