Abstract: Aimed at the attitude tracking control problem of spacecraft with uncertain parameters, a datadriven composite adaptive finitetime control method was proposed. Through construction of the filtered dynamics of the spacecraft, the unknown inertial parameters were obtained from the attitude angle and angular velocity signals. To relax the constraints of the persistent excitation condition required for achieving precise estimation, a learning term was introduced into the parameter estimation, to form a composite adaptive parameter estimator. This estimator can achieve precise estimation of unknown parameters within finite time under relaxed excitation conditions. Additionally, a finitetime stable controller was designed, and the effectiveness of the proposed composite adaptive finitetime control method was verified by the stability analysis of the closedloop system. Numerical comparative simulation results demonstrate the effectiveness and superiority of the proposed method.