Open Access
Add to BibSonomy
Abstract
After entering the 21st century, asteroid exploration has become one of the key research directions in the field of deep space exploration. In this article, we present a navigation, imaging, and hyperspectral acquisition integrated optical system designed for asteroid exploration. Based on the pixel-level light modulation capability of the digital micromirror device (DMD), this system introduces a triple-pass total internal reflection (TIR) prism to overcome the limitation of small DMD deflection angles, achieving the co-aperture design of the high-resolution imaging branch and the hyperspectral acquisition branch. In view of the actual usage scenarios, we design a new triple-pass TIR prism and optimize it, correcting the on-axis point coma and astigmatism it introduces, and designing it to be lightweight and compact. Compared to existing optical systems for asteroid exploration, this system offers advantages such as compact structure, no moving parts, high spatial resolution, high spectral resolution, and more flexible imaging modes.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
After entering the 21st century, deep space exploration has gradually become a new focal point in the aerospace field worldwide, representing another frontier in space development following the exploration of near-Earth satellites and manned spaceflight. Deep space exploration encompasses missions to the Moon, as well as the exploration of planets such as Mars, Mercury, Jupiter, and more, along with missions targeting asteroids and comets. Among these, the exploration of asteroids holds the potential to further our understanding of the universe, enhance our knowledge of the solar system, and investigate the formation and evolution of Earth's environment. As a result, it has progressively emerged as one of the key research directions within the field of deep space exploration [1–6].
Asteroid exploration spacecraft is facing a series of challenges, including vast distances, extended flight durations, limited data transmission rates, and the complexity of the unknown deep space environment. Ground control has many limitations like real-time and emergency response, necessitating spacecraft with optical autonomous navigation capabilities [7–9]. Additionally, in the intricate environment of deep space, it is challenging for any single sensor to provide an accurate description of the surroundings. Therefore, the ability to achieve high-resolution imaging and spectral analysis of multiple celestial bodies such as encountered asteroids during the mission is essential for asteroid exploration spacecraft [10]. The requirements for the navigation camera, high-resolution imaging camera, and spectral camera on asteroid exploration spacecraft are similar in terms of optical system parameters. From an integrated and miniaturized perspective, the design of an integrated navigation, imaging, and spectral camera is an inevitable trend in asteroid exploration [11,12].
In the field of navigation, imaging, and spectral integrated optical systems for asteroid exploration, there are currently two main optical configurations. 1) Filter wheel configuration: This approach involves switching between panchromatic and spectral channels using a filter wheel. Such as the NavCam camera on the Stardust-Next spacecraft, the FC camera on the Dawn spacecraft, AMICA on the Hayabusa spacecraft, and ONC-T on the Hayabusa 2 spacecraft [13–17]. These systems are relatively simple and can achieve both imaging and spectral analysis with a single detector. However, they have limitations such as a limited number of spectral channels, low spectral resolution, the ability to only capture stare images, and the impact of mechanical moving parts on navigation accuracy. 2) Visible light cameras combined with prisms or gratings: This approach is used in cameras like MICAS on Deep Space 1, HRI on Deep Impact [18–22], and others. These systems effectively improve the spectral resolution of deep space exploration instruments and do not rely on moving parts. However, they can only achieve two-dimensional spectral analysis in a push-broom manner. Deep space exploration, especially asteroid detection, has complex imaging conditions, making it difficult to achieve push-broom imaging. Therefore, satellites usually use line field of view sampling for spectral analysis, resulting in poor detection flexibility. In addition, these systems use a beam splitter to achieve co-aperture design of multiple optical paths, which leads to conflicts between the navigation branch and the spectral branch, making it impossible to perform spectral analysis on the navigation band.
Here is a new integrated optical system for navigation, imaging and hyperspectral acquisition in this paper. The system combines multiple optical paths tightly using a DMD and splits adjacent optical paths using a triple-pass TIR prism. The article studies the operational state of the triple-pass TIR prism, provides its design methodology, analyzes and corrects for the aberrations it causes, and performs lightweight design optimization. With the DMD and the triple-pass TIR prism as its core components, the design of the integrated optical system for navigation, imaging and hyperspectral acquisition is completed. An experimental setup is constructed to validate the feasibility of this system.
