Driving Conditions Research Articles

Real-Time Driver State Detection Based on OpenCV and MediaPipe Libraries

Empowering street security requires being able to distinguish occupied driving. A real- time way to keep an eye on a driver's condition is to utilize driver-state diversion. We will screen varieties within the driver's eye and look developments by utilizing the eye-aspect proportion (EAR) and the look score. A dashboard-mounted camera gives us with video, which we record and prepare utilizing the OpenCV and Mediapipe modules. Deciding the driver's sign of languor is made simpler by the EAR. Discover out in the event that the drivers are occupied utilizing the look score. As a implies of instantly distinguishing and settling such issues, the framework persistently checks the driver's confront. With broad testing, we appear that our approach can effectively caution drivers, lessening the chance of collisions and progressing everyone's security when driving

Analysis of Influencing Factors of Level 3 Automated Vehicle Takeover: A Literature Review

Level 3 automated vehicles (L3 AVs) enable the driver to perform non-driving tasks, taking over in an emergency. In recent years, studies have extensively discussed the influencing factors of L3 AV takeovers. Extensive literature review shows that L3 AV takeovers are affected by human factors, traffic environment, and automatic driving systems. On this basis, this study proposes a conceptual framework of L3 AV takeovers. The main findings of this study include the following: (1) non-driving tasks, non-driving posture, individual characteristics, and trust have an impact on takeover behavior; (2) high traffic density, poor road geometry, and extreme weather have a negative impact on the takeover; (3) multimodal interaction design can improve collection performance. Although the existing research has made rich achievements, there are still many challenges. The influence of human factors on takeover performance is controversial, the quantification standard of takeover influencing factors is insufficient, and the prediction accuracy needs to be improved. It is suggested to refine the criteria of driver participation in NDRT, formulate an effective measurement standard of driver fatigue, and develop a takeover prediction model combining driver status and traffic environment conditions. It provides a research basis for the formulation of laws, infrastructure construction, and human–computer interaction design.

Advanced Necklace for Real-Time PPG Monitoring in Drivers.

Monitoring heart rate (HR) through photoplethysmography (PPG) signals is a challenging task due to the complexities involved, even during routine daily activities. These signals can indeed be heavily contaminated by significant motion artifacts resulting from the subjects' movements, which can lead to inaccurate heart rate estimations. In this paper, our objective is to present an innovative necklace sensor that employs low-computational-cost algorithms for heart rate estimation in individuals performing non-abrupt movements, specifically drivers. Our solution facilitates the acquisition of signals with limited motion artifacts and provides acceptable heart rate estimations at a low computational cost. More specifically, we propose a wearable sensor necklace for assessing a driver's well-being by providing information about the driver's physiological condition and potential stress indicators through HR data. This innovative necklace enables real-time HR monitoring within a sleek and ergonomic design, facilitating seamless and continuous data gathering while driving. Prioritizing user comfort, the necklace's design ensures ease of wear, allowing for extended use without disrupting driving activities. The collected physiological data can be transmitted wirelessly to a mobile application for instant analysis and visualization. To evaluate the sensor's performance, two algorithms for estimating the HR from PPG signals are implemented in a microcontroller: a modified version of the mountaineer's algorithm and a sliding discrete Fourier transform. The goal of these algorithms is to detect meaningful peaks corresponding to each heartbeat by using signal processing techniques to remove noise and motion artifacts. The developed design is validated through experiments conducted in a simulated driving environment in our lab, during which drivers wore the sensor necklace. These experiments demonstrate the reliability of the wearable sensor necklace in capturing dynamic changes in HR levels associated with driving-induced stress. The algorithms integrated into the sensor are optimized for low computational cost and effectively remove motion artifacts that occur when users move their heads.

Enhancing Road Safety: Fast and Accurate Noncontact Driver HRV Detection Based on Huber-Kalman and Autocorrelation Algorithms.

