Thesis supervisor: Tamás Ruppert
co-supervisor: János Abonyi
Location of studies (in Hungarian): University of Pannonia, Engineering Faculty, Department of Process Engineering Abbreviation of location of studies: UPPE
Description of the research topic:
Short description the topic:
This research aims to use multimodal techniques with operator 4.0 and artificial intelligence to monitor and classify human intention. The proposed idea will use the following parameters as inputs to the deep learning neural network. Speech signals and gesture dynamic: this is user-dependent and based on specific body gesture that reflects the emotional status. Face expression and gaze tracking, and finally apply Galvanic Skin Response GSR (Skin Conductance)
Several elements of operator 4.0 are available in the production systems. The main operator 4.0 elements that need to use are the Analytical operator, Augmented operator, Healthy operator, Smarter operator and Virtual operator. Several cases could be studied during the research time. The human intention will be classified for each input separately for the first time. Then all the inputs will be fed together (speech signals, face expression, GSR, etc.) into the system to classify the intentions precisely. Classification of intentions will be achieved using the data from the participants, and also available free data.
The tasks of the applicant while using process mining would be:
• Develop a framework for monitoring human intention parameters.
• Gather data from physiological data, and vital sign of human operators
• Develop an AI system for processing the collected data individually and together
• Study the efficiency of the designed system for each input separately (Speech signals, face expression, GSR, etc.)
Antecedents of the research topic:
Operator 4.0 is a brand new concept which utilizes the Industry 4.0 existing technology, as well as the data from sensory in Industry 4.0, to monitor and control the utilization of human resources, in collaboration with each other and with machines. This concept aims at a real-time and dynamic pace, rather than the abovementioned static method. With the tremendous amount of acquired data from sensors, process -mining and -simulation are the most favorite techniques are deployed due to its forecast ability. By using this approach and the integration of the waste-elimination principles of Lean 4.0 concept, process abnormalities and deficiencies are detected sooner. Thus the improvement and adjustment can be made in a faster manner, without any physical and mental harm or damage to human operators, while tracking their performance and health continuously.
Number of students who can be accepted: 1
Deadline for application: 2021-08-31
2024. IV. 17. ODT ülés Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).
Login
