Aaron J Rosengren
 Assistant Professor, AerospaceMechanical Engineering
 Assistant Professor, Applied Mathematics  GIDP
 (520) 6212235
 Aerospace & Mechanical Engr., Rm. 301
 Tucson, AZ 85721
 ajrosengren@email.arizona.edu
Biography
Is currently an Assistant Professor in the Aerospace and Mechanical Engineering Department at the University of Arizona, focusing on SSA and Astrodynamics research. Formerly, he was a Postdoctoral Researcher at the Aristotle University of Thessaloniki in the Department of Physics as part of the EU H2020 Project ReDSHIFT. He also spent two years as an Experienced Researcher at the Institute of Applied Physics "Nello Carrara" of the Italian National Research Council as part of the Stardust Asteroid and Space Debris Network. He earned a B.S. in Mathematics and a B.S. in Mechanical Engineering from the University of Missouri, and was awarded an M.S. and a Ph.D. in Aerospace Engineering Sciences from the University of Colorado at Boulder, working with Prof. Dan Scheeres as a member of the Colorado Center for Astrodynamics Research. Research interests include space situational awareness, orbital debris, planetary science, celestial mechanics, and dynamical astronomy
Awards
 Invited keynote speaker at the IAU Symposium 364, Multiscale (time and mass) dynamics of space objects
 Summer 2020
 2018 Junior Faculty Award for Excellence at the Student Interface
 AME, UA, Spring 2018
Interests
No activities entered.
Courses
202021 Courses

Advanced Astrodynamics
AME 559 (Fall 2020) 
Orbit Mech+Space Flight
AME 457 (Fall 2020) 
Orbit Mech/Space Flight
AME 557 (Fall 2020) 
Research
AME 900 (Fall 2020)
201920 Courses

Dynamics
AME 250 (Spring 2020) 
Graduate Seminar
AME 696G (Spring 2020) 
Research
AME 900 (Spring 2020) 
Thesis
AME 910 (Spring 2020) 
Directed Research
AME 492 (Fall 2019) 
Independent Study
AME 599 (Fall 2019) 
Independent Study
PHYS 599 (Fall 2019) 
Orbit Mech+Space Flight
AME 457 (Fall 2019) 
Orbit Mech/Space Flight
AME 557 (Fall 2019) 
Research
AME 900 (Fall 2019)
201819 Courses

Independent Study
PHYS 599 (Spring 2019) 
Research
AME 900 (Spring 2019) 
Thesis
AME 910 (Spring 2019) 
Advanced Astrodynamics
AME 559 (Fall 2018) 
Orbit Mech+Space Flight
AME 457 (Fall 2018) 
Orbit Mech/Space Flight
AME 557 (Fall 2018) 
Research
AME 900 (Fall 2018)
201718 Courses

Dynamics
AME 250 (Spring 2018) 
Independent Study
AME 599 (Spring 2018) 
Independent Study
AME 599 (Fall 2017) 
Orbit Mech+Space Flight
AME 457 (Fall 2017) 
Orbit Mech/Space Flight
AME 557 (Fall 2017)
Scholarly Contributions
Journals/Publications
 Amato, D., Bombardelli, C., Baú, G., Rosengren, A. J., & Morand, V. (2019). Nonaveraged regularized formulations as an alternative to semianalytical orbit propagation methods. Celestial Mechanics and Dynamical Astronomy.
