Most riding simulators adopt the motorcycle model that uses steering torque as an input. In this paper, we propose a motorcycle model that uses steering angle as an input. This is because normally, examinees require a long time to get used to the common steering torque type riding simulator. We checked the validity of our steering angle model by analyzing the frequency response of the model and comparing it to the linearized BikeSim model. The results show that the value of yaw rate and the roll angle coincides with the linearized BikeSim model results of a motorcycle constantly circling, when the steering torque input period is under 0.1 [s^-1]. We also measured the steering angle, the speed, the roll angle and the yaw rate of the motorcycle in the practical riding test. These parameters lead us to our algorithm calculating the roll angle and the yaw rate from steering angle and speed. We applied the algorithm to the riding pattern data that simulates the situation of turning in an intersection. The results showed that the calculated data of roll angle and yaw rate are almost the same as the measured data. From these results, this proposed model can be said that it is appropriate to be used in a riding simulator for analyzing rider behavior.

This paper presents improvement of a rotational surface-contact tire friction model, which has been proposed by Kikuuwe et al. (2006), for driving simulators. This paper extends the previous model by including the anisotropy of viscoelasticity and the dependency on the normal force. The parallel viscoelastic element is proposed to reproduce smooth transitions from the static friction to kinetic friction, which have been empirically observed and have been reproduced by Pacejka’s Magic Formula. This proposed model leads to better results for both longitudinal friction force and lateral friction force as compared to the previous model.

In recent years, the traffic accident shows gradual decreasing in the number of fatalities, whereas slowly decreasing the total number of accidents and injuries. In order to decrease the total number of car-pedestrian accident, it is necessary to investigate what kind of accident has happened and evaluate driver’s behavior. So the feature of accident and near-crash incident involving pedestrian was analyzed from near-miss incident database. And to estimate driver’s behavior on driving simulators, near-miss incident database data was fitting on the scnarios of driving simulator.

We aim to construct a driver assistance system that is able to detect driver’s deviations. The system uses time-series head motion information during safety confirmation behaviors at non-regulated intersections. Therefore the system needs quantification of head postures. In this paper, we evaluate a head posture quantification method using one of the time-series data analysis method: Recurrent Self-Organizing Maps (RSOMs). RSOMs are added an ability of categorizing according to time changes to SOMs. Specifically, we discuss categorized results of safety confirmation behaviors at two intersections. A driving simulator is used for this experiment and we obtain driver’s upper body images. Input data is calculated with edge extraction and coarse graining using driver’s images. Number of output layer of RSOMs is 10. We compare graphs of categorized results of two intersections. From the results of this experiment, it is showed that same intersection graphs have similar trend, and this result means that there is same patterns in same intersection’s safety confirmation behaviors.

In recent years， concern with autonomous driving technology is heightened around the world. To realize the technology， it is necessary to consider about the ecosystem. In this study， we focused on the truck platoon system as the most important part among autonomous driving， evaluate the acceptability to the spec of it by using driving simulator. As a truck forms a platoon with short gap in this system， the truck platoon may adversely affect a general traffic. At first， we show results of experiment about general driver's acceptability， moreover， discuss that about platooning truck driver's acceptability and future policy.

As an effort to support the earthquake disaster reconstruction of the northeast (Tohoku) region, the project has collaborated to promote the research and development of a mobility integrated energy management system (EMS) that supports the sustainable development of a disaster resilience region. This study analyzes the charging behavior of the driver using driving simulator. As the results of experiments, we describe the effectiveness of navigation system for EV..

Most of two-wheeled vehicle accidents occur in passing and the crossing including the crossing right turn at intersections. In this study, we proposed a system which presents information regarding approaching vehicle to a motorcycle rider at intersections and analyzed the driving behavior of the rider using a riding simulator that we constructed. When a rider used this information presenting system, we observed that the braking reaction time was shortened and the deceleration behavior became safer. In addition, we constructed a rider model based on experimental data and we simulated the collision avoidance behavior of riders at intersections. As a result, we confirmed that the collision probability was decreased and the system is effective in accident avoidance support.

