An Investigation of Aerogravity Assist at and for Capture into Orbit About and 2nd International Planetary Probe Workshop August 2004, USA

Philip Ramsey(1) and James Evans Lyne(2)

(1)Department of Mechanical, Aerospace and Biomedical Engineering The University of Tennessee Knoxville, TN 37996 USA [email protected]

(2)Department of Mechanical, Aerospace and Biomedical Engineering The University of Tennessee Knoxville, TN 37996 USA [email protected]

ABSTRACT proposed that an aerogravity assist maneuver at the moon Titan could be used to capture a probe Previous work by our group has shown that an into orbit about Saturn, using an aeroshell with a aerogravity assist maneuver at the moon Titan with a low to moderate lift-to-drag ratio (0.25 to could be used to capture a spacecraft into a 1.0).1 This approach provides for capture into closed orbit about Saturn if a nominal orbit about the giant, while avoiding the atmospheric profile at Titan is assumed. The very high entry speeds and aerothermal heating present study extends that work and examines environment inherent to a trajectory thru the the impact of atmospheric dispersions, variations of Saturn itself. in the final target orbit and low density aerodynamics on the aerocapture maneuver. Titan is unique among moons in the solar Accounting for atmospheric dispersions system in that it has an atmosphere considerably substantially reduces the entry corridor width for thicker than ’s, with a ground level density a blunt configuration with a lift-to-drag ratio of of about 5.44 kg/m3. Moreover, its atmosphere 0.25. Moreover, the choice of the outbound extends much higher above the surface than hyperbolic excess speed (with respect to Titan) Earth’s atmosphere, with the density at 800 km strongly influences the corridor width. Given above the surface being approximately the same the influence of these two parameters, certain as that on Earth at 132 km. mission scenarios may be feasible using a blunt aeroshell, while other mission designs would Titan has a near-circular, equatorial orbit about likely require a biconic vehicle with a higher lift- Saturn at a radius from the planetary center of to-drag ratio. Preliminary simulations indicate 1.22 (106) km and an orbital velocity of that the same technique may be feasible for approximately 5.57 km/s. This orbit is well capture into orbit about Neptune using the outside the ring system, which extends in the tenuous atmosphere of Triton. equatorial plane to a radius of approximately 480,000 km. A wide range of target orbits about Saturn can be achieved by means of a 1. INTRODUCTION Titan aerogravity assist maneuver. The final orbit will depend on both the orientation and the Aerocapture and aerogravity assist maneuvers magnitude of the outbound hyperbolic excess have long been recognized as methods by which speed with respect to Titan after the AGA otherwise impractical interplanetary missions maneuver. could be accomplished. Our group recently

49

Vinf = 1.5 km/s

Titan’s orbit r = 1.22(106) km Titan's orbit r=1.22(10 6) km

SATURN

Departure V inf antiparallel to Titan's velocity vector 5 Departure V parallel to Titan's velocity vector SemiSemi-major-major axis axis 8.33(108.33(10) km5) km inf 6 Semi-majorSemi-major axis axis3.13(10 3.13(10) km 6) km

Figure 1. Potential variation in Final Saturn Orbit for an Outbound Vinf of 1.5 km/s Figure 6. Potential Variation in Final Orbit for an Outbound V inf of 1.5 km/s

By lining up the outbound Vinf with Titan’s virtually in the same plane as Saturn’s rings, any orbital velocity vector, an orbit can be reached mission using this strategy will be complicated with a periapsis radius equal to that of Titan’s by the need not to fly through the debris field. orbit and a higher apoapsis. If the outbound If Titan’s orbital velocity vector and the

Vinf is in the opposite direction, the final orbit outbound Vinf are co-linear, the final spacecraft will have an apoapsis radius equal to that of trajectory about Saturn would lie very near or in Titan’s orbit and a lower periapsis radius (Fig. the ring plane. Directing an outbound Vinf of 3 1). An orbital periapsis of approximately km/s 15 degrees above the ring plane would 160,000 km would allow the probe to pass result in a final orbit about Saturn with its through the gap between rings F and G, a apoapsis coincident with its ascending node at maneuver which the Cassini spacecraft recently Titan’s orbital distance, its periapsis coincident accomplished without any apparent damage. with the descending node and in the Cassini Achieving this periapsis radius would require an division (between rings F and G), and an orbital

outbound Vinf of approximately 2.89 km/s, inclination of approximately 16 degrees (Fig. 2 opposite in direction to Titan’s orbital velocity and 3). vector. However, since Titan’s orbit lies Final orbital plane of probe Final probe Outbound Vinf = 3 km/s velocity about Saturn

