Research Article  Open Access
Jingji Wang, Chunyang Liu, "The Relative Perturbation Analysis of Satellite Formation under the Requirement of Relative Position Maintenance with MillimeterLevel Accuracy", International Journal of Aerospace Engineering, vol. 2021, Article ID 6687468, 9 pages, 2021. https://doi.org/10.1155/2021/6687468
The Relative Perturbation Analysis of Satellite Formation under the Requirement of Relative Position Maintenance with MillimeterLevel Accuracy
Abstract
With highprecision DEM (Digital Elevation Model) and GMTI (Ground Moving Target Indicator) as the demand background, the influence of zonal harmonic term perturbation on the relative motion of the millimeterlevel shortrange leaderfollower satellites in nearcircular orbit is studied through the relative perturbation method. An equation of motion that can describe the motion of the leaderfollower satellites under the influence of perturbation in nearcircular orbit is derived, and the characteristics of the trajectory of inplane periodic motion are analyzed. A study shows that under the influence of the relative perturbation of the term, the inplane periodic motion of the leaderfollower satellites in nearcircular orbit is a symmetrical closed “dropshaped” trajectory with a period of . By comparing with the results of numerical simulations, the correctness of the conclusions obtained in this paper is verified. According to the research results, it can be known that only using a thruster as the actuator to maintain the relative position can no longer meet the requirements of the longterm mmlevel relative position maintenance. In the future, a new technical approach needs to be explored to achieve the longterm relative position maintenance with millimeterlevel control accuracy.
1. Introduction
The satellite formation system [1] refers to a space system composed of multiple physically disconnected satellites in order to complete the same mission. Satellite formation can not only complete complex tasks that cannot be completed by a single satellite but also have more advantages than a single satellite in terms of reliability, flexibility, and economy. Therefore, satellite formation systems have become a research hotspot in the aerospace field and are widely used in many other fields, such as earth observation, astronomical observation, and deep space exploration. This article takes the highprecision DEM (Digital Elevation Model) and GMTI (Ground Moving Target Indicator) tasks as the background. As shown in Figure 1, the leaderfollower satellite means that the deputy satellite and the chief satellite operate in the same orbit in a certain sequence, and the orbit elements of leaderfollower satellites are only different in argument of latitude. Therefore, the leaderfollower satellite is equivalent to a sparse antenna array, using ground echo interferometric imaging principles and formation flying satellite control technology to complete DEM and GMTI tasks.
Whether the satellite formation system can meet the DEM mission measurement accuracy and the minimum detectable speed of the GMTI mission not only depends on the performance of payload, length of baseline, and data processing but also depends on the control accuracy of relative position maintenance of the payload of the leaderfollower satellites. In lowprecision observation tasks, it only needs to control the relative position accuracy of the interference baseline at the meter or decimeter level. Therefore, in the study of relative dynamics, only the secular and longperiodic effects of various perturbations on the relative position need to be paid attention to. As users have higher and higher requirements for performance and accuracy, the accuracy requirements for the relative position maintenance of formation flying satellite have also been increased from the meter level and decimeter level to the millimeter level, which raises the higher control accuracy of the relative position maintenance of the satellite formation. Therefore, it is necessary not only to offset the effects of the secular and longperiodic contributions due to perturbation, but also to compensate for the shortperiodic impact of perturbation. Hence, it is important to conduct a detailed analysis of the shape of the trajectory of formation flying satellites under the impact of perturbation. Among the many perturbations, the influence of the term perturbation occupies a major position. So this paper’s aim is mainly to conduct a detailed analysis of the influence of the perturbation on the relative position of the leaderfollower satellites.
A lot of research efforts had been done in the field of satellite relative motion under the influence of perturbation based on state transition matrices (STMs) and relative orbital elements (ROEs) [2]. Schweighart and Sedwick [3] derived a linear model by incorporating the firstorder term of the perturbation into the CW equation with a meterlevel position error when compared with a numerical integration. Izzo et al. [4] considered the atmospheric drag and perturbation in the extension of the CW model and used the Floquet theory to predict the relative motion under the perturbations with a higher accuracy than that of the CW equation. Stringer et al. [5] used the quadratic Volterra method developing a nonlinear model for including the perturbation with submeterlevel accuracy.
