This paper proposes an admittance control algorithm for lower-limb exoskeletons to assist healthy wearers in energy-efficient stand-to-squat locomotion. A two-degree-of-freedom (DOF) model dedicated to the stand-to-squat locomotion is derived, which takes into account the wearer's torso inclination angle and the vertical displacement of the hip joint. The torso inclination angle is estimated from an inertial measurement unit (IMU) located at the backpack of the exoskeleton through a second-order complementary filter. Then a model-based disturbance observer (DOB) is used to estimate the wearer's joint torques as an indication of the motion intention. The exoskeleton complies with the intention by following the desired joint angular velocity which is generated from the estimated joint torque through an admittance function. Experiments with different admittance parameters are conducted, and the energy consumption is evaluated in terms of the normalized energy consumption index (NECI) proposed in previous works. The results show that the proposed method allows the wearer to conduct the stand-to-squat locomotion freely with tunable energy consumption by setting the admittance parameters appropriately.