Lane-keeping Assistance Research Articles

Correlation Between Traffic Locus of Control and Advanced Driver Assistance System (ADAS) Use

Inappropriate use of advanced driver assistance systems (ADAS), such as adaptive cruise control (ACC) and lane keeping assistance (LKA), may cause safety issues. In a driving simulator experiment, we examined how a specific driver characteristic, locus of control in traffic (T-LOC), influences the use of ADAS. It was found that without any ADAS training, the Self dimension of T-LOC, measuring the extent to which one believes they are responsible for their driving safety, was negatively correlated with time spent on secondary tasks after environmental cues became visible indicating a possible need for takeover. Further, with a responsibility-focused training, drivers with higher Self scores tended to take control over faster when necessary. The findings highlight the role of individual differences in the formation of driver behaviors associated with ADAS use. Investigation of other driver characteristics and incorporating them to the design of training paradigms are suggested.

A Comprehensive Analysis of AI/ML-enabled Predictive Maintenance Modelling for Advanced Driver-Assistance Systems

Advanced Driver-Assistance Systems (ADAS) are changing driver-vehicle interactions to improve road safety and reduce distractions. Technological advances like ADAS and AI in cars present societal challenges and opportunities. It shows how AI aids human-machine communication by improving motor skills. The auto industry is interested in ADAS because it can increase energy efficiency, safety, and comfort. Numerous studies have shown its benefits. ADAS and vehicle networking show promise, but establishing a sound control system is challenging. Model Predictive Control (MPC) is one answer to these problems. To manage higher-level connectivity and automation, the paper analyses and implements key research. It also finds issues and recommends solutions. The latest driverless car ADAS improvements have dramatically increased passenger safety. These systems are safer and more automated using sensors and ECUs. Most ADAS have RADAR, cameras, ultrasonic, and LiDAR. This work uses AI/ML-enabled Predictive Maintenance modelling to improve ADAS safety and longevity. AI and ML in Advanced Driver Assistance Systems (ADAS) are significant vehicle safety and reliability advances. AI/ML-enabled predictive maintenance detects and fixes ADAS component faults. ADAS predictive maintenance using AI/ML can detect issues, improve driver safety, and boost vehicle efficiency. Advanced sensor arrays and control units are needed for adaptive cruise control, traffic sign recognition, and lane-keeping assistance. AI/ML algorithms discover issues and enable early interventions in predictive maintenance models. Predictive maintenance is examined utilizing classical machine learning, deep learning, and reinforcement learning. Integration of numerous AI/ML models, real-time data processing, customization based on vehicle usage patterns, scalability, and adaptability of predictive maintenance models to new ADAS technologies are research gaps.

A Method to Develop the Driver-Adaptive Lane-Keeping Assistance System Based on Real Driver Preferences.

To satisfy the preference of each driver, the development of a Lane-Keeping Assistance (LKA) system that can adapt to individual drivers has become a research hotspot in recent years. However, existing studies have mostly relied on the assumption that the LKA characteristic aligned with the driver's preference is consistent with this driver's naturalistic driving characteristic. Nevertheless, this assumption may not always hold true, causing limitations to the effectiveness of this method. This paper proposes a novel method for a Driver-Adaptive Lane-Keeping Assistance (DALKA) system based on drivers' real preferences. First, metrics are extracted from collected naturalistic driving data using action point theory to describe drivers' naturalistic driving characteristics. Then, the subjective and objective evaluation method is introduced to obtain the real preference of each test driver for the LKA system. Finally, machine learning methods are employed to train a model that relates naturalistic driving characteristics to the drivers' real preferences, and the model-predicted preferences are integrated into the DALKA system. The developed DALKA system is then subjectively evaluated by the drivers. The results show that our DALKA system, developed using this method, can enhance or maintain the subjective evaluations of the LKA system for most drivers.

