In this work, we present a self-balancing bike system based on the open-source Robot Operating System (ROS) and the momentum flywheel, and has designed a bicycle control system that can achieve autonomous balance and autonomous driving. This bike uses flywheel to maintain balance and uses lidar to achieve autonomous driving. The drive part control core is TC377 microcontroller. The motor driver is Odriver, which directly control the flywheel motor. The corresponding SLAM algorithms are developed and designed in Linux based on the ROS distributed operating system framework. It collects position information via gyroscopes, commands body balance by rotating the flywheel with Cascade PID controllers and FOC controller, and uses lidar and depth cameras to localize the position and posture of the bicycle body in real-time. It generate an optimal path based on the given target point to automatically identify and avoid surrounding obstacles. And Qt interface is used to adjust parameters and display data.