BACKGROUND OF THE INVENTION

In the past, high-performance missile configurations have been constrained almost exclusively to bodies of revolution employing cruciform control. The desire to utilize the higher maneuver-duty-cycle capability (g-sec) of lifting bodies presents the problem of controlling their nonlinear, cross-coupled motions. The unique feature of lifting-body control is the requirement to align one missile-maneuver plane with the desired inertial-maneuver direction. Hence, high-performance roll maneuvers are required that meet such criteria as time response, etc. Cruciform control techniques using arbitrary maneuver directions are obviously not suitable for lifting bodies where the high-lift missile plane must be used for efficient maneuver performance.

Some problems in lifting-body control design are: (1) crosscoupling terms are difficult to suppress and are sometimes destabilizing, (2) nonlinearities make conventional linear design procedures less valid, and (3) to meet response-time criteria, roll-control torque requirements may be quite large. The roll-torque penalty is striking if required roll capability for cruciform and lifting-body control are compared on a configuration suited for cruciform control. For typical lifting bodies, a reduced yaw-control requirement and increased rollmoment arm tend to reduce the total control-force penalty.

The cross-coupling phenomena may be divided into four categories, according to source: (1) geometric, (2) gyroscopic, (3) control induced, and (4) aerodynamic. Categories 1 and 2 are important because of high roll rates. Categories 3 and 4 are configuration dependent. Of particular interest in terms of performance is yaw/roll aerodynamic coupling. Without proper yaw-autopilot maneuver strategy, roll-maneuver time is significantly increased by this phenomenon.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram illustrating the preferred embodiment of the invention; and

FIG. 2 is a diagrammatic showing of the coordinate system of the missile.

DESCRIPTION OF THE PREFERRED EMBODIMENT

The prime function of the control system is to maneuver the vehicle to achieve the commanded acceleration and inertial roll angle, as computed by the airborne guidance system. The secondary function of the control system is to maintain vehicle stability as critical plant parameters vary during transition through the flight environment. These basic functions are preformed by a control system. The control system receives acceleration, roll angular error, and gain commands from the airborne guidance system based on ground guidance acceleration magnitude, roll position, and gain commands.

Because of the elliptical cone shape of the UPSTAGE vehicle (see FIG. 2), the maneuver strategy which best achieves the desired results is to achieve the acceleration command with the high lift (flat) side of the ellipse. This is defined as a pitch plane maneuver. The acceleration vector is steered by rolling the vehicle to align the acceleration vector in the commanded direction. The control system must perform the maneuver in as short a time as possible, minimize overshoot, and keep residual errors in acceleration and angular position at a minimum. The control system incorporates a number of body mounted instruments to obtain essential inertial information. These instruments include a three axis laser angular rate sensor and four accelerometers (not shown). Control mechanism actuation status is also monitored and used by the control system. The instrument and command signals are blended and filtered by the autopilot shaping networks to provide signal conditioning and autopilot stabilization. The resultant control signals provide commands to the control mechanism logic which converts the analog control signals to discrete control mechanism actuation signals.

The bank-to-turn control system differs from cruciform control in that there is no symmetry between the steering axes as there is in cruciform control. Bank-to-turn requires steering with the pitch axis and roll axis while decoupling the steering axes via yaw control. Further, the coupling between pitch and roll can be very strong. Inadequate yaw control quickly lends to unsatisfactory pitch and roll control. Throughout the UPSTAGE program the pitch control system has been viewed as one control problem, and the yaw-roll system as another. That pattern will be adhered to in the following.

To clarify the contrast between cruciform and lifting-body control, it may be explained that in cruciform control, pitch-and yaw controllers are used to steer the vehicle. Often, roll control is employed only to decouple the steering axes. On the other hand in lifting-body control, pitch and roll are the steering axes. The primary function of the yaw axis control is to decouple the steering axes.

The UPSTAGE Control System shown in FIG. 1 consists of three autopilots 4, 5 and 6, each controlling one of the three vehicle axes, pitch, roll or yaw, and sufficient command and control logic to effectively control a lifting body. The control system is specifically designed to overcome the strong aerodynamic and geometric coupling between controlled axes. In general, it is the nature of the coupling to directly resist any change of vehicle conditions, and/or to induce a disturbance in the control of one axis when the state of one or both the remaining axes is changed.

