RASPBOT-V2 is an open source AI vision car for beginners. It uses a Mecanum wheel chassis, a Raspberry Pi 5 as the main control, and a high-definition wide-angle camera. With the OpenCV image processing library and the Media-Pipe machine learning framework, it can realize color recognition, target tracking, license plate recognition, autonomous driving, face recognition, gesture recognition and other functions. RASPBOT-V2 is equipped with a two-dimensional gimbal, which can control the camera to move flexibly in the horizontal and vertical directions, and can realize AI visual tracking functions. In addition, it also supports the expansion of multiple sensors to achieve a richer perception and interactive experience. RASPBOT-V2 uses Python programming, is equipped with the ROS robot operating system, is equipped with detailed tutorial materials, and is open source. From basic drivers to advanced visual applications, it takes you to easily learn robots and AI visual technology, allowing you to quickly build a robot programming environment and explore the infinite possibilities of AI technology.