Content uploaded by James Horton
Author content
All content in this area was uploaded by James Horton on Apr 11, 2019
Content may be subject to copyright.
131
AAS 18-034
EARTH TO MARS ABORT ANALYSIS FOR
HUMAN MARS MISSIONS
C. Russell Joyner II,* James F. Horton,† Timothy Kokan,‡
Daniel J. H. Levack§ and Frederick Widman**
Future human exploration missions to Mars are being studied by NASA and in-
dustry. Several approaches to the Mars mission are being examined that use var-
ious types of propulsion for the different phases of the mission. The choice and
implementation of certain propulsion systems can significantly impact mission
performance in terms of trip time, spacecraft mass, and especially mission abort
capability. Understanding the trajectory requirements relative to the round-trip
Earth to Mars mission opportunities in the 2030’s and beyond is important in
order to determine the impact of trajectory abort capability. Additionally, some
propulsion choices for the crew vehicle can enable mission abort trajectories
while others will most likely provide less flexibility and increase mission risk.
This paper focuses on recent modeling of Earth to Mars abort scenarios for hu-
man missions to determine the capability to provide fast returns to Earth. The
modeling assumed that the abort would occur after the Mars crew vehicle has
been injected along the path to Mars (i.e., after the Trans Mars Injection (TMI)
burn). These aborts have been defined as well as the timing of fly-by aborts to
quickly return crew to Earth.
These abort trajectory studies are based on missions NASA defined during the
Evolvable Mars Campaign (EMC) with crew going to Mars in 2033, 2039, 2043
and 2048. Detailed trajectory analysis was performed with the NASA Coperni-
cus program for the several crew missions that were in the EMC as well as other
new missions being considered using finite-burn low thrust electric propulsion.
The goal was to determine how the heliocentric trajectory elements change and
the “abort trajectory” impulse requirements.
Abort scenarios that were studied included fast returns N-days after TMI as well
as fly-by aborts and multiple revolution cases, using all available propellants
(e.g., main propulsion system and reaction control system (RCS)) to provide the
required abort velocity change. Trajectories were investigated for impulsive ma-
neuvers and for finite burn cases and the abort timelines for each are examined
and compared.
This paper and presentation will focus on the Copernicus trajectory analysis re-
sults that were performed to determine the abort trajectories that altered the pri-
mary mission to return to Earth as soon as possible.
* Fellow, Mission Architecture, Aerojet Rocketdyne, 17900 Beeline Hwy, M/S 712-67, Jupiter, Florida 33478, USA.
† Specialist Engineer, Mission Architecture, Aerojet Rocketdyne, P.O. Box 7922, Canoga Park, California 91309, USA.
‡ Specialist Engineer, Mission Architecture, Aerojet Rocketdyne, 555 Discovery Dr., Huntsville, Alabama 35806, USA.
§ Sr. Manager, Advanced Space and Launch, Aerojet Rocketdyne, P.O. Box 7922, Canoga Park, California 91309, USA.
