A study on the integration of Control Engineering and Psychology
What kind of information do drivers perceive, and what strategy do they use for their driving?
In order to answer these questions, driver behaviour such as steering input has been analysed through various psychology experiments. On the other hand, we can analyse driver behaviour without the experimental environments if driver's mathematical models can be constructed. We have constructed driver's mathematical model inspired by the knowledge of various results in the psychological experiments, and then we have analysed driver behaviour based on the driver's mathematical models. I believe we will be able to answer "What is human-like driving?" by these approaches which tackle with both the psychological experiments and the driver models.
Briefly, it is interesting that we can construct a mathematical model that can represent driver behaviour. My research aim is to find new knowledge through the fusion of both Control Engineering and Perceptual Psychology.
Optical Flow Control (Engineering)
Humans can perceive the direction of self-motion based on optic flow, and they can track a target path by matching their direction to a point of "they want to go" on the target path. This is a human (animal)-like control method. Therefore, we made a mathematical model of optic flow and introduced this human-like method into the vehicle steering control based on a nonlinear control method. From the results of the simulations and vehicle experiments, we confirmed this method is effective for the automated steering controller and can reproduce human's steering behaviour in terms of steering accuracy.
Examine Influence of Optic Flow (Psychology)
From the results of our proposed method, we got a hypothesis: the usefulness of a region of optic flow (for instance, far and near, or central and peripheral vision) for driver's steering behaviour is not consistent. Then, we made a unique simulation environment which can selectively mask either the optic flow or the road edge information. Then, we verified how this two information affects the steering performance of the driver within the scope of Two-point steering model. Amazingly, a part of the results is the same as the results produced by the optical flow control. This conclusion could give us some intuition: driver models capturing driver characteristics can figure out new driver behaviour even if we do not conduct the psychological experiments.
PROFESSION / EDUCATION
[Oral] Y. Okafuji, T. Fukao, H. Inou, “Theoretical Interpretation of Driver’s Gaze Considering Optic Flow and Seat Position”, IFAC Symposium on Analysis, Design, and Evaluation of Human-Machine Systems, Tallinn, Estonia, September 16-19, 2019 (to appear)
[Oral] C. Mole, G. Markkula, O. Giles, Y. Okafuji, R. Romano, N. Merat, R. Wilkie, “Drivers fail to calibrate to optic flow speed changes during automated driving”, Driving Assessment Conference, New Mexico, USA, June 24-27, 2019 (to appear)
[Demo] Y. Okafuji, J. Baba, J. Nakanishi, “Face-to-Face contact method for humanoid robots using face position prediction”, ACM/IEEE International Conference on Human-Robot Interaction, Daegu, Korea, March 2019
Domestic Conference (in Japan)
[Poster] 岡藤勇希，深尾隆則，伊能寛，”オプティカルフローに基づいた自動操舵システム”，ロボティクス・メカトロニクス講演会 (ROBOMECH2014)，2014年5月
AWARDS / GRANTS