Driver Warning System Research Articles

The prototype design and tests of vibration controlled driver warning system from the steering wheel

In this study, drivers are warned according to detected adverse driving conditions by miniature vibration motors positioned on the steering wheel by obtained information from sensors such as acceleration sensor placed in the center gravity of vehicle, the proximity sensor placed on the rear bumper and rain sensor placed in fender. For this purpose, two different electronic circuit boards are used including firstly center gravity placed in vehicle that has collected signals from sensors and secondly center steering wheel that controlled the miniature vibration motors placed on the steering wheel cover. Wireless communication was established between the two electronic circuit boards. Tests was carried out in real road conditions; it was determined that drivers are warned on steering wheel with vibrations at the different durations (180 Hz frequency) according to the scenario developed for adverse driving conditions using accelerometer sensor, proximity sensor and the rain sensor. It was seen that the developed prototype can be used on available vehicles with small modifications, with a simple montage for a vehicle in the production process.

Fuzzy Neural Network based Real-Time Driver Fatigue Detection and Warning System using EEG Waves

Driving fatigue has been considered as a significant risk factor in transportation accidents, and the development of the human cognitive State based on Electro Encephalo Gram (EEG) has become a major focus in the field of driving safety for many drivers. In adaptive online prediction framework which is capable of capturing subject-specific predictive patterns autonomously by constructing a subject-specific pattern library, based on which a probabilistic prediction rule is established. A real-time wireless Electro Encephalo Gram (EEG) based Brain Computer Interface (BCI) system for drowsiness detection has been proposed to reduce the time delay. A new method for driver fatigue detection based on EEG waves using Recurrent Self-Evolving Fuzzy Neural Network (RSEFNN)is proposed. The model uses Mindwave to collect the attention waves of EEG from the left prefrontal lobe of the subject, and measures the data when the subject is in the State of concentration, relaxation, fatigue and sleep. The Mindwave device and Matlab is interfaced using the Think Gear Library module. The correlation coefficient of the driver attention is classified to detect driver fatigue. Thus the system using brain wave sensor warns the driver when he/she feels drowsy.

Pavement distress detection and avoidance for intelligent vehicles

Pavement distresses and potholes represent road hazards that can cause accidents and damages to vehicles. The latter may vary from a simple flat tyre to serious failures of the suspension system, and in extreme cases to collisions with third-party vehicles and even endanger passengers' lives. The primary scientific aim of this study is to investigate the problem of road hazard detection for driving assistance purposes, towards the final goal of implementing such a technology on future intelligent vehicles. The proposed approach uses a depth sensor to generate an environment representation in terms of 3D point cloud that is then processed by a normal vector-based analysis and presented to the driver in the form of a traversability grid. Even small irregularities of the road surface can be successfully detected. This information can be used either to implement driver warning systems or to generate, using a cost-to-go planning method, optimal trajectories towards safe regions of the carriageway. The effectiveness of this approach is demonstrated on real road data acquired during an experimental campaign. Normal analysis and path generation are performed in post-analysis. This approach has been demonstrated to be promising and may help to drastically reduce fatal traffic casualties, as a high percentage of road accidents are related to pavement distress.

Robust representation and recognition of facial emotions using extreme sparse learning.

Recognition of natural emotions from human faces is an interesting topic with a wide range of potential applications, such as human-computer interaction, automated tutoring systems, image and video retrieval, smart environments, and driver warning systems. Traditionally, facial emotion recognition systems have been evaluated on laboratory controlled data, which is not representative of the environment faced in real-world applications. To robustly recognize the facial emotions in real-world natural situations, this paper proposes an approach called extreme sparse learning, which has the ability to jointly learn a dictionary (set of basis) and a nonlinear classification model. The proposed approach combines the discriminative power of extreme learning machine with the reconstruction property of sparse representation to enable accurate classification when presented with noisy signals and imperfect data recorded in natural settings. In addition, this paper presents a new local spatio-temporal descriptor that is distinctive and pose-invariant. The proposed framework is able to achieve the state-of-the-art recognition accuracy on both acted and spontaneous facial emotion databases.

