Driving Simulators for the Evaluation of Human-Machine Interfaces in Assisted and Automated Vehicles (Transportation)

Description

Driving Simulators for the Evaluation of Human-Machine Interfaces in Assisted and Automated Vehicles is a concise reference work on driving simulators, which conveys the technology behind simulator systems used to test driver assistance systems and automated vehicles, including electric vehicles. Coverage includes architecture, computer graphics, evaluation parameters and applied examples. A driving simulator is a device that has the function of presenting similar visual, auditory and force perceptions to those experienced during driving, creating the illusion that the driver is driving an actual car. The advantage of tests using a driving simulator is that it can reproduce dangerous traffic situations and tests safely. Driving simulators are also valuable in research and development into intelligent driving systems, allowing for testing and evaluation in a simulation environment rather than on the road. With its concise selection of relevant material and applied focus, this book will be of use to research and development professionals in industry and academic researchers whose work involves automotive systems and technologies in general, and particularly those working on driving simulators and automated driving. Cover Contents About the editors Introduction 1 Overview 1.1 Introduction 1.2 Objectives of DS 1.3 History, evolution, and challenges of DS 1.3.1 Driving experiment using DS 1.3.2 Belt conveyor system 1.3.3 Six-axis motion system 1.3.4 Six-axis motion system and translator system 1.3.5 Recent large-scale DS 1.3.5.1 DS of Toyota [3] 1.3.5.1.1 Overview 1.3.5.1.2 Motion equipment 1.3.5.1.3 Video equipment 1.3.5.2 DS of Nissan [4] 1.3.5.2.1 Motion equipment 1.3.5.2.2 Image equipment 1.4 Conclusion References 2 Present driving simulators 2.1 Introduction 2.2 DS SIoTDS-O (simple type at Omiya campus in SIT) 2.3 DS SIoTDS-T (advanced type at Toyosu campus in SIT) 2.3.1 Introduction of SIoTDS-T 2.3.2 Motion device on the SIoTDS-T 2.3.3 IG of the SIoTDS-T 2.3.4 System configuration of the SIoTDS-T 2.3.5 Cockpit of SIoTDS-T 2.4 DS UoLDS (full-scale type at Leeds University) [3] 2.4.1 Introduction of UoLDS 2.4.2 UoLDS motion device 2.4.3 IG of the DS 2.4.4 System configuration of the UoLDS 2.4.5 Cockpit of UoLDS 2.5 Conclusion Acknowledgements References 3 Architecture of driving simulators 3.1 Architecture principles 3.1.1 Introduction 3.1.2 Hardware/software general architecture 3.1.3 System integration 3.1.3.1 XIL and standardization 3.1.3.2 Data collection and simulation replay 3.1.4 Hardware architecture 3.1.4.1 Introduction 3.1.4.2 The cabin 3.1.4.2.1 The frame structure and its variations 3.1.4.2.2 Immersive cockpit design 3.1.4.2.3 Full-scale simulator using a real car 3.1.4.3 The main display system 3.1.4.3.1 The display size 3.1.4.3.2 Brightness and contract 3.1.4.3.3 Large FOV and multi-display systems 3.1.4.3.4 Displays in a system with motion 3.1.4.3.5 Head-mounted displays 3.1.4.4 Head-up displays 3.1.4.4.1 Background 3.1.4.4.2 Use of HUD in DS 3.1.4.4.3 Using a real HUD hardware 3.1.4.4.4 Simulation of HUD devices in VR 3.1.4.5 Steering wheel 3.1.4.6 Motion, haptic feedback 3.1.4.7 Physiological measurement 3.1.4.7.1 Neural activity measurement, EEG, MRI, fMRI, and NIRS 3.1.4.7.2 Eye-tracking 3.1.4.7.2.1 Types of eye trackers 3.1.4.7.2.2 Video base eye-tracking: principle 3.1.4.7.2.3 Bright pupil