Massimizzare la precisione del pick & place: il ruolo essenziale della retroilluminazione.

Robot guidance applications require vision systems to deliver accurate, repeatable pose estimation at production speed. The camera sees what the light reveals. When illumination is inconsistent — due to ambient light interference, thermal drift, or incorrect geometry — the vision algorithm receives ambiguous image data, and robot placement errors follow. Selecting and positioning the correct LED illuminator is as important as selecting the camera and lens in any robot vision system.

The three main categories of robot guidance application — pick-and-place from a defined feeder or fixture, bin picking from a randomly filled container, and collaborative robot (cobot) guidance in shared workspaces — each impose different requirements on the illumination system. Working distance, part geometry, ambient light conditions, safety constraints, and imaging modality all influence the correct illuminator choice.

Perché un’illuminazione coerente è essenziale per la stima della posa del robot

Robot pose estimation algorithms — whether rule-based template matching or deep learning-based detection — calculate part position and orientation from image features. The reliability of this calculation depends on the contrast, sharpness, and repeatability of those features in the image. Illumination directly controls all three parameters.

Any change in illumination intensity, colour temperature, or spatial distribution between calibration and production causes the image feature set seen by the algorithm to differ from the training or reference data. This manifests as increased localisation uncertainty, higher pick failure rates, and reduced throughput. In practice, the most common cause of robot vision instability in production is not algorithm failure — it is lighting drift or ambient light contamination.

Guida robot 2D: telecamere area scan e illuminazione LED diretta

Two-dimensional robot guidance systems use area scan cameras to capture a top-down or angled image of the part. The vision algorithm calculates the 2D position and rotation of the part within the field of view. The robot corrects its approach trajectory based on this information before picking.

Illuminatori ad anello per la visione pick-and-place

LED ring lights are the most widely used illumination solution for 2D pick-and-place robot guidance. Mounted coaxially with the camera lens, a ring light provides directional illumination from a consistent angular position relative to the camera optical axis. This geometry produces repeatable shadow patterns that enhance part edge contrast for template matching and feature extraction algorithms.

Low-angle ring lights direct illumination at grazing incidence to the part surface, accentuating surface texture and edge relief. This technique is effective for inspecting parts with raised features, logos, or surface markings that define the pick orientation. For flat, smooth parts on diffuse backgrounds, high-angle ring lights or direct matrix illuminators provide more uniform field illumination and better overall contrast.

Barre luminose e illuminatori a matrice per ampi campi visivi

When the robot picks from a wide conveyor or pallet area, a single ring light cannot illuminate the full field of view uniformly. Bar lights or large-format matrix LED illuminators positioned at controlled angles provide the uniform, directional illumination required across extended fields. Multiple bar lights arranged symmetrically around the field eliminate directional shadow asymmetries that cause orientation errors in vision algorithms.

Guida robot 3D: luce strutturata e proiezione di pattern

Three-dimensional robot guidance systems recover the full 6-DOF pose of a part — position in X, Y, Z and rotation around three axes. This capability is required for picking parts from fixtures, trays, or mixed-orientation presentations where 2D guidance cannot resolve ambiguity in depth or tilt.

Structured light 3D systems project a known pattern — typically fringe patterns, grids, or coded light sequences — onto the part surface. A camera captures the distortion of the projected pattern caused by part surface geometry, and a reconstruction algorithm calculates the 3D point cloud. Illumination for structured light must deliver high contrast between the projected pattern and the part background. This requires a high-intensity, stable LED projector with a narrow emission angle.

For time-of-flight (ToF) and active stereo 3D sensors, LED illuminators operating in near-infrared (NIR) wavelengths — typically 850 nm or 940 nm — project the reference pattern. The camera incorporates a narrowband optical filter matched to the illuminator wavelength to reject ambient light. High-power NIR LED illuminators with stable peak output are essential for reliable 3D reconstruction at robot guidance distances.

Bin picking: illuminazione per l’ispezione di cumuli disordinati

Bin picking requires the vision system to locate and identify individual parts from a randomly filled container, with parts at arbitrary orientations, partially occluded, and at varying heights. This is the most demanding robot guidance application from an illumination perspective.

