Yuji Sagawa

Quantitative measurements of driver's motion after-effect

In this article, the motion after-effect (MAE) is treated, which takes place when one drives on a highway for a while. Ones sensitivity to motion in the direction of moving stimulation is reduced, resulting in misjudgements of cursing speed and sometimes in car accidents, especially when the driver intends to slow down the speed of the car. We have developed a system foe systematic measurement of MAE under various conditions; temporal frequency of stimulus, spatial frequency of stimulus, stimulus with depth cues, complex stimulus for the central as well as for the peripheral parts of the retina, and so on. The degree of MAE, the duration of MAE, was measured in terms of the duration of MAE. We obtained the following results: the degree of MAE is strongest around 5Hz; the degree of MAE extended lip to some value as the period of stimulation got longer; for low spatial frequencies, the duration of MAE decreased as the speed increased, and rose for high spatial frequencies; and so on. We paid attention to the two mechanisms of the visual system; central vision and peripheral vision. We synthesized a complex stimulus pattern consisting of two regions, one of which is to stimulate the central region of the retina and the other the peripheral region. It was possible to reduce the duration of MAE by showing a stimulus moving in the opposite direction of the periphery.

Implementing driver's behaviours to evaluate how to offer road information

An information offering system in a highway attracts attention as a method of a course instruction to an effective driver for a current traffic jam problem. However, when the effect of the course instruction by these information offering is evaluated from drivers behaviour etc, it is difficult to acquire the behaviour data of an actual driver. Therefore, the system which virtually evaluates the effect is needed. Then, the computer simulation is used before the method of inducing the route is introduced into an actual road. In order to evaluate the effect virtually, the driver model to imitate the behaviour of the driver on the simulation is needed. In the article, we suggest a driver model for the evaluation of the effect of these informations given to a driver. We focus on the number of cars in the circumference that can be recognized by a driver as well as the distance between the car in front and the drivers car itself. We conducted simulations on various combinations of the above two factors. We compared the results of the simulations with a previous studies on drivers perception of traffic jam. It turned out that both studies yielded reasonably similar results.

Behavior patterns emerging among mobile robots with a diversity of personalities cooperating in collection and cleaning-up tasks

We investigated the behavior patterns emerging among autonomous mobile robots with diverse personalities. They cooperate with one another in collection and cleaning-up tasks. First, we emphasize the rationale for introducing personalities into robots. Second, we present our assumptions on the common functions of the robots, i.e., visual functions, mobility, and the capturing and conveying of target objects. Third, we define a function which is specific to each robot; its personality. Fourth, we explain the collision avoidance behavior, which is very dependent on the personalities. Finally, we present the results of simulation studies of robots with diverse personalities cooperating in collection and cleaning-up tasks. We discuss the influence of the distribution of various personalities on the performance of the tasks.

Character Input by Gesture Performed with Grasping Both Ends of Tablet PC

We developed a method to enter Hiragana (cursive in Japanese) characters and alphanumeric characters by thumb gestures for Japanese people. This method is optimized to be used while holding the left and right ends of a tablet PC with hands. Each gesture is designed by concatenating a few horizontal and/or vertical strokes, but path of the gesture remains inside the one-stroke square centering its starting position. A Hiragana character is input by selecting a row of the table of Japanese syllabary and then a column of the table. The alphabet characters are segmented into 6 groups in order of letters. Just like Hiragana, an alphabet character is entered by selecting a group and then selecting a member. However, in the alphanumeric mode, to reduce the cost of modifying the mistakenly selected group, an operation to move to the previous or next group have been added.

SliT: Character Input System Using Slide-in and Tap for Smartwatches

One character is entered with two steps: firstly, selecting a group of five characters, and secondly selecting one character from the group. For Japanese Hiragana letters, each row of a table of the Japanese syllabary, is used as the group. For alphabets, every five letters in alphabetical order are gathered to one group.