Fuel spray characteristics are affected by cavitation flows in a fuel injector nozzle. Cavitation in nozzles with various holes lengths and jets are visualized. As a result, we find that different flow characteristics of nozzles with a short hole length and a long hole length result in different characteristics of cavitation in the nozzles and the liquid jets. Especially in a super cavitation condition, cavitation characteristics of the nozzle with the short hole length are unstable from the hole inlet to the exit and large scale discharges of cavitation near the nozzle hole exit influence large scale deformations of the liquid jet.

View full abstract

Fuel spray characteristics are affected by cavitation in fuel injector nozzles. We calculated the gas volume fractions of cavitation in nozzles and the liquid volume fractions of liquid jet based high-speed images of cavitation with two hole lengths nozzles and liquid jet. We quantitatively analyzed the sequential gas and liquid volume fractions in order to clarify deformation processes of liquid jet affected by cavitation. As a results, the difference of the nozzle hole lengths affects scale of cavitation clouds shedding near the hole exit and also scale of deformation of liquid jet boundaries between liquid jet and the ambient.

View full abstract

The estrangement of the emission measurement of an approval test mode and the real world is a problem, and Portable Emissions Measurement System (PEMS) measurement is proposed. The exhaust gas flow measurement when the low load that is low flow rate and high pulsation is difficult. Therefore, measurement accuracy of PEMS becomes the problem. In this study, the exhaust gas flowmeter using the laminar flow element was developed. It was evaluated influence on flow rate, temperature and pulsation. From the above, ExLFM showed measurement performance applicable to exhaust gas measurement.

View full abstract

For fuel efficiency improvement, contributing to further reduction in mechanical loss is required to the piston-ring which is powertrain component. However, there are also a lot of cases of lubricating oil consumption increase and abnormal burning caused by oil upward flow. The cylinder bore shape under engine operating condition has much effect, and the sliding surface pressure distribution between the piston ring and the cylinder boa is important about oil transport via the piston ring sliding surface. For clarifying the mechanism of lubricating oil consumption, changes in sliding condition of the piston ring by the cylinder bore shape was evaluated by using of the motoring test device, and details of the piston ring conformability to the cylinder bore was considered in this report.

View full abstract

Piston design approach for automotive engines has been advanced from both experimental and calculation analysis. However, the developments of experimental analysis method that can verify the accuracy of the calculation analysis are required. In this paper, multi-point thin-film pressure sensor for piston pin-boss part (hereinafter pin-boss) was sputtered on piston pin sliding surface and oil-film pressure distribution was measured in axial and circumferential direction using the pin-boss fatigue tester. Additionally, oil-film pressure measurement of the piston pin-boss portion was carried out without changing the environment of the measurement portion by using the thin-film pressure sensor. We also analyzed the application of elastic fluid lubrication theory using CAE analysis, compared with measured values．

View full abstract

To investigate the effect of the fuel injection pressure on the cooling loss in a diesel engine, the present study evaluated the turbulent characteristics of diesel spray combustion in an optically-accessible compression and expansion machine. Images of the diesel spray flame were taken by a direct photography using a CMOS camera, and the flame velocity field was estimated from the pairs of the images with an image correlation method. The velocity was separated into the high and low frequency components by a frequency separation method, and the turbulent intensity was examined. The flame temperature was determined with a two-color method in the compression and expansion machine, and the heat balance was obtained in a diesel engine with a cylinder bore and a stroke identical with the compression and expansion machine. The correlation between the turbulent properties, the flame temperature and the cooling loss was investigated while changing the fuel injection pressure.

View full abstract

Multi Stage Hybrid Transmission has adopted the multi-stage approach with the input split mode to enable the concept of multiple high efficiency points. This approach has more high efficiency points than previous hybrid transmission to improve acceleration performance, fuel economy dramatically. In this paper, the reasons to adopt the multi-stage approach and the result of evaluation will be introduced.

