The complex phenomena of pre-chamber jet combustion and their mechanisms were discussed using the CFD Converge. The second half of the combustion period (MFB50-90) was more dominant for knocking than MFB10-50 due to the higher in-cylinder temperature. Pre-chamber jet combustion shortened MFB50-90 by forming a complex flame front due to jet penetration rather than strong turbulence caused by jet streams. The larger the pre-chamber volume, the faster the combustion in the pre-chamber, and the higher the pressure ratio between the main and prechamber. Even if the jet intensity was increased, there was a limit to the complexity of the flame front.

View full abstract

In the development of a mechanical Atkinson cycle engine, a lightweight, compact sliding mechanism with high resistance to seizure was necessitated. The design was fundamentally based on a sliding bearing as its primary structure. Subsequently, the optimal profile of the sliding surface was acquired. This refinement led to a reduction in local surface pressure during operation, thereby enhanced seizure resistance.

View full abstract

The "single-tooth pinion gear", which achieves high efficiency and high speed reduction, can be applied to speed reduction mechanisms with various specifications, just like the involute gear. In this paper, we examined the application of the single-tooth pinion gear to the "planetary gear mechanism", which is often used in automobile parts. As a result, we confirmed that it is possible to make the input and output shafts coaxial, increase the speed reduction, and reduce the size.

View full abstract

Recently, a safety concern of human exposure of electromagnetic field (EMF) for occupants of electric vehicles is increasing. The spatial resolution of the magnetic field sensor required for the validation was determined to be 5 mm based on the simulation. The repeatability of the small magnetic field sensor was verified using a system for generating AC magnetic field standards developed in National Institute of Advanced Industrial Science and Technology (AIST). A simulation model of a real component (battery) installed in a vehicle was improved based on measurements, and it was shown that the magnetic field level could be simulated within 3 dB.

View full abstract

Lateral forces acting on the tires generate jack-up/down forces and the forces raise or lower the vehicle body depending on the arm angles of the suspension. However, the effect and mechanism of suspension jacking force on vehicle behavior had not been clear, and it had not been possible to determine whether the jacking characteristics are appropriate or not. In this paper, a vehicle model that incorporates suspension's kinematics equivalent 3-link model into a 5-DOF model was constructed. Using the model, simulation analysis were conducted with various combinations of roll center height and virtual arm lengths and the effects of the jacking characteristics caused by transient steering on the planar motion were clarified.

View full abstract

In the TM Tire Model, we altered the tread model with the aim of improving the cornering stiffness approximation accuracy for a wide range of tire sizes. The altered tread model made it possible to identify the tread stiffness with a value closer to that of a more realistic tire. However, the altered of the tread model not only improved the identification accuracy of the tread part, but also affected the approximation accuracy of the belt part and sidewall part. We considered the effect of the altered of the tread model on the accuracy of identification of the belt and side parts, and verified the effect on the calculation accuracy of the vehicle motion characteristics.

View full abstract

Road traffic noise is one of the most important issues. A more accurate prediction model is necessary to incorporate measures for reducing this noise. Current simulation models like the ASJ-RTN utilize fixed vehicle sound models. To enhance accuracy, an updated vehicle sound model, is recommended. This paper proposes a method for creating vehicle sound models with minimal measurements, applicable to all vehicle categories. Additionally, it presents a case study comparing the impact on exterior noise of a 100% electric drive hybrid system vehicle and an ICE vehicle, using the sound model developed through this method.

View full abstract

Since the millimeter wave sensor can detect micro vibrations，it is possible to acquire the respiration and heartbeat of the human body. As a preliminary step to acquire such information inside the vehicle，we developed a device that detects seats where people sit. In this paper，we describe the configuration of this device and the way to deal with range resolution and multipath，which were problems in detection.

View full abstract

In order to study the relation of safety level between automated driving vehicle and human driver, analysis of driving behavior of human drivers is necessary. In this research, in order to study driving behavior of human drivers in some conceivable traffic scenes in which it is difficult for automated driving vehicle to avoid a collision, the experimental test by using the driving simulator was conducted.

View full abstract

To clarify the elderly drivers’ usage situation of Advanced Safety Vehicles (ASV), and their characteristics, we analyzed the result of the questionnaire survey of elderly drivers nationwide (N = 1,657). The major findings are as follows. 1) The indifferent group, which had experienced a minor collision in the past 10 years and had no intention to change to ASV, accounted for 11%. 2) The indifferent group tended to think their driving was unsafe and did not follow traffic rules. 3) Result of the characterize elderly drivers by using CCDC, we confirmed that the indifferent group has the characteristics of risk-taking.

View full abstract

A new method for predicting vehicle energy consumption corresponding to a reference driving cycle using OBD signals as real-world driving data has been developed. Based on the scatter plot of vehicle speed and acceleration of on-road driving data and WLTC, “feature plots” which are the on-road data points that match the refence driving data points are extracted, then the WLTC equivalent energy consumption can be predicted by accumulating the instantaneous energy consumption at the individual extracted feature points. In an evaluation test, WLTC energy consumption predicted by this method was 2% greater than the WLTC reference value.