 Lara, M., Rosengren, A. J., & Fantino, E. (2019). Nonsingular recursion formulas for thirdbody perturbations in mean vectorial elements,. Astronomy and Astrophysics. doi:10.1051/00046361/201937106
 Rosengren, A. J., Skoulidou, D., Tsiganis, K., & Voyatzis, G. (2019). Dynamical cartography of Earth satellite orbits. Advances in Space Research.More infoWe have carried out a numerical investigation of the coupled gravitational and nongravitational perturbations acting on Earth satellite orbits in an extensive grid, covering the whole circumterrestrial space, using a suitably modified version of the SWIFT symplectic integrator, which is suitable for longterm ($\sim$120~y) integrations of the nonaveraged equations of motion. Hence, we characterize the longterm dynamics and the phasespace structure of the Earthorbiter environment, starting from low altitudes ($\sim$400~km) and going up to the GEO region and beyond. This investigation was done in the framework of the ECfunded ``ReDSHIFT'' project, with the purpose of enabling the definition of passive debris removal strategies, based on the use of physical mechanisms inherent in the complex dynamics of the problem (i.e., resonances). Accordingly, the complicated interactions among resonances, generated by different perturbing forces (i.e., lunisolar gravity, solar radiation pressure, tesseral harmonics in the geopotential) are accurately depicted in our results, where we can identify the regions of phase space where the motion is regular and longterm stable and region for which eccentricity growth and even instability due to chaotic behavior can emerge. The results are presented in an ``atlas'' of dynamical stability maps for different orbital zones, with a particular focus on the (dragfree) range of semimajor axes, where the perturbing effects of the Earth's oblateness and lunisolar gravity are of comparable order. In some regions, the overlapping of the predominant lunisolar secular and semisecular resonances furnish a number of interesting disposal hatches at moderate to low eccentricity orbits. All computations were repeated for an increased areatomass ratio, simulating the case of a satellite equipped with an onboard, areaaugmenting device. We find that this would generally promote the deorbiting process, particularly at the transition region between LEO and MEO. Although direct reentry from very low eccentricities is very unlikely in most cases of interest, we find that a modest ``deltav'' ($\Delta V$) budget would be enough for satellites to be steered into a relatively shortlived resonance and achieve reentry into the Earth's atmosphere within realistic timescales.
 Skoulidou, D. K., Rosengren, A. J., Tsiganis, K., & Voyatzis, G. (2019). Medium Earth Orbit dynamical survey and its use in passive debris removal. ADVANCES IN SPACE RESEARCH, 63(11), 36463674.
 Daquin, J., Gkolias, I., & Rosengren, A. J. (2018). Drift and its mediation in terrestrial orbits. Frontiers in Applied Mathematics and Statistics.More infoThe slow deformation of terrestrial orbits in the medium range, subject tolunisolar resonances, is well approximated by a family of Hamiltonian flow with$2.5$ degreeoffreedom. The action variables of the system may experiencechaotic variations and large drift that we may quantify. Using variationalchaos indicators, we compute highresolution portraits of the action space.Such refined meshes allow to reveal the existence of tori and structuresfilling chaotic regions. Our elaborate computations allow us to isolate preciseinitial conditions near specific zones of interest and study their asymptoticbehaviour in time. Borrowing classical techniques of phase spacevisualisation, we highlight how the drift is mediated by the complement of thenumerically detected KAM tori.
 Skoulidou, D. K., Rosengren, A. J., Tsiganis, K., & Voyatzis, G. (2018). Dynamical Lifetime Survey of Geostationary Transfer Orbits. Celestial Mechanics and Dynamical Astronomy.More infoIn this paper, we study the longterm dynamical evolution of highlyelliptical orbits (HEOs) in the mediumEarth orbits (MEO) region around the Earth. The real population consists primarily of Geosynchronous Transfer Orbits (GTOs), launched at specific inclinations, Molniyatype satellites and related debris. We performed a suite of longterm numerical integrations (up to 200 years) within a realistic dynamical model, aimed primarily at recording the dynamical lifetime of such orbits (defined as the time needed for atmospheric reentry)and understanding its dependence on initial conditions and other parameters, such as the areatomass ratio (A/m). Our results are presented in the form of 2D lifetime maps, for different values of inclination, A/m, and drag coefficient. We find that the majority of small debris (> 70%, depending on the inclination) can naturally reenter within 2590 years, but these numbers are significantly less optimistic for large debris (e.g., upper stages), with the notable exception of those launched from high latitude (Baikonur). We estimate the reentry probability and mean dynamical lifetime for different classes of GTOs and we find that both quantities depend primarily and strongly on initial perigee altitude. Atmospheric drag and higher A/m values extend the reentry zones, especially at low inclinations. For high inclinations, this dependence is weakened, as the primary mechanisms leading to reentry are overlapping lunisolar resonances. This study forms part of the ECfunded (H2020) “ReDSHIFT” project.
