2

Active Electro-Optical Sensing Approaches

In this chapter the committee discusses mature and fielded active electro-optical (EO) sensing technologies. Emerging or future EO technologies are discussed in Chapter 3.

RANGE MEASUREMENT TECHNIQUES

One of the most widely exploited features of active EO sensing is the ability to measure range very accurately. While some lidars and ladars (such as 2-D imagers) do not make use of accurate range measurement capability, most of them do. Range can be measured using active EO sensors in many fashions. The earliest, and in many ways simplest, method is to emit a short laser pulse and measure the time when it returns. This method is simplest, but it has limitations. This is the method of range measurement used with direct detection ladars.

With a single pulse, there is a trade-off between the precision of a range measurement and the precision of a velocity measurement, since a high-resolution range measurement requires a short pulse, and a high-resolution Doppler velocity measurement requires precise measurement of frequency, which takes time. Range measurement precision for a simple pulsed ladar is based on a convolution of the detector response and the return pulse shape. To have both precise measurement of range and velocity at the same time, a series of laser pulses can be used. The range precision comes from the rise time and width of a single pulse. The velocity precision comes from the pulse trains, assuming phase coherence across the pulses in the measurement.

High range precision does not necessarily require a short pulse. Any waveform that has wide bandwidth can provide high range resolution. In addition to short pulses and an array of short pulses, range can be measured with great precision by using a linear frequency modulated (LFM) chirp or a pseudo random coded waveform with sufficient bandwidth. The range resolution will be approximately the inverse of the bandwidth, multiplied by the speed of light, and then divided by a factor of 2 because the light has to go both ways. A good rule of thumb is that 1 nsec provides one-half foot—or 0.15 m—of range resolution, due to the two-way path.

To measure frequency modulation requires heterodyne detection. One major benefit of LFM as a method of range measure is it enables chirping of the local oscillator (LO) along with the emitted signal. Chirping the LO time delay compared to the linear chirp of the master oscillator is sometimes called stretch processing. The time delay of the linear chirp limits the difference in frequency between the LO and the return signal without limiting the ladar bandwidth associated with range resolution. For example, Bridger Photonics recently published results1 with 3.6 THz LFM, resulting in 50 µm range resolution, but there are no detectors capable of bandwidths even close to that level. Of course, if one only needs range resolution consistent with available detector bandwidth then one does not need this advantage. Even if one has available detectors with sufficient bandwidth, use of stretch processing may save money by allowing use of cheaper, lower bandwidth detectors.

____________________

1 R.R. Reibel et al., 2009, “Ultrabroadband optical chirp linearization for precision length metrology applications,” Optics Letters, 34: 3692.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 25
2 Active Electro-Optical Sensing Approaches In this chapter the committee discusses mature and fielded active electro-optical (EO) sensing technologies. Emerging or future EO technologies are discussed in Chapter 3. RANGE MEASUREMENT TECHNIQUES One of the most widely exploited features of active EO sensing is the ability to measure range very accurately. While some lidars and ladars (such as 2-D imagers) do not make use of accurate range measurement capability, most of them do. Range can be measured using active EO sensors in many fashions. The earliest, and in many ways simplest, method is to emit a short laser pulse and measure the time when it returns. This method is simplest, but it has limitations. This is the method of range measurement used with direct detection ladars. With a single pulse, there is a trade-off between the precision of a range measurement and the precision of a velocity measurement, since a high-resolution range measurement requires a short pulse, and a high-resolution Doppler velocity measurement requires precise measurement of frequency, which takes time. Range measurement precision for a simple pulsed ladar is based on a convolution of the detector response and the return pulse shape. To have both precise measurement of range and velocity at the same time, a series of laser pulses can be used. The range precision comes from the rise time and width of a single pulse. The velocity precision comes from the pulse trains, assuming phase coherence across the pulses in the measurement. High range precision does not necessarily require a short pulse. Any waveform that has wide bandwidth can provide high range resolution. In addition to short pulses and an array of short pulses, range can be measured with great precision by using a linear frequency modulated (LFM) chirp or a pseudo random coded waveform with sufficient bandwidth. The range resolution will be approximately the inverse of the bandwidth, multiplied by the speed of light, and then divided by a factor of 2 because the light has to go both ways. A good rule of thumb is that 1 nsec provides one-half foot—or 0.15 m—of range resolution, due to the two-way path. To measure frequency modulation requires heterodyne detection. One major benefit of LFM as a method of range measure is it enables chirping of the local oscillator (LO) along with the emitted signal. Chirping the LO time delay compared to the linear chirp of the master oscillator is sometimes called stretch processing. The time delay of the linear chirp limits the difference in frequency between the LO and the return signal without limiting the ladar bandwidth associated with range resolution. For example, Bridger Photonics recently published results 1 with 3.6 THz LFM, resulting in 50 μm range resolution, but there are no detectors capable of bandwidths even close to that level. Of course, if one only needs range resolution consistent with available detector bandwidth then one does not need this advantage. Even if one has available detectors with sufficient bandwidth, use of stretch processing may save money by allowing use of cheaper, lower bandwidth detectors. 1 R.R. Reibel et al., 2009, “Ultrabroadband optical chirp linearization for precision length metrology applications,” Optics Letters, 34: 3692. 25

OCR for page 25
26 LASER RADAR Digital holography, or spatial heterodyne (see Chapter 3), can measure range by using more than one wavelength. As the two or more wavelengths walk off from each other in phase, range can be determined. Usually the two wavelengths are separated by large bandwidths, resulting in very fine range resolution, but also very fine unambiguous range. LASER RANGE FINDERS Laser range finders are very simple forms of laser radar. A range finder determines the range to a target object based on the round-trip time-of-flight of a laser pulse to and from the object. Most laser range finders are direct-detection ladars, meaning the return pulse hits the detector and triggers a range measurement. A combination of the return pulse shape and the detector response determines how accurately one can measure the round trip time. An early application of a laser range finder was the determination of the precise distance from the Earth to the Moon. 2 “The first laser ranging retroreflector was positioned on the Moon in 1969 by the Apollo 11 astronauts (see Figure 1-10).” 3 By beaming a 60-Joule ruby laser pulse at the reflector from Earth, scientists were able to determine the round-trip travel time, which gave the distance between the two bodies at any time to a precision of about 3 cm. 4 While the U.S. military pioneered laser range finders, 5 they are now available for construction, surveying, golf, hunting, or many other applications where one wants to know the range to an object. Laser range finders, also called laser rangers, are widely available commercially and cost from $100 up to tens of thousands of dollars. Usually, commercial range finders are direct detection pulsed ladars.6 The main limitation on range finders is the atmosphere. If the laser beam cannot be transmitted to the target, or received from the target, then one cannot use a laser ranger. For long-range measurements, slight variations in the index of refraction of the atmosphere can slow the speed of light, introducing slight inaccuracies in the range measurement. ONE-DIMENSIONAL RANGE PROFILE IMAGING LADAR A pulsed laser and a single detector can provide a range profile from an object with a range resolution that depends on the transmitter pulse rise time, pulse width, or in the case of coherent ladars, the width of the frequency chirp (e.g., how short the laser pulses are). Of course one must match the transmitter pulse with how fast the detector can resolve the pulse, and a wide frequency chirp must be matched by a detector with sufficient bandwidth. The range profile does not depend on the size of the receiving optical apertures, except for signal-to-noise considerations. This can be very useful in environments that are not very cluttered and for objects that are far away, where the transverse dimensions of the object are not resolved by the optical system and the only spatial information about the object comes from the range profile. A high-range-resolution profile is most useful when the orientation of the object is known; for example, one generally assumes that an airplane is oriented so the nose is pointed in the direction that the plane is traveling. Figure 2-1 shows how an airplane’s scattering signal will vary with range, depending on what scatterers are in a given range bin. Starting from the time when the detector is first triggered by 2 P.F. McManamon, M. Huffaker, and G. Kamerman, 2010, “A history of laser radar in the United States,” Proceedings of SPIE, 7684: 76840T. 3 Lunar Planetary Institute, 1994, “Apollo Laser Ranging Experiments Yield Results,” LPI Bulletin 72. http://eclipse.gsfc.nasa.gov/SEhelp/ApolloLaser.html. 4 Ibid. 5 J.T. Correll, 2010, “The emergence of smart bombs,” Air Force Magazine, March: 60. 6 See, for example, patent publication number US5359404 A.

OCR for page 25
ACTIVE ELECTRO-OPTICAL SENSING APPROACHES 27 FIGURE 2-1 Range profile of an airplane viewed in one dimension (1-D) from the front, although the image shown is from the side. SOURCE: Courtesy of Matthew Dierking, Air Force Research Laboratory, Dayton, Ohio. the large scattering signal from the nose, the dimensions of the rest of the airplane can be calculated by the relative time delay. Tomographic Imaging Using Multiple 1-D Profiles from Various Angles Tomographic methods, which reconstruct an image from a set of its projections, are well known in radio astronomy and medical imaging. The goal of reflective tomography is to estimate an object’s surface features using a set of reflective projections (as opposed to transmissive projections). These projections are measured in angular increments around the object to obtain 3-D information. 7 Each angular resolution cell in the projection represents the energy reflected off the corresponding illuminated surface of the object at that angle. For any laser radar, the received signal represents information about the surface of the object illuminated by the radar from a given line of sight. A series of signals along the range resolution coordinate produces a reflective projection of the object that can be used to reconstruct the object. 8 Figure 2-1 shows one such 1-D image. 7 F.K. Knight, S.R. Kulkarni, R.M. Marino, and J.K. Parker, 1989,” Tomographic techniques applied to laser radar reflective measurements,” Lincoln Laboratory Journal 2 (2): 143. 8 X. Jin and R. Levine, 2009, "Bidirectional reflectance distribution function effects in ladar-based reflection tomography," Appl. Opt. 48: 4191-4200.

OCR for page 25
28 LASER RADAR FIGURE 2-2 Two-dimensional projection reconstruction of an F-4 fighter for (a) measurements from 0 to 180 degrees and (b) measurements within 20 degrees around broadside. SOURCE: M.P. Dierking, F. Heitkamp, and L. Barnes, 1998, “High temporal resolution laser radar tomography for long range target identification,” OSA Signal Synthesis & Reconstruction Conference, June. In this technique, a pulse whose physical length is short compared to the range extent of the object is first reflected off an object. The resulting time-dependent return signal is collected by single detector in a 1-D optical system (single-element detector), which provides signal as a function of range. 9 This 1-D signal relates to a 1-D slice of the spatial 3-D profile of the object. 10 When the object rotates, different slices of the 3-D profile can be obtained. 11 If a sufficient number of 1-D slices are collected, a 2- D image can be reconstructed. A scanning ladar can be employed as long as the 2-D images are registered in angle. 12 Figure 2-2 shows a 2-D image, obtained by Air Force researchers in 1998, using range profiles determined at multiple angles. Various algorithm approaches can be used to combine multiple range profiles to make a more complete image from multiple 1-D profile images at different angles to the target. Wider-angle reconstruction yielded better detail, as expected. Range profiles from various angles can be obtained when the object moves with respect to the 1- D ladar. The object may be stationary, as a target on the ground, or moving, as an airplane, or rotating, as a satellite. In each case suitable mathematics has been developed to provide images. The more complex the object, the more interest there is in obtaining range profile information over multiple angles, but even with a single range profile 1-D imaging can be useful for recognizing some objects, as discussed in the previous section. 9 J.B. Lasche, C.L. Matson, S.D. Ford, W.L. Thweatt, K.B. Rowland, V.N. Benham, 2009, “Reflective tomography for imaging satellites: Experimental results,” Proc. SPIE 3815: 178, Digital Image Recovery and Synthesis IV. 10 Ibid. 11 J. B. Lasche, C. L. Matson, S. D. Ford, W. L. Thweatt, K. B. Rowland ; V. N. Benham, 2009, “Reflective tomography for imaging satellites: experimental results,” Proc. SPIE 3815: 178, Digital Image Recovery and Synthesis IV. 12 R.M. Marino, T. Stephens, R.E. Hatch, J.L. McLaughlin, J.G. Mooney, M.E. O’Brian, G.S. Rowe, J.S. Adams, L. Skelly, R.C. Knowlton, S.E. Forman, and W.R. Davis, 2003, “A compact 3-D imaging laser radar system using Geiger-mode APD array,” Proceedings SPIE, 5086: 1.