View full abstract

In addition to measurement by on-board emission measurement system, analysis by a traffic flow simulator is one of the most effective ways to identify pollution locations on actual roads. In this study, the real-world emission behavior of a direct-injection gasoline vehicle was analyzed and a emission prediction model for a traffic flow simulator was constructed based on this analysis. The result of the emission prediction using vehicle speed, acceleration, and road gradient showed that the CO2 prediction was accurate, but the NOx prediction needed to consider the fuel cut timing to improve prediction accuracy.

View full abstract

Oxidation enhancement by a Three-Way Catalyst (TWC) porous particles membrane filter was investigated through a comparison with a TWC solid particles one. The TWC porous particle was fabricated through agglomeration of TWC and PMMA nanometer-sized particulates and then pyrolysis of PMMA. It was confirmed that some pores are connecting from the surface to the center of the porous TWC particle by observation of the cross-sectional area using a scanning electron microscope. Using a temperature programed oxidation experiment by only CO (44 ppm, N2 balance) supply, the light-off temperature defined as a 50 % oxidation temperature by the TWC porous particles membrane filter is 149 °C which is lower than 182 °C by the TWC solid particles membrane filter though the total weight of TWC in the case of porous particles membrane is approximately 5 times smaller than that of solid one. This is because the specific surface area of the porous particle is more than 2 times larger than that of the solid one and all TWC material will be used effectively while the TWC material emerged inside of the particle is not used in the case of solid particle.

View full abstract

It is expected to realize safer autonomous driving control by collecting CAN data from autonomous vehicles in a Cyber-Physical System to achieve advanced self-driving. Therefore, real-time CAN data collection from autonomous vehicles to CPS is one of the most challenging problems in the current IoT. In this paper, I propose an MEC (Multi-access Edge Computing) enhanced network Redundancy Elimination service for CAN data collection. As a result of simulation evaluation, the proposed method can reduce the network traffic of CAN data by 63.2% in the mobile network and by 86.9% in the Internet.

View full abstract

Mechanical Atkinson cycle engines, which are composed of mechanisms and structures different from conventional mechanisms, have issue with friction and vibration depending on the force characteristics of the mechanism. An analysis model was constructed to reduce friction and vibration. As a result of manufacturing a prototype engine and verifying the actual machine, we confirmed the reduction as predicted, and we reduced friction and vibration significantly.

View full abstract

The atomization method by spraying and droplet impingement of the CO₂-absorbing liquid into the exhaust pipe was investigated as a new lowcost emission technology and a new approach for CO₂ absorption. First, it confirmed the theoretical possibility of CO₂ absorption by an aqueous MEA solution. Next, CO₂ absorption experiments were carried out with the surface dimple texture on impingement plates. Furthermore, the relationship between the absorption rate of the aqueous MEA solution and the total surface area was investigated from the droplet size and particle number analysis results after atomization, and the possibility of CO₂ absorption by droplet spraying into the exhaust pipe was investigated.

View full abstract

The use of exhaust gas recirculation (EGR) is necessary to improve the thermal efficiency of spark-ignition engines. However, deposits formed in the EGR valve or cooler cause valve sticking and reduced heat exchange performance of the cooler, which are serious problems. In this study, the formation mechanism of EGR deposits was investigated using the gasoline engine exhaust and an EGR deposit model was constructed. As a result, it was found that the deposit was formed under the conditions where the wall temperature was below 200 ℃, and the amount of the deposit increased with decreasing the wall temperature between 80 and 200 ℃. The developed model predicted the experimental results well.

View full abstract

Battery management, which controls battery temperature and SOC, is important for safe and efficient long-term use with low degradation of lithium-ion battery in the electric vehicle. To design a suitable battery management system, it is effective to use a vehicle system 1D simulation model. We studied an appropriate battery degradation model for the vehicle system 1D simulation model based on lithium iron phosphate battery test data. Implemented physical model of anode SEI layer growth and electrode structural disorder to predict calendar and cycle degradation respectively. After implementing the degradation models, we analyzed the long-term degradation behavior by assuming a real-world use case of electric vehicle driver. Evaluated the contribution of calendar and cycle degradation to the real-world degradation.

View full abstract

Driving on rough roads in off-road vehicles, the suspension characteristics that contribute to ride comfort may conflict with the magnitude of the load from the road surface due to overcoming rocks. In order to balance multiple different performance indicators, the optimal combination of design variables has been searched by trial and error. In the early stage of vehicle development, it is necessary to consider not only the characteristics of each part in the vehicle but also the external environment of the vehicle in the design variables. In this paper, Bayesian Active Learning is adopted to obtain the feasible region about the design variables for off-road vehicles. A practical numerical example of a multi-disciplinary vehicle design problem is demonstrated, and it is represented by the effectiveness of the proposed method.

View full abstract

The authors previously developed an intermediate evaluation index for transfer characteristics of the structural system of the completed vehicle to enhance the sound quality of combustion noise. Based on the results, a method was developed to obtain an index for the power unit similar to the above intermediate evaluation index by a shaker test on a stationary power plant. It was found that this shaker test and transfer characteristics evaluation method can approximately estimate the frequency band in which the structural system amplifies half-order components of the combustion noise related to sound-quality deterioration without conducting an actual engine operation test.