Enhancing road safety by monitoring a driver's physical condition is critical in both conventional and autonomous driving contexts. Our research focuses on a wireless intelligent sensor system that utilizes millimeter-wave (mmWave) radar to monitor heart rate variability (HRV) in drivers. By assessing HRV, the system can detect early signs of drowsiness and sudden medical emergencies, such as heart attacks, thereby preventing accidents. This is particularly vital for fully self-driving (FSD) systems, as it ensures control is not transferred to an impaired driver. The proposed system employs a 60 GHz frequency-modulated continuous wave (FMCW) radar placed behind the driver's seat. This article mainly describes how advanced signal processing methods, including the Huber-Kalman filtering algorithm, are applied to mitigate the impact of respiration on heart rate detection. Additionally, the autocorrelation algorithm enables fast detection of vital signs. Intensive experiments demonstrate the system's effectiveness in accurately monitoring HRV, highlighting its potential to enhance safety and reliability in both traditional and autonomous driving environments.

ARDUINO BASED ALCOHOL SENSE ENGINE LOCK USING GPS & GSM

This project introduces an innovative approach to improving road safety by developing an Arduino-based alcohol detection and engine lock system, integrated with GPS and GSM technologies. The main goal is to prevent driving under the influence of alcohol by disabling the vehicle's engine upon alcohol detection. The system employs an MQ-3 alcohol sensor to measure the driver's breath alcohol content. When alcohol is detected, the Arduino microcontroller processes the data from the sensor and activates the engine lock mechanism, thus preventing the vehicle from starting. This functionality is enhanced with GPS technology to provide precise location tracking of the vehicle. In the event of alcohol detection, the system utilizes the GSM module to send alerts to pre-configured contacts, supplying real-time information about the driver's condition and the vehicle's location. By incorporating GPS, the system ensures continuous location monitoring, adding an extra layer of security and enabling swift emergency response. The GSM module allows for immediate communication, dispatching SMS alerts to family members or authorities, thereby increasing the safety of both the driver and the public. This comprehensive solution aims to reduce incidents of drunk driving by leveraging Arduino for sensor interfacing and control, and combining it with GPS and GSM technologies for real-time tracking and communication. The proposed system is cost-effective, easy to implement, and scalable, offering a practical tool for enhancing road safety and saving lives. Key Words: Arduino Alcohol sensor, Engine lock, MQ- 3 sensor, GPS tracking, GSM module etc.

Design of a Multidimensional Intelligent Safety Driving Monitoring System Based on Machine Vision

This study introduces a multidimensional driving safety monitoring system that integrates millimeter-wave radar and an Arduino Uno-based alcohol sensor. The system aims to enhance driving safety and prevent accidents by real-time monitoring and analysis of the driving environment, driver's condition, and vehicle dynamics using machine vision technology. Utilizing millimeter-wave radar, the system precisely detects obstacles and pedestrians around the vehicle, facilitating efficient distance and speed measurements to support automatic emergency braking and collision warnings. Additionally, the Arduino Uno-based alcohol sensor monitors the driver's breath for alcohol concentration, ensuring the driver is in a suitable state for driving. Combined with machine vision technology, the system further analyzes video stream data from inside and outside the cockpit, monitoring driver behavior and external traffic conditions to enhance the identification of potential risks. The success of this research illustrates the potential of combining machine vision with advanced sensor technologies, offering new insights for the development of comprehensive driving assistance systems.

