US3517447A - Optical-reimaging system - Google Patents

Description

June 30, 1970y P. L. Fox

OPTICAL-@IMAGING SYSTEM 3 Sheets-Sheet 1 Filed Jan. 5, 1968 INVENTOR. Haz L. Fax BY g/ N. 66ml rrmA/IJG June 30, 1970 P. l.. Fox 3,517,447

OPTICAL-REIMAGING SYSTEM Filed Jan. 5. 1968 5 Sheets-Sheet 2 INVENTOR.

794m z. Fax

g BY @7W/Mg ya@ Ni GALL Arranz/2K5' June 30, 1970 P. l.. Fox

OPTICAL-REIMAGING SYSTEM 3 Sheets-Sheet S Filed Jan. 5, 1968 www WWW; L@ u m n United States Patent O ABSTRACT OF THE DISCLOSURE Y An optical-reimaging system for converting the image n from a rotating disk target of a target and ground speed simulator into a rectilinearly moving target having a pair of angled reflectors positioned between the target and a collimator. One of the angled refiectors having a concave cylindrical configuration while the other reflector is a compound aspheric.

BACKGROUND OF THE INVENTION This invention relates generally to target and ground speed simulators and, more particularly, to an opticalreimaging system used in conjunction with Such a simulator.

The infrared target and ground speed simulator provides test targets for the dynamic evaluation of parameters that establish the performance characteristics of IR reconnaissance sensors or cameras.

This invention provides a means for exposing a high speed moving target to certain types of infrared reconnaissance cameras for the purpose of testing the same cameras. The target image in this application must be presented as an object located at a great distance (optical infinity) and for this purpose a collimator is used.

Collimators can be designed in a wide variety of configurations and may exhibit diverse optical properties, but the choice narrows quickly to an all-reflective type of system when the Simulator requirements are applied. This choice satisfies the requirement for simulating targets in the broad spectral region from l to microns. ln addition, the all-refiective collimator has the advantage of causing no chromatic aberration; with the use of mirrors, energy at all wavelengths is focused at the same point. Furthermore, the reflective efiiciency of metallic mirror surfaces is very high throughout the infrared region. Once the collimator parameters are established, the simulator-system concept becomes totally defined upon the selection of target types, sources and methods of display and control.

Essentially, this target should move across the field in a straight line. Also, it is required to be continuous, such as might be obtained with a moving belt carrying the target configuration.

The following concepts and arrangements of moving target systems have been heretofore considered: (1) continuous belt, (2) open-end rotating drum With target pattern on periphery, (3) rotating disk with slots and bars in radial configuration, (4) rotating disk with optical corrector for radial slots and bars, (5) linear moving slides (circulating), and (6) rotating carrier with mechanically oriented slots and bars.

The use of a target motion simulating a ground speed (v./h.) target of 7 radians/sec. and a 24-ft. focal length from the collimator yields a requirement to provide a linear target speed of 158 ft./sec. A 20-in. diameter drum or disk will revolve at 158/ (10/ 12) 189.6 radians/ sec. [or rev./sec. (r.p.s.) or 1490 rev./min. (r.p.m.)] and generate a centrifugal force of 932 g.

This consideration alone rules out the use of concepts (1), (2) and (6) listed above. In addition, the continuous 3,517,447 Patented June 30, 1970 "ice belt cannot be implemented because no strong fiexible material exists that will transmit infrared radiation. Openslot construction is not practical, because the slot spacing could not be maintained within the accuracy limits dictated by the resolution requirements.

The rotating drum with peripheral slots and bars is rejected because these bars are subject to distortion and probable fracture under the 932 g centrifugal acceleration. Furthermore, the drum curvature is critically close to the limit of field curvature, which will cause defocusing to the maximum permissible resolution. Any other contributions to loss of resolution will therefore put the system beyond the specified limits.

The centrifugal force of 932 g is sufficient to render unfeasible second mechanical motion, such as required for-the functioning of concept (6). In addition, the bars are subject to deformation.