Which partially automated driving function do drivers prefer? Results from two field studies on public highways

Partial driving automation consists of several subsystems that drivers need to trust to an appropriate level and interact with in a secure manner. To investigate whether drivers adapt their trust functional specific, how they experience the subsystems, and what their preferences are, we conducted two field studies with two partially automated (SAE level 2) vehicles and involving a total of 132 drivers. In both studies, participants drove on a public highway with support from partial driving automation systems for longitudinal control (adaptive cruise control, ACC) and lateral control (lane-keeping assistance, LA). Participants in Study 2 were additionally able to use a lane change assistance (LC) system. The drivers' trust generally increased over time and reflected in how much a participant used the automation (Study 1). Hereby, in both studies drivers had functional specific trust: in Study 1 drivers trusted the ACC more than the LA. In Study 2 drivers in addition had lower trust in the LC than in the LA and the ACC. In their thinking-aloud statements, drivers were more positive about ACC and more critical about LA but reported more interaction difficulties with the ACC (Study 1). In Study 2 participants on average preferred the ACC over the LA and the LA over the LC. The interview responses suggest that the lower preference was due to the perceived reliability (LA) and usefulness (LC). In summary, our results indicate that drivers adapt their trust functional specific when using partial driving automation and have an overall preference for ACC that persists despite a higher number of interaction problems.

Driver-Centric Lane-Keeping Assistance System Design: A Noncertainty-Equivalent Neuro-Adaptive Control Approach

Vehicle roadway departure accidents are a major traffic safety concern as they oftentimes result in severe injuries and fatalities. To address such an issue, this article originates a novel driver-centric and neuro-adaptive-control-based lane-keeping assistance system (LKAS). The proposed control strategy synergizes a noncertainty-equivalent adaptive control design scheme, an adaptive radial-basis-function-based neural network (RBFNN) that captures the human driver's lane-keeping steering behavior, and a Gudermannian-function-based smooth parameter projection operator. The benefit and uniqueness of the proposed solution are threefold. First and foremost, the noncertainty-equivalent adaptive control design, which leverages the immersion-and-invariance-like methodology, ensures the asymptotical convergence of the parameter-estimation-error-induced perturbation despite the reference signal's persistency of excitation condition. Second, the LKAS is devised to be driver-centric, i.e., an adaptive RBFNN-based human driver steering model is embedded inside the LKAS's algorithm such that a human driver is assisted in a personalized and adaptive manner. Third, the Gudermannian-function-based smooth parameter projection operator ascertains the prescribed boundedness of the control parameters while maintaining the control action's smoothness. A pilot human-subject study using a high-fidelity moving-base driving simulator is conducted to validate the proposed LKAS. Further, its performance is compared with a baseline certainty-equivalent neuro-adaptive controller.

Investigation of an Autonomous Vehicle's using Artificial Neural Network (ANN)

This paper presents a study of autonomous vehicles on a normal highway in UAE, using MATLAB/Simulink® 2022. There is great potential for autonomous vehicles to improve the safety and efficiency of transportation, enhance the quality of time spent in cars, and make transportation more accessible to everyone. The vehicles utilized in the study have three different speeds 40, 80, and 120 km/hr. All vehicles modelled are representative of that available in the UAE. In the model, lane following and lane-keeping assistance functions and Simulink block which are described using artificial neural networks are selected. Simulation is validated with existing published results of physical vehicle models. In the simulations, it is assumed that vehicles have minimal steering angles as the system is in an autonomous collision free environment, selected from MATLAB. Results are obtained as velocities, accelerations, and safe distance with respect to the preceding vehicle. The following results are critically analyzed and validated.

Stability Analysis of Lane-Keeping Assistance System for Trucks under Crosswind Conditions

To enhance the control accuracy of lane-keeping assistance systems for trucks encountering crosswind-induced lateral deviations to improve the lateral stability of the vehicle, this study proposes a control strategy based on a linear quadratic regulator (LQR) using a path-tracking preview model. First, the lateral deviation is calculated using the path-tracking preview model. Then, an observer for the vehicle’s sideslip angle is designed using a vehicle lateral tracking deviation model and a Kalman filter controller, and this is used to solve the deviation of the sideslip angle. Finally, a feedforward controller is designed based on the LQR controller and a linear two-degrees-of-freedom vehicle model to eliminate steady-state errors arising from LQR optimization, thereby obtaining the steering angle of the vehicle when subjected to crosswind conditions. Comparing the test results of the sideslip angle, yaw rate, and lateral acceleration demonstrates that this strategy effectively improves the control accuracy of lane-keeping under crosswind conditions. The proposed method is validated through hardware-in-the-loop experiments on a test bench, yielding results consistent with simulations.

