Control of the air-handling units (AHU) is required to maintain satisfactory comfort conditions with low energy consumption. In the case of failure in sensor fusion systems of AHUs, full-order observers can be used as the supportive tool to provide an acceptable estimation of state variables. In this paper, a multivariable nonlinear model of the AHU is considered in the presence of uncertainties. The indoor temperature and relative humidity are controlled via manipulation of valve positions of air and cold water flow rates. In proposed hybrid control system, full and minimum-order observers are designed for the estimation of indoor temperature and relative humidity. Also, a regulator for disturbance rejection and a servo system for tracking objectives are presented. Minimum-order observers are designed when one or two states are available. The efficiency of full-order and two minimum-order observers in estimation, regulation and tracking objectives is compared for the nominal and uncertain models. Moreover, it is investigated that how the amount of model uncertainty affects the AHU performance and control inputs. It is generally predicted that when a minimum-order observer with two sensors is used, less control inputs are required. In addition, using full or minimum-order observers guarantees the robust performance of AHUs. (C) 2015 Elsevier B.V. All rights reserved.