Structure-durability mechanism of Three-Way-Catalyst (TWC) particles membrane filter was investigated through TEM images of contact areas between two agglomerated TWC particles. After those agglomerated TWC particles were sintered at 700, 800, 900 and 1100℃, a thin sample around the contact area was cut off out using an FIB-SEM system. From the TEM image analysis, the contact area becomes larger by thermal expansion of both TWC particles with temperature. Moreover, nanometer-sized primary particulates such as alumina and ceria-zirconia which are elements of the single TWC particle cross over the apparent contact line between two agglomerated TWC particles. As a result, the percolation structure becomes stronger as the sintering temperature increases because the crossover rate increases.

View full abstract

Emission regulations for vehicles are continuously being strengthened in various countries. Additionally, the electrification of vehicles leads to lower exhaust gas temperature and limits on board space. To overcome those matters, we have developed a compact and low temperature dependence after treatment system using plasma and HC adsorption. We investigated PN and HC removal performance under cold start condition. As a result, PN removal rate and pressure drop of the plasma reactor were maintained during continuous driving. Ag ion-exchanged BEA indicated high HC adsorption rate of 80%. HC emissions were reduced by 50% from exhaust gas after the three-way catalyst.

View full abstract

We are developing a “compact water-cooled fuel cell system” that can be used to refuel all types of mobility. The FC stack adopts compact FC separators manufactured using our core precision press technology. The cooling device adopts a “thermal management system” that utilizes the thermal characteristics of the hydrogen tank and the water-cooled FC stack. These technologies have enabled the miniaturization of the entire system.

View full abstract

Thermal management of the lithium-ion battery in the hybrid and battery electric vehicle, which controls the battery temperature, is important to counteract battery degradation and overheating. In the previous research, we proposed the thermal management system for the battery electric vehicle, where the battery refrigerant and chiller are connected to the long life coolant circuit based on the vehicle system 1D simulation model. In this report, we improved the thermal control performance of this system by modifying the system structure from the battery electric system to the series hybrid electric system. Then, we optimized the thermal and power management control using the optimization method.

View full abstract

Expanding the commercial use of electric vehicles is crucial for achieving carbon neutrality in the automobile transportation sector. This study conducted questionnaire surveys targeting two groups: general freight transport companies and those already using electric vehicles, to compare their perceptions of electric vehicles and usage practices. The findings identified external factors that could be challenges to the broader adoption of commercial electric vehicles, such as performance of existing electric vehicles, availability of chargers, and concerns by transport companies.

View full abstract

Electric wheelchairs, unlike cars, do not have dedicated lanes and operate in shared spaces with pedestrians and bicycles, leading to collision accidents. In this study, we propose a method to estimate collision risk by considering the positional relationship between the wheelchair and pedestrians, as well as the state of the pedestrians. The proposed method involves detecting pedestrians using a stereo camera, estimating their distance and speed, and then estimating the collision risk based on the pedestrians' direction and speed of movement.

View full abstract

For safe autonomous driving, localization is essential. Conventional point cloud matching methods using convergence calculations have the problem of initial position dependency. While methods that predict the vehicle's position as a likelihood distribution reduce initial position dependency, but they have difficulty handling the 3D information of the point cloud. This study proposes a matching method that retains the features of LiDAR point clouds in a pillar structure and converts them into pseudo-images, using deep learning to estimate the vehicle's position as a likelihood distribution. The proposed method aims to achieve more robust localization compared to conventional point cloud matching methods.

View full abstract

This research examined the differences in pedal-misapplication accidents between passenger and light passenger vehicles. The number of accidents caused by pedal misapplication in passenger vehicles was higher than that in light passenger vehicles. The pedal-misapplication accident ratio, considering traffic exposure effects, was similar between the two vehicle types. Drivers aged 75 and older were more often involved in these accidents across both vehicle types. In single-vehicle accidents caused by pedal misapplication, the proportion of fatal injuries to drivers was similar for the two vehicle types. However, serious and slight injuries were more frequent in light passenger vehicles, indicating a higher risk of injury.

View full abstract

Automobile electrification requires miniaturization and power saving of cabin air conditioners and battery cooling systems. For further performance improvement, it is important to improve the performance of heat exchanger by effectively using turbulence produced by blower. However, the influence of fan turbulence on fine fins of heat exchanger is not clear. In this study, the effect of flow turbulence on heat transfer of the louver fin was investigated using large-scale LES. As a result, it was found that the separation between louvers was suppressed by the effect of fan turbulence, and the amount of heat radiation increased.

View full abstract

When wipers are operated while driving, washer fluid flows over side windows, affecting the driver's visibility. To predict this phenomenon, this study introduces a CFD method for analyzing airflow and washer fluid behavior using the Lattice Boltzmann method and the Lagrangian particle tracking. The wind tunnel test validated this method, showing consistent results with the experiment, including the washer fluid spread from the upper to the center of the A-pillar and the position of the washer fluid reaching the side window. This method allows for early-stage design improvements in vehicle performance related to washer fluid, even without an actual vehicle.

View full abstract

While tailpipe emissions have been significantly reduced by stricter pollutant limits, the proportion of non-exhaust particle emissions such as brake wear has become noticeable. Therefore, introduction of regulations on brake particle emissions in the next European regulation (Euro 7) is currently under discussion. In this study, the measurement system for brake particle emissions has been developed. Validity of the measurement systems and emissions behavior are discussed based on measurement results of brake particle emissions during WLTP-Brake cycles.

View full abstract

In this study, we focus on improving the performance of raindrop removal. Conventional methods for still images and videos have limited available information, leading to performance constraints. Therefore, we propose a novel approach focusing on camera placement in autonomous driving, utilizing images from other cameras captured at the same time. In the proposed method, the camera images without raindrops are used for raindrop removal by transforming the viewpoint. The proposed method significantly improves the baseline method in terms of subjective evaluation and object detection, and the effectiveness of the proposed method is demonstrated.

View full abstract

The objective of this study is to improve the decision performance for lane change feasibility by planning trajectories that consider future variations. We propose a method that imposes constraints on the probability of maintaining safe distances from other vehicles in trajectory generation using chance- constrained model predictive control. In merging scenarios where the behavior of other vehicles is highly variable, the success ratio of merging improved by 25.9% and false positive ratio decreased by 30.3%. Additionally, the performance of the proposed method was evaluated by applying it to a real vehicle.

View full abstract

Based on the system simulations, we showed the effect of the aluminum electrolytic capacitors' low-temperature behavior on the actuator's control in the previous works. We assembled an actuator system that followed the previous system simulation to confirm whether the same phenomena appear in a low-temperature condition. We will show the details of these results.

View full abstract

This study proposes a novel video analysis technique utilizing multimodal large language models (MLLM) to aid AD/ADAS logic development. By combining rule-based algorithms and MLLM corrections, the precision of lane change detection improved from 0.74 to 0.90. Evaluations using monocular camera footage from Japanese roads revealed that MLLM effectively identified error-prone scenes, aiding logic improvement. However, challenges such as spatial misinterpretations suggest limitations in MLLM's temporal and spatial reasoning. Future work aims to address these issues and expand applications to diverse driving events.

View full abstract