Attitude and Orbit control

The purpose of this paper is to give a broad introduction to attitude and orbit control of geostationary satellites and, in particular, to discuss the connections between the mission requirements imposed upon a broadcast satellite and the design of the corresponding attitude and orbit control system (AOCS). Thus, while no fundamentally new knowledge is presented here, the main objectives are to highlight system-level interface problems that are rarely discussed in the literature and to show people involved in the planning of future broadcast missions some of the impact of their requirements.

In order to define the class of satellites under discussion and the modes of attitude control suitable for it, it is appropriate to make a brief review of past and present satellites taking the Intelsat family of communications satellites shown in figure 1 as a point of reference.

Intelsats I and II were essentially cylindrical in shape and they were 'spinstabilised'. This means that the orientation in space of only one axis of the body (i.e. the spin axis) is controlled. Provided that the required accuracy for this orientation is not better than 3- to i degree, which was the ease for Intelsats I and II since the antennabeam width was 17 degrees, this leads to very simple on-board attitude control hardware. For Intelsat III, the same attitude control system was employed for the main body, but an increase in communications capability was

obtained by mechanically despinning the antenna with respect to the body. Plainly, this requires an additional control loop for antenna pointing.

The extension of the technique of spin-stabilisation to give more communications capacity (which implies larger antennas and more solar cells) was limited by a fundamental law of physics.

It follows from Newton's laws that the angular momentum of a body is constant unless torques are imposed upon it from outside. Internal torques cannot change the overall momentum, but they can affect the amount of kinetic energy in the body. Specifically, if they dissipate energy (e.g. in heat) then the body will move to the state of minimum energy. This minimum energy state for spin-stabilised satellites is the desired mode of spin only if the inertia about the desired spin axis is greater than the inertia about any other axis. That is, the body has to be roughly disc-shaped or ' oblate '. If any transverse axis has a greater inertia, then the body will br stable in spin only about this axis.

Plainly, adding large despun antennas above the spinning body will contravene this condition, i.e., it will make the body prolate (pencil-shaped), so that it will spinstably about a transverse axis and not about the axis of symmetry. Before Intelsat IV could be accepted, therefore, a solution to this stability problem had to be found. Hughes Aircraft Company found the solution and termed it the Gyrostat (Iorillo 1967) although now it is more generally referred to as the ' dual-spin' configuration.

In this configuration, an energy dissipating device (e.g. a damped pendulum) is placed on the despun part, and this has the effect of stabilising the complete prolate satellite about the spin axis of its rotating section. Hughes exploited this principle very successfully in Intelsat IV and IVA in which 4~ ~ spot beams are employed, and a dualspin configuration was one of the two winners selected by Comsat for Intelsat V in the first technical evaluation.

From the point-of-view of AOCS design, it has proved to be a fine example of high performance with simple hardware. Now to consider the European side, dual-spin configurations received a great deal of attention at the time of the first broadcast satellite proposals in 1969. They were studied in some depth (Brewer et al 1970, 1974) but finally they were eliminated mainly on the basis of lack of growth capability due to the inefficiency of their cylindrical solar arrays. The first European geostationary broadcast satellite was the Franco-German satellite Symphonie shown in figure 2. This satellite is not spinstabilised but rather it has 'three-axis attitude stabilisation', i.e. the orientation in space of all three axes is controlled.

In retrospect it does appear to have some similarity to spin-stabilised configurations (the body-fixed solar arrays and the general disc-shape) but from the control system point-of-view the major step has been made. ESA's first satellite in this field is the Orbital Test Satellite (OTS) illustrated in figure 3. In contrast to Symphonie, the solar arrays rotate to track the Sun while the body-fixed antennas track the Earth. This three-axis stabilised configuration has virtually become the ' classical' pattern for all present communication and broadcast satellites. Certainly, it is being adopted for most current missions including Intelsat V and the Indian communications satellite APPLE. The remainder of this paper is concerned solely with this configuration and refinements thereof.