Sfide di illuminazione nel bin picking

Parts in a bin present multiple challenges simultaneously. Metallic parts create specular reflections from any directional illumination. Overlapping parts produce complex shadow patterns. The vertical depth variation within a bin may span 200 mm or more, causing significant changes in illumination intensity and shadow geometry across the working volume. No single illumination geometry resolves all these challenges for all part types.

For 3D bin picking systems, the structured light projector or active stereo illuminator must provide sufficient contrast for reliable 3D reconstruction across the full bin depth range. High-power LED matrix illuminators with adjustable intensity allow the system integrator to optimise the illumination level for each specific bin geometry and part reflectivity. Diffuse or dome illumination is often combined with the 3D sensor to reduce specular artefacts on metallic parts.

Combinare modalità di illuminazione per un bin picking robusto

Many production bin picking cells use multiple illumination sources in sequence. A structured light projector captures the 3D scene for part localisation. A separate directional LED illuminator fires during 2D image capture for grasp point selection and quality verification. This multi-stage approach optimises each lighting condition independently.

Illuminazione per robot collaborativi: sicurezza, ingombro e funzionamento flicker-free

Collaborative robots operate in shared workspaces alongside human workers. Lighting for cobot vision systems must satisfy requirements that do not apply to fully guarded robot cells: photobiological safety, compact form factor compatible with the cobot end-of-arm tool, and flicker-free operation that does not cause discomfort or hazard to nearby operators.

Sicurezza fotobiologica per l’illuminazione cobot

LED illuminators used in proximity to human workers must comply with IEC 62471 photobiological safety limits. This standard defines Risk Group 0 (exempt), Risk Group 1 (low risk), and Risk Group 2 (moderate risk) classifications based on measured optical radiation levels. For cobot applications, Risk Group 0 or Risk Group 1 illuminators are appropriate to ensure operator safety without requiring additional protective barriers.

High-intensity blue LED illuminators at short working distances can present blue light hazard risk and must be assessed against IEC 62471 limits before deployment. Infrared illuminators beyond 780 nm are invisible to the human eye and do not trigger the blink reflex, requiring particular attention to emitted power levels relative to IEC 62471 infrared radiation limits.

Funzionamento LED flicker-free per ambienti uomo-cobot

LED illuminators driven at mains frequency (50 or 60 Hz) or at low PWM frequencies produce visible flicker that causes eye strain for human operators working nearby. In collaborative workspaces, LED illuminators should operate in true DC continuous mode with regulated constant-current drivers, or in high-frequency PWM mode above 1 kHz that places flicker above the human perception threshold.

Illuminazione montata sul robot vs. fissa: pro e contro

Illuminators for robot guidance can be mounted on the robot end-of-arm tool (EOAT), moving with the camera, or fixed relative to the workspace. Each mounting strategy has specific advantages and limitations that influence system design decisions.

Illuminazione montata in EOAT (end-of-arm)

EOAT-mounted illuminators maintain a constant geometric relationship between the light source, camera, and part surface regardless of robot position. This ensures consistent illumination geometry throughout the workspace. The constraints are size and weight: illuminators for EOAT mounting must be compact and lightweight to stay within the robot payload budget. Cable routing to a moving EOAT requires careful management to prevent fatigue failures over the illuminator service life.

Illuminazione fissa

Fixed illuminators are positioned at a defined location in the robot cell. The robot moves the part or camera to the illuminated zone for image capture. Fixed illumination decouples the illuminator from robot payload constraints, allowing larger, higher-power illuminators. For well-defined pick positions and single-zone illumination, fixed mounting is simpler and more reliable.

Eliminazione della luce ambientale nelle celle di visione robotizzate

Factory ambient light — from overhead fixtures, welding arcs, or sunlight through skylights — contaminates robot vision images when its intensity is comparable to the LED illuminator output at the part surface. Effective ambient light rejection strategies include: high-intensity strobed LED illuminators that overpower ambient light during the camera exposure; darkfield hoods or shrouds enclosing the capture zone; narrowband LED illuminators paired with matching bandpass optical filters on the camera lens; and NIR illumination at 850 nm or 940 nm where factory ambient light levels are lower than in the visible spectrum.

Prodotti e tecnologie

Domande frequenti