| Περίληψη: | This dissertation aims to create dynamically reconfigurable production systems that can successfully meet the changing demand of consumers for personalized products. To achieve this goal, the dissertation proposes the introduction of autonomous Cooperating Robots that work independently but also in collaboration with human operators. There robots are autonomous, stationary, and mobile dual arm robots that are able to perceive their environment and through reasoning, cooperate with each other and with other production resources including human operators. The dynamically reconfigurable behavior of these robotic systems is achieved through an "intelligent" control system that integrates the decision-making process at three different levels.
This “intelligent” control system has been implemented through the deployment of a multi-level reconfiguration architecture that integrates: a) The Digital Twin of the factory, b) multi-level decision making tools and c) tools for controlling and dynamically re-organizing the execution of the manufacturing processes.
The factory Digital Twin employs "smart" digital models that contain geometrical and semantic information about the environment, including human operators, robots, sensors, products, and processes. The Digital Twin is updated in real time using information from multiple sensors mounted on the robots and the surrounding area.
The first level of decision-making concerns the optimal task allocation to the available human operators and Cooperating Robots at the production line level. The method is based on a multi – criteria decision making framework, allowing the automatic generation of task plans whenever a change is needed in the process or in case of unexpected failures. The different task plan alternatives are evaluated against user defined criteria based on the end user’s requirements. The method is integrated in a 3D physics-based simulation engine that is used to capture the required measurements for the criteria calculation (human ergonomics, cycle time, resources’ utilization etc.).
The second level concerns the optimization of the autonomous execution of manufacturing tasks, by integrating tools that allow the Cooperating Robots to dynamically perceive: a) the production environment, b) the parameters of the manufacturing processes and c) the behavior of the human operators involved in the assembly. These tools receive information from the Digital Twin to detect static and moving obstacles as well as the real time location of the parts and the human operators involved in the production process.
The third level concerns decision-making methods for the adaptive behavior and control of the robotic system. More specifically, based on the real time information extracted from the Digital Twin, decisions are made about: a) the mobile robots’ path planning for autonomous navigation and localization in the factory, b) the robot arms’ motion planning for collision free trajectories generation, c ) the safe and easy coexistence and cooperation of the Cooperating Robots with human operators.
The above system was demonstrated and validated in three cases of production systems of the automotive industry. The performance of the system shows that it may achieve an increase resources’ utilization, improve human operator ergonomy and enable faster adaptability to changes against current industrial practice.
|
|---|
