1. The general problem solved by Aether 

Alteia enables and automates the visual and digital observation of physical systems in order to measure their resilience to change, assess their productivity, or plan their maintenance efficiently.

Systems of interest can be industrial assets (pipelines, electricity grid, mines, roads, or bridges), or any other high-value object that needs to be monitored over time like a production line or a forest.

The Alteia software allows the automated aggregation and contextualization of visual data on these systems over time, and generates actionable insights thanks to the monitoring of the system representation.

As you get started in the Alteia software, it's helpful to think about data in the software living in two places: the data layer and the object layer.

2. The data layer

In the data layer, data is stored inside datasets. Datasets usually come from visual sensors: satellites, drones, or lidar sensors.

Every dataset in the Alteia software maintains a record of how it was produced so that the origins of data are always preserved and accessible. This concept is known as data lineage.

• The Alteia software tracks which input datasets were used to produce which output datasets. This allows you always to know where a piece of data came from, and understand how data is used.
• The software tracks the logic that was applied to produce each output dataset. For example, an input dataset might be filtered to produce a smaller output dataset; that filtering logic is preserved and visible in the software. There are many ways to write logic in the platform, ranging from code calculation tools to custom tools.

All datasets related to the same local area are regrouped under a project.

Projects serve as permission boundaries for grouping and organizing work around a specific area. We'll cover the details of how to access and use Projects in one of the upcoming sections.

3. The system layer

3.1 What is a "system representation"?

A system representation is defined by a series of attributes that quantify the properties to be monitored. Altogether they describe the system at a specific point in time and how it is performing. In fact, it is a purpose-oriented description. For systems whose productivity needs to be monitored, the system representation will describe the various means of production with the appropriate level of detail. For other types of systems where the focus area is risk exposure or lifecycle management, the state representation will change accordingly.

Depending on the problem being addressed, a state with different attributes can be defined. 

As an exemple, a pipeline can be defined by structural attributes or thermodynamic attributes depending on the type of condition that is measured and analyzed. 

The system representation is also not static over time. In order to understand the system performance and its evolution, it is important to calculate its state at different timeframes, with a chosen frequency, for example each minute, each month, each year, etc...

The general objective of Alteia is to allow the creation of any system representation, and to evaluate it at different moments in time thanks to heterogeneous data collected along the physical system.

The history of the system (previous states) can be extrapolated to make predictions about future system states.

Artificial intelligence is used throughout the process to make sense of the heterogeneous data that are collected, and analyze it to update and predict the state of the system.

3.2 How is such analysis made possible? 

There are two ways to evaluate the state of a system. 

• Propagation: it is a set of rules (based on physics or statistics for example) that allow - from a given state - the calculation of a future state. Example: dune movement, vegetation growth modeling, etc. 
• Update: An update is the action of correcting the system state representation at a given time, based on a measure that is made at that time. Updates are made each time a new measure is performed thanks to a sensor.  Different types of measurements can be used to update part of all attributes of the state representation.