This paper presents a new approach to trajectory optimization for a working sample of Eclipse-II motion simulator via model predictive control scheme. A number of so-called "washout algorithms" have been proposed to produce the realistic sensation without violating both spatial and dynamical constraints on a motion simulator. However, it is well known that the existing algorithms are not only conservative for implementing aggressive motions but also too inadequate for computing trajectories of a motion simulator in real-time. The approach presented in this paper addresses these shortcomings via model predictive control (MPC) scheme. The trajectory design problem is formulated as an optimization problem to minimize a quadratic cost function with linear inequalities by using the primal-dual interior point method. Furthermore, the structures of underlying matrices are exploited to ensure fast implementation of the proposed algorithm. The extensive simulations using the working sample of Eclipse-II motion simulator are carried out with a variety of window horizon lengths and input trajectories, which guarantees the compatible performance of the proposed algorithm.