Meet Turium Edge AI
The world is seeing explosive growth in the amount of data being collected from IoT sensors at the edge — with billions of new sensors and zettabytes of additional data coming online in the near future. Many of the valuable decisions— whether in combat or in manufacturing — are time-dependent.
An Inside Look
Designed for situations where time, availability and efficiency matter, operating in disconnected, low-bandwidth conditions — including on drones, aircraft, ships, robots, buildings, and satellites.
Integrates data of any type or complexity
— Turium Edge Ai ingests high-volume sensor and streaming data, including video, images, radar, lidar, acoustic, GPS, radio data, and more — The solution handles a variety of sensor input formats, such as RTSP/RTP, NITF, GeoTIFF, and MPEG-TS. Finally, it supports outputs in open standard formats — such as Parquet, CoT, MISB 0601/0903 KLV, MPEG-TS, and GeoJSON — which enables data and insights to be sent downstream to other subsystems with little to no integration work required. The result is a virtuous cycle, offering continuous integration / continuous delivery (CI/CD) of models all the way to the edge.
Extremely lightweight and power-efficient to deploy
— Turium Edge AI compresses and minimises the data that needs to be stored and transmitted — enabling low-latency, real-time decision-making at even the device level. In more constrained situations, it can be configured to transmit a “metadata-only” stream (e.g., any detection data, along with positional lat/long/elevation data), which uses less available bandwidth. Designed for the DIL situation Critically, users can hot swap models in real time without breaking the flow of sensor data through the system. This also means that if a model crashes, it does not impact downstream users who rely on that sensor output.
Provides a configuration interface that allows all users
— Turium Edge AI provides a configuration interface that allows all users — including non-technical ones — to configure highly-complex pipelines on real-time sensors. The interface assists with a wide variety of use cases, including video stabilisation, object detection, geo-registration, and movement detection and prediction.
Strong data governance practices
— Turium Edge AI provides strong data governance practices that are necessary to model performance and reliability. Turium also captures a detailed history of integration, including time of ingestion, source, and revision history. This metadata is used to both track data and model provenance as well as manage compliance with data auditing and retention policies.
Get Started in 4 Steps
See Impact in Days
Connect to your edge systems and start extracting data immediately.
Explore your enterprise data and import only what is relevant to your problem space.
Automatically build your pipeline and generate business objects at scale.
Turn your data into immediate impact with out-of-the-box use-cases.
Built For Security
Cybercrime has become a well-oiled and financed global industry, fueled by the value of the data that is being generated. Cybercriminals can employ the increased threat surface that comes from the billions of connected users on edge devices. The intrinsic fragmentation and inherent complexity of these infrastructures can open many doors. Turium Layered security solutions enable a seamless flow of data: it is an end-to-end solution that looks at the entire system holistically and addresses security at each potential point of failure on edge.
Different protocols at different levels
Layered above TCP/IP, it employs communication protocols that address the specific needs of edge devices that are unattended, geographically dispersed, and often mobile. Turium supports different protocols, but we advocate the use of MQTT (Message Queue Telemetry Transport), which is not only a lightweight protocol optimised for M2M device communications, but also a heavyweight technology that enables IM-type messages to be used and files exchanged. MQTT provides secure messaging between Everyware Software Framework (ESF), which is embedded in the gateway. All MQTT traffic is encrypted over an SSL/TLS connection. Device status and health monitoring, which is layered on top of MQTT, is an important additional Turium security feature to allow edge devices to be monitored at lower levels.
Device and server authentication
To ensure that edge devices, systems and applications can trust their respective counterparts it is necessary to identify and authenticate the connected devices in the field as well as on the server side. Secure authentication can be achieved in many ways. But when looking for a standards-based, proven and solid technology approach, Turium solutions leverage 10 digital certificates to provide the highest level of security. One of the important standards is X.509, a cryptography standard for a public key infrastructure (PKI). X.509 certificate-based authentication is an effective way of identifying individual devices. They can also be used to sign application code that is deployed and executed in the IIoT devices and gateways in the field. This provides integrity, authenticity and nonrepudiation of the data’s origin.
