The development of computers has led to much research on plant-based systems. But not many have focused on flowers, one of the most fascinating aspects of plants. In this study, we focused on the morning glory, which has the characteristic of changing flower color, and created an interactive flower dome that changes the flower color of the morning glory when carbon dioxide is added. Although it is generally known that the flower color of morning glory changes, the conditions necessary for the change have not been investigated. Therefore, in this paper, investigated and examined the conditions under which the flower color of morning glory changes too. The results showed that the addition of 100 ml/10 s of carbon dioxide to a container of approximately 900 ml capacity for one minute and two minutes, and waiting for three minutes for the former and one minute for the latter, caused the flower color to change to red. It was also found that the flower color returned to blue after waiting approximately 20 minutes after the change. In addition, comparison of the color differences showed that it is possible to distinguish between the altered red state and the altered blue state.

In general, the delivery planning problem used to obtain efficient routes for multi-agents requires all nodes to be visited, so the cost per agent (travel distance, time, etc.) becomes very high when the number of agents is small and the number of nodes is large. Therefore, there are cases where the route obtained is not practical, even if it is the optimal solution in the delivery planning problem. For example, in a time-limited battle royale game, the optimal route cannot be obtained by implementing an existing method for the delivery planning problem in the AI of a battle royale game because there is not enough time to visit all the buildings. Therefore, we propose a formula for obtaining a travel route that can select nodes under cost constraints in order to obtain a practical and efficient route even under the above circumstances. We then compare the efficiency of the existing solution methods with that of VRP, a general delivery planning problem, and DCVRP, which allows cost constraints to be set. The results show that the proposed method is more efficient than the others when not all nodes can be visited under cost constraints.

This paper proposes a visual simulation method for fog with a large amount of dry ice, such as that produced when dry ice is added to a liquid such as water. The proposed method uses a polygonal model of a simulation source and a vessel that serves as a fog collision and adds noise animated in the vertical direction to the simulation source to reproduce the upward motion of bubbles, as would occur when dry ice is thrown into the water. A lattice-based fluid simulation is performed on the created source to generate a dry ice mist. The original buoyancy equation proposed in this paper is used to generate the velocity field. The disappearance of the fog is reproduced by controlling the density using the phase transition equation based on the saturated water vapor content. This method can efficiently reproduce, for example, fog generated from dry ice overflowing from a vessel.

The global spread of COVID-19 has highlighted the importance of avoiding the Three Cs (closed spaces, crowded places, and close contacts) to reduce the risk of infection. As a result, research on analyzing human flow to recognize walking states and patterns has gained traction. Visualizing people flow has been used in several studies as a means of analysis. From the perspective of infection prevention, reducing congestion during events is a crucial issue. Furthermore, decreasing the proximity between individuals is particularly effective. However, there are few and far between the visualization of human flow based on proximity methods. In this study, we propose a method for visualizing proximity status through the creation of a network formed by connecting pedestrians in proximity. We aim to visualize walking patterns with a high risk of infection by analyzing the walking paths of proximity clusters detected from the network. Additionally, we present a case study of the application of the proposed method for visualizing pedestrian proximity using real-world human flow data collected at an event site.

With recent advent of numerous 3DCG production software, the difficulty of 3D motion production has been greatly reduced, because production processes and user interfaces supporting them have become more sophisticated. However, the basis of motion production still lies in keyframe manipulation, and thus, considerable skill is required to exude the naturalness of even simple motions. On the other hand, actual dance motion production processes are much simpler. Especially in house dance, which is a compilation of dance cultures from past to present, dancers have created various choreographies by considering each motion of body parts as an element and going through a constructive production process to link them together. In this study, we clarify the relationship between various motions and operations used in actual choreography production to propose a visual programming environment that enables intuitive production of a wide variety of house dance motions. The proposed system reproduces the actual process of choreography production by connecting nodes that represent operations on elementary motions interactively, and further reproduces the actual process of sequence creation by smoothly connecting the created choreographies in order. Empirical evaluations proved the naturalness of the created dance motions and the usefullness of the proposed environment.