OCR for page 25
ACTIVE ELECTRO-OPTICAL SENSING APPROACHES 29 When a coherent detection system is employed and data are collected from multiple views, an object’s Doppler spectrum can be used to help determine the angle from which a given 1-D image has been collected. 13 With Doppler techniques, the rotation speed of an object can be precisely measured. The ability of range-resolved reflective tomography to image satellites was demonstrated in 14 2001. The researchers used a CO 2 laser mode-locked to produce nanosecond pulses, and the data were collected using a coherent laser radar system. “Key components of the image reconstruction process included the calculation of tomographic angles and the alignment of the intensity projections to an appropriate center of rotation.” 15 Researchers from the Electronic Engineering Institute of the State Key Laboratories of Pulsed Power Laser Technology and Electronic Restriction in Anhui Province, Hefei, China recently reported their verification by computer simulation of ladar reflective tomography imaging of space objects. 16 This paper concluded that atmospheric turbulence has little effect on the coherent detection signal at 10.6 µm, because the atmospheric coherence length is on the order of meters. They found that chirped pulse signals are advantageous for reflective tomography imaging of space targets and that the “optimal pulse repetition is decided by the target altitude, target size, and the cutoff frequency in the spectrum domain.” 17 Background work on reflective tomography has been ongoing at Key Laboratory of Space Laser Communication and Testing Technology at the Shanghai Institute of Optics and Fine Mechanics in China. 18 Most early work on 1-D ladar, and tomographic imaging with 1-D ladar used low-duty-cycle pulsed ladar sensors. More advanced range measurements techniques can be used, such as linear frequency-modulated (FM) chirp or pseudo-random coded range measurements. This provided another method of achieving both 1-D imaging and tomographic imaging. 19 TWO-DIMENSIONAL ACTIVE/GATED IMAGING Two dimensional (2-D) gated imaging is one of the more straightforward and mature methods of active EO imaging. It is a useful imaging technique for seeing through haze and smoke, for seeing objects in shadow, for imaging at higher angular resolution in darkness and at longer range (approximately 10-20 km) than can be achieved with passive MWIR systems, and for seeing objects more clearly against a cluttered background. “Range-gated active imaging is widely used in night vision, underwater imaging, three-dimensional scene imaging, and other applications because of its ability to suppress backscatter from fog and other obscurants, as well as its high signal-to-noise ratio, high angular resolution, long detection range, and direct visualization.” 20 13 C. Matson and D. Mosley, 2001, “Reflective tomography reconstruction of satellite features? Field Results,” Appl. Opt. 40(14): 2290-2296. 14 C. Matson and D. Mosley, 2001, “Reflective tomography reconstruction of satellite features? Field Results,” Appl. Opt. 40(14): 2290-2296. 15 C. Matson and D. Mosley, 2001, “Reflective tomography reconstruction of satellite features? Field Results,” Appl. Opt. 40(14): 2290-2296. 16 F. Qu, Y. H., and D. Wang, 2011, “Lidar reflective tomography imaging for space object,” Proc. of SPIE 8200: 820015. 17 Ibid 18 X. Jin, J. Sun, Y. Yan, Y. Zhou, and L. Liu, 2012, “Imaging resolution analysis in limited-view laser radar reflective tomography, Optics Communications, 285 (10-11): 2575. 19 J. Murry, J. Triscari, G. Fetzer, R. Epstein, J. Plath, W. Ryder, and N.V. Lieu, 2010, “Tomographic Lidar,” OSA Conference Paper, Applications of Lasers for Sensing and Free Space Communications, San Diego, CA, February. 20 X. Wang, Y. Zhou, S. Fan, Y. Liu, H. Liu, 2009, “Echo broadening effect in the range-gated active imaging technique.” Proc. SPIE 7382:738211, International Symposium on Photoelectronic Detection and Imaging 2009: Laser Sensing and Imaging.

OCR for page 25
30 LASER RADAR 2-D gated imagers are typically used in conjunction with laser illuminators and often share the same pulsed laser. They use the pulsed laser to illuminate a scene, then collect the returning reflected light at a predetermined delay time based on the range of the target. These sensors are gated both to eliminate the reflection from obscurations and to reduce the noise coming into the detector (Figure 2-3). The range information is typically first acquired with the co-located laser illuminator. The gated imager uses this information to set the start of the integration window of a high-speed focal plane receiver to coincide with light returning from the range of interest in the scene. This provides a range slice image each time the scene is illuminated. The scene can be interrogated at a series of ranges by varying the delay in turning on the receiver. An example is shown in Figure 2-4. Collecting imagery using a 2-D gated camera has three primary advantages over passive imaging: 1. For scenes with clutter (such as particles in the air or haze) or obstructions near the camera or clutter in the background behind the area of interest, 2-D gated imaging effectively eliminates the visually confusing effects of clutter by only imaging the slice of the scene of interest. In these cases the object of interest is either not visible or very indistinct when viewed with standard visible or IR cameras. 2. 2-D gated imaging is sometimes referred to as “bring your own flashlight.” It effectively brings out features in shadowed or recessed areas, such as inside doorways or other openings or under overhanging structures. By shining the laser into those areas, it is easy to see what is not otherwise visible. It also works when thermal imaging has limited signal, such as thermal crossover. 3. An active EO eye-safe laser 2-D gated camera operating at 1.55 µm will have approximately 3 times better angular resolution than an MWIR sensor. Higher F-number optics allow the user to make use of the improved diffraction limit resulting from operating at shorter wavelengths to see finer details, perform better target detection and recognition, and to see details that have contrast at the laser wavelength but not in the thermal region. FIGURE 2-3 A gated active EO imaging system can eliminate clutter and obscurants that might otherwise degrade the image. Although laser returns from image clutter, smoke, mist and foreground obstructions can be range-gated, effects of atmospheric turbulence remain. SOURCE: I. Baker, S. Duncan, and J. Copley, 2004, “A Low Noise, Laser-Gated Imaging System for Long Range Target Identification,” Proc. of SPIE, Vol. 5406. Courtesy of Selex ES Ltd.

OCR for page 25
ACTIVE ELECTRO-OPTICAL SENSING APPROACHES 31 FIGURE 2-4 Illustration of how the outline of a man can be extracted from the deep gate by knowledge of the individual pixel ranges. SOURCE: I. Baker et al., 2008, “Advanced infrared detectors for multimode active and passive imaging applications,” Infrared Technology and Applications XXXIV, edited by Bjørn F. Andresen, Gabor F. Fulop, Paul R. Norton, Proc. of SPIE 6940: 69402L. Courtesy of Selex ES Ltd. 2-D gated imagers have processing requirements equivalent to passive imaging systems, which are significantly less than those required for 3-D imaging. An inexpensive 2-D active imager that can form both passive and active images has recently been reported. 21However, moderate spatial resolution 3- D images can be constructed using gated imaging in the case where a large number of images are sequentially collected and integrated.22 2-D gated imagers are currently fielded in the United States and in Europe. Performance characteristics of two types are given in Table 4-4. In addition to U.S. Air Force work in the late 1990s on the ERASER Program, 23 much of the early development work was done in the United Kingdom at BAE Systems and in Denmark. 24 Commercial units are relatively compact and are used on almost all types of platforms, from handheld and tripod-mounted viewers to ground vehicles and ships, and in aircraft. Current providers of component hardware and systems include Intevac (see Figure 2-5), which is working on an export-approved unit, Sensors Unlimited, SELEX (UK), DRS Technologies, and Northrop Grumman. China’s interest is shown in publications referred to above, which are mostly theoretical and do not appear yet to be state of the art. 21 R.H. Vollmerhausen, 2013, “Solid state active/passive night vision imager using continuous-wave laser diodes and silicon focal plane arrays,” Optical Engineering, 52: 043201, April. 22 J. Busck and H. Heiselberg, 2004, “Gated viewing and high-accuracy three-dimensional laser radar,” Applied Optics 43 (24): 4705. 23 P.F. McManamon, M. Huffaker, and G. Kamerman, 2010, op. cit. 24 McManamon et al., op. cit., and Busck and Heiselberg, op. cit.