For concept (5), it has been possible to conceive an arrangement of plates that match each other in a row to make a continuous train that (after passing the target exposure area) can be displaced laterally, decelerated, recirculated to the front end, and again accelerated to the running speed. Accelerating motions can be designed to avoid slot and bar deflections.

In review, however, this system would be prohibitively complex mechanically and would probably cause undue vibration in the whole target cart. Only a rotating disk appears feasible for displaying a moving target. If this is used directly, however, it has two drawbacks: (1) the path of the target motion is curved, and (2) the resolution slots and bars are radially tapered and converging.

Because an airplane or other moving vehicle that would normally carry the sensor will be moving substantially in a straight line when the sensor is in use, a rotating-disk target simulation would not be a true analog of the real moving-target condition without artificial conversion.

To more closely simulate the real condition, in accordance with this invention, an optical-reimaging or optical correction system was conceived that will present a rectilinearly moving and parallel-bar-spaced target to the collimator. This arrangement is capable of satisfying all the criteria of fidelity that may be presented.

SUMMARY OF THE INVENTION The preferred configuration comprises a moving target disk designed for the kind of measurement or tests on IR sensors or cameras desired and used in conjunction with the optical correction system of this invention, thus presenting to the collimator a rectilinearly moving target with parallel bars and slots. The optical corrector of this invention will not deteriorate the resolution of the system beyond the 10 limit set previously for the overall system. The corrector establishes an accurately controlled background-radiation level and is also intended for use in conjunction with the hole-pattern target during sensitivity tests.

The corrector optics and target-disk mount are arranged to permit the disk to be used optionally with or without the corrector so that the target may be presented directly to the collimator, if so desired.

It is therefore the primary object of this invention to provide an optical-reimaging system which will reimage a target such that it appears to a camera or sensor to be of rectilinear configuration and moving across the field in a straight line.

It is a further object of this invention to provide an optical-reimaging system which may be retracted from its operative position of reimaging to an inoperative position wherein the target is viewed directly by the collimator.

It is still a further object of this invention to provide an optical-reimaging system which is economical to produce and which utilizes conventional, currently available components that lend themselves to standard mass production manufacturing techniques.

For a better understanding of the present invention, together with other and further objects thereof, reference is had to the following description taken in connection with the accompanying drawing and its scope will be pointed out in the appended claims.

DESCRIPTION OF THE DRAWINGS In the drawings:

FIG. 1 represents a side elevation, partly in section of a target and ground speed simulator which includes the reimaging system of this invention;

FIG. 2 represents a top view of the target and ground speed simulator of FIG. l wherein the regular lines represent the operative position of the reimaging system of this invention and phantom lines represent the inoperative position thereof;

FIG. 3 represents a schematic view of the reimaging system of this invention showing the direction of travel of the light rays from the image to the collimator;

FIG. 4 shows the direction of the light rays from the image as they reflect from the concave mirror of the reimaging system of this invention; and

FIG. 5 shows the direction of the light rays from the image as they reflect from the convex mirror of the reimaging system of this invention.

DESCRIPTION OF THE PREFERRED EMBODIMENT Referring to FIGS. 1 and 2 of the drawing, the IR target and ground speed simulator includes an optical collimator 11 and enclosure assembly 12, a remote-controlled movable-target-turret assembly 14 providing the required target types and ground speed-simulation mechanism, and auxiliary equipment comprising the operators control console, an air-conditioning unit, a window-heat-l ing assembly, a recessed window assembly, and a bellows assembly (not shown).

The optical collimator includes two beam-deviating plane mirrors 16 and 18 which reflect patterns generated by the target turret assembly 14 into an aspheric, primary collimating mirror 20. The second plane mirror 18 refleets the collimated beam formed by the primary mirror 20 through aperture 40 to the sensor or camera (not shown) for testing. By varying the position of the target turret assembly 14 along the optical axis of the collimator the IR target patterns are made to appear at varying altitudes. The collimating optical elements are housed in an enclosure assembly 12 such as a cylindrical tank-type structure that provides a mounting for the target turret assembly 14 and thermal-enclosure panels (not shown).

