In real-world industrial environments, robotic performance is often affected by changing ambient light, reflective surfaces, and surrounding clutter. Among these, high reflectivity is a common challenge. In industries such as automotive, heavy machinery, and home appliances, robots must handle a wide range of reflective parts, including discs, shafts, small components, and large body panels, often in high-mix production scenarios with frequent part variations. After machining, these parts often have highly reflective surfaces, which makes reliable detection and recognition by vision systems difficult, and can potentially compromise the operational stability of the robotic system.
Mech-Mind “Eye + Brain” combines high-precision 3D cameras, AI vision software, and multiple AI models into a unified system for robotic perception and decision-making. It enables robots to handle reflective objects reliably across a wide range of part types. With built-in generalization capability, the system can adapt to new parts and scenarios without task-specific programming or extensive customization. With strong generalization capabilities, the system adapts to different geometries and production scenarios without extensive reconfiguration, providing a scalable solution for complex industrial environments. It is widely used in industries such as automotive and heavy machinery for tasks including picking, loading, sorting, and assembly, helping improve efficiency, accuracy, and reliability.
Mech-Mind “Eye + Brain” enables handling of diverse reflective parts in real-world production
Point clouds of reflective parts in real-world scenarios
Reflective Ring & Disc Parts
Reflective ring- and disc-shaped parts, such as brake discs, bearing sleeves, and inertia rings, are widely used in automotive and related industries. Their smooth, highly reflective surfaces and thin structures make reliable imaging difficult. Stacking, nesting, and partial occlusion further challenge stable robotic picking.
· Picking and Loading of Highly Reflective Tubes
On-site photo
Reflective tubes present challenges due to curved surfaces, thin walls, deformation, and nested stacking. Mech-Eye PRO M-GL captures accurate 3D data for thin-walled reflective tubes with high speed and precision. Combined with AI-based recognition algorithms, the system detects object poses reliably, handles random stacking and nesting, and enables stable picking.
· Picking and Loading of Reflective Brake Discs
On-site photo
Brake discs are critical automotive components with highly reflective machined surfaces. Oil, water, and residues in production further reduce visibility.
Point cloud of brake discs
Mech-Eye LSR L-GL delivers high-quality 3D imaging under strong reflections and ambient light. Combined with self-developed AI-based recognition and planning algorithms, the system accurately detects each disc, optimizes picking sequences, and handles tight stacking and occlusions for reliable automated handling.
· Random Picking of Reflective Bearing Sleeves
On-site photo
Bearing sleeves have smooth, curved surfaces that cause strong reflections, and random stacking increases picking difficulty. Combined with AI vision algorithms, Mech-Eye PRO M-GL generates high-quality 3D point clouds and accurately detects part poses. Robots can plan and optimize picking paths in real time, ensuring stable feeding, reducing collisions, and improving bin clearing efficiency.
Reflective Shaft & Rod Parts
Reflective shaft and rod parts, such as drive shafts and spline shafts, have curved, highly reflective surfaces that make light reflection unpredictable, often causing incomplete or distorted 3D data. This issue is further exacerbated when the 3D camera is oriented perpendicular to the parts.
Equipped with self-developed AI-powered anti-reflection technology, Mech-Eye high-precision 3D cameras can generate complete point clouds without imposing constraints on workpiece orientation. This significantly minimizes point cloud loss and deformation along the edges of shafts, enabling the precise reconstruction of the workpiece’s geometry and contours.
· Bin Picking of Highly Reflective Spline Shafts
On-site photo
Spline shafts are key components in automotive transmission systems. Mech-Mind AI technologies handle reflections, packaging, and ambient light interference, enabling stable, accurate robotic loading and bin stacking, improving automation efficiency.
· Precision Loading of Highly Reflective Drive Shafts
On-site photo
Drive shafts require precise pick planning due to their reflective surfaces and geometry. Mech-Eye LSR L-GL delivers high accuracy, wide field of view, and strong anti-reflection performance, capturing reflective shafts even with oil contamination. Combined with AI-based recognition and planning, the system selects optimal pick points and ensures stable, collision-free operation in confined spaces.
· Loading of Highly Reflective Metal Shaft Ends
On-site photo
Metal shaft ends often have thin, highly reflective surfaces that reduce detection accuracy and loading efficiency. Mech-Eye NANO ULTRA-GL generates high-quality 3D data, even for thin and reflective end faces. Combined with AI-based imaging and recognition, the system precisely localizes the center of the shaft end inner diameter and ensures stable, accurate picking in production.
Small Reflective Parts
Small reflective parts, such as screws, bolts, and outer races, are difficult to detect and handle due to their small size and subtle features. Reflective surfaces further reduce visibility, while high-mix, randomly stacked conditions increase the difficulty of reliable recognition and picking.
· Sorting of Small Reflective Screws
On-site photo
Screws are widely used standardized parts, but their small size, variety, and random stacking make automated handling challenging. Mech-Mind “Eye + Brain” enables robots to efficiently handle large quantities of small reflective parts. Its strong recognition and generalization capabilities ensure accurate sorting and operation, even under highly reflective and mixed conditions.
· Picking and Loading of Small Reflective Outer Races
On-site photo
Outer Race is a common small part in automotive production. Mech-Eye PRO M-GL with AI vision software captures high-quality 3D data of these small, highly reflective parts. This allows robots to accurately detect parts and separators, locate bins with ±50 mm positioning deviations, and achieve high picking success and reliable bin clearing.
Reflective Sheet Metal and Automotive Body Parts
Reflective sheet metal and automotive body parts vary in shape, size, and geometry, often with multiple reflective surfaces. Curved and angled areas, occlusions, and slight deformations make stable picking challenging.
· Picking and Loading Thin Sheet Metal Parts
On-site photo
Thin sheet metal parts have high variability and tight packing, making detection and handling difficult. Mech-Mind “Eye + Brain” enables robots to accurately handle highly reflective, thin, and tightly packed parts, supporting flexible production and improving efficiency.
· Picking and Loading Floor, Side, and Roof Panels
On-site photo
Automotive body panels, including floor, side, and roof panels, are often reflective, flexible, and prone to placement deviations, limiting production efficiency.
Point cloud of a floor panel
Mech-Eye PRO S-THR-GL with AI vision software delivers ±0.5 mm positioning accuracy, supports robot-mounted and rail-mounted configurations, avoids camera-gripper interference, and ensures stable operation with built-in thermal compensation.
Mech-Mind “Eye + Brain” offers strong generalization and adapts flexibly to reflective part handling in diverse scenarios.
Across these use cases, robots must handle reflective objects under challenging conditions. Mech-Mind “Eye + Brain” provides a consistent and scalable solution, supporting reliable picking, loading, sorting, and assembly. With strong generalization capability, the system adapts to diverse parts and changing production scenarios without extensive reconfiguration. As adoption expands across industries, these solutions help manufacturers improve efficiency, maintain quality, and move toward flexible, intelligent automation.