Hydrogen has combustion characteristics such as a wide flammable range, fast burning velocity and low minimum ignition energy. The objective of this study is to reduce cooling losses and improve brake thermal efficiency due to hydrogen addition using a PI system SI engine in which hydrogen is injected into the intake pipe. This paper investigates the effects of rapid combustion by the PI system on combustion and exhaust characteristics. The results show that brake thermal efficiency increases with increasing hydrogen addition ratio, but cooling losses increase at higher hydrogen addition ratios.

View full abstract

For improving thermal efficiency of K-car engine under usual operating conditions, the authors explored the optimized engine parameters applying quality engineering tool for 0-D engine cycle simulation model. In this simulation study, the potential for improving thermal efficiency was clarified by exploring various engine geometrical parameters such as bore and stroke, compression ratio, connecting rod length and intake valve timing, etc. regardless of the engine displacement standard for K-cars. The results indicated that super high expansion ratio engine which has the piston swept volume beyond 660 cm3 significantly improved the thermal efficiency for K-car engines.

View full abstract

The effects of amplitude and frequency on purification characteristics were investigated using a perturbation technique that switches between rich and lean at high speeds. It is thought that increasing the amplitude increased the amount of oxygen inflow, reducing the amount of available sites and decreasing the purification rate. It is also thought that increasing the frequency kept the amount of free sites high and reduced the amount of slip of the three chemical species.

View full abstract

Waste heat recovery (WHR) is important to improve fuel efficiency of hybrid electric vehicles. In this study, WHR systems with phase change cooling (PCC) concept were applied to a series hybrid electric vehicle with a lean burn SI engine. The effects of combining organic Rankine cycle and PCC (PCC-ORC) and combining thermoelectric generators and PCC (PCC-TEG) on brake thermal efficiency and fuel efficiency were evaluated by using one-dimensional numerical simulation. PCC-ORC was more effective than PCC-TEG; the improvement of brake thermal efficiency was 2.78%pt at the steady condition and the improvement of fuel efficiency was 4.87% in WLTC assumed mode.

View full abstract

Hydrogen is mentioned as a power source toward carbon neutrality, and dual-fuel combustion system is effective to apply hydrogen to engine system because of its redundancy. In order to improve performance and avoid abnormal combustion of hydrogen engines, it is necessary to clarify the combustion forms. In this study, the analysis of hydrogen/diesel combustion process was carried out by using the hydrogen/diesel dual-fuel engine and conducting 3D-CFD calculations based on experimental results.

View full abstract

A piston made of aluminum alloy needs monitoring and control system of the temperature for long lifetime. A method to estimate piston temperature for diesel engines was developed. It revealed by CAE and temperature measurement that change of piston temperature is linearly and independently by influential factors as start of fuel injection, EGR, cylinder intake gas temperature and cooling oil jet cooling. The method can express transient temperature change by 1st order lag. Using this method, piston stress and lifetime on a market are predictable. Moreover, piston temperature control combined with this method and oil path circuit control can reduce fuel consumption at during transient operation.

View full abstract

Hydrocarbons with blending octane number (BON) higher than the octane number (ON) measured as a single component fuel are considered to have a large effect on improving the octane number of the mixed fuel when mixed with other hydrocarbons. In this study, the characteristics of the reaction of high BON hydrocarbons in mixed fuel were discussed based on the analysis of reaction products in a flow reactor. When mixed with primary reference fuels, hydrocarbons whose BON is higher than ON suppress the reaction of the mixture. This effect may be caused by the production of reaction intermediates at lower temperatures than when used as a single component fuel, or by the production of reaction intermediates that were not observed when used as a single component fuel.

View full abstract

The carbon deposit in gasoline engines have various influences on engine performance and hinder long-term operation. They are generated from carbon in fuel oils and lubricating oils. It is important to classify the cause of carbon deposit broadly in research & development and investigate the countermeasures to reduce them at an early stage. In this study, we report on the development of a method to quantitatively indicate the specific causes in gasoline engines for all deposit that affect engine performance.