The target turret assembly 14 includes the optical reimaging system of this invention.

This reimaging system comprises a pair of angled reflectors such as mirrors 22 and 24 (shown in FIGS. 2-5) having a special contour designed to make the tapered slots (not shown) of the resolution-target plate 26 appear parallel and of uniform width.

Referring now to FIG. 3, mirror 22 is shown to have a concave cylindrical curvature having a radius of approximately 100 in. The maximum deviation from plane is 0.020 in. Mirror 24 is a compound aspheric. The sagittal curvature varies along a central longitudinal-plane intercept with the mirror according to the second derivative of y with respect to x, where y is the deviation from a 10G-in. reference circle for values of x displacement along the central plane (plane of the paper). In the direction normal to the central plane, the sagittal curvature is uniform to the edges of the mirror.

The tangential deviation (y) from a 10G-in. circle (which defines the actual curvature of the mirror along the central plane) can be found by integrating twice and multiplying the resultant deviation values by 75". (The sagittal and tangential curvatures are related approximately by cos2 30=0.75.)

It is found that maximum deviation of the aspheric mirror from plane is 0.070 in.

Certain constraints on the design and function of these optics have been established on the basis of consideration of the resolution requirements of the overall collimator system.

The collimator error on axis is permitted to be 4" of arc. If the collimator is successfully constructed with an error of only 2 of arc, the remaining 2 could be allowed for the reimaging optics. For aspheric optics the requirement would appear formidable, but the error is defined with respect to the extended focal length of the collimator (24 ft.) and the reimaging optics are operating only about 1 ft. from the target-image plane (focal plane). The allowable angular error is thus (24) (2)-=48. It is also expected that mirror 24 will be figured to match mirror 22 after mirror 22 has been finished, so that all the error will be applied to mirror 214. Because mirror 24 is only 81/2 in. from the targetimage plane, the allowable error becomes 68".

Actually, a more meaningful concept of accuracies is obtained if the image errors are stated directly in linear dimensions at the target plane:

(24 ft.) (2) (24) (12 in./radian) (0.00001 radian) :0.002188 in.=0.003 in.

Because the collimator f/No. is 11.5, the combined values of coma, astigmatism, and field curvature can be considered to be limited to (11.5) (0.003)=0.034 in. The resolution lines on the target actually extend in only one direction on the plate, and the focal plane can therefore be selected to bring to zero the error due to astigmatism. In addition, the type of coma generated by the mirrors is such that its effect also will be more when astigmatism is zero, leaving the field-curvature error allowance at 0.034 in.

When sensitivity measurements are made with a holepattern target, the astigmatism and coma errors are present but only the energy out of each hole (not its apparent size) is of primary importance. Therefore, these errors do not significantly affect the measurement.

`On the basis of the foregoing analysis, the required characteristics of the reimaging optics are established as follows: (a) Tapered lines of the target must be reimaged parallel within 0.003 in., (b) barrel distortion is not to exceed 0.003 in., (c) astigmatism is not to exceed 0.030 in., (d) coma is not to exceed 0.1 in., and (e) field curvature is not to exceed 0.030 in.

A first-order design of the reimaging optics was established, using a combination of graphical and mathematical techniques. The system has the following characteristics: (a) lateral distortion 0.010 in.; (b) field curvature 0.030 in., causing 0.003 in. of defocusing; (c) astigmatism 0.060 in. (estimated) (d) coma 0.100 in. (estimated for a tangential image); and (e) longitudinal distortion 15%, which can easily be tolerated because it can be precisely Icompensated in the target-pattem design for this application.

A program for precision computation of the subject aspheric curvature is outlined below. The layout for the central ray of the system is shown in FIG. 3.

