Introduction of internal EGR (Exhaust Gas Recirculation) is effective for improving the efficiency of ICE (Internal Combustion Engine), though the combustion stability usually limits its maximum amount. In this study, we modelled the dynamics of the cycle-to-cycle variation of the HCCI(Homogeneous Charge Compression Ignition) combustion, designed a controller that actuates fuel injections based on the dynamics, and evaluated it on real engines. The experiment of the controller showed the improvement of the fuel efficiency and possibility of extending the limit of the Internal EGR amount.

View full abstract

In addition to the conventional three-way catalyst (TWC) for CO, NOx, and HC, a gasoline particulate filter (GPF) is now required for filtration of gasoline soot in the exhaust aftertreatment system of gasoline direct injection (GDI) engines. For the space reduction in the engine room especially of hybrid cars which is combined with both an internal combustion engine and a battery-powered electric motor, we need the aftertreatment system coupled with TWC and GPF. However, deposited soot inside the filter may degrade catalytic activity. In this study, by using a lattice Boltzmann method (LBM), we conducted numerical simulations of catalyzed GPF to discuss the effect of soot layer on the conversion rate of TWC. We investigated the effects of the inlet temperature and the filtration velocity on the catalyst performance. Results show that as the inlet temperature is higher or the filtration velocity is lower, the conversion rates of CO, HC and NO are enlarged. When the soot layer is formed on the surface of the filter wall, these conversion rates are all reduced. the filter. These findings would be important for determining the period of filter regeneration.

View full abstract

We proposed a method to detect the opening/closing timing of the needle valve of diesel injector by measuring the vibration occurred during the operation with an accelerometer attached to the injector-body surface. Simultaneous measurements of the vibration and needle lift were conducted on the injection rig test equipment. Obtained results show that by measuring vibrations due to the mechanical contacts of a control valve or a needle valve in a solenoid diesel injector using an accelerometer, it is possible to detect the typical timings of needle valve motion.

View full abstract

We developed a driving energy simulator to study fuel cell output control of a hybrid test train. The simulator takes into account that the powering and regenerating performance of the test train depends on the SOC of the battery. The simulator was validated by confirming that it matched the driving test data. And the running simulation was carried out on a 15 km virtual route to show the efficiency and fuel consumption of the fuel cell.

View full abstract

When a drone is used to suspend and transport a load, the residual vibration that occurs when the drone stops is a problem. In this report, residual vibration is suppressed by elimination of the natural frequency component from the apparent external force, which considers the effects of non-linearity and damping. Numerical simulations and experiments have verified the effectiveness of this method for a suspended load of a multi-rotor drone.

View full abstract

A comfortable gearshift operation feeling is created by designing suitable gearshift sound for feedback to gearshift operation by paddle shifting, etc. In addition, the acceleration feeling was improved by designing gearshift sound for each gearshift stage, taking into account changes in the sound environment during acceleration driving. In this study, we created a gearshift operation feeling by sensitive balance design of geashift sound, gearshift shock and gearshift operation force.

View full abstract

Based on the need to improve quietness in the automotive industry associated with the spread of xEVs, we are developing laminated sound-absorbing materials that combine non-woven fabrics and foam materials. In this study, we calculated the sound absorption characteristics of a large number of lamination patterns using a homogenization method, and attempted to show material design guidelines by combining them with machine learning.

View full abstract

In this paper, the application of participation factor focusing on response at seat point for vibration evaluation of motorcycle frame in the early stage of development is described. Using extended participation factor called partial participation factor, which is an index showing the importance of vibration of specific parts or specific positions, major vibration modes are efficiently screened, and design change proposals are evaluated. In addition, the effects of these design changes are validated.

View full abstract

First of all, the definition of “Unit Mode” is a smallest unit component mode which has only one loop and cannot be decomposed any further. The resonance frequency of an unit mode is almost the same as that of whole system. An unit mode is a disassembly of eigenmodes of each part, and eigenmodes are obtained by assembling the unit modes. The energy correlated with the response of interest on unit modal coordinates is called “Power Flow”. By visualizing this, it is possible to draw a mode path diagram that shows vibration transmissions from the excitation point to the evaluation point. The path can be changed by shifting the eigenvalues. This paper introduces the concept of power flow and its application to vibration transmitted through the vehicle body.

View full abstract

In this paper, we propose a system that immediately informs a driver that he/she is pressing the accelerator pedal by means of a tactile vibration of the pedal. An experiment has been conducted with a driving simulator to investigate whether the pedal vibration system is effective in preventing pedal misapplication accidents and whether it is accepted to drivers. We induced pedal misapplications on a driving simulator and observed reactions of experiment participants. As a result, it was found that the pedal vibration system allowed drivers to notice pedal misapplication more quickly and to correct their foot position more rapidly.

