Laboratory For The Development Of Artificial Intelligence For Colleges (DL AI-BASE)

ELEMENTS THAT COMPOSE IT

SENSING MODULE - IT ALLOWS YOU TO:

• Collect essential data to train AI models.
• Provide real-time information for decision-making.
• Simulate real-world conditions to solve practical problems.
• Connect the physical world with AI applications.
• Facilitate experimentation with different types of data.
• Allow measuring, monitoring, and controlling environmental variables.
• Help understand the impact of AI on physical systems.
• Drive hardware and software integration in educational projects.
• Promote active learning through direct manipulation.
• Encourage innovation by exploring new AI applications.

ACTUATOR MODULE - IT ALLOWS YOU TO:

• Transform AI decisions into physical actions.
• Enable tangible interaction with the environment.
• Demonstrate the applicability of AI in real systems.
• Facilitate validation of AI predictions and models.
• Teach how to automate processes using AI.
• Simulate practical solutions for industrial problems.
• Convert processed data into useful responses.
• Integrate theoretical learning with visible results.
• Foster creativity in designing intelligent systems.
• Promote understanding of AI’s impact on automation.

POWER SUPPLY MODULE

• Universal voltage input: Compatible with 110 - 240Vac.
• Regulated outputs: Provides 5V and 3.3Vdc simultaneously.
• Current capacity: Supports up to 2A per outp ut.
• Overload protection: Prevents damage from excess current or short circuits.
• Compact design: Easy integration into prototypes or mounted systems.
• Standard connectors: Screw terminals and pins for easy connections.
• LED indicators: Shows power status in real time.
• Energy efficiency: Reduces energy loss during conversion.
• Compatibility: Ideal for powering microcontrollers like Arduino, NodeMCU, and sensors.
• Easy mounting: Can be fixed to surfaces or educational racks.

EMBEDDED PROGRAMMING MODULE - (ARDUINO MEGA)

• Basic Level – Fundamentals of Embedded Programming:

• Introduction to embedded systems and their applications.
• Configuration of digital inputs and outputs.
• Timing and control of simple events.
• Reading analog signals.
• Generation of pulse-width modulated (PWM) outputs.
• Use of conditional structures and control loops.
• Handling variables, constants, and data types.
• Introduction to using the serial monitor for debugging.

• Intermediate Level – Sensing and Basic Control:

• Reading physical environment variables through sensors.
• Implementation of conditional logic based on analog and digital inputs.
• Use of control structures to automate processes.
• Generation of visual, sound, or mechanical responses to events.
• Display of local data (screens or indicators).
• Use of interrupts for efficient event handling.
• Introduction to temporary data storage.
• Design of programs with basic proportional control.

• Advanced Level:

• Implementation of serial communication between devices.
• Transmission and reception of data for monitoring or remote control.
• Design of autonomous systems with decision-making.
• Reading and interpreting industrial signals (0 ÷ 10 V, 4 ÷ 20 mA).
• Design of control sequences for automated processes.
• Modularization of code and reuse of functions.
• Process control using timers and logical counters.
• Integration of multiple subsystems into a functional project.

• Intermediate Level – Sensing and Basic Control:

• Reading physical environment variables through sensors.
• Implementation of conditional logic based on analog and digital inputs.
• Use of control structures to automate processes.
• Generation of visual, sound, or mechanical responses to events.
• Display of local data (screens or indicators).
• Use of interrupts for efficient event handling.
• Introduction to temporary data storage.
• Design of programs with basic proportional control.

• Advanced Level:

• Implementation of serial communication between devices.
• Transmission and reception of data for monitoring or remote control.
• Design of autonomous systems with decision-making.
• Reading and interpreting industrial signals (0 ÷ 10 V, 4 ÷ 20 mA).
• Design of control sequences for automated processes.
• Modularization of code and reuse of functions.
• Process control using timers and logical counters.
• Integration of multiple subsystems into a functional project.

INTERNET OF THINGS (IOT) MODULE - (NodeMCU)

• Basic Level – Fundamentals of IoT:

• Introduction to the Internet of Things concept.
• General architecture of an IoT system.
• Communication between local devices.
• Configuration of wireless networks.
• Sending and receiving data via Wi-Fi.
• Publishing data to local interfaces (serial monitor or console).
• Use of basic protocols for data transmission (HTTP).
• Programming periodic routines for data capture and sending.