In the mirror 22 (see FIG. 4) a set of ten chief rays is selected to trace and locate the tangential-image points relative to a set of rectangular coordinates. Because the mirror is cylindrical, the sagittal-image distance remains unchanged. Only three rays are shown in FIG. 4.

The geometry of the reflection Will determine the direction of the reflected chief rays.

sagittal-image distances are transferred directly to the reflected rays. Tangential-image distances are calculated from the following equation (Southall, Principles and Methods of Geometrical Optics, p. 363):

where S is image distance of the original bundle along the chief ray, S is the image distance of the reected bundle along the chief ray, R is the radius of curvature of the mirror, and a is the angle of incidence of the chief ray. The routines for the computation are readily derived. The result of this initial computation should be a set of coordinates locating the tangentialand sagittal-image points on each of ten chief rays, plus the angles those rays make with the reference axes.

The chief rays from the collimator are initially spreading 1 from one side of the eld to the other; after reflection, the chief rays are converging.

Referring now to FIG. 5, mirror 24 is convex and will tend to cancel the effects produced by the concave mirror 22. In particular, it will substantially restore the original stigmatic image at M. It is desired, however,-that an image (i) shall be produced that is inclinedto the direction of the rays and is distorted so that a normally rectilinear pattern at M will be trapezoidal at i. Mirror 24 must be made aspheric to accomplish this.

The coordinate dimensions originally computed for' the S and T images should be transferred to a new set of axes as shown.

Again, the directions of the chief rays are computed as they are reected from a spherical mirror. These will intersect the established image plane, and the coordinate points of these intersections are computed, as are the image-point distances to the mirror surface. The trigonometric routine for this computation is readily derivable. In the next step, employ S+S 2"? cos where S represents the image distance to T, S the image distance to i, and the incidence angles of the chief rays on mirror 24, Use this equation to compute the radii of curvature of the required aspheric at the assumed point of reflection of the chief rays.

The following equation indicates that for small values of the slope dy/dx, the numerator remains nearly constant at unity. Therefore,

Plot the reciprocals of the radius of curvature and draw a smooth curve through these.

Integrate to obtain the slope of the aspheric contour, dy/dx.

Integrate a second time to obtain y=f(x), or the aspheric contour. This is only an approximate value, because it was derived from reflected directions off the circular arc contour.

It is now necessary to replace the circular contour with the given aspheric and to compute the coordinates of the incidence points of the chief rays, the angles of reflection (making use of dy/dx), and the new intersections of chief rays at the image.

A new set of image distances (S) and incidence angles are substituted in and new radii of curvature are derived. Plot again and integrate to obtain dy/dx.

Cit

Because more accurate values of y are now required, the dy/ dx values should be substituted in the Rc formula:

and a new plot of amy/dx2 should be made. The integration of this plot twice will yield the desired aspheric contour.

If there is a significant difference between this contour and that originally derived, one more iteration of the above procedure may be required. It has been observed in preliminary processing that two iterations will reduce the data to a degree of coincidence that cannot be resolved by the graphical technique. It is therefore concluded that a similar iteration by purely mathematical techniques will bring the data to a satisfactory degree of accuracy.

The sagittal curvature of the aspheric consists of a set of radii that sweep across the mirror from a center anchored in the central plane, but moving along the length of the mirror. The radii vary in length from about 30 in. to innity as the center moves longitudinally in the central plane.

At each cross section of the mirror the radius (Re) is determined from 1 l 2 cos SJFS'" R.

where S is the image distance to S, and S' is the distance to along the chief rays. (Note: This formula is different from the one used for tangential Rc.)

It is finally necessary to check the distortion of the image pattern. All lines oriented originally in the longitudinal direction (parallel to the plane of the paper) must remain straight in the subject image, although they are now converging.

For this purpose, project the sagittal centers of curvature and the S image points as well as the i image points into a at plane normal to the paper and containing the x reference axis.

