Recent years have seen a rapid development of AI-based IoT systems and applications like real-time object detection systems, face identification, AI-based cameras, etc. Many of these utilize a compact-form computing device based on SoC embedded modules from different providers like Qualcomm, Renesas, Quectel, etc. One of the most popular and easy-to-access platforms both practically and academically is the Raspberry Pi computing device and other PI-like devices such as: Orange Pi, Banana Pi, Jetson Nano, etc. However, we realize that the existing platforms strongly focus on the computation aspect, while other important features such as communication, location, ease of integration with high-level operating systems are lacking or poorly supported (through the use of so-called HAT). Therefore, in this work we propose a design of a multi-purpose IoT platform with full integration of computation module, IO ports, communications, location and high-level operating system (Android). We also utilize this IoT platform in a variety of applications including data collection, location-based services and biometric identification. Those sample applications will be presented in our work together with the algorithm and testing result to demonstrate the practical use of our platform.