** Specialist Engineer, Mission Architecture, Aerojet Rocketdyne, 555 Discovery Dr., Huntsville, Alabama 35806, USA.
132
NOMENCLATURE
AR = Aerojet Rocketdyne
DSM = Deep Space Maneuver
EOI = Earth Orbit Insertion (event where the Mars vehicle stops in orbit around Earth)
EMC = Evolvable Mars Campaign
ISP = Specific Impulse, a measure of efficiency (thrust per pound of fuel burned),
km = Kilometers
lbf = pounds force (force measurement for thrust)
LEO = Low Earth Orbit
LEU = Low Enriched Uranium
LDHEO = Lunar Distant Highly Elliptical Orbit
MOI = Mars Orbit Insertion (the event where the Mars vehicle stops in orbit around Mars)
MSC = Mars Study Capability
mT = Metric ton (1,000 kilogram or approximately 2,200 pounds)
MTV = Mars Transfer Vehicle (in reference to the crew vehicle for the Mars mission)
NASA = National Aeronautics and Space Administration
NTP = Nuclear Thermal Propulsion
SLS = Space Launch System
SOL = Term for 1 Solar Day at Mars (approximately 24.65 hours)
TEI = Trans-Earth Injection (the event that send the Mars vehicle back to Earth)
TMI = Trans-Mars Injection (the event that send the Mars vehicle from Earth to Mars)
VGA = Venus Gravity Assist
INTRODUCTION AND BACKGROUND
Engineering for crewed spaceflight is the art of designing affordable solutions that mini-
mize risk to human health and safety above all else. Any long duration transits beyond low earth
orbit to Mars will force astronauts to deal with the negative health effects of zero gravity and
deep space radiation. Beyond the physiology, logistical and engineering risks also exist during the
mission in relation to any of the supporting spaceflight systems before the crew reaches their des-
tination. For NASA’s Evolvable Mars Campaign (EMC), described in Percy et al.1, that could be
a failure in the prepositioned descent/ascent vehicle, the return propellant or any critical system
during flight of the crewed Mars Transfer Vehicle (MTV). When traded early in the architectural
and conceptual design phase, a choice of higher-efficiency MTV propulsion has the ability to
minimize those risks for the crew in terms of enabling shorter transit time, reduction in preposi-
tioned elements, or providing in-flight abort capability. One promising near-term technology is
Nuclear Thermal Propulsion (NTP) with its extensive test history, specific impulse (ISP) around
900 seconds, and affordability through a Low Enriched Uranium (LEU) construction2. Recent
internal studies have shown that an in-line MTV using a cluster of NTP engines, as conceptually
envisioned in Figure 1, can be assembled in orbit using only three to four Block 2 Space Launch
System (SLS) launches in the 8.4m payload fairing configuration. However, any selection of pro-
pulsion beyond traditional chemical systems will require investment and development before they
fly. For that reason, Aerojet Rocketdyne (AR) and NASA are examining the impacts of NTP ad-
vanced propulsion designs to mitigate risk, and cost (e.g. reduction in campaign launches), when
compared to alternate architectures AR is also investigating that use chemical rocket engines.
Ultimately any engineering evaluation against the baseline requires a detailed understand-
ing of the impulsive requirements (i.e., delta-v or Δv) of the nominal and corresponding abort
maneuvers. An MTV utilizing a higher ISP propulsion system, with increased impulsive capabil-
ity, will have the ability to take a faster nominal path to Mars or use its propellant reserve to un-
dergo a propulsive maneuver to return to Earth. With the impulsive requirements well understood
133
for nominal trajectories over the Earth-Mars synodic period using Lambert’s problem and other
Astrodynamics codes3, this paper focuses on understanding the impact of return-to-earth abort
trajectories on a mission’s delta-v budget. As described in previous AR abort modelling4, the au-
thors used NASA’s Copernicus trajectory design and optimization system to assess the heliocen-
tric deep space maneuvers (DSM) and Mars flyby trajectories. The Copernicus software is ideally
suited to model higher fidelity interplanetary trajectories as it can account for n-body gravitation-
al perturbations5, whereas a simple Lambert solution will under predict the required impulse of a
maneuver due to propagated errors in the position and velocity of the MTV over long flight dura-
tions under perturbing third body accelerations.