Because each ray (both incident and reflected portions) lies in a plane that contains the sagittal center of curvature, the S image point, and the intersection of the ray in the mirror (this is the plane of the dihedral), the i image point is also in this plane. Furthermore, each S image point, each corresponding i image point, and the center of curvature lie in a straight line.

Take a set of S image points from the edge of the field (i.e., not in the central plane) and project the corresponding image points as above. These should fall in a straight line. If they do not, adjust the field curvature of the image to compensate; then obtain a new set of sagittal radii of curvature.

It has already been determined that the correction needed will be suiciently small to ensure that the resulting amount of field curvature is within tolerable limits. Tangential curvature of the aspheric needs no further correction.

Coma in the tangential direction should be checked, but does not require compensation because it has already been established that the given aspheric will substantially correct the coma to tolerable limits.

Although the reimaging system of this invention is usually in the operative position (regular lines shown in FIG. 2) the system may be optionally placed in an inoperative position (phantom lines shown in FIG. 2) if so desired wherein the target may be presented directly to the collimator.

MODE OF OPERATION Referring now to FIGS. 1 and 2 in particular, the Operation of the reimaging system is as follows. The background radiation emitting from background source 30 is bounced off the front face of the target plate 26 at an incident (and reflected) angle of about 38. The direct target radiation is also coming through the target plate from another source 32 at a similar incident angle. The two beams mix at the target plane 34, which is also the focal plane of the collimator (see FIG. 2). This mixing is possible because the focal plane of the collimator and reimaging-system combination is inclined relative to the direction of the beam at 5 (38 with the normal plane). This inclination results from a characteristic of the reimaging optics.

The emerging beam then reflects oil mirrors 24 and 22, respectively (shown by the regular lines in FIG. 2).

Referring now to FIG. 1, we can see the beam directed to the optical collimator 11. The beam bounces off plane mirrors 16 and 18, respectively, into the aspheric, primary collirnating mirror 20. The second mirror 18 then reflects the collimated `beam formed by the primary mirror 20 through aperture 40 to the `sensor or camera (not shown) for testing.

The reimaging syste-m of this invention is mounted in such a manner that the target may be presented directly to the collimator if so desired. This is accomplished by retracting reflectors 22 and 24 and repositioning target plate 26 and :sources 30 and 32 in the manner shown by the phantom lines in FIG. 2.

Although the invention has been described with reference to a particular embodiment, it will be understood to those skilled in the art that the invention is capable of a variety of alernative embodiments Within the spirit and scope of the appended claims.

I claim:

1. In a target and ground speed simulator having an optical collimator, an enclosure assembly and a remote controlled movable target turret assembly providing a rotating disk target, the improvement therein comprising an optical reimaging system positioned between and in optical alignment with said rotating disk target and said collimator, said reimaging system comprising a pair of angled reflectors, the first of said angled reflectors being convex and in optical alignment with said rotating disk target whereby the rays from the image of the rotating disk target impinge upon said first angled reflector and are reflected therefrom, said iirst angled reflector being a compound aspheric, the second of said pair of angled reflectors being in optical alignment with said first angled reflector so that the reflected rays from said first --angled reflector are reflected towards said second angled reflector, and said second angled reflector having a concave cylindrical curvature, whereby the optical reimag'ing system converts the curved image from the rotating disk target into 'a rectilinearly moving and parallel bar spaced image for presentation to said collimator.

2. In a target and ground speed simulator as defined in claim 1 wherein said pair of reflectors comprise mirrors.

3. In a target and ground speed simulator as defined in claim 1 wherein said rst reflector has a maximum deviation from plane of approximately 0.070".

4. In a target and ground speed simulator as defined in claim 3 wherein said second reflector has a radius of approximately 100 in. and a maximum deviation from plane of approximately 0.020.

References Cited UNITED STATES PATENTS 3,080,483 3/1963 Jaffe et al. 250-84 EUGENE R. CAPOZIO, Primary Examiner P. V. WILLIAMS, Assistant Examiner U.S. Cl. X.R.

Images