This paper proposes a design method to suppress weave and wobble which are self-excited vibrations of a motorcycle. This method is to illustrate the stability region on the plane of design variables. As this stability determination, Hurwitz's stability determination is partially used. Partial use means that capsize instability is not targeted for this stability determination. As an example of this proposed method, this paper shows how to set the damping coefficient of the steering damper to prevent weave and wobble.

In this paper, control inputs and vehicle parameters are optimized simultaneously for the purpose of reducing power consumption of electric vehicles. The electric power consumed by the vehicle running resistance and motor copper loss is defined as an evaluation function, the optimal control problem is formulated using the front-wheel steering angle and front-wheel driving force as control inputs and the front and rear cornering stiffness as vehicle parameters. The optimal control problem is solved by a numerical optimization method called the direct method. From the results, the optimal control input and optimal front and rear cornering stiffness for minimizing power consumption in a predetermined course are discussed.

Single-vehicle accidents are known to occur because of many factors such as internal (e.g., operation mistakes, drowsy driving, and looking away) and external ones (e.g., traffic flow). Dangerous situations wherein an animal dashes out in front of a running car, sometimes occur and are often reported as a cause of single-vehicle accidents. This study refers to this situation as the “near-miss incident with dashing-out animals” (NIDA). While many studies have been conducted on near-miss incidents, only a few of them have examined NIDA. Since the construction of near-miss incident database in 2005, more than 160,000 near-miss incident data sets and approximately 700 accidents documented by drive recorders have been registered. This report statistically analyzes factors such as traffic situation, driver reaction, and others based on 358 NIDA data sets from this database.

In this study, we proposed a method of detecting pedestrians at blind spots from the vehicle side, such as intersections with poor visibility, by detecting space potential fluctuations around the human body caused by contact separation between the shoe sole and the road surface during walking. It was clarified that the walking motion of the subject in the blind spot area can be detected under the conditions determined by simulation and experiment. It was also shown that the sensor can be easily configured at low cost using FET. Furthermore, as a measure for improving the characteristics of walking detection, we proposed the application of assist electrodes and optimization of sensor circuit constants, and investigated their usefulness. As a result, the S/N ratio was improved by about 18 [dB] by applying the assist electrode at a distance of 5 [m] from the sensor to the subject. The optimization of the sensor circuit constants confirmed an improvement in the S/N ratio of 3.22 [dB] on average up to 5 [m].

According to the White Paper on the Aged Society of the Cabinet Office in 2017 year, the proportion of the population aged 65 years or older (the aging rate) in the total population of Japan continues to increase, and in 2060 it is predicted that the aging rate will reach 38.4%. So, an increase in traffic accidents involving the elderly is also expected. Therefore, urgent measures are required to prevent accidents caused by elderly drivers. In this regard, recently, accidents caused by the pedal operation error by the elderly drivers have become a social problem. In the previous report, the authors conducted an experiment in which the driver operated the pedal, in a posture to twist the upper body to the right and look backwards, assuming back-drive. As a result, it was clarified that the pedal layout is one of the factors affecting the pedal operation error. In this study, the authors focused on the vertical height of pedal step among the brake and accelerator pedal layout. And the authors experimentally analyzed the behavior of the toes in the pedal-switching operation from the accelerator to the brake by elderly drivers. It was clarified that not only the physical decline due to aging but also the height of pedal step is one of the factors that affect the pedal operation error.-

In this study, The authors analyzed the drive recorder data, focusing on the driver's response to the risk in an accident between four-wheeled vehicles. The authors analyzed the accident factors after classifying them according to the drivers’ position in the accidents, and investigated the driver's behavior tendency that led to the accident. As a result, the following tendency is found. In many of the interruption accidents, the driver could not react to the existence of other vehicles or the risk of the accident until the accident occurred. In particular, the lack of risk perception of interrupting car’s driver was a major factor leading to the accidents .In the case of leading to the accident, the interrupting car’s driver did not respond to the risk of the accident. On the other hand, in many rear-end collision accidents, the response to the risk of the accident tended to be slow, and the delay in the perception of the risk had a great influence on the avoidance of the accident. The following car was able to avoid the accident by stopping at the time when the risk of the accident became recognized. However, in most cases, it was difficult to avoid accidents due to delayed recognition.

The wheel flat is a phenomenon in which a part of the wheel tread is cut away to become a flat surface. It is known that wheel flats increase the noise during traveling and that the high impact acceleration generated by flats has a major impact on vehicle and track damage, deterioration, and life. Therefore, it is important to analyze the generated impact acceleration with high accuracy. In this study, a rolling contact analysis was performed using a bogie model with a flat, and the impact acceleration generated and the distribution of the normal contact force on the contact surface was evaluated. As a result, it was found that the impact acceleration and normal contact force due to the flat affect not only the wheel with the flat but also the opposite wheel through the axle. It was also found that the stress distribution on the contact surface was asymmetric between the wheel and the rail due to the rotation of the wheel and that the rail generated higher stress.

