論文ID: 19-00145
Automated driving vehicle technology is expected to solve several road traffic issues. Therefore, this technology has been receiving increased attention and several relevant experiments have been conducted. This paper describes the development and verification of target trajectory and speed control target calculation. The control target calculation system of the autonomous vehicle algorithm, which is the focus of this study, mimics an expert driver's direction and speed control in detail. The proposed algorithm consists of a driver model, which considers the risk potential, and vehicle dynamics model like a human driver. We think that a human driver drives under the influence of some obstacles and the oncoming vehicle on the road. And they change a driving action so that their risk is always kept low. At first we measured the traveling path and vehicle speed as a driving action of the human driver by the experiment in the real traffic condition. Next, we reproduced the driving action of the human driver by the proposed algorithm. The parameter of the risk potential driver model was estimated by the experiment results. In addition, we performed inspection by the simulation about calculated control target. As a result, it was confirmed that direction control and the vehicle speed control by proposed algorithm simulated the driving of the human driver in detail.
