Baseline memory for robotics hardware
Robotics and physical AI product in developmentStreamlineddrift detection.
A Runtime dashboard to review working state, check current hardware, inspect evidence, and know whether the robot is still safe to trust.
Runtime health
Drift view for the current camera and sensor behavior against the accepted baseline.
Health score
98%
ok · May 22, 8:41 PM
Sensor sync
+1ms
baseline 12ms · latest 13ms
Camera latency
-2%
baseline 83ms · latest 81ms
Sensor sync trend
IMU and camera timing compared with the accepted baseline.
latest +1ms
Camera latency trend
First-frame latency sampled across recent checks.
latest 81ms
When drift started
May 22, 8:41 PM
Likely cause
Nearest setup event: imu-json
Evidence clue
No specific change recorded
Recommended action
Keep automatic checks enabled
Baselines and checks
Raw health observations
One idea,different machines,same drift problem.
The runtime starts where the hardware already is.
Oplut runs beside the robot on existing boards, laptops, and lab computers. It reads the local OS, ports, services, and connected devices before deciding what should be measured.
First prompt
What are you working with?
Replacingthe blankstatus light.
One local loop for the question that matters: this worked before, what changed?
Memory gap
When hardware changes, know what changed
The baseline turns a vague field failure into a specific difference from the version that worked.