Unlike the above research, Schaub and Alfriend [6] and Gim and Alfriend [7] considered that the ROEs vary slowly with time and developed the ROEbased equations for relative motion under perturbation. Schaub and Alfriend [6] used the perturbed Hamiltonian deriving a relative state equation defined by Delaunay element differences and explored the impact of perturbation on relative orbits. Gim and Alfriend [7] developed a state transition matrix (STM) including the effect of secular, long, and shortperiodic contributions due to to the first order. Gim and Alfriend [8] then developed a model that can avoid the singularity on equatorial orbits by using differential equinoctial elements. D’Amico [9] developed a linear relative motion model which captures the time evolution of quasinonsingular ROE subjects (relative semimajor axis, the relative mean longitude, the relative eccentricity, and inclination vectors) to firstorder perturbations in nearcircular orbits. By using the same ROE state, Gaias et al. [10] developed a more complete relative motion model which corrects the original formulation shortcomings of D’Amico’s [9] work and includes the effects of atmospheric drag, with a higher accuracy than that of D’Amico’s [9] model. In addition, Mahajan et al. [11] developed a STM for relative motion under the complete zonal perturbations based on the studies of Gim and Alfriend [7]. More recent work by Russell et al. [12] developed an analytical STM for relative motion including and perturbations by using the Vinti theory. Burnett et al. [13] derived a new linearized differential equation model including zonal harmonic perturbations and corresponding perturbation solutions for relative motion without averaging the perturbing acceleration or its gradient. The perturbation solution of their model depends only on time, the initial chief orbital elements, and the deputy initial conditions in the chief Hill frame and has a similar error to the GASTM for zero initial chief orbit eccentricity.
As a nonlinear extension to the ROEbased linear relative motion, Alfriend [14] and Alfriend and Yan [15] derived a nonlinear model for propagating the mean OED in arbitrarily eccentric orbits while including secondorder effects. Based on the GimAlfriend’s studies, Sengupta et al. [16] developed a secondorder state transition tensor (STT) and a nonlinear solution including the effects of perturbation by combining the Keplerian STT with GimAlfriend’s [7] linear model. Then, Yang et al. [17] derived a more complete secondorder STT for relative motion under the perturbed elliptic orbits based on the Sengupta et al.’s [16] work. The two studies above [16, 17] did not consider the effect of perturbation of the secondorder transformation from the rectilinear relative state to the osculating ROEs and the secondorder propagation of the mean ROEs. Similar to the methods in [16, 17], Zhen et al. [18] developed a secondorder analytical STT for relative motion under the perturbed elliptic orbits by using the geometric method, and the result is more accurate than that of previous linear or nonlinear analytic methods. Moreover, Mahajan et al. [19] considered the zonal harmonic and sectorial and tesseral harmonics in their work, presenting a STM for perturbed satellite relative motion with the position errors below 1% of the formation size after two days in both the lowly and highly eccentric reference orbits. Recently, Gaias et al. [20] presented an analytical framework for the precise modeling of the relative motion in low earth orbits. Their model includes the firstorder expansion of the effects due to any even zonal harmonics and the secondorder expansion of the unperturbed and terms and can achieve highprecision mean/osculating orbital element conversions.
In addition, in order to achieve highprecision relative orbit maintenance, some scholars studied the hovering formation and orbit determination on the basis of the above dynamic research in recent years. Rao et al. [21] obtained a “teardrop” hovering formation by designing a set of relative orbit elements and proposed a new impulsive control strategy to keep the deputy satellite in the hovering pattern for a long time. Sun et al. [22] proposed an innovative orbit determination method which makes use of gravity gradients for low earthorbiting satellites. Bai et al. [23] proposed a teardrop hovering formation design for the elliptical reference orbit with a perturbation.
