In this paper, we provide a new approach to classify and recognize the acoustic events for multiple autonomous robots systems based on the deep learning mechanisms. For disaster response robotic systems, recognizing c...In this paper, we provide a new approach to classify and recognize the acoustic events for multiple autonomous robots systems based on the deep learning mechanisms. For disaster response robotic systems, recognizing certain acoustic events in the noisy environment is very effective to perform a given operation. As a new approach, trained deep learning networks which are constructed by RBMs, classify the acoustic events from input waveform signals. From the experimental results, usefulness of our approach is discussed and verified.展开更多
Many smaller manufacturers utilize production systems such as ＂individual production＂ and the ＂production of many models in small quantities.＂ In the processes for the ＂production of many models in small quantit...Many smaller manufacturers utilize production systems such as ＂individual production＂ and the ＂production of many models in small quantities.＂ In the processes for the ＂production of many models in small quantities＂ by multi-skilled workers, a manufacturer cannot make good use of the management method with standardized work tables used by large enterprises. Thus, it is necessary to develop ways to f＇md standardized work that is suitable for smaller enterprises. Therefore, we investigated two kinds of factories that utilized flexible manufacturing procedures for the ＂production of many models in small quantities,＂ and considered their work instructions, work contents, changeover methods, workers＇ abilities, etc. One of the compared processes was a case where there was defined standardized work, which was not observed. In the other, a QA （quality assurance） network was created to guarantee the quality. Based on the investigation results, we formulated improvement plans, had the supervisors review them, and clarified the characteristics of their problems. The results revealed areas where much attention was required to find standardized work for such processes. In addition, we made it possible to support the specification of difficult tasks and the renewal of standardized work by extending a standardized work table for the entry of workers＇ actions. This paper reports a method to define and make good use of standardized work tables and standardized work combination tables, i.e., a way to find and use standardized work based on the results of the previously mentioned factory investigation.展开更多
文摘In this paper, we provide a new approach to classify and recognize the acoustic events for multiple autonomous robots systems based on the deep learning mechanisms. For disaster response robotic systems, recognizing certain acoustic events in the noisy environment is very effective to perform a given operation. As a new approach, trained deep learning networks which are constructed by RBMs, classify the acoustic events from input waveform signals. From the experimental results, usefulness of our approach is discussed and verified.
文摘Many smaller manufacturers utilize production systems such as ＂individual production＂ and the ＂production of many models in small quantities.＂ In the processes for the ＂production of many models in small quantities＂ by multi-skilled workers, a manufacturer cannot make good use of the management method with standardized work tables used by large enterprises. Thus, it is necessary to develop ways to f＇md standardized work that is suitable for smaller enterprises. Therefore, we investigated two kinds of factories that utilized flexible manufacturing procedures for the ＂production of many models in small quantities,＂ and considered their work instructions, work contents, changeover methods, workers＇ abilities, etc. One of the compared processes was a case where there was defined standardized work, which was not observed. In the other, a QA （quality assurance） network was created to guarantee the quality. Based on the investigation results, we formulated improvement plans, had the supervisors review them, and clarified the characteristics of their problems. The results revealed areas where much attention was required to find standardized work for such processes. In addition, we made it possible to support the specification of difficult tasks and the renewal of standardized work by extending a standardized work table for the entry of workers＇ actions. This paper reports a method to define and make good use of standardized work tables and standardized work combination tables, i.e., a way to find and use standardized work based on the results of the previously mentioned factory investigation.
