Driving Simulation Software Research Articles

Driverless holographic simulation test integrated environment design

Abstract Aiming at the design problem of simulation test integrated environment for unmanned car, starting from the four aspects of unmanned simulation test [1], real scene quiz, scene analysis and layout, and data visualization and processing, through the combination of software and hardware, the use of various types of sensors, controllers, communication equipment, etc. for collaborative management of unmanned car driving, and the use of intelligent networked cars to collaboratively verify the multi-class algorithms and driving programs, and continuously improve the Overall environment design [2] [3], the integrated environment design is getting better and better. This overall environment is tested based on carla’s Apollo autonomous driving simulation software, and the test results show that the recognition accuracy rate reaches 87% and the simulation rate reaches 91%, etc. It shows that the unmanned holographic simulation test integrated environment technology program is feasible.

A Dynamic-Bayesian-Network-Based Approach to Predict Immediate Future Action of an Intelligent Agent

Predicting immediate future actions taken by an intelligent agent is considered an essential problem inhuman-autonomy teaming (HAT) in many fields, such as industries and transportation, particularly toimprove human comprehension of the agent as their non-human counterpart. Moreover, the results of suchpredictions can shorten the human response time to gain control back from their non-human counterpartwhen it is required. An example case of HAT that can be benefitted from the action predictor is partiallyautomated driving with the autopilot agent as the intelligent agent. Hence, this research aims to develop anapproach to predict the immediate future actions of an intelligent agent with partially automated drivingas the experimental case. The proposed approach relies on a machine learning method called naive Bayesto develop an action classifier, and the Dynamic Bayesian Network (DBN) as the action predictor. Theautonomous driving simulation software called Carla is used for the simulation. The results show that theproposed approach is applicable to predict an intelligent agent’s three-second time-window immediate futureaction.

A Federated Learning Monitoring Tool for Self-Driving Car Simulation (Student Abstract)

We propose CARLA-FLMon, which can monitor the progress of running federated learning (FL) training in the open-source autonomous driving simulation software, CARLA. The purpose of CARLA-FLMon is to visually present the status and results of federated learning training, and to provide an extensible FL training environment with which FL training can be performed repeatedly with updated learning strategies through analysis. With CARLA-FLMon, we can determine what factors have positive or negative influences on learning by visualizing training data. Then, we can optimize the parameters of the FL training model to improve the accuracy of FL. With preliminary experiments of CARLA-FLMon on lane recognition, we demonstrate that CARLA-FLmon can increase the overall accuracy from 80.33% to 93.82% by identifying lowly-contributing clients and excluding them.

Research on safety of the intended functionality of automobile AEB perception system in typical dangerous scenarios of two-wheelers

The statistical analysis was conducted on data of accident scenarios between cars and two-wheelers from National Automobile Accident In-depth Investigation System (NAIS) database in order to study safety of intended functionality of Autonomous emergency braking (AEB) perception system in typical dangerous scenarios of cars and two-wheelers. 11 scenario-related variables were selected, and 6 types of typical scenarios were obtained through cluster analysis and manual classification. The 6 types of typical scenarios were built by the automatic driving simulation software PreScan, and the AEB longitudinal control algorithm was built in Matlab/Simulink. Batch simulation script files were written, and the relative location distribution of car and two-wheeler with different time to collision (TTC) was obtained by batch simulation. Furthermore, the effects of car velocity, two-wheeler velocity and cyclist casualties on the parameter configuration of the perception system were analyzed. Under the premise of satisfying safety of intended functionality of the perception system, the optimal sensor detection scheme at different TTCs was obtained by comprehensively considering the death accident detection rate, detection area, and standard deviation. The results show that when the detection rate is 90%, the AEB system can adopt the detection scheme of long-range radar and short-range radar. The field of view (FOV) and detection range of the short range radar are 133.6° and 38.1 m, and those of the long range radar are 84.5° and 74.3 m. And when the detection rate was close to 100%, a single sensor can be used, and the detection parameters are 150° and 77.6 m. It provides reference for parameter optimization of AEB perception system.

Analysis of hazard perception characteristics based on driving behavior considering overt and covert hazard scenarios.

