| 中部 仁 | M, 2回目発表 | ユビキタスコンピューティングシステム | 安本 慶一, | 清川 清, | 諏訪 博彦（特任准教授） | |
|
title: Creating Optimal Schedule for Multiple Parallel Recipes Under Arbitrary Cooking Environment
abstract: In recent years, the number of people who cook their own food has been increasing, and there is a demand for the presentation of optimal cooking procedures to enable efficient preparation of a menu that includes multiple dishes. On the other hand, it is difficult for the cooks themselves to determine the optimal cooking procedure because the cooking environment differs for each cook and the optimal cooking procedure varies from cook to cook. Therefore, in this study, we propose a method of creating optimal cooking procedures that captures the parallel cooking of multiple recipes as a task scheduling problem by creating a task graph for each recipe. Based on the optimization algorithm DF/IHS method for the task scheduling problem, we extend the pre-processing part and the pruning part to take into account the washing of dishes specific to cooking, to search for the optimal cooking procedure efficiently. In the evaluation experiment, the proposed algorithm was applied to an actual recipe, and it was confirmed that the proposed algorithm was able to search for an optimal cooking procedure that reduced the total cooking time by 15%, including washing up, compared with a manual cooking procedure. language of the presentation: Japanese | ||||||
| 松田 明大 | M, 2回目発表 | ユビキタスコンピューティングシステム | 安本 慶一, | 清川 清, | 諏訪 博彦（特任准教授）, | 松田 裕貴 |
|
title:Detecting Street Parking Using Dashboard Camera Videos
abstract:In recent years, street parking in prohibited areas has become a social problem, especially in urban and tourist areas. In addition, because street parking can cause traffic congestion and accidents, real-time detection is required. The detection of street parking has been previously implemented on the basis of comparisons of videos recorded by fixed-point cameras. However, this approach has a limited detection area and low accuracy. To overcome these problems, this study aims towards a real-time street parking detection system that uses dashboard camera videos. We propose a machine learning method based on the characteristics of on-street parked vehicles derived by transforming images into text. The object detection model YOLOv3 was used to analyze videos. We created a dataset based on the coordinate information of 1765 vehicles and the recording vehicle information. We also created a model using random forest and logistic regression algorithms and evaluated it using the holdout and stratified 5-fold validation methods. F-measure values of up to 92% and 89% were obtained for the two types of model, respectively. These results confirm the effectiveness of the proposed street parking detection method based on bounding boxes and recording vehicle data. language of the presentation: Japanese | ||||||
| 熊本 光 | M, 2回目発表 | ロボティクス | 小笠原 司, | 清川 清, | 高松 淳, | GARCIA RICARDEZ GUSTAVO ALFONSO |
|
title:An Underwater Suction Device for Underwater Robot Manipulation
abstract: In recent years, UVMS (underwater Vehicle Manipulator Systems) has been used for underwater tasks. Many existing systems are equipped with parallel grippers and multi-finger hands as end effectors for manipulators. However, the difficulty of gripping planning and the time required for grasping are issues. In this research, I propose to develop a suction device that solves these issues and apply it to underwater manipulation tasks. The proposed suction device is designed to be small and lightweight so that it can be mounted on an underwater robot. In addition, by using a thruster drive for the suction system, the suction force can be controlled to match the grasping object. In the experiment, a physical model of underwater suction was verified in order to evaluate the grasping performance of the prototype suction device. Furthermore, as a qualitative evaluation, the grasping of various objects was verified. In the future, I will work on the application to underwater tasks using underwater robots in real. language of the presentation: Japanese | ||||||
| USAWALERTKAMOL BUNYAPON | M, 2回目発表 | ロボティクス | 小笠原 司, | 清川 清, | 高松 淳, | GARCIA RICARDEZ GUSTAVO ALFONSO |
|
title: Human-Robot InteractionSystem for Non-expert users by using Behavior Tree
abstract: In the present-day that robot has taken the important roles in human life, service-robot is one of the most popular types of robots used its ability to interact with humans or HRI. It is used in various situations, such as for smart homes, industrial factories, or convenience stores. Most of the service robots are programmed from the distributor or the factory to do the fixed-job, and it is difficult for non-expert users such as store staff to edit or make change. This research proposes the system for non-expert users to use, edit, and create the robot behavior for the situation they are in by using realtime human action recognition to indicate the robot reaction. The recognized action will be putting in the Behavior Tree in order to create the desired human-interactive system also, by providing user interface non-expert users will comfortable for make their own Behavior Tree. language of the presentation: English | ||||||