View full abstract

This study aims to develop and research technology that reduces cooling loss without increasing smoke emissions by applying partial insulation on the combustion chamber wall surface. Cooling loss and engine volumetric efficiency in engine cylinders were analyzed by simulation to find suitable insulation material and identify the areas to insulate. Experimental study at full and part load was conducted to confirm the above conclusion. These results indicated partially insulated piston is useful as a technology for reducing cooling loss of internal combustion engine.

View full abstract

Along with the recent increase of the interest in lower exhaust emissions and better fuel economy of automobiles, it has been required to improve the accuracy of the engine air-fuel ratio control. In improving the accuracy of the air-fuel ratio control, how to improve the estimation accuracy of in cylinder intake air amount is the challenge. For preceding studies, the estimation accuracy has been steadily improved, but the estimation error caused by the transient state remains. In this study, we propose to use Gaussian Process (GP), a statistic model, to estimate the in cylinder intake air amount takes to the transient state into account. To improve the estimation accuracy using GP, it is required to obtain the exhaustive data of engine operation, but it is not easy because of cost constraint. Therefore, we have studied a method to extract data which contributes to the estimation accuracy out of the limited measurement data. We have verified the impact on the estimation accuracy where modeling is done by extracting the training data from engine operation data according to a rule and the results are reported in this paper．

View full abstract

In order to respond to the strict demands for reduction of environmental impact on a global scale, a new sheet metal turbine housing has been developed to improve fuel efficiency performance and exhaust emission performance of a diesel vehicle. It contributes to significant improvement of diesel engine emission under real world operation. It can realize earlier light-off of catalyst compared to a conventional solid casting housing through less heat mass by its hollow structure using sheet metal. Thermal deformation at a turbine shaft carrier caused by replacing with sheet metal is minimized by a casting and sheet metal hybrid structure. Increase of turbine noise is measured by adopting a damping plate. As a result, performance of a sheet metal turbine housing can be improved against casting one with keeping same liability of it.

View full abstract

When cylinder pressure diagram sampling data number is reduced, calculation results of indicated mean effective pressure (IMEP) and mass burnt fraction (MBF) are influenced. These factors are aliasing component. In this paper, investigated IMEP calculation method and MBF 50% crank angle estimation method by how to be hard impacted of the sampling data number reduction. As the results, the validity of the method has been verified.

View full abstract

Targeting thermal efficiency improvement of diesel engines with injection rate shaping, TAIZAC (TAndem Injectors Zapping ACtivation) injector consisting of commercially available two common-rail injectors was developed as a research tool. Control of injection timing and duration of the upper injector realized (1) originally targeted inversed-delta injection rate shaping, (2) sharper rise of initial injection rate, and (3) modulation of injection pressure among and during multi-injection pulses, ex. low pressure pre-injection and high pressure main injection.

View full abstract

By shifting natural frequencies of the internal transmission system of combustion impact power in a diesel engine and by changing the fuelinjection timings in the case with two-stage combustion, we made a natural frequency of the internal transmission system coincided with a combustion-impact cancelling frequency and investigated the influence of the present method on the radiated noise characteristics. We confirmed that we can make the natural frequency of the piston-connecting rod vibration higher and reduce the engine noise by shortening thinned portion of connecting rod. We also confirmed that the combination of changing the natural frequency of the piston-connecting rod vibration and noise cancelling spike combustion is effective in the engine noise reduction.

View full abstract

In an air path control system, a good transient response is required and a feedforward controller plays an important role to improve the transient response. In this paper, a nonlinear feedforward controller taking account of the response delay of the turbocharger is designed based on the obtained model. The exhaust manifold temperature is estimated by using a discrete combustion model to improve the control performance. The effectiveness of the proposed method is evaluated by conducting a reference tracking experiment under transient mode operation using a real diesel engine.

View full abstract

This study investigated a correlation between ignition delay time and cetane index using nine types of Light Cycle Oil (LCO). In addition, the effects of the pilot fuel spray on the ignition behavior of the main fuel spray were closely discussed. It was shown that the old cetane index (JIS K 2204:1983) represented the fuel ignitability of the LCOs accurately compared to the current cetane index (JIS K 2280-5:2013). The effect of the pilot injection became less effective as the ignition delay of the main fuel spray became longer.

