ERTIS (Embedded Real-Time Systems) focuses its research activity on improving the management, accessibility and integration of embedded devices in the context of the Internet of Things (IoT). ERTIS also investigates the integration of Edge, Fog and Cloud paradigms in order to optimize response times, fault tolerance and security in the IoT. They also work on the use of deep and distributed neural networks in this context. The fruit of his research has been successfully applied in the monitoring of Critical Infrastructures and Smart Grids.
Main Research areas
Open Digital Twins Platforms
Although digital twins have recently emerged as a clear alternative for reliable asset representations, most of the solutions and tools available for the development of digital twins are tailored to specific environments. Furthermore, achieving reliable digital twins often requires the orchestration of technologies and paradigms such as machine learning, the Internet of Things, and 3D visualisation, which are rarely seamlessly aligned. In this research line, we aim at developing a generic framework for the development of effective digital twins combining some of the aforementioned areas. In this open framework, digital twins can be easily developed and orchestrated with 3D-connected visualisations, IoT data streams, and real-time machine learning predictions.
Streaming Deep and Distributed Neural Networks over the IoT/Edge/Fog/Cloud
Deep neural networks have been widely used in applications such as image and video recognition and classification and anomaly detection, generally designed to be used in large processing systems such as Cloud solutions. Currently, there is a trend to implement and distribute these networks in the computing that goes from the IoT to the Cloud in order to facilitate the generation of critical actions, the reduction of bandwidth, and the improvement of their precision, among other aspects. ERTIS investigates how to integrate these networks with message distribution systems to enable their distribution in Edge, Fog and Cloud Computing frameworks, and manage their deployment on these paradigms. Our clear example is our framework Kafka-ML which manages the whole pipeline of ML/AI models through data streams.
Middleware and Platforms for the Development and Deployment of Applications over the IoT/Edge/Fog/Cloud
Edge and Fog Computing are paradigms that have a place between Cloud Computing and the IoT and aim to move the computing capacity as close as possible to where the information is produced in order to reduce response time and bandwidth and increase security in critical and distributed applications. This paradigm has special application in environments that generate a large amount of information and whose latency requirements and/or existing network limitations do not allow this information to be processed in cloud environments. In this research line, ERTIS focuses on the development of middleware and platforms that allow managing and deploying the life cycle of these applications to meet the established requirements and their perfect adaptation to the IoT, in addition to adapting current solutions to this type of environments to optimize their computing cycle.
Critical Systems Applications and expert systems (infrastructure monitoring, smart grids, Industry 4.0)
The application of our research work is one of the fundamental pillars of the ERTIS research group. This research has been funded and applied through regional, national and European projects in the following areas:
- Data analysis and expert systems through machine and deep learning. Development of virtual analysers for complex processes in Industry 4.0
- Monitoring of structural health of civil infrastructures. The continuous monitoring and protection of these infrastructures is a priority for multiple government institutions and for the safety of the population.
- Smart grids. Improving flexibility and endurance in electrical distribution networks through energy balancing and the IoT are essential to accommodate new energy demand factors such as electric cars.