Skip to main content

Frequently Asked Questions

Is Atlas another log management system?​

No.

Atlas is not a log collection platform.

Atlas provides runtime governance for robots, including:

  • Runtime observation
  • Evidence organization
  • Runtime investigation
  • Historical knowledge reuse
  • Cross-organizational collaboration

Logs may serve as one source of runtime evidence, but Atlas is not designed to replace existing log management systems.

Does Atlas replace a fleet management system?​

No.

Atlas does not replace fleet management platforms.

Fleet management systems focus on:

  • Device management
  • OTA updates
  • Remote monitoring
  • Task scheduling
  • Robot operations

Atlas focuses on:

  • Runtime investigations
  • Evidence Packs
  • Historical RGA
  • Investigation workflows
  • Runtime governance

The two systems are complementary.

Does Atlas replace a ticketing system?​

No.

Atlas does not replace ticketing platforms such as Jira or ServiceNow.

Atlas integrates with existing ticketing workflows.

Its responsibility begins after a Runtime Execution Failure (REF) has been reported.

Does Atlas automatically determine the root cause?​

No.

Atlas does not automatically determine:

  • Root cause
  • Liability
  • Product defects
  • Customer fault

Atlas organizes runtime evidence and investigation context.

Engineering teams remain responsible for technical conclusions.

Does Atlas support AI-assisted investigations?​

Yes.

Atlas provides structured runtime evidence that can be used by AI-assisted analysis.

However, Atlas does not depend on AI.

All runtime investigation workflows remain deterministic and fully reviewable.

What types of robots does Atlas support?​

Atlas is platform-independent.

It can be deployed across different:

  • Robot platforms
  • SBC platforms
  • Linux environments
  • ROS-based systems
  • Custom robot software

Deployment is based on runtime surface mapping rather than any specific robot architecture.

What sensors does Atlas support?​

Atlas is sensor-agnostic.

Typical integrations include:

  • Cameras
  • LiDAR
  • IMUs
  • GNSS
  • Radar
  • Encoders

Additional sensors can be integrated through runtime surface mapping.

Does Atlas require ROS?​

No.

ROS is optional.

Atlas can observe runtime surfaces on systems with or without ROS.

Does Atlas require a cloud connection?​

No.

Atlas supports multiple deployment models, including:

  • Cloud deployment
  • Headquarters servers
  • On-premises networks
  • Offline deployment
  • Manual export
  • Dock-and-upload workflows

Continuous Internet connectivity is not required.

Does Atlas require us to replace our existing infrastructure?​

No.

Atlas is designed to integrate with existing infrastructure.

It coexists with:

  • Fleet management platforms
  • Cloud platforms
  • Ticketing systems
  • Object storage
  • Existing monitoring platforms

No large-scale infrastructure replacement is required.

Can Atlas support multiple robot models?​

Yes.

Atlas is designed as runtime governance infrastructure rather than a project-specific integration.

A single deployment can gradually expand to support multiple:

  • Robot models
  • Sensor products
  • Deployment environments
  • SBC platforms

Can Atlas work with multiple sensor manufacturers?​

Yes.

OEM runtime investigations can collaborate with different sensor manufacturers through the Sensor Engagement Pack™.

Each sensor manufacturer maintains its own:

  • Runtime Profiles
  • Historical RGA
  • Investigation workflows

Atlas provides the collaboration framework.

What is a Sensor Runtime Profile?​

A Sensor Runtime Profile describes how a sensor behaves under known deployment environments and known runtime disturbances.

It is not:

  • A certificate
  • A passport
  • A compliance document

It is reusable runtime investigation knowledge.

What is Historical RGA?​

Historical Runtime Governance Asset™ (Historical RGA) stores reusable investigation knowledge.

Typical contents include:

  • Runtime patterns
  • Investigation paths
  • Investigation Results (IR)
  • Lessons Learned (LL)
  • Runtime Surface references

Historical RGA enables future investigations to begin with accumulated knowledge rather than from scratch.

What is an Evidence Pack?​

An Evidence Pack is a standardized package of runtime evidence generated from a Runtime Dataset.

It is associated with a specific Runtime Execution Failure (REF) and provides structured evidence for runtime investigations.

An Evidence Pack is not a log archive.

Does Atlas permanently store all runtime data?​

No.

Atlas follows configurable retention policies.

Typical deployments use:

  • Rolling buffers
  • Retention policies
  • Dataset locking upon REF
  • Time-range exports

Customers control their own data retention policies.

Who owns the runtime data?​

Customers own their runtime data.

OEMs retain ownership of:

  • Runtime Datasets
  • Evidence Packs
  • Historical RGA

Sensor manufacturers retain ownership of:

  • Sensor Historical RGA
  • Sensor Runtime Profiles

Atlas does not claim ownership of customer runtime data.

Can Atlas share customer data across organizations?​

No.

Atlas never automatically shares customer runtime data.

Cross-organizational collaboration always requires customer authorization.

For industry-wide knowledge reuse, Atlas supports anonymized knowledge sharing through Assist Vault™.

Is Atlas a SaaS platform?​

Not necessarily.

Atlas supports multiple deployment models, including:

  • On-premises deployment
  • Private cloud
  • Customer-controlled infrastructure
  • Hybrid deployment

Deployment depends on customer requirements.

Is Atlas open source?​

No.

Atlas is a commercial runtime governance infrastructure developed by SensorDeck.

Customer deployments are delivered through professional integration and customized implementation projects.

Is Atlas designed for only one robot project?​

No.

Atlas is designed as reusable runtime governance infrastructure.

It can scale across:

  • Multiple robot SKUs
  • Multiple sensor products
  • Multiple deployment environments
  • Multiple customers

without redesigning the runtime investigation framework.

How is Atlas typically deployed?​

Deployment typically follows four stages:

Pilot

↓

Controlled Deployment

↓

Production Ready

↓

Full Deployment