In this research, we investigate an adaptability of drivers to a personal mobility vehicle (PMV) using lean mechanism. In order to evaluate PMV with a lean mechanism considering driver's feeling, the purpose is to establish an evaluation method in human-car system. Experiments were conducted with multiple vehicle characteristics using a driving simulator. In this report, we analyzed for the steering angle focused on the input of the driver. First, it was confirmed that the steering angle really obtained with DS depends on the steering characteristics. Next, we confirmed that the tendency of the trajectory was different from the difference in equivalent preview time. Finally, the frequency characteristics of the adjustable steering were examined. As a result, there was a difference in reproducibility. And this difference may be a method of expressing driver characteristics.

In recent years, research and development of personal mobility, which is said to be a compact single-seat mobility support device as one of the next generation transportation, has been promoted by companies and research institutions. Many of the vehicles that have been developed have a distance between the driver's boarding position and the road surface, and there are riding methods such as standing and sitting. In this study, wheelchairs and self-propelled vehicles that the driver himself drives are defined as a part of personal mobility, and the vibration analysis that affects the driver rather than driving and the effect on ride comfort are examined. In addition, the natural frequency obtained from the impact test is compared with the theoretical value.

Rail of railway is one of the most important elements in constructing railway systems. Railheads are subjected severe contact with wheels during repeated passage of vehicles. As a result of severe contact with wheels, wear of rail or rail defect have been occurred on railheads. Rail profile will be changed due to wear development. The wear of rail, which does not occur in a short period of time, has a close relation to crack initiation. And there is a possibility that it influences the running stability or the curving performance of vehicle due to the change in the rail profile attributable to it. Therefore, it is very important to analyze the wear development of the rail. In the previous research, we conducted the wear experiment under the longitudinal or the lateral slip generated by use of full scaled test equipment. In this research, we conducted the wear experiment under the mixed slip conditions that the longitudinal slip and the lateral slip were generated at the same time in order to construct the wear prediction model under the mixed slip condition by use of multibody dynamics.

Wheel/rail wear not only decreases running safety and stability, but also causes vibration and noise. However, measures against the wear require a great deal of cost and labor. In this study, numerical procedure using Multi-body Dynamics Software is proposed to predict the wear profile of rail in the curved section. In the curved section, the wheel passes with an attack angle, causing longitudinal and lateral slips at the wheel/rail contact patch. Longitudinal and lateral slips affect the wear development. Therefore, in this study, in order to research the influence for wear development when longitudinal and lateral slips occur simultaneously, we constructed numerical analysis model for wear prediction, and compared between the numerical and the experimental results.

This paper presents a basic study for developing a new adhesion improvement method for railway vehicles, focusing on the tangential force characteristics between wheels and rails. In our past studies, the following phenomena have been confirmed. First, the tangential force of the wheel/rail is increased by repeated rolling and sliding frictional force, and the tangential force is stabilized by the oxide film covering the contact surface. In addition, the tangential force under the condition where the humidity around the contact surface is low tends to be larger than that under high humidity condition. And finally, these phenomena indicate that the friction coefficient of wheel/rail can be increased by instantaneously reducing the humidity around the contact surface. Therefore, in this paper, we propose a method for improving the frictional force that reflects the above-mentioned results in our pas studies. In the proposed method, assuming that it is applied to re-adhesion control of railway vehicles, the frictional force of the wheel/rail contact is temporarily reduced by lowering the humidity around the contact surface by injecting dry nitrogen gas. To verify the proposed method, using a pair of the small cylindrical specimen, tangential force was measured under the conditions with different humidity. In addition, in order to confirm the effect on the braking performance of actual railway vehicle, numerical analysis was carried out. As a result, the experimental result shows the frictional force increases up to approximately 40% under high humidity condition, and was confirmed by the numerical analysis that the proposed method has an improvement effect on the reduction of braking distance during dry nitrogen gas injection even if the humidity condition around the contact surface is high.

The vehicle behavior under force control is stable near the straight line, but may become unstable near limit cornering. This phenomenon was explained by adding the nonlinearity of the tire characteristics and the nonlinearity of the power steering system to the conventional three-degree-of-freedom model.

In the design of modern railway vehicles, it is general way to predict and evaluate the vehicle vibration by vehicle dynamics simulation under the vehicle running on the assumed traveling section. To realize the desired vehicle vibration characteristics, it is important to clarify the mechanism of vibration transmissibility from track irregularity to vehicle. In many previous studies, the vehicle lateral vibration on curve section was evaluated by vehicle dynamics simulation. However, the vibration transmissibility on curve section from lateral track irregularity to vehicle was not sufficiently investigated. In this study, the vibration transmissibility characteristics on both curve section and straight section is compared by vehicle dynamics simulation with sine-wave lateral track irregularity excitation. As a result, the mechanism of vibration transmissibility of vehicle lateral vibration for curve section is clarified.