Figure 1: AR’s In-line Mars Transfer Vehicle concept using Low-Enriched Uranium NTP
BASELINE EMC ARCHITECTURE
For this assessment, AR recreated in Copernicus the nominal trajectories outlined in
NASA’s EMC architecture, as seen in Figure 2, by anchoring to epoch and delta-v values provid-
ed in Percy et al1. In the EMC a crew of four would travel to Phobos in 2033 with follow on mis-
sions to Mars surface starting in 2039, 2043, and 2048. All crew opportunities utilize conjunc-
tion-class trajectories with a long surface stay on Mars. The astronauts would spend a minimum
of 300 days in Mars vicinity and be constrained to a maximum Earth-Mars round trip of 1100
days. After earth departure using a Trans Mars Injection (TMI) burn, 3-4 propulsive burns are
performed to complete the trip. At Mars there is a 1-SOL Mars Orbit Insertion (MOI) maneuver,
a Trans-Earth Injection (TEI) maneuver to leave Mars, and finally an Earth Orbit Insertion (EOI)
maneuver into Lunar Distant Highly Elliptical Orbit (LDHEO) on return to Earth. Depending on
the opportunity, there is either a midflight inbound or outbound broken plane6 DSM to lower the
overall MTV propellant load. This maneuver must be taken into account during mission planning
as it lowers the usable propellant for an abort maneuver. One major aspect of the architecture is
that key mission elements, such as ascent/descent vehicles or return-trip propellant, is ‘split’ from
the crew transit and prepositioned using unmanned robotic spacecraft. The prepositioned propel-
lant is due to the use of chemical (LOX/Methane) propulsion, with an ISP of ~360 s, for the base-
line EMC crew spacecraft known as the Methane Cryogenic Propulsion Stage (MCPS)7. When
134
NTP propulsion is used there is no prepositioned propellant. The higher specific impulse (900 s
versus 360 s) allows all the propellant to be carried on the MTV with modest penalty. Figure 2
shows the resulting delta-v’s available for abort for the chemical and the NTP approaches.
Figure 2: Post-TMI available delta-v comparison between EMC chemical1 and NTP MTVs
It should be noted that using a conjunction-class trajectory, as juxtaposed to an opposi-
tion-class mission with a short surface stay, lowers the chance of being able to perform beneficial
Venus gravity assist (VGA) without a propellant-consuming powered Mars flyby during an abort
scenario. Recently, NASA has been reevaluating the aforementioned trajectories under a follow-
on assessment known as the Mars Study Capability (MSC). Preliminary, this review could add
opportunities for crewed transit in 2037 and 2041. However, as that study is in-work, and there
has been no publically released trajectory information, those years were excluded for this paper.
Additionally, it is assumed that any abort scenario would return into a similar LDHEO
used in the nominal EMC return. To increase the timeframe of feasible aborts and to minimize the
EOI maneuver delta-v stored on the MTV, the final EOI inclination was allowed to float with re-
spect to the vehicle’s incoming hyperbolic trajectory. In a rescue situation, mission planning for
the crew must address the associated plane change if needed.
ABORT TRAJECTORY DESCRIPTIONS AND OPTIONS
Three general abort options exist during the interplanetary journey to Mars: direct return,
free return, and powered flyby8. A direct return utilizes a heliocentric DSM to lower the energy of
the nominal transfer orbit and set up intersection with Earth at a later date. Since Earth and Mars
are not coplanar, a portion of the maneuver is used for plane change as well. In Copernicus, the
burns were modeled as a single impulsive combined maneuver. The authors refer to this class of
aborts as type “A” and it is further described below. A free return uses the natural alignment of
the outbound Earth-Mars trajectory to return to earth if no MOI is performed. However free re-
turn trajectories are rare and not always economical in terms of launch energy8; none were identi-
fied in the modeling during this abort study for the EMC opportunities. Outside the scope of this
examination of EMC trajectories, if the launch date is flexible, Wooster et al.9 illustrates the im-
135
pulsive abort requirements for two and three year free return trajectories in 2033 and 2037. In-
cluding a burn during the Mars encounter is known as a powered flyby10 and can increase the
abort opportunities for the mission. In Copernicus this powered flyby was conservatively mod-
elled. If the decision to abort was made well in advance of the planned MOI, the orbital geometry
at encounter could be changed to further take advantage of the Mars gravitational assist to lower
the impulse required. In this investigation that class of aborts are known as type “B” and further
described below.