DRIVER DROWSINESS DETECTION SYSTEMS

Car accidents are frequently caused due to many reasons like drowsiness, drunkenness or tiredness, which has serious consequences for traffic safety. Numerous factors, such as fatigue, intoxication, or drowsiness, commonly contribute to car accidents, which have detrimental effects on traffic safety. Many fatal collisions might be avoided if sleepy drivers were warned beforehand. Many sleepiness detection technologies are available to identify and warn drivers of any indications of inattention while driving. Sensors in self-driving cars must be able to detect whether a driver is sleepy, agitated, or experiencing sharp emotional fluctuations. Thus far There have been numerous recommendations for enhanced warning systems to prevent sleepy driving. When these systems detect a significant lane departure, a change in the headway distance, or other indications of subpar driving, they sound an alarm. These systems rely heavily on machine learning algorithms, which allow them to learn and adjust to unique driving patterns and characteristics. In supervised learning, models are trained on labelled datasets to discriminate between awake and sleepy states. These approaches are widely used for classification tasks. Unsupervised learning techniques are also applied for anomaly detection, which finds departures from typical driving behaviour. The incorporation of cutting-edge sensors like facial recognition cameras and brainwave sensors, which offer more thorough insights into the driver's condition, has recently improved driver sleepiness detection systems. Moreover, these systems are more accurate and responsive when real-time data processing and cloud-based analytics are used. KEYWORDS Drowsiness-detection, Driver-Fatigue, Computer Vision, Real Time, survey, Eye Tracking Technology, Automotive Safety Systems, Driver Assistance Systems, Real-time Drowsiness Detection, Alertness Monitoring.

Development of an Integrated Longitudinal Control Algorithm for Autonomous Mobility with EEG-Based Driver Status Classification and Safety Index

During unexpected driving situations in autonomous vehicles, such as a system failure, the driver should take over control from the vehicles in SAE Level 3 to cope with unexpected situations. Therefore, it is necessary to develop reasonable takeover technologies to ensure safe driving. In this study, an electroencephalogram (EEG)-based driver status classification model and a safety index-based integrated longitudinal control algorithm considering the takeover time and driving characteristics are proposed. The driver status is classified into two states: road monitoring and non-driving-related tasks. EEG data are acquired while the driver performs certain tasks. The driver status classification model is presented using the EEG data based on a machine learning method. It is designed such that the desired takeover time is determined based on the classified driver state. To design the integrated longitudinal control algorithm, a safety index is designed and calculated based on the vehicle state and driver’s driving characteristics. The desired clearances based on the desired takeover time and driver characteristics are calculated and selected based on the safety index. A sliding-mode control algorithm is adopted to allow the vehicle to track the desired clearance reasonably. The performance of the proposed control algorithm is evaluated using the MATLAB/Simulink R2019a (Mathworks, Natick, Massachusetts, U.S.A) and CarMaker software 8.1.1 (IPG Automotive, Karlsruhe, Germany).

Abstract 2429: Longitudinal ctDNA monitoring using a high sensitivity tumor-informed assay in patients with metastatic HR+/HER2− breast cancer receiving endocrine therapy and CDK4/6 inhibitors