tracking vs. dark pupil tracking 3.1.4.7.2.4 Applications 3.1.4.7.2.5 Limitations 3.1.4.8 ECUs and sensor modules 3.1.5 Software architecture 3.1.5.1 Vehicle modeling and simulation 3.1.5.1.1 Role and integration of the vehicle simulation 3.1.5.1.1.1 Simulate the dynamic movement of the vehicle accurately to test its performance 3.1.5.1.1.2 Respond to user driving input, ADAS, and other ECU commands 3.1.5.1.1.3 Reproduce the movement and other information of the vehicle represented in the virtual environment 3.1.5.1.1.4 Provide data to control motion and haptic feedback 3.1.5.1.1.5 Integrate with the simulation of the environment, surrounding vehicles, and pedestrians 3.1.5.1.2 Vehicle dynamics simulation 3.1.5.1.2.1 Vehicle body model 3.1.5.1.2.2 Tire and wheel model 3.1.5.1.2.3 Suspension model 3.1.5.1.2.4 Powertrain dynamic model 3.1.5.1.2.5 Brake system model 3.1.5.1.2.6 Steering dynamics 3.1.5.1.3 ADAS simulation 3.1.5.1.4 Sensors 3.1.5.1.4.1 Ideal sensor and physical models 3.1.5.1.4.2 Camera sensors 3.1.5.1.4.3 Radars 3.1.5.1.4.4 Light detection and ranging 3.1.5.2 Environment modeling and simulation 3.1.5.2.1 3D environment 3.1.5.2.2 Road modeling 3.1.5.2.3 Traffic modeling 3.1.5.2.4 Scenario and interactive content 3.2 Motion cueing and haptic feedback 3.2.1 The human motion perception 3.2.1.1 Visual sense 3.2.1.2 Vestibular senses: sense of gravity, rotation, and movement 3.2.1.3 Proprioception (somatosensory) 3.2.1.4 Auditory sense (hearing) 3.2.1.5 Motion sickness 3.2.2 Reproduction of the motion stimulus in the simulator 3.2.2.1 Visual sense: displays 3.2.2.2 Auditory sense: multi-speakers or headsets to reproduce 3D sound 3.2.2.3 Proprioception sense 3.2.2.4 Vestibular sense: motion platforms 3.2.2.5 Types of motion platforms 3.2.2.6 Applications 3.2.2.7 Limitations 3.2.3 Motion cueing algorithm 3.2.3.1 Algorithm principles on a Stewart platform 3.2.3.2 What changes in the algorithm by extending the motion platform? 3.2.3.3 The classical motion cueing algorithm of the filter-based algorithm 3.2.3.4 Discussions and strategies for improvement 3.2.3.4.1 Use of a human perception model 3.2.3.4.2 Optimum algorithm 3.2.3.4.3 Use of a kinematic model of the motion hardware 3.3 The evolution of simulators with VR 3.3.1 Driving simulation and transportation 3.3.2 VR and cockpit HMI 3.3.3 AI and machine learning References 4 Computer graphics in driving simulators 4.1 Principles of computer graphics 4.1.1 Objectives 4.1.2 Basic concepts 4.2 Modeling 4.2.1 Sky modeling 4.2.2 Lamp and lamp pattern modeling 4.2.3 Modeling the road surface 4.2.4 Transparent and semi-transparent surfaces 4.2.4.1 Concepts of transparency in computer graphics 4.2.4.2 Trees and vegetation 4.2.4.3 Windows and glass 4.2.5 Optimization for real-time application 4.2.5.1 Geometry simplification and texturing 4.2.5.2 Normal mapping 4.2.5.3 Texture baking 4.2.5.4 Multiple levels of detail 4.2.6 Particle systems: rain, snow, and smokes 4.2.7 Water on road surface and windshield 4.3 Shading 4.3.1 Rendering of light sources 4.3.2 Physically based rendering 4.3.3 Material layering 4.3.4 Color range and tone mapping 4.3.5 Data and processing flow 4.3.6 Multitarget rendering 4.4 Hardware 4.4.1 Rendering hardware architecture 4.4.1.1 Role of the GPU and CPU 4.4.1.2 A simple comparison between existing GPU solutions 4.4.1.3 Double-precision versus single-precision floats with GPUs 4.4.2 Synchronization over multiple displays 4.4.3 Synchronization of driver’s eye with VR viewpoint