Driver Assistance Systems Research Articles

Application of CNN-LSTM Model for Vehicle Acceleration Prediction Using Car-following Behavior Data

Accurate vehicle acceleration prediction is useful for developing reliable Advanced Driving Assistance Systems (ADAS) and improving road safety. The existence of driver heterogeneity magnifies the variations in acceleration data, leading to consequential impacts on the precision of vehicle acceleration prediction. However, few studies have fully considered the driver heterogeneity when predicting vehicle acceleration. To model the characteristics of individual drivers, this study first identifies the driving behavior semantics which is defined as the underlying patterns of driving behaviors. The analysis results from the coupled hidden Markov model (CHMM) are used to evaluate the driving behavior differences between different drivers by Wasserstein distance. Then the convolutional neural network (CNN) and long short-term memory (LSTM) network are applied to predict vehicle acceleration. To validate the accuracy of the proposed prediction framework, vehicle acceleration data in car-following conditions is extracted from the safety pilot model deployment (SPMD) dataset. The segmentation results indicate that the CHMM possesses a robust capacity for modeling driving behavior. The prediction results demonstrate that the proposed framework, which incorporates driver clustering before prediction, significantly improves the accuracy of predictions. And the CNN-LSTM outperforms the LSTM in predicting vehicle acceleration during car-following scenarios. The findings from this study can enhance the development of personalized functionalities within ADAS to promote its deployment, thereby improving its acceptance and safety.

Enhanced Driver Drowsiness Detection System using Deep Learning and Image Processing Techniques

Driver drowsiness remains a critical factor contributing to road accidents worldwide. To address this issue, we present an enhanced driver drowsiness detection system leveraging advanced deep learning and image processing techniques. Our research focuses on developing a robust and accurate model capable of detecting drowsiness indicators in real-time from driver facial images. We conducted a comprehensive review of existing literature on driver drowsiness detection systems and identified key challenges and opportunities in the field. Leveraging this knowledge, we propose a novel methodology that integrates convolutional neural networks (CNNs) for feature extraction and classification, coupled with sophisticated image processing algorithms for facial recognition and eye state analysis. We describe the experimental setup, data collection process, and model training procedures, followed by a detailed presentation of results and performance evaluation metrics. Our findings demonstrate significant improvements in drowsiness detection accuracy, with the proposed system achieving promising results in both laboratory and real-world driving scenarios. The implications of our research extend to the development of more effective driver assistance systems and the enhancement of road safety measures. We conclude by discussing future research directions and potential applications for advancing driver drowsiness detection technology. Keywords— Driver Drowsiness Detection, Deep Learning, Image Processing, Convolutional Neural Networks (CNNs), Facial Recognition, Road Safety, Real-Time Monitoring, Driver Assistance Systems, Accident Prevention

Adaptive Advanced Emergency Braking on Combined Road Friction Coefficients

In this study, an Advanced Emergency Braking System (AEBS) is adopted to the combined road friction coefficients by proposing an emergency steering maneuver. AEBS is one of the significant driver assistant systems to avoid rear-end collisions. AEBS supports the drivers with acoustic and optical warnings before applying the full braking maneuver. However, in special cases, such as the sudden decrease of road friction coefficient during braking, a single hard braking maneuver without a proper steering assist may not be sufficient to prevent a rear-end collision. Therefore, AEBS may work with the autonomous emergency steering systems to prevent inevitable rear-end collisions. The originality of this study arises from the necessity of the adaptation of the AEBS to work in cooperation with the autonomous emergency steering systems. A predictive controller was used to support the emergency braking maneuver with an autonomous steering maneuver via observing the yaw rate of the vehicle. The predictive controller was developed in MATLAB software. The implementation of the controller was performed in a non-linear four-wheel vehicle model in the IPG/CarMaker simulation environment. The communication between the simulation environment and MATLAB software was established in the Simulink interface of MATLAB. The results proved that the predictive controller maintained the vehicle lateral stability without causing any type of collisions.

MMoNet3D: Enhancing monocular 3D object localization with modern advancements

MMoNet3D, an innovative project, addresses monocular multi-object detection and 3D localization in real-time, surpassing the limitations of the existing MoNet3D system. Utilizing a custom deep learning model inspired by Centernet, it excels in oriented car detection across static images, videos, and live webcam feeds. The system, powered by the KITTI dataset, showcases its potential impact on real-time monocular multi-object detection. Its unique features, including 3D bounding box visualization and a user-friendly GUI, set MMoNet3D apart, making it suitable for embedded advanced driving-assistance systems. This advancement promises heightened accuracy and efficiency, contributing to enhanced vehicular safety and operational efficacy. MMoNet3D stands as a beacon in computer vision, paving the way for intelligent transportation systems.