Sistema de Alerta al Conductor Basado en Realimentación Vibro-Táctil

This paper proposes the design and build of a driver warning system, based on vibro-tactile feedback for preventing accidents through the generation of tactile stimuli. The system has two vibro-tactile grips devices which are easily mounted on the steering wheel of any vehicle and an electronic system based on location sensing as well as inter-vehicles communication, from which a risk level is computed on line in order to warn the driver about dangerous situations and risk zones. A model focalized on human factors is proposed and it is employed to justify the advantages of using artificial stimuli. Besides, the hardware, communication between vehicles and embedded software, are described too. Finally, experiences using the device in a racing car simulator and tests using two ordinary cars are shown.

Research of Fatigue Driving Warning System Based on Android

This paper proposed a method for detecting fatigue detection Android smart phone system, and applied to the Android system. The system monitors the state of fatigue by smart phones photographs. Face detection method is used to localize the eyes of the driver and the eye region is extracted to monitor the movement of eyelids. An alarm rule is designed based on the PERCLOS standard to detect drowsy driving.Experiments show that the method is accurate rate, running speed, and can be used to monitor driver fatigue during the day.

Assessing Response Time Effects of Forward Collision Warning HMIs with Central Tendency, Variance, and Dominance Tests

Reaction times (RTs) to driver warning system human-machine interface (HMI) designs are often used to assess their impact on driver detection performance. This paper introduces three different ways to statistically compare alternative HMIs in terms of RTs. The most commonly reported way is to compare central tendencies such as RT means or medians. A second approach, less commonly used, compares the variability in RT distributions associated with different HMIs as an assessment of the consistency of response effects elicited by an HMI across participants. The third method is to compare the RT distributions from alternative HMIs by means of ordinal dominance statistics. This assessment directly addresses the question: “Do people tend to respond faster with HMI A than HMI B (or vice versa)?” These alternative assessments are applied to RT data from a simulator study of several alternative forward collision warning (FCW) HMIs and the insights provided by each are discussed.

Tyre contact length on dry and wet road surfaces measured by three-axial accelerometer

We determined the tyre contact length on dry and wet roads by measuring the accelerations of the inner liner with a three-axial accelerometer. The influence of the tyre pressure, driving velocity, and tread depth on the contact length was studied in both types of road surface conditions. On dry asphalt the contact length was almost constant, regardless of the driving velocity. On wet asphalt the presence of water could be detected even at low driving velocities (e.g. 20km/h for a worn tyre) as the contact length began to decrease from that found in the dry asphalt situation. In addition to improving the performance of active safety systems and driver warning systems, the contact length information could be beneficial for classifying and studying the aquaplaning behaviour of tyres.

Task analysis of transit bus drivers’ left-turn maneuver: Potential countermeasures for the reduction of collisions with pedestrians

Left-turn maneuvers by transit buses are associated with a particularly high rate of collisions with pedestrians. The primary objective of the current study was, through task analysis, to gather a detailed outline of all mental and perceptual tasks that transit bus drivers complete when making a left turn at a signalized intersection. As a secondary objective, based on the task analysis examination, we proposed several countermeasures for the reduction of collision between transit buses and pedestrians. The task analysis revealed that transit bus drivers engage in various tasks during all stages of a left-turn maneuver. The results also showed that the task demand is at a peak during the performance of one of the most critical tasks – detecting pedestrians. Non-driving tasks that transit bus drivers are required to perform (e.g., collecting fares, providing directions to passengers) further increase the tasks demands. Based on the results of the task analysis, we proposed several potential solutions for the reduction of collisions between transit buses and pedestrians. The proposed countermeasures included a technology-based (collision warning system for both drivers and the pedestrians) and infrastructure-based (protected left turn signal for buses only) solutions, as well as examination of individual factors (e.g., stress coping ability) and the relationship with different aspects of driving performance.