2. System composition and working principle
The optical path configuration of the integrated optical system for navigation, imaging, and hyperspectral acquisition is shown in Fig. 1. This system consists of three closely coupled parts: the telescope, the high-resolution imaging branch, and the hyperspectral acquisition branch. The DMD is located at the image plane of the telescope, and each micromirror on it corresponds precisely to a target point within the two-dimensional field of view, a pixel on the detector in the high-resolution imaging branch, and a spectral line in the hyperspectral acquisition branch. As needed, any pixel can be controlled to enter either the high-resolution imaging branch or the hyperspectral acquisition branch [23–28].
Fig. 1. Composition of the integrated optical system for navigation imaging and hyperspectral acquisition.
In this system, the combination of the telescope and the high-resolution imaging branch is responsible for observation and navigation, while the combination of the telescope and the hyperspectral acquisition branch is responsible for spectral analysis of the targets. During operation, all micromirrors on the DMD are redirected towards the high-resolution imaging branch, creating a high-resolution geometric image on the detector of the high-resolution imaging branch. This image can be used for navigation and target analysis. If the analysis indicates a need for spectral analysis of specific targets, the corresponding DMD micromirrors are controlled to redirect towards the hyperspectral acquisition branch. When selecting targets, attention should be paid to the spacing between them to avoid spectral overlap between adjacent targets. When conditions permit, a single row of DMD micromirrors can also be used to form an entrance slit for the hyperspectral acquisition branch, allowing for spectral analysis of all targets within the two-dimensional field of view through push-broom scanning.
Due to the small swinging angles of the DMD micro-mirrors (typically at +12 degrees and -12 degrees), when the light beam is incident perpendicular to the DMD chip, the angle between the reflected light beam's main ray and the optical axis of the telescope is only 24 degrees, as shown in Fig. 2. This results in a mutual constraint between the linear field of view and the aperture angle of the telescope. From a spatial layout perspective, due to the relatively small angles between the high-resolution imaging branch, the hyperspectral acquisition branch, and the optical axis of the telescope, the three components are very close to each other in terms of spatial structure. This directly leads to mutual constraints on their system apertures.
To address the aforementioned issues, a triple-pass TIR prism has been introduced into this optical system [29–32]. This prism splits the light paths of adjacent systems, allowing both the avoidance of mutual constraints and occlusion between different systems while folding the optical paths to make the system compact.
3. Design and analysis of TIR
3.1 Total reflection angle of TIR
The triple-pass TIR prism consists of three prism components, as shown in Fig. 3. Among them: Prism 1, situated near the telescope, receives the converging light from the telescope. Prism 2, positioned near the relay lens, is responsible for redirecting the light reflected by the DMD into the high-resolution imaging branch. Prism 3, located near the DMD chip and the collimating lens, is responsible for redirecting the light reflected by the DMD into the hyperspectral acquisition branch. The three prisms are positioned with their interfaces parallel to each other and in close proximity, with an air gap of approximately 10µm between the interfaces. This configuration allows for total internal reflection of light beams at these interfaces at specific angles.
The total reflection capability of the triple-pass TIR prism relies on the angles ${\theta _1}$ and ${\theta _2}$ of surfaces 1 and 2. According to its working principle, light rays from the telescope pass through surfaces 1 and 2, and the rays reflected by the DMD undergo total reflection when passing through surfaces 1 and 2 again. The operating wavelength range of this system is 400 nm to 800 nm. Due to the different angles required for total reflection at different wavelengths, it is necessary to ensure that even the shortest wavelength light can pass through when it comes from the telescope. Similarly, when the light comes from the DMD, it is essential to ensure that even the longest wavelength light can undergo total internal reflection.
Therefore, for the incident light from the telescope, the constraints on ${\theta _1}$ and ${\theta _2}$ are as shown in Eqs. (1) and (2). Here, F# represents the F-number of the telescope, ${n_{400}}$ is the refractive index of the material of the triple-pass TIR prism at a wavelength of 400nm, and ${\theta _{T400}}$ represents the total reflection angle corresponding to a wavelength of 400 nm, and its calculation is given by Eq. (3).