Titan orbital velocity = 5.57 km/s Ring plane Figure 2. Velocity vector diagram for Figure 3. Edge view of possible final insertion into inclined orbit about Saturn orbital geometry

50 2. METHODOLOGY biconic, a base radius of 1 m was used, giving a reference area of 3.14 m2. The atmospheric entry corridor of an aerocapture maneuver is bounded by the overshoot and 3. RESULTS undershoot limits. These represent the shallowest and steepest angles at which the The corridor width for the capsule with an L/D spacecraft can enter the atmosphere and of 0.25 is shown in Fig. 5 as a function of entry successfully complete a capture into the desired velocity for both the nominal and extreme orbit. The width of the corridor depends on atmospheric models and for three target values of many factors, including the entry velocity, the the outbound Vinf. In general, the high density atmospheric density profile, the vehicle’s lift-to- atmosphere produced a shallower undershoot drag ratio and limits which may be imposed on bound than the nominal atmosphere did, and the the vehicle’s deceleration or aerodynamic low density atmosphere produced a steeper heating. For this study, trajectory simulations overshoot bound than was allowed with the were run using the three-dimensional version of 2 nominal atmospheric profile. Since the density POST. No constraints were placed on the profile which will be encountered is not known vehicle’s deceleration or aerothermal heating. prior to entry, it is necessary to consider the All simulations were begun at an altitude of 905 corridor width with these narrower limits. The km at zero degrees latitude and with a due east corridor defined in this manner must be wide azimuth. Entry angles are measured at the 905 enough to allow for off-nominal atmospheric km altitude. entry angles, knowledge errors and uncertainties in the expected aerodynamic performance of the Overshoot boundaries were found by directing vehicle. Current estimates indicate that insertion the vehicle’s lift vector toward Titan’s surface (a angle errors of +/- 0.9 degrees are to be vehicle roll angle of 180 degrees) and adjusting expected.5 Therefore, a corridor width of 2.0 the entry angle until the desired outbound V inf degrees is the minimum considered acceptable was achieved. Various target values of the for this study. Improvements in interplanetary outbound V were evaluated, corresponding to inf navigation capabilities that result in a more different final orbits about Saturn. The precise insertion angle would allow this corridor undershoot boundaries were found by flying a requirement to be relaxed. full lift up trajectory and adjusting the entry angle until the target outbound Vinf was achieved. Our original study considered only an outbound V of 1.5 km/s in determining the corridor Nominal, minimum and maximum atmospheric inf density profiles seen in Fig. 4 were based on the bounds (Ref. 1). From Fig. 5 it is apparent that 3 an increase in the targeted outbound V from work by Yelle. Atmospheric winds were not inf considered, nor were horizontal density 1.5 to 3.0 km/s substantially reduces the dispersions. aerodynamic corridor width. This finding follows the trend seen in conventional For Titan, four vehicle configurations were aerocapture studies, where corridor widths often evaluated. The first was a blunt aeroshell with decrease as the target apoapse (and orbital energy) an L/D of 0.25 that has been considered by other increase. This reflects the decreased control the investigators for use on a conventional Titan vehicle’s aerodynamics can exert over the 4 trajectory as the duration of the atmospheric pass aerocapture maneuver. An ellipsled with an and the required energy loss are reduced. L/D of 0.39, an Apollo-derived capsule flying at a high angle of attack (L/D of 0.482), and a Thus, if we assume that two degrees of corridor biconic with an L/D of 1.0 were also evaluated. width are required to allow for off-nominal entry (It is recognized that flank heating may prevent angles and aerodynamic dispersions, the choice an Apollo configuration from flying at such a of entry velocity and outbound Vinf will high alpha, but the configuration was chosen determine whether a given L/D provides a simply as a “place holder” to determine the sufficient corridor. If a target orbit about Saturn corridor available to a vehicle with an L/D in with a periapsis in the Cassini division is this range.) The vehicle mass was assumed to be desired, the outbound V will need to be near 3 600 kg, and a reference area of 12.56 m2 inf km/s, and a lift-to-drag higher than 0.25 will be (corresponding to a base radius of 2 meters) was required. used for all three blunt configurations. For the

