In the last decade, the industry is striving to keep up with modern market requirements and embrace mass customization in a productive and cost-efficient manner. This suggests the introduction of Information and Communication Technologies and Industry 4.0 schemes in the manufacturing lines.
Subsequently, operators need to adapt to constantly changing environments, either in terms of workstation resources or in terms of manufactured products. Additional factors, involving fenceless coexistence, sophistication of modern tools, data acquisition and monitoring procedures, etc. indicate an increased volume of information that needs to be learned before operators become “masters of their work”. A significant challenge is how operators can be trained in shorter periods of time and obtain more knowledge than ever.
Studies reveal that despite advances in manufacturing technologies, the training methods remain the same and involve traditional practices (e.g., manuals, seminars). The main reason is that the enhancement of training practices requires significant funding and businesses cannot formulate pioneer training frameworks under the constantly changing conditions of Industry 4.0.
Sharework introduces pioneer means of training using Augmented Reality technology, supported by learning based on object recognition tools, towards steeper learning curves and gamification of training practices.
Within Sharework, a novel training framework has been introduced based on Augmented Reality technology. It supports direct interaction between the operator and the surroundings, allowing a straightforward familiarization with hybrid workstations, IoT technologies and customized assembly operations.
It provides an intuitive way to record and visualize training material that would be difficult to explain using conventional manuals, or video guiding approaches. A gamification of the application environment suggests increased operator interest and better comprehension of the training sessions.
The implemented framework serves a twofold purpose under the scope of introducing operators either in new assembly routines or new assembly workstations. Therefore, two training modes are designed, namely:
- Training mode: offers a walk-through experience, where augmented items provide information about the available resources; and instructions about their respective usage or operation.
- Assistive mode: gives a hands-on experience through a step-by-step augmented simulation of the assembly process. The former mode is supported by learning based on object detection in order to highlight the objects of interest with augmented notifications on each assembly step.
Augmented reality (AR) has proven to be a powerful tool in industrial automation. The introduction of an extra dimension to the operator’s point of view promises greater levels of productivity; by exposing information from the system’s digital twin to the physical workspace.
Practically, AR is the enrichment of visual senses with overlaid data, as of instructions, guiding images, notifications, and models on top of the physical world.
Sharework use cases present an opportunity to implement and evaluate the developed framework in collaborative assembly scenarios where state-of-the-art multi-modal interfaces (between the robot and the human) are applied. The modes of the training application introduce users to the workstation functionalities and interaction mechanisms, as well as the assembly routines.
Early experimentation with users indicates a great improvement in training efficiency through steeper learning curves at reduced cognitive loads. This is also reflected during system’s runtime, where novice operators can keep up with the “robot colleagues” and avoid assembly errors.
For more information about the proposed methodology and system, refer to “G. Apostolopoulos, D. Andronas, N. Fourtakas, S. Makris, Operator training framework for hybrid environments: An Augmented Reality module using machine learning object recognition, Procedia CIRP, Volume 106, 2022, pg. 102-107.”
The Robotics, Automation and Virtual Reality in Manufacturing research group from the Laboratory for Manufacturing Systems & Automation of the University of Patras, Greece, is involved in the development, software implementation and use of advanced tools based on AI methods for industrial robotics applications. The group will transfer its experience for the development of human-robot interfaces for an effective collaboration
Dionisis Andronas, Research Engineer, LMS, University of Patras.
Dionisis Andronas holds a Bachelor’s and Master’s in Mechanical Engineering and Aeronautics from the university of Patras (Greece) in 2018. He works as a research engineer at the “Robots, Automation and Virtual Reality in Manufacturing” group of Laboratory for Manufacturing Systems and Automation (LMS). His research topics involve the design and development of hybrid production systems and cognitive mechatronic devices for reconfigurable manufacturing systems. His involvement in FP7 and H2020 projects concerns the designing of collaborative workstations, human system interfaces, model-based deformable object co-manipulation planners, in addition to innovative systems for material handling and assembly.