Marca - Vision

Descubre el mundo de la robótica con nuestro programa completo, diseñado para estudiantes y educadores interesados en explorar los fundamentos de la automatización, la programación y las aplicaciones robóticas. Este curso guía a los alumnos desde los conceptos básicos de la historia de la robótica hasta la programación práctica y la automatización industrial utilizando el robot colaborativo NED2 (cobot) y la programación Blockly.

¿A quién va dirigido este curso?

✔️ Estudiantes de secundaria y universitarios en programas CTE o STEM
✔️ Educadores que buscan un currículo estructurado de robótica
✔️ Principiantes que desean aprender automatización robótica sin necesidad de programar
✔️ Cualquier persona interesada en la robótica industrial y la automatización

¡Únete hoy y da tu primer paso hacia el dominio de la robótica y la automatización! 🚀

Este curso permite a los estudiantes obtener las certificaciones SACA C-103 y C-215

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.

Scenario

The scenario enables interaction between an operator and a Robotic arm NED2​, using interfaces based on artificial intelligence. In particular, the operator can use gestures to designate the part to be picked up by the Robotic arm NED2, one gesture corresponding to a part from a collection arranged in the alpha zone. The Robotic arm NED2 places the gripped part in the operator’s hand, after identifying its position using the vision set camera, above the loading zone. This is a pick-and-place sequence, where the pick and place points are provided in real time by the operator’s gesture commands. Gesture and hand position recognition is performed by deep learning tools.

The objective of the scenario is to perform this operation by following the steps illustrated in the following algorithm:

Lab Contents

Chapter 1: Pick and Place

• Define points of interest
• Create movements for pick and place

Chapter 2: Gesture recognition

• Know how to use Teachable Machine to train a model
• Obtain predictions based on gestures
• Create a filter to validate a gesture

Chapter 3: Hand detection

• Detect the hand in the camera image
• Calibrate the camera
• Obtain the coordinates of the drop point in the middle of the hand

Chapter 4: Integration

• Integrate the subprograms into a complex and functional program

Prerequisite knowledge

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

Reference frames and transformation: Understanding how Cartesian coordinate systems work and the principle of transformations

Required Equipment