View full abstract

For accident prevention and damage mitigation, this study mainly aims to understand the kinematic behavior and injury risk of children in collisions involving bicycles equipped with child seats through experiments. Rear-end collision tests were conducted between a car and a bicycle equipped with a child seat, and the effects of the seat belt and the headrest of the child seat were investigated. As results of the tests, there were risks of head injury when the child's head impacted the vehicle, the adult occupant, or the road surface. It was clarified that the headrests have a large effect on head injuries.

View full abstract

Impact SEA(ISEA) is a method to evaluate impact performance based on energy propagation. ISEA is used for the design of impact structures considering dynamic progressive buckling. In this paper, from the load-stroke diagrams used in vehicle body design, crash members are designed via the ISEA model and verified by FEM analysis that presents the development and validation of a design procedure using ISEA.

View full abstract

To enhance a driving experience, we focused on the driver's perceived sense of vehicular control, such as a sense of unity in driving. This study explores the relationship between the perception and vehicle characteristics or driver's attributes. In the experiment, participants were divided into three groups by their frequency of driving and performed some tasks on a driving simulator under conditions of delayed visual response, followed by a sensory evaluation questionnaire. The results show that visual response delay and individual driving frequency affect the formation of a sense of unity in driving.

View full abstract

A purpose of this study is to model ocular movement for stabilization with eyes and is to enable an evaluation about visibility due to the instability of eyes based on the model. Relations of visual acuity with retinal slip have been formulated in this paper. The dynamic visual acuity and the ocular movement of a participant seated on a 6-axes motion base have been measured simultaneously under whole-body vibration. As the result, the relation of the dynamic acuity with the retinal slip under the random input ( acceleration of the 6-axes motion base floor : ±0.5 m/s2 , ±1.0 m/s2 ) is according to V = V0 + 0.35 * S / (14.6 + S), where V is the dynamic visual acuity, V0 is the static visual acuity and S is the ratio of the visual target image on the retina beyond the central fovea range.

View full abstract

The framing effect is a phenomenon in which a change in the way the conditions are presented can significantly change the decision-making process, even if the presented conditions are objectively equivalent. It has been reported that negative framing tends to lead to stronger motivation. The purpose of this study is to examine the alerts that promotes safe driving. We investigated the effects of pre-driving and in-driving alerts on driving behavior. As a result, it was confirmed that alerts using negative frames, rather than positive frames, before driving and in the pause-situation could promote more cautious driving behavior.

View full abstract

Our previous research showed that it is important to project a visual target on fovea centralis stably for good dynamic visual acuity while riding in a car. In this research, to enable a numerical evaluation about dynamic visual acuity in a car, we propose a new single eye movement model based on physiological knowledge about eye movement. Our eye movement model has three characteristics as follows: (i) to stabilize eye motion by feedforward controllers using head motions, (ii) to decrease a retinal slip by feedback controller using the difference between a future position of visual target and current eye position, (iii) to estimate model parameters by regression equation of variance of head rotational angular velocity. And we show effectiveness of our proposed model by validation using the comparison between numerical result and experimental result of eye movement.

View full abstract

This research aimed to mitigate injuries caused by pedal misapplication accidents involving elderly drivers through the traffic accident analysis. The results revealed that, compared to younger drivers, elderly drivers had a higher proportion of pedal misapplication accidents in the following situations: in parking lots, with two or more collision marks on their own vehicles, during reversals, and with impacts to pedestrians or objects. Additionally, the impacted pedestrians tended to result in more serious injuries. Therefore, the development and widespread adoption of the acceleration control system for pedal error, which can effectively operate in such situations and mitigate injuries, are necessary.

View full abstract

Automobile exhaust emission has a serious impact on the environment. Driver models are currently used to predict the amount of CO2 emitted from vehicles, but the current models are inadequate as they do not capture the real road traffic environment. So, it is necessary to clarify the effects of traffic environments on driving operation under real-world driving conditions. In this study, how the driver’s accelerator operation, which has a large impact on CO2 emission behavior, was analyzed. Actual driving experiments using hybrid vehicles on multiple routes on public roads and highways were conducted in the Tokyo metropolitan area, and various data on traffic environments were collected and analyzed. To conduct the analysis, a driving model of a hybrid vehicle were created and which information about the traffic environment influences the driver’s driving operation were examined. Then, a multiple regression model to predict the amount of accelerator operation by the driver was created based on this information. In order to improve the accuracy of the model, a modified version of the model that takes into account the driver’s reaction time and the decision to accelerate off was developed. Finally, the created models were fitted to test data on general roads and highways to verify their accuracy and to analyze the magnitude of the influence of each traffic environment.

