Tag - Vision

Can a robot truly understand what it sees?
This hands-on lab takes you on a journey from raw images to precise robotic actions. Using only a single camera mounted on the Niryo Ned2, you will build a full perception, pipeline from camera calibration to object detection and real-time grasp execution.

The current vision system relies on a monocular camera mounted on the robot's wrist and a calibrated workspace using visual markers (tags). Here are the technical specifications of this onboard camera:

• Sensor type: IMX322 (CMOS)
• Resolution: 640 × 480
• Pixel size: 12.8 × 11.6 μm
• Field of view (FOV): approximately 104°
• Focal length: approximately 3 mm

This setup, while functional, presents several technical and methodological limitations:

• Need for visual tags to guide the robot.
• Demanding workspace configuration that must remain fixed and precisely calibrated.
• High sensitivity to errors, whether from robot movements or setup changes.
• Vulnerability to lighting conditions, shadows, or reflections.
• Detection limited to a few predefined rigid objects.

When a human looks at an object, they can estimate its position, distance, and orientation almost instinctively. Our brains have been trained since birth to perceive depth, interpret shadows, relative sizes, movements, and reconcile information from both eyes. It’s a rich, complex, yet entirely transparent process for us.

A camera, however, doesn’t see like a human. It captures only a 2D projection of a 3D world. It has no sense of depth or understanding of the scene’s context. Estimating an object’s pose (its position and orientation in space) from a single image is therefore a fundamentally ambiguous and ill-posed task.

This is compounded by several challenges specific to the industrial context of this project:

• Exclusive use of a monocular camera, without direct access to depth.
• Camera is mobile, as it is mounted on the robot, making localization difficult.
• No additional sensors (stereo, LIDAR, etc.).
• Camera noise, lighting, resolution.

In response to these challenges, the goal of this lab is to build a robust, reliable, and coherent image processing pipeline in which the only source of information is a monocular onboard camera—and the end goal is to guide a robotic arm to interact correctly with one or more objects in its environment.

The lab thus aims to design a system that:

• Operates without a preconfigured workspace or visual tags,
• Adapts to a mobile, onboard camera,
• Enables flexible recognition of a wide range of objects using their CAD models,
• Provides 6D pose estimation (position + orientation) in the robot's base frame,
• Detects multiple objects simultaneously with dynamic selection,
• Remains robust to environmental variations.

Entdecke die Welt der Robotik mit unserem umfassenden Lehrplan, der für Studierende und Lehrkräfte entwickelt wurde, die die Grundlagen der Automatisierung, Programmierung und robotischen Anwendungen erforschen möchten. Dieser Kurs führt die Teilnehmenden von der Geschichte der Robotik bis hin zur praktischen Programmierung und industriellen Automatisierung mit dem kollaborativen Roboter NED2 (Cobot) und der Blockly-Programmierung.

Für wen ist dieser Kurs geeignet?

✔️ Studierende in MINT- oder technischen Ausbildungsprogrammen
✔️ Lehrkräfte, die einen strukturierten Robotik-Lehrplan suchen
✔️ Einsteiger:innen, die Robotik und Automatisierung ohne Programmierkenntnisse lernen möchten
✔️ Alle, die sich für industrielle Robotik und Automatisierung interessieren

Melde dich noch heute an und mach den ersten Schritt zum Profi in Robotik und Automatisierung! 🚀

We want to create a loading and unloading scenario for a machine tool. The robotic arm NED2 will have to coordinate with a virtual machine tool to enable a part to be loaded into the machine, machined, unloaded onto the conveyor and then cleared from the area by the conveyor. The machine tool will simulate machining and cleaning cycles in a loop, and the robotic arm NED2 will act in coordination.

Laboratory Contents

Chapter 1: Pick and place

• Practice the notions of reference points and points, associated with a tool and a base, in robotics.
• Understand and master articular and linear movement sequences.
• Perform pick-and-place operations in a workspace using a specific tool.

Chapter 2: Vision

• Understand and control an RGB camera's settings in a specific environment.
• Put into practice object detection with the camera.

Chapter 3: Generating trajectories

• Create the virtual environment of a robotic cell using robotic simulation software.
• Simulate a trajectory to check its feasibility and extract the information needed to program a robot.
• Compare robotic arm NED2 trajectories with simulation and optimize.

Chapter 4: Synchronization with a PLC

• Integrate operator controls into the robotic arm NED2’s environment.
• Be able to simulate robotic arm NED2’s operations with on-screen displays to develop complex behaviors.
• Deepen your programming knowledge with thread implementation in Python.
• Apply the synchronization of the robotic arm NED2's movements with external commands.
• Learn how to define the main stages of a machine tool loading and unloading operation.

Chapter 5: Integration

• Be able to create partial integrations constituting sub-systems.
• Be able to integrate all subsystems.

Prerequisite knowledge

Python: Basic syntax + simple data and control structures + loops + simple function calls

Required components