When the DMD micromirrors are in the “on” state, all micromirrors are turned toward the high-resolution imaging branch, and the light undergoes total reflection at surface 2. At this time, ${\theta _2}$ is subject to the following constraints (as can be seen from Fig. 3, the incident angle of light on surface 1 is smaller than the incident angle of light from the telescope, preventing total reflection; hence, no specific constraint is placed on ${\theta _1}$):
When the DMD is in the “off” state, micromirrors are turned toward the hyperspectral acquisition branch, and the light undergoes total reflection at surface 1. At this time, ${\theta _1}$ is subject to the following constraints:
By using Eqs. (1–5), when we have determined the material of the triple-pass TIR prism and the F-number of the telescope, we can obtain the range of total reflection surface tilt angles that meet the application requirements.
3.2 Aberration analysis and correction of TIR prism
As depicted in Fig. 3, the triple-pass TIR prism can be effectively considered as parallel plates relative to the telescope. However, its impact on image quality is relatively minor and can be easily corrected. In contrast, in both the high-resolution imaging branch and the hyperspectral acquisition branch, the triple-pass TIR prism can be equivalently modeled as a single prism. Due to the fact that the incident light beam's main ray makes a certain angle with the prism's entrance interface, this refraction process breaks the symmetry of the light beam, introducing more complex aberrations and resulting in a degradation of image quality. This effect is particularly pronounced for points along the optical axis, and since the subsequent optics are coaxial, the aberrations introduced by the prism cannot be corrected.
To analyze the impact of the prism-induced aberrations, we established a vector ray tracing model for the prism, as shown in Fig. 4. In this model, the light rays emitted from points on the optical axis can all be represented by their optical vectors, denoted as $\overrightarrow {{L_1}} $. $\overrightarrow {{L_1}} $ is a unit vector, and its direction is determined by the aperture angle u and the azimuthal angle φ. It can be expressed as follows:
The refracted light ray corresponding to the optical vector $\overrightarrow {{L_1}} $ after passing through the prism surface can be determined using the vector refraction law:
In the equation, ${n_i}$ and ${n_{i + 1}}$ represent the refractive indices of the incident medium and the exiting medium of the light beam, respectively. $\overrightarrow {{K_i}} $ signifies the unit normal vector of each refractive surface, directed from the incident medium to the exiting medium. ${\alpha _i}$ and $\alpha _i^\mathrm{^{\prime}}$ respectively denote the incident angle and the refracted angle of the light beam at the refractive surface.
In this model, the air gap ${d_1}$ between the DMD and the prism, the thickness ${d_2}$ of the prism, and the angle ${\beta _2}$ of the prism's exit surface are treated as variables. Using the vector ray tracing method, we convert the aberrations introduced by the prism into parameters such as the root mean square radius of a spot diagram, the focal position, and the differences in focal positions for different rays in order to quantitatively analyze the aberrations.
To keep the system compact, the distance between the DMD and the TIR prism is set to a minimum value of 3 mm. Therefore, for the equivalent prism, there are only two remaining variables: the thickness of the prism and the angle of its exit surface. The analysis of the impact of these two factors on coma and astigmatism was conducted using vector ray tracing, and the results are shown in Figs. 5 and 6.
Fig. 5. Effects of prism thickness and inclination of exit surface on coma and astigmatism. (a) Effect on coma. (b) Effect on astigmatism.
The analysis reveals that by altering the thickness of the prism and the angle of its exit surface, coma and astigmatism introduced by the prism can be individually eliminated. However, the conditions for their elimination are not the same. Considering that astigmatism can be corrected by adding a cylindrical lens, the combination of prism thickness and exit surface angle is employed to eliminate coma. The results of the aberration analysis were input into ZEMAX for simulation, and they are compared to the state where the chief ray exits the optical system perpendicular to the exit surface. The full-field astigmatism and coma maps are shown in Fig. 7. The simulation results demonstrate that adjusting the tilt of the exit surface in conjunction with a cylindrical lens can effectively correct the coma and astigmatism at on-axis points, bringing the center of the aberration field symmetry back to the center of the field of view.