In response to the abovementioned background and research status, this paper studies the shape of relative motion of the leaderfollower satellites in nearcircular orbit under the effect of perturbation. On the basis of the predecessors, an equation of motion that can describe the relative motion of the leaderfollower satellites in nearcircular orbit under the influence of perturbation is derived, and the characteristics of inplane periodic motion of the leaderfollower satellites under the influence of term perturbation are analyzed. The equation of motion derived in this paper has a very simple form and is more desirable and achievable in onboard computers. By comparing with the result of numerical simulations under the same conditions, the relative position error of this equation is less than 20 mm in the radial direction and 60 mm in the tangential direction in a day. The major contributions of this study are summarized as follows:(1)We derive a highprecision equation of motion that can describe the relative motion of the leaderfollower satellites in nearcircular orbit under the influence of perturbation. The equation has a very simple form and is more desirable and achievable in onboard computers, so it can be used as a state equation for highprecision relative navigation(2)We analyze the characteristics of relative motion of leaderfollower satellites under the influence of perturbation, which can provide a theoretical basis for the design of controllers of mmlevel relative orbit maintenance
The outline of this paper is as follows: the coordinate system and the mathematical description of formation flying are introduced in Section 2. An equation of motion that can describe the relative motion of the leaderfollower satellites in nearcircular orbit under the influence of perturbation is derived, and the characteristics of inplane periodic motion of the leaderfollower satellites under the influence of term perturbation are analyzed in Section 3. Numerical simulations used to validate the results in this paper are shown in Section 4. In Section 5, we summarize the study and propose a new method for future relative orbit maintenance.
2. The Mathematical Description of Formation Flying
In order to derive the equation of the motion of leaderfollower satellites under the influence of the relative perturbation, it is first necessary to establish a coordinate system and a mathematical model to describe the relative motion [24] of the satellite formation. As shown in Figure 2, denote the chief satellite as and the deputy satellite as . The orbital coordinate system of the chief satellite is defined as follows: the origin is located at the center of the chief satellite and moves with it, the axis takes the radial direction of the reference satellite and points from the center of the earth to , the axis points to the positive normal direction of the chief satellite orbital plane, and the axis is determined by the righthand rule.
Assuming that the chief satellite is moving in a nearcircular orbit, the following CW equation [25] can be obtained to describe the relative motion while ignoring highorder small quantities (since the distance between the chief satellite and the deputy satellite is much smaller than the geocentric distance of the chief satellite , we can ignore the and higher power terms):where is the relative position of the deputy satellite in the orbital coordinate system of the chief satellite , is the average angular velocity of the chief satellite , is the geocentric gravitational constant, and is the semimajor axis of the chief satellite. , , and are the differences between the perturbation acting on the chief satellite and the deputy satellite, that is, the relative perturbation.
Taking the initial moment as the moment when the chief satellite is located at the RAAN (Right Ascension of Ascending Node), in a nearcircular orbit, the argument of latitude of the chief satellite is
Then, there is
In the context of this paper, the effect of the perturbation on the argument of latitude of the satellite is secondary, so we can neglect it and substitute formula (3) into formula (1); then, the independent variable of the CW equation can be changed to , and we have
In order to facilitate the analysis of the motion deviation of the deputy satellite in the relative motion orbit under the influence of relative perturbation, let replace in the above formula; then, equation (4) can be changed towhere , , , is the position of the deputy satellite in the relative motion orbit only under the influence of twobody gravity, and is the actual position of the deputy satellite in the relative motion orbit under the influence of relative perturbation and twobody gravity. The solution of equation (5) can describe the relative motion only under the influence of relative perturbation.
3. Relative Perturbation Analysis
In this section, the relative perturbation method is used to derive the equation of relative motion of the leaderfollower satellites under the influence of perturbation. The gravitational potential function of the zonal harmonic term of the earth [26] iswhere is the radius of the earth, is the geocentric distance and geocentric latitude of the spacecraft in the groundfixed coordinate system, and , , and are the coefficients of zonal harmonic term perturbation.
Then, the perturbation potential function of (the central gravitational term is omitted) is
The components of the perturbation force in spherical coordinates are
From the relationship between satellite spherical coordinates and orbital coordinates, it can be seen that the radial perturbation forces under satellite spherical coordinates and orbital coordinates are the same, and the tangential and normal perturbation forces and have the following conversion relationshipwhere is the angle between the orbital plane and the meridian plane of the point where the satellite is located. From the spherical triangle relationship, we have
From equations (6), (7), (8), (9), and (10), we have the perturbation force of the zonal harmonic term in orbital coordinates:where and and are the inclination and argument of latitude of the spacecraft in the groundfixed coordinate system.
In equation (11), by taking partial differentiation of , , and , we can obtain the following equation:
In the above formula, is the difference of the geocentric distance between the deputy satellite and the chief satellite, is the difference of the inclination between the deputy satellite and the chief satellite, and is the difference of the argument of latitude between the deputy satellite and the chief satellite.
