1. Home
  2. Electronic & Communication Trainers
  3. Embedded Trainers
  4. Edge Device Development Board (VPL-EDGE)

Edge Device Development Board (VPL-EDGE)

Edge Device Development Board • Processor: ARM Cortex-M4 32-bit RISC core optimized for low-power, real-time operation • Clock Speed: 80 MHz • Memory: Internal RAM 128 KB or better, Internal Flash 1 MB or more

SKU: VPL-EDGE Category:

Hardware Overview

  • Processor: ARM Cortex-M4 32-bit RISC core optimized for low-power, real-time operation
  • Clock Speed: 80 MHz
  • Memory: Internal RAM 128 KB or better, Internal Flash 1 MB or more
  • Connectivity:
    • ARDUINO® Uno V3 Compatibility:
      • Digital I/O
      • I2C
      • SPI
      • UART
      • Analog I/O
      • PWM
  • On-Board Capabilities:
    • Debugger/Programmer: On-board debugger/programmer with USB re-enumeration capability (mass storage, Virtual COM port, debug port)
    • User Input/Output:
    • User programmable LEDs and button
    • MCU current measurement point
    • USB Power Management: Efficient USB power management for easy integration

Key Concepts – Edge AI Board:

  1. Edge AI: The processor is designed to handle complex AI algorithms at the edge, enabling real-time processing without cloud dependency.
  2. Low-Power Operation: The ARM Cortex-M4 core is optimized for low-power applications, making it ideal for battery-operated edge devices.
  3. Connectivity and Expansion: With ARDUINO® Uno V3 support, the board is equipped for a wide range of peripherals and connectivity options, including I2C, SPI, UART, and PWM.
  4. Real-Time Processing: The high clock speed and internal memory ensure efficient handling of AI models and data processing tasks in real-time.
  5. USB Management: USB power management and re-enumeration enhance flexibility in device connection and data handling.

Experiment List

  1. Basic STM32 Programming on In-Built LED
    • Objective: Learn the basics of STM32 programming by toggling an onboard LED.
    • Key Concepts: STM32CubeIDE setup, GPIO configuration, LED control, delay functions.
  1. Toggling a LED using a USR Button in STM32CubeIDE
    • Objective: Configure and program the onboard user button to control an LED.
    • Key Concepts: GPIO input configuration, button state detection, LED toggling.
  1. Toggling a LED using Interrupt in STM32CubeIDE
    • Objective: Implement interrupt-based input handling for the user button.
    • Key Concepts: External interrupts, interrupt service routines (ISR), event-driven programming.
  1. Serial Communication Protocol (UART)
    • Objective: Implement UART communication for data transfer between the Edge AI board and a PC.
    • Key Concepts: UART configuration, data transmission, serial debugging.
  1. Serial Communication Protocol (UART) with Printf
    • Objective: Use printf statements to send formatted data from the Edge AI board to a PC via UART.
    • Key Concepts: UART data formatting, debugging techniques, serial communication monitoring.
  1. Sensor Data Logging Methodology (Method 1)
    • Objective: Design a data logging methodology for acquiring sensor data.
    • Key Concepts: Sensor initialization, data logging, and storage.
  1. Sensor Data Logging Methodology (Method 2)
    • Objective: Implement an alternative data logging method for acquiring sensor data and optimizing it for AI training.
    • Key Concepts: Real-time sensor data processing, data optimization.
  1. Running a Data Logger Code and Building an AI Model
    • Objective: Collect data and train an AI model for classification.
    • Key Concepts: Data collection, ML model training, real-time classification.
  1. Image Classification with Pre-Trained Models
    • Objective: Classify images using pre-trained deep learning models.
    • Key Concepts: Image classification, transfer learning, model evaluation.
  1. Real-Time Emotion Recognition
    • Objective: Recognize and classify emotions from facial expressions using the camera module.
    • Key Concepts: Emotion classification, facial feature extraction, AI model deployment.
  1. Audio Scene Classification Using Machine Learning
    • Objective: Implement a machine learning model to classify various sounds.
    • Key Concepts: Feature extraction, sound classification, machine learning model.
  1. Gesture Recognition Using Motion Sensor Data
    • Objective: Train and deploy a gesture recognition model based on motion sensor data.
    • Key Concepts: Gesture classification, motion sensor data, TensorFlow Lite.
  1. Vibration-Based Predictive Maintenance Using Machine Learning
    • Objective: Develop and deploy an ML model to detect and classify abnormal vibration patterns.
    • Key Concepts: Anomaly detection, predictive maintenance, AI-powered fault detection.
  1. 1-Class Model for Ultrasonic Sensor Data
    • Objective: Use NanoEdge AI to detect irregularities based on ultrasonic sensor data.
    • Key Concepts: One-class classification, ultrasonic sensor data, anomaly detection.

 Platform and Workstation Details

Training Environment Setup:

  • Display: 4 Inch capacitive touch LCD for user interaction with the Edge AI system.
  • Connectivity: USB OTG, I2C, SPI, UART for various external device interfaces.
  • Power Supply: Powered via USB or external sources for stable operation.
  • Enclosure: Plastic enclosure for durability and protection in industrial environments.