ADAS USING TOUCH SENSOR

EPS systems integrate Advanced Driver Assistance Systems (ADAS) while the driver controls the steering wheel. Touch sensors, torque, and position sensors improve driver involvement, safety, and control in EPS. ADAS features like lane-keeping assistance and adaptive cruise control are activated by the driver’s steering wheel touch. EPS with driver-enabled ADAS integration, using touch sensors, empowers drivers while leveraging enhanced automation. The vehicle model design incorporates touch sensors, Arduino, and motor drivers to enable Advanced Driver Assistance Systems (ADAS) exclusively when the driver holds the steering wheel. The touch sensors detect the driver’s touch on the steering wheel, triggering the ADAS functionalities and allowing the vehicle to move. The Arduino microcontroller, equipped with embedded C code, processes the touch sensor inputs and communicates with the motor drivers to initiate vehicle propulsion. The embedded C code facilitates real-time monitoring and control of the touch sensor inputs, enabling a seamless integration between driver interaction and vehicle movement, thereby enhancing both safety and user experience

Ground vehicle lane-keeping assistance system via differential flatness output feedback control and algebraic derivative estimation

Vehicle run-off-road is one of the most frequent and fatal traffic accidents in the United States. Various lane-keeping assistance (LKA) systems have been developed in the last decade to help drivers stay on the road. Most of them are built upon linear driver–vehicle–road (DVR) models and treat road curvature as a disturbance. Albeit effective, their control performance would degrade if road curvature varies rapidly. This paper proposes a novel nonlinear DVR model by integrating a driver steering model into vehicle–road kinematics, which explicitly considers road curvature. Particularly, this nonlinear DVR system has been proven to be differentially flat, and a flatness-based LKA system is designed. Additionally, Model-Free Control is introduced to compensate for system modeling errors. Hardware-in-the-loop simulations and diver-in-the-loop experiments validate the proposed control framework and demonstrate the performance enhancement with respect to a representative linear robust LKA system.

Synthesizing fatal crashes involving partially automated vehicles and comparing with fatal crashes involving non-automated vehicles

Automated vehicles (AVs) are expected to improve safety by gradually reducing human decisions while driving. However, there are still questions on their effectiveness as we transition from almost 0% AVs to 100% AVs with different levels of vehicle autonomy. The focus of this study, therefore, is on synthesizing and identifying risk factors influencing fatal crashes involving partially automated vehicles (PAVs) i.e., level 1 AVs and level 2 AVs, in the United States. Fatal crashes involving non-AVs (level 0 vehicles) within their vicinity were used for comparison, to minimize unobserved heterogeneity and randomness associated with the influencing risk factors. The fatal crash data for the years 2016 to 2019 is used for the analysis. A partial proportional odds model is developed using crash, road, and vehicle characteristics as independent variables while the fatal crash involving a vehicle with a specific level of automation (0, 1, or 2) is used as the dependent variable. The level of automation was captured by developing tools and identifying advanced features in each vehicle based on the vehicle identification number (VIN). The odds ratios varied for PAVs compared to non-AVs. PAVs are safer but are more likely to be involved in a fatal crash with non-motorists. Pedestrian automatic emergency braking (PAEB) and lane-keeping assistance (LKA) were observed to improve safety by reducing possible collision with a pedestrian and roadside departure, respectively. Contrarily, vehicles with other smart features are still highly likely to be involved in fatal crashes, demanding further research and attention.

