Study on predicted reaching position of moving body using video stimulus

摘要

Remote-controlled robots, which are expected to perform useful tasks, are required to move as intended by operators. The information used to perceive the remote environment including the robot is limited and the operators usually have to perceive or predict motions of the robot through visual information included in a video image. The limitation makes the remote operation difficult. In order to investigate influences of the limitation, we measured operator’s predictions about a position of a remote robot. We had operators do tasks where they navigate the robot to a target. The task is divided into two parts. In the first part, the operators are requested to start the robot to move and given video or non-video stimuli. In the latter part, they are requested to stop the robot at the target without any visual information and the reaching position at which the robot has stopped is measured. Consequently, it was shown that the video stimuli modulate the reaching position of the robot. This means the operators' predictions are modulated. This effect should be caused by the difference between the velocity of the robot on the video stimuli and that on the non-video stimuli. In addition, we investigated influences of optic flows related to robots' motions by a similar task to the prior one. We gave the operators optic flow stimuli during the latter part of the task, while the video stimuli are given during the first part. Then, it was shown that the optic flow stimuli have additional effect on modulating the predicted reaching position which is already modulated by the video stimuli in the first part. It seemed that optic flows related to a robot's motion navigated in a remote environment may modulate a prediction of its operator. We may use this effect well when we operate robots remotely.