Development of dead-reckoning sensor system for indoor environments

This paper presents a method for constructing a dead-reckoning sensor system for indoor environments that enables autonomous control of a mobile robot. The proposed technique includes a method for achieving accurate autonomous control of robots. Using existing electronic equipment, we propose a system to measure the actual position and azimuth of mobile robots. The synthesis of information from sensor data into time series data of actual transition movement record of the mobile robot, and the algorithm of programming installed in a microcomputer and a PC for controlling peripheral devices around sensors influences the accuracy in estimating its position/posture. The dynamic characteristics of the mobile robot can be derived using induction theory for the system that installs a mouse device. The objective of the study for the mobile robot corresponds to a novel autonomous robot as an assistant without any guideline or other induction by GPS indoors. Here we discuss the adequacy of the sensor system that determines the positional accuracy of the robot. The position and orientation of the robot can be determined using a gyroscope and azimuth sensors. Finally, we investigate the performance of the robot indicated by the sensor system for a dead-reckoning strategy with optical sensors, orientation sensors, and gyroscope sensors to achieve highly accurate self-position estimation for a mobile robot moving indoors in the experiment.