This error is then used to determine the rate of decreasing the control step-size dC. The gradient technique described here is an efficient means of solving nonlinear optimal control problems. The algorithm converges quickly during the early stages of the computation and slows down as it approaches the optimal solution. CONCLUSIONS AND DISCUSSIONS This paper described an efficient attitude control mode for maintaining nominal solar orientation of large solar collectors. It has been proved that in the presence of damping, the quasi-inertial attitude cannot be maintained without compensating for the system's energy loss. To this end an efficient attitude control method, the power-assisted quasi-inertial attitude control system, is proposed. The controller is an open-loop optimal control computer which is operated only when a phase correction is required. Using the parametric curves, the ideal quasi-inertial attitude and attitude rate at any orbit position can be obtained for given spacecraft configurations. These values, along with the sensor-measured information, can be used to determine the drift error which in turn determines whether the controller should be activated. The parametric data are again used to determine the terminal state for given orbital positions. This paper presents only the theoretical aspects of the quasi-inertial attitude control problem. There are many other aspects of the problem yet to be investigated. In the following, some of the important aspects are briefly discussed. In our application, all the quantities in Eqs. 39 through 42 are scalar. U(r) is a selected weighing function. From Eqs. 29, 31, and 37, The quantity e in Fig. 6 is a measure of the validity of the linear approximation; it may be defined as the relative prediction error e}.
RkJQdWJsaXNoZXIy MTU5NjU0Mg==