OCR for page 25
32 LASER RADAR FIGURE 2-5 Intevac Laser Illuminated Viewing and Ranging (LIVAR) camera. SOURCE: Courtesy of Intevac, Inc. The fundamental limit to this technology is atmospheric transmission, which limits the range of the sensor. Increasing the laser pulse energy and detector sensitivity and/or reducing the illuminated area can extend the range of the sensor. Atmospheric turbulence is not usually a significant concern as long as the transmission is adequate to close the link budget. The snapshot time duration is short enough that the atmosphere is essentially frozen for each image, and the atmospheric distortion is usually not the limitation in angle/angle resolution. In general the short-pulse, high-energy lasers used for illuminators operate, by their nature, on a number of longitudinal (and often spatial) modes, and thus may be broadband enough in wavelength to eliminate speckle in a single pulse. Techniques such as broadband seeding of the high-energy source by a low-power semiconductor laser may be used to further spoil the laser coherence. If necessary, any remaining speckle can be averaged out using multiple 2-D images. While not a physical fundamental limit, gated imaging requires high per-pulse energy to illuminate the scene (tens to hundreds of mJ) at any useful range. Therefore, the operating wavelength must be in the eye-safer region of the spectrum where optical radiation cannot focus on the retina (see Box 1-4), requiring solid state lasers with high per-pulse energy and detectors in the 1.55-2 µm range, which are more challenging to manufacture than 1 µm or visible lasers. Single fiber lasers cannot achieve this illuminator performance. Intevac cameras have limited life due to plasma etching of internal components, which limits them to low-repetition-rate systems. Over the next decade, the committee expects to see the proliferation of 2-D gated imagers, particularly for marine and airborne systems. Their advantages are relative simplicity, technical maturity, good utility, and the ability to utilize existing laser rangefinder transmitters that are already in targeting systems for added functionality. Improvements in detector sensitivity and laser efficiency will extend the useful range of 2-D active imaging systems. THREE-DIMENSIONAL DIRECT-DETECTION ACTIVE IMAGING Ladar has been used extensively to create accurate and precise high-resolution angle/angle/range- resolution digital elevation models of tactically significant geographic areas. These can provide true 3-D point clouds 25 with a high density of information. The digital terrain elevation data (DTED) point spacing depends practically on the position of the sensor: 1-10 m angle/angle resolution for airborne laser 25 A point cloud is a set of data points in some coordinate system, often intended to represent the external surface of an object.

OCR for page 25
ACTIVE ELECTRO-OPTICAL SENSING APPROACHES 33 scanning (ALS), and millimeter to centimeter angle/angle resolution for terrestrial laser scanning (TLS). Helicopter-based ALS can give a higher angle/angle resolution than aircraft-based ALS and allows orienting the scanner in all directions. A recent book 26 provides a comprehensive compilation of the principles and fundamentals of ladar technologies as well as the state-of-the-art of the performance aspects of the sensors (components, calibration, waveform analysis, quality control of ladar data, filtering and feature extraction techniques). In most references the term lidar has been used for 3-D mapping, but since the convention in this study is to use the word ladar for active EO imaging of surface reflecting targets, the committee will use ladar for 3-D mapping. Scanning 3-D Ladar A scanned laser radar is one technology for collecting 3-D images. Figure 2-6 is a block diagram of a typical bistatic, 27 direct-detection scanned ladar. A laser generates an optical pulse that is shaped and expanded to reduce its divergence. It is then directed toward the scene to be interrogated by a scanner. The backscattered light is directed into a collector (e.g., a telescope), where it is focused onto a photosensitive element. The resulting electronic signal is filtered to remove noise and analyzed to determine the time of arrival of the reflected optical signal. The distance to the scene at the point where the transmitted light was reflected is determined by the round-trip time of the radiation. The scanner repositions the interrogation point and the process is repeated. FIGURE 2-6 Block diagram of a typical bistatic, scanned 3-D imaging laser radar. 26 J. Shan and C.K. Toth, eds., 2008, Topographic laser ranging and scanning: principles and processing, CRC Press. 27 Bistatic means that separate apertures are used to transmit and receive the optical radiation.

OCR for page 25
34 LASER RADAR FIGURE 2-7 One scanning 3-D ladar scan pattern. If the scanner follows a raster or line pattern, as shown in Figure 2-6, then the data will resemble a television image. Here, however, instead of the data consisting of two angles and intensity, it is composed of two angles and range. It is a representation of the geometric shape of the scene portrayed in spherical coordinates originating at the sensor. By trigonometric coordinate transformations, these data can be converted into another format such as Cartesian coordinates. If the location and orientation of the sensor are known, then the Cartesian representation closely resembles a high-resolution digital topographic map. Such “flying spot” scanned systems are currently the rule for commercial airborne topographic mapping as well as ground-based (tripod) systems. These commonly use a line scan in one dimension, typically across the aircraft flight path. The physical motion of the aircraft itself accomplishes scanning along the flight path (Figure 2-7). A swath is then mapped. Subsequently, overlapping swaths are stitched together to map wider areas, as shown in the figure. Frequently, the ground spatial resolution or ground sampling distance in flying spot systems is limited by the transmitter spot size at the target, and not by the receiver optics. The instantaneous field of view (IFOV) of the receiver optics is often designed to be larger than the transmitter spot size in order to simplify transmitter to receiver alignment. This is in contrast to the focal plane systems to be described below. A typical commercial airborne topographic mapping system is shown in Figure 2-8. This class of instrument typically uses only a single photosensitive element to receive the reflected radiation. The acquisition cost of sensors that employ this “one pulse-one pixel” methodology is lower than sensors with multiple detector elements but the methodology limits the data collection rate, which in turn limits the area coverage rate at the desired ground sampling distance. The commercial collection strategy requires multiple passes and additional flight time to cover a given collection area. This is acceptable for commercial topographic mapping systems but may not be acceptable for military