Proceedings Publications
 Rosengren, A. J., Amato, D., Bombardelli, C., & Jah, M. (2019, October). RESIDENT SPACE OBJECT PROPER ORBITAL ELEMENTS. In SPACEFLIGHT MECHANICS 2019, VOL 168, PTS IIV, 168, 23532364.
 Amato, D., Furfaro, R., Rosengren, A. J., & Maadani, M. (2018, September). Attitude propagation of resident space objects with recurrent neural networks. In Advanced Maui Optical and Space Surveillance Technologies Conference.
 Rosengren, A. J. (2018, September). Passive debris removal using orbital resonances. In Advanced Maui Optical and Space Surveillance Technologies Conference.
 Rosengren, A. J., Amato, D., & Bombardelli, C. (2018, 20180111). THALASSA: a fast orbit propagator for nearEarth and cislunar space. In AIAA SciTech Forum.More infoWe present THALASSA, a fast nonaveraged orbit propagation Fortran code for the EarthMoon system implementing regularized formulations of dynamics. THALASSA uses a variable stepsize and order solver to integrate Newtonian equations of motion, the KustaanheimoStiefel formulation, and a regularized set of orbital elements. It includes perturbations from a 15 x 15 geopotential, atmospheric drag, the Sun, the Moon, and SRP. Close encounters with the Moon are handled efficiently through a trajectory splitting algorithm, in which the primary body is switched during the integration. We compare our code to the semianalytical orbit propagator STELA in several numerical experiments. In the 75year propagation of a HEO orbit with large semimajor axis, the nonaveraged regularized element set implemented in THALASSA is able to accurately recover the position along the orbit with a CPU time of about 10 s, while the Cowell formulation and STELA fail to do so. For the 200year propagation of the IBEX spacecraft, in a 3 : 1 meanmotion resonance with the Moon, the qualitative evolution of the orbit can only be re produced with the nonaveraged formulations. The code also displays a satisfactory performance for navigational satellite orbits, in which STELA is more efficient. We expect additional gains in efficiency for Molniyatype orbits and in the estimation of lifetime for LEO orbits.
 Rosengren, A. J., Correa, J. R., & Scheeres, D. J. (2018, August). Mean values in elliptic motion: averaging the Legendre polynomials. In AAS/AIAA Astrodynamics Specialist Conference.
 Rosengren, A. J., Amato, D., Daquin, J., & Gkolias, I. (2017, 20170620). The Dynamical Placement of Satellite Constellations and Designing for Demise. In 9th International Workshop on Satellite Constellations and Formation Flying.More infoThe dynamical environment occupied by satellite constellations is subject to mo tions that have widely disparate timescales: the earthly day, the lunar month, the solar year, and various precession frequencies ranging from a few years to nearly 26 000 years for the equinoxes. This provision of frequencies in the EarthMoon Sun system gives rise to a diverse range of complex resonant phenomena associ ated with orbital motions. Resonances can have a profound effect on the longterm dynamics of the system, giving rise to a rich spectrum of highly complicated be haviors. Such resonant perturbations are the most important mechanism for deliv ering distant Earthorbiting satellites into the regions where atmospheric drag can start their decay. We have previously found that even in the mediumEarth orbit (MEO) region, without the destabilizing influence of drag, there exist many dy namical pathways that can be used to effectively clear this distant region of space from any future collision hazard. Building on the lessons learned in MEO, we introduce here a new prospective into the mission analysis and design of satellite constellations, showing how a more holistic approach that incorporates dynamics into the early design phase can aid efforts in space debris mitigation and remediation. We emphasize the importance of conducting (a priori) detailed dynamical surveys of the neighboring operational regions, and incorporating lifetime estimates as a new constraint in the launch window analysis. Such dynamical assessments could have a profound and tangible influence on constellation design, perhaps attacking the debris problem at its source
Presentations
 Amato, D., & Rosengren, A. J. (2018, 20180625). A nonaveraged approach to the numerical cartography of the LEO region. 5th European Workshop on Space Debris Modelling and Remediation. CNES HQ, Paris, France: CNES.