According to a survey in Japan, the number of all type of traffic accidents during traffic congestion is approximately 8 times greater than that of during usual traffic conditions. Particularly as regards the number of rear end collisions, it increases by a factor of 16. The purpose of this study was to improve drivers’ condition during traffic congestion by supplying aroma compounds to reduce traffic accidents. In this research, (R)-(-)-linalool which has a sedative effect was chosen as an aroma compound and the effect of it on the mental stress was analyzed. Electrocardiogram was employed and R-R Interval (RRI) was used as a stress index. Although there was individual difference, the sedative effect of (R)-(-)-linalool was shown.

Recently, as a result of the weight saving and the simplification of the structure of railway carbody due to satisfy the demand of the speed up and energy saving, ride quality tends to get worse from the effect of elastic vibration of the carbody. In order to reduce the vibration of carbody, new structure that the inner shell is hanged from the wall of the outer shell is proposed. In this paper, new structure is modeled as a double shell structure and vibration analysis is carried out using transfer matrix method and building block approach. In order to verify the result of the simulation, experiment is also carried out. Finally, vibration characteristics of the structure are evaluated from the results of the simulation and the experiment.

In the analytical model of the VDS (Vehicle Dynamics Simulator) developed by RTRI, the rails just under the wheels are laterally and vertically supported by springs and dampers connected to a road bed so that they have two degrees of freedom independently of each other. While the lateral supporting stiffness is mainly corresponding to rail tilting resistance, the angle of the rail tilting has been ignored and the rails have been considered to be able to move horizontally and vertically independently. In addition, the roll vibration of track is described by vertically moved left and right rails. In this study, therefore, new wheel / rail contact model, considering rail tilting angle due to lateral force and roll vibration of track, is adopted to the VDS. The effects of the rail tilting angle on the results of running safety during an earthquake have been investigated. As a result, it has been found out that by taking the rail tilting angle into consideration, running safety limits are changed within several percents. The proposed model can be used for calculation requiring high accuracy such as derivation study on a cause of a real derailment accident.

Vibration suppressors are used to resolve the problem of collisions and entanglement to the hoistway equipment caused by the rope sway due to the resonance of the building and the elevator rope. To gain an understanding of rope sway, the free vibration analysis of the string with vibration suppressors has been studied. However, the vibration of the string equipped with vibration suppressors encounters geometric nonlinearity, and hence, its characteristics have been studied under a few conditions. Furthermore, in the case in which the vibration suppressors are located in the center part of the string, no exact solution has yet been obtained for the free vibration of the string. In this paper, an exact solution to the free vibration of the rope with vibration suppressors located except for both ends part is presented in the case where rope length is divided equally into N (=p+q+p), when p is an even number and center position of the rope is pulled. The rope is modeled with string. Finite difference analysis of the rope vibration with vibration suppressors is also performed. The calculated results of the finite difference analyses are in good agreement with those of the exact solution.

Elevator ropes in a high-rise building are excited by deflection of building caused by earthquake and wind. Especially, the long-period ground motion causes the resonance between elevator ropes and building. As it causes severe damage for elevator, reducing the displacement of elevator rope is demanded. When the rope is long, it is difficult to ignore that the interaction between horizontal and vertical vibrations of rope. Currently, there are some methods to analyze the rope sway. But, the existence of universal method that can analyze the interaction between horizontal and vertical vibrations, and coincide with the experiments using actual elevator rope is very few. Therefore, in this study the new rope model is constructed as a multi rigid body pendulum, and I set the equation of motion by following the multi-body dynamics. And, the coincidence of the proposed model and the result of experiment are shown by numerical simulation. Furthermore, vertical or horizontal direction damping is added at the lower end of rope, and its effectiveness is confirmed.

We are developing new devices available in the speed range of high-speed bullet train operated through Shinkansen lines and regular train lines to improve the curving performance and running stability. The numerical analyses using a linear vehicle dynamics model showed that the axlebox support damping in the lateral direction is effective to increase the critical speed of hunting. In addition, we made some trial liquid-seal bushes and performed characteristic tests to investigate the values of stiffness and damping coefficients which is possible to put into practice. As a result, we got the prospect of providing a new bogie with liquid-seal bushes, which is superior in curve negotiation and stability.

