抄録
The total length of sewer pipelines in Japan is over 420,000 kilometers. Quantitative measurement methods for regularly scheduled inspection of those pipelines have been desired instead of experienced observation on in-pipes animation images, which are obtained from an autonomous vehicle robot equipped with a CCD camera. In this study, we developed vehicle robots equipped with integrated sensing system using two types of sensors, so that measuring unevenness of the sewer pipe precisely and quantitatively is achieved by a simple measurement way at comparatively low cost. To accomplish these missions, two vehicle robots are designed and manufactured respectively. The first vehicle is specialized in the function of driving to pull a rope, which is connected to the second vehicle. The second one is specialized in the function of measurement to be towed. Two methods to integrate the data on the evenness, which are obtained by a gyro sensor and an inclinometer, are proposed so that each sensor faults are mutually compensated. The first integration method is to estimate the attitude angle applying the Kalman filter algorithm, and the second method is to assign well-performed sensor data to static and dynamic state each other. The experimental study shows that the integrated algorithms obtain the unevenness effectively and precisely.
