We present in this paper an improvement of a nonlinear control algorithm based on the Lyapunov analysis and the saturation functions, to realize autonomous navigation of a quadrotor vehicle. The algorithm is analyzed and the convergence of the states is ameliorated, in addition, the stability analysis in closed-loop system is proved with these new conditions. In order to locate the aerial vehicle a new position estimation algorithm is developed using the dead-reckoning technique with the Extended Kalman Filter (EKF). This algorithm is based in the data fusion of the classical sensors; an Inertial Measurement Unit (IMU), an ultrasonic sensor and a vision system. The algorithms are validated onboard in flight tests to realize autonomous navigation of the quadrotor vehicle. The most important results are depicted in some graphs.