View full abstract

Measuring of in-cylinder pressure of diesel engine during cranking period under low-temperature condition was attempted by using a strain gauge. Variation of dynamic strain of glow plug associated with the in-cylinder pressure was obtained by means of a strain gauge attached to the glow plug housing. In order to convert the dynamic strain to pressure, relationship of in-cylinder pressure to dynamic strain was investigated by using a piezoelectric pressure sensor. Calibration coefficient, which converts dynamic strain to pressure, was derived from the values of dynamic strain and in-cylinder pressure at the compression end in a motoring operation. As the result, in-cylinder pressure of diesel engine during cranking period can be obtained, and rate of heat release can be derived from the obtained in-cylinder pressure. Both the in-cylinder pressure and the rate of heat release reveal that ignition of fuel spray of each cycle was remarkably unstable during the cranking period under the low-temperature condition.

View full abstract

Emission standards for PM (particulate matter) in the world have been stricter year by year. Although a gasoline direct injection engine has an advantage of lower fuel consumption, it emits more PMs than a port fuel injection (PFI) engine. In Europe, a gasoline particulate filter (GPF) has been introduced for the filtration of gasoline soot. Similar to a diesel particulate filter (DPF), PM is collected by GPF when the PM passes through the porous filter walls. It would be better to have better GPF filtration efficiency, but at the same time, its large pressure rise should be avoided. In this study, using carbon particles as model soot, we evaluated the filtration performance of GPF. Using three GPFs of different length, we examined the pressure drop (filter backpressure) and the filtration efficiency experimentally, compared with DPF.

View full abstract

This paper describes the effect of methanol addition to urea solution for the freezing point depression on the performance of SCR catalyst. In experiments, the effect of methanol addition to urea solution on the NOx reduction performance of SCR catalyst was investigated by using a small-scale catalytic reactor with a simulated exhaust gas. The results show that NOx conversion rate is deteriorated by adding methanol to a reductant even if the mass content of urea in a solution is kept constant.

View full abstract

Kinetic parameters of ammonia adsorption/desorption, NO reduction, and ammonia oxidation are measured, and applied for the steady-state ammonia-SCR Cu-zeolite catalytic models in the simulations. As the oxygen concentration changes due to the EGR and engine load in the real drive operation, the effects of oxygen concentration on the reactions are considered in the model. The calculated results in steady-state conditions are compared with the surrogate gas experiments and the accuracy of the model is verified.

View full abstract

Gasoline engines have the trace-knock phenomena induced by the faster combustion which happens a few times during 100 cycles. And that constrains the thermal efficiency improvement due to limiting the ignition timing advance. Therefore this trace-knock was simulated with a turbulent combustion model in which a cycle-by-cycle variation model and a chemical calculation subprogram were included. The combustion zone of this combustion model is considered to obtain proper turbulent combustion speed through wide range of engine speed. A reduced elementary reaction model of a gasoline PRF (primary reference fuel) for the chemical calculation was customized to the trace-knock prediction. Some advantages of the chemical reaction calculation were found through this simulation. These are described in this paper.

View full abstract

As is well-known, high efficiency is an important advantage of lean burn mode of combustion engines. However, combustion variation is a bottleneck that constraints thermal efficiency improvement in lean combustion. In this paper, a combustion variation restrain control strategy with thermal efficiency optimization consideration is proposed which consists of dual control loops, the inner-loop targets on cycle-to-cycle and cylinder-to-cylinder variation evaluated by IMEP and CA50, and the out-loop is exploited to provide extremum seeking of spark advances of individual cylinder. The effectiveness of the proposed method and variation restrain performance are validated on a commercial gasoline engine test bench..