 Amato, D., Amato, D., Rosengren, A. J., Rosengren, A. J., Furfaro, R., & Furfaro, R. (2018, July). Solving the main problem in satellite theory through recurrent neural networks. 42nd Scientific Assembly of the Committee on Space Research (COSPAR). Pasadena, California.
 Namazyfard, H., Amato, D., & Rosengren, A. J. (2018, July). Lifetime analysis in launch window maps: designing satellite orbits for demise. 42nd Scientific Assembly of the Committee on Space Research (COSPAR). Pasadena, California.
 Namazyfard, H., Rosengren, A. J., & Amato, D. (2018, November). Lifetime analysis in launch window maps: designing satellite orbits for demise. CODER 2018 Workshop on Orbital Debris Education and Research. College Park, Maryland.
 Reiland, N., Amato, D., Bombardelli, C., & Rosengren, A. J. (2018, November). The dynamical placement of megaconstellations. CODER 2018 Workshop on Orbital Debris Education and Research. College Park, Maryland.
 Reiland, N., Aschenbrenner, M., Amato, D., & Rosengren, A. J. (2018, July). The dynamical placement of megaconstellations. 42nd Scientific Assembly of the Committee on Space Research (COSPAR). Pasadena, California.
 Rosengren, A. J. (2018, 20180615). Orbital Resonances and Averaging in the Motion of Satellites. Invited Seminar. Politecnico di Milano, Dipartimento di Scienze e Tecnologie Aerospaziali.More infoMany physical systems can be modeled as having an underlying dynamical skeleton that organizes and governs how all the possible behaviors are related. The global properties of multidimensional, nearly integrable Hamiltonian systems are determined by the relative location and size of the predominant resonances. The dynamical model governing satellite motion (assuming noncommensurate orbital frequencies) is referred to in the astrophysical and celestial dynamics communities as the quadrupolar, secular, hierarchical restricted fourbody problem with an oblate primary. In the nonautonomous case, this model degenerates to either the classical KozaiLidov mechanism or the critical inclination resonances. In the timedependent model, brought about in this case by the Moon's perturbed motion, secular resonances involving the frequencies of perturbed motions become woven throughout the inclination, eccentricity, and semimajor axis space in an exceedingly complicated weblike structure, emanating from the classical critical inclinations. In this talk, I will review this 2.5 degreeoffreedom Hamiltonian system from both a Gauss averaging and LaplaceLagrange secular theory perspective. It is the structure of the satellite and the nature of its orbit that determine which perturbations are significant and which are negligible. In this sense, every distinct problem conditions its particular scheme of computation, and many refinements, sometimes reducing the always elaborate calculations in a marked degree, depend on a careful examination of the dynamical situation. I will show the breakdown of our basic dynamical model in the presence of resonances of a nonsecular origin and close encounters with the Moon.
 Rosengren, A. J. (2018, December). Dynamical phenomena in the Earth orbiter problem and their implications for celestial mechanics (Keynote). XIX Coloquio Brasileiro de Dinamica Orbital, CBDO, 2018. Sao Jose dos Campos, SP, Brazil.
 Rosengren, A. J. (2018, July). Artificial Earth satellites as a natural dynamical laboratory. invited talk presented at the CMS Summer School 2018 in Applied Mathematics. Technion, Haifa, Israel.
 Rosengren, A. J. (2018, November). Circumterrestrial orbital dynamics and new results for space situational awareness and space debris. invited research seminar in the Daniel Guggenheim School of Aerospace Engineering, Georgia Institute of Technology. Atlanta, Georgia.
 Rosengren, A. J. (2018, September). Physical aspects of satellite orbit prediction. invited research seminar in the Department of Physics Colloquium, University of Arizona. Tucson, Arizona.