• Intermediate Level – Interaction with Remote Services:

• Client-server structure in IoT.
• Sending data to remote databases.
• Displaying variables on web interfaces.
• Introduction to lightweight messaging protocols (Socket).
• Use of graphical interfaces for real-time monitoring.
• Storage and retrieval of historical records.
• Configuration of dashboards for remote visualization.
• Error handling and automatic network reconnection.

• Advanced Level – Distributed Automation and Decision Making:

• Implementation of cloud logic and remote control.
• Receiving commands from web or mobile interfaces.
• Event-driven programming for real-time response.
• Security in IoT systems (validation, basic encryption).
• Synchronization of states between devices.
• Development of integrative applications: monitoring, alert, automatic response.

AI DEVELOPMENT MODULE - (Jetson NANO NVIDIA)

• Basic Level – Fundamentals of AI and Data Science:

• Difference between symbolic AI and machine learning.
• Introduction to supervised and unsupervised learning.
• Python fundamentals applied to AI (syntax, structures, libraries).
• Data representation: vectors, matrices, and normalization.
• Reading data from sensors or local files.
• Data visualization and exploratory data analysis (EDA).
• Building simple classification models with labelled data.

• Intermediate Level – Applied Machine Learning:

• Architecture of basic artificial neural networks.
• Preparation of custom datasets (physical sensors, CSV, images).
• Training, validation, and testing of models.
• Pattern detection in physical data (temperature, humidity, light, pressure).
• Application of regression and classification models with multiple inputs
• Introduction to specialized libraries: TensorFlow, scikit-learn, Keras.
• Exporting and inference of trained models on embedded systems.
• Logging and analyzing prediction results for decision-making.

• Advanced Level – Real-time AI and Physical Integration:

• Real-time image processing and computer vision.
• Object detection and classification using pre-trained models.
• Integration of AI with physical data acquisition systems.
• Model optimization for fast inference on embedded systems.
• Implementation of intelligent alerts for anomalous events.
• Design of autonomous AI-based solutions (quality, safety, maintenance).
• Intelligent decision-making in physical environments (IoT + AI).
• System performance evaluation: metrics, errors, accuracy, and continuous improvement.

SOFTWARE APPLICATIONS

INCLUDED SOFTWARE

• MySQL Database

• Logging and querying of sensed variables, events, and AI decisions.

• Database Cloning

• Students working in parallel.
• Provides knowledge of database administration.

• Block-based Programming

• Control and programming in a block-based environment to acquire knowledge in programming.

• Web Services

• Integration with external systems, dashboards, or remote AI.

MAIN BASIC CONFIGURATION (20 students)*

• 1 x DL AI-BASE – AI Main configuration (Hardware system, Server, Software’s, and Applications) (for 3 students).
• 17 x DL AI- BASESW. (for 17 students)

EXAMPLES FOR OTHER CONFIGURATIONS*

If the laboratory includes less than 20 students (i.e. 15 students), it will be necessary to have:

• 1 x DL AI- BASE. (for 3 students)
• 12 x DL AI- BASESW. (for 12 students)

If it includes 30 students, it will be necessary to have:

• 2 x DL AI- BASE. (for 6 students)
• 24 x DL AI- BASESW. (for 24 students)

If it includes 40 students, it will be necessary to have:

• 2 x DL AI- BASE. (for 6 students)
• 34 x DL AI- BASESW. (for 34 students)

If it includes 50 students, it will be necessary to have:

• 3 x DL AI- BASE. (for 9 students)
• 41 x DL AI- BASESW. (for 41 students)

If it includes 55 students, it will be necessary to have:

• 3 x DL AI- BASE. (for 9 students)
• 46 x DL AI- BASESW. (for 46 students)

If it includes 60 students, it will be necessary to have:

• 3 x DL AI- BASE. (for 9 students)
• 51 x DL AI- BASESW. (for 51 students)

*: even the three students who work with the HW need their own PC, and DL AI-BASESW is a single license for the expansion of the basic hardware configuration.