To reduce the flexural vibration of railway vehicle carbody, a high-damping elastic support for underfloor equipment is being developed. This method is a kind of dynamic vibration absorber (DVA) utilizing underfloor equipment as mass element. The primary focus in this report is the influence of the difference of elastically supported mass on the vibration reduction effect. A series of excitation tests was conducted at the rolling stock testing plant using a Shinkansen type test vehicle by applying the proposed method. As a result, two dominant peaks in the acceleration power spectrum densities of floor corresponding to different flexural vibration modes were reduced simultaneously when the proposed method was applied and the reduction effect became large according to the increase of the mass of the elastic support. Compared with the case of the rigid support, reduction of the ride quality level was about 3 dB in case where the elastically supported mass was 1040kg.

To prevent bogie hunting motion and reduce carbody vibration in railway vehicles, a new mounting structure for yaw dampers has been proposed in this report. The new mounting structure enable to reduce the excitation force transmitted from a bogie frame to a carbody through yaw dampers. The prototype device was installed in a full-scale test vehicle and bogie hunting test was carried out at the Railway Technical Research Institute’s rolling stock testing plant to examine the running safety. Moreover, we performed vehicle dynamics simulation using versatile multibody dynamics analysis software SIMPACK. In this paper, the outline of the new mounting structure for yaw damper is presented and results of the bogie hunting test and the vehicle dynamics simulation are also described.

E6 series Shinkansen train has the convenient feature which can operate seamlessly through Shinkansen section and conventional-line section. In order to operate at higher speed, both the running stability at high-speed operation and the curving performance during the sharp curve passage are required. In terms of the environment noise reduction, it is also necessary to reduce the noise from bogies by mounting soundproof covers. The authors studied on proper arranging position of yaw dampers fit into bogie covers. As a result, the authors clarified that the running stability of the proposed serial mount method was equal to the current method (parallel mount) and the riding quality of the serial mount was slightly better than the current one.

Large overshoot sometimes occurs during use of air-spring car-body tilting by which a car-body is tilted using inflated or deflated air-springs. The large overshoot causes the contact loss between the pantograph and the contact wire. Therefore, we developed a controlling method of the overshoot by using the vertical motion damper equipped in parallel with the air-spring. The detail of the vertical motion damper, the controlling effect of the damper on the overshoot, and the overview of the running test is described in this paper.

Although the number of patent applications filed by Japanese companies is the largest in the world, it must be assumed that regarding patent applications, German companies are comparable to Japanese companies when taking German’s GDP etc. into consideration. In addition they are fiercely competing with each other in developing automobile preventive safety technologies. The competition is expected to become fiercer because IT companies will also be involved in developing such technologies. Japanese automakers must work out technology development strategies for their core technologies in which they have advantage over others, in order to be well prepared for the future.

Analysis using near-miss database, which is composed of drive recorder data, is expected to promote active safety researches to reduce traffic accidents. Automatic classification techniques for such big data have been eagerly studied. In the most of the preceding researches, they took notice of acceleration variation near the trigger time. For the better scenario understanding, it is important to observe the data of several seconds before the trigger. In this paper, two approaches are reported; firstly, smooth/rough road classification by power spectrum of vehicle’s vertical acceleration, and secondly, vehicle behavior pattern classification by vehicle’s speed profiles.

This paper discusses simulation for estimation of multiple effects to prevent traffic accident by diffusion of Advanced Driver Assistance System(ADAS) and Automated Vehicle(AV). In order to reproduce realistic traffic environments, it is effective to model the various drivers’ behavior and error based on agent based simulation. Driver model and ADAS model are implemented for evaluation in car-following situation. The provisional estimation of the effectiveness of collision warning system and automatic brake system is executed by utilizing drive video recorder data.

