For underwater vehicles to be self-sufficient in an {\em a priori} unknown environment, feedback from the environment through altitude sensors is essential. This paper presents how this feedback can be efficiently achieved for fully actuated holonomic autonomous underwater vehicles (AUVs) in which different simultaneous primary tasks have to concur to the achievement of a mission. These vehicles are submitted to constrained motion due to the operational requirements of the load sensor(s) that is(are) carried for a mission, and are equipped with acoustic range altitude sensors. The approach relies on the task-function concept which is based on the reduction of the control problem to the regulation to null value of a task-function defined by the user. Two ways of taking into account different primary tasks in the determination of the task-function to be regulated to zero are presented~: a minimization of a cost function and a smooth switching between tasks. The ! presented control law appears to be robust against sensory data noise and efficient for the following of an a priori unknown and profile-varying environment