View full abstract

A new idea of describing ignition delay time τ for hydrocarbon fuels has been proposed, in which ignition process is divided into two phases, H₂O₂ loop induction time τ₁ and H₂O₂ loop dominant time τ₂, whether it induces a cool flame with low initial temperatures or does not with high initial temperatures. A new idea for predicting ignition timing in the cylinder with high accuracies has been proposed, in which the integral of 1/ τ₁ predicts the end of τ ₁, and then, the integral of 1/ τ₂ starts to predict the ignition timing. Ignition delay time equations for a premium-gasoline surrogate fuel have been developed, which can reproduce the temperature-, pressure-, and equivalence ratio-dependences of τ₁ and τ produced using a detailed reaction mechanism. H₂O₂ loop induction time with low initial temperatures τ_{1_low}, H₂O₂ loop induction time with high initial temperatures τ_{1_high}, and ignition delay time with high initial temperatures τhigh have been expressed using Arrhenius equations. H₂O₂ loop induction time τ₁ and ignition delay time τ have been expressed as τ ₁ = τ_{1_low} + A(τ_{1_high} – τ_{1_low} ) and τ = C{τ_{1_low} + B(τ_high – τ_{1_low} )}, respectively. Equations for A and B have been developed using Gompertz functions to express τ₁ and τ with middle initial temperatures, respectively. An equation for C has been developed using a Gompertz function to express τ with very high initial temperatures.

View full abstract

Among the parts used in gasoline engines, the thermal load on exhaust parts such as turbochargers is very high. In this study, we used nonlinear general-purpose analysis software and built a technology in three-dimensional model to predict thermal fatigue cracks occurring in products by identifying the process progressing from the occurrence of cracks by fatigue tests.

View full abstract

In recent years, reducing the number of cylinders of an engine has been advanced for fuel saving. However, as the cylinder number is reduced, the torque fluctuation from the engine increases and the NV performance deteriorates. Therefore, the importance of torsional vibration reduction technology has been rapidly increased, and active research and development is under way. Among them, CPVA is attracted to attention because it effects in the full revolution range. This study focuses on the torque generated by CPVA as the vibration reduction performance of CPVA and a theoretical expression for calculating the generated torque is derived. Pendulum trajectory covers circular path and epicycloid path. Next, in order to verify the effectiveness of the theoretical analysis result, a multibody dynamics model corresponding to the theoretical analysis model is developed and the results of both are compared. Further theoretical analysis will clarify the influence of excitation order, excitation amplitude, pendulum damping coefficient and pendulum path for CPVA generated torque.

View full abstract

This study clarified heat value that was derived from effects of magnetic field in space by metal objects with perturbative approach. Magnetic flux could be divided into 3 areas (outer metal object, inner metal object and boundary). This study derived each degenerated expression from Maxwell’s equations. Outer area should be considered magnetic flux distortion due to metal objects, on the other hand it could be assumed quasi steady state and none of electric current. Inner area could limit the range of frequency, magnetic permeability and electric conductivity. At boundary area, magnetic flux density or magnetic field strength is continuing, and the continuity could link the equations on inner and outer area.

View full abstract

Our research group designed and made a trial creation of a medium-sized electric bus for expressways and conducted a demonstration test on Metropolitan Expressway Bay Shore Route for approximately one year. In this paper, analyzed results focused on the regenerative energy made by electric heavy-duty vehicles running on the expressway will be summarized.

View full abstract

To consider an appropriate requirement of a burst pressure of compressed-hydrogen tanks for fuel cell vehicles, an influence of hydraulic sequential tests on the burst pressure of the tanks has been evaluated in the previous paper. In this 2nd paper, a relationship between the damage state of compressed-hydrogen tank and the degradation of the burst pressure is investigated by ultrasonic flaw detection and CT scanning. As a result, it has been clarified that the delamination of CFRP in the drop test and the expansion of the delamination in the pressure cycling test are main factors of degradation of the burst pressure.

View full abstract

Hydrogen filling to a small capacity container by variable pressure ramp rate based on the MC Formula is carried out. An initial pressure of the container is estimated by pre-shot. It is found that the hydrogen filling based on the MC Formula is an suitable filling method for the small container. The initial pressures are estimated properly by theoretical model of pre-shot.

