docs: make K1 realtime map the primary lidar source
This commit is contained in:
@@ -61,6 +61,22 @@ The physical Lab 002 capture now provides stronger evidence than the prior
|
||||
official-material inference: a compressed left/right preview leaves the device.
|
||||
It still does not prove a full-resolution raw panorama stream.
|
||||
|
||||
### Architecture clarification · 2026-07-25
|
||||
|
||||
The processed nature of `lio_pcl` does not make the stream non-real-time or
|
||||
unsuitable as the current K1 perception source. Mission Core received the
|
||||
point/pose reports while the rig moved; K1 had already performed onboard
|
||||
LIO/modeling before publication. The accepted roadmap therefore keeps K1 as the
|
||||
physical starting sensor and produces separate short-TTL perception and
|
||||
persistent-reconstruction derivatives from the immutable report stream.
|
||||
|
||||
Current LiDAR work is passive. Static firmware evidence may explain fields and
|
||||
processing order, but does not authorize a firmware change, onboard exporter,
|
||||
internal-topic invocation or new application command. The handheld `1.27 m`
|
||||
estimate is retained only in its historical diagnostic result; operational
|
||||
surface height is time-varying and must be estimated from pose-relative
|
||||
geometry when the selected platform profile requires ground.
|
||||
|
||||
### Still unsupported until measured
|
||||
|
||||
- Local `map.las` layout resembles the network stream.
|
||||
|
||||
Reference in New Issue
Block a user