Floor plans are widely used for representing residential spaces. This study aims to propose a feature representation method for floor plans, namely unstructured image data, using a Random Visibility Graph and Graph Neural Network. In this method, the feature representation is performed using Invariant Information Clustering. To prepare a dataset, the Random Visibility Graph is designed to derive multiple graphs from a single floor plan effectively. The obtained features formed shape-oriented clusters, rather than ordinary variables such as area. In addition, the similarity of features corresponds to the similarity of plans, including slight differences in the connectivity of rooms.

In this study, we aim to enhance the productivity and utilization of robots in construction sites by developing an optimal route planning system for autonomous mobile robots. The system generates lightweight graph-based maps from BIM and city models, enabling rapid pathfinding in large environments like university campuses. It maintains map data on a server and distributes route information via a web API. Validation with wheeled robots, humanoid robots, and drones confirmed efficient execution of point-to-point, multi-point, and detour routes. Updating server-stored map data improved route planning efficiency. Future work will focus on incorporating dynamic information to further enhance pathfinding.

On construction sites, robots are expected to move around and perform tasks. Although it would be efficient if the robot movements could be generated using the design data as is, there may be temporary structures at the sites that are not in the design, or the work may not proceed as planned. The robot’s motion plan must be modified to cope with such situations. In this research, we developed a method to realize stable and efficient machining by recognizing the position and posture of the work surface, which are difficult to predict in advance, and adjusting the robot motion on-site.

By using a gripping device with a positioner, we have successfully achieved efficient processing with minimal material defects, even for small-diameter, irregularly shaped materials like thinned wood, resulting in high-quality building components. We developed and utilized a unique multi-axis processing machine with 9 degrees of freedom for this purpose. In a demonstration test, we produced a frame fence from high-design-quality heptagonal cross-section materials. Each joint of this fence uses traditional Japanese joinery techniques, showcasing the feasibility of intricate processing for polygonal cross-section timber.

We were conducting literature surveys on damaging earthquakes after the Meiji era in the Northern Nagano Prefecture. In this paper, we focused on the 1941 Nagano Earthquake, affected at northeast of Nagano city and its vicinities with magnitude 6.1. The details of the damage were clarified through investigation the reports and newspaper articles from that time. It was pointed out that most of the damage occurred in lowland areas along the Chikuma River. Areas along this river are prone to flooding and including recent one in 2019. The relationship between them was also discussed.