On your drone
Edge AI Companion
Software running on your device’s onboard computer. Run detection models, retain observations and telemetry while disconnected, and synchronise with the ground node when links return.
APPLICATION
AI for drones on Scaleout Edge.
Deploy and operate onboard AI, even when connectivity is limited.
Scaleout Edge provides software infrastructure for AI onboard drones. Run models on sensor feeds, retain observations and telemetry during disconnection, and synchronise with your ground infrastructure when links return.
The onboard workflow
Keep onboard AI working through connection changes.
A continuous ground link should not be a prerequisite for onboard inference. Connect model deployment, local processing and retained observations in a workflow that can synchronise when connectivity returns.
Improve from collected observations.
Review data and evaluate model updates on your ground infrastructure before approving them for devices.
-
Deploy
Stage approved models to selected drones without redeploying the surrounding application software. Match models to each device’s computing resources.
-
Run
Process sensor feeds on onboard compute and produce detections locally. Inference continues when the connection to ground infrastructure drops.
-
Retain
Cache selected observations, detection records and telemetry onboard during disconnection, keeping them available for later review on your ground infrastructure.
Synchronise
Synchronise retained records with your ground node when links return. Make observations and reported device state available to connected systems.
-
Improve
Use collected observations to train and evaluate models on ground infrastructure. Distribute approved improvements to participating devices when connectivity allows.
Where the software runs
On your drone.
At your ground node.
Across your fleet.
Three connected software roles bring onboard inference, local data review and model management into the same workflow.
At your ground node
Vision Ground Node
A software workbench on local GPU-equipped infrastructure. Review drone observations, annotate data, train and evaluate models, and stage approved updates back to devices.
Across your fleet
Scaleout Edge
Manage model versions and stage updates to all or selected devices. Review reported telemetry and coordinate federated learning across participating ground nodes when connectivity allows.
Drone ↔ Ground nodeObservations return to the ground node for review and model improvement. Approved models are staged through it to the Companion when links are available.
Ground node ↔ Scaleout EdgeEach site trains on local footage. Federated learning combines model updates to improve a shared model without transferring raw footage between sites.
The software runs on hardware supplied by you or your chosen partners. Scaleout Edge is deployed within your chosen infrastructure. Scaleout provides the software infrastructure and engineering support for this workflow, integrated with your sensor feeds, models, computing infrastructure and operational systems.
How it fits your operation
Connect with the systems you already use.
Integrate onboard AI with your drone hardware, sensor feeds and ground applications. Scaleout Edge connects model deployment and learning with your existing development tools. Hardware compatibility, interfaces and computing requirements are assessed during evaluation.
Your onboard hardware
Connect supported cameras and sensors to your companion computer. Assess inference speed, memory, power and storage requirements against the equipment and operating conditions you choose.
Your models
Bring models from your team or chosen suppliers. Establish a compatible onboard workflow, then use observations collected by your devices to evaluate improvements on ground infrastructure.
Your operational systems
Make detection records and telemetry available to your ground applications through agreed interfaces. Monitor reported connectivity and device state to understand when information was last synchronised.
For the data selection, annotation and training workflow, explore tactical computer vision.
Bring onboard AI and local model improvement together with integration and engineering support. Explore the Tactical AI Network solution.
Your decisions.
Your models. Your data.
Your models. Your data.
Your team retains control of operational decisions and model approvals. Your operational data and trained models remain yours.
Get started
Evaluate it in your environment.
The drone module has completed internal integration testing on representative hardware and is ready for structured field evaluation. Work with Scaleout engineers to assess it against your equipment and operational requirements.
Define the scope
Choose an onboard AI task, identify hardware and sensor feeds, and agree on success criteria for inference performance, model deployment and operation with limited connectivity.
Test the workflow
Run the workflow on selected hardware. Check onboard inference, retained observations and model deployment through connection loss and recovery within the agreed test environment.
Plan the next step
Review results and define the integration, field testing and support needed for your platform. Agree how your team will manage observations and future model updates.
Scaleout provides software, engineering support and knowledge transfer within the agreed scope. Security requirements and accreditation support are scoped around your deployment environment. Hardware is supplied by you or your chosen partners.
Discuss an evaluation