Abstract BACKGROUND: ctDNA burden is prognostic and its dynamics during therapy may permit innovative strategies in metastatic breast cancer (mBC). Previous analyses have focused on driver genes, with limited sensitivity to quantify depth of response or detect early molecular progression. To enable sensitive, quantitative ctDNA monitoring, we analyzed samples from a cohort of HR+/HER2− mBC pts treated with endocrine therapy (ET) and CDK4/6 inhibitors (CDKi) using RaDaR, a patient-specific assay that tracks tumor-informed truncal variants independent of driver status with a limit of detection of 0.001%. METHODS: HR+/HER2− mBC pts receiving standard ET + CDKi were enrolled in a prospective observational cohort from 2018. Plasma samples were collected at baseline (BL), within 30 days (d), and ~q3 months (mo) with radiological scans. RaDaR was performed on all samples for pts with tissue available for WES (as per assay). Clinico-pathological variables were collected. Endpoints were time to treatment failure (TTF) and overall survival (OS). RESULTS: 51 pts were available for analysis. Median age was 60 years (range 38-88), 63% had visceral disease, and most pts were treated as first (75%) or second (20%) line. RaDaR was successful on 43 pts, with the rest dropping out due to failed WES (1) or panel failure (7). Archival tissue WES was used to design bespoke assays targeting a median 48 (23-52) variants in 248 samples, with a median of 5 samples/pt (2-14) collected over a median follow up of 27.8 (0.9-64.3) mo. ctDNA was detected in 91% (39/43) of BL and 70% (174/248) of all samples, with a median estimated variant allele fraction (eVAF) of 0.05%. Higher BL eVAFs were associated with liver disease (p=0.02) and with shorter TTF (p<0.01 adjusted by age and visceral disease) and OS (p<0.01). 36/39 pts had ≥1 sample with ctDNA decrease from BL, occurring within the first mo of therapy in 77% (24/31) of pts with available early samples. In 13 pts ctDNA became undetectable (‘suppressed’) at a median time (T) of 157 d (14-343). Among 25 pts with treatment failure (TF), the median T from the last negative/decreasing eVAF to TF was 163 d (0-943). Only 3 pts with TF had prior negative ctDNA, occurring with a median lead T of 198 d (119-595). Among ongoing pts, the median T from the last ctDNA decrease to last follow up was 317 d (163-1508), and from the last negative ctDNA sample was 352 d (163-993). CONCLUSION: RaDaR enabled sensitive ctDNA monitoring in 91% HR+/HER2− mBC pts on ET + CDK4/6i. ctDNA levels were prognostic and fell rapidly with therapy, but suppression occurred in less than one-third of pts, taking a median of 5.2 mo. Pts with negative/decreasing ctDNA may require less frequent clinical/radiographic tumor evaluations. Highly sensitive ctDNA tracking may inform prognostic assessments, follow up strategies, and innovative interventional trial designs. Citation Format: Jesus Fuentes-Antras, Mitchell J. Elliott, Philippe Echelard, Aaron Dou, Zachary Veitch, Philippe L. Bedard, Eitan Amir, Michelle B. Nadler, Nicholas Meti, Nancy Gregorio, Elizabeth Shah, Emily Van de Laar, Celeste Yu, Lisa Gates, Clodagh Murray, Christopher G. Smith, Amber Chevalier, Lillian L. Siu, Hal K. Berman, David W. Cescon. Longitudinal ctDNA monitoring using a high sensitivity tumor-informed assay in patients with metastatic HR+/HER2− breast cancer receiving endocrine therapy and CDK4/6 inhibitors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2429.

Driver Drowsiness Detection System

The Driver Drowsiness Monitoring System (DDMS) stands as a crucial advancement in automotive safety, specifically designed to mitigate accidents stemming from drowsy driving. Employing a sophisticated blend of sensors and advanced algorithms, it continuously monitors the driver's state, identifying signs of drowsiness in real-time. Key elements encompass infrared cameras, facial recognition technology, steering angle sensors, and biometric sensors such as heart rate monitors, collectively offering a comprehensive insight into the driver's behavior and physiological condition. The DDMS software meticulously analyzes data from these sensors, evaluating parameters like eye movement, blink frequency, facial expressions, and steering patterns. Upon detecting signs of drowsiness, the system promptly issues audible and visual alerts, thereby averting potential accidents caused by driver fatigue. Its adaptability to diverse vehicles and driving conditions establishes it as an indispensable tool for bolstering road safety, effectively curbing drowsy driving-related incidents.

Evaluation of Traffic Accidents in the Province of Albay: Basis for Sustainable Traffic Management Framework

The traffic safety situation around the world is still not optimistic. Accidents occur due to a combination of factors and are rarely caused by a specific reason. Risk factors are a better explanation for why traffic accidents occur. This study determined the traffic accidents in the province of Albay. It identified the number of reported traffic accidents in the Province of Albay covering Calendar Year 2020 to 2022. It also determined the causes and risk factors of traffic accidents regarding driver's behaviors, road conditions, vehicle conditions, traffic signs, and traffic management. Moreover, it inferred a significant agreement among the ranks regarding the causes and risk factors of traffic accidents. It proposed a traffic management framework to prevent traffic accidents in the province of Albay. The respondents of this study are 60 personnel from the three CPS in Albay, 30 from the Local Government Unit, and 47 community, representing the riders and commuters in the Poblacion area. The study revealed a concerning prevalence of traffic accidents, underscoring a pressing public concern due to the high number of casualties and severity of outcomes. Despite this, road safety has not received sufficient attention, attributed to a lack of awareness regarding the problem's extent, associated costs, and effective interventions. Implementing a comprehensive traffic management framework involving collaboration among stakeholders is crucial for understanding and addressing the multifaceted causes of accidents, emphasizing the importance of community-wide initiatives and cooperation among various entities to enhance public awareness and promote responsible driving behavior.

