Masami Aga

Collision-mitigation level of collision-avoidance braking system

In this study, a methodology for estimating the collision mitigation ratio using collision-prevention support braking system is discussed. A method for analysing the collision-mitigation ratio is discussed through a time series Monte-Carlo simulation based on the integrated error of driver and system. First, driver performance in terms of braking timing and deceleration level for collision avoidance was investigated in a driving simulator when the control timing of the braking-support system was changed. In this analysis, the braking behaviour of the driver during malfunction of the system and this behaviour when the driver was informed about the malfunction of the system were also investigated. Next, a driver model simulating braking operation, when a preceding vehicle started slowdown, was constructed. Through time series Monte-Carlo simulations using this driver model, the frequency of collisions to a preceding vehicle and the collision velocity were estimated.

Accident analysis on intersection right turning by using driving simulator

We simulated an accident prone traffic scene with our highly realistic driving simulator. In this research, the scene features a motorcycle suddenly appearing from a behind a line of vehicles in a congested street into the path of an oncoming car turning right at an intersection. As a result, the driver often only has less than one second before applying brake, that is to say, the permitted avoidance time is short. To avoid a collision between a vehicle turning right and an oncoming vehicle, it is important to predict the possibility of danger in the intersection and lower the vehicle speed while driving through the intersection. Driving behaviour in a collision was compared among different ages of drivers and manoeuvre differences before collision were confirmed. The analysis of the collision patterns and driver collision avoidance behaviour described above suggests that enhancing forward collision avoidance support systems.

Human centered design of a pre-collision system

Human error such as distraction and inattention while driving is one of the major causes of the rear-end collisions. In order to help reduce those accidents, a pre-collision system (PCS) has been developed and spread. A PCS basically consists of the warning, the brake-assist, and the autonomous brake functions. The total effectiveness of the system depends on not only the autonomous brake performance but also the drivers reaction to the warning and the braking maneuver. Earlier activation timing can make the system more effective in terms of collision mitigation/avoidance; however, the drivers may feel nuisance if the timing is too early. Thus, human factors issue should be considered in designing and evaluating a PCS. This paper describesthe human-centered design of a PCS from object recognition to the effectiveness estimation method.

Estimating damage-mitigation level of collision-prevention support braking

In this study, collision-prevention support braking was used as an example, and a methodology for estimating the collision mitigation ratio using this system is discussed. A method for analyzing the collision-mitigation ratio is discussed through a time series Monte-Carlo simulation based on the integrated error of driver and system [1][2]. First, driver performance in terms of braking timing and deceleration level for collision avoidance was analyzed in a driving simulator when the control timing of the braking-support system was changed. Next, a driver model simulating braking operation, when a preceding vehicle started slowdown was constructed. Through time series simulations using this driver model, the frequency of collisions with a preceding vehicle and the collision velocity were estimated. One result analyzed through this simulation study based on the integrated error of driver and system considering the drivers risk taking behavior during the use of the system estimates that the frequency of collisions decreased from 6.64×10−1 to 1.00×10−5 when the driver used the driving support system for collision avoidance. Although further verification regarding the driver error and system error will be necessary, a methodology for evaluating effectiveness of driving-support system is shown in this study.