OCR for page 25
ACTIVE ELECTRO-OPTICAL SENSING APPROACHES 35 FIGURE 2-8 RIEGL LMS-Q780 Airborne Laser Scanner. SOURCE: Courtesy of Riegl USA. applications. Area coverage rate is a key attribute of military sensors as multiple passes are frequently not permissible. Multiple passes are not practical in a hostile airspace or where this collection strategy might alert opposing forces. Commercial flying spot scanners, or airborne line scanners, usually operate at lower altitudes than do military systems, which must operate at higher altitudes in order to avoid hostile fire. Operating at lower altitudes permits the use of lower sensitivity detectors, smaller aperture sizes, and lower power lasers. Overall, this reduces the acquisition cost of the laser radar and limits eye-safety concerns. This lower cost is highly desirable in a commercial environment but these commercial sensors cannot be applied to many military applications as they require altitudes lower than the effective altitude of many common air defense threats. Commercial, ground-based survey laser radar systems generally use a second, orthogonal scan element instead of gross sensor motion to generate a raster. The area coverage rate of these ground-based laser radar systems is also limited by the one-pulse-one-pixel collection strategy. It is notable that even though flying spot scanned 3-D imaging laser radars were first demonstrated in the United States, U.S. industries have not taken advantage of this early technological advantage. The majority of commercial airborne mapping and all ground-based surveying laser radars are currently produced outside the United States. Based on total operating units, the commercial market is currently dominated by Optech (Canada), Riegl (Austria), and Leica Geosystems (overall headquarters in Switzerland).28 3-D Flash Imaging As illustrated in Figure 2-9, flash imaging is achieved by flood-illuminating a target scene or a portion of a target scene. The receiver collects the backscattered light and directs an image of the target 28 M.S. Renslow, ed., 2012, Manual of Topographic Lidar, American Society for Photogrammetry and Remote Sensing, 34.

OCR for page 25
96 LASER RADAR FIGURE 2-54 Schematic of coherent-detection wind-sensing (Doppler) lidar based on aerosol scattering. accurately measure both position and velocity, such as using a series of laser pulses, called a pulse doublet or poly pulse measurement. They have a high time × bandwidth product, as will be discussed below. To measure horizontal wind from a ground-based system, several strategies have been developed. In the velocity-azimuth Doppler (VAD) technique, the beam is directed at an angle from zenith and scanned 360 degrees in a cone. Returns are collected and the velocity components {u, v, w} are calculated. Here, {u, v, w} are the streamwise, cross-stream, and vertical components of the velocity. Another scan strategy is the range-height, or RHI, technique in which the beam is aligned with the average u vector and scanned in a small range of angles to probe the related range of heights downstream. As is evident from the numbers, wind measurements require very-frequency-stable, narrow- linewidth laser sources. For a 1,500-nm laser, the center frequency is 2 × 1014 Hz, and a source with a 1- MHz linewidth must have a purity of 5 × 10-9. Even if one locks the source laser to the local oscillator, the time delay in the returned signal places constraints on how much absolute drift of the operating frequency is allowable. Historically, CO 2 gas lasers were first used to demonstrate coherent wind-sensing lidar systems, due to the relative ease of obtaining, for use in the local oscillator, stable, single-frequency operation from low-pressure, CW-discharge CO 2 lasers, given the narrow gain linewidth for the CO 2 vibrational- rotational transitions and the relatively minor perturbation of the laser resonator by the gas medium. The main challenge was to make stable optical cavities, which was accomplished through the use of good engineering design and the use of low-thermal-expansion materials such as Invar to set the cavity length. Unfortunately, as noted above, the aerosol backscatter level around the 10.6-µm wavelength of CO 2 lasers is low and varies widely with atmospheric conditions, requiring multi-Joule-level pulsed transmitters based on high-pressure, pulsed CO 2 lasers for even modest ranges. Early solid state lasers were not good candidates for coherent lidars, owing to the considerably higher frequency instability from lamp-pumping-induced fluctuations in the optical path length of the laser medium compared to CO 2 lasers. With the development of diode-laser pump sources, it became

OCR for page 25
ACTIVE ELECTRO-OPTICAL SENSING APPROACHES 97 possible to make much more frequency-stable, single-frequency, low-power lasers used for the local- oscillator function, and through development of injection-seeding techniques based on the low-power lasers it became possible to make high-energy sources for use as the transmitter. Although the first solid state systems employed Nd:YAG lasers at 1,064 nm, concern over eye-safety has led to the recent development of sources based on Er-doped solid state lasers around 1,600 nm, and Tm and Ho-doped lasers for use around 2,000 nm. While fiber-format solid state lasers cannot yet generate the high-energy pulses needed for very long-range sensors, they have been employed for shorter-range systems and leverage the development of fiber components used for the telecom industry. One of the most sought-after coherent wind sensors has been a space-based system that would provide global coverage and mapping of all atmospheric winds. Such a system would be of major value to science as well as of significant practical use for weather forecasting and commercial aviation routing, which could take advantage of favorable winds at different altitudes to speed up flights and save on fuel consumption. For military applications, such a system would provide greatly improved theater weather forecasts. It would also help with the determination of improved mission flying routes and support precision reconnaissance instruments and precise missile launch and reentry calculations, as well as improved ballistic wind estimates for conventional munitions. 148 NASA began developing plans for space-based systems that would have required 20-J/pulse CO 2 lasers but in the 1990s moved to development of the 2,000-nm region with Ho-doped solid state lasers and now has system designs that can function from space with 250 mJ of energy. 149 At present, NASA is testing a diode-pumped, 10-Hz, 250-mJ, Tm, Ho:LuLiF 4 solid-state-laser-based system, the Doppler Aerosol Wind Lidar (DAWN) instrument, on aircraft platforms and has generated data over a 10-km vertical range from a DC-8, with a modest 15-cm-aperture receiver. Development of laser technology for space-based missions is in the planning stage. In the next section molecular-scattering-based wind sensing is discussed. This is also being developed for space-based sensors, generating data from atmospheric regions where aerosol densities are low. Ground-based coherent wind sensors have been developed for a number of applications. Following several crashes of airplanes on takeoff or landing approaches, there was interest in fielding sensors to detect wind-shears around airports, particularly so-called “dry” wind shear not associated with rainstorms and thus undetectable by conventional radar systems. Other aviation applications have included detection of wake vortices generated by large aircraft on takeoff, a hazard to smaller aircraft in particular. Interest has lessened somewhat with improvements in air-traffic control procedures, but, in general, having a wind sensor as an adjunct to other meteorological sensors is a useful asset at an airport. (Attempts to put wind sensors directly on aircraft have met with resistance due to cost considerations, not the least of which is finding a location for the sensor.) An example of a ground-based wind sensor for airport applications is the WindTracer system developed by Lockheed Martin Coherent Technologies. The device employs a pulsed, milliJoule-level solid state laser, which has evolved from a Tm-doped, 2,000-nm region laser to the present Er:YAG- based source at 1,617 nm. Specifications for the WindTracer product appear in Table 2-4, and a photograph of the shed-size unit is presented in Figure 2-55. Notable is the operating range of unit, which extends to as much as 33 km. According to the vendor, since the initial deployment in 2002, WindTracer successfully operates at Hong Kong, Tokyo, Osaka, London, New York, San Francisco, and Las Vegas airports, and is planned for installation at airports in Bangkok and Sicily. In addition, systems are in use at the Charles de Gaulle International Airport near Paris and at the Frankfurt International Airport, primarily for wake-vortex detection and ultimately determination of safe timing for departing aircraft. The goal is to attempt to speed up departure rates. A sample of the data product from the WindTracer system appears in Figure 2-56. 148 S.B. Alejandro, “Application of LIDAR Against Soft Targets,” presentation to the committee, March 5, 2013. 149 U.N. Singh, “Emerging and enabling lidar technologies and techniques for NASA’s space-based active optical remote sensing missions,” presentation to the committee, April 17, 2013.

