An inertial navigation system has been developed to obtain information on the orientation and position of an unmanned surface vehicle (USV). The prototype consists of a GPS and low-cost microelectromechanical sensors, such as accelerometers, gyroscopes and magnetometers. In order to determine the attitude and heading, a sensor fusion algorithm is proposed using the standard Kalman filter, which runs on the MSP432P401R microcontroller with ARM Cortex-M4F 32-bit architecture. The experimental results in the laboratory and at sea showed the viability of the system.