 Rosengren, A. J., Daquin, J., & Gkolias, I. (2018, 20180419). Chaotic Transport in Circumterrestrial Orbits. AAS/DDA 2018 Meeting. Pasadena, CA: American Astronomical Society Division on Dynamical Astronom.More infoThe slow deformation of circumterrestrial orbits in the medium region, subject to lunisolar secular resonances, is well approximated by a Hamiltonian system with 2.5 degrees of freedom. This dynamical model is referred to in the astrophysical and celestial dynamics communities as the quadrupolar, secular, hierarchical threebody problem, and, in the nonautonomous case, gives rise to the classical KozaiLidov mechanism. In the timedependent model, brought about in our case by the Moon's perturbed motion, the action variables of the system may experience chaotic variations and large drifts due to the possible overlap of nearby resonances. Using variational chaos indicators, we compute highresolution portraits of the action space, revealing the existence of tori and structures filling chaotic regions. Our refined and elaborate calculations allow us to isolate precise initial conditions near specific areas of interest and to study their asymptotic behavior in time. We highlight in particular how the drift in phase space is mediated by the complement of the numerically detected KAM tori. Despite their reputed normality, Earth satellite orbits can possess an extraordinarily rich spectrum of dynamical behaviors, and, like the small body remnants of Solar system formation, they have all the complications that make them very interesting candidates for testing the modern tools of chaos theory.
 Rosengren, A. J., Skoulidou, D., Tsiganis, K., & Voyatzis, G. (2018, 20180626). Dynamical Systems Approach to Debris Mitigation and Remediation. 5th European Workshop on Space Debris Modelling and Remediation. CNES HQ, Paris, France: CNES.More infoThis work focuses on the evolution of satellites and space debris as dynamical systems with the overall intent being to identify and study the different `life cycles' which these bodies go through. Recent advances in our understanding of the dynamical processes that act on these artificial celestial bodies predict that they may undergo significant chaotic drifts in circumterrestrial phase space throughout their existence. We study the implications of coupled gravitational and nongravitational perturbations on their longterm orbital dynamics, and show how secular and semisecular resonances can profoundly affect the behavior of these bodies, in both dissipative and Hamiltonian settings. In this talk, we will present the most complete to date dynamical atlas of the entire usable circumterrestrial space, characterizing the longterm evolution of Earth satellites from LEO to GEO and beyond, for the purposes of passive debris mitigation and remediation. In some regions, the overlapping of the predominant resonances furnish a number of interesting disposal hatches at moderate to low eccentricity orbits. For satellites equipped with an onboard, areaaugmenting device to increase their areatomass ratios at the end of life, solar radiation pressure was found to generally promote the deorbiting process, particularly at the transition region between LEO and MEO. Here we will link these cartographic stability maps to the appropriate disposal strategy or deorbiting device for any operational orbit, and we highlight in particular how such dynamical assessments can have a profound and tangible influence on space debris remediation though the passive debris removal ideology. Although direct reentry from very low eccentricities is very unlikely in most cases of interest, we find that a modest ``deltav'' budget would be enough for satellites to be steered into a relatively shortlived resonance and achieve reentry into the Earth's atmosphere within realistic timescales. The solution to the debris proliferation problem throughout all space regions can only be found by coupling a deep understanding of the dynamical environments occupied by artificial satellites and space debris with technical, political, and legal solutions.
 Rosengren, A. J., Skoulidou, D., Tsiganis, K., & Voyatzis, G. (2017, 20170704). Killing Satellites with Resonances: The Dynamics of Passive Debris Removal. 13th Hellenic Astronomical Conference. Heraklion, Crete.More infoOne of the main goals of the space debris community is to determine how to prevent debris from becoming so populous that it adversely affects operational satellites. Recent efforts have explored more passive means to curtail the growth rate of the debris population, by seeking to cleverly exploit the dynamical instabilities brought on by resonant perturbations to deliver retired Earthorbiting satellites into the regions where atmospheric drag can start their decay. With a modest deltav budget, satellites can be steered into a shortlived resonance, or passive systems can be deployed at the end of life like solar sail devices to enhance solar radiation pressure to bring satellites down earlier than would otherwise be the case. Having previously characterized the dynamical architecture of the circumterrestrial environment, from very lowaltitude orbits up to the geostationary region and beyond, this talk will link these cartographic stability maps to the appropriate disposal strategy or deorbiting device for any operational orbit. We will highlight in particular how our previous dynamical assessments can have a profound and tangible influence on space debris mitigation though the passive debris removal ideology.