OCR for page 25
98 LASER RADAR TABLE 2-4 Specifications for WindTracer Doppler Wind Sensor. Measurement Typical range 400 m to 18 km Maximum range 33 km Radial wind velocity range ±35 m/s Minimum range resolution 100 m Scanner Azimuth range 0 to 360 degrees Elevation range -5 to 185 degrees Resolution 0.001 degrees Pointing accuracy ±0.1 degrees Optical clear aperture 12 cm Transceiver Laser wavelength 1,617 nm Pulse energy 2.5 mJ ±0.5 mJ Pulse duration 300 ns ±150 ns Pulse repetition frequency 750 Hz Beam diameter 9.6 cm (e-1 intensity width) Shelter Environment All weather Weight 2600 kg Dimensions 197 × 244 × 329 (h) cm (clearance with lightning rods) Power specification 200-240 VAC single phase, 50 or 60 Hz (specified at time of purchase) 50 A service required. SOURCE: See © 2013 Lockheed Martin Corporation. All Rights Reserved. FIGURE 2-55 WindTracer system. SOURCE: © 2013 Lockheed Martin Corporation. All Rights Reserved.

OCR for page 25
ACTIVE ELECTRO-OPTICAL SENSING APPROACHES 99 FIGURE 2-56 Sample WindTracer color-coded radial wind velocity map at a 1- degree elevation tilt and out to a range of ~11 km. Overlaid on the radial velocities are wind barbs depicting the vector velocity field. SOURCE: S. Hannon, K. Barr, J. Novotny A. Oliver, J. Bass, and M. Anderson, “Large scale wind resource mapping using a state-of-the-art 3-D scanning lidar,” WINDPOWER 2008, Houston, June 1-4, 2008, Available at http://www.res- americas.com/media/918770/awea-2008_large-scale-wind-resource-mapping- using-a-state-of-the-art-3-D-scanning-lidar.pdf. Mitsubishi Electric (Japan) claims manufacture of a similar-sized Doppler wind sensor, employing a 1,500-nm-region pulsed solid state laser 150 and lists a 20-km horizontal range. Based on a recent publication, the laser final stage may use a state-of-the-art, Yb, Er:glass planar waveguide amplifier, which has been shown capable of generating an output power of 3.3 kW with a pulse width of 580 ns, 1.9 mJ of pulse energy, a repetition rate of 4 kHz, and a high beam quality, with an M2 factor of 1.3. 151 Applications of the device are listed as follows: • Meteorological measurements (3-D profile of atmospheric wind, turbulence). • Air traffic safety (low-level wind shear around airports, sudden gusts, and aircraft wake turbulence). • Air environment consultation (urban heating, monitoring of substances released in emissions from factories, application in wind power generation). 150 See http://www.mitsubishielectric.com/bu/lidar/products/coherent/index.html. 151 S. Kameyama, T. Yanagisawa, T. Ando, T. Sakimura, H. Tanaka, M. Furuta, and Y. Hirano, 2013, “Development of wind sensing coherent Doppler lidar at Mitsubishi Electric Corporation from late 1990s to 2013,” CLRC 2013, 17th Coherent Laser Radar Conference, Barcelona, Spain, June 17-20. Conference papers are available at http://www.tsc.upc.edu/clrc/.

OCR for page 25
100 LASER RADAR (a) (b) FIGURE 2-57 (a) Calculations of operating range for Doppler wind-sensing lidars as a function of visibility, with 1,550-nm laser pulse energy as a parameter. (b) Leosphere system in field installation at wind-turbine farm. SOURCE: (a) See http://esamultimedia.esa.int/docs/technologytransfer/Leosphere-Windcube.pdf. (b) See http://www.upwind.eu/media/576/D6.1.1.pdf. The emergence of a significant number of wind-turbine power systems has led to the development of a number of commercial short-range wind sensors. The economics of first determining good locations for turbines and then developing sensors that provide look-ahead data for turbine control justifies expenditures for laser-based systems rather than simple anemometers. The devices tend to be based on 1,550-nm region, all-Er:fiber architectures, given that only 0.1- to 1-km ranges are needed, and systems are advertised from Mitsubishi Electric, HALO Photonics (U.K.), Leosphere (France), and ZephIR (U.K.). Leosphere employs technologies initially developed at the Office National d’Etudes et RecherchesAérospatiales (Onera), the French national aerospace research center. Figure 2-57a shows data developed by Onera for a 1,550-nm based system with a 30-m range resolution, showing operating range under different atmospheric visibility conditions, with laser pulse energy as a parameter. 152 The LeosphereWindcube product for wind-turbine power generation facilities runs with 10-µJ (20-kHz-rate, 200-ns-duration) pulse energies and provides condition-dependent ranges of 40-200 m. 153 A photograph of the device appears in Figure 2-57b. The unit emits four separate beams at an angle to the vertical, spaced 90 degrees apart to develop wind-direction data. The ZephIR wind sensor is unique since, in contrast to the sensors just discussed, it employs a CW, 1,550-nm laser provided by NKT Photonics (Denmark). The source is based on a distributed feedback (DFB) single-frequency Er:fiber laser, amplified to approximately a 1-W power level. 154 According to the ZephIR manufacturer, 155 the device measures wind speeds at five different heights at altitudes in the 10-200 m range. The beam is projected at 30 degrees to the vertical and rotated to form a cone. The beam is focused to different heights in that cone and measures 50 points of data every second at each height. Speed measurements fall in the <1 to 70 m/s range, with <0.5 percent speed accuracy 152 See http://esamultimedia.esa.int/docs/technologytransfer/Leosphere-Windcube.pdf. 153 Ibid. 154 See http://esamultimedia.esa.int/docs/technologytransfer/Leosphere-Windcube.pdf. 155 “Fibre lasers stretch from cells to wind farms,” Optics and Lasers Europe, March 2009, 15, optics.org/ole.

