Adaptive Real-time Anomaly Detection & Identification in Sensored Systems (ARADISS)

ABSTRACTION | DETECTION | IDENTIFICATION

What is ARADISS?

ARADISS is an innovative fault detection/anomaly detection method for cyber-physical systems. It leverages AI/machine learning models to correlate component behavior with battery voltage/current consumption. By creating 1-to-1 mappings, the models can pinpoint faults or anomalies in the system by detecting discrepancies between the anticipated battery consumption based on the components’ activity and the real-time battery usage.

How does it work?

1. Data Extraction

ARADISS taps into real-time battery voltage and current data, extracting features that form the foundation of its learning models.

2. Mapping

These models, often referred to as “norm maps,” establish independent relationships between battery metrics and each system’s operational variables.

3. Anomaly Detection

When there’s a deviation from the established norm, it signifies an anomaly. ARADISS quantifies these deviations using five critical parameters from its unsupervised anomaly detection algorithm.

4. Self-Adapting Mechanisms

What sets ARADISS apart is its ability to self-adapt to gradual system changes. This ensures highly accurate and sensitive diagnostics over time.

5. Cost-Effective Fault Management

Not only does ARADISS provide a tailored fault management solution, but it also facilitates the quick implementation of mitigation strategies. This combination makes it a cost-effective choice for various platforms.

Why is this model important?

A growing variety of crewed and uncrewed aerial vehicles utilize batteries as their primary power source. These vehicles comprise various interacting components and sensors for safe operation and their respective missions.

As their interactions, complexity, and numbers increase, the risk for anomalies such as degradation of components, sensor faults, and erroneous controls increases. These anomalies pose significant risks for vehicles flying over densely populated areas or conducting critical missions. It is, therefore, crucial to detect and mitigate these anomalies.

Cheaper and faster to implement than traditional fault management while being transparent

On-board or off-board (e.g. ground control, telemetry-based) deployments

Provides comprehensive and robust fault coverage

Can be rapidly and automatically adapted to new platforms

Low computational requirements

Improves autonomy

Can act as a “second line of defense” on top of traditional fault detection for highly risk-averse missions

Information and data assurance

Root cause identification

Sensor and component faults

Software bugs and cyber threats

Environmental disturbance

Space systems, including satellites and robotic systems

Unmanned aerial, ground, and underwater vehicles

Other cyber-physical systems, such as automobiles

PROJECTS
A Reusable Fault Management Tool for Satellite Health Monitoring for NASA

All software solutions.

Advanced Sensor Signature Analysis
Early Warning Fault Detection & Mitigation Strategies
IoT (Fullstack IoT Development)
Modeling & Simulation
Business Intelligence & Advanced Data Analytics Services

Get in touch.

Get in touch with our team today to discuss how we can help solve your problems.

Stay in touch.

Stay up to date with the latest news, updates, and tech advances in the industry

Stay in touch