View full abstract

In order to realize automatic drive equivalent to an expert driver, it is necessary to design a target route on the premise of acceleration / deceleration coordinated to lateral motion. Generally, clothoids and sinusoidal half-wavelength gradual relaxation curves designed on the assumption that the traveling speed is constant, however we adopted relaxation curves considering the decrease in speed by G-Vectoring control. The result of comparing this relaxation curve with the vehicle locus of the expert driver will be reported.

View full abstract

In this paper, sensitivity improvement method of magnetostrictive-type torque sensor for EPS was examined. Ni-Fe film which had stronger reverse magnetostrictive effect was selected. However, impedance of the Ni-Fe film decreased in both rotatory direction when detecting impedance gradient with a solenoid coil, therefore the rotation's direction couldn’t be decided. Hence a new method was devised for doing high frequency heating while twisting a steering shaft after Ni-Fe plating processing to it to move a peak point by finding a stress-relaxation phenomenon of Ni-Fe film by the heat-treatment. It could improve not only sensitivity but also hysteresis characteristics of it.

View full abstract

The effects of lead time constants, which transfer from steering to lateral acceleration, are investigated by a driver model-based handling quality evaluation. The modeled driver parameters are identified by the use of steering angle and lateral position during lane change, which are measured with a driving simulator. It is found that a handling quality improves if both the first order lead time constant is large and the second order one is small. The constants for the driver are determined by the longitudinal driver position and also are influenced by both weight distribution and yaw inertia moment.

View full abstract

Previous studies have suggested that every response parameter has a preferred value. There are complicated relations between response parameters and vehicle specifications. This research offers guidelines for design decisions which lead to preferred response parameters. First, design parameters are uniquely defined for designers. With these design parameters, response parameters are calculated, and sensitivity analysis is adapted for evaluating how design parameters impact response parameters. The impacts of design decisions are evaluated quantitatively. Moreover, sensitivity of sensitivity is defined for evaluating a transient status. It is analyzed in a quasi-steady state.

View full abstract

Running resistance is based on measurement on the condition outdoors, and weather environment on test impacts it greatly to measured data. Running resistance are corrected to standard temperature (20℃) using air temperature under measurement by the present regulation correction. We created correction technology which took tire temperature into consideration as the new correction method aiming at high precision. As a result, it became clear that variation of rolling resistance value reduces by using rolling resistance correction formula of tire as compared with correction result by normal regulation.

View full abstract

In recent years, the demand for lightweight and crashworthiness of automobiles has been increasing, and steel-aluminum multi-material automobile structures have been expected because of excellent crash performance. In this study, a multi-material(MM) beam made of aluminum extrusion and high strength steel was proposed instead of side sill of automobile to improve the specific energy absorption(SEA) which stands for the energy absorbed per unit mass and minimize the maximum crash force during pole side impact. The bending behavior of MM beam was investigated by numerical simulation and compared with that of three kinds of double-hat beams (steel-steel, aluminum-aluminum, aluminum-steel). It is found that MM beam has better performance of energy absorption and deformation stability compared to double-hat beams under bending loading. Subsequently, the MM beam was parameterized and a parametric study was performed to evaluate the effect of geometric parameters of the MM beam on crashworthiness. It is found that crashworthiness of the MM beam was significantly affected by the geometric parameters. Finally, in order to obtain the optimal design of MM beam, multi-objective optimization was conducted by combining the radial basis function (RBF) and the non-dominated sorting genetic algorithm II (NSGA-II). The optimal solution obtained from Pareto frontier indicates that multi-material beam which with appropriate geometric parameters is superior in crashing behavior to initial design. And the optimization design data is a good guidance for the crashworthiness design of MM beam.

View full abstract

For compatibility of low vibration and light weight in high-frequency problem, a design guideline is proposed based on a structural design process using experimental SEA for reducing structure-borne sound on machinery. The basic concept of the proposed method is taking consideration into reduction of vibration energy at target subsystem by preventing propagation to the target and concentration of vibration energy into another subsystem simultaneously. In this paper, the structural design process is reviewed, and compatible design guideline for low vibration and light weight is proposed. Then the proposed method is applied to reduce vibration and lighten power control unit used in hybrid automobile.

View full abstract

In this study, we considered a method for selecting high contributing CAE vibration mode to the vehicle interior noise at the running condition through a combination analysis between operational TPA (OTPA) and CAE technique. The high contributing body vibration mode to the road noise was estimated by associating the high contributing principal component mode by OTPA with CAE vibration modes. The countermeasure was then considered by using CAE technique. As a result, the road noise at the target frequency band could be reduced well and the analytical result about the high contributing mode was verified.

View full abstract

In our previous study, a new control method for ride comfort that is called the triple skyhook control has been proposed. This control method perform as if the vehicle’s sprung mass is hooked by three elements; the spring, damper, and inerter. Since the roll motion is coupled with lateral motion through tire side force, the triple skyhook control as pitch and heave motions is not simply applicable to roll motion. This paper applied the triple skyhook control to roll motion by considering roll-lateral coupled motion, and confirmed the control effect of the proposed method.

View full abstract

Self-localization is important technique for autonomous driving. In our laboratory, we are developing an autonomous vehicle, and we implemented two-dimensional self-localization algorithm using LIDAR and orthophotos for the vehicle. Therefore, there is a problem that the method cannot identify the layer when the vehicle is traveling in multi-layer structure road. In this work, we extend the method and suggest selflocalization technique in three-dimensional space based on elevation map. We carried out practical experiments in urban road, and the results showed that our proposed method can be used for estimating the traveling layer in multi-layer structure road.

View full abstract

ADAS (Advanced Driver Assistance System) has been developed recently, and it is desirable that the driver assistance system assists driver to avoid or prevent the high risk situation. Therefore, it is necessary to find situations with high driver’s risk and high assistance effect. In complex traffic situations where other traffic participants exist, the risk of overlooking pedestrians is increased by distributing driver’s attention on steering maneuver. Drive recorder analysis shows that the risk of overlooking pedestrians increases while turning at an intersection. In this paper, experiment was conducted in a test course by using a special test vehicle equipped with steering control assistance as the driver assistance. By this experiment, it was found that the steering control assistance can double the safety confirmation time for detecting pedestrian under the existence of the lead vehicle so that it may become safer even when following the lead vehicle.

View full abstract

This study focuses on the high-risk scene for bicyclist accidents wherein vehcile drivers overtake a bicycle. In this experiment, using a test vehicle called JARI-ARV, the behaviors of drivers with different driving skill, including expert, young-aged, middle-aged, and senior drivers, were examined. The driving indices of potential collision risk to a bicyclist suddenly change the travelling direction were examined in the traffic situations where a bicycle travels in front of the vehicle. Based on our analyses, we clarified that the expert drivers overtake a bicycle with better attention to both risk magnitude and risk duration in the high-risk scene for bicyclist accidents.

View full abstract

This study proposes a lateral acceleration for collision avoidance (LACA) which represents the minimum required lateral acceleration to avoid collision with a preceding vehicle. Moreover, it derives a control law to avoid a collision by using of braking and steering with consideration for tire force saturation, based on the values of deceleration for collision avoidance (DCA) and LACA. Finally, driving simulator experiments are performed to verify the effectiveness of the proposed method and its acceptability.

View full abstract

Ignition conditions and risks of ignition on a permissible value of hydrogen concentration in purge gas prescribed by HFCV-GTR were reevaluated. Experiments were conducted to investigate burning behavior and thermal influence of continuous evacuation of hydrogen under continuous purge of air / hydrogen premixed gas, which is close to an actual purge condition of FCV, and thermal evacuation of hydrogen. As a result of the reevaluation, it was shown from the viewpoint of safety that the permissible value of hydrogen concentration in purge gas prescribed by the current HFCV GTR is appropriate.

View full abstract

For the driver monitoring system in autonomous driving, it is effective not only to detect low-arousal level but also to keep adequate arousal condition by predicting low-arousal level in advance. We focused on driver’s acclimatization caused by monotonous visual stimulus during supervising a vehicle autonomous driving. In this paper, we quantified stimulus in visually recognizing such as a driving environment and display information so that we proposed a model that can predict the arousal level in advance.

