Robotics / Concepts and practice

Robot vision

Robot vision converts image or depth data into information for a robotic task, such as an object position or an inspection result.

Prepared by MALONI Robotics · Updated

How it works

Calibration relates camera coordinates to the robot workspace. Image processing may use geometric methods or learned models, depending on the task.

Where to start

Part picking requires both locating an object and choosing a suitable grasp. Quality inspection instead requires defined defects and a way to verify the result.

Robot applications →

Limits and what to verify

Reflections, transparent surfaces, occlusion and lighting changes can affect results. Detecting an object does not establish the safety of the next movement.

Before requesting a quote

Collect representative images, including difficult cases, and define acceptable error. Evaluate the camera and lighting together with the robot and tool. A linked robot platform does not imply an included, ready-to-use vision system.

Related MALONI platforms

These links help you explore the portfolio. The technologies described are not automatically included in delivery.

Related concepts

Sources and further reading

MALONI offers robot platforms. This guide separates general principles, research examples and product offers; it is not an independent test of the linked models.

Your next project step

Discuss your application with MALONI →