Robot Operations
Robot Operations focuses on day-to-day production workflows after robots are live. This phase is about maintaining high availability, ensuring safety, and driving continuous improvement through data review.

Suggested shift workflow
A structured operational routine ensures that potential issues are identified before they impact business services.
- Fleet Scan: Check the global robot list for online status, heartbeat freshness, and active alert markers.
- Health Drill-down: For critical units, review real-time metrics including CPU load, memory usage, storage capacity, and network latency.
- Teleops Verification: Periodically open Teleops to verify media streams and sensor data. Take brief manual control to ensure movement responsiveness.
- Post-Incident Review: Analyze recent conversations, scene execution outcomes, and review the event timeline using the Replay feature to identify root causes of any deviations.
High-priority metrics
Track these KPIs to measure the health and efficiency of your robot fleet:
- Availability: Ratio of scheduled uptime vs. actual online duration.
- Sensor Reliability: Percentage of time video and perception streams are fully functional.
- Execution Success: Success rate of automated commands and scene transitions.
- MTTR (Mean Time To Recovery): Average time taken to restore a robot to service after a failure or alert.
Monitoring & Intervention
When a robot encounters an edge case it cannot resolve autonomously, operators can intervene through Teleops. This “human-in-the-loop” approach ensures safety while the system continues to learn from manual corrections.
Related docs
- Fleet List and Status — Start from the fleet list and drill into one robot
- Teleops — Real-time remote monitoring and control
- Scenes — Managing autonomous mission execution
- Events and Replay — Post-incident analysis with the event timeline