UNIHIKER K10 AI Learning Board Features ESP32-S3 Xtensa LX7 for TinyML, Image Detection, and Voice Recognition

DFRobot has introduced the UNIHIKER K10, a cost-effective AI learning device based on the ESP32-S3 development board, designed for STEM education. It integrates on-device machine learning and artificial intelligence with vector extensions for tasks like image detection and voice recognition.

The device is powered by an ESP32-S3 Xtensa LX7 microcontroller with 512KB SRAM and 16MB flash storage. It supports Wi-Fi 2.4G and Bluetooth 5.0 and includes a 2.8-inch 240 x 320 color display, a 2MP camera, a 2-microphone array, a speaker, and sensors for temperature, humidity, light, and acceleration. Additionally, it features an edge connector similar to the BBC Micro:bit for power signals and GPIOs, along with microSD card support.

Compared to the Linux-based UNIHIKER M10, which uses the Rockchip RK3308, the UNIHIKER K10 is a more affordable alternative optimized for IoT applications. It enables long-term data collection and remote monitoring, making it suitable for embedded AI projects and educational use.

UNIHIKER K10 specifications:

• Core Module: ESP32-S3-WROOM-1
  • Processor: ESP32-S3N16R8 dual-core Tensilica LX7 @ up to 240 MHz with 512KB SRAM, 8MB PSRAM, and 16MB flash
  • Wireless Connectivity:
    • Wi-Fi 4
    • Bluetooth 5.0 / BLE
    • PCB antenna
• Storage: MicroSD card slot
• Display: 2.8-inch 240 x 320 color LCD (ILI9341 driver)
• Camera: 2MP camera with 80°FOV (GC2145 sensor)
• Audio: 2W speaker and 2x MEMS microphones
• USB: 1x USB-C port for power and programming
• Sensors:
  • AHT20 temperature (-40 to 85°C) and humidity (0 to 100% RH) sensor
  • LTR303ALS ambient light sensor (up to 64k Lux)
  • SC7A20H accelerometer
  • 2x MEMS microphones
• Expansion:
  • 3-pin Gravity connector (2x GPIO)
  • 4-pin Gravity I2C connector
  • Edge connector:
    • 3x GPIO (banana clip)
    • 15x Digital IO
    • 1x I2C
    • 3.3V and GND
• Misc.:
  • A, B, RST, and BOOT buttons
  • 3x WS2812 RGB LEDs
• Power Supply:
  • 5V via USB Type-C
  • 3.3V via edge connector
  • 2-pin battery connector
• Dimensions: 51.6mm x 83mm x 11mm

The UNIHIKER K10 comes with four pre-installed AI vision and audio models, including face detection, image recognition, cat/dog detection, QR code recognition, motion detection, offline speech recognition, and custom voice commands. It supports both graphical programming and MicroPython, with TinyML capabilities for training custom vision and sensor models.

The device is available for purchase on DFRobot for $28.90 plus shipping. The package includes the UNIHIKER K10 and a Type-C USB cable. A limited-time bundle featuring both the K10 and M10 models is available for $79.90 until April 15. Documentation, code samples, and setup resources can be accessed on the official wiki page.

Please follow and like us: