Search

Periodic inspections are necessary in maintaining concrete infrastructures. As efficient and low-cost inspection methods, visual inspection robots and hammering inspection robots to detect cracks and internal defects have been developed, respectively. However, these robots can perform only a single task. To realize multiple tasks by a robot, multiple measurement sensors must be installed considering the stability of the robot on the wall and measurement performance. Parameters, such as weight and size, must be considered to ensure the stability of the robot on the wall. We proposed a propeller-type wall-climbing robot that can capture images and perform hammering simultaneously while moving on the wall. The static model of robot was established by considering various parameters, such as the direction of the propeller’s thrust force, weight, and placement of the measurement sensors. This model increases the payload for sensor installation and enables movement on dusty concrete walls with a low friction coefficient. The image captured by the inspection robot was employed for crack detection and width estimation. We confirmed that the hammering sounds generated by the robot showed different frequency characteristics on clear and defective areas. The developed inspection robot can collect adequate data for concrete health monitoring.

Abstract In this paper, we propose a wearable tactile sensor for measuring the shape of micro-undulations on a hard surface. The proposed sensor has two layers. The inner one is a thin rubber layer into which a strain gauge is embedded that is formed around a user’s finger. The outer layer is a flexible structure that consists of numerous pins on a flexible sheet. The shape of micro-undulations on an object surface can be measured when a user wearing the sensor traces the surface. The results of an experiment, in which we compare the cases with and without the flexible structure of the sensor, show that our proposed sensor is sufficiently accurate to measure the shape of micro-undulations due to its flexible structure, which contributes significantly to improving its signal-to-noise ratio.

In this paper we propose an algorithm for learning from positive examples only, based on a modified cognitive control model. This problem arises when positive examples are readily available while negative examples are difficult or impossible to obtain. In particular, we are motivated by an application of the proposed algorithm to learning strategies from real-life optimisation problem solutions provided by skilled humans. Using a packing problem as an example, we performed an experiment in which the performance of the proposed algorithm was compared to that of one-class Support Vector Machines and auto-encoding neural network algorithms. The results of the experiment show that the proposed algorithm compares favourably to the other two algorithms.