To study the reduction of damage of the railway vehicle caused by natural disasters, a 1 to 10 scale model vehicle is made for various tests such as an overturning test, which is difficult to be performed with a full-scale model. Shaking tests are carried out in lateral and vertical direction to find out the amplitude-frequency response of this model vehicle. It is found that the natural frequencies exist around 4.5Hz and 40Hz in the vertical direction, 2Hz and 10Hz in the lateral direction. Comparing the results with the calculation results of a full-car vibration model, we tune the parameters in the model to explain the results well.

In recent years, railway vehicles have been more lightweight because of improving the running speed. On the other hand, the vertical vibration of car body sensed uncomfortably by passengers has become a problem because of weight reduction. Therefore, there is a lot of research to reduce those vibrations. In previous study, the effectiveness of the vertical vibration reduction was examined by controlling internal pressure of air spring using H_∞ control theory. As a result of this examine, the controller by H_∞ control theory was effective for vertical vibration reduction. However, several problems have been identified with the controller by H_∞ control theory. In this study, the author aimed to reduce vertical vibration by controlling internal pressure of air spring using μ - synthesis. The merit of μ - synthesis is that this control theory can make higher potential controller than controller by H_∞ control theory and guarantee robust performance. At the beginning, the author made the controller by μ - synthesis for 1/6 scale railway model. Moreover, the author experimented with 1/6 scale railway model to verified nominal performance and robust performance. In conclusion, the controller by μ - synthesis is effective for vertical vibration reduction of 1/6 scale railway model.

It is necessary for pantographs of Japanese high-speed trains to satisfy both low aero acoustic noise characteristics and insensitive lift force characteristics with respect to the variation in wind direction and in the shape of pantograph head due to wear. Taking particular note of the profile of the pantograph head, streamline profile generally leads to low noise but sensitive lift force characteristics, whereas rectangular profile leads to less sensitive lift force characteristics but generates larger noise. It is therefore difficult to achieve the two characteristics at the same time with a passive way. The authors propose a solution for this trade-off by applying the pantograph equipped with smoothed profile pantograph head, which gives better performance in terms of aerodynamic noise, and actuator to actively control the lift force. This paper describes the new lift force control method with the sparse modelling technique for the lift force estimation by the pantograph working height and the static holes which are equipped on smoothed profile multi-segment pantograph head and these wind tunnel test results. From test results, we confirm that it is possible to reduce the lift force variations by using proposed technique.

When developing a control law for a vehicle as a plant it has become possible to design while paying attention to the state variables in the plant with the development of modern control theory. In this context, in order to analyze the static stability and dynamic stability of the plant at a higher level, the definition of static stability and dynamic stability, and the static stability and dynamic stability. The motivation for this research is to clarify the relationship between stability and state variables. In general, static characteristics in automotive engineering often refer to US and OS characteristics that are determined by the positive and negative recovery moments. On the other hand, static stability in control engineering is determined by the constant term of the characteristic equation. Comparing the two judgment methods, especially assuming an OS below the critical velocity, the former recovery moment judgment is negative to be static instability. But the latter characteristic equation is positive in the constant term judgment to be statically stable. This is the contradictions. In this Research, after clarifying this contradiction, the relationship between static stability and dynamic stability is examined again, and the results of investigation and arrangement of stability from both the automotive engineering and control engineering reports are reported. Furthermore, regarding the stability of the Plant (vehicle) required in the design of steering (steer-by-wire, SBW) control laws to enhance or compensate for vehicle motion, new definitions of static stability and dynamic stability.

This paper proposes a control target generation algorithm for autonomous vehicles. The algorithm is applied to the risk potential method, in which some risks that human drivers are influenced by the surrounding environment information of the vehicle are defined by exponential functions. In addition, we proposed a method for estimating the parameters constituting the risk potential function in the algorithm. The proposed method uses driver's driving behavior data, vehicle status, and environmental information around the vehicle. Using this method, the possibility of traveling in urban areas where oncoming traffic, obstacles, and pedestrians exist on road traffic is also explained.

In recent years, the movement for the development and commercialization of autonomous driving vehicle is rapidly getting higher. In order to spread autonomous driving vehicles, it is necessary to evaluate safety, and we must create a scenario for that. Therefore, we aim to extract parameters for scenario creation from near-miss incident data. In this research, we propose a method of pitch angle estimation of monocular camera using drive recorder video for the measurement relative distance. First, estimate initial pitch angle estimation from lane parallelism. Next, in a scene where the pitch angle changes due to sudden braking, the estimation method is switched. Here, the pitch angle is estimated from feature point matching.