CARV2021
Deep Learning as a mean for Enabling Self-Learning and Self-Optimizing Capabilities in Real-World Industrial Applications
Recent progress in Deep Learning e.g. Deep Reinforcement Learning and Computer Vision can enable self-learning and self-optimizing capabilities in robotic and manufacturing systems. Nowadays in the context of Industry 4.0, manufacturing companies are faced by increasing global competition and challenges, which require them to become more flexible and able to adapt faster to rapid market changes. Advanced robotic and manufacturing systems are enablers for achieving greater flexibility and adaptability, however, integrating such systems also becomes increasingly more complex. Thus, new methods for programming and optimizing the systems to accommodate the natural variation and complexity exhibited in real-world tasks are needed. Deep Learning may provide the means to enable these “self-x” capabilities.
Topics of interest include, but are not limited to:
- Robot programming through Imitation Learning or Inverse Reinforcement Learning
- Industrial process optimization with Deep Learning approaches
- Robot control optimization through Deep Reinforcement Learning
- Object and defect detection with Deep Learning
- Smart autonomous vehicles in industry (air, land, sea)
- Simulation modelling for Deep Learning training and validation
- Real-world case studies
Session chair
Associate Prof. Simon Bøgh (sb@mp.aau.dk)
Department of Materials and Production
Aalborg University
Transforming Traditional Production Systems into Smart Production Systems
The exponential growth in digital technologies provides manufacturing companies with an increasing number of new possibilities for the development of new products, processes and services – many of these with the potential of being disruptive. The expected benefits related to the implementation of these new digital technologies range from the improvement of operational efficiency to the enabling of new value creation possibilities, catalyzed by new product capabilities as well as by entirely new business models. However, to operationalize the transformation of a production system into an integrated, digital and smart production system remains a challenging task.
Topics of interest include, but are not limited to:
- How can companies benefit from novel digital technologies?
- How can companies integrate the novel digital technologies into existing (brown field) and new (green field) production environments?
- How does the digital transformation impact and change businesses and organizations
- How is the digital transformation process managed over time and how should staff members be further qualified?
Session chair
Associate Prof. Yang Cheng (cy@mp.aau.dk)
Department of Materials and Production
Aalborg University