Recently, car manufacturers are developing automobile active safety systems to prevent traffic accident. Active safety systems include automated driving system, AEB (Automatic Electric Brakes), ESC (Electric Stability Control), LAKS (Lane Keeping Assist System), IPAS (Intelligent Parking Assist System). It is important for the system to make drivers understand the system easily and operate appropriately. Driver’s acceptability of the system will increase when the system functions in the same way as a driver. It is necessary to research driving behavior to emerge the systems that enable us to use comfortably and unmistakably. In this paper, we study behavior of a vehicle when the vehicle passes a pedestrian on the local street. We measured some quantities of vehicle behavior such as trajectory, acceleration, steering angle using a vehicle on the test course which simulates the local street. In the experiments, the subjects are chosen with various driving record and frequency of driving. As a result, it is revealed that characteristics of vehicle behavior depend on driver‘s skill and frequency.

This paper describes an investigation on the nonlinear filter that can realize its gain and phase independently. In the driver-vehicle closed loop system， it is reported that there is an optimal value in vehicle phase delay. Therefore excessively large or small delay deteriorates vehicle maneuverability. However， there is a strict relationship between the gain and phase delay in linear system. It is difficult to evaluate the gain and phase delay of vehicle response independently. In this investigation， we tried to design a non-linear filter that can set the relationships between the gain and phase delay arbitrarily. We investigate effect of the gain and phase delay corresponding to the steering angle input on the ease of driving by using driving simulator，and evaluate the control performance on the lane change task. We verified the ease of driving by reducing a phase delay and decreasing a high frequency gain.

One of the cause of traffic accidents or incidents is drivers' looking aside behavior. The authors analyzed the features of drivers' looking aside behavior by using the drive recorder data. We found that this behavior occurs when three conditions stated below are satisfied. The first one is that the time head way is long. The second is that the required deceleration for collision avoidance is supposed to be small. The third is that there is enough time for the drivers to push brake pedal and stop safely. Drivers’ looking aside behavior under dangerous situation doesn’t always lead to an accident, but when the looking aside period is too long, this behavior tends to cause an accident.

In recent years, increase of the accident an old driver causes is a problem. Therefore it's necessary to ascertain driving continuation aptitude of an elderly driver. It is reported that self-assessment of driving skill is important for elderly drivers. However, there are no indexes which estimate it objectively. In this report we made the index which aimed at motor coordination to express self-assessment of driving skill and inspected relationship of motor coordination, self-assessment of driving skill and driving behavior of elderly drivers. In this paper, we have developed systems to evaluate the motor coordination required to continue driving. We decided to deal with timing and a grading as the item to do a movement adjustment necessary to driving, and Repeat reaction task, Prediction of motion task, Chase task were made. Next we did the experiment for 40 people over the age of 65 for the purpose of the thing confirmed about independence of cognitive function tests and relation with uneven reaction. The results showed cognitive function and motor coordination are the function which became independent and uneven reaction may affect the grading exercise coordination.

Aiming at the realization of agile autonomous flight for the extension of the flying range of single-rotor helicopters in the limited flight time, the flight control system design and the control performance verification for the high-speed turn flight are conducted. The control system is composed of the main-rotor thrust and body torque centralized controller based on the MIMO (Multiple-Input, Multiple-Output) helicopter model including the blade motion dynamics, rotor speed proportional-integral controller, the attitude controller designed via the backstepping method based on the quaternion attitude model and the guidance controller based on translational motion model. Simulation results show that the hovering and 10m/s straight cruise flight followed by over 45-degree bank turn is feasible by using the proposed control system. In addition, flight tests have been conducted using the experimental small unmanned electric helicopter equipped with the flight control computer, and the high speed turn agile autonomous flight was successful in the real environment.

For commercial applications of small unmanned aerial vehicles such as infrastructure monitoring, agricultural missions, and small-package deliveries, a tolerance for gusty wind conditions is one of the most important performance factors that need to be considered to make them practical to use. However, there is no standard for the performance evaluation of gust tolerance. In this study, free flight gust response testing method using a motion capture based precise flight control is proposed. A prototype of the tesing system using JAXA's low-speed wind tunnel (LWT2) was constructed, and gust response free flight demonstration was carried out successfully.