As we all know, the orbit elements of leaderfollower satellites are only different in argument of latitude. Therefore, in a nearcircular orbit, there are no differences in the geocentric distance and inclination between the chief and deputy satellite, that is, . Then, equation (12) can be changed to
Assuming that the deputy satellite has no relative position and velocity deviation between the influence of relative perturbation and twobody gravity at the initial moment, the initial value is
Substituting equations (14) and (15) into equation (5), we can get the following analytical solution:
In formulas (16), (17), and (18), is the inplane motion trajectory of the leaderfollower satellites in nearcircular orbit only under the influence of perturbation, and reflects the outofplane motion only under the influence of perturbation. It can be seen from the above formula that the relative position deviation in the three directions under the influence of perturbation is positively correlated with the difference of argument of latitude between the deputy satellite and the chief satellite. Therefore, the greater the distance between the leaderfollower satellites, the greater the relative position error of the formation flying satellite.
Because the relative motion of shortrange leaderfollower satellites in the direction under the influence of perturbation is less than 1 mm, we mainly discuss inplane motion. For inplane motion, let
Then, formulas (16) and (17) can be simplified as
The parametric equation represented by equation (20) describes the inplane motion of leaderfollower satellites in nearcircular orbit only under the influence of perturbation. In equation (20), by finding the firstorder and secondorder derivatives for , we can obtain the velocity and acceleration of the inplane motion only under the influence of perturbation as follows:
By analyzing equations (20), (21), and (22), it can be obtained that the inplane motion of leaderfollower satellites in nearcircular orbit only under the influence of perturbation has the following law:(a)The curve described by the parametric equation (20) is a symmetrically closed “dropshaped” line as shown in Figure 3, and is the symmetry axis of the curve. Let be the starting time of the inplane motion, which corresponds to in the figure. The deputy satellite moves periodically in the clockwise direction from point , and the period is . The detailed parameters at the special points in the motion are shown in Table 1(b)In formula (20), the parameter “” determines the magnitude of the inplane motion only under the influence of perturbation. When the distance between the leaderfollower satellites is certain, the magnitude of “” is also related to the orbital altitude and the inclination of the chief satellite. As the orbital altitude increases, “” continues to decrease, and the inplane relative position deviation of the leaderfollower satellites also decreases. When , “” increases with the increase in inclination , and the inplane relative position deviation of leaderfollower satellites also increases. When , the inplane relative position deviation reaches its maximum value. When , “” decreases as the inclination increases, and the inplane relative position deviation of the leaderfollower satellites also decreases(c)Figures 4 and 5 show the velocity and acceleration of the deputy satellite’s inplane motion in nearcircular orbit only under the influence of relative perturbation at an orbital altitude of 500 km. It can be seen from the figure that in order to control this inplane motion within the mm level, the required acceleration is of the order of , and the thruster is required to provide a continuous thrust of about 0.1 mN. In order to eliminate this movement caused by the relative perturbation of , more frequent thruster control must be relied on. The longterm frequent jets of the thruster will increase fuel consumption and greatly reduce the service life of the satellites. Therefore, at this stage, simply using the thruster as the actuator cannot achieve mmlevel longterm relative position maintaining control

4. Simulation
According to the background, the simulation parameters select the commonly used orbital parameters for earth observation using synthetic aperture radar satellites. The two satellites are, respectively, selected for simulation in the form of forward and backward flying formation with a distance of 100 m and 200 m. Select a lowearth circular orbit with an orbit altitude of 500 km, and the remaining initial orbit parameters are shown in Tables 2 and 3. The numerical simulator takes the initial analytic equations of motion as described in equation (23) and integrates the absolute motion of each satellite in time. The motion of each satellite is then differenced and converted to a coordinate system that was mentioned in Section 2.


In order to eliminate the error of integration, we take the initial analytic equations of motion as described in equation (24) as a numerical simulator and integrate the absolute motion of each satellite in time once more and perform the same processing as above. By subtracting the above two results, the relative motion result of the satellites only under the perturbation can be obtained, which can be compared with the analytical solution result in this paper. By comparing the conclusions obtained in this paper with the results of the numerical simulation, the correctness of the results obtained in this paper is verified.