The drivers’ hazard perception plays an important role in preventing and reducing the occurrence of traffic accidents. In order to explore the drivers’ hazard perception and their behavioral characteristics in overt and covert hazards, hazardous events of three traffic conflict types (vehicle to vehicle, vehicle to cyclist and vehicle to pedestrian) were designed for overt and covert hazards based on the UC-win/Road driving simulation software, respectively. 35 drivers were organized to conduct the driving simulation tests. The data of driving behavior was collected when they were driving. A comparative analysis of drivers’ hazard perception ability and driving behavior characteristics was carried out for hazardous scenarios and traffic conflict types. The result has shown that drivers are more likely to take slowing measures or brake earlier in overt hazard scenarios to ensure safe driving. And drivers are more likely to be involved in collisions in covert hazard scenarios. The types of traffic conflict have a significant effect on the hazard perception ability of drivers (F = 5.92, p < 0.01). Drivers have the strongest hazard perception for cyclists and the weakest hazard perception for pedestrians. Traffic conflict types has a significant effect on drivers’ average braking depth (F = 32.31, p < 0.01), average speed (F = 13.78, p < 0.01), and average acceleration (F = 9.26, p < 0.01).

Classifying Driving Fatigue by Using EEG Signals.

Fatigue driving is one of the main reasons for the occurrence of traffic accidents. Brain-computer interface, as a human-computer interaction method based on EEG signals, can communicate with the outside world and move freely through brain signals without relying on the peripheral neuromuscular system. In this paper, a simulation driving platform composed of driving simulation equipment and driving simulation software is used to simulate the real driving process. The EEG signals of the subjects are collected through simulated driving, and the EEG of five subjects is selected as the training sample, and the remaining one is the subject. As a test sample, perform feature extraction and classification experiments, select any set of normal signals and fatigue signals recorded in the driving fatigue experiment for data analysis, and then study the classification of driver fatigue levels. Experiments have proved that the PSO-H-ELM algorithm has only about 4% advantage compared with the average accuracy of the KNN algorithm and the SVM algorithm. The gap is not as big as expected, but as a new algorithm, it is applied to the detection of fatigue EEG. The two traditional algorithms are indeed more suitable. It shows that the driver fatigue level can be judged by detecting EEG, which will provide a basis for the development of on-board, real-time driving fatigue alarm devices. It will lay the foundation for traffic management departments to intervene in driving fatigue reasonably and provide a reliable basis for minimizing traffic accidents.

Analysis on hazard perception ability of drivers in plateau areas: by different elevations

The low-pressure and low-oxygen environment of plateaus results in a greater workload for drivers, which contributes to the serious road safety situation in plateau areas in China. This study conducted four hazard perception experiments in Nanjing (50 m above sea level (asl)), Nyingchi (2995 m asl), Lhasa (3650 m asl), Nagqu (4460 m asl), and Yanghu Scenic Spot (4998 m asl) using UC-WIN/ROAD driving simulation software. A total of 31 drivers (23 males, 8 females) were recruited in this study, with a mean age of 28.0 years old and mean driving experience of 6.5 years. A bivariate correlation test was adopted to analyze the impacts of the elevation, age, acclimation period, gender, and driving experience on the perception–reaction time. Then, the drivers were divided into three groups (Good, Medium, and Poor) using K-means clustering. Finally, the marginal effects of the linear regression models for the three groups were calculated to augment the comparison. As expected, the elevation showed a positive correlation with the perception–reaction time but indicated a variance in influence effects in the three groups. Consistent with previous evidence, experienced drivers had better perception ability in the plateau. These findings could help in the design of high-elevation roads and guide the definition of suitability to drive in the plateau, thereby improving driving safety.

Using Simulation-software-generated Animations to Investigate Attitudes Towards Autonomous Vehicles Accidents

Road accidents involving autonomous vehicles are inevitable and have the potential to damage the public's confidence in the technology and ultimately result in its disuse. It's important to understand how people react to such incidents and the influencing factors of blame attribution and trust restoration. Research in this field has started to grow but faces a huge methodological challenge, which is to develop high-fidelity experimental stimuli as realistic representations of accident scenarios in order to elicit valid reactions from human participants. The present paper reviews and evaluates several existing methods used in the research field before proposing an alternative method of generating animated accident sequence using driving simulation software. It is argued that this method strikes a good balance of fidelity, versatility and cost-effectiveness. We also present some preliminary evidence for the effectiveness of variable manipulation using such a methodology.