In order to realize automatic investigation of MAVs in indoor environment, an independent navigation method which does not use exterior system such as the satellite navigation system is required. Such a method is called “SLAM”, and the authors are now focusing on applying it to MAVs. Hector SLAM was applied as the basic SLAM algorithm and it was fused with IMU by EKF. The brief concept of SLAM algorithm and implementation of EKF are introduced. An MAV equipped with a LIDAR, an IMU and a PC was used for 2 flight experiments where first one was at non-ordered environment and second one was at well-ordered environment, which is more preferable for the SLAM algorithm. Motion capture system was used to evaluate the navigation performance. Both experiments succeeded to automatically flight control the MAV by the proposed navigation method in real-time. Second flight showed more stable response and accurate navigation compared with first flight. This implies that the environment configuration will affect the SLAM navigation accuracy.

The research trends in recent years, large and various amount of digital data set called big data are focused on. For the purpose of promoting research in air traffic management, flight track data of Air Traffic Control have been provided as CARATS Open Data by Ministry of Land, Infrastructure, Transport and Tourism. The CARATS Open Data which is newly opened to the public in August 2016 includes Haneda terminal radar control data. In order to increase the Haneda airport departure and arrival slot, such technology that increases the prediction accuracy of ground speed of Haneda airport approach trajectory considered to contribute. We can expect new findings by analyzing the trajectory data, though the deterioration of accuracy of ground speed is pointed out by calculating the speed from the difference of position data and time. In this paper, ground speed in Haneda airport approach trajectory is investigated using CARATS Open Data, and the variation of approach speed by aircraft type is discussed.

Collisions sometimes occur when two vehicle are on a sudden encounter at such an intersection. In this paper we deal with the case of the vehicle lieaving a garage to a public road. The robot car is used in the experiment. The laser range finder of the robot car detects the coming vehicle. The automatic lieving a garage program made by ous is verified. We have confirmed that there is the enaough headway distance between the own car and the other car by this algorithm. An active safety method in the case of two vehicle cncountering suddenly when passing across the road may be obtained.

If drivers take their hands off the steering wheel during automated driving, drivers may easily lose situation awareness. In order to establish really useful level 2 driving automation, the automated driving system should be designed to assure that drivers can maintain their awareness of the system’s operating condition. Thus, we propose a method to support drivers’ situation awareness during the automatic mode by providing the system’s intention and situation understanding through vehicle behavior. Drivers may easily recognize something is wrong when an expected vehicle behavior is not observed due to system’s functional limitation, and the drivers may be able to intervene into control smoothly even when their hands are off the steering wheel. An experiment using a driving simulator was conducted in order to evaluate the effectiveness of the proposed method. The results from 20 drivers showed that the proposed method was effective in supporting drivers who did not hold the steering wheel to understand the system’s operating condition just like the drivers who held the steering wheel.

Autonomous vehicles have been developed rapidly in recent years for the purpose of decreasing traffic accidents, reducing traffic jam and increasing human life quality. Environment perception is an important key issue in order to bring such vehicles into reality. Accordingly, a robust object classification is needed in terms of identifying obstacles and providing safe path planning. This paper describes an object classification method based on 3D point cloud data generated by LIDAR. Objects are classified into car, pedestrian, bicyclist and background using AdaBoost and Real AdaBoost classifiers. These classifiers have some advantages such as easy implementation, high performance and low processing time. Features such as the object’s size, shape, intensity and velocity calculated from 3 dimensional coordinates are used as inputs to the classifiers. The proposed strategy is to use decision tree as a weak classifier for AdaBoost and 2-dimension probabilistic distribution for Real AdaBoost. A hug database has been used for training, testing and evaluating the performance of the implemented classifier. The results have verified that the proposed method provides promising accuracy in the range of 85%~95% depending on some object conditions such as appearance, mass and number of available object surfaces.

This paper presents a method of accurately refining 3D laser scan data from in-vehicle multilayer laser scanner. The vehicle estimates its own 3D pose (position and attitude) based on Normal-Distributions-Transform (NDT) scan-matching method. Based on the pose information, the vehicle pose when the laser scan data is captured is extrapolated and interpolated under the assumption that the vehicle moves at an almost constant linear and turning velocities. The extrapolated and interpolated pose of the vehicle is applied to mapping laser scan data into the voxel map represented on the world coordinate frame. Subsequently, the vehicle again estimates its own 3D pose based on NDT scan matching. Such an iteration enable to compensate the distortion of laser scan date and to accurately map the data on the world coordinate frame. Experimental results show the performance of the proposed method.

