The KQ-V9 projects a structured light pattern and reconstructs a dense point cloud of the scene, giving the robot the depth information that random bins and mixed pallets demand. Parts lying at angles, overlapping or half-buried become individually gripped candidates with a computed approach vector.
The sensor ships with the matching software runtime: CAD-based part matching for known geometry, or surface-based picking for sacks and organic shapes. Collision checking against the bin walls and neighboring parts runs before every pick command leaves the sensor.
Mounting patterns cover static overhead frames and on-arm brackets for the KQ-A12.