View full abstract

To predict vehicle occupant injuries based on accident information obtained from a video-recorded drive recorder, injury prediction models were constructed by machine learning using information currently available from drive recorders and information that is expected to be acquired hereafter. Light Gradient Boosting Machine and Bayesian Networks were used for the machine learning models. The results showed that it was difficult to predict injuries with only the information currently available from a video-recorded drive recorder, while the prediction model with additional information expected to be available hereafter improved the prediction accuracy.

View full abstract

In order to handle free speech while driving, we constructed spoken dialogue system combining a question-and-answer database and a largescale language model. We analyzed user behavior and the usefulness of the system by collecting utterances using an actual vehicle. Among the utterances that could not be answered by the conventional system, the large-scale language model was able to respond to more than half of them. This indicated the effectiveness of the introduction of the large-scale language model.

View full abstract

This study explored multitasking among elderly drivers, whose cognitive function declines with age. Experiment 1 demonstrated that elderly drivers struggled more with simultaneous calculation and driving tasks compared to younger drivers. Experiment 2 analyzed individuals unaffected by multitasking and found that the elderly drivers in this group experienced more mind-wandering when driving alone. The findings highlight the unique challenges faced by elderly drivers in managing multitasking situations, providing insights into their cognitive demands and performance.

View full abstract

We proposed a method to evaluate the degree of safe driving based on a cognitive test utilizing the Trail Making Test-J Part B (TMT-J) and a minimum TTC using a driving simulator. We also proposed a training program to acquire compensatory behavior through metacognition. The effectiveness of these methods was tested on 30 elderly drivers. It was confirmed that a combination of TMT-J and driving ability was effective in judging safe driving. In addition, the use of charts comparing driving ability with others and a bird's-eye view of driving improved metacognitive abilities and realized compensatory behaviors for safe driving.

View full abstract

Quietness is one of the key factors of quality and competitiveness of a vehicle, and noise reduction is one of the main objectives in vehicle development. In recent years, with the increasing popularity of HEV (Hybrid Electric Vehicle) and BEV (Battery Electric Vehicle), innovations in vibration and noise reduction technology have led to significant reductions of engine noise, road noise and motor noise. As a result, contribution of exterior wind noise has become more dominant, particularly in high-speed cruising conditions. Exterior wind noise is generated by acoustic pressure fluctuation induced by turbulent flow around a vehicle, and it propagates and transmits inside the cabin. In this paper, we propose a method for predicting exterior wind noise and identifying noise sources based on simulation results of the flow field around vehicle and show its validation by comparison with the results of wind tunnel evaluation.

View full abstract

In this research, we construct a machine learning model that can predict the drag coefficient based on an existing physics-based machine learning model. The base model can predict velocity and pressure fields accurately thanks to the physical knowledge embedded. Our novelty is to add a model that computes the drag coefficient inside it rather than postprocessing for more accurate results. Also, we generated a dataset using aerodynamic simulation with various shapes generated based on the DrivAer model. The model shows high accuracy with an error of 0.0025 in the drag coefficient for the considered dataset.

View full abstract

In recent years, demand for high-tensile strength steel sheets has been increasing from the viewpoints of energy conservation through weight reduction and the upgrading of product functions. However, since such high tensile strength steel sheets contain martensite, it has been reported that material strength is greatly reduced in the Heat Affected Zone (HAZ) due to tempering during laser welding. In other words, when laser welding is applied to high-strength materials, there is a possibility that rupture will be apparent in the HAZ softened zone. In this study, we evaluate the strength of a 1500 MPa steel sheet, which is subject to HAZ softening, by applying tension in any direction to the laser weld line and rupturing it. The results show that a welding angle of 20 ° significantly improves elongation and strength. If it is possible to minimize the strength loss due to HAZ softening by using these results, the welding portion of high-strength materials can be located where tensile deformation occurs during forming or collision, thereby increasing design flexibility.

View full abstract

In vehicle system development, complying with ISO 21448 and ISO 26262 grows in demand. However, the challenge lies in realizing an efficient safety analysis method that can comply with both standards by effectively sharing their respective parts. To address this challenge, we focus on STAMP/STPA. STAMP/STPA is applicable to safety analysis in the concept phase of the product development process, and examples of the application of STAMP/STPA are provided in the appendix of ISO 21448. In this paper, we propose an extended STAMP/STPA that can efficiently comply with both international standards.

View full abstract

Logical execution time (LET) paradigm has been proposed as efficient design method for multicore software development. It is also well known that LET has a disadvantage to increase latency from input to output. In this research, we propose a solution named Flexible LET, which is extension of LET paradigm. Flexible LET enables inter-LET communication based on data age required by LET-based applications to reduce latency. We developed a prototype on AUTOSAR Adaptive Platform and evaluated the performance. The evaluation results showed that our proposed method is feasible.

View full abstract