Abstract

Since 1965, a numerous number of cities implementing sewerage systems have increased rapidly throughout Japan, and sewerage development is considered to be becoming more widespread in various regions. However, with the increase of management facilities, the aging of facilities for long-term use is becoming more and more apparent. The standard expected durability of these pipes is approximately 50 years, but there is a tendency and a risk that the number of collapsed roads will increase rapidly 30 years after the pipes are laid. Against this background, maintenance of drainage and sewage pipes is critical and must be carried out continuously. Therefore, in recent years, investigation using robots have been actively conducted in order to reduce manual workload of the workers. However, these robots have a large-scale system as a whole, and as a result, they are poorly maintainable and expensive. Therefore, in this research, I have developed an autonomous and portable pipe inspection robot through the know-how on rescue robots which I have studied so far. However, for inspections using a pipe inspection robot, there is always the risk that the robot itself will tip over due to steps or small gaps at the joints of the pipes or slips caused by sludge. Therefore, to prevent tumbles and rollovers of the robot, I propose a localization method only by straight-driving control without relying on hardware. In addition, taking possible slips inside pipes into account, this method utilizes only acceleration sensor. In this study, localization method using only accelerometer mounted on the robot, which focuses on the relation between the pipe and the contact point of the tires, was shown as well as presenting a method using numerical analysis to derive the estimated values. Furthermore, it was confirmed that the estimation was stable as a result of an estimation experiment using autonomous small pipe inspection robot with and without a gradient (approx. 4/100) of a pipe, with a diameter of 189mm.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call