Driving in the Virtual: Towards Large-Scale Software-Defined Transportation Scene

Image dataset plays an essential role in training the computer vision models. Compared with realistic datasets that are expensive to label and inflexible in number and scene, virtual datasets gradually show advantages and begin to act as a supplement in the training and validation of computer vision models. In this paper, we propose a pipeline for constructing artificial traffic scenes and generating virtual datasets based on an automatic driving simulation software from the perspective of parallel vision. Firstly, map data and simulation modeling elements like buildings are used to model the 3D artificial scene built upon a real environment. Furthermore, we use those artificial transportation scenes to automatically generate a large-scale and diverse dataset with the ground-truth label, which can provide service for computer vision tasks such as object detection and semantic segmentation. The environmental conditions and sensor properties in artificial scenes can be controlled flexibly and repeatedly. Finally, we show the detail and performance of creating virtual artificial scenes and generating the virtual dataset with low modeling time and highly accurate labeling, which does not depend on manual labeling and achieves a diverse dataset automatically and conveniently.

Research on drivers' hazard perception in plateau environment based on visual characteristics

A good visual search mode is an important prerequisite for a driver to perceive the hazard in the traffic environment timely and accurately. However, hypoxia conditions in the plateau environment might affect drivers' cognitive and judgment ability, posing potential threats to safe driving. This study analyzed the eye movement behaviors of drivers when faced with traffic hazards in the plateau environment, to examine the impact level of altitude on drivers’ perception and reaction. Nine typical traffic hazard scenarios were investigated at four locations with four different altitudes, including Linzhi, Lhasa, Naqu, and Yanghu Scenic Area based on UC-WIN / ROAD driving simulation software. Then, drivers’ visual search modes were analyzed according to drivers’ eye movement data collected by ASL Mobile Eye monocular eye tracker. As the altitude increased, the drivers’ first fixation time and the average saccade amplitude decreased, while the fixation duration percentage increased. Drivers with fewer years of driving experience had a larger percentage of fixation duration and a smaller saccade amplitude. In addition, a shorter acclimation period also negatively influenced the percentage of fixation time. The increase of altitude would weaken the drivers' visual sensitivity and cognitive processing ability of hazard information, which would reduce drivers' hazard perception skills, and the increase of driving experience might help alleviate such negative impacts to some extent. Based on the visual characteristics of traffic hazard scenarios obtained in this paper, the training of highly accident-prone drivers can be guided specifically to improve their visual search strategies, thereby improving driving safety.

Impact of antiepileptic drugs on simulated driving in patients with epilepsy

ObjectiveResults of observational investigations have demonstrated that the risk of a traffic accident is independent of use of AEDs. However, no reports of driving tests conducted with patients administered AEDs have been presented. This study examined this scenario in a simulated driving setting. MethodsDriving performance of 43 patients with epilepsy (PWE) and prescribed an AED, who were licensed to drive and drove regularly (subject group), was assessed, with the results compared to 40 age- and gender-matched healthy volunteers (control group). Daily driving skills associated with a traffic accident were examined using two different tests provided by a driving simulator software package, road-tracking and car-following. Standard deviation of lateral position (SDLP) and distance coefficient of variation (DCV) were determined as primary and exploratory outcomes, respectively. ResultsThere was no statistically significant difference for primary outcome shown by SDLP between the subject and control groups (p = 0.906), nor for exploratory outcome shown by DCV (p = 0.063). Multiple regression analysis revealed that age (ß=0.967, p = 0.001), female gender (ß=0.469, p<0.001), and duration of driving experience (ß=-0.583, p = 0.038) were correlated with SDLP. SignificanceThe present results demonstrated that the driving performance of PWE taking AEDs was not different from that of healthy volunteers.

Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research.

Collision warning system plays a key role in the prevention of driving distractions and drowsy driving. Previous studies have proven the advantages of tactile warnings in reducing driver's brake response time. At the same time, tactile warnings have been proved effective in take-over request (TOR) for partially autonomous vehicles. How the performance of tactile warnings can be optimized is an ongoing hot research topic in this field. Thus, the presented low-cost driving simulation software and methods are introduced to attract more researchers to take part in the investigation. The presented protocol has been divided into five sections: 1) participants, 2) driving simulation software configuration, 3) driving simulator preparation, 4) vibrating toolkit configuration and preparation, and 5) conducting the experiment. In the exemplar study, participants wore the tactile vibrating toolkit and performed an established car-following task using the customized driving simulation software. The front vehicle braked intermittently, and vibrating warnings were delivered whenever the front vehicle was braking. Participants were instructed to respond as quickly as possible to the sudden brakes of the front vehicle. Driving dynamics, such as the brake response time and brake response rate, were recorded by the simulation software for data analysis. The presented protocol offers insight into the exploration of the effectiveness of tactile warnings on different body locations. In addition to the car-following task that is demonstrated in the exemplar experiment, this protocol also provides optionsto apply other paradigms to the driving simulation studies by making simple software configuration without any code development. However, it is important to note that due to its affordable price, the driving simulation software and hardware introduced here may not be able to fully compete with other high-fidelity commercial driving simulators. Nevertheless, this protocol can act as an affordable and user-friendly alternative to the general high-fidelity commercial driving simulators.