Preliminary Evaluation of Drivers’ Responses to a Roadside Animal Warning System

The problem of vehicle-animal crashes is one of the few areas in which the number of incidents has been climbing steadily in recent years. Many strategies and technologies have been investigated over the years to try to reduce the number of vehicle-animal crashes, and each have had their own strengths and shortcomings. Thus far, no single strategy or technology has proven to be an effective solution. In this project, a 0.6 mi (1 km) site in Northern California, along State Route 3, near the city of Fort Jones, was selected based on a high frequency of vehicle-deer collisions to test a new animal detection and driver warning system. An animal detection system used microwave break-a-beam technology to detect objects along side or crossing the roadway, and warnings were provided to drivers using dynamic, LED, warning signs. These signs were typically blank in the off position and were only illuminated when the animal detection system detected a beam break. Coupled with the animal detection and driver warning system was a traffic monitoring system and data collection system, and a total of 10 months of data were collected between August 2011 and April of 2012. An analysis of the data showed that when illuminated, the warning signs reduced the traffic speeds by an average of 3.1 mph (5 km/h) to 5.1 mph (8.2 km/h), depending on the analysis method.

An integrated driver warning system for driver and pedestrian safety

A semi-integrated system for driver assistance and pedestrian safety is presented. This system is composed of a single camera which focuses on the driver for picking up visual cues and a stereo rig that focus on the road ahead for the detection of road obstructions and pedestrians. While the car is in motion, the driver's viewing direction is obtained and analyzed along with information of road condition and any moving vehicle ahead in order to determine if the current driving condition is safe. In addition, when the vehicle is moving slowly, the system can also detect the existence of a pedestrian ahead and warns the driver if the pedestrian moves in front of the car. This system contains algorithm-based safety analysis as well as fuzzy rules-based analysis for interaction between variables. Our experimental results show that the condition for driver safety can be accurately classified in 94.5% of the tested driving conditions, and the pedestrians can be identified in 93.18% of the tested cases. These were compared to the results of similar systems and shown to be superior.

Vehicle Dynamics Approach to Driver Warning

This paper discusses a concept for enhanced active safety by introducing a driver warning system based on vehicle dynamics that predicts a potential loss of control condition prior to stability control activation. This real-time warning algorithm builds on available technologies such as the Electronic Stability Control (ESC). The driver warning system computes several indices based on yaw rate, side-slip velocity, and vehicle understeer using ESC sensor suite. An arbitrator block arbitrates between the different indices and determines the status index of the driving vehicle. The status index is compared to predetermined stability levels which correspond to high and low stability levels. If the index exceeds the high stability level, a warning signal (haptic, acoustic, or visual) is issued to alert the driver of a potential loss of control and ESC activation. This alert will remain in effect until the index is less than the low stability level at which time the warning signal will be terminated. A vehicle speed advisory algorithm is integrated with the warning algorithm to provide a desired vehicle speed of a vehicle traveling on a curve. Simulation results and vehicle tests were conducted to illustrate the effectiveness of the warning algorithm.

