Project ideas - Digitized products & working vehicles
Join our project pitches on Digitized products & working vehicles
09h30: Intelligent Model Reuse (iMore_IRVA)
This project aims to manage the growing number of complex models in various engineering fields by enabling model re-use and decreasing the cost of creating/maintaining them, which can improve the efficiency of the design process. The project proposes a strategy to semantically search required models in the repository of available models, clone/pattern detection in the product models to find product families, and classification of models by learning available models and detecting the type/label of new models to help with maintenance and the re-use of models. Overall, this project aims to increase development productivity and support long-term model management.
10h00: Continuous deployment of cyber-physical software (ContinueD_SBO)
The manufacturing of high-tech cyber-physical systems is becoming increasingly reliant on software due to the IoT connectivity. The adoption of the DevOps development process, used by IT companies, is a promising solution for these manufacturers, but it is also challenging. This project aims to establish a basis for a DevOps development process for cyber-physical systems, with goals such as a modular and flexible software architecture, a digital twin of the cyber-physical system for continuous testing and deployment, and an overall DevOps strategy and orchestrator component for automation and decision making.
10h30: Reconstructing the Distributed Thermal Field using PINNS (DTF-Pinn_SBO)
The DTF-PINN project aims to develop real-time virtual sensing techniques for monitoring and controlling industrial machines and manufacturing processes. These techniques will help address the challenge of accurately reconstructing temperatures and transients at critical points of interest, using a Physics-Inspired Neural Network approximator. By combining distributed simulations and localized measurements, this project will provide a more accurate solution for understanding the complex 3D thermal phenomena in many industrial machines and components.
11h00: Transformer-based Interpretable Deep Forecasting with Uncertainty in Mechatronics (INTERTWINE_SBO)
INTERTWINE aims to improve forecasting accuracy in mechatronics by leveraging digital twins and (hybrid) machine learning. The project aims to address current forecasting system limitations that lack the ability to handle different types of data and provide explanations for predictions. By incorporating transformer models and innovative techniques such as hybrid transformers and quantile regression, INTERTWINE aims to provide insights into system dynamics and enhance the performance and efficiency of industrial products.
11h30: SITuational Aware NAVigation and semantic mapping (SITANAV_SBO)
The SITANAV project aims to create a framework for situational awareness, semantic mapping, and context-aware navigation, enabling automated systems to adapt to changes in unstructured environments, such as warehouses and manufacturing facilities. Currently, such systems require re-programming and re-mapping for every change in the environment, making them unsuitable for such settings. The SITANAV framework aims to address this challenge by developing a system that can automatically update the semantic map and apply desired behaviors for certain situations, allowing for autonomous navigation and dispatching of objects.