References 5 Tools for evaluating HMI 5.1 Introduction 5.2 Gaze detection 5.2.1 Introduction of gaze detection 5.2.2 Measurement method 5.2.2.1 Contact type 5.2.2.2 Non-contact type 5.2.3 Issues 5.3 Response time evaluation 5.4 Electroencephalograph – brain wave detection 5.4.1 Introduction of electroencephalograph – brain wave detection 5.4.2 Measurement method 5.4.2.1 Electrode configuration 5.4.2.2 Electrode 5.4.2.3 Measurement method 5.4.2.4 Data processing 5.5 Cerebral blood flow – brain blood detection 5.6 Electrocardiograph – heartbeat detection 5.7 Driving performance 5.8 Steering wheel angle 5.9 Simulator sickness evaluation 5.10 ADAS evaluation by DS 5.10.1 FVCWS evaluation by DS 5.10.2 Automated breaking evaluation by DS 5.10.3 ACC evaluation by DS 5.10.4 LDWS evaluation by DS 5.10.5 LKA evaluation by DS 5.11 Trust evaluation 5.11.1 Concept of validity 5.11.2 Visual information processing 5.11.3 Vestibular information processing 5.11.4 Auditory information processing 5.11.5 Physical consistency between perceptual information 5.12 Automated driving evaluation by DS 5.12.1 Take-over evaluation by DS 5.12.2 Ethics evaluation by DS 5.12.3 Communication method evaluation by DS 5.13 Conclusion References 6 Applications using driving simulators 6.1 Introduction 6.2 A study on the effect of HUD information on driving operation 6.2.1 Introduction of evaluation for HUD information 6.2.2 Method 6.2.2.1 Driving simulator 6.2.2.2 Conditions for the experiment 6.2.2.3 Evaluation method 6.2.3 Result 6.2.4 Conclusion of evaluation for HUD information 6.3 Study on AEBS applying driver models 6.3.1 Introduction of evaluation for AEBS 6.3.2 AEBS taking into account the individual characteristics of drivers 6.3.3 Method 6.3.4 Experiment on braking operations by drivers 6.3.5 Experiment on evaluation of alarm timing 6.3.6 Results and Discussion 6.3.7 Conclusion of evaluation for AEBS 6.4 Effect of unconscious learning for driver attention 6.4.1 Introduction of unconscious learning 6.4.2 Experimental method 6.4.2.1 Outline of the experiment 6.4.2.2 Learning effect experiment by guidance sound 6.4.2.3 Learning effect experiments with guidance sound and driving habits 6.4.3 Experimental results 6.4.3.1 Learning effect experiment by guidance sound 6.4.3.2 Learning effect experiments with guidance sound and driving habits 6.4.4 Summary of unconscious learning 6.5 Comparison of the effects among the keeping awakening tasks for the driver during automated driving using EEG analysis 6.5.1 Introduction of keeping awakening tasks 6.5.2 Experimental tasks 6.5.2.1 Conventional method (1): Holding steering wheel 6.5.2.2 Conventional method (2): Pushing button 6.5.2.3 Conventional method (3): Saccadic eye movement 6.5.2.4 Proposed method (1): Information provision 6.5.2.5 Proposed method (2): Single light blinking 6.5.3 Experiment 6.5.3.1 Experiment overview 6.5.3.2 Subjects and experimental conditions 6.5.3.3 Experimental apparatus 6.5.3.4 Evaluation method 6.5.4 Experimental results 6.5.4.1 Number of subjects who fell asleep during the experiment 6.5.4.2 Subjective evaluation 6.5.4.3 Electroencephalographic analysis 6.5.5 Discussion 6.5.6 Summary of keeping awakening tasks 6.6 Estimation of driver drowsiness change in automated driving using heart beat analysis 6.6.1 Introduction of heart beat analysis 6.6.2 Heart rate analysis 6.6.2.1 Autonomic nervous system 6.6.2.2 RRI (R-R interval) 6.6.2.3 Heart rate variability analysis 6.6.3 Experiment 6.6.3.1 Experiments using