The airbag recovery system is a kind of unmanned underwater vehicle (UUV) recovery methods, with the controller playing a crucial role in executing the recovery procedure. This study introduces a novel smart controller for the airbag recovery system, designed for seamless integration with both the recovery system and the UUV platform. The controller is designed in the form of embedded system, capable of autonomously monitoring the operational status of the UUV and triggering the airbag recovery system to elevate the UUV to the surface of the water for recovery upon detecting UUV platform failure. Moreover, the controller can send information to operational staffs via remote communication technologies, aiding in UUV location once the recovery function is activated. Additionally, the controller incorporates a Lithium-ion battery state of charge (SOC) prediction feature based on a long short-term memory (LSTM) network. Which can increase the intelligence and maintainability of the airbag recovery system.