Fig. 7. Full field of view aberration map. (a) Coma map in the vertical exit state. (b) Astigmatism map in the vertical exit state (c) Coma map under aberration elimination conditions. (d) Astigmatism map under aberration elimination conditions
It's important to note that the method described above only eliminates coma and astigmatism at points on the optical axis. Coma and astigmatism at off-axis points will need to be corrected by subsequent optical elements. The prism also introduces some lateral chromatic aberration, which can be corrected by introducing eccentricity and tilt to the cylindrical lens.
3.3 Lightweight design
The size of the triple-pass TIR prism can affect the aperture of the subsequent optical system, which in turn impacts the system's volume and mass. For instruments designed for asteroid exploration, larger launch mass leads to higher costs. Therefore, while ensuring that all reflected light passing through the DMD can smoothly enter the subsequent optical paths, it is essential to minimize the size of the TIR prism.
To ensure the smooth operation of the triple-pass TIR prism, all incoming light from the telescope must pass through surface 5 to enter the TIR prism. When the DMD is in the “on” state, all reflected light rays undergo total reflection at surface 2 and exit through surface 4. When the DMD is in the “off” state, all reflected light rays undergo total reflection at surface 1 and exit through surface 4, as shown in Fig. 8.
Based on this foundation, a lightweight design is applied to the triple-pass TIR, and the critical conditions that need to be satisfied are as follows:
- a) The intersection point of marginal ray 1 with surface 2 coincides with frontier point 1.
- b) The intersection point of marginal ray 2 with surface 2 coincides with frontier point 2.
- c) The intersection point of marginal ray 3 with surface 1 coincides with frontier point 3.
- d) The intersection point of marginal ray 4 with surface 1 coincides with frontier point 4.
- e) The intersection point of marginal ray 5 with surface 5 coincides with frontier point 5.
- f) The intersection point of marginal ray 6 with surface 5 coincides with frontier point 6.
As described in section 3.2, under the conditions for coma correction, the size and tilt angle of the exit surface of the TIR prism are interrelated. Therefore, during the lightweight design of the triple-pass TIR prism, it is necessary to simultaneously adjust the tilt angle of the exit surface to eliminate coma at on-axis points. The final design result of the triple-pass TIR prism is shown in Fig. 9.
4. Design and optimization of optical system
The designs of the telescope, high-resolution imaging branch, and hyperspectral acquisition branch were completed separately, and the final design result, achieved by integrating them using the triple-pass TIR prism and DMD, is shown in Fig. 10. The main parameters of the system are shown in Table 1.
Fig. 10. Optical path diagram of the integrated optical system for navigation imaging and hyperspectral acquisition.
Table 1. The main parameters of the telescope
The telescope is a coaxial refractive-reflective system with a telecentric structure in the image space. This design not only shortens the optical system's length but also reduces the difficulty of correcting chromatic aberration and secondary spectrum. It also facilitates the integration of subsequent optical systems. The first lens material of the correction lens group is fused quartz because it has good radiation stability and provides some protection for subsequent lenses when facing various radiations in outer space, such as gamma rays. The DMD is located on the focal plane of the telescope and consists of 1920 × 1080 micromirrors, with each micromirror measuring 7.56µm × 7.56µm.
The high-resolution imaging branch, labeled as “Multi-structure 1”, is designed with an object-side telecentric optical path. Due to the special structure of DMD, it is a tilted surface relative to the subsequent optical path. According to the Scheimpflug Principle, the image surface needs to introduce a certain amount of tilt to ensure that the best imaging effect is achieved for each point on the DMD, as shown in Fig. 11. The amount of tilt can be obtained by Eq. (8):
The hyperspectral acquisition branch, labeled as “Multi-structure 2”, employs an object-side telecentric optical path design for the collimating lens. It achieves a spectral resolution of 3 nm and uses a prism-grating-prism combination as the spectroscopic element, which has the advantages of high diffraction efficiency and coaxial optical path. Similar to the high-resolution imaging branch, the image surface of the hyperspectral acquisition branch also needs to introduce a certain amount of tilt, so the PGP needs to be placed horizontally to disperse the light from the DMD along the x direction, ensuring that the light of each wavelength can achieve the best imaging effect, as shown in Fig. 12.