AR계수와 SVM을 이용한 뇌파 기반 운전자의 졸음 감지 시스템

운전 중 운전자의 졸음은 교통 사망사고를 일으키는 중요한 요인이며 음주운전보다도 더 위험할 수 도 있다. 이러한 이유로 운전자의 졸음을 판별하고 경고하는 시스템 개발이 최근에 매우 중요한 이슈로 떠올랐다. 그중에서도 졸음과 가장 밀접한 관련이 있는 생체 신호 분석이 많이 적용되는데 그중에서도 뇌파(Electroencephalogram, EEG)와 안구전도(Electrooculogram, EOG)를 분석하는 연구가 주류를 이루고 있다. 본 논문에서는 실험 프로토콜를 바탕으로 측정된 뇌파를 주파수별로 분석하여 운전자의 상태별 뇌파 데이터베이스를 구축하였고 선형예측(Linear Predictive Coding, LPC) 계수와 Support Vector Machine(SVM)을 이용한 운전자 졸음 감지 시스템을 제안한다. 실험결과로 졸음의 뇌파분석에서 알파파가 감소하며 세타파가 증가하는 추세를 보였으며, LPC 계수가 각성, 졸음 및 수면상태의 특징을 잘 반영하였다. 특히 제안한 시스템은 적은 샘플(250ms)에서도 96.5%의 높은 분류 결과를 얻어 짧은 순간에 일어날 운전시 돌발 상황을 실시간으로 예측할 수 있는 가능성을 보였다. One of the main reasons for serious road accidents is driving while drowsy. For this reason, drowsiness detection and warning system for drivers has recently become a very important issue. Monitoring physiological signals provides the possibility of detecting features of drowsiness and fatigue of drivers. One of the effective signals is to measure electroencephalogram (EEG) signals and electrooculogram (EOG) signals. The aim of this study is to extract drowsiness-related features from a set of EEG signals and to classify the features into three states: alertness, drowsiness, sleepiness. This paper proposes a drowsiness detection system using Linear Predictive Coding (LPC) coefficients and Support Vector Machine (SVM). Samples of EEG data from each predefined state were used to train the SVM program by using the proposed feature extraction algorithms. The trained SVM program was tested on unclassified EEG data and subsequently reviewed according to manual classification. The classification rate of the proposed system is over 96.5% for only very small number of samples (250ms, 64 samples). Therefore, it can be applied to real driving incident situation that can occur for a split second.

The application of the random regret minimization model to drivers’ choice of crash avoidance maneuvers

This study explores the plausibility of regret minimization as behavioral paradigm underlying the choice of crash avoidance maneuvers. Alternatively to previous studies that considered utility maximization, this study applies the random regret minimization (RRM) model while assuming that drivers seek to minimize their anticipated regret from their corrective actions. The model accounts for driver attributes and behavior, critical events that made the crash imminent, vehicle and road characteristics, and environmental conditions. Analyzed data are retrieved from the General Estimates System (GES) crash database for the period between 2005 and 2009. The predictive ability of the RRM-based model is slightly superior to its RUM-based counterpart, namely the multinomial logit model (MNL) model. The marginal effects predicted by the RRM-based model are greater than those predicted by the RUM-based model, suggesting that both models should serve as a basis for evaluating crash scenarios and driver warning systems.

Automatic drowsiness detection system using autoregressive coefficients and neural network

One of the main reasons for serious road accidents is driving while drowsy. For this reason, drowsiness detection and warning system for drivers has recently become a very important issue. Monitoring physiological signals provides the possibility of detecting features of drowsiness and fatigue of drivers. One of the effective signals is to measure electroencephalogram (EEG) signals. The aim of this study is to extract drowsiness-related features from a set of EEG signals and to classify the features into three states: alertness, drowsiness, sleepiness. This paper proposes a neural-network-based drowsiness detection system using Autoregressive (AR) coefficients as feature vectors and Multi-Layer Perceptron (MLP) as a classifier. Specifically, the proposed method estimates AR coefficients using EIV (Errors-In Variables) providing an accurate estimation in a noisy process and linear predictive coding (LPC) analysis not considering noise. Samples of EEG data from each predefined state were used to train the MLP program by using the proposed feature extraction algorithms. The trained MLP program was tested on unclassified EEG data and subsequently reviewed according to manual classification.

Evaluating the Impact of Processing Spoken Words on Driving

The potential impact on driving of the processing of a single, auditorily presented word is analyzed in this work. Because driving is a complex cognitive activity that involves the integration and coordination of multiple subprocesses, the authors narrowed the scope of the research to concentrate on one critical task involved in driving: driver braking response. If two tasks have to be performed concurrently and both of them require access to a capacity-limited system, then performance in one or both of the tasks will dramatically worsen because the two processes will compete for access to cognitive resources. It has been shown that both word recognition and driving require central resources; therefore, these tasks are likely to interfere with each other. In the experiments, participants were required to perform two tasks during simulated driving. In the word recognition task, participants had to categorize auditorily presented words. In the braking task, participants depressed a brake pedal in response to the lead car's brake lights. The interval of time between the onset of the tasks’ stimuli was varied. Braking responses were substantially slower as the overlap between tasks increased. This finding demonstrates that the processing of a single word hinders driving performance. The experiments carried out have significant implications in the field of road safety. Many situations, such as cell phone ringing, cell phone conversations, auditory tips from navigation systems, and auditory alerts from driver warning systems, are similar to those studied. The experiments suggest that all these situations can negatively affect a driver's response time, increasing the likelihood of near misses and accidents.

