In Japan, aging of infrastructure and shortage of the labor have become social problems. It is a socially demanded, to realize efficient inspections and maintenance. In this paper, we mention on the “infra-robot evaluation project'' conducted by MLIT in 2014-2015. Then, based on its results, we consider the superiority of the robot-based inspection of the bridges, and proper procedure for practical inspection. Finally, we propose the performance evaluation method to promote the practical use.

Exploration UAVs (Unmanned Aerial Vehicles) are required to fly autonomously, because GPS cannot be used on Mars. Hence, localization is one of the major topics to be developed. In general, visual odometry is available for relative position estimation. However, it is difficult to estimate absolute position. This paper deals with the path planning in consideration of the absolute localization. This paper uses Mutual Information for image matching, which is robust to scene changes. MCL (Monte Carlo Localization) is applied for the absolute localization. Then, this paper proposes a new RRT (Rapidly-exploring Random Tree) extension method to generate a path with a large effect of the absolute localization. The simulation results show that the proposed method reduces the uncertainty of the position and generates a path that can increase the probability of reaching the goal.

In recent years, multi-rotor UAV has been used in various fields. Among them, what is expected to be particularly active is the delivery of packages in the field of logistics and the transportation of relief supplies in the event of a disaster. In order to realize the transportation business using multi-rotor UAV, robust flight without sudden posture change is required. In this research, we focus on the design of the attitude and angular rate control system. We aim to realize an attitude control system of a multi-rotor UAV that is highly robust against disturbances that occur in the outdoor environment and modeling errors of controlled objects by using super twisting sliding mode (STSM) control. As a result of numerical simulations and demonstration experiments, it was shown that STSM control has excellent tracking performance and high robustness.

In recent years, many deep learning methods have been proposed, but the annotation process for creating datasets is a time-consuming and labor-intensive task. In this study, we propose a fluorescent texture to generate a 2-3D dataset that can be used in visible light. The fluorescent texture uses fluorescent paint, which is transparent under visible light but can be recognized under UV light. Target object can be made measurable by applying the texture. The fluorescent texture is an extensible method and can change the annotation data depending on the representation of the texture. In this study AR markers and grid textures are given to target objects using fluorescent textures. By applying existing methods such as marker recognition algorithms and stereo vision algorithms to the fluorescent texture, we can automatically annotate 3D information such as object position, orientation, and point cloud, as well as image region for semantic segmentation. Fluorescent textures can be applied to not only general objects but also objects that are difficult to recognize. The accuracies of the point cloud were as follows, general objects 1.7[mm], transparent containers 1.9[mm], and metal plates 1.7[mm].

Soft robots have been attracting a lot of attentions in recent years because they have completely different characteristics from conventional robots having rigid structures. We have proposed a soft microrobot using a thermoresponsive gel as a soft actuator. This robot can be driven by being irradiated with light for controlling its temperature. In this study, we drive a robot having three gel actuators with a straight shape with peristaltic motion by scanning the robot with a laser. We evaluated actuation timing of each actuator and a displacement of the robot during the peristaltic motion.