Fig. 12. The influence of PGP orientation on the hyperspectral acquisition branch image plane. (a) Dispersion direction is y-direction. (b) Dispersion direction is x-direction.
The detector pixel size used in both the high-resolution imaging branch and the hyperspectral acquisition branch is 7.56µm, which is equivalent to the size of the DMD micromirror, facilitating the subsequent system calibration.
The image quality of the integrated optical system for navigation, imaging and hyperspectral acquisition is shown in Figs. 13 to 14. From the figures, it can be observed that the modulation transfer function (MTF) for all fields of view in the high-resolution imaging branch is above 0.43 at 66 lp/mm. In the hyperspectral acquisition branch, the energy concentration of different wavelength dispersion spots within a 2 × 2 pixel range exceeds 90%. These results demonstrate that the optical system exhibits good imaging quality, providing design-level evidence of the feasibility of the dual-path co-aperture navigation, imaging, and hyperspectral acquisition integrated optical system.
Fig. 14. MTF and energy concentration curve of hyperspectral acquisition branch. (a) 400 nm. (b) 600 nm. (c) 800 nm.
Figure 15 presents a point plot chart corresponding to the DMD micromirrors in the hyperspectral acquisition branch. The chart uses the four corners and the center of the micromirrors as feature points to represent the size of the micromirrors. The results show that when the wavelength interval is 3 nm, the system exhibits excellent resolving capability for the point plots formed by the micromirrors, meeting the design requirements for spectral resolution.
5. Comparison to a previous design
As mentioned in the introduction of this article, there are some drawbacks to filter wheel configuration asteroid exploration instruments and visible light cameras combined with prisms or gratings asteroid exploration instruments. In comparison, the navigation, imaging, and hyperspectral acquisition integrated optical system based on DMD and triple-pass TIR prism has the advantages of high spectral resolution, no moving parts, a large field of view for hyperspectral acquisition branch, and solves the problem of conflict between the navigation branch and the spectral branch of visible light cameras combined with prisms or gratings asteroid exploration instruments. Most importantly, the pixel-level light modulation capability of DMD makes the system have a more flexible imaging mode, which is sufficient to deal with various application scenarios encountered in asteroid detection missions.
Under normal circumstances, the system can analyze the images of the high-resolution imaging branch to control the DMD for spectral analysis of the target of interest; under low-light conditions, the DMD micromirrors that the target point passes through over a period of time can be flipped sequentially to achieve multiple exposures of the same target, thereby improving the signal-to-noise ratio of the hyperspectral acquisition branch.
During the observation period of gazing at the asteroid, the DMD can be controlled to flip line by line, achieving spectral analysis of all targets within the two-dimensional field of view.
In the push-broom mode, the optimal imaging effect can be achieved by adjusting the slit width through the analysis of ambient light. However, it should be noted that increasing the slit width will decrease the spatial resolution and spectral resolution of the hyperspectral acquisition branch.
6. Experiment and analysis
To validate the feasibility of the integrated optical system for navigation, imaging and hyperspectral acquisition based on DMD and triple-pass TIR prism in an engineering context, this paper constructed an experimental optical setup to simulate its imaging scenario, as shown in Fig. 16.
In this experiment, a star simulator was used as the light source to simulate the actual operation of the system. When the star simulator emitted a group of transparent star point arrays, the imaging situation on the imaging detector is shown in Fig. 17(a). By adjusting the DMD, some of the star points were diverted to the spectral branch, and the signal received by the spectral detector is depicted in Fig. 17(b). Due to the relatively small image plane of the spectral detector, these spectral lines are not complete. The complete spectral representation is indicated within the red box in Fig. 17(b).
