AlphAI Artificial Intelligence Educational Robot

AlphAI: Artificial Intelligence learning robot

The tiny and friendly AlphAI robot is equipped with several sensors, including a camera for image recognition, which will allow it to learn various actions.

• Sensors: Wide angle camera / Ultra-sound / 5 infra-red line tracking sensors
• Actuators: 2 wheel drive / 4 multi-color LEDs / buzzer
• Onboard computer: Raspberry Pi Zero (linux, Python) / WiFi connection to PC
• Hull: nice, shock absorbing, removable.

AlphAI is a robot that learns in real time. In supervised learning, the student trains the robot by piloting it, then the robot reproduces the behavior independently: robot races, choreographies, etc. In reinforcement learning, the student defines a reward system and the robot learns by trial and error: avoiding obstacles, chasing a ball, following a circuit, etc.

Artificial intelligence with AlphAI

• Classical programming: like any other robot, AlphAI can be programmed classically with instructions. Whether in Python or via the graphical interface, your new robot will obey at a glance. If it bumps into something, you'll know who to blame!
• Supervised learning: Create data to train the robot's AI starting with remote control. It will then be able to recognize people, compete in a robot race, or perform a choreography. Here, it looks like it likes you!
• Reinforcement learning: At first he can't do anything. He bumps into himself, spins around, hesitates... Then, he becomes more confident and develops his interaction skills. Look, he is now able to avoid obstacles, follow a path and even play soccer. Bravo AlphAI!

Because the quality of learning also depends on the robot's environment, Learning Robots has designed an arena to provide you with an optimized turnkey solution: good floor/wall contrast for fast learning, good detection that the robot is in motion, size 80x80cm, fits on a large classroom table and modular: possibility to enlarge, mazes, etc

AlphAI software: the black box of AI

The AlphAI software license is required only once for several AlphAI robots and is sold separately.

AlphAI is the only program that makes learning methods (deep learning, KNN, etc.) accessible through its graphical interface. It shows you how to use them in an easy and visual way, from the beginner mode for a "plug and play" start to the expert mode for writing AI in Python!

This visualization and control software allows to configure different types of learning, from beginner mode to load already optimized configurations, to expert mode to control all AI parameters. It also takes you into the algorithms, including K-nearest neighbors and Deep Learning, displaying their details and workings in an intuitive way.

Enter the Deep Learning network

Simple network to understand everything in detail: On simple learning scenarios, understand and teach step by step how artificial neural networks work. Trace connections inside the network yourself and observe the consequences on the robot's movements. Thanks to multiple displays (activity values, thickness of connections, colors for learning, etc.), explain in pictures the learning equations that automatically optimize these connections.

Complex networks for scaling: Activate the camera and the intermediate layers of neurons and see how the simple principles detailed above allow the learning of complex behaviors (obstacle avoidance, pushing a balloon, etc.) after scaling to thousands of neurons and hundreds of thousands of connections. You have in front of you all the calculations made.

Other algorithms: Discover still very visually other Machine Learning algorithms. In particular, the K-nearest neighbor algorithm, which is part of the high school curriculum, is particularly intuitive and adapted to introduce supervised learning. Then compare its results to those of neural networks, on the same training database.

Take the controls and experiment

Beginners, let us guide you: Select the beginner display mode to see only the most common settings. Most importantly, load demo configurations to effortlessly launch pre-set supervised or reinforcement learning scenarios.

Intermediates, launch your own learning: Test your own learning by selecting the sensors used as inputs, the authorized actions, the reward mode for reinforcement learning. Finally, select and configure the AI algorithm, for example to test what the intermediate neurons are used for...

Experts, benefit from a professional environment: Use the AlphAI platform to debug your learning model, test the most advanced parameters, study the properties of individual neurons, etc...

Program and code in Python

Classic programming: A dedicated Python library allows to take control of the robot from the computer with simple instructions to direct the motors, retrieve sensor data, and program displacements, exit mazes, etc., without necessarily requiring AI. It is also possible to run Python code directly on the Raspberry Pi for faster reaction times.

Programming AI with a modular approach: Choose the "student code" algorithm option and you code the one of your choice in the boxes provided: the AlphAI platform takes care of all the I/O machinery and you focus on the most interesting! Even easier than AI, program your own reward function to make AlphAI learn new behaviors. Our CLF worksheets provide examples of how to program the program.

Other systems: Learning Robots is working on making the AphAI software compatible with other robots and automated systems, so that they can be driven by learning. You will also be able to carry out your own projects to develop new hardware to be controlled with AlphAI.