Figure 3: Type A fast return abort trajectories with annotated Copernicus modeling callouts
As seen in Figure 3, AR studied two cases of direct returns. An “A1” abort is a fast return
utilizing a non-tangential burn that occurs well before one revolution of the sun. As seen later in
the results, impulse curves are created by allowing the post DSM coast time to float and minimize
the phase and plane change required. Type “A2” aborts lower the heliocentric plane change cost
at the expense of time (~1 revolution around the sun). Total time of flight (TOF) for these trajec-
tories is defined as TMI to the abort event (e.g., DSM) plus the resulting coast time to EOI.
Figure 4 illustrates two of the three flyby cases studied. A type “B1” abort utilizes a sin-
gle powered Mars flyby to set up return to earth. However, two forms of it exist. The first “B1”
variation is where the spacecraft lowers heliocentric orbital energy and, typically, dips below the
orbit of Earth and Venus before EOI. The second variation is where the spacecraft increases its
orbital energy to fly beyond the orbit of Mars before its Earth encounter. As seen later in the re-
sults Tables 1 and 2, this allows the spacecraft to trade Earth entry velocity at the expense of the
Mars flyby maneuver. Type “B3” (not depicted) is similar to the aforementioned tactic but allows
~2 revolutions around the Sun post flyby to await improved alignment for Earth entry. However,
this comes at the expense of TOF. As studied in prior trajectory assessments (in Tartabini et al.8,
Hughes et al.11, and Okutsu & Loguski12), a type “B2” trajectory attempts to use a purely gravita-
tional unpowered VGA to lower the required abort delta-v. This requires targeting Venus during
the Mars flyby which may not be in optimal alignment compared with an opposition-class trajec-
tory not considered for EMC.
136
Figure 4: Type B Mars flyby abort trajectories with annotated Copernicus modeling callouts
RESULTS
Figure 5 provides an impulse requirement comparison over all the crewed EMC opportu-
nities to perform a type “A1” abort 10 and 30 days after TMI. The delta-v presented in all the re-
sults is any propulsive maneuver performed post TMI including the EOI burn. Additional time-
dependent curves for each individual EMC opportunity can be found in earlier publications (i.e.
Joyner et al.4). However a comparison across opportunities illustrates the natural fluctuation of
the data with the 7-8 year synodic cycle between Earth and Mars3. A vehicle sized to 3.5 km/s,
using both the main propulsion system and reaction control system (RCS) propellants, would al-
low fast returns up to 1 month post TMI. This is currently feasible for NTP powered spacecraft
sized to the 8.4m payload fairing configuration of the Block 2 SLS. Figure 6 shows the type “A2”
abort requirements across the EMC crewed opportunities. The figure illustrates a vehicle sized to
3.5 km/s would allow aborts up to 80 days after TMI while getting the crew home within 10 to 14
months after departure. This is far less than the ~3 year nominal mission duration. The 2039 op-
portunity struggles in both “A” cases with the ecliptic inclination differences between Earth and
Mars.
Impulse requirements for the various type “B” trajectories are presented in Table 1 and 2. The
data illustrates that only “B3” and select “B2” type trajectories come close to the delta-v’s that
can be packed on MTVs utilizing chemical and NTP propulsion using multiples SLS launches1,2.
However these suffer from trip times near, or exceeding, the nominal mission time ~1000 days. In
137
this instance it would make sense for the crew to perform MOI and wait to fly back on the nomi-
nal preplanned return trajectory. This would allow the crew to get partial radiation protection
from Mars/Phobos while in orbit instead of full exposure in deep space. Also there may be some
value-added science that can be performed in while in the Mars vicinity. This tactic would also
reduce the need to add additional radiation shielding for “B” trajectories that approach closer to
Venus and the Sun.
Figure 5: Type A1 fast return aborts options using nominal EMC trajectories
138
Figure 6: Type A2 fast return aborts options using nominal EMC trajectories
Table 1: Type B Mars flyby aborts options using nominal 2033 & 2039 EMC trajectories
Type
B1
Type
B1
Type
B2
Type
B2
Type
B3
Type
B1
Type
B1
Type
B2
Type
B2
Type
B3
Venus Gravity Assist? No No Yes Yes No No No Yes Yes No
Num ber of Revs ~1 ~1 ~1.5 ~1.5 ~2 ~1 ~1 ~1.5 ~2 ~2
Eart h-Mars TOF day s 231 231 231 231 231 350 350 350 350 350
ΔV at Mars (Fly-By, No
Mars Gravity Assist)
km/s
3.15 8.68 3.30 2. 84 1.64 8.96 9.63 7. 90 4.65 0. 44
Mars-Venus TOF days --- --- 171 229 --- --- --- 165 599 ---
ΔV at Venus km/s --- ---
DV pe r
VGA
DV pe r
VGA
--- --- ---
DV pe r
VGA
DV pe r
VGA
---
Mars- Earth TOF day s 241 579 -- - --- 914 180 467 -- - - -- 708
Venus-Earth TOF days --- --- 209 167 --- --- --- 111 168 ---
ΔV at Earth ( LDHEO) km /s 4. 19 1.15 2.66 0.57 0.60 15.25 0.89 0.62 1.08 1.51
Total Abort ΔV km/s 7.34 9.84 5.96 3. 41 2.24 24. 21 10.53 8.53 5. 73 1.96
Total TOF days 472 810 611 627 1145 530 817 626 1117 1058
2033
2039
139
Table 2: Type B Mars flyby aborts options using nominal 2043 & 2048 EMC trajectories
Type
B1
Type
B1
Type
B2
Type
B2
Type
B3
Type
B1
Type
B1
Type
B2
Type
B2
Type
B3
Venus Gravity Assist? No No Yes Yes No No No Yes Yes No
Num ber of Revs ~1 ~1.5 ~1.5 ~2 ~2 ~1 ~1.5 ~1. 5 ~2 ~1.5
Eart h-Mars TOF day s 289 289 289 289 289 235 235 235 235 235
ΔV at Mars (Fly-By, No
Mars Gravity Assist)
km/s
5.46 9.46 5.09 4. 93 4.63 3. 68 8.97 5. 48 2.83 3.51
Mars-Venus TOF days --- --- 270 261 --- --- --- 180 601 ---
ΔV at Venus km/s --- ---
DV pe r
VGA
DV pe r
VGA
--- --- ---
DV pe r
VGA
DV pe r
VGA
---
Mars- Earth TOF day s 217 534 -- - - -- 862 241 579 - -- - -- 607
Venus-Earth TOF days --- --- 156 456 --- --- --- 251 171 ---
ΔV at Earth (LDHEO) km/s 9.05 1.02 1.76 0.49 0.55 5.41 1.16 2.79 1.41 5.25
Total Abort ΔV km/s 14.51 10.47 6.85 5.42 5.18 9.09 10.13 8.27 4.23 8. 76
Total TOF days 506 823 715 1007 1151 476 814 666 1007 842
2043
2048
CONCLUSIONS
Building in the capability to perform type “A” fast aborts during the crewed transit to
Mars is one step engineers can take to minimize the inherent risks of interplanetary travel. The
impulse requirements presented in this study show that advanced technologies, such as NTP, are
potentially capable of adding that capability to NASA’s planned journey to Mars when compared
to the chemical baseline. The benefits of developing technology for abort trajectories goes beyond
just worse case scenarios; even if the technology is not used on the outbound journey, an NTP
system could reduce the overall mission time and health risks by shortening the inbound time on
the return home. Additionally the architecture would benefit from the cost savings of reducing
SLS launches for prepositioning propellant that would not be needed with the NTP MTV. Type
“B” aborts are feasible for only some EMC opportunities (e.g., 2033) if 1.5 years to return is ac-
ceptable. When low trip time (e.g., less than 1 year) is required and low delta-v cost is considered,
Type “B” aborts cannot be performed with a minimum delta-v optimized MTV. There may be
compromises that allow the use of type “B” aborts if the launch date is allowed to change and the
vehicle is designed to accommodate the optimum Type “B” for a particular mission opportunity;
however that optimization is outside the scope of this paper and should be a future topic for study.