Nowadays, for practical use of automated driving of a vehicle, the governments and manufacturers have developing actively. Most existing automated driving systems are need to use digital road map data that is included lane position information. So, digital road map needs to prepare at all roads. Currently, the lane position information data includes errors caused by development process of manually, and the errors affects vehicle stability and lane followability. Therefore, this study aims to the decreasing the effect that caused by the errors by using a smoothing method. However, the result of common smoothing methods like low-pass filter is not always appropriate for automated driving systems. Thus, this study focuses on an effect of smooth driving by human drivers, and proposed a smoothing method by driving simulation with driver model. The effectiveness of the proposed method is verified by using lane position information data.

It is important for railway vehicles to get compatibility between curving performance and running stability. Utilizing independently rotating wheels may be an effective solution. The EEF bogie proposed by Frederich in 1990th shows good curving performance making use of gravity restoring force generated by the tread gradient of independently rotating wheels. However, the bogie gives rise to a kind of hunting motion as the vehicle running speed increases. In this paper, an effective modification of the EEF bogie which solves the hunting motion is mentioned. The solution is to incline both wheel axles while adjusting tread gradient of each wheel. In this paper, the EEF bogie with inclined wheel-axles is firstly proposed and analytically evaluated by modeling and MBD simulation. From the result of linear stability analysis, proposed bogie can dramatically improve the high speed stability as compared to the conventional EEF bogie. In addition, the proposed bogie has excellent curving performance in tight curved section equivalent to the conventional EEF bogie. As those results, it is possible to achieve both high speed stability and high curving performance utilizing the proposed bogie unit in the vehicle.

The driving force control improves the off-road vehicle traveling performance and driving efficiency. However, the studies on the driving force control have been developed for the vehicle mainly on rigid roads. This paper describes a method of driving force control for the independent wheel drive vehicle on dry sand. The proposed method is based on the tire model on dry sand. In this method, tire sinkage is estimated by measuring slip ratio and vertical load. Using the estimated sinkage value, the slip ratio which generates the target driving force is calculated and controlled. Straighttraveling tests on flat sandy ground were carried out by using a model vehicle in order to examine the proposed method. The results of proposed method were compared with the results from even torque control and even wheel rotation speed control. As the results, it was found that the proposed method was the most efficient on the flat sandy ground.

This paper proposes an optimization procedure of handling performance of road vehicles by using a multi-objective optimization technique that considers driver’s subjective rating for handling performance. The paper applies the procedure to examine the effects of battery location of electric vehicles on handling performance. With this procedure, Pareto solutions are first obtained with respect to four objective performance functions using constraints method. Then, the Pareto solutions are narrowed down by comparing them with the optimal domain of driver’s subjective rating. To clarify the effects of battery location on the handling performance, proposed optimization procedure is applied to the optimization of front and rear tire cornering stiffness of linear vehicle model with varying battery location. From the optimization results, following two findings are obtained: 1) the proposed procedure is useful to contract the Pareto solutions, 2) the battery location that minimizes yaw inertia is preferable from an improvement in handling performance.

In this paper, a novel driving force and lateral force distribution method is proposed. Four wheel independent driving and front and rear active steering enable vehicle to have a redundant control input. This redundancy should be used so that maximum tire workload becomes minimizing for vehicle stabilization. However, conventional distribution method has been used minimizing sum of squares method. Therefore, a fast distribution method using the property of an infinity norm, Equal Magnitude Property, minimizing maximum tire workload is proposed. The effectiveness is verified by simulations and experiments.

Electric vehicles have an advantage over internal combustion vehicles in that the forces applied to the wheel can be estimated from the wheel velocity and input torque. By exploiting this feature, the authors propose to apply force control, which is often used for industrial robots, to EVs in order to achieve much more human friendly EVs. Many applications of force control for EVs are possible. In this paper, novel hand parking assist control is proposed for in-wheel-motor with reduction gear. The position of the vehicle body is adjusted by the human hand outside vehicles. The high backdrivability is achieved by using impedance control and driving force control. External forces applied by human hands and driving forces are estimated by disturbance observer with high-resolution load-side encoder.

