Smarts start from the Edge, Machine Learning for Critical Infrastructure applications!
What is EDGE-AI?
EDGE-AI is the Critical Infrastructure Machine Learning (ML) technology R&D division of InProgress Research Inc. (IPR), focused on developing ML products & solutions optimized for Critical Infrastructure applications. With the Critical Infrastructure applications focus, EDGE-AI has the mission of driving innovation in Machine Learning and Artificial Intelligence for the Critical Infrastructure domain through introducing new ML methods, concepts and technologies best optimized for critical infrastructure environments, applications and limitations.
The integration of EDGE-AI machine learning products & solutions will be carried out through DESIAGO, IPR’s consulting & integration division. The solutions are intended to address existing domain challenges rule based programming can not currently tackle, and to complement our existing IIoT partners offerings.
EDGE-AI will work with IPR’s Industrial Internet of Things (IIoT) partners on integrating its new ML technologies into their products as well as supporting their ML product transformation through R&D collaboration. Our goal is to become the trusted go to partner, Critical Infrastructure vendors count on to assist with their ML transformation.
Our focus verticals are:
- Critical Infrastructure applications
- Industrial applications
- Smart Buildings
- Smart Cites
What products and solutions will EDGE-AI develop?
EDGE-AI will focus on developing products in the following categories:
- Predictive Maintenance using ML
- Physical Security and Access Control using ML
- Capacity Planning using ML
- Industrial Automation using ML
- Anomaly Detection using ML
- Network Security using ML
Why is Machine Learning for Critical Infrastructure different?
With the isolated nature of critical infrastructure networks, its very limited or no connectivity to the Internet, no Cloud access, no centralized global repository of data and not enough historical data to train ML models, comes the need to develop new unique Machine Learning approaches and techniques to better fit the environment and overcome its limitations. The use of current available ML technologies applicable for enterprise and commercial applications under the limitations described can yield ML models that fail to generalize properly with accuracy levels below what is acceptable. Due to the inherent issues caused by the isolated environment and limited availability of data, highly biased systems can develop with very low model generalization accuracy. The limitations can also make it not commercially feasible and/or be technically prohibitive to develop and train ML models.
EDGE-AI will focus on developing new non-orthodox approaches and techniques to address the limitations Machine Learning faces within Critical Infrastructure Applications, while benefiting from existing standard ML technologies developed. This will require changes in the principle approaches used to develop ML models, methods training, training duration as well as pushing some training and analytics to the edge. The Changes will call for next generation industrial grade edge processing hardware that we will coin as Edge Processing Unit (EPU) as well as Critical Infrastructure Machine Learning models that we will coin as Critical Infrastructure-ML (CI-ML) Models.
Why is Machine Learning Vital for Critical Infrastructure Applications?
Critical Infrastructure applications are generally slow to adapt new technologies, reliability is always in focus and availability is a key consideration. If you have been around long enough, you have many times heard Protection & Control engineers say things like, “if it works and proven reliable don’t change it” , “let the technology mature enough for us to consider it”, and “Keep it simple, simple works and we need this”. All are very rightfully said, and justified by the extremely high reliability and predictability requirements. The lights need to stay on and that is priority one!
When adapting new technologies for Critical Infrastructure applications the math is simple. For a new technology to be quickly adapted, the technology has to prove a natural fit, be mature, reliable, predictable in behavior and mostly does not interfere with the critical aspects of Protection & Control. In short, the benefits have to outweigh the risks while respecting Operational Technology (OT) priorities. Now lets take a closer look at Machine Learning (ML) and see how it fits the bill.
Without getting into deep technical details, machine learning is about looking at past behaviors, and learning enough to be able to protect future events. taking action based on those predictions is another level and is a step into Artificial Intelligence (AI). AI takes things up a few notches, it requires the system to be more aware of its surroundings and be able to make decisions based on new predictions, presumably in a similar fashion to what a human would do. With this being said, right out of the gate, AI would take much longer time to be considered for Critical Infrastructure networks, just because of the fact that it can action things on its own and this can result in protection, availability and/or safety hazards. That is not to say never, but it will take a long time for adaption to take place. On the other hand ML is much simpler, ML only uses past state to predict future state and flags anomalies for a human to look into taking action, while only if desired, a predefined safe course of action can be automatically triggered. The beauty is, it can do all of this without the need for cloud or internet connection, if designed and implemented right.
As things stand today, Critical Infrastructure applications are faced with challenges conventional technologies aren’t able to overcome opening the door for ML to become a core mainstream technology for future Critical Infrastructure deployments. One of the biggest challenges is that critical infrastructure networks are kept in islands, nearly isolated from the outside world with very limited or no connection between the OT world and the internet. If a connection exists, it is secured and data flow is highly restricted to the bare minimum. This presents a challenge for some existing applications to function properly and securely. Any application that requires frequent updates or internet or cloud access, under the circumstances, will be deemed of very limited or no value, and here comes the need for ML based application to close the gap. Examples of those application are network security applications utilizing cloud analytics or that are signature based requiring frequent updates. Same goes for Predictive Maintenance, Anomaly Detection and Physical Security applications that require cloud access.
A properly designed Critical Infrastructure Machine Learning (CI-ML) application would facilitate an offline capability to perform Network Security, Physical Security , Anomaly Detection, and Predictive Maintenance tasks with high accuracy, while isolated, something that wouldn’t be possible to achieve with current traditional rule-based and/or cloud based programs.
Email us at: