For welding of conventional structural steels semiautomatic technologies (MIG, MAG) are widely used, whereas for welding of Al-alloys and stainless steels the TIG welding method is the most common. The current study concentrates on welding of thin sheet metal products from stainless steel and aluminium by using a novel cold metal transfer (CMT) process. The CMT technology is an alternative to TIG, providing advantages, such as reduction of distortions and increased productivity. This is mainly due to low heat input, achieved by controlled movement of the electrode. In order to realize these advantages, optimization of the CMT welding process is essential. The aim of this study was the optimization of the process using the existing welding equipment (robot, manipulator, etc) and validation ofthe CMT process. The limiting factors for the increase of the productivity are the reduction of quality (increase of porosity, distortions and inacceptable shape of the welding bed). As a result, practical recommendations are given for the implementation ofthe CMT technology for robotic welding.

Collaborative work between human and industrial robot in different areas is developing rapidly. High payload industrial robots that can harm humans execute their tasks in separated protected areas and are regulated by safety standards. The main aim of this study was to design experiments, analyze, and model the collaborative work and safety monitoring systems by using small industrial robots with inbuilt power and force limiting. Different methods of collaboration are standardized by the ISO/TS 15066:2016. Human co-work with robots gives us a variety of options that can be used in manufacturing and other related areas. The advanced multi-level monitoring system of co-work processes can reach the highest level of the human safety with minimum robot downtime. Several methods were analyzed to find the optimal solution for the online monitoring processes. As a result, a safety multi-level monitoring system was designed, tested in simulation software with real-room conditions with an industrial robot. Safety multi-level system, as a part of the control and monitoring online system consist of different types of sensors and a microcontroller that can be attached to the Cyber Physical Production System (CPPS). Those control and monitoring methods and processes were implemented in experimental setup in Robot Operating System (ROS).

Digital Twin (DT) concept nowadays is shown via the simulations of the manufacturing systems and included those production processes and parametric 3D models of the product. It is the primary method for planning, analysing and optimising the factory layout and processes. Moreover, work on management via the simulation in real-time is already done using Virtual Reality (VR) tools from a safe and remote environment. However, there is a list of limitation of such kind of digital systems, as connectivity speed and precision of the digital environment. The primary goal of this study is to access second listed limitation and on the example of the fully synchronised physical with its digital replica industrial robot, increase the level of precision of the developed DT environment. The proposed approach introduces transfer of the mathematical model to the virtual environment, thus creating a precise and scaled visual model of the Industrial Robot.

Modern Industrial Robot (IR) programming process is mainly performed by using three different methods — manual, offline, and online programming. Each of these methods has various advantages and disadvantages. Prominent automotive industries often use a combination of them, as there is no way to avoid one or another form of programming on one factory. However, the use of a combination of different programming methods is time-consuming and demands the operator’s presence on site for reconfiguration of the IR. The primary goal of this study is to introduce and test the concept of a hybrid IR programming method, which combines both: offline and online robotic cell design, programming, and re-configuration methods. Testing of this method is based on fully synchronized robotic cell’s Digital Twin (DT), developed in Industrial Virtual and Augmented Laboratory of Tallinn University of Technology. Usage of the virtual replica allows to plan and program robotic cell on the means of telepresence and interfere with the predefined path of the robot by online programming method. Moreover, this approach reduces the time for robotic cell design and re-programming, enables to minimize downtime of the robotic cell on the factory shop floor. Included Virtual Reality (VR) environment allows simulating a full-scale operator presence on site. Thus, the proposed approach supports an immersive and safe environment for the IR and similar equipment programming purposes.

During recent years the automation of production processes in small and medium enterprises (SME-s) has been a subject of growing interest. The economy of scale and increased volume of production can be achieved by selecting the right strategy for the automation. The automation systems are as a rule complex and their implementation is resource consuming for SME-s. In the present paper we study implementation of robot welding cells in several enterprises. It is shown that introducing robot welding cells in SME-s is a difficult task because of the limited resources and lack of the needed competence in SME-s. For successful realization of automation projects the complex systems must be divided into smaller and simpler parts using modular approach. The success of the project can be achieved through suitable definition of the modules. This makes it possible to implement the project steps in parallel by involving the needed resources.