Lateral Shared Sliding Mode Control for Lane Keeping Assist System in Steer-by-Wire Vehicles: Theory and Experiments

The lane-keeping assistance design for steer-by-wire road vehicles is a multi-objective control problem that addresses lane tracking, improvement of driver comfort and ensuring vehicle stability. The designed architecture must be sensitive to various driver behaviors, provide assistance to drivers whenever required and ensure a smooth transition of authority between manual and automated driving modes. In this work, we propose a novel shared lane-keeping controller-based on quasi-continuous high-order sliding mode control dealing with the above challenges. The parameters of the controller are characterized by an upper bound estimate of the lateral winds force and the road curvature which acts as disturbances. Using the sharing parameter dependent on the monitored driver attribute, the authority is smoothly transitioned between various modes while accounting for driver behaviors. Experimental results on the SHERPA vehicle simulator shows the real-time implementation feasibility. Extensive results and discussions are provided to demonstrate the effectiveness and the applicability of the proposed controller.

Prediction-Uncertainty-Aware Threat Detection for ADAS: A Case Study on Lane-Keeping Assistance

Advanced driver assistance systems typically support the driver in cases where the driver is likely to fail the driving task. The challenge, from a system perspective, is to accurately detect those cases. Recently, machine learning-based prediction models that are able to estimate the prediction uncertainty in real-time have successfully been introduced for this purpose. However, very little effort has been made on using the prediction uncertainty in the decision-making logic to improve the system's robustness, especially in cases where the input data is affected by noise or anomalies that are not presented in the training data. In this work, four threat-detection methods using uncertainty estimates are proposed and evaluated using a real-world data set. The methods use different strategies for leveraging uncertainty information, where the goal is to ensure that the intervention decision is based on trustworthy predictions. The threat-detection methods' performances are evaluated, using five different learning-based prediction models, in the context of a lane-keeping assistance application.

Evaluation of macroscopic fundamental diagram characteristics for a quantified penetration rate of autonomous vehicles

BackgroundThe availability of private vehicles with autonomous features is widespread nowadays. Various car manufacturers are providing attributes like collision warning, city automatic emergency braking, adaptive cruise control, pedestrian detection, lane-keeping assistance and lane departure warning, rear cross-traffic and blind-spot warning in their high-end models.PurposeSuch features can automatically manage the macroscopic fundamental traffic parameters such as speed, headway, etc adaptively. Consequently leading to a heterogeneous traffic stream with diverse car-following behaviour comprising completely manual/traditional (TVs) and autonomous vehicles (AVs). This questions the applicability of classic traffic flow theory relationships on such heterogeneous traffic streams.MethodologyThis paper focuses on developing the macroscopic fundamental diagram for such heterogeneous traffic streams based on the quantified penetration rate (QPR) for autonomous vehicles. The penetration rate is devised by taking into account user demographics, land usage and road network properties. QPR is used as an input for heterogeneous urban traffic stream scenarios to calculate the aggregated urban traffic network dynamics of flow and density for the same network. Travel time versus flow characteristics is evaluated based on calibrated hyperbolic urban link travel time function for both interrupted and uninterrupted flows following the aggregated speed and density output from MFDs for heterogeneous traffic streams. Also, two scenarios are generated for comparison to explain the improvement in the network characteristics together with a sensitivity analysis.ResultsCompared to the base scenario there could be 25-35% of AVs on the road networks based on the analysis in coming fifteen years. This increment in usage impacts the capacity of road networks positively by increasing it up to 59%.ConclusionsResults obtained after the application of the suggested model approach to the real network can be used to define a realistic method for multi-vehicle equilibrium assignment models for heterogeneous traffic streams including autonomous vehicles instead of approximating the penetration rates.

What do drivers prefer? A field study on longitudinal and lateral driving automation

Vehicle automation aims to improve safety and comfort, but using partial driving automation safely can challenge drivers. To investigate how drivers trust and rely on the automation, we conducted a field study with 100 participants. While driving on a public highway, they rated their trust and commented on their interaction with the automation systems for longitudinal (adaptive cruise control, ACC) and lateral control (lane-keeping assistance, LA). Our results show that trust in partial driving automation increases continuously by using it. Comparing the automation systems, we found that drivers trust and use the ACC more than the LA. While driving, drivers express more liking statements about the ACC and more disliking and criticism statements about the LA. However, the number of mentioned interaction difficulties is higher for the ACC. We discuss possible reasons for the observed differences between the longitudinal (ACC) and lateral (LA) automation.

Driver Assisted Lane Keeping with Conflict Management Using Robust Sliding Mode Controller

Lane-keeping assistance design for road vehicles is a multi-objective design problem that needs to simultaneously maintain lane tracking, ensure driver comfort, provide vehicle stability, and minimize conflict between the driver and the autonomous controller. In this work, a cooperative control strategy is proposed for lane-keeping keeping by integrating driving monitoring, variable level of assistance allocation, and human-in-the-loop control. In the first stage, a time-varying physical driver loading pattern is identified based on a relationship between lateral acceleration, road curvature, and the measured maximum driver torque. Together with the monitored driver state that indicates driver mental loading, an adaptive driver activity function is then formulated that replicates the levels of assistance required for the driver in the next stage. To smoothly transition authority between various modes (from manual to autonomous and vice versa) based on the generated levels of assistance, a novel higher-order sliding mode controller is proposed and closed-loop stability is established. Further, a novel sharing parameter (which is proportional to the torques coming from the driver and from the autonomous controller) is used to minimize the conflict. Experimental results on the SHERPA high-fidelity vehicle simulator show the real-time implementation feasibility. Extensive experimental results provided on the Satory test track show improvement in cooperative driving quality by 9.4%, reduction in steering workload by 86.13%, and reduced conflict by 65.38% when compared with the existing design (no sharing parameter). These results on the cooperative performance highlight the significance of the proposed controller for various road transportation challenges.

Stability and [formula omitted] performance of human-in-the-loop control systems through hidden semi-Markov human behavior modeling

In this paper, a stochastic framework of analysis and synthesis is proposed for the stability and H∞ performance of linear human-in-the-loop (HiTL) control systems which involve both human’s discrete internal states and machine’s continuous-time dynamics. Initially, a hidden semi-Markov model (HS-MM) is used for the human behavior modeling, which takes into account the random nature of human internal state (HIS) reasoning and the sojourn time of HIS, as well as the uncertainty from HIS observation. Subsequently, by integrating human model, machine model, and their interaction, a hidden semi-Markov jump system (HS-MJS) model is constructed to describe the HiTL control system. By the tools of stochastic Lyapunov functional and linear matrix inequalities (LMIs), a sufficient and necessary condition for the stochastic stability of the HiTL control system via state feedback is then given on the basis of the HS-MJS model. Besides, an LMI-based human-assistance stabilizing control design is developed. For the prescribed disturbance attenuation level, a human-assistance H∞ controller synthesis method of the HiTL control system is proposed in terms of LMIs. Finally, the developed theoretical results are applied to the steering control of the lane-keeping assistance system.

Implementation Resource Allocation for Collision-Avoidance Assistance Systems Considering Driver Capabilities

Various collision-avoidance assistance (CAA) systems, such as automatic emergency braking (AEB) and lane-keeping assistance (LKA), have been developed in the last decades to enhance the active safety of ground vehicles. Meanwhile, more electronic computing units (ECUs) have been embedded inside a vehicle to support the diversified CAA systems, which complicate the automotive electrical/electronic architecture and increase the cost. Instead of adding <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">extra</i> ECUs, we propose to allocate the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">existing</i> implementation resources, i.e., the available processor time and memory space to the CAA systems, per individual driver’s maneuver capabilities. As an illustrative example, we first show that two drivers can exhibit distinct maneuvers in a pre-crash situation on highway, according to which they can be classified as either steering-oriented or braking-oriented. Then, we design two CAA systems: an AEB and an LKA, based on the ultra-local model predictive control method. Furthermore, we show that by adjusting the prediction horizons of the two controllers, the implementation resources can be allocated to the two CAA systems in different fashions, which yields three control modes: standard mode, steering-enhanced mode, and braking-enhanced mode. Finally, by comparing the control performance of each driver-type/control-mode pair through both CarSim-Simulink joint simulations and driver-in-the-loop simulator experiments, we demonstrate that by allocating more resources to compensate for the weakness of a driver’s maneuver, the CAA systems can provide enhanced driving safety by consuming the same overall amount of the implementation resources.

