This paper proposes a reference path generation method for upper-limb rehabilitation exoskeletons considering the scapulohumeral rhythms of the shoulder. The developed method is based on Central Nervous System's (CNS) governing rules for coordination of arm motions, and to the best of our knowledge is the first computational model to consider the motion of the inner shoulder in path generation. Existing reference generation methods which utilize computational models such as minimum jerk, minimum torque, etc, are based on the assumption that the shoulder joint does not move, and the origin of the reference frame is defined at the center of the glenohumeral (GH) joint. These computational methods are generally developed for simple point-to-point reaching movements with limited range of motion (RoM) which justifies the assumption of fixed shoulder center. However, most upper limb motions such as Activities of Daily Living (ADL) tasks include larger scale inward and outward reaching motions, during which the center of shoulder joint moves significantly. The proposed motion planning method can be used in upper-limb exoskeletons with 3 Degrees of Freedom (DoF) in shoulder and 1 DoF in elbow which are capable of supporting the motion of the shoulder girdle by moving the center of shoulder joint. The outputs of the proposed model are compared with the natural motion of arm during ADL tasks, recorded via a motion capture system. Comparison of the results show that the proposed model is able to reproduce human ADL motions, and can effectively be used for reference generation. The results of this study also confirm that neglecting the fine manipulations with wrist and fingers, ADL tasks can be modeled as large RoM reaching tasks from the perspective of elbow-shoulder coordination.