Aerobatic maneuvers of an aircraft are useful for quickly execution of flight tasks or recovery maneuvers from dangerous state. However, such aerobatics are performed out of trimmed state and autonomous position control law based on near-trimmed flight is inapplicable. In this paper, position compensator is proposed that enables the trajectory tracking of agile maneuvers requiring attitude rotation and the reverse of main rotor thrust of single-rotor helicopters. The outer loop position compensator calculates desired values of attitude and thrust required to position feedback compensation by utilizing a dynamical relation established in local model of reference trajectory. To achieve high speed tracking of lateral-longitudinal angular rates in the inner loop, two-degrees-of-freedom servo system is designed. The control system with the position compensation is applied to autonomous control of flip maneuvers in which the helicopter is transferred between normal and inverted hover flight. Verification by simulation and flight tests of the autonomous flip maneuvers show the possibility of the position feedback compensation that reduce error of position tracking during flight state including out of trimmed state.

Flapping-Wing Micro Air Vehicle (FWMAV) have lately attracted considerable attention since it can achieve complex flights such as hovering, high speed flight and steep turn. These characteristics would enable FWMAV to achieve missions in hazardous locations where people cannot enter. However, few FWMAVs, which can fly stably, have been developed because the UAV is greatly influenced by wind disturbances. To overcome this problem, we apply H_∞ control theory to the flight system so as to have strong robustness against wind disturbance. We confirm the effectiveness of the proposed control system for FWMAVs by numerical simulation.

Intelligent Transport System (ITS) has been intensively studied in the last decade to solve environmental and energy problem by improving traffic flow. Along with the development of ITS and autonomous driving technologies, vehicle velocity will be decided by ITS. In this paper, range extension autonomous driving (READ) system which minimizes consumption energy for electric vehicle (EV) is proposed. The proposed method optimizes velocity profile and direct yaw moment simultaneously in consideration of cornering. The effectiveness of the proposed method is verified by simulations and experiments.

It is difficult to represent by physical quantities all phenomena that affect riding comfort. Whereas, it is known that the most important factor is the vibration acceleration of a vehicle. Until now, the riding comfort evaluation has executed through a test run by a real train. Testing conditions are apt to change test after test. We used a simulator to reproduce the acceleration of the railway vehicle. We can repeatedly evaluate riding comfort by using this simulator. The simulator was devised by combining the universal shaking table with a translation table and an inclination table. Depending on the frequency, a program can distributes the acceleration to the above components. A subject presses a switch when the subject feels bad ride. We got two types of data from simulations. One is a subjective data, namely, the time that subjects press the switch, and the proportion of the number of subjects who pushed the switch in bad ride. The other is an objective data such as vehicle’s running acceleration, jerk, and the tensile force of a strap. In this study, we clarify the correlation between the subjective data and the objective data. Then Correlation will lead us to evaluate the riding comfort from physical quantities. However, difference between two types of data will exist. The difference is called “delay time”. In this study, we will report about the delay time.

We evaluated the ride comfort of high-speed railway vehicle for the section between Morioka and Shin-Aomori in Tohoku Shinkansen-line by using the optical topography. Although the result shows the unclear correlation between the ride quality level and the stress brain index (LIR) estimated by the difference between left and right cerebral blood flow, we obtained the clear correlation between the tendency of ride quality level variation and LIR variation for the each section. For the correlation parameters between differences of LIR and ride quality level, 0.72 for the vertical vibration and 0.69 for the lateral vibration, respectively, were obtained. In addition, we found that the objective estimation of the optical topography has a high potential to evaluate the riding comfort for the total experimental condition in detail under the time series and/or the running section.

In this paper, we study road crossing behaviors of pedestrians in order to develop pedestrian agent. The purpose of the agent is to evaluate safety system, which focus on reducing vehicle-pedestrian accidents. Pedestrians who collided with vehicles are often seriously injured or killed because they are not protected with collision safety devices. Therefore, it is important to prevent pedestrians from colliding with vehicle to reduce the number of fatalities by traffic accident. To prevent pedestrian accidents, it is necessary to predict the movement of pedestrian. For this reason, this study sets its final goal to develop a pedestrian model, which represents the actual pedestrian movement of the traffic environment. This paper describes our field survey and result of analysis of behaviors of pedestrians crossing roads without crosswalk. As a result, we find the feature of pedestrian behavior, including crossing angle, speed and relationship of vehicle movement.

