Our founders invented Cyber-Physical AI
Research is the foundation of Exalens. By unifying innovative data science and engineering, we are bridging the cyber-physical monitoring gap and democratising digital transformation for everyone with the click of a button. It's what sets us apart and makes our technology revolutionary.
Our Background
Exalens are thought leaders in cyber-physical systems monitoring and security.
Over the last 10 years, our people have developed state-of-the-art technologies and delivered innovative solutions for emerging and digitally transforming industries world-wide.
UK/EU R&D projects
We have delivered ground-breaking innovation for industries undergoing significant digital transformation such as manufacturing, energy, maritime, transport, smart cities, healthcare, and more.
PhDs and Masters
Our team includes experts with PhDs and Masters in disciplines spanning Data Science, AI, Mathematics, Engineering and Cybersecurity.
Published Research Articles
The people of Exalens have published over 300 research articles on advances in AI for Cyber-Physical Anomaly Detection, Digital Twins, Cyber Security, Industrial and Robotic Security, and other areas.
Our Research Partners comprise of some of Industry’s and Academia’s foremost technology pioneers and innovators
Research Summary
Self-configuring classification of discrete process variables states
Accurately classifying and tracking discrete time-series process states from one or more process variables (e.g., input/output sensors, actuators, etc) is a key data modelling challenge that addresses many critical use cases in Industrial and Manufacturing monitoring operations. Common use cases include identifying causal relationships between process variables to understand dependencies in an automated production line, or detecting when vibration sensors exhibit different behavioural patterns that indicate an early warning of component failure. The challenge can be formulated as a multidimensional time-series clustering task for univariate or multivariate time-series data, accounting for arbitrary sampling volume and rates, process states that emerge as complex time-dependent shapes in continuous process variables (e.g., sensor value as a univariate dataset) or mixed continuous and stateful process variables (e.g., sensor and actuator/program counter as a multivariate dataset), and state classification latency. By employing a novel temporal multidimensional feature matrix, a composite of time-orientated shapes may be provided as input to fit an unsupervised "state learner" model operating in Euclidean space. Combined with an optimiser for automatic determination of classification window and robust state identification, the unsupervised state learner model may be used to train a supervised time series classification model on univariate or multivariate process variables with highly scalable and accurate state classification in sub-millisecond latency.
Impact-Aware Response
Autonomous incident response workflows that continuously assess and connect cyber and physical incidents while measuring and tracking risk to provide impact-aware recommendations for response Exalens combines multiple AI approaches to autonomously link and classify correlated cyber, physical, and cyber-physical incidents.
Predictive Condition Monitoring
Real-time, adaptive monitoring for physical machine and process anomalies that require no human intervention or understanding of underlying data Securely reads and upcycles physical machine data from existing OT systems, IoT platforms, devices, and sensors. An anomaly detection engine that self-configures and calibrates its behavioural analysis to best fit any physical machine and process, regardless of data type, frequency, or context.
Sequential Analysis
A detection approach that intelligently replicates human analysis strategies and processes. Combines heuristics, and statistical analysis with state-of-the-art AI methods such as deep learning, and unsupervised and supervised ML, to detect cyber and physical system threats with high efficacy.Provides explainable detection indicators in natural language that supports easy interpretation and actionability for both IT and OT teams.
Dynamic Structural and Lexical embeddings for AGD Detection
Applying Deep Natural Language learning to classify domains names similar to those used malware command and control channels
Recursive multipass semantic analysis for device similarity and change detection
Identify similar devices and determine when there are sudden changes in behaviour between them
SDN-based Resilient Smart Grid Architecture
An architecture for dynamic risk assessment, intrusion detection / correlation, and self-healing in Smart Grids