View full abstract

By dividing combustion chamber, the homogeneous charge compression ignition (HCCI) combustion can be achieved the multi-stage. The combustion system has two parts of combustion chamber, one is small inside piston and the other is large. At first, ignition takes place in the small chamber inside piston, and then the burned gas ejected into the large chamber to ignite the mixture in the large chamber. This combustion system is promising to reduce maximum pressure rise rate so can be achieved high load operation without knocking. The 3D-CFD results have shown that over 50% (maximum 51.4%) of indicated thermal efficiency at IMEP 1.1 MPa operation with exhaust gas recirculation (EGR) and fueled with iso-octane and iso-octane mixed with ethanol 10vol.% and 15vol.% (E10 and E15, respectively).

View full abstract

Thermal management for the lithium-ion battery in the electric vehicle, which controls the battery temperature, is important to counteract power output, degradation and overheating of the battery. To design an optimal thermal management system, it is effective to use a vehicle system 1D simulation model. We studied several thermal management systems for the battery electric vehicle and evaluated the performance of the system based on this simulation model. First, we implemented battery thermal model using battery temperature measurement test in the vehicle. After implementing the battery thermal model, we analyzed the battery warm-up behavior at cold start with each thermal management system proposal and evaluated battery warm-up performance in terms of warm-up time, electrical energy and heat transfer energy.

View full abstract

During the initial planning stages for selecting HEV systems, fuel efficiency is a crucial metric. Predicting this performance through simulations is vital. However, the vast number of different system configurations, each requiring unique plant and control models, makes comprehensive simulation impractical due to the extensive workload. Therefore, in this paper, we introduce a case where the simulation model was simplified based on the inherent functions of HEVs, and target allocation was conducted.

View full abstract

The fuel cell hybrid test train requires methods to control fuel cell output power. The acceleration and regenerative-brake performance of the train varies with the SOC of the battery. And the energy consumption also varies with the passenger occupancy and the ambient temperature. Therefore, we studied the traction performance and fuel consumption depending on the difference of the control methods and the number of parameters to be considered with using a driving energy simulator.

View full abstract

This study investigates the optimal trajectory of competition vehicles as an optimal control problem and generates it through optimization computations. For a course layout with a mixture of straight and curved shapes, we apply an intra-convex domain constraint method to ensure the vehicle remains within the given course. Curved shapes, which are concave domains, map to convex domains by complex mapping. Numerical results confirm the effectiveness of our approach in generating optimal trajectories.

View full abstract

Automated driving control technology has introduced as a mobility solution. This research shows a real-time vehicle control technology that considers conditions on a driving route several kilometers ahead using model predictive control to improve movement efficiency with avoiding deviations from operational design domain, which is defined for the autonomous vehicle, due to changes in driving environment conditions.

View full abstract

The deflection behavior of a 660mm square, 13.6mm height sandwich panel was investigated by the finite element method for application to a slope of a welfare vehicle. When loads were applied simultaneously to multiple positions on the panel assuming that a wheelchair with a care recipient in it and a caregiver were moving on the panel, the deflection at each loading position was greatest at the center of the front and rear support edges of the panel. The maximum deflection occurred at the position of the caregiver’s feet, regardless of the weight of the wheelchair. This maximum deflection decreased as the inclination angle of the panel (φ) increased, and had a linear relationship with cos2φ. From the deflection when only the caregiver’s feet are loaded, it was possible to estimate the maximum deflection when multiple loads are applied.

View full abstract

To design the vibration performance of a car, it is effective to divide the structure into multiple subsystems. And the kernel Compliance Analysis (kCA) based on the frequency based substructuring is known as an effective analysis method to understand the coupling among plural subsystems. When the kCA is applied to the body shell of a car, there are many degrees of freedom in the coupling region, which makes the analysis difficult. In this paper, we proposed an analytical method and a design guideline for vibration reduction of body shells. In the method, the transmitted forces at the coupling region are expressed using wavenumber spectrum as a simple indicator to quantify the magnitude of vibration transmission. By using the indicator, complex vibration transmission can be understood in terms of the spectrum of transmitted forces in wave number space. This is useful for simplifying and understanding the mechanism of the coupling phenomena and linking them to design guidelines.

View full abstract

This paper employs the SlowFast deep neural network to classify near-miss incident videos, significantly enhancing accuracy through emulation of the human visual system's separation into slow (P-cells) and fast (M-cells) processing two streams. By delaying the analysis time of near-miss incident, more obvious features of near-misses are exposed, making it easier for the model to learn near-miss features. This methodology demonstrates the potential for early driver notification to prevent accidents, thereby encouraging progress in traffic safety.

View full abstract

Retrieving relevant traffic scene data from existing database is essential in the development of advanced driver-assistance systems but such task is time consuming and computationally expensive. This study proposes a traffic scene retrieval system that utilizes a vision-language model and clustering techniques. The proposed system is capable of executing data retrieval task by inputting an image data or text as a search query. Evaluation results showed that the system was able to retrieve complex scene data(e.g., traffic congestion) from a driving video database under 3 seconds. Overall, the results indicate that the prosed system is feasible for practical applications.

View full abstract

In the present study, we proposed two types of e-HMI that promotes safe and smooth pedestrian behavior by measuring pedestrians' crossing and eye movement at pedestrian crossings where multiple vehicles are approaching. "Crossing icon" on the windshield encourages pedestrians to cross, and "Brake light" turns green when a car slows down. The experiment was conducted with 36 participants using a VR pedestrian simulator. It was found that the proposed e-HMI increased the sense of safety for many of the participants, and some participants were able to cross the street smoothly.

View full abstract

This study aimed to classify elderly drivers based on subjective evaluations using the Simplified Questionnaire Comprehensively Captures Driving Characteristics (SQ-CCDC). A web-based survey of 1,377 participants aged 60-80 was conducted. Multiple correspondence analysis of the SQ-CCDC subscale ranks revealed two dimensions: compensatory behavior awareness and safe driving aptitude. Cluster analysis identified five distinct groups of elderly drivers. The results suggest that safety-oriented attitudes and emotional control ability may influence traffic offences experiences. Gender differences were also observed in driving patterns. This classification approach using SQ-CCDC provides insights for developing personalized and effective safe driving support strategies for elderly drivers.

View full abstract

FVM analysis technology is a simulation technology for dew condensation, but although this analysis method can predict the occurrence of dew condensation, it cannot reproduce the flow of condensed water caused by factors such as vehicle tilt and driving acceleration/deceleration. We developed the reproducing method of condensation water on the air conditioner ducts by adapting MPS method. The simulation results of the condensation water flow are similar as the actual vehicle test cases with vehicle tilts and driving situations.

View full abstract

To prevent Global warming caused by CO 2, multi material automotive structures including CFRP are expanding due to the vehicle weight reduction. Therefore, “CFRP joining Working Group” of Fatigue Reliability Committee in JSAE is studying CFRP structure with the bonding joint. In this report, the fatigue strength of urethane adhesive joint is studied by lap shear joints and specime ns simulated car body structure. As results, fatigue properties such as crack initiation and propagation at adhesive joint were clarified. Fur thermore, the criteria of fatigue strength were studied to evaluate the adhesive joint of body structures.

View full abstract

This study provides a current dispersion method that avoids local current concentration in automotive power network to reduce weight of power supply line for automotive power network compatible with zone architecture. This paper shows an algorithm that optimizes power supply route to each zone using dijkstra's algorithm. As a result of simulation evaluation, this method can reduce peak current of the power supply line by 30%.