Tire Track Identification: A Method for Drivable Region Detection in Conditions of Snow-Occluded Lane Lines

&lt;div class="section abstract"&gt;&lt;div class="htmlview paragraph"&gt;Today’s Advanced Driver Assistance Systems (ADAS) predominantly utilize cameras to increase driver and passenger safety. Computer vision, as the enabler of this technology, extracts two key environmental features: the drivable region and surrounding objects (e.g., vehicles, pedestrians, bicycles). Lane lines are the most common characteristic extracted for drivable region detection, which is the core perception task enabling ADAS features such as lane departure warnings, lane-keeping assistance, and lane-centering. However, when subject to adverse weather conditions (e.g., occluded lane lines) the lane line detection algorithms are no longer operational. This prevents the ADAS feature from providing the benefit of increased safety to the driver. The performance of one of the leading computer vision system providers was tested in conditions of variable snow coverage and lane line occlusion during the 2020-2021 winter in Kalamazoo, Michigan. The results show that this computer vision system was only able to provide high confidence detections in less than 1% of all frames recorded. This is an alarming result, as 21% of all crashes in the U.S. are weather-related. To increase the capabilities of ADAS when snow-occlusions are present, a tire track identification system was developed by comparing various supervised machine learning models. A custom dataset was collected using the Energy Efficient and Autonomous Vehicles lab’s research platform from Western Michigan University. A data preparation pipeline was implemented to label tire tracks and train the machine learning models. The best model achieved high confidence detections of tire tracks in 83% of all frames of which tire tracks were present, an 82% increase in detections than the leading computer vision system provider.&lt;/div&gt;&lt;/div&gt;

Improving the performance of unsignalized t-intersections within CAVs mixed traffic

The rapid growth in population and the increase in the number of vehicles on the road have resulted in severe traffic congestion over the last two decades. However, intersections, where different flows intersect, are among the major cause of traffic congestion besides bottlenecks. Past decades have seen major technological advancements in road vehicles aimed at making vehicles traveling securely and comfortably. Current connected and automated vehicles (CAV) are packed with lane-keeping assistance and adaptive cruise control to ensure that vehicles do not collide and reduce traffic congestion. In this research, we developed a control algorithm that utilizes CAVs to help generate additional usable gaps for the minor road vehicles to enter the intersection without affecting the mainline traffic flow. Simulation results showed that the delay and queue length of the minor road approach is minimized without causing a significant delay to the mainline. The minor road delay was reduced by 72% when the percentage of CAVs on the major road is 70% compared to the benchmark with no CAVs on the major road.

Individualizable Vehicle Lane Keeping Assistance System Design: A Linear-Programming-Based Model Predictive Control Approach

Model predictive control (MPC), a control technique that can systematically and explicitly cope with system constraints, has become increasingly prevalent in the field of automotive engineering. Respecting the advanced driver-assistance systems (ADAS) design, the MPC has found its applications in various kinds of ground vehicular lane-keeping assistance systems (LKAS). In this paper, an MPC-based LKAS is synthesized by leveraging the linear programming (LP) methodology. Compared to the conventional MPC-based LKAS that is based on quadratic programming (QP), the LP alternative is less computationally demanding, which is desired for commercial vehicular electronic control units. Besides, the LP scheme enables the designer to formulate the MPC's performance criterion in the sense of L1/L∞ norms, which differs from the QP's L2-norm-based measure. The proposed LP-based MPC-LKAS is examined in CarSim-Simulink joint simulations and its performance is compared with a QP-based solution.