51 of Triton is extremely tenuous, with a surface Fig. 6 shows corridor widths for all four pressure of approximately 16 microbars and a vehicles, assuming an outbound V of 3.0 km/s pressure at 48 km altitude of 2 microbars (Fig. inf 8 (this is considered the most likely target value of 8). Despite the very thin atmosphere, Vinf, since it provides a close approach to trajectory simulations show that either a blunt Saturn, opportunities for repeated Titan flybys aeroshell (L/D = 0.25) or a ballute could be used and a close view of the ring system). From this to accomplish a capture into orbit about Fig., it is apparent that the Ellipsled (with an Neptune, assuming the vehicle encounters L/D of 0.39) can provide a satisfactory corridor atmospheric conditions similar to those shown only at entry speeds of 9.5 km/s or more. The in Fig. 8. Figures 9 shows the corridor width for a 600 kg aeroshell (L/D of 0.25, reference modified Apollo configuration at an angle of 2 attack of 30 degrees (L/D of 0.482) provides a area of 12.56 m ), with the vehicle targeted to an sufficient corridor width at an entry speed of 8 atmospheric exit velocity of 4.8 km/s. As can km/s or higher. As the entry speed decreases, be seen, the corridor width for this vehicle is vehicle configurations with higher lift-to-drag unlikely to prove adequate for the maneuver in ratios (such as a biconic) become necessary light of the high degree of temporal variability in unless better targeting of the atmospheric entry Triton’s atmospheric density profile, In addition, angle can be achieved. The arrows in Fig 6 the vehicle often flies to altitudes less than 10 indicate the reduction in corridor width caused km above Triton’s surface during its atmospheric by atmospheric uncertainty, with the solid lines trajectory. Therefore, we also considered the use of a non-releasing ballute with a reference area of representing the nominal atmosphere and the 2 dashed lines showing the results for Yelle’s low 100 m . Without modulation of the release and high density profiles. time, variation in the atmospheric entry angle produces alterations in the outbound energy as While all the results presented up to this point shown in Fig 10. The ballute has a significant have assumed continuum aerodynamics, we advantage in that it flies at very high altitudes, conducted an evaluation of the impact of free thereby providing a greater “margin of error” molecular and transitional aerodynamics on with regard to surface impacts. vehicle trajectories and corridor bounds. For the vehicles considered here, there was no 4. CONCLUSIONS AND FUTURE WORK appreciable difference between results obtained neglecting and accounting for low-density The use of an aerogravity assist maneuver at influences on aerodynamic characteristics. We Titan to capture a vehicle into orbit about Saturn assume this results from the fact that the appears to offer the potential for a less severe vehicles’ deceleration almost exclusively occurs aerothermal enviroment than a direct capture in the continuum flow regime (Fig. 7) using Saturn’s atmosphere, while allowing for a variety of final orbital geometries. Depending on the choice of final Saturn orbit and the 4. POTENTIAL APPLICATION AT NEPTUNE atmospheric entry speed at Titan, blunt AND TRITON aeroshells or vehicles with mid-range lift-to-drag ratios (such as biconics) offer adequate entry NASA has conducted fairly detailed studies in corridor width, even when accounting for recent years of aerocapture at Neptune;6,7 an atmospheric dispersions. Low density aerodynamics seem to have minimal impact on important conclusion of this work has been that typical trajectories, but the choice of outbound the required aeroshell would have a much higher mass fraction than those considered for use at Vinf strongly influences entry corridors. , Earth or Titan, with 50% or more of the total probe mass being devoted to the aeroshell. Future studies of the topic must address optimal This situation greatly decreases the appeal of approach geometries, aerodynaic heating during aerocapture for Neptune missions. the Titan atmospheric trajectory, thermal protection system requirements and TPS mass Preliminary calculations indicate that the fractions. Guidance algorithms must be approach discussed in this paper for use at developed that can target both the magnitude and Titan/Saturn may also be feasible for capturing a direction of the outbound hyperbolic excess probe into orbit about Neptune using an speed in order to achieve the desired orbit about aerogravity assist maneuver at the moon Triton. Saturn. In addition, the use of ballutes for This maneuver would be substantially different capture at both the Titan/Saturn and from the Titan scenario, in that the atmosphere

52 Neptune/Triton systems should be further examined.

5. REFERENCES

1) Ramsey, P. and Lyne, J.E., “An Investigation of Titan Aerogravity Assist or Capture into Orbit About Saturn,” AAS Paper AAS-03-644, presented at the AAS/AIAA Astrodynamics Scecialists Conference, Big Sky, Montana, Aug. 2003.