View full abstract

This study reports on the prediction of 4G LTE uplink throughput in vehicles using LSTM (Long Short-Term Memory) networks. To make accurate predictions, data was collected less than two years in Japan, enriching the dataset. LSTM was chosen for model selection due to its proficiency in time series processing. The prediction method used the last 15 seconds of data and historical data of the location in the next 20 seconds to predict the throughput in the next 20 seconds. The relatively intuitive MAE and MAPE were used as the prediction evaluation metrics.

View full abstract

Supercharging is very efficient to improve thermal efficiency and reduce emissions of diesel engines. Especially, electrical supercharger development has significantly progressed in recent years and it enables high supercharging pressure operation which could not be realized by conventional turbo engines. However, it is suggested that excessive supercharging has a negative impact on indicated thermal efficiency by zero-dimensional combustion simulation. In this study, verification experiments using a single cylinder test engine were carried out to confirm the decrease of thermal efficiency by excessive supercharging arise. Furthermore, the configuration of superchargers was investigated to realize the best excess air ratio condition on the test engine.

View full abstract

In the cornering behavior of a vehicle, the differential connecting the left and right wheels is not negligible. In this study, to clarify the handling characteristics of a rear-wheel-drive vehicle in the locked and slipping states of the differential including lateral load transfer, derivatives for yaw damping during cornering and yaw moment change when a driving force is applied were derived from a linear 2 degrees of freedom 3-wheels vehicle model, and their effects on handling characteristics were discussed.

View full abstract

Vehicle electrification induces changes in specifications, including a lower center of gravity, reduced pitch and yaw inertia moments, and a rearward shift in the center of gravity. This paper investigates the impact of differences in specifications between internal combustion engine vehicles and electric vehicles on dynamic characteristics. Results from comprehensive vehicle simulations and driving simulator experiments assessing steering performance are presented, providing insights for the improvement of electric vehicle design and driving dynamics.

View full abstract

In the turning limit, front-wheel-drive vehicles tend to plow, and rear-wheel-drive vehicles tend to spin, resulting in consistent vehicle behavior. On the other hand, four-wheel-drive vehicles have a small difference in the workload between the front and rear tires, and their vehicle behavior is inconsistent. In this paper, we proposed a design method for front/rear driving force distribution based on tire workload in order to control vehicle behavior in the turning limit.

View full abstract

The matrix-based methodology "Design Structure Matrix (DSM)" which enables to evaluate dismantling performance of automotive structures is introduced. After the component of DSM and the numerical operation of its elements to make the evaluation graph shall be explained, the systematic scheme of matrix re-ordering and expressions of order constraint would be proposed. Finally, practical dismantling example of automotive door structure with modular design shall make us convinced that the DSM is reasonable and intuitionally comprehensible tools.

View full abstract

Combination of complex driving operations and vehicle behavior by expert drivers are necessary for drivability performance improvement. In this study, an evaluation method using Machine Learning was developed to reduce extensive testing efforts and sharing knowledge of experts. Furthermore, the method is applied for engine bench system. This utilization involves acceleration behavior predicting from ECU data, conducting independently from the actual vehicle evaluations. Additionally, this method has also been applied to actual vehicle evaluations.

View full abstract

Aiming at the harmony with the global environment, exhaust gas clean of the ICE has to promote becoming it parallel to EV, HEV and CN fuel development. However, if the catalyst is warmed up early for the environment, the combustion stability decreases, and torque change occurs, and the vibration of the car body gets worse. We took a lot of person-hours for these performance coexistence in the past. Therefore, we aimed at MBD for car body vibration during catalyst warm-up, and we created physical models of the exciting force and the sensitivity of the vehicle body and verified them with the actual device to analyze factors. By integrating these physical models, we achieved the optimization of development by constructing the comprehensive MBD from combustion waveforms to vehicle body vibrations analyzed by frequency. This paper introduces the details.