Deciphering principles of nucleosome interactions and impact of cancer-associated mutations from comprehensive interaction network analysis.

Nucleosomes represent hubs in chromatin organization and gene regulation and interact with a plethora of chromatin factors through different modes. In addition, alterations in histone proteins such as cancer mutations and post-translational modifications have profound effects on histone/nucleosome interactions. To elucidate the principles of histone interactions and the effects of those alterations, we developed histone interactomes for comprehensive mapping of histone-histone interactions (HHIs), histone-DNA interactions (HDIs), histone-partner interactions (HPIs) and DNA-partner interactions (DPIs) of 37 organisms, which contains a total of 3808 HPIs from 2544 binding proteins and 339 HHIs, 100 HDIs and 142 DPIs across 110 histone variants. With the developed networks, we explored histone interactions at different levels of granularities (protein-, domain- and residue-level) and performed systematic analysis on histone interactions at a large scale. Our analyses have characterized the preferred binding hotspots on both nucleosomal/linker DNA and histone octamer and unraveled diverse binding modes between nucleosome and different classes of binding partners. Last, to understand the impact of histone cancer-associated mutations on histone/nucleosome interactions, we complied one comprehensive cancer mutation dataset including 7940 cancer-associated histone mutations and further mapped those mutations onto 419,125 histone interactions at the residue level. Our quantitative analyses point to histone cancer-associated mutations' strongly disruptive effects on HHIs, HDIs and HPIs. We have further predicted 57 recurrent histone cancer mutations that have large effects on histone/nucleosome interactions and may have driver status in oncogenesis.

Adaptive Intervention Algorithms for Advanced Driver Assistance Systems

Advanced driver assistance systems (ADASs) have recently gained popularity as they assist vehicle operators in staying within safe boundaries, helping them thereby to prevent possible collisions. However, despite their success and development, most ADAS use common and deterministic warning thresholds for all drivers in all driving environments. This may occasionally lead to the issuance of annoying inadequate warnings, due to the possible differences between drivers, the changing environments and driver statuses, thus reducing their acceptance and effectiveness. To fill this gap, this paper proposes adaptive algorithms for commonly used warnings based on real-time traffic environments and driver status including distraction and fatigue. We proposed adaptive algorithms for headway monitoring, illegal overtaking, over-speeding, and fatigue. The algorithms were then tested using a driving simulator. Results showed that the proposed adaptive headway warning algorithm was able to automatically update the headway warning thresholds and that, overall, the proposed algorithms provided the expected warnings. In particular, three or four different warning types were designed to distinguish different risk levels. The designed real-time intervention algorithms can be implemented in ADAS to provide warnings and interventions tailored to the driver status to further ensure driving safety.

Comprehensive Survey of Driver Drowsiness Systems

Driver monitoring systems have been improved over time as artificial intelligence and computer technology have advanced. Several experimental studies have collected real-world driver drowsiness data and used various artificial intelligence algorithms and feature combinations to dramatically improve the real-time effectiveness of these systems. This study presents an updated assessment of the driver sleepiness detection systems implemented over the last decade. In modern automobiles, assessing the driver's cognitive condition is an important aspect of passenger safety. The term cognitive state refers to a driver's mental and emotional state, which has a substantial impact on their ability to drive safely. Drivers' cognitive states may be altered by factors such as fatigue, distraction, stress, or disability. Intelligent automotive technology may be able to adapt and aid the driver by identifying varied conditions in real-time, reducing the frequency of accidents. The face, being an integral component of the body, communicates a significant quantity of information. The facial expressions, such as blinking and yawning patterns, exhibit changes in a driver when they are experiencing fatigue.

