Sub-Category

AI-Powered Robots

Sub-Category · Science & Innovation Competition

Learn the logic of Image Processing using ready-made modules and smart cameras. Object recognition, sign reading or gesture control — choose one of three concepts and compete.

Apply Now All Categories

Smart Robotics with Image Processing

The purpose of this category is for students to grasp the logic of "Image Processing" using ready-made modules and cameras instead of complex code. The use of smart cameras like HuskyLens, PixyCam or software like mBlock, Teachable Machine, Python (MediaPipe/OpenCV) is completely free.

Teams can compete by choosing one of the following 3 tasks (concepts). Each concept has a different difficulty level, field dimensions and scoring system.

HuskyLens PixyCam mBlock Teachable Machine Python (MediaPipe / OpenCV)

Task Concepts

Each team competes by choosing one of the three concepts below. Concepts are ranked from easy to hard.

Field Info
Field: 150 × 150 cm flat white surface
Total Time: 3 minutes
Delivery: Red and Blue zones
Concept 1 Easy

Smart Sorter (Object Sorting)

The easiest level. The robot is stationary or moves very little, it only recognizes the object and decides.

The referee places randomly colored cubes (or cards with shapes like Apple, Banana) in front of the robot's camera.

The robot uses its camera and AI model to decide whether the object belongs to the Red or Blue zone, then autonomously pushes the object to the correct zone.

Scoring
Correct zone delivery+20 pts
Wrong zone delivery0 pts
Total time3:00
Field Info
Track: 200 × 200 cm
Signs: 10 × 10 cm (STOP, TURN RIGHT etc.)
Track with black road line
Concept 2 Medium

Smart Traffic (Sign Reading)

An updated version of the classic line-following robot with "Autonomous Vehicle" technology. The robot scans roadside signs while following the black line.

The robot advances normally following the black line. Meanwhile its camera continuously scans traffic signs on the roadside.

The robot must stop its motors for exactly 3 seconds when it sees a red STOP sign. At intersections, it must read the direction arrow sign and autonomously turn in that direction.

Scoring
Track completion+50 pts
Each correctly detected sign+30 pts
Sign violation (penalty)-10 pts
Field Info
Course: 210 × 280 cm labyrinth
Control: Hand/body gestures (Pose Tracking)
Obstacle collision: +5 sec penalty
Concept 3 Hard

Jedi Mode (Gesture Control)

The most interactive concept that puts on a show for the audience, without any external controller. The competitor directs the robot using arms and hands.

The competitor (Operator) stands in front of the computer camera set up outside the field. No phone or controller may be in their hands.

The camera maps the person's hand/body skeleton (Pose/Hand Tracking). Raising the right hand gives a turn right command, opening the palm brakes the robot.

Scoring
Shortest time winsAgainst the clock
Each obstacle hit+5 sn

Ready to combine AI with robotics?

Apply now to showcase your image processing skills on the international platform.