Secure transmission of all data via encryption over an SSL/TLS connection is essential. All Console and REST API access is only available over an encrypted HTTPS connection. All databases are protected from external access through strict firewall rules. And data is segregated by account. Confidentiality and integrity is ensured through a role-based access control model and access control lists that follow the Principle of Least Privilege and are enforced through all the layers of the architecture. A strict segregation of tenants down to a data level is another important element ensuring that other parties cannot access data and infrastructure. Turium lets you define granular access control policies at the data integration stage, then propagate those policies intelligently across the system. This enables models to be promoted to collaboration confidently with granular data security and transparent data governance.
Independent, certified security firms perform remote vulnerability assessments, including network/host and applications. Vulnerability scanning is conducted regularly and after any major changes to the infrastructure and environment. From the most remote device in the field to the backend business system, every part of the Turium Edge AI solution remains secure.
Turium Edge AI solutions are compliant and vigilant: when new threats emerge, so does our response to swing the pendulum away from the attackers and towards the Turium users.
Turium Edge AI for Assured Positioning,
Navigation and Timing (APNT)
Turium Edge AI ensures accurate, reliable and precise position and time measurements for successful military operations for land, air, sea, space and cyberspace.
Our technology provides proven Assured Positioning, Navigation and Timing (APNT) in delayed/ disconnected, intermittent, and limited (D/DIL) conditions. OEM components and commercial off-the-shelf items are engineered to ensure continuous positioning providing Situational Awareness by detecting, locating, and characterising interference sources, allowing operations within contested environments.
At the heart, Turium Edge AI is a software "sensor fusion and knowledge engine" that pulls all the available data together, using artificial intelligence, to determine how accurate and trustworthy the different types of data are, and what to infer through the graphical output to get the most precise and reliable decision possible. Results are updated continuously throughout the mission.
Enable spoofing detection for a comprehensive understanding of your RF environment to identify when your position, navigation and timing (PNT) are at risk. Edge AI Interference Toolkit identifies and characterises interference frequencies in your area, protecting you from unintentional and malicious interference.
Time-tagged snapshots of analog to digital data samples allow users to characterise the RF environment and develop their own interference location algorithms.
Knowledge graphs to build relationships and find anomalies in the ruggedised environment
Working with information gleaned from cameras, compasses, and other sensors readily available on the vehicles, Turium Edge AI weaves the data together into an accurate navigation solution critical to mission safety and success — even when a GPS signal isn’t available. Turium Edge AI finds ways to wring accurate locations from the information on and around vehicles.
Street-level matching within Turium Edge AI relies on third-party databases of 3D global maps that can be used to render landscapes as they would look from the ground, rather than being limited to the birds-eye view of most map databases. It is looking for a match between the view from the vehicle’s camera and the database’s 3D view of nearby terrain. The images in the 3D databases are geotagged within Turium Edge AI with longitude and latitude, getting a match provides the vehicle with an absolute location fix on the map, which is key to minimising the drift that otherwise accumulates from inertial sensors.
When a vehicle is in an area where 3D database images aren’t available, Turium Edge AI uses Comparative Localization and Mapping. With this, the system examines the relative speeds at which different objects and features captured by the camera, such as trees, hills, or buildings, move across the image frame. That motion information can be enlisted by the software to calculate the relative locations of surrounding features. Tracking those relative locations over time can help determine how far the vehicle has moved since the last time an absolute location fix was found. Contributions from multiple vehicles can be combined to assemble a larger, more detailed, more accurate knowledge graph that all the vehicles can share.
Turium Edge AI can also monitor the vehicle’s speedometer and magnetometer to supplement the image data and better keep track of the location between exact fixes. The multiple information streams coming in from cameras and other instruments are integrated by Turium Edge AI to get the best mix of accuracy and reliability. Along the way, the software provides calibration for misaligned instruments to generate a single location estimate that is as reliable as possible.