Deep learning approach for accurate and stable recognition of driver's lateral intentions using naturalistic driving data

Accurate and stable recognition of a driver's lateral intention is a crucial prerequisite for the proper functioning of advanced driver-assistance systems (ADAS). Existing studies usually rely on auxiliary sensor signals, such as cameras and eye trackers; however, this reliance poses challenges in applying these methods to vehicles lacking such auxiliary sensors. Furthermore, existing studies have not fully leveraged the inherent temporal dependence of lateral intentions, leading to difficulties in avoiding erroneous recognition interruptions. Thus, this study proposes a deep-learning-based lateral intention recognition method to achieve accurate and stable recognition of lateral intention using onboard sensor signals. First, a real vehicle is used to collect a vast amount of driving data, and thus guarantee the robustness and practicality of the recognition model. Subsequently, vehicle trajectories are extracted, and a trajectory clustering method is used to label lateral intentions of the driving data; these intention labels and a feature selection algorithm are utilized to select the most representative recognition features. Therefore, a lateral driving intention recognition model is constructed using double convolutional neural networks with a long short-term memory layer (CNN-LSTM). This network architecture can fully utilize the temporal dependence of lateral intentions. Finally, the recognition performance of the designed double CNN-LSTM networks is validated using the existing driving data and real-world vehicle tests. The results indicate that the double CNN-LSTM networks can achieve stable recognition of lateral intention in real-time and the accuracy reaches 98.64% in the experiment.

Method of brake mechanisms thermal load control under the influence of automated braking system load modes

Introduction (problem statement and relevance). Most of the kinetic energy of a vehicle equipped with an anti-lock braking system (ABS) is damped due to friction work in the brakes. Besides the ABS function, the automated braking system implements a lot of functions of active safety and driver assistance systems by using regular service brake mechanisms as an actuator. Overheating of brakes, namely of their friction pairs, leads to the phenomenon of critical fading, which is accompanied by a sharp decrease in braking torque. Reducing the impact of this phenomenon is a very difficult task in terms of both taking into account the cost of the braking mechanism and its minimum complexity. The proposed procedure makes it possible to assess the influence of operation of various active safety and driver assistance systems on the thermal load of the brake mechanisms. The prospect of its expansion when creating new vehicle designs is also covered.The purpose of the study is to increase the reliability of brake mechanisms, braking characteristics stability by improvement of the procedure for calculation and reduction of the brake mechanism thermal load of the wheeled vehicles equipped with the automated brake system.Methodology and research methods. Theoretical and experimental studies were conducted in the paper. In the theoretical study, the main computational tool is the finite element method (FEM) with using of licensed software packages Abaqus CAE and SolidWorks Simulink.Scientific novelty and results. The thermal load estimation procedure for brake mechanisms of vehicles that are equipped with an automated brake system has been created. The calculation procedure developed takes into account the increase in the proportion of the vehicle kinetic energy that is extinguished in the brake mechanism of the vehicle equipped with different subsystem functions.Practical significance. Application of the developed complex calculation procedure with any thermal calculation complexes on the FEM basis will allow brake system developers to reduce the costs for intermediate tests for thermal load at the design stage, and upon that, the influence of functional subsystems of the automated brake system will be taken into account. Application of the suggested improvement of the monitoring and control systems for the automated brake system will reduce the probability of fading phenomenon occurrence.

Synthesis of ISO 26262 and ISО 21448 methods to increase reliability of hazard analysis and risk assessment of intelligent driver assistance systems operation

Introduction (problem statement and relevance). Application of sensors to increase automation of vehicles is a growing global trend. However, as the number of these devices in vehicles increases, so does the number of risk factors caused by their application.The purpose of the study is to reduce the areas of unknown potentially dangerous situations of possible impact of these risks on a highly automated vehicle using a synthesis of the risk analysis methods.Methodology and research methods. This paper considers the system analysis methodologies set forth in GOST R ISO 26262, ISO 21448:2022.The scientific novelty is the interference of approaches to the analysis that are set forth in the standards.Practical significance. The results of application of the synthesized approach to risk analysis will form the basis of studies of the prospective intelligent driver assistance systems thus allowing the development of test programs for these systems within highly automated vehicles.