Exploring the interrelations and driving factors among typical ecosystem services in the Yangtze river economic Belt, China

Exploring the spatiotemporal characteristics of ecosystem services (ESs) and their drivers is crucial for managers to develop significant scientific policies that further sustainable development. We used the Yangtze River Economic Belt (YREB) to explore the trends, hotspots, and drivers of water yield (WY), soil conservation (SC), carbon sequestration (CS), and food supply (FS) between 2000 and 2020. Similarly, we analyzed relationships among ESs and drivers of the multiple ecosystem services landscape index (MESLI). We used the self-organizing map method to obtain the types and distribution of the ES bundles, revealing the bundles, trade-offs, and synergies among ESs. The four ESs had an increasing trend, with CS having the highest increase; ES hotspot analysis showed differences among upper, middle, and lower reaches. Constraint lines among ESs and drivers were diverse; the corresponding SC and WY reached thresholds when CS values were 1477.81 and 460.5 t km−2, respectively. When FS values were 67.34 and 86.17 × 104 Yuan·km−2, CS and WY reached their thresholds. All critical drivers of the four ESs were natural factors. The thresholds that the MESLI reached with driver status were 1000 mm (evapotranspiration), 2121 mm (precipitation), 2.42° (slope), 1.46% (soil organic matter), 36.08% (sand), 30.75% (proportion of non-agricultural population), 18.57% (cropland proportion), 1.05 × 104 persons·km−2 (population density), and 84.84% (proportion of non-agricultural industries in total gross domestic product), respectively. FS, water supply, and ecological conservation bundles changed over the 20 years, and trade-offs and synergies among ESs within bundles differed. We revealed the complexity of ESs from multiple perspectives, which will enable the development of ecosystem management and conservation recommendations for the YREB and large-scale economic zones worldwide.

The Effect of Diesel Fuel Additives on the Exhaust Gases of Agricultural Tractors During Plowing

The research aims to study the effect of the fuel type, plow types, and engine speeds on carbon monoxide (CO), hydrogen sulfide (H2S), and oxygen (O2) gases. The field experiments were conducted on an experimental farm in Al-Hamdaniya, Nineveh, Iraq. The investigation was factorial, using a Randomized Complete Block Design with Split –Split plot Design. The three types of fuel were placed in the main plot, the plow types in the secondary split plot, and the speed engine in the Split –Split action. The results showed significant differences between the fuel, plows, and engine speeds in most studied traits. There were no significant differences in the interaction between the plow types and the engine’s speed regarding CO emissions at the two levels. Still, the rate at 1000 rpm recorded the lowest emission value. At the two levels (the level of the exhaust nozzle - the status of the tractor driver’s seat), it amounted to 39.833-18.767 ppm, respectively. Moreover, the highest reading of H2S gas emissions was recorded with diesel without added fuel and the moldboard plow. Finally, the results showed no significant difference, as diesel fuel with super additives from Manol and an engine speed of 2000 rpm recorded the lowest emission rate for the emission rate of O2 gas, amounting to 20.550%.

Driving behavior assessment: A practical study and technique for detecting a driver's condition and driving style

New research confirms that driving mistakes account for almost 75 % of all traffic accidents. Recently, many researchers have studied the recognition of driving behavior, including aggressive, fatigued or drowsy driving. The principal aim of this work is to present a combination approach for evaluating driver behavior, focuses on the analysis of numerous driving-related parameters and the detection of driver fatigue and drowsiness. Experimental research is carried out using a monitoring unit which can be mounted in any vehicle fitted with CAN (Control Area Network) technology, and which monitors current driving actions. This work combines the analysis of driving parameters such as engine speed, vehicle speed, accelerator pedal position, steering wheel angle, engine noise intensity, fuel consumption and, finally, the quantity of exhaust gases generated to define whether driving is normal or aggressive. Analysis of the driver's face enables us to detect fatigue and know whether the driver is asleep on the basis of facial features. This proposed detection method achieved an average accuracy of 99.10 %. The information gathered for each driving style is recorded, in order to build up our own data and analyze it carefully. In-depth study of these parameters reveals the boundaries that define driving status and style. Experimental and simulation results confirm that the proposed system can effectively detect driving states and driver reactions, with the aim of guaranteeing a sufficient and acceptable level of safety. This study can be used to improve policies and design more robust driver training and education programs.

