AiVibe

Robotics & Physical AI

3D Machine Vision

3D machine vision captures depth as well as appearance, producing point clouds or depth maps of parts and scenes using techniques such as stereo cameras, structured light, laser triangulation and time-of-flight sensing.

Stereo systems match features between two cameras to calculate depth, and active stereo adds a projected texture to help matching. Structured-light sensors project known patterns and compute depth from their distortion. Laser triangulation profilers measure a line profile as parts move past, building up a 3D surface, and time-of-flight cameras measure the travel time of modulated or pulsed light. The output is a point cloud or depth image calibrated in physical units.

3D vision is used for bin picking, depalletising mixed cases, measuring volume, height and flatness, inspecting weld beads and seals, and guiding robots to parts whose position varies in all directions. It also supports mobile robot navigation and obstacle detection.

Shiny, transparent and very dark surfaces challenge most 3D techniques, as do occlusions and strong ambient light. Resolution, accuracy, working distance, field of view and acquisition time must suit the task, and the sensor must be calibrated to the robot through hand-eye calibration. Deep learning is increasingly combined with point-cloud processing to recognise parts.

Key points

Where AiVibe comes in

AiVibe designs and manufactures the AiAmbA AI Factory, whose edge devices and AI agents let people talk to robot controllers in plain language. Robotics perception is an AiAmbA use case, and robot safety functions follow ISO 10218 and never depend on the AI layer.

Explore AiVibe’s work in Robotics & Physical AI →

Related terms

Terms that refer to 3D Machine Vision

Ask AiMuruga can explain 3D Machine Vision for your plant, product or security programme, and draw how it fits.