Examining the effects of stress on the driving abilities of paramedic students

Background: Research has suggested that stress may have a negative effect on paramedics' clinical performance. Stress has also been demonstrated to negatively affect the driving abilities of the general population, increasing the number of driving errors. However, no studies have explored stress and its potential impact on the non-clinical performance of paramedics, particularly their driving abilities. Methods: Paramedic students underwent emergency driving assessment in a driving simulator before and after exposure to a stressful medical scenario. The number and type of errors were documented before and after through the use of both driving simulator software and observation by two members of the research team. The NASA Task Load Index (TLX) was used to record self-reported stress levels. Results: Thirty-six students participated in the study. Following exposure to a stressful medical scenario, paramedic students had no rise in overall error rate, but demonstrated increases in three critical driving errors: namely, failure to wear a seat belt (three baseline versus 10 after stress); failing to stop for red lights or stop signs (seven versus 35); and losing control of the vehicle (two versus 11). Self-reported stress levels also increased after the clinical scenario, particularly in the area of mental (cognitive) demand. Conclusion: Paramedics are routinely exposed to acute stress in their everyday work, and this stress could affect their non-clinical performance. The critical errors committed by participants in the present study closely matched those considered to be contributory factors in many ambulance collisions. These results illustrate the need for further research into the effects of stress on non-clinical performance in general, and highlight the potential need to consider additional driver training and stress management education to mitigate the frequency and severity of driving errors among paramedics.

Non-intrusive vehicle-based measurement system for drowsiness detection

The purpose of this study is for prototyping a non-intrusive vehicle-based measurement system for drowsiness detection. The vehicle-based measurement system aims to achieve the non-intrusive drowsiness detection. The non-intrusive vehicle-based measurement achieved by placing sensors on the steering rod, gas pedal, and brake pedal. Drowsiness can be detected by comparing the position of the steering angle to the desired target angular position, especially when the difference in value of both is greater. Some sensors have been tested to obtain the actual steering angle position. From the test results, sensors that meet the criteria of accuracy are MPU6050 and HMC5883L. Both sensors have been tested in the prototyping of a vehicle-based drowsiness detection system with sufficient results. Furthermore, the prototype of non-intrusive vehicle-based drowsiness detection system has been integrated with interesting driving simulation software. The result has been able to show the actual condition of the steering position, the gas pedal and the brake pedal precisely. Moreover, this prototype opens opportunities to support the study of drowsiness detection using vehicle-based driving simulator.

Assessment of driving skills of indian adults on adriving simulator: A pilot study

Background: Driving is associated with independence, but it also exposes a person to harmful situations while on road. Safety becomes an important aspect while assessing an individual's driving ability. This study examined the driving performance of Indian drivers, both young- and middle-aged adults on driving simulator (DS). Three tests were selected on the DS: Ability test 1: clutch, accelerator, brake, and steering assessment; Ability test 2: stopping, precise stopping, hard stopping, and reaction time; and Ability test 3: city driving for 5 min. Objective: The objective of the study was to assess and compare the driving performance in young- and middle-aged Indian adults. Study Design: Observational, analytical study design was chosen for the research - a pilot study on DS. Methods: A total of thirty normal adults were included in this study. Among them, 15 normal adults who were of 20-40 years were called young adults and other 15 who were of 40-60 years were called middle-aged. The inclusion criteria were normal adults with experience of driving four-wheeler for ≥1-year, both male and females, and age >20 years. Exclusion criteria were normal adults who had readily noticeable visual, cognitive, or locomotors impairment which can compromise on the driving ability. The outcome measures were scores from DS software. The test scores were analyzed using unpaired t-test. Results: There were 30 normal adults included in the study (n = 30). In Ability 1, performance on clutch (P = 0.13), brake (P = 0.90), and accelerator (P = 0.97) did not show significant difference between the young and middle age groups; steering (P = 0.00) performance had highly significant difference between young- and middle-aged adults. Middle-aged adults performed better than young adults. In Ability 2, performance for precise stopping (P = 0.01) has shown a significant difference, whereas reaction time (P = 0.008) is highly significant. On the other hand, stopping (P = 0.6) and hard stopping (P = 0.85) do not have a significant difference between the two groups.In city driving (P = 0.00), middle-aged adults performed better than young adults. The level of significance was set at α =0.05 and the confidence interval was at 95%, and the degree of freedom was 28. Conclusion: The performance of driving skills in normal Indian adults on DS is an achievable functional measure. Middle-aged adults are at par with young adults in the simulated operations when checked on the DS for Ability 1, Ability 2, and city driving.