The simulation results shown in Figures 6(b) and 7(b) show that under the influence of the relative perturbation of , the inplane periodic motion of the leaderfollower satellites is a symmetrical closed “dropshaped” trajectory, which is coincident with the results of numerical simulations (the results shown in Figures 6(a) and 7(a)), and the relative position error of this equation is less than 20 mm in the radial direction and 60 mm in the tangential direction in a day, as shown in Figure 8.
(a) The result of numerical simulation
(b) Analytical solution result
(a) The result of numerical simulation
(b) Analytical solution result
5. Conclusions
In this paper, the relative perturbation method is used to study the influence of the relative perturbation of zonal harmonic term perturbation on the relative motion of leaderfollower satellites under the requirement of millimeterlevel accuracy in nearcircular orbit. Studies have shown that under the influence of the relative perturbation, the inplane periodic motion of the leaderfollower satellites in nearcircular orbit is a symmetrical closed “dropshaped” trajectory with a period of . In order to eliminate this motion caused by the relative perturbation, more frequent thruster control must be relied on. The longterm frequent jets of the thruster will increase fuel consumption and greatly reduce the service life of the satellite. Therefore, only using a thruster as the actuator to maintain the relative position can no longer meet the requirements of the longterm mmlevel relative position maintenance. In the future, we can follow the principle of flywheel angular momentum exchange and use the principle of linear momentum exchange for relative position control. By installing a twodimensional sliding mechanism on the two satellites and controlling the sliding mechanism to move, the linear momentum exchange caused by the continuous motion of the sliding mechanism can change the relative position of the formation satellites. Since the relative motion under the influence of perturbation is a continuous closed periodic trajectory and the motion amplitude is limited, this method can well offset this “dropshaped” motion. Moreover, this method does not need to carry additional fuel and can provide continuous control power only by relying on electric energy. It has advantages over the use of thrusters in terms of accuracy and fuel saving.
Data Availability
The data used to support the findings of this study are available from the author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This study was supported by the National Defense Science and Technology Project Foundation of China (Grant No. 0303002).
References
 Y. L. Zhang, G. Q. Zeng, and Z. K. Wang, Theory and Application of Distributed Satellite System, Science Publication, Beijing, China, 2008.
 J. Sullivan, S. Grimberg, and S. D’Amico, “Comprehensive survey and assessment of spacecraft relative motion dynamics models,” Journal of Guidance, Control, and Dynamics, vol. 40, no. 8, pp. 1837–1859, 2017. View at: Publisher Site  Google Scholar
 S. A. Schweighart and R. J. Sedwick, “Highfidelity linearized J model for satellite formation flight,” Journal of Guidance, Control, and Dynamics, vol. 25, no. 6, pp. 1073–1080, 2002. View at: Publisher Site  Google Scholar
 D. Izzo, M. Sabatini, and C. Valente, “A new linear model describing formation flying dynamics under J2 effects,” in Proceedings of the 17th AIAA Congress, pp. 15–19, Rome, Italy, January 2003. View at: Google Scholar
 M. T. Stringer, B. Newman, T. A. Lovell, and A. Omran, “Analysis of a new nonlinear solution of relative orbital motion,” in The 23rd International Symposium on Space Flight Dynamics, pp. 1–18, Pasadena, CA, USA, 2012. View at: Google Scholar
 H. Schaub and K. T. Alfriend, “J2 invariant relative orbits for spacecraft formations,” Celestial Mechanics and Dynamical Astronomy, vol. 79, no. 2, pp. 77–95, 2001. View at: Publisher Site  Google Scholar
 D. W. Gim and K. T. Alfriend, “State transition matrix of relative motion for the perturbed noncircular reference orbit,” Journal of Guidance Control, and Dynamics, vol. 26, no. 6, pp. 956–971, 2003. View at: Publisher Site  Google Scholar
 D. W. Gim and K. T. Alfriend, “Satellite relative motion using differential equinoctial elements,” Celestial Mechanics and Dynamical Astronomy, vol. 92, no. 4, pp. 295–336, 2005. View at: Publisher Site  Google Scholar
 S. D’Amico, Autonomous formation flying in low earth orbit, [Ph.D. thesis], Delft University of Technology, Delft, The Netherlands, 2010.