View full abstract

In diesel engines, soot deposition on lubricating oil causes a decline in lubrication performance, but the mechanism is not well understood. The purpose of this study is to construct an equation for predicting the amount of soot contamination in lubricating oil. The effects of the temperature difference between the spray flame and the wall surface on the amount of soot deposition were investigated for measuring the soot concentration by the LII method, the flame temperature by the two-color method, and the amount of soot adhesion by the transmitted light attenuation method.

View full abstract

The purpose of this study is improving fuel economy at part load operation for light commercial vehicles. 0-D engine simulation results show that increasing displacement and removing turbocharger solve multiple trade-offs between component technologies that are barriers to improving thermal efficiency. A prototype engine was produced to verify the effectiveness of this right-sizing concept. As a result of prototype demonstration, the concept engine can meet Japan’s post-Post New Long-Term emissions regulations while improving thermal efficiency.

View full abstract

In the first report, right-sizing concept was proposed. Essence of the proposed concept is a naturally aspirated engine eliminating EGR. Also, this concept obtains low density in-cylinder condition and low Pmax operation. In this report, the combustion parameters were optimized under low density condition in-cylinder using 3-D simulation. Furthermore, friction reduction technologies were developed based on advantage of low Pmax operation. As the results, the prototype engine with combining all technologies demonstrated that the maximum thermal efficiency can be obtained as intended at partial load.

View full abstract

Rh performance may not correlate with Rh particle size in automotive catalysts. Therefore, we studied the relation between Rh surface state and catalyst light-off performance by XPS analysis to find the cause. As a result, we found that there were metal Rh state, oxidized Rh state and highly oxidized Rh state in Rh surface state. In addition, we found that more metal Rh particles were produced by reduction treatment when the catalysts had both more oxidized Rh particles and less highly oxidized Rh particles. Furthermore, we found that Rh performance correlated with proportion of metal Rh particles.

View full abstract

Regarding thermal efficiency and emissions, potential of a dual-fuel engine of ammonia ignited using diesel pilot was explored by 3D-CFD. The SAGE with detailed reaction models of ammonia and diesel (n-heptane) was used in the CONVERGE software. For zero-level emissions, stoichiometric gas condition controlled by EGR ratio is applied to purify emitted NOx and unburned species by a 3way-catalyst. The CFD results have shown that ignition timing of ammonia can be well controlled by optimizing injection timing and quantity of diesel pilot, and mild burn can be obtained by the adequate amount of EGR gas. In spite of the stoichiometric condition, indicated thermal efficiency of the dual-fuel combustion is the almost same level of that of a conventional diesel engine using 100% diesel oil, mainly thanks to lower heat-loss and more efficient combustion with higher degree of constant volume (ηglt).

View full abstract

Various approaches have been employed to model rider behaviors, but replicating complex movements or diverse behaviors requires further consideration. In this study, the aim is to develop a model that can more in detail replicate the behaviors of real riders by leveraging the rapidly evolving technology of deep reinforcement learning. In this initial report, examples are demonstrated of constructing a rider model. Additionally, by utilizing technologies to evaluate riding skills, this study assesses the proficiency of the learning model and attempts to establish a connection between the learning model and reality.

View full abstract

Brain strain indices calculated using human finite element models have recently been used to evaluate the brain injury risk in automobile accidents. However, the calculation using a finite element model of a human body is computationally expensive and time-consuming, making it impossible to evaluate brain strain indices immediately after a crash test. Therefore, this paper develops a deep learning model to predict the brain strain waveform from the angular velocity waveform of the head that can be measured by a crash dummy. The results of a comparison between the brain strain waveform obtained by finite element analysis and the waveform predicted by the developed deep learning model showed that the CORA evaluation exceeded 0.9, indicating that the model predicted the waveform with a high degree of accuracy.

View full abstract

Automated vehicles are expected to be as a mobility service for securing transportation in rural area of Japan as the communities age. As a first step, this study analyzed the unsafe overtaking cases between a low-speed automated service car and its following on the obtained results and investigated the following driver’s judgement on his/her overtaking and understanding toward the intention when using an external HMI and lateral driving position of an automated service car near curve sections.