Multi-level and Metrics Evaluation Approach for Data-Driven Based Sensor Models

Nowadays, with increased sensor perception performance for Advanced Driver Assistance Systems (ADAS), scenario-based simulation is becoming more frequent to manage the complexity of reality in terms of cost and time. The perception system provides the basis for the vehicle guidance algorithms calculation, but the simulation of ADAS sensors is a challenging task in virtual testing. Literature reports the magnitude of relevant modelling approaches and data-driven models becoming increasingly important. A basic method is to fit the sensor output in the virtual environment with high-fidelity measurements of real-world scenarios, thus a direct relation can be established between real and synthetic sensor data. To prove the suitability of a method, it is necessary to quantify the gap between simulation and reality to determine the performance of different models. In this work, authors address this problem and visualize the gap by introducing a multi-level evaluation approach that combines Model Generalization Ability Evaluation and Case Implicit Performance Evaluation. The former directly evaluates the model’s overall performance, while the latter is used for specific cases in simulation. The study shows that this combined evaluation approach provides an in-depth framework for evaluating sensor models to make the differences apparent.

An investigation of ADAS testing scenarios based on vehicle-to-powered two-wheeler accidents occurring in a county-level district in Hunan province

Abstract To meet the requirements for high-risk testing scenarios in the safety assessment of advanced driver assistance systems (ADAS), this study selected 239 vehicle-to-powered two-wheeler accidents occurring in Xiangtan County from the China In-depth Mobility Safety Study—Traffic Accident (CIMSS-TA) database. Seven scenario elements were selected from the levels of static road elements, dynamic traffic participants and environment elements. The accident characteristics were obtained through statistical analysis. The results showed 75.73% of accidents occurred at T-junctions/intersections, and 34.73% of accidents were caused by the visual obstructions. Six typical vehicle-to-powered two-wheeler crash scenarios were established by cluster analysis, including two straight road scenarios (scenario 3 and 6), two T-junction scenarios (scenario 1 and 5), and two intersection scenarios (scenario 2 and 4). Furthermore, six vehicle-to-powered two-wheeler testing scenarios, which represent the traffic environment characteristics of Chinese county-level districts, were constructed by analyzing the vehicle/PTW speed, the first collision point of vehicle, obstacle type and location, etc. The results showed the vehicle speed in scenario 4 was higher than that in scenario 2, with the average values of 34.32 km/h and 21.99 km/h, respectively. The results of this study provided testing scenarios for the safety assessment of ADAS, and contributed to improving the vehicle active safety technology.

A Fast and Accurate Lane Detection Method Based on Row Anchor and Transformer Structure.

Lane detection plays a pivotal role in the successful implementation of Advanced Driver Assistance Systems (ADASs), which are essential for detecting the road's lane markings and determining the vehicle's position, thereby influencing subsequent decision making. However, current deep learning-based lane detection methods encounter challenges. Firstly, the on-board hardware limitations necessitate an exceptionally fast prediction speed for the lane detection method. Secondly, improvements are required for effective lane detection in complex scenarios. This paper addresses these issues by enhancing the row-anchor-based lane detection method. The Transformer encoder-decoder structure is leveraged as the row classification enhances the model's capability to extract global features and detect lane lines in intricate environments. The Feature-aligned Pyramid Network (FaPN) structure serves as an auxiliary branch, complemented by a novel structural loss with expectation loss, further refining the method's accuracy. The experimental results demonstrate our method's commendable accuracy and real-time performance, achieving a rapid prediction speed of 129 FPS (the single prediction time of the model on RTX3080 is 15.72 ms) and a 96.16% accuracy on the Tusimple dataset-a 3.32% improvement compared to the baseline method.

Synthetic Displays and Their Potential for Driver Assistance Systems

Advanced visual display technologies typically supplement the out-of-window view with separate displays (e.g., analog speedometer or artificial horizon) or with overlays (e.g., projected speedometer or map). Studies on head-up displays suggest that altering the out-of-window view itself is superior to supplemental displays, as sensor-based information not normally visible to the driver can be included. Such novel synthetic displays have been researched for cockpit implementation but less so for driving. We discuss such view-altering synthetic displays in general, and camera–monitor systems (CMS) designed to replace rear-view mirrors as a special instance of a novel synthetic display in the automotive domain. In a standard CMS, a camera feed is presented on a monitor, but could also be integrated into the windshield of the car. More importantly, the camera feed can undergo alterations, augmentations, or condensations before being displayed. The implications of these technologies are discussed, along with findings from an experiment examining the impact of information reduction on a time-to-contact (TTC) estimation task. In this experiment, observers judged the TTC of approaching cars based on the synthetic display of a futuristic CMS. Promisingly, TTC estimations were unaffected by information reduction. The study also emphasizes the significance of the visual reference frame.

