<p>This paper proposed an adaptive event-triggered model predictive control scheme based on the error covariance of an extended Kalman filter to reduce the computational burden of model predictive control to apply it effectively to actual satellite formation control. In a realistic operational environment that includes system uncertainties and sensor noise, a state estimation algorithm is essential, and utilizing this, the proposed scheme adaptively adjusted the event-triggering threshold online according to the uncertainty of the state estimation through the error covariance of the Extended Kalman Filter. This approach actively adapted to changes in the system and operational environment with a small parameter tuning effort, while reducing the computational burden. To validate the effectiveness, Monte Carlo simulations were conducted in a realistic environment with deterministic perturbations, system uncertainties, and sensor noise, and it was demonstrated that it showed control performance comparable to conventional Model Predictive Control while reducing the computational burden.</p>