ROBOTICS & AVs
Warehouse Spatial Mapping
Warehouse Spatial Mapping. What it takes to run it.
Dynamic updating of facility maps for AMR fleets.
// 01 · The question
“What is where in this facility right now?”
// 02 · Why it is hard
The facility is being changed by the people working in it all day. The map has to be updated continuously, and the ground truth it is measured against is stale by the time it is collected.
// 03 · What the labels have to be
3D cuboid
Decided before the first frame
The map is metric and the fleet plans paths through it, so an obstacle needs extent and position in world coordinates. A pallet's footprint decides whether an aisle is passable, which is a volume question.
This is the decision that is expensive to reverse. The geometry has to match what the answer contains, and finding out it does not means labelling the set a second time.
// 04 · How we run it
Three parts of one loop, on this job.
LexAnnotate
Pixel-Perfect Data
Pixel-level semantic segmentation, LiDAR point cloud annotation, sensor fusion labeling, and temporal event tracking for complex perception tasks. Expert annotators, accountable for every label.
LexInsight
Spatial Intelligence
Build spatial understanding from flat labels. Map object relationships, track movement patterns, identify zero-day anomalies that standard training data misses.
LexAlert
Perception Alerts
Continuously validate perception performance in production. Detect when sensor drift or environmental changes degrade navigation. Route failures back for retraining before they cause incidents.
// 05 · What breaks it after launch
Racking, stacked pallets and other robots block the view of the space being mapped. The map reports clear on the aisles it could see and confident-but-empty on the ones it could not.
It is not the only one that can get this use case, it is the one that usually gets it first. All ten conditions.
// 06 · In robotics
Dynamic Spatial Navigation
The Challenge
Flat bounding boxes failed to provide spatial context for autonomous material handling in shifting environments.
Our Execution
Deployed pixel-level semantic segmentation and temporal event tracking for high-fidelity spatial datasets.
The Result
Enabled autonomous systems to adapt to changing environments safely. Reduced manual overrides.
95%+
Navigation reliability in changing environments