The prevention of road accidents in non-expert drivers: Exploring the influence of Theory of Mind and driving style

BackgroundDrivers consider external and hypothetical behaviours of other drivers, but internal factors also impact road safety. ObjectivesThis study aims to examine the connection between Cognitive and Affective Theory of Mind (ToM) and Driving Style in road safety. It hypothesizes that a higher level of ToM corresponds to a greater ability to avoid accidents and to assume virtuous driving behaviour. The study investigates how ToM impacts driving behaviour, directly correlating with assessing potential accidents' probability and severity. Method207 non-expert drivers, including 164 females, participated in assessing Cognitive and Affective ToM through two tasks. They also completed self-measured questionnaires to assess their driving style and behaviours. In addition, they evaluated 12 videos depicting cars, motorcycles, trucks, and pedestrians to determine the probability of a road accident and the level of danger in each scenario. ResultsThe results of the mediation models clearly indicate a relationship between ToM and the reduction of road accidents. Specifically, Cognitive ToM plays a crucial role in assessing the probability of risky and dangerous situations related to Risky and Angry Driving Styles. However, it was observed that Cognitive ToM does not significantly affect the prediction of actual driving behaviours. ConclusionsFindings are discussed within the theoretical framework of the Task-Capability Interface Model and the Embodied Simulation Model based on mirror neuron research. Our results suggest the importance of creating drive-assistance systems considering both the Cognitive ToM and Driving Style to reduce road accidents among non-expert drivers.

Predicting Driver's mental workload using physiological signals: A functional data analysis approach

This study investigates the impact of advanced driver-assistance systems on drivers' mental workload. Using a combination of physiological signals including ECG, EMG, EDA, EEG (af4 and fc6 channels from the theta band), and eye diameter data, this study aims to predict and categorize drivers’ mental workload into low, adequate, and high levels. Data were collected from five different driving situations with varying cognitive demands. A functional linear regression model was employed for prediction, and the accuracy rate was calculated. Among the 31 tested combinations of physiological variables, 9 combinations achieved the highest accuracy result of 90%. These results highlight the potential benefits of utilizing raw physiological signal data and employing functional data analysis methods to understand and assess driver mental workload. The findings of this study have implications for the design and improvement of driver-assistance systems to optimize safety and performance.

Proactive Braking Control System for Collision Avoidance during Right Turns with Occluded Vision at an Intersection

This paper describes the development of an Advanced Driver Assistance System (ADAS) which will allow drivers to avoid collisions with an oncoming vehicle from an occluded area when turning right at intersections in left-hand traffic. Connected vehicles, in coordination with infrastructure, represent one of the commercialized ADAS technologies for collision avoidance. However, the coverage of the ADAS will be limited to designated intersections only, as communication equipment needs to be installed in both the vehicle and infrastructure to enable the assistance. This paper proposes an ADAS using on-board sensors, independent of infrastructure facilities, to control the vehicle velocity to avoid collisions. Most current intersection assistance, by using an Autonomous Emergency Braking System (AEBS), allows the driver to avoid a collision with oncoming vehicles when there is clear vision without occlusion. However, many accidents occur when the vehicle detects the oncoming vehicle too late because of occlusion in the intersection, such as a vehicle in the opposite lane. This system calculates the hazardous speed criteria of the ego vehicle, which might result in a high risk of collision when darting out occurs, and provides speed control assistance to allow the driver to escape from the hazardous speed region. The simulation results reveal that the proposed system effectively reduces the possibility of collisions compared to conventional AEBS.