In the above-mentioned experiment, the imaging detector successfully received images, demonstrating its ability for array imaging. When coupled with navigation algorithms, it can provide navigation functionality for asteroid exploration spacecraft. The spectral detector successfully collected spectral information, validating the spectral analysis capability of the spectral branch. In conclusion, the dual-path co-aperture navigation, imaging, and hyperspectral acquisition integrated optical system based on DMD and triple-pass TIR prism is engineering feasible and can meet the requirements for a combined navigation imaging and spectral camera.
7. Conclusion
In this paper, considering the research status of asteroid exploration in various countries worldwide, we analyzed the advantages and inevitable trends of integrated instruments for navigation, imaging and hyperspectral acquisition. We proposed a design solution for such an integrated instrument that employs DMD and a triple-pass TIR prism to achieve dual-path co-aperture design. The paper elucidated the working principles and design methodology of the triple-pass TIR prism and analyzed the aberrations introduced by TIR, with a particular focus on the correction of coma and astigmatism at on-axis points. In the end, an integrated optical system for navigation imaging and hyperspectral acquisition was designed. This system operates in the wavelength range of 400 nm to 800 nm, has an instantaneous field of view of 18.9µrad, a F-number of 4, and a spectral resolution of 3 nm. The system exhibits good imaging quality, compact structure for easy installation and adjustment, and it meets the requirements of practical applications.
Funding
National Natural Science Foundation of China (41974210).
Acknowledgments
The authors would like to thank the anonymous reviewers for their valuable comments.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. R. Zhang, J. Huang, R. He, et al., “The development overview of asteroid exploration,” Journal of Deep Space Exploration 6(5), 417–423 (2019). [CrossRef]
2. H. Wang, J. Yan, M. Fan, et al., “Dynamics of asteroid probe orbit calculations,” Chinese Space Science and Technology, (2023), https://kns.cnki.net/kcms2/detail/11.1859.v.20230531.1350.002.html
3. W. Wu, W. Liu, D. Qiao, et al., “Investigation on the development of deep space exploration,” Sci. China Technol. Sci. 55(4), 1086–1091 (2012). [CrossRef]
4. J. Qiu, Q. Wang, and J. Ma, “Deep space exploration technology (Invited),” Infrared and Laser Engineering 49(5), 20201001 (2020). [CrossRef]
5. L. Xu, Y. Zou, and Y. Jia, “China’s Planning for Deep Space Exploration and Lunar Exploration before 2030,” Chin. J. Space Sci. 38(5), 591–592 (2018). [CrossRef]
6. V. G. Shevchenko and R. A. Mohamed, “Spacecraft Exploration of Asteroids,” Sol. Syst. Res. 39(1), 73–81 (2005). [CrossRef]
7. X. Xu, J. Sang, and H. Liu, “Discussion on optical autonomous navigation methods for deep spacecrafts,” Journal of Navigation and Positioning 9(1), 1–4 (2021). [CrossRef]
8. Z. Ye, Q. Zhou, and H. Wang, “Research on autonomous navigation method for deep space exploration spacecraft in cruise phase,” Chinese Space Science and Technology 40(5), 53–60 (2020). [CrossRef]
9. W. Zhang, J. Xu, Q. Huang, et al., “Survey of autonomous celestial navigation technology for deep space,” Flight Control & Detection 3(4), 8–16 (2020).
10. X. Liu, F. Wu, and Y. Wu, “Development trend of optical navigation sensor technology for asteroid exploration,” Optics and Precision Engineering 25(12), 1–17 (2017). [CrossRef]
11. N. Thomas, H. U. Keller, E. Arijs, et al., “OSIRIS-The optical, spectroscopic and infrared remote imaging system for the Rosetta Orbiter,” Adv. Space Res. 21(11), 1505–1515 (1998). [CrossRef]
12. D. L. Hampton, J. W. Baer, M. A. Huisjen, et al., “An Overview of the Instrument Suite for the Deep Impact Mission,” Space Sci. Rev. 117(1-2), 43–93 (2005). [CrossRef]
13. K. P. Klaasen, D. Brown, B. Carcich, et al., “Stardust–NExT NAVCAM calibration and performance,” Icarus 222(2), 436–452 (2013). [CrossRef]
14. H. Sierks, H. U. Keller, R. Jaumann, et al., “The Dawn Framing Camera,” Space Sci. Rev. 163(1-4), 263–327 (2011). [CrossRef]
15. A. Fujiwara, J. Kawaguchi, D. K. Yeomans, et al., “The Rubble-Pile Asteroid Itokawa as Observed by Hayabusa,” Science 312(5778), 1330–1334 (2006). [CrossRef]
16. Y. Tsuda, M. Yoshikawa, M. Abe, et al., “System design of the Hayabusa 2- Asteroid sample return mission to 1999 JU3,” Acta Astronaut. 91, 356–362 (2013). [CrossRef]
17. Y. Tsuda, M. Yoshikawa, T. Saiki, et al., “Hayabusa2-Sample return and kinetic impact mission to near-earth asteroid Ryugu,” Acta Astronaut. 156, 387–393 (2019). [CrossRef]
18. S. Bhaskaran, S. D. Desai, P. J. Dumont, et al., “Orbit determination performance evaluation of the deep space 1 autonomous navigation system,” Spaceflight Mechanics 1(2), 1295–1314 (1998).
