In this paper, an adaptive control allocation algorithm is proposed for nonlinear autonomous vehicles. Parameter uncertainty of effectors is considered in the design of allocation algorithm. By designing a nonlinear parameter estimator, it needs not to measure the control moments, i.e., forces and torques. Sensors and corresponding supporting devices to measure and transit the control moments are not needed, thus the overall cost of the control system is reduced, and the hardware system is simplified in implementation. Stability of the overall system is analyzed utilizing the Lyapunov method and sufficient conditions are derived. Digital simulation based on a spacecraft simulator is implemented. The results of the simulation validate the effectiveness of the control scheme.