Dynamic model based integrated navigation for a small and low cost autonomous surface/underwater vehicle

The civil operation in narrow waters puts forward the requirements of small volume, light weight, flexible movement and low cost for vehicles. A new small autonomous surface/underwater vehicle (S-ASUV) with six thrusters is investigated in this study. To realize the reliable navigation of S-ASUV under the limitation of size and cost, a set of model-aided surface/underwater integrated navigation system is presented. For surface navigation, a measurement adaptive Kalman filter combining the output of GPS and the vehicle's dynamic model is proposed to address the problem that the GPS receiver may lose signal or produce large positioning deviation due to terrain or other interference factors. While GPS cannot be used in underwater environment and the single inertial navigation system (INS) leads to a growing navigation error with time, a new dynamic model-aided underwater integrated navigation is studied. In particular, the coefficients of the thrust model are corrected on line by S-ASUV surfacing. The field experiments were carried out to testify the dynamic model-aided integrated navigation by using the S-ASUV prototype. Especially, the adaptive Kalman navigation for GPS outliers was tested by using the track data collected from the field experiment. The results verified the effectiveness and practicability of the proposed method.