REFERENCES
1. Percy, T., McGuire, M., and Polsgrove, T., "In-space Transportation for NASA’s Evolvable Mars Campaign,”
AIAA-2015-4519, AIAA Space Conference and Exposition, Pasadena, California, August 31st - September 2nd
2015.
2. Joyner II, C. R., Levack, D. J. H., Crowley, J., “Determining Mars Mission NTP Thrust Size and Architecture
Impact for Mission Options”, AIAA-2015-3956, 51st AIAA/ASME/SAE/ASEE Joint Propulsion Conference and
Exhibit, Orlando, Florida, July 28-30 2015.
3. Burke, L. M., Falck, R. D., and McGuire, M. L., “Interplanetary Mission Design Handbook: Earth-to-Mars Mis-
sion Opportunities 2026 to 2045”, NASA TM-2010-216764, Glenn Research Center, Cleveland, Ohio, October
2010.
4. Joyner, Claude R., James F. Horton, Timothy S. Kokan, Daniel J. Levack, and Frederick Widman. "Enabling Mul-
tiple Abort Strategies Using the NTP Approach for Human Mars Missions", AIAA-2017-5273, AIAA SPACE and
Astronautics Forum and Exposition, AIAA SPACE Forum, Orlando, FL, 2017.
140
5. Willams, J., Senent, J.S., Ocampo, C., Mathur, R. and Davis, E.C., “Overview and Software Architecture of the
Copernicus Trajectory Design and Optimization System”, 4th International Conference on Astrodynamics Tools
and Techniques, Madrid, Spain, May 2010.
6. Abilleira, F., "Broken-Plane Maneuver Applications for Earth to Mars Trajectories", NASA/CP-2007-214158,
20th International Symposium on Space Flight Dynamics, September 2007.
7. Percy, T., McGuire, M., and Polsgrove, T., "Combining Solar Electric Propulsion and Chemical Propulsion for
Crewed Missions to Mars." IEEE Aerospace 2015, Big Sky, Mt., March 7th -14th 2015.
8. Tartabini, P. V., Striepe, S. A., Powell, R. W., “Abort Options for Potential Mars Missions”, Journal of Spacecraft
and Rockets, Vol. 31, No. 4, July-August 1994, pp. 543-549
9. Wooster, P. D., Braun, R. D., Ahn, J., Putnam, Z. R., “Mission Design Options for Human Mars Missions”, The
International Journal of Mars Science and Exploration, doi:10.1555/Mars.2007.0002, 2007.
10. Ishimatsu, Takuto, Jeffrey Hoffman, and Olivier de Weck. "Method for Rapid Interplanetary Trajectory Analysis
using ΔV Maps with Flyby Options." Journal of the British Interplanetary Society 64 (2011), pp. 204-213.
11. Hughes, Kyle M., Peter J. Edelman, James Longuski, Mike Loucks, John P. Carrico, and Dennis A. Tito. "Fast
Mars Free-Returns via Venus Gravity Assist", AIAA-2014-4109, AIAA/AAS Astrodynamics Specialist Confer-
ence, AIAA SPACE Forum, San Diego, California, 2014.
12. Masataka Okutsu and James M. Longuski. "Mars Free Returns via Gravity Assist from Venus", Journal of Space-
craft and Rockets, Vol. 39, No. 1 (2002), pp. 31-36.
