Reinforcement learning with function approximation for cooperative navigation tasks