Haptic Seat Interfaces for Driver Information and Warning Systems

The implementation of haptic interfaces in vehicles has important safety and flexibility implications for lessening visual and auditory overload during driving. The present study aims to design and evaluate haptic interfaces with vehicle seats. Three experiments were conducted by testing a haptic seat in a simulator with a total of 20 participants. The first experiment measured reaction time, subjective satisfaction, and subject workloads of the haptic, visual, and auditory displays for the four signals primarily used by vehicle navigation systems. The second experiment measured reaction time, subjective satisfaction, and subjective workloads of the haptic, auditory, and multimodal (haptic + auditory) displays for the ringing signal used by in-vehicle Bluetooth hands-free systems. The third experiment measured drivers' subjective awareness, urgency, usefulness, and disturbance levels at various vibration intensities and positions for a haptic warning signal used by a driver drowsiness warning system. The results indicated that haptic seat interfaces performed better than visual and auditory interfaces, but the unfamiliarity of the haptic interface caused a lower subjective satisfaction for some criteria. Generally, participants showed high subjective satisfaction levels and low subjective workloads toward haptic seat interfaces. This study provided guidance for implementing haptic seat interfaces and identified the possible benefits of their use. It is expected that haptic seats implemented in vehicles will improve safety and the interaction between driver and vehicle.

The Development of Parameters and Warning Algorithms for an Intersection Bus-Pedestrian Collision Warning System

This study presents the conceptual design of an intersection bus-pedestrian collision warning system for bus drivers approaching an intersection. The basic parameters of the proposed design concept include the bus drivers’ perception-reaction time, the emergency deceleration rate of the bus, and pedestrian walking speed. A bus driving simulation was designed and conducted to analyze bus drivers’ responses to unexpected pedestrians crossing unsignalized intersections or signalized intersections during a green light interval for parameter analysis. The timings of auditory warnings and visual warnings, the locations for vehicle detectors and pedestrian detectors, and the locations for visual warning devices were also developed after analyzing the experimental results. The experimental results also highlight some important characteristics of bus driving behavior at intersections. Moreover, bus drivers really pay attention to the warning messages. Finally, this study develops and discusses some warning algorithms.

運転状況を推定し学習データを選択する適応的ブレーキ警報システム

The paper proposes a rear-end collision warning system for drivers, where the collision risk is adaptively set from driving signals. The system employs the inverse of the time-to-collision with a constant relative acceleration as the risk and the one-class support vector machine as the anomaly detector. The system also utilizes brake sequences for outliers detection. When a brake sequence has a low likelihood with respect to trained hidden Markov models, the driving data during the sequence are removed from the training dataset. This data selection is confirmed to increase the robustness of the system by computer simulations.

가속도센서를 이용한 운전패턴 인식기법

In this paper, we proposed an algorithm to detect aggressive driving status by analysing six kinds of driving patterns, which was achieved by comparing for the feature vectors using mahalanobis distance. The first step is to construct feature matrix of size using frequency response of the time-series accelerometer data. Singular value decomposition makes it possible to find the dominant eigenvalue and its corresponding eigenvector. We use the eigenvector as the feature vector of the driving pattern. We conducted real experiments using three drivers to see the effects of recognition. Although there exists differences from individual drivers, we showed that driving patterns can be recognized with about 80% accuracy. Further research topics will include the development of aggressive driving warning system by improving the proposed technique and combining with post-processing of accelerometer signals.