Compared to conventional visual- and auditory-based assisted driving technologies, haptic modality promises to be more effective and less disturbing assistance to the driver. However, in most previous studies, haptic assistance systems were evaluated from safety and stability viewpoints. Moreover, the effect of haptic assistance on human driving behavior has not been sufficiently discussed. In this paper, we introduce an assisted driving method based on haptic assistance for driver training in reverse parking, which is considered as an uncertain factor in conventional assisted driving systems. The proposed system assists the driver by applying a torque on the steering wheel to guide proper and well-timed steering. To design the appropriate assistance method, we conducted a measurement experiment to determine the qualitative reverse parking driver characteristics. Based on the determined characteristics, we propose a haptic assistance calculation method that utilizes the receding horizon control algorithm. For a simulation environment to assess the proposed assistance method, we also developed a scaled car simulator comprising a 1/10 scaled robot car and an omnidirectional camera. We used the scaled car simulator to conduct comparative experiments on subjects, and observed that the driving skills of the assisted subjects were significantly better than those of the control subjects.

In this article, we introduce a novel method of assisted driving system intended to help drivers to improve the reverse parking skill. The system is able to assist the driver by applying the haptic instruction on the steering wheel in order to prompt the driver to make the adequate operation. This is helpful to train the driving technique in an intuitive manner. To determine the assistance means, we conducted the preliminary experiment in order to investigate the quantitative characteristics of human driving. Afterwords, we developed 1/10 Scale Car Simulator as a simulation environment on which we installed the proposed assistance method. Finally, we conducted the experiment with subjects by using the simulator. In consequence, we validated that the proposed assisted driving system could accelerate the learning of the humans' driving skills.

No abstract

No abstract

No abstract