USE CASES

FLUIDOS has three main use cases: 

The introduction of small and medium sized non-programmable distributed energy sources (e.g., solar, wind) requires an ever-increasing ability to measure and control the most important grid parameters in real-time and across multiple domains. This, in turn, requires better ways to measure and control the power grid in real-time and across different areas. This means having to use a lot of information and communication technology (ICT) like measurement tools, processors, and big networks in the grid. All the data we gather helps us (1) keep a close watch on how the grid is doing in real-time and (2) make sure we can provide energy to users even if power demand and supply change quickly. 

FLUIDOS aims to extend current cloud-based intelligence with edge-based capabilities to:

1. Reduce the latency in the control loop decisions (e.g., for the grid state estimation at the distribution level);  
2. Enable local processing to avoid error escalation in case of communication breakdown with the remote ICT infrastructure (e.g., lack of connectivity)  
3. Scale the solution to support the introduction of thousands of low-cost local Phasor Measurements Units (PMUs) for the distribution grid. 

Terraview OS, a unified platform for viticulture, enables the grower to manage information from many sources, returning high-value info such as yield estimation, smart irrigation, and disease prediction and diagnosis. In its current form it is a centralised but cloud native service. From the perspective of FLUIDOS and a strategy of Terraview, the service is to be offered as an edge computing service in order to enable cost-effectiveness and increase data privacy and security.  

FLUIDOS aims to: 

1. Leverage its solutions to simplify the deployment and development process, making Terraview OS independent from the actual running location, i.e., in cloud or on edge systems, without requiring Terraview OS to support the (heterogeneous) hardware available on each site;
2. Use FLUIDOS to simplify the interactions between different components of the Terraview OS (data collection, AI algorithms and predictions, dashboard, etc.) that look and behave the same independently from their actual location;
3. Distribute AI training and model inferencing across customer site and cloud.

Mobile robots for industry 4.0, smart logistics and retails are, de-facto, resource-constrained and battery-powered computing devices that roam in a space shared with humans and other ground IoT devices. The productivity and effectiveness of such a system is directly proportional to the capacity of robots to carry out complex tasks, the duration of their battery  and the coordination among them. 

Managing a Robotic Fleet in a logistic environment can benefit from FLUIDOS project in several ways increasing its performance. The underlying idea of this use case is to increase productivity by extending the working time of the robots and giving them more computation resources without draining the battery. 

FLUIDOS aims to manage the computational capacity of a fleet of robots with the following goals: 

1. Increase computational efficiency by dynamically orchestrating workloads associated to logistics tasks, in coordination with the robotic fleet management system;
2. Enable the execution of heavy workloads while avoiding battery draining, by using the fleet as a unique, fluidly connected, device, leveraging the capacity of the swarm to serve the single and vice versa;
3. Significantly reduce costs (travel and personnel) associated with the deployment and maintenance of logistic robots.