OCR for page 25
ACTIVE ELECTRO-OPT EL TICAL SENSIN APPROACH NG HES 101 FIGURE 2-58 The ZephIR un in the field. SOURCE: Ze T nit ephIR 300 in nstalled, image courtesy of Ze e ephIRLidar. variation and <0.5 degr rees of directi accuracy variation. Th manufactur claims tha “ZephIR ha ion he rer at as also now amassed mor than 3 milli hours of operation acro 650+ depl re ion o oss loyments glob bally spannin a ng decade of commercial experience.”156 A photogra of the tra f 1 aph ash-can-sized unit appears in Figure 2-5 58. Conclusio 2-14: Win sensor sys on nd stem complex has been an issue for long-range systems but has xity n r been addressed and solved for sho nsors, based on a market pull. ort-range sen Conclusio 2-15: Any country tha develops a space-based wind sensor could use it for commer on y at d r t rcial and milit tary advantag ges. herent Detec Incoh ction of Mole ecular Scatter Figure 2-59 sh hows a schema of an inco atic oherent-detecction, wind-se ensing lidar sy ystem. In con ntrast to the coh herent-detectio system, the device emp on ploys narrow-l line filters an multiple de nd etector channe to els determine the shift in the center wav e t velength of th scattered li he ight. Figure 2 2-55 shows the positioning and e response of three band-pass filters (Edge1, Edge2 and Lockin Filters) em o 2, ng mployed in the so-called do e ouble- edge detecction techniqu which has evolved from the original edge-detecti concepts for incoheren ue, s m l ion nt- detection wind sensing 157,158,159According to Gen g. ntry, Chen, an Li, nd Th separation of the two edge filters is chos to be 5.1 G he o e sen GHz, so the sensitivity of the b broader molecular signal is equal to tha of the narrow aerosol sig m at wer gnal. The locki etalon peak ing k is located 1.7 GH from the Ed Hz dge1 filter peak so the crosso k, over point of th two edge-fil band- he lter paasses is aligned to the half-he d eight point of th Locking Fil band-pass. Actively locki the he lter ing la and the eta aser alon at this poin on the Locki Filter ensu symmetry of the edge-filt nt ing ures ter ch hannels for win measuremen nd nt. 156 Fred Olson Ltd. Ten years at the top, measurin right to the to with ZephI Lidart. d T e ng top, IR http://www w.fredolsen.co.uk/news/ten-ye ears-top-measuuring-right-top p-zephir-lidar. A Accessed on M March 14, 20144. 157 C.L Korb, B.M. Gentry, and C. Weng,” Ed technique: T L. G .Y. dge Theory and appplication to the lidar measure e ement of atmosph heric wind,” Ap Opt. 31 42 (1992). ppl. 202 158 C.L Korb, B.M. Gentry, S.X. Li, and C. Flesia,” Theory of t double-edg technique fo Doppler lida L. G L the ge or ar wind meas surement,” App Opt. 37 3097 (1998). pl. 159 B.M Gentry, H. Chen, and S.X. Li, “Wind me M. C easurements w 355-nm mo with olecular Doppl lidar,” Opt. Lett. ler 25, 1231 (2 2000).

OCR for page 25
102 LASER RADAR FIGURE 2-59 Schematic of incoherent-detection wind-sensing (Doppler) lidar based on molecular scattering. FIGURE 2-60 Profiles of wind speed and direction measured by the molecular lidar system compared with data from collocated balloons on November 17, 1999. SOURCE: B. M. Gentry, H. Chen, and S. X. Li, 2000, “Wind measurements with 355-nm molecular Doppler lidar,” Opt. Lett. 25: 1231.

OCR for page 25
ACTIVE ELECTRO-OPTICAL SENSING APPROACHES 103 Differential comparison of the signals from the two-edge filters allows determination of the Doppler shift from the backscattered signal. Although, in principal, one could apply the edge technique to sense aerosol Doppler shifts, current efforts center on the use of the scheme for molecular backscatter, since it provides data for regions of the atmosphere having low aerosol content and thus not suited for coherent sensing. Figure 2-60 shows validation data 160 taken by NASA from a ground-based, double-edge lidar that used a 45-degree elevation and four azimuthal angles for wind-direction determination. The system employed a single-frequency, tripled Nd:YAG laser that produced 15-ns pulses at a 10-Hz repetition rate, with 70 mJ of energy and a 80-MHz linewidth. The receiver telescope had a 45-cm aperture. Lidar data were compared with a conventional balloon-borne wind sensor and show excellent agreement in terms of both wind velocity and direction, with evident larger scatter at the highest altitudes. Currently, NASA is fielding the TWiLiTE (Tropospheric Wind Lidar Technology Experiment) designed for airborne platforms, and flight tests have been carried out on DC-8, WB57, and ER-2 aircraft, with plans for long-duration tests on a Global Hawk UAV. 161 The European Space Agency (ESA) has been funding development of the Atmospheric Dynamics Mission (ADM-Aeolus) effort, to employ an incoherent wind-sensing lidar (Aladin) to sense the lowermost 30 km of the atmosphere. The system was originally set for launch in October 2007 but experienced significant issues with performance of the 355-nm laser due to optics contamination and damage issues, and the launch date is now set for 2015. An alternative approach to incoherent wind sensing is under NASA-funded development, the Optical Autocovariance Wind Lidar (OAWL),162 which uses a double-edge sensor for the molecular- backscatter channel but is able to sense the Doppler shift of the aerosol channel through the use of a high- spectral-resolution interferometer in another detection channel. The advantage of the technology is that one sensor, at one wavelength, would be able to carry out the entire wind-sensing mission from space. Ground-based testing validation of the technology has been carried out, and aircraft-based measurements are in the planning stages. COMMERCIAL LASER/LADAR PRODUCTS As discussed in this chapter, there is a wide variety of uses for ladars, ranging from mapping and imaging to velocimetry. 163 While there is no rigid dividing line between scientific, military, and commercial applications, the first two domains are characterized by small numbers of high-performance devices, while the third is performance-constrained by marketplace affordability. Nonetheless, the compelling advantages of ladar in terms of distance and velocity measurement have created an active marketplace with a variety of products in two principal application areas: (1) speed detection for law enforcement and (2) distance sensing for positioning and obstacle avoidance. Police use ladar because it precisely measures target distance and velocity within a small field of view. For example, the illuminator (laser) can be used to select a specific vehicle (even in traffic) from which reflections can be measured. The devices (one is illustrated in Figure 2-61) have measurement filters to overcome errors introduced by surface irregularities such as mirrors and may rely on features on license plates (e.g., retroreflective backing materials) intended to support their use. The police ladar 160 B.M. Gentry, H. Chen, and S.X. Li, 2000, “Wind measurements with 355-nm molecular Doppler lidar,” Opt. Lett. 25: 1231. 161 See http://twilite.gsfc.nasa.gov/index.php. 162 C.J. Grund, S. Tucker, R. Pierce, M. Ostasziewski, K. Kanizay, D. Demara, and J. Howell, 2010, “Development and demonstration of an optical autocovariance direct detection wind lidar,” Paper B3P5, 2010 Earth Science Technology Forum (ESTF2010), June 22-24, Arlington, Virginia. http://esto.nasa.gov/conferences/estf2010/papers/Grund_Chris_ESTF2010.pdf. 163 See http://www.lidar-uk.com/usag EOf-lidar/ for a listing.