 G. Gaias, J. S. Ardaens, and O. Montenbruck, “Model of J2 perturbed satellite relative motion with timevarying differential drag,” Celestial Mechanics and Dynamical Astronomy, vol. 123, no. 4, pp. 411–433, 2015. View at: Publisher Site  Google Scholar
 B. Mahajan, S. R. Vadali, and K. T. Alfriend, “Analytic solution for the satellite relative motion: the complete zonal gravitational problem,” Advances in the Astronautical Sciences, vol. 158, pp. 1–24, 2016. View at: Google Scholar
 A. D. Biria and R. P. Russell, “A satellite relative motion model including J2 and J3 via Vinti’s intermediary,” Celestial Mechanics and Dynamical Astronomy, vol. 130, no. 3, pp. 1–23, 2018. View at: Google Scholar
 E. Burnett, E. Butcher, A. J. Sinclair, and T. A. Lovell, “Linearized relative orbital motion model about an oblate body without averaging,” Advances in the Astronautical Sciences, vol. 167, no. AAS 18218, pp. 691–710, 2018. View at: Google Scholar
 K. T. Alfriend, “Nonlinear considerations in satellite formation flying,” in AIAA/AAS Astrodynamics Specialist Conference and Exhibit, pp. 2002–4741, Monterey, CA, USA, 2002. View at: Google Scholar
 K. T. Alfriend and H. Yan, “An orbital elements based approach to the nonlinear formation flying problem,” in Proceedings of the International Formation Flying Conference: Missions and Technologies, pp. 44–58, Toulouse, France, October 2002. View at: Google Scholar
 P. Sengupta, S. R. Vadali, and K. T. Alfriend, “Secondorder state transition for relative motion near perturbed, elliptic orbits,” Celestial Mechanics and Dynamical Astronomy, vol. 97, no. 2, pp. 101–129, 2007. View at: Publisher Site  Google Scholar
 Z. Yang, Y. Z. Luo, V. Lappas, and A. Tsourdos, “Nonlinear analytical uncertainty propagation for relative motion near J2perturbed elliptic orbits,” Journal of Guidance, Control, and Dynamics, vol. 41, no. 4, pp. 888–903, 2018. View at: Google Scholar
 Y. Zhen, L. YaZhong, and Z. Jin, “Secondorder analytical solution of relative motion in J2perturbed elliptic orbits,” Journal of Guidance, Control, and Dynamics, vol. 41, no. 1, pp. 1–13, 2018. View at: Google Scholar
 B. Mahajan, S. R. Vadali, and K. T. Alfriend, “Statetransition matrix for satellite relative motion in the presence of gravitational perturbations,” Journal of Guidance, Control, and Dynamics, vol. 42, no. 7, pp. 1–17, 2019. View at: Google Scholar
 G. Gaias and C. Colombo, “Analytical framework for precise relative motion in low earth orbits,” Journal of Guidance, Control, and Dynamics, vol. 43, no. 1, pp. 1–13, 2020. View at: Google Scholar
 Y. Rao, J. Yin, and C. Han, “Hovering formation design and control based on relative orbital elements,” Journal of Guidance, Control, and Dynamics, vol. 152, no. 2, pp. 1–12, 2015. View at: Google Scholar
 X. Sun, P. Chen, C. Macabiau, and C. Han, “Lowearth orbit determination from gravity gradient measurements,” Acta Astronautica, vol. 123, pp. 350–362, 2016. View at: Publisher Site  Google Scholar
 S. Bai, C. Han, X. Sun, H. Zhang, and Y. Jiang, “Teardrop hovering formation for elliptical orbit considering J2 perturbation,” Aerospace Science and Technology, vol. 106, article 106098, 2020. View at: Google Scholar
 X. N. Xi and W. Wang, Orbital Basis for NearEarth Spacecraft, National University of Defense Technology Publication, Hunan, China, 2003.
 W. H. Clohessy and R. S. Wiltshire, “Terminal guidance system for satellite rendezvous,” Journal of the Aerospace Sciences, vol. 27, no. 9, pp. 653–658, 1960. View at: Publisher Site  Google Scholar
 Z. W. Zhang, Satellite Orbit Attitude Dynamics and Control, Beijing University of Aeronautics and Astronautics Publication, Beijing, China, 1998.
Copyright
Copyright © 2021 Jingji Wang and Chunyang Liu. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.