 Skoulidou, D., Rosengren, A. J., Tsiganis, K., & Voyatzis, G. (2017, 20170704). Dynamical Study of the NearEarth Space Environment for Passive Debris Removal. 13th Hellenic Astronomical Conference. Heraklion, Crete.More infoOnly during the past decade has the precarious state of the Earth's orbiting environment, permeated by clouds of space debris, been fully appreciated and understood. Both active and passive debris mitigation and removal techniques are designed to curtail the growth of the debris environment by limiting the amount of mass in preferential space regions that may lead to future collisions. The solution to the debris proliferation problem, however, can only be found by coupling these mitigation and remediation methods with a deep understanding of the dynamical environments occupied by artificial satellites and space debris. An attempt at this heuristic approach is presented herein. This work uses tools and practices that are common in celestial mechanics and dynamical systems theory, which have been applied successfully in studies on the longterm dynamics in the Solar System and on the design of satellite orbits, but have only just started to be applied to the dynamics of space debris in recent years, as did the need for extending our knowledge of these artificial celestial bodies on intervals longer than mission timescales. Though a cartography of stability maps, obtained using a suitably modified version of the SWIFT symplectic integration package (Levison \& Duncan, 1994), we characterize the whole circumterrestrial space from LEO to GEO for the purposes of passive debris removal using resonances. The overlap of the predominant lunisolar secular and semisecular resonances furnish a number of interesting disposal hatches at moderate and low eccentricity orbits. We show, furthermore, that increasing the satellite's areatomass ratio using a solar sail helps promote the deorbiting process, through coupled gravitational and radiation pressure perturbations.
Poster Presentations
 Amato, D., Rosengren, A. J., & Baù, G. (2018, 20180520). What happened to Luna3? A numerical exploration of cislunar dynamics. John L. Junkins Dynamical Systems Symposium. Texas A&M University.More infoLuna3 was the first spacecraft to perform a lunar flyby and to image the far side of the Moon. Launched in October 1959, it collided with the Earth in late March 1960. Such a short dynamical lifetime for a highaltitude orbit has often been attributed to an increase in eccentricity due to a LidovKozai cycle. We study the evolution of the Luna3 trajectory by generating accurate numerical solutions through the THALASSA regularized orbit propagation code, and comparing them to single and doubleaveraged integrations. Lunar close encounters, which cannot be reproduced by semianalytical integrations, affect the trajectory decisively by suppressing the doubleaveraged secular dynamics. Their manifestation is evident in the LidovKozai representations of the flow, where close encounters cause impulsive changes in the osculating elements that cannot be predicted by the doubleaveraged dynamics. In the quadrupolar approximation with only the Sun as a perturber, large amplitudes of shortperiodic terms (with respect to the Sun) cause flips of the orbital plane. Ultimately, we find that the demise of Luna3 was caused by a complex interaction between lunar close encounters, shortperiodic terms, and doubleaveraged dynamics. The spacecraft never described LidovKozai cycles in a rigorous sense, since its argument of perigee did not enter a libration regime.
 Rosengren, A. J. (2018, 20180520). Dynamical Systems Approach to AstrodynamicsBased Space Situational Awareness. John L. Junkins Dynamical Systems Symposium. Texas A&M University.More infoThis work focuses on the evolution of resident space objects as dynamical systems with the overall intent being to identify and study the different `life cycles' which these bodies go through. Recent advances in our understanding of the dynamical processes that act on these artificial celestial bodies predict that they may undergo significant chaotic drifts in circumterrestrial phase space throughout their existence. Here was the birth of a new ideology to remedy the space debris proliferation problem, based on a judicious use of the resulting instabilities to prescribe natural Earth reentry itineraries and navigate the phase space. We present here recent results on the passive debris removal paradigm, and show how we can `kill' satellites using resonances. This work uses tools and practices that are common in astrophysical and celestial dynamics, but have only just started to be applied to the dynamics of space debris in recent years, as did the need for extending our knowledge of these bodies on intervals longer than mission timescales.