19. M. D. Rayman, P. Varghese, D. H. Lehman, et al., “Results from the Deep Space 1 technology validation mission,” Acta Astronaut. 47(2-9), 475–487 (2000). [CrossRef]
20. R. G. Walker, W. B. Weaver, W. W. Shane, et al., “Deep Impact: Optical spectroscopy and photometry obtained at MIRA,” Icarus 191(2), 526–536 (2007). [CrossRef]
21. R. B. Frauenholz, R. S. Bhat, S. R. Chesley, et al., “Deep Impact Navigation System Performance,” J. Spacecr. Rockets 45(1), 39–56 (2008). [CrossRef]
22. J. Oberst, B. Brinkmann, and B. Giese, “Geometric calibration of the MICAS CCD sensor on the DS1 (Deep Space One) spacecraft: laboratory vs. in-flight data analysis,” International Archives of Photogrammetry and Remote Sensing XXXIII(B1), 221–230 (2000).
23. A. Travinskya, D. Vorobieva, Z. Ninkova, et al., “Evaluation of digital micromirror devices for use in space-based multi-object spectrometer application,” J. Astr. Telescopes, Instruments, and Systems 3(3), 035003 (2017). [CrossRef]
24. P. Spanò, F. Zamkotsian, R. Content, et al., “DMD multi-object spectroscopy in space: the EUCLID study,” UV/Optical/IR Space Telescopes: Innovative Technologies and Concepts IV 7436, 74360O (2009). [CrossRef]
25. K. D. Fourspring, Z. Ninkov, and J. P. Kerekes, “Scattered light in a DMD based multi-object spectrometer,” Modern Technologies in Space- and Ground-based Telescopes and Instrumentation 7739, 77390O (2010). [CrossRef]
26. N. A. Riza and F. N. Ghauri, “Hybrid analog-digital MEMS fiber-optic variable attenuator,” IEEE Photon. Technol. Lett. 17(1), 124–126 (2005). [CrossRef]
27. P. Wu, “Research on Imaging Technology of High Dynamic Range Pixel-level Light Adjusting based on Digital Micromirror Device,” University of Chinese Academy of Sciences, (2017).
28. Y. Wang, H. Chen, Y. Liu, et al., “Application of DMD to high dynamic scene image detection system,” Optics and Precision Engineering 22(09), 2508 (2014). [CrossRef]
29. Y. Qiao, X. Xu, Y. Pan, et al., “Research on DMD infrared scene projector with a high contrast ratio infrared prism design,” Optik 125(22), 6854–6859 (2014). [CrossRef]
30. Y. Huang and J. Pan, “High Contrast Ratio and Compact-Sized Prism for DLP Projection System,” Opt. Express 22(14), 17016–17029 (2014). [CrossRef]
31. J. Pan, C. Wang, W. Sun, et al., “Portable digital micromirror device projector using a prism,” Appl. Opt. 46(22), 5097–5102 (2007). [CrossRef]
32. J. Pan and H. Wang, “High contrast ratio prism design in a mini projector,” Appl. Opt. 52(34), 8347–8354 (2013). [CrossRef]