View full abstract

In this study, the characteristics of the power plant eigenmodes that deteriorate the sound quality of engine combustion noise are analyzed by experimental modal analysis. The results showed that the power plant eigenmodes amplifying the half- and odd-order components of engine combustion noise are caused by elastic deformation of the engine, especially deformation of the main bearing caps. Moreover, when the elastic deformation of the engine is coupled with the deformation of the mounting bracket, which is the evaluation point for the order-based frequency response function, the half- and odd-order components of combustion noise are further amplified.

View full abstract

The time-frequency analysis images are classified based on a machine learning model that combines a Convolutional Neural Network and a Quasi-Recurrent Neural Network, and features in each frequency component of engine sound with time variability are visualized by using Gradient-weighted Class Activation Mapping, which is a middle layer extraction method of a neural network.

View full abstract

The Emergency Driving Stop System (EDSS), which can stop a vehicle when a driver's sudden illness occurs while driving a vehicle by pressing the switch, is beginning to be adopted for large commercial bus etc. This study analyzed the operation behavior of pressing the emergency stop switch using a three-dimensional human postural skeletal detection system, assuming the posture of a driver in a sudden physical condition, and clarified the postures that affect the ease of pressing the switch from an ergonomic viewpoint. As a result, the multiple regression analysis for upper body posture on ease of pressing indicated that the influential factors were forearm angle, upper arm movement, and wrist movement in the z-direction. Considering the situation of sudden change in physical condition and the posture that facilitates pressing, the EDSS switch should be located as close as possible to the driver, and the center of the handle would be suitable.

View full abstract

Accurate self-localization is important to realize safe autonomous driving. Conventionally, millimeter-wave radar has been used for self-localization to ensure robustness against changes in the surrounding environment, such as snow and heavy rain. But there is an issue of estimated accuracy due to low resolution. This study aims to improve accuracy of self-localization while maintaining high robustness against environmental changes by implementing localization module which uses 4D imaging radar with high resolution.

View full abstract

Localization is important to enable safe autonomous driving. In previous studies, localization based on infrared reflectance by LiDAR has been proposed and tested. However, when environmental changes occur, the reflectance from the road surface decreases due to rainy weather and blurring of white lines. Then, the localization accuracy is reduced. To solve this problem, previous methods use sensor fusion with LiDAR and a camera or Millimeter Wave Radar (MWR). However, they do not increase the robustness of map matching itself. Therefore, the objective of this research is to achieve robust matching against environmental changes by using Convolutional Neural Network (CNN).

View full abstract

The purpose of this study was to clarify the occurrence and mechanism of thoracic injuries in frontal collisions at moderate impact velocities. In the accident case study, AIS 2-3 sternum and rib fractures were identified in frontal collisions at 30 to 40 km/h. The results of FE simulations showed there were differences in the predilection site of plastic strain in the ribs due to differences in body size even for the same crash pulse. The plastic strain distribution was almost the same for the same body size even at lower impact velocities, and the plastic strain values were also similar. The left third rib was subjected to not only bending around the superior-inferior axis of the rib but also torsion around the longitudinal axis, which caused plastic strain in the rib even at relatively low impact velocity.

View full abstract

This study deals with road friction in snowy countries in winter, which is one of the major issues in traffic safety, as a first step toward building a forward road friction estimation system. To achieve this purpose, in previous research we proposed a new method that can continuously measure the peak μ in μ-s characteristics and presented measurement results on different road surfaces. Based on these results, in this paper we measured the friction characteristics of snowy and icy surfaces maintained at proving ground, and showed the continuous μ-s characteristics on the surfaces and the peak μ region on each road surface. Furthermore, measurements are taken on ordinary roads under snow-covered conditions, and comparisons with these areas showed that the dynamics of μ-s characteristics should be considered.

View full abstract

Accident data shows that pedestrians can sustain injuries from both the vehicle and the ground in vehicle-pedestrian collisions. To mitigate ground-related injuries, multi-body simulations of vehicle-pedestrian collisions were conducted, and vehicle braking deceleration after contact was optimized to minimize pedestrian injuries from impact with the ground. The optimized vehicle braking deceleration profile allowed the pedestrian’s torso rotation angle to be less than 90 degrees, depending on the vehicle shape.