An experimental paradigm for the assessment of realistic human multitasking

Human multitasking has been evaluated with paradigms that administered two—rarely three—concurrent tasks. In everyday life, however, we usually face an ever-changing sequence of distinct concurrent tasks. Available studies therefore provided valuable insights into our ability for dual tasking, but they did not address the natural interplay of dual tasking and task switching. The present study was undertaken to explore the feasibility of two new paradigms which replicate that interplay in virtual reality. We used car driving simulator software to implement a virtual car-driving task as well as a virtual street-crossing task. Either task was administered alone, as well as concurrently with a battery of loading tasks that mimicked activities of everyday life. The loading tasks used different sensory modalities, different cognitive processes, and different output channels and were presented in an ever-changing sequence. Cronbach’s alpha scores of key registered variables were high, which indicates that our approach is reliable. Driving and street-crossing performance deteriorated under multitask conditions, which indicates that our approach is sensitive to multitasking. This is the first study to demonstrate the feasibility of an experimental paradigm for the assessment of natural multitasking, i.e., of combined dual tasking and task switching. This paradigm could be of interest for basic science as well as for prevention and rehabilitation settings.

Magnetoencephalogram recording during simulated driving: Towards an ecologically-valid paradigm

Magnetoencephalogram recording during simulated driving: Towards an ecologically-valid paradigm

Sensitivity of traffic safety to the pavement and road environment conditions using driving simulations

ABSTRACTThis article aims (1) to validate the influence of pavement skid resistance on traffic safety in most important road environments and (2) to demonstrate the usefulness and reliability of driving simulations in traffic safety analysis of real scenarios taking into account pavement skid resistance. For this purpose, driving simulation software was used to assess the influence of surface properties on vehicular movement. Collision and overturning were the selected types of accidents in the road segments under study. Traffic safety was evaluated in three possible scenarios: (1) curve driving, (2) straight-line emergency braking, and (3) braking on a curve. The selection of the most relevant road environments and accidents to the sensitivity analysis was validated by previous research. Simulations have confirmed that road environments where braking maneuvers are more common and environments with small radii of curvature and high speeds require a higher level of skid resistance. The driving simulations performed as it was demonstrated and validated in the article are easy and simple to operate and can be useful in the design phase of safe solutions.

A Pedestrian Simulator for Urban Crossing Scenarios

In order to evaluate critical urban road crossing scenarios from the point of view of a pedestrian in a safe and reproducible manner, a setup for a pedestrian simulator - analog to a driving simulator - was elaborated using a head-mounted display, a motion capture system and a driving simulator software (for the virtual environment). This paper explains the motivation for such a simulator and delineates the integrated components. Further, a first small-scale study was conducted, delivering insights into the acceptance of the technical setup and the perception of the virtual environment, followed by a larger study evaluating the subjective immersiveness of the simulator using the Presence Questionnaire (PQ) in the case of a gap-acceptance study. The participants were asked to cross the virtual road on two different conditions: an easy and a difficult scenario, varying the density and velocity of passing vehicles. The comparison of the PQ scores of these two conditions revealed no significant difference neither regarding the total PQ score nor its factors, suggesting no influence of the traffic density/velocity on the user’s immersive impression. The discussion provides a lookout for possible improvements regarding the setup and further studies.

The Effects of Dextromethorphan on Driving Performance and the Standardized Field Sobriety Test.

Dextromethorphan (DXM) is abused most commonly among adolescents as a recreational drug to generate a dissociative experience. The objective of the study was to assess driving with and without DXM ingestion. The effects of one-time maximum daily doses of DXM 120 mg versus a guaifenesin 400 mg dose were compared among 40 healthy subjects using a crossover design. Subjects' ability to drive was assessed by their performance in a driving simulator (STISIM® Drive driving simulator software) and by conducting a standardized field sobriety test (SFST) administered 1-h postdrug administration. The one-time dose of DXM 120 mg did not demonstrate driving impairment on the STISIM® Drive driving simulator or increase SFST failures compared to guaifenesin 400 mg. Doses greater than the currently recommended maximum daily dose of 120 mg are necessary to perturb driving behavior.