This study focuses on dash out behaviors of pedestrians, which is one of the most unsafe behaviors with the object of traffic safety. Since passengers’ fatalities of automobiles have been decreased remarkably thanks to passive safety technology and preventive safety technology, it is desired to reduce the pedestrian fatalities for the future. For this reason, this study sets its final goal to develop an active safety technology to prevent pedestrian accidents. In this study, we focused on the near-crash incident scene with a car by dash out behaviors of pedestrians. Therefore, the dash out behaviors of pedestrians are first extracted from near-crash incident database and then analyzed the feature of dash out behaviors of pedestrians quantitatively. Next, a mathematical model expressing dash out behaviors is proposed based on the analysis results. As a result, it is confirmed that the pedestrian model reproduces the behavior of the pedestrian, capturing the feature of the behavior of the real pedestrian.

Recently, increasing numbers of very small-sized vehicles like a personal mobility have been developed. As the mass of such small-sized vehicle is often comparable to that of their drivers, the coupling behavior between them is increasing in comparison with traditional vehicle types. Therefore, when developing small-sized vehicles, it is necessary to give ample consideration to the dynamics of the human body riding inside them. In this research, a model of a human body inside a small-sized personal vehicle is proposed. However, whole body model has multiple degrees of freedom and property. Furthermore, the influence of parameters of human is uncertainty and it is difficult to set accurate parameters. Accordingly, the dynamics of only trunk of the human body on a vehicle has been studied. And numerical simulations were used to investigate conditions that exist when lateral acceleration is applied. Then the parameters of numerical simulation are identified by Genetic Algorithm (GA). Consequently, Findings of human motion gained by identification result.

In order to establish really useful automated driving systems, it is necessary to take into account that both humans and machines are imperfect. The fact that they are not perfect looks obvious, but it is sometimes ignored. In this paper, we discuss issues to be addressed for mutually complementary human-machine systems. Among several topics, there are two important ones, i.e., (1) enhancing manual driving, and (2) designing systems that could helpful to improve human’s driving skill. These topics will be presented in this organized session.

This paper discusses design of control authority shifting between a driver and an assistance system while changing lanes under shared control. Two means are distinguished for system’s intervention into steering maneuver during lane change: (i) to increase the steering friction torque to resist hazardous lane changing maneuver (haptic system), and (ii) to cancel the steering input by the driver in order to prevent hazardous lane change (mixed system). Results of an experiment with a driving simulator showed: 1) Drivers’ reaction time was earlier and the drivers were less error-prone when supported by the haptic system, 2) Both systems improved safety up to a certain level, 3) More collisions were observed when drivers had to avoid sudden traffic changes while receiving assistance, and 4) There was a positive relationship between drivers’ willingness to cooperate with a system and their ability to regain the control.

When The aim of this study was to investigate effects of different methods of handing control back to drivers on drivers' performance. A driving simulator with thirty-six participants was used for this. The following tree methods were evaluated; ( Method A) handing all control of driving back to the drivers immediately after the limits of system functioning were reached caused by heavy fog, (Method B) continuing automated driving at a less accurate level by using facilities unaffected by heavy fog, (Method C) decreasing vehicle speeds in addition to Method B. Results showed that there was a possibility that compared to Method A drivers could resume control safely and drive more smoothly with Method B and C. Potential implication on system design of automated driving will be discussed in response to differences in the taking-over methods.

Excessive use of driver assistance systems might reduce the chances of developing drivers’ skill thought such assistance systems have greatly succeeded to reduce drivers’ workload and collision risks. Effective collaboration between a human operator and an automated system can be achieved by the tuning controller, which is introduced to facilitate the improvement of human operators’ skills. We discuss a methodology to decrease drivers’ workload and improve their driving skills using shared control methods. As an example, backward parking assist by haptic shared control is presented to increase performance during assistance and skill improvement after assistance.