OCR for page 25
104 LASER RADAR devices are very affordable. Online auction sites have them for $1,500-2,000. Interestingly, the devices have stimulated active countermeasures by motorists, such as so-called “ladar jammers” which emit light in the same frequency range as the speed guns. The effectiveness of these countermeasures is unclear. The marketplace for the second application area, distance sensing, also seems primarily oriented toward automotive applications. In this case, rather than law enforcement, the primary motivations are safety and navigation. The devices are characterized by multiple lasers and detectors organized in an array format to achieve a wide field of view; the array can be cyclically rotated to achieve 360 degree swaths of real-time information. These devices can in principle be used to provide greater situational awareness or even driver assistance for a human driver. A more intriguing application with substantial commercial investment is underway that has a potentially much larger market than speed guns: driverless automobiles. In this application, the ladar serves as part of a package of sensory apparatus needed for the autonomous vehicle to avoid stationary obstacles as well as moving objects such as pedestrians and other vehicles. The sensors used in this application (e.g., for the Stanley 164 self-driving car, which won the 2005 DARPA Grand Challenge) include those from SICK, examples of which are shown in Figure 2-62 were also used atop an autonomous vehicle that completed the DARPA Urban Challenge in 2007. FIGURE 2-61 UltraLyte LTI 20-20 speed enforcement LiDAR. SOURCE: http://www.lasertech.com/UltraLyte-Laser- Speed-Guns.aspx. 164 S. Thrun, M. Montemerlo, H. Dahlkamp, D. Stavens, A. Aron, J. Diebel, P. Fong, J. Gale, M. Halpenny, G. Hoffmann, K. Lau, et al., 2006, “Winning the DARPA Grand Challenge,” Journal of Field Robotics.

OCR for page 25
ACTIVE ELECTRO-OPTICAL SENSING APPROACHES 105 FIGURE 2-62 Ladars from SICK atop “Little Ben,” an experimental self-driving car developed for the 2007 DARPA Urban Challenge. Credited to Lehigh University. The ladar type highlighted by the ovals in Figure 2-62 is intended for short-range (e.g., 20-50 m) outdoor operation. It uses a 905 nm light source, has a 270 degree field of view, fog correction, a 25-50 Hz scanning frequency, and an operating range (depending on model) of 20-50 m, with an error of about 1 cm. Retail price ranges from $4,000 to $6,000. Significant indicators would include dramatic reductions in cost and/or performance. The thousand dollar range of commercial ladar products listed above does not cover integration into a system vehicle. The integration is a barrier to entry but one that many are striving to overcome commercially. The achievable performance could be improved by signal processing methodologies such as algorithmic advances or advances in signal processing hardware such as a higher degree of parallelism or integration. Any advances that are applicable in the commercial lidar domain are equally useful in more specialized systems, such as those for defense applications. Also, improvements in size, weight, and power are generally indicative of manufacturing capabilities that are applicable across technology applications, including defense. Incentives for the commercial lidar technologies include (for the producer) general advantages that stem from active and profitable instrument manufacturing capabilities, including the ability to allocate some profits to further research and development. Disincentives for the commercial ladar technologies include trade-offs that come from commercialization, such as trading away performance to hit an attractive price point, or overoptimizing for a particular problem domain such as speed detection. Improvements in laser range finder technologies will enhance advanced manufacturing capabilities such as autonomous, flexible manufacturing and sensor-actuator models. The key technology improvements will increase performance (including measurement accuracies, scan rates, and system reliability) while decreasing cost. The likely means of achieving desirable new positions in the cost/performance trade space is integration (and the associated miniaturization) of components rather than a particular component technology such as a better laser, detector, or processor. Experience with both applications of the technologies and the processes used to manufacture them will present opportunities for technology advances.

OCR for page 25
106 LASER RADAR There are significant foreign capabilities in commercial ladar. Vendors exist abroad, such as SICK in Germany and Hokuyo in Japan as well as in the United States (e.g., Velodyne). Conclusion 2-16: Commercial applications for active EO sensing will become widespread, dramatically increasing the use of active EO sensing and significantly lowering the cost. The performance and precision of active EO sensors are compelling, but there is currently a cost trade-off that may limit commercial penetration in very cost-sensitive settings. For example, ultrasonic sensors that aid in automobile parking exist and only cost a few dollars; an active EO sensor in this application could be considered overkill. However, the importance of the human lives involved in autonomous vehicles will put pressure on the sensor technologies for accuracy, and with sensor fusion, diversity and redundancy. The “overkill” of an active EO sensor could in fact be a desirable attribute if the sensor is multifunctional and can be made cheaply. Conclusion 2-17: Robotics—for example, autonomously navigated vehicles—is likely to be a dominant commercial and military application for active EO sensing. Conclusion 2-18: Active EO sensing is poised to significantly alter the balance in commercial, military, and intelligence operations, as radar has done over the past seven decades. Conclusion 2-19: Most active EO sensors used for military and security application will fall into one of two categories: (1) adaptation of inexpensive commercial devices to military application or (2) development of unique ladar systems for the military. The first category is accessible to nonstate actors and resource-poor countries. The second category is only accessible to countries with significant resources. Conclusion 2-20: Development programs for high-end active EO sensors will involve observable activities or indicators of the direction these programs are taking. Conclusion 2-21: Widespread, commercially available active EO technology offers a low barrier to entry for asymmetric adversaries with limited resources as well as to those seeking to adapt the technology for large-volume applications. Recommendation 2-1: Analysts assessing states of technological development and projected timelines for developments of military significance should pay close attention to activity in the commercial sector.