Data-Driven Controller for Drivers’ Steering-Wheel Operating Behaviour in Haptic Assistive Driving System

An advanced driver-assistance system (ADAS) is critical to driver–vehicle-interaction systems. Driving behaviour modelling and control significantly improves the global performance of ADASs. A haptic assistive system assists the driver by providing a specific torque on the steering wheel according to the driving–vehicle–road profile to improve the steering control. However, the main problem is designing a compensator dealing with the high-level uncertainties in different driving scenarios with haptic driver assistance, where different personalities and diverse perceptions of drivers are considered. These differences can lead to poor driving performance if not properly accounted for. This paper focuses on designing a data-driven model-free compensator considering various driving behaviours with a haptic feedback system. A backpropagation neural network (BPNN) models driving behaviour based on real driving data (speed, acceleration, vehicle orientation, and current steering angle). Then, the genetic algorithm (GA) optimises the integral time absolute error (ITEA) function to produce the best multiple PID compensation parameters for various driving behaviours (such as speeding/braking, lane-keeping and turning), which are then utilised by the fuzzy logic to provide different driving commands. An experiment was conducted with five participants in a driving simulator. During the second experiment, seven participants drove in the simulator to evaluate the robustness of the proposed combined GA proportional-integral-derivative (PID) offline, and the fuzzy-PID controller applied online. The third experiment was conducted to validate the proposed data-driven controller. The experiment and simulation results evaluated the ITAE of the lateral displacement and yaw angle during various driving behaviours. The results validated the proposed method by significantly enhancing the driving performance.

The role of driver head pose dynamics and instantaneous driving in safety critical events: Application of computer vision in naturalistic driving

This paper investigates the role of driver behavior especially head pose dynamics in safety–critical events (SCEs). Using a large dataset collected in a naturalistic driving study, this paper analyzes the head pose dynamics and driving behavior in moments leading up to crashes or near-crashes. The study uses advanced computer vision and mixed logit modeling techniques to identify patterns and relationships between drivers’ head pose dynamics and crash involvement. The results suggest that driver-head pose dynamics, especially poses that indicate distraction and movement volatility, are important factors that can contribute to undesirable safety outcomes. Marginal effects show that angular deviation for head pose dynamics indicated by yaw, pitch and roll increase the likelihood of crash intensity by 4.56%, 4.92% and 8.26% respectively. Furthermore, traffic flow and lane changing also contribute to increase in likelihood of crash intensity. These findings provide new insights into pre-crash factors, especially human factors and safety–critical events. The study highlights the importance of considering human factors in designing driver assistance systems and developing safer vehicles. This research contributes by examining naturalistic driving data at the microscopic level with early detection of behaviors that lead to SCEs and provides a basis for future research on automation.

STA-Net: A Spatial–Temporal Joint Attention Network for Driver Maneuver Recognition, Based on In-Cabin and Driving Scene Monitoring

Next-generation advanced driver-assistance systems (ADASs) are a promising direction for intelligent transportation systems. To achieve intelligent security monitoring, it is imperative that vehicles possess the ability to accurately comprehend driver maneuvers amidst diverse driver behaviors and complex driving scenarios. Existing CNN-based and transformer-based driver maneuver recognition methods face challenges in effectively capturing global and local features across temporal and spatial dimensions. This paper proposes a Spatial–Temporal Joint Attention Network (STA-Net) to realize high-efficient temporal and spatial feature extractions in driver maneuver recognition. First, we introduce a two-stream architecture for a concurrent analysis of in-cabin driver behaviors and out-cabin environmental information. Second, we propose a Multi-Scale Transposed Attention (MSTA) module and Multi-Scale Feedforward Network (MSFN) to extract features at multiple scales, addressing receptive field inadequacies and combining high-level and low-level information. Third, to address the information redundancy in multi-scale features, we propose a Cross-Spatial Attention Module (CSAM) and Multi-Scale Cross-Spatial Fusion Module (MCFM) to select essential features. Additionally, we introduce an asymmetric loss function to effectively tackle the issue of sample imbalance across diverse categories of driving maneuvers. The proposed method demonstrates a remarkable accuracy of 90.97% and an F1 score of 89.37% on the Brain4Cars dataset, surpassing the performance of the methods compared. These results substantiate the fact that our approach effectively enhances driver maneuver recognition.