View full abstract

The European Union (EU) has mandated the installation of the driver drowsiness and attention warning systems (DDAW) in motor vehicles. DDAW must be approved based on the Karolinska Sleepiness Scale (KSS). To validate the Japanese version of the KSS conforming to EU law, participants rated the KSS while performing the psychomotor vigilance test (PVT) and a simulated driving task, during measurement of sleepiness measures. All of the correlation coefficients between the KSS and the sleepiness measures were significant in both the PVT and simulated driving task. These results support the construct validity of the Japanese version of the KSS as a sleepiness measure.

View full abstract

Autonomous highway merging is one of the challenges for achieving “High Driving Automation”. This study aims to investigate human driver behavior when manually merging into a highway with different traffic speeds and densities for the development of safe autonomous merging. Using a driving simulator, we designed various merging situations based on real-world highway settings in Japan. Considering the UN Regulation (No. 157) for Automated Lane Keeping Systems (ALKS), we examined the feasibility of autonomous merging when applying the current vehicle lane change requirements and further examined human driving characteristics for autonomous highway merging.

View full abstract

A three-dimensional unsteady flow around the Ahmed body at a low Reynolds number ( Re) was numerically investigated for the cases of slant angles 𝜃 = 29° and 31°. The critical Reynolds numbers are determined as Re𝑐 = 1049 for 𝜃 = 29° and Re𝑐 = 1090 for 𝜃 = 31°. The oscillatory flows appearing near the critical Reynolds number have a single frequency of Strouhal number St ≈ 0.4, and St increases up to St ≈ 0.6 with the increase for Re. At certain large Re, the oscillation of the flow changes to the multi-frequency behavior. The time-averaged flows are mainly composed of a pair of longitudinal vortices shed from the sides of the slant surface and another pair of longitudinal vortices shed from the spanwise center of the slant surface. For the relatively small Re cases, the flows are separated from the slant surface, whereas for the relatively large Re, the separated flow re-attaches downstream and center of the slant surface. The value of Re where the re-attachment takes place is smaller for the case of 𝜃 = 29°. From the energy analysis based on the Reynolds-Orr equation, the dominant terms and their spatial distribution are investigated. The regions where the energy is strongly supplied from the time-averaged field to the deviated oscillation field appear along the two pairs of longitudinal vortices of the time-averaged field. It was found that the energy supply along vortices at spanwise centers becomes relatively large with the increase of Re.

View full abstract

A vehicle-mounted measurement system (SEMS) which included sensors such as NOx sensor and NH3 sensor directly inserted into the exhaust pipe was installed on a heavy-duty diesel vehicle that complies with the 2016 exhaust gas regulations. Then, we acquired the data in the transportation business driving. Based on these data, the NOx purification rate, space velocity, etc. were analyzed as the parameters to investigate the NOx purification performance of the Cu-based zeolite SCR catalyst. As an example, during DPF regeneration, the purification rate gradually decreased to about 40% due to the oxidation of NH3 in the process of temperature rise.

View full abstract

To reduce the driver’s workload on parking, Parking Assistance System is required to be easy to use in the usual parking slots. This paper shows the proposed system can detect target parking positions without complicated user operation by using road surface patterns around the parking spot memorized after the driver parks manually.

View full abstract

In a software-defined vehicle (SDV), where software controls all of the vehicle behaviors, heterogeneous communication flows with different characteristics share a single in-vehicle network. While Time-Sensitive Networking (TSN) is promising to prevent interferences among the flows, conventional network design methods for TSN require huge computational effort. To reduce computational complexity, we propose “time and bandwidth division flow assignment method” where the communication resource is partitioned based on the characteristics of the communication flows. By the evaluation, we confirmed our method could generate network design within one second satisfying all of the communication requirements supposing autonomous driving level 2.

View full abstract

Recently, as the upcoming limitation of operational speed enhancement in conventional silicon device technologies, quantum computing methodologies are realized one of the promising candidates that could potentially instead conventional multi-core supercomputer platforms with GPUs. The functional material optimization problem, that idealize high performance electric vehicle powertrain systems, would be a suitable target for benchmarking the efficiency of this novel high performance computing technologies. Regarding this direction, for the purpose of elucidating latent powerfulness of quantum annealing algorithms in functional material design, we performed quantum inspired study for the optimization of compound compositional design with substituting heterogeneous elements. (Quantum inspired computing is as the simulation technology for solving specific Ising Hamiltonian that reproduce energetic behavior of systems under the latest GPU platforms.) Carbon material design, in relation to fuel cell catalyst or electrode material for Li ion battery in automotive application, was studied as the representative case in functional material investigation. Notable optimization performance improvement was observed.

View full abstract