Return to the range 
AlphAI Robot
SI/STI2D
Doc. commercial Fr
AlphaAI - brochureRessources
AlphaAI - Ressources pédagogiquesVidéo
Related ranges :
SI/STI2D
The AlphAI robot and the AlphAI software platform explain and demonstrate the use of artificial intelligence through the robot's real-time learning, with all the algorithmic details (artificial neural network, etc.) displayed in the graphical interface.
The AlphAI solution teaches Artificial Intelligence in a concrete and visual way. The robot learns in real time. In supervised learning, the student trains the robot by controlling it, then the robot reproduces the behavior autonomously: robot races, choreography, etc. In reinforcement learning, the student defines a reward system and the robot learns on its own through trial and error: avoiding obstacles, chasing a ball, following a circuit, etc.
The AlphAI software for visualization and control allows you to configure different types of learning, from beginner mode for loading pre-optimized configurations to expert mode for controlling all AI parameters.
It also introduces algorithms, such as K-nearest neighbors and deep learning, displaying their details and operations in an intuitive way.
The AlphAi robot and its interface will introduce a new way of programming where, instead of giving instructions, we rely on machine learning algorithms. This is why the graphical interface does not resemble graphical programming environments such as Scratch.
The modular training track (arena) for the AlphAI robot has been specially designed in terms of surface and color choices to enable the AlphAI robot to learn in a reproducible manner in arenas of varying sizes.
The AlphAI learning track has been optimized for effective demonstrations with the AlphAI robot. It simplifies the robot's environment with high color contrast, enabling rapid learning. It also protects the equipment: the floor mat protects against dust, while the flexibility of the wood absorbs shocks and allows the robot to be pushed to higher speeds. In particular, the robot can be trained to roll on the printed track. An individual track, measuring 80 cm on each side, can be placed on a large table and can accommodate one or two robots. By combining several tracks, large arenas can be formed on the floor to accommodate numerous robots (e.g., with two tracks forming a square measuring 160 cm on each side, a maximum of six robots).
The learning track consists of: