Empowering Cobots with Energy Models: Real Augmented Digital Twin Cobot with Accurate Energy Consumption Model

The concept of a Digital Twin has proved its worth over the past two decades, establishing itself as a cornerstone of contemporary industry. Augmented Reality, an emerging technology, enhances the interaction between humans and machines, including computers and robots. Today, numerous examples exist of the union of these two technologies to create real-augmented digital-twin models of collaborative robots. However, these models often lack data on motor currents and power consumption. In this study, we propose a real-augmented digital-twin model that accurately estimates energy consumption. This additional energy information equips the tool for various applications such as robot optimization, commissioning, and troubleshooting. We employ our real-augmented digital-twin model to test methods for reducing Cobots' energy consumption, using the tool to demonstrate and train Cobot practitioners on these techniques' applications. The model is also useful for anomaly detection (troubleshooting) when the robot's consumption statistically deviates from the ideal model. Moreover, the model can anticipate the robot's power consumption during the commissioning phase, prior to its installation. Through a series of experiments and a practical demonstration at a robot fair for practitioners, we illustrate the benefits and training capabilities of our approach.