2) Brauer, G.L. at al, “Program to Optimize Simulated Trajectories (POST),” NASA contract report NAS1-18147, Sept. 1989.

3) Yelle, R.V., Strobel, D.F., Lellouch, E. and Gautier, D., “Engineering Models for Titan’s Atmosphere.” In Huygens: Science, Payload and Mission,” European Space Agency, ESA SP 1177, pp.243-256, 1997.

4) Way, D.W., Powell, R.W., Edquist, K.T., Masciarelli, J.P. and Starr, B.R., “Aerocapture Simulation and Performance for the Titan Explorer Mission,” AIAA paper 2003-4951, presented at the 39th AIAA/ASME/SAE/ASEE Joint Propulsion Conference, July 2003.

5) Private communication, Jeff Hall, Jet Propulsion Laboratory, August 2004.

6) Lemmerman, L. and Wercinski, P., “Small Neptune Orbiter Using Aerocapture,” Space Technology and Applications International Forumn (STAIF-97), Albuquerque, NM, Jan. 1997.

7) Wercinski, Paul, Chen, Y.K., Loomis, M., Tauber, M., McDaniel, R., Wright, M., Papadopoulos, P., Allen, G., and Yang, L., “Neptune Aerocapture Vehicle Preliminary Design,” AIAA Atmospheric Flight Mechanics Conference, Monterey, CA, Aug. 2002.

8) Elliot, J.L. et al, “The Prediction and Observation of the 1987 July 18 Stellar by Triton: More Evidence for Distortion and Increasing Pressure in Triton’s Atmosphere,” Icarus, 148, pp. 347-369, 2000.

53 10 1

10 0

10 -1 -2 10 High density profile kg/m^^3 10 -3 Nominal profile 10 -4

10 -5 Density, -6 Low density profile 10

10 -7 -8 10

10 -9 Atmospheric -10 10

10 -11 0 200 400 600 800 1000 Altitude, km Figure 4. Titan atmospheric density

profiles from Yelle (Ref. 4)

5 Outbound Vinf, km/s 1.0

4 2.0

degrees 3.0 3 nominal atmosphere Width,

2 1.0 Corridor 2.0 1

Extreme 3.0

0 5 6 7 8 9 10 11 Entry Velocity, km/s Figure 5. Corridor width at Titan vs entry velocity for a probe with an L/D of 0.25

54

14

12 Nominal Atmospheres Extreme Atmospheres 10 dth (deg)

Wi 8 L/D = 1 6

Corridor 4 L/D = 0.48 L/D = 0.38 Entry 2 L/D = 0.25 0 6 6.5 7 7.5 8 8.5 9 9.5 10 Entry Velocity (km/s) Figure 6 Titan/Saturn AGA maneuver corridor width vs. entry angle for various vehicles

10 9

G's) 8 7 (Earth 6

Load 5 Biconic 4 Bluint 3 2

Accerleration 1 0 0.00001 0.0001 0.001 0.01 0.1 1 Knudsen Number Figure 7. Acceleration Load vs. Knudsen Number

55 10 -3

10 -4 kg/m^^3 10 -5 Density,

10 -6

10 -7 0 50 100 150 200

Altitude, km

Figure 8. Triton atmospheric density profile

-20 L/D = 0.25 Mass = 600 kg Reference area = 12.54 square meters -21

-22 Corridor

-23 Entry Angle (deg) -24

-25 7000 8000 9000 Entry Velocity (m/s)

Figure 9. Corridor Bounds for Triton/Neptune AGA Maneuver

56 6000 70

m/s Minimum altitude km 65 5000 velociy, altitude,

exit 60

4000 Amospheric exit velocity Minimum 55 Mass = 600 kg Reference area = 100 m^^2 Atmospheric Cd = 1.46

3000 50 15 16 17 18 Entry Angle, degrees

Figure 10. Aerocapture of a ballute at Triton as a function of entry speed

57