Perancangan Sistem Pemantauan Kelelahan Driver Berbasis IoT (Internet Of Things) Yang Adaptif Untuk Transportasi Makanan Segar

In the logistics industry, fresh food transportation requires effective management of driver fatigue to maintain safety and service quality. With the increasing digital connectivity, Internet of Things (IoT) technology has become a promising solution to monitor driver fatigue in real-time. This research aims to design an adaptive IoT-based driver fatigue monitoring system specifically for fresh food transportation. Through a case study in the logistics industry, we identified specific challenges and needs in driver fatigue management. Based on the identification results, we developed a system that utilizes IoT sensors integrated in vehicles and drivers' smartphones. The system automatically monitors the driver's physical condition and behavior and provides real-time alerts if signs of fatigue are detected. With an adaptive approach, the system can customize recommendations and interventions according to the measured fatigue level. The results of the case study show that the designed system is able to reduce the risk of driver fatigue, increase transportation safety, and improve service quality in fresh food transportation. This research contributes to the practical understanding of the implementation of IoT technology in driver fatigue management, particularly in the context of the logistics and fresh food transportation industries.

Wearable and self-powered triboelectric sensors based on NaCl/PVA hydrogel for driver multidimensional information monitoring

Dangerous behaviors during driving such as fatigue or making phone calls may seem common in daily life. However, these behaviors are the actual "culprit" of many traffic accidents and pose a serious threat to traffic safety. Therefore, it is necessary to effectively detect dangerous driving behaviors by scientific and technological means. The present detection methods still face many bottlenecks, including individual differences, complex lighting changes, battery power supply, poor wearability, et al. Here, we designed a green and stretchable triboelectric sensor (TES) to monitor dangerous driving behaviors. The performance of the sensor was improved by about 9.99 and 3.58 times through doping sodium chloride solution in PVA hydrogel and introducing a curved contact surface between the electrode and friction layer, respectively. The proposed sensor has a high sensitivity of 1.95 V/kPa in the linear range of 0–11.28 kPa. By employing different machine learning models, we developed an intelligent neck ring based on the proposed sensor array to recognize different neck movements, which has achieved the highest accuracy of 96.10%. Finally, the intelligent neck ring was used to construct a sensing system for driver status monitoring. By collecting detailed driver information, the system can detect dangerous driving behaviors, monitor drivers’ health conditions, and provide appropriate reminders to improve driving safety and prevent the spread of viruses.

A multilayer stacking method base on RFE-SHAP feature selection strategy for recognition of driver’s mental load and emotional state

The driver state monitoring is becoming one of the research hotspots in the field of traffic and vehicle safety, which can ensure driving safety by monitoring the driver’s state. Therefore, this work makes an attempt to recognize driver's mental load and emotional states. However, the reliability and accuracy of driver status detection largely depend on the extracted features and the detection algorithm. The existing methods mainly improve accuracy by increasing the number of features, but for the problem with limited training samples, the increase in features may lead to an increase in model variance, and resulting in overfitting. In addition, single-model based methods may fall into local optimal solutions when performing local search. To alleviate these, we firstly propose an RFE-SHAP algorithm that improves the recursive feature elimination algorithm by using the Shapley Additive exPlanning value, and paired it with the eXtreme Gradient Boosting (XGBoost) algorithm to screen out the subset of features that best represent the driver's mental load and emotional state. Secondly, we proposed a multilayer stacking ensemble learning method based on different optimal feature subsets to further improve the accuracy of the driver’s mental load and emotional state recognition. For each base models used for integration learning, the corresponding optimal feature subsets are selected using the RFE-SHAP algorithm, and then the different optimal feature subsets are combined with multilayer stacking ensemble learning. The validation via experimental data demonstrates that the XGBoost-RFE-SHAP algorithm achieves superior model performance with fewer feature combinations than the classic algorithm. The proposed driver state detection model based on multiple optimum feature subsets and multilayer stacking integration methods would more accurately identify the driver's mental load and emotional state than the single model, commonly used machine learning and integrated learning algorithms.