View full abstract

Towards reduction of right-turn accidents at intersections involving pedestrians, we investigated whether change in driving behavior is effective to reduce collision risk against crossing pedestrians. Using a driving simulator, we conducted an experiment changing the driving behavior of participants intentionally by applying operation intervention methods while drivers made right turns at intersections. Experiment results revealed that collision risk against crossing pedestrians can be reduced by the change of drivers’ behavior in accordance with driving behavior indices which can evaluate pedestrian-collision-risk of right-turn maneuvers at intersections.

View full abstract

Three groups of drivers who were given different explanations respectively about driving support systems (DSS) drove a passenger car equipped with both ACC and LKA on an actual expressway. Drivers’ subjective evaluations showed that mental workload for using DSS increased with detailed explanations of DSS limitations, while explanations of visual and auditory displays for monitoring the DSS status decreased workload. These results suggest influence of introductory explanations of DSS to mental workload, and effectivity of HMI for monitoring DSS status to lighten it.

View full abstract

This study focuses on the effect of gradual transition from automated to manual driving on drivers’ behavior (at a time chosen by the driver). The experiments used a driving simulator; three types of gradual transitions were conducted based on the differences between the deactivation timings of the lane-keeping assist (LKA) and adaptive cruise control (ACC). In the first experiment, participants were instructed to conduct each transition while performing a visual task. In the second experiment, based on each of the transitions, drivers had to avoid a collision in congested traffic, which was found to occur just after switching to manual driving. In the present study, we discuss how each transition influenced driving performance and visual behavior just after the deactivation of automated driving.

View full abstract

A passenger seated on a 6-axis motion base has been exposed to vehicle sway that simulates lane-changing with the seat which changed the torso angle of the back rest and the position of the arm rest. The logistic regression result of the relationship between the subjective evaluation and the motion of the physical parts of the passenger measured with a motion capture system suggests that the angular velocity of skull is a major factor that influences subjective evaluation. In addition, it has been demonstrated that the armrest position influences the angular velocity of skull.

View full abstract

How to estimate the driver's emotion is an important topic and has attracted much attention. Some basic research results have been reported in literature, but many problems are still unsolved especially towards the practical application. In this paper, we propose a new method to estimate the driver’s emotion in positive, negative and neutral state via deep learning by driver’s biological signals, where the ECG(ElectroCardioGram) data sequence is changed to an image. The estimating accuracy is up to 90% by our initial experiment results. Furthermore, we show the features of emotion with the Grad-CAM method and find that they appear in T and U wave of ECG instead of R wave, which means a new explanation.

View full abstract

A previous study showed that some drivers felt a haptic display was not enough to inform them of the take-over request（TOR） in a highly automated driving system. This study aimed to verify the result of the previous study after increasing the frequency of haptic stimulus. Thirty drivers participated in a driving simulator experiment to investigate the objective of this study. The driver’s vehicle could automatically adjust the headway distance to a lead vehicle and keep its lateral position in a lane. In automated driving, drivers had to perform a non-driving task and avoid colliding with a stopped vehicle because of automated driving system malfunction. The results of the driving simulator experiment did not produce the same result as the previous study. The results also revealed that the haptic display might induce drivers to look at a frontal road scene before they checked a visual display presented in a car indicator with the haptic one. The human machine interface in highly automated driving was discussed in terms of adequate sensory modalities to inform driver of the take-over request.

View full abstract

This driving simulator study investigated a drivers' ability to resume manual control of their level-3 automated vehicle in response to system failure. Participants were asked to engage in non-driving task which required them to input text on a screen. A take-over request (TOR) was then provided under conditions where the TOR was and was not displayed on the task screen, and the effects with respect to system operation during the TOR were investigated. Regarding lateral operation, amount of lane departure was larger under conditions of suppressing control force rather than under conditions of continuing control force. Contrastingly, risk of collision was lower due to the driver’s suppression of longitudinal operation when the TOR was provided.

View full abstract