This study presents a motion control system for a coaxial tilt-rotor (CTR) unmanned aerial vehicle (UAV) equipped with two CTR modules and a tail rotor. The existing adaptive control strategies for CTRUAVs fail to guarantee the theoretical convergence of estimated parameters to their true values. Additionally, the existing mixer requires frequent and inefficient adjustments of the tilt angles for motion control. To address these issues, this work proposes a control strategy that integrates a robust integral of the sign of the error (RISE)-based immersion and invariance (I&I) adaptive controller with segmented gains and an improved mixer. The RISE-based adaptive controller is theoretically capable of estimating and compensating for external disturbance torques and forces with bounded derivatives. Furthermore, a model of the CTR module that accounts for differences between the upper and lower rotors is introduced, and the proposed mixer is designed to realize efficient control at varying tilt angles of the CTR modules. Experimental results demonstrate that the proposed control scheme significantly improves stability, transient response speed, disturbance rejection performance, and parameter estimation accuracy compared to existing control strategies for the CTRUAV.