情報処理学会 第79回全国大会講演要旨
1ZD-02
高速車線変更シーンにおける個人リスク知覚の分析
More and more Advanced Driver Assistance Systems (ADASs) are entering the market for improving both safety and comfort. However, although there are many researches about analyzing and prediction general risky level of driving scenes according to many features, but few of them were about analyzing individual risk perception. In our study we first describe some risk perception features, such as TTC (Time- To-Collision), THW (Time-Headway), and PICUD (Possibility Index for Collision with Urgent Deceleration), then we propose a feature accessing method based on random forest to analyze the importance of all related driving features.
We assume different drivers should differ in some degree on judging risky scenes during various driving situations, we use the features importance result to predict the risk level for target experiment participant.
In order to know which driving phase product the risky feeling, we briefly introduce the relationship between driving environment and driving assist system.
We can see from the result that risky feeling comes from the process between environment perception and situation judgment, and in this process different user usually have slightly different reaction. Actually we conducted subjective risky feeling experiment before, the details is shown in Table1. The flowchart of proposed method is shown in this paper. we will introduce risk perception features in details, introduce details about random forest ,report the experiment result and summarize finally.
We assume different drivers should differ in some degree on judging risky scenes during various driving situations, we use the features importance result to predict the risk level for target experiment participant.
In order to know which driving phase product the risky feeling, we briefly introduce the relationship between driving environment and driving assist system.
We can see from the result that risky feeling comes from the process between environment perception and situation judgment, and in this process different user usually have slightly different reaction. Actually we conducted subjective risky feeling experiment before, the details is shown in Table1. The flowchart of proposed method is shown in this paper. we will introduce risk perception features in details, introduce details about random forest ,report the experiment result and summarize finally.