Advanced Data Science Applications in Vehicles: A Comprehensive Review

Purpose: This research paper aims to explore the transformative impact of data science applications within the automotive industry, with a focus on the evolution of vehicles into intelligent, connected, and autonomous entities. Through an extensive literature review and analysis of recent developments, the paper seeks to provide a comprehensive understanding of the current state and future prospects of data science in vehicles.
 Methodology: Advancements in sensor technologies, such as LiDAR, HD maps, radars, location tracking and cameras, are discussed, highlighting their role in data collection for applications like advanced driver assistance systems (ADAS) and autonomous driving. Relevant citations are provided. This subsection covers the collection and transmission of real-time data through telematics systems, showcasing their importance for predictive maintenance, fleet management, and personalized insurance programs. Citations support the presented information. The deployment of edge computing for real-time data processing in vehicles is discussed, emphasizing its significance for safety-critical applications like collision avoidance. Citations are provided to support the information. This section explores the application of machine learning algorithms to predict vehicle failures, optimize fuel efficiency, and analyze driver behavior. The importance of leveraging historical data to create accurate models is highlighted with supporting citations.
 Findings: The pivotal role of data science in enabling autonomous vehicles to navigate complex environments is discussed, emphasizing the use of machine learning models for real-time decision-making. Citations support the presented information. This subsection explores the integration of data science in traffic management systems, covering dynamic traffic signal control, congestion prediction, and route optimization. Citations support the findings related to traffic management applications. The paper discusses challenges associated with widespread implementation, including data privacy concerns, cybersecurity risks, and the need for standardized communication protocols. Additionally, potential future directions are outlined, such as the integration of blockchain for secure data sharing and the development of advanced human-machine interfaces.
 Unique Contribution to Theory, Practice and Policy: This research paper provides a well-rounded contribution by seamlessly integrating theoretical concepts, practical applications, and policy considerations in the realm of advanced data science applications in vehicles.

Assessing Training Methods for Advanced Driver Assistance Systems and Autonomous Vehicle Functions: Impact on User Mental Models and Performance

Understanding the complexities of Advanced Driver Assistance Systems (ADAS) and Autonomous Vehicle (AV) technologies is critical for road safety, especially concerning their adoption by drivers. Effective training is a crucial element in ensuring the safe and competent operation of these technologies. This study emphasises the critical role of training methodologies in shaping drivers’ mental models, defined as an individual’s cognitive frameworks for understanding and interacting with ADAS and AV systems. Their mental models substantially influence their interactions with those technologies. A comparative analysis of text-based and video-based training methods has been conducted to assess their influence on participants’ performance and the development of their mental models of ADAS and AV functionalities. Performance is evaluated in terms of the accuracy and reaction time of the participants as they interacted with ADAS and AV functions in a driving simulation. The findings reveal that video-based training yielded better performance outcomes, more accurate mental models, and a deeper understanding of ADAS functionalities among participants. These findings are crucial for policy makers, automotive manufacturers, and educational institutions involved in driver training. They underscore the necessity of developing tailored training programs to facilitate the proficient and safe operation of increasingly complex automotive technologies.

Modelling yielding dilemma zone for drivers at non-signalised crosswalks

In developing countries like China, non-signalised crosswalks represent a considerable risk to pedestrian safety. This study introduces a “yielding dilemma zone” model to tackle the decision-making challenges encountered by drivers at these crosswalks. Field data collected via drones at three non-signalised crosswalks in Xi'an (China) facilitated the analysis of 1687 instances of driver yielding behaviour. The study employed Support Vector Machine (SVM), Artificial Neural Network (ANN), and Binary Logistic Regression (BLR) models to establish the boundaries of the dilemma zone. Additionally, it utilized surrogate safety metrics, such as the deceleration rate required to stop (DRS) and time to crossing (TC), to compare risks between dilemma and non-dilemma zones. Each modelling approach provided insightful findings. The SVM and ANN models identified the speed-distance dilemma zones, while the BLR model determined the multifactorial dilemma zones by considering characteristics such as pedestrian, vehicle, and traffic condition. The findings underscore the significance of the “yielding dilemma zone” for understanding driver-pedestrian interactions at non-signalised crosswalks. The innovative multi-model approach uncovered that pedestrian speed, vehicle type, vehicle approach speed, adjacent vehicle number, driving directions, and pedestrian crossing stage significantly influence this interaction. indicated a higher probability of collisions within dilemma zones compared to non-dilemma zones across all models. The study suggests that the SVM model could be instrumental in creating passive physical warning signs, while the ANN and BLR models could significantly contribute to real-time pedestrian and driver assistance systems. The “yielding dilemma zone” model, with its comprehensive approach, presents a promising avenue for enhancing pedestrian safety, intelligent transportation systems, and traffic management at non-signalised crosswalks.