a DS 6.6.3.2 Experimental apparatus 6.6.3.3 Experimental results 6.6.4 Results of heart rate variability analysis 6.6.5 Classification results 6.6.6 Summary of heart beat analysis 6.7 Driving characteristics of low awakening drivers during transition from automatic driving to manual driving 6.7.1 Introduction 6.7.2 Method 6.7.2.1 Scenario in the experiment 6.7.2.2 Evaluation of arousal level 6.7.3 Results 6.7.3.1 Reaction time of brake operation (comparison by arousal levels) 6.7.3.2 Maximum brake pedal force (comparison by arousal levels) 6.7.3.3 EEG in low-arousal states 6.7.4 Conclusion 6.8 Driving characteristics of low awakening drivers during transition from automatic driving to manual driving 6.8.1 Introduction of evaluation for transition from automatic driving to manual driving 6.8.2 Experimental device 6.8.2.1 Biological signal measurement device 6.8.2.2 EEG measurements 6.8.3 Experimental methods 6.8.3.1 Low awakening driver 6.8.3.1.1 Experimental scenario 6.8.3.1.2 Experiment participants 6.8.3.1.3 NEDO evaluation 6.8.3.2 Sleeping driver 6.8.3.2.1 Experimental scenario 6.8.3.2.2 Experiment participants 6.8.4 Experimental results 6.8.4.1 Low awakening driver 6.8.4.1.1 Proportion of driving operation regarding collision avoidance by the driver 6.8.4.1.2 Comparison of reaction time until brake operation according to alertness level 6.8.4.1.3 Comparison of maximum brake pedal force according to alertness level 6.8.4.1.4 Comparison of time to maximum brake pedal force according to alertness level 6.8.4.1.5 Relationship between reaction time and maximum brake pedal force according to alertness level 6.8.4.1.6 Relationship between time to maximum brake pedal force and maximum brake pedal force according to alertness level 6.8.4.1.7 Steering action in driver who conducted steering avoidance 6.8.4.1.8 EEGs of drivers in low awakening state 6.8.4.2 Sleeping driver 6.8.4.2.1 Proportion of collisions 6.8.4.2.2 Comparison of reaction time until performing braking operations 6.8.4.2.3 Comparison of maximum brake pedal force 6.8.4.2.4 Comparison of time until maximum brake pedal force 6.8.4.2.5 Comparison of speed reduction rates 6.8.4.2.6 Relationship between reaction time and maximum brake pedal force 6.8.4.2.7 Relationship between time to maximum brake pedal force and maximum brake pedal force 6.8.4.2.8 EEG during sleep 6.8.5 Conclusions 6.8.5.1 Low awakening driver 6.8.5.1.1 Proportion of driving operations for collision avoidance performed by the driver 6.8.5.1.2 Reaction time until the driver steps on brake 6.8.5.1.3 Maximum brake pedal force 6.8.5.1.4 Time to maximum brake pedal force 6.8.5.1.5 EEG of low awakening driver 6.8.5.2 Sleeping driver 6.8.5.2.1 Proportion of participants colliding their operating vehicle with pedestrian 6.8.5.2.2 Time to stepping on brake pedal after participant visually recognized pedestrian 6.8.5.2.3 Maximum brake pedal force of the participant in braking operations 6.8.5.2.4 Time until the maximum brake pedal force was measured during braking operations of the participant 6.8.5.2.5 Proportion of vehicles with reduced speed when colliding with pedestrian 6.8.5.2.6 Relationship between the time until operating brakes after visually recognizing the pedestrian and maximum brake pedal force 6.8.5.2.7 Relationship between time until maximum brake pedal force and maximum brake pedal force 6.8.6 Conclusion of evaluation for transition from automatic driving to manual driving 6.9 Conclusion References Index Back Cover