The command and control logic 2 initiates a maneuver in such a way that the aerodynamic coupling moments generated aid the desired maneuver. The desired maneuver can thus be completed in less total time and with less control effort than when the desired maneuver is commanded directly. In addition, roll activity is minimized by command logic 2 which executes either a commanded maneuver to a specified amplitude and direction on the pitch axis oriented to a specified inertial roll angle or a maneuver to the specified amplitude but oppositely sensed on the pitch axis, coupled with a roll maneuver through the supplemental change in roll angle. These two maneuvers result in an identical inertial maneuver. The logic chooses the maneuver requiring the smallest roll angular change.

The commands are transmitted in polar coordinates rather than the conventional Cartesian form. Thus, the command is sent as a maneuver magnitude N_g to be achieved at an inertial roll angle φ_c. Based on the difference between the present vehicle roll attitude φ and φ_c the command and control logic 2 selects the smaller of the two angle (φ_c -φ) or (180° - φ_c + φ) and issues this angle as an error signal φ_e to the roll and yaw autopilots. Based on the angle decision, the command and control logic 2 sets the sign of the command magnitude and transmits this signed command to the pitch autopilot 4.

A yaw command is computed based on present vehicle state, and the roll maneuver to be executed, and is sensed so as to aid the commanded roll maneuver. The coordinate system used in the following discussion is shown in FIG. 2. A complete cycle of operation is as follows. The inertial roll command φ_g is received and compared to the vehicle's present inertial roll attitude φ by the subtractor 1. The roll error magnitude |φ_g - φ| is tested and found to be greater than or less than 90° as the first operation of the command logic 2.

1. If |φ_g - φ| is less than 90°, that is -90° ≦ φ_g - φ < 90°, the command logic 2 sets the roll angle error φ_e to φ_g - φ and P_L to false.

2. If |φ_g - φ| is greater than 90°, that is φ_g - φ < -90° or φ_g - φ ≧90°, the command logic 2 sets the roll angle error φ_e to the supplement of ##EQU1## and sets P_L to true.

The command φ_g, the vehicle inertial angle φ, computed by the attitude reference computer 8, the comparison in the subtractor 1 and the manipulations in the command logic 2 are all handled computationally so that φ_g, φ, and φ_g - φ always lie between -180° and +180°, and φ_e always lies between -90° and +90°. Thus, no ambiguities exist in the manipulations of 1 and 2.

The pitch command N_g is received by the pitch polarity switch 3. N_g is always a positive number. If the comparison of 2 results in P_L being false, the polarity switch 3 sets the pitch command N_Zc equal to N_g. If the comparison of 2 results in P_L being true, the polarity switch 3 sets the pitch command N_Zc equal to -N_g.

The pitch command N_Zc is set to the pitch autopilot 4 and executed. The roll angle error φ_e is sent to the roll autopilot 5 and executed. The above manipulations of the commands assure minimum roll activity since the roll error angle is always less than or equal to 90° in magnitude.

The roll angle error φ_e is also sent to a multiplier 7 where it is multiplied by the vehicle estimated angle of attack α. This product (the output of multiplier 7) is the acceleration command applied to a yaw autopilot 6. In executing the commanded yaw acceleration, a yaw angle-of-side-slip is generated which aids the concurrent rolll maneuver via the strong lifting body yaw-roll aerodynamic coupling. That is, the roll moment aerodynamically induced by the yaw angle-of-side-slip accelerates the vehicle about the roll axis in the desired roll direction, aiding the roll maneuver.

The maneuvers about the three vehicle axes, pitch, yaw, and roll, are executed simultaneously. The computed yaw command varies as the maneuver progresses, returning to zero when the roll angle error φ_e is reduced to zero.

The command logic 2 may take the form of any of the well known logic devices properly programmed in accordance with the information given. The computer 8, multiplier 7, pitch polarity switch 3, and autopilots 4-6 may be the same as any of the well known devices. The vehicle shown in FIG. 2 may be any of the known missiles or vehicles which have a favorite axis along which aerodynamic lift can be generated.