View full abstract

Most injury prediction algorithms used for determining injury severity in Advanced Automatic Collision Notification Systems (AACN) do not include the effect of emergency transport time, that is, the time taken to transport an injured person from a road crash to an emergency trauma centre. Therefore, in this study, we developed an injury prediction algorithm that incorporates the influence of emergency transport time as a risk factor. The base model of the injury prediction algorithm was constructed by applying logistic regression analysis to the South Australian Traffic Accident Reporting System (TARS) data. The TARS data, which are statistical data on traffic accidents, do not contain time-related data. Therefore, we quantified the effect of transport time on the fatality and serious injury rate as odds ratios using the Serious Injury Database (SID) derived from a trauma centre. Finally, the odds ratios were converted into regression coefficients and mixed with the base model to construct an injury prediction algorithm that takes into account the transport time.

View full abstract

The purpose of this study is to evaluate the primary energy consumption and GHG emissions of hydrogen, synthetic fuels, and biofuels in the Well-to-Wheel (WtW) process. The passenger cars considered in Tank-to-Wheel include gasoline vehicles (ICEV-G), diesel vehicles (ICEV-D), hybrid electric vehicles (HEV), battery electric vehicles (BEV), and fuel cell electric vehicles (FCEV). The results show that Ethanol from sugar beet has the lowest GHG emissions of the other fuels. It can reduce GHG emissions by 93% compared to conventional petroleum-based fuels. In addition, ICEV (ICEV-G, ICEV-D, HEV) powered by e-fuel or biofuels had lower GHG emissions in WtW than BEV. Therefore, when using carbon- neutral fuels, ICEV is a more effective means of reducing GHG emissions than BEV.

View full abstract

In order to develop a retrofit compatible ammonia mixed combustion engine for automobiles, we utilized the property of ammonia to dissolve easily in water and used a "water mixed fuel generation device" developed by the National Fisheries University to produce ammonia water mixed fuel. An experiment was carried out in which ammonia was injected as a liquid at normal temperature under atmospheric pressure. To clarify the combustion characteristics and DPF performance when ammonia is used in a small high-speed engine, 214kW/3,101min-1 diesel engine with a DPF was operated using a fuel made from a mixture of gas oil and ammonia water, and the engine performance and characteristics of NOx, PM, and DPF were analyzed. As a result, it was clarified that ammonia can also be used in small high-speed engines.

View full abstract

The body weight of heavy trucks varies greatly depending on the amount of cargo. In order to achieve platooning driving, the vehicle must run safely without being affected by changes in vehicle weight. In this study, a control method that constructs constraints using a control barrier function is used, and constraints that suppress acceleration to ensure that this method can withstand the effects of vehicle weight changes are investigated. Computer simulations were also carried out to confirm the degree to which this method can withstand vehicle weight changes during formation driving, and the effects of body weight changes were discussed.

View full abstract

In order to accelerate the social implementation of autonomous driving, it is important to repeat the process of technological development, environmental improvement, and evaluation of social acceptance in a short demonstration cycle. In this study, we present an installation assistance system that automatically calibrates the installation posture of infrastructure sensors to speed up integration of geofenced autonomous driving. The calibration pipeline is composed of multi-point search and iterative closest point. Additionally, we propose grid-based hidden point removal, which improves robustness of the multi-point search while reducing execution time. The proposed algorithm is evaluated with actual infrastructure sensor data.

View full abstract

In order to improve the prediction accuracy of noise and vibration performance of vehicles with two-motor hybrid systems during the performance design phase, a 1D system model that can calculate the driving conditions of the vehicle was developed. In addition, an interior noise model was developed that can reproduce power unit noise/road noise/wind noise in an integrated manner. By passing engine speed, engine torque, and vehicle speed from the 1D system model to the interior noise model and coupling them, it is possible to predict the interior noise in conjunction with the power unit control study.

View full abstract