Narvina
AI-based telemetry monitoring assistant that learns spacecraft systems, detects anomalies, forecasts trends and explains root causes.
Technical specifications
- Core capabilities
- Anomaly detection, pattern matching, data correlation, trend monitoring, forecasting
- Interface
- Intuitive UI with REST API for integration into existing infrastructure
- Learning model
- Continuously learns from operator feedback on flagged events
- Analysis type
- Multi-parameter combined/relational analysis, not isolated single-parameter thresholds
- Flight heritage
- Built on 20+ years of Solenix telemetry analysis experience, incorporating lessons from the MUST system
- Availability
- Early-Adopter Partner Program with free/low-cost trials and post-trial licence discounts
About
Narvina is a suite of AI applications that simplifies telemetry data analysis by automating a wide range of critical tasks for spacecraft operators. Critical anomalies often don’t appear where they are expected: early signs may be hidden in parameters that are not typically in focus, and slow, consistent changes in sensor readings can start months or even years before a major failure occurs. Narvina acts as a quiet, autonomous observer monitoring all parameters and tracking long-term trends, raising early flags that prompt timely investigation.
With Narvina, operators can detect potential anomalies early, identify similar behaviours, match patterns, monitor trends, and predict future telemetry evolution. These features work together to enhance preventive maintenance, support anomaly investigations, and keep operations running smoothly, while significantly reducing operational costs by streamlining performance assessments.
Narvina does not just look at individual parameters — it understands system relationships. For example, in a simple setup with a switch (Parameter A) and a light bulb (Parameter B), each state might look normal in isolation, but if the switch is on and the light is off, something is wrong; Narvina catches this kind of mismatch through combined analysis across all system parameters, not just isolated values. As a machine-learning teammate, Narvina learns continuously: if it flags something that is confirmed as expected behaviour, it remembers and refines its model; if it is a real issue, this is reinforced too, so it gets better over time. It is designed for human-AI collaboration, freeing engineers and controllers from chasing blind patterns or being buried in plots, so humans can focus on decisions while Narvina ensures they are looking in the right place at the right time.
Narvina automates housekeeping telemetry analysis via a set of applications: anomaly detection, pattern matching, data correlation, and forecasting. It is designed with an intuitive user interface and REST APIs, making it easy to integrate into existing infrastructure, either used directly or as part of a broader system. Narvina results from Solenix’s 20+ years of experience in telemetry analysis for spacecraft operations, building on the heritage of the MUST system, and is offered through an Early-Adopter Partner Program including free/low-cost trials, discounts on post-trial licences, roadmap planning input, customer feedback surveys and Q&A sessions.
Documentation
No public datasheet yet — request the datasheet / ICD from the supplier.