Rosetta Nominal and Backup Operational Scenarios to Land on the Comet Wirtanen F. Dufour, J. Bernard

1
Rosetta Nominal and Backup Operational Scenarios
to Land on the Comet WirtanenF. Dufour, J.
Bernard, P. Gaudon, CNEST. Ceolin, S. Kerambrun,
CS-SI

• Space Ops 2002
• Houston, TX, USA
• 9 - 12 October 2002

2
Contents

• 1. Introduction
• 2. Rosetta Mission
• 3. Cometary Environment Modeling
• 4. Landing Phase
• 5. Descent Maneuvers
• 6. Numerical Results
• 7. Conclusion

3
1. Introduction

• ESA mission that aim at a rendezvous with the
  comet 46P/Wirtanen
• First ever landing of a probe on a comet to
  perform is-situ measurements of the cometary
  materials
• CNES is in charge of the landing maneuvers
• We must propose and validate a robust scenario to
  land on the comet in the best conditions of
  precision and terminal velocity
• The landing trajectory will be under the
  influence of complex forces
• gravitational forces greatly dependent on the
  comet nucleus shape,
• aerodynamic forces induced by the outgassing.
• Backup landing scenarios must also be prepared to
  cope with a failure of the nominal release
  mechanism.

4
2. Rosetta Mission

• Launch January 2003
• Rendezvous with Wirtanen November 2011
• Landing on Wirtanen October 2012

5
3. Cometary Environment Modeling 3.1. A Complex
Task

• Lack of information concerning the comets shape
  and environment
• At this stage, we can only count on astronomical
  observations done during previous perihelion
  passes
• The one-year long observation period should
  greatly improve our knowledge of Wirtanen
• Till then, we have to make many assumptions about
  Wirtanens physical characteristics

6
3. Cometary Environment Modeling 3.2. Comet
Nucleus Shape

• Forces acting in the vicinity of a comet are
  strongly linked with the shape of the nucleus
• At first, we assumed simple ellipsoidal shapes (a
  sphere or and ellipsoid)
• But a precise analysis requires more realistic
  nucleus shapes
• Muinonens statistical model allows very complex
  shapes

7
3. Cometary Environment Modeling 3.3. Comet
Attitude

• The attitude of the comet is defined by the
  orientation of the main inertia axes of the
  nucleus
• The poles motion is derived from the rotation of
  these axes
• At least 3 ways to define the attitude
• 1- polar axis fixed in the reference frame and
  nucleus motion restricted to a constant rotation
  around that axis
• 2- the 3 Euler angles that define the attitude
  evolve linearly with the time
• 3- the attitude is precisely interpolated from
  observation data
• Up to now, we assumed a fixed polar axis and a
  constant rotation

8
3. Cometary Environment Modeling 3.4.
Gravitational Forces (1/2)

• Gravitational forces are one of the 2 main acting
  forces in the vicinity of a comet
• 3 methods have been used to model the
  gravitational force field
• 1- Spherical Harmonic Expansion
• divergence problems with very elongated bodies
• 2- Ellipsoidal Harmonic Expansion
• no sign of divergence outside the smallest
  fitting ellipsoid
• but a divergence possible between ellipsoid and
  nucleus

9
3. Cometary Environment Modeling 3.4.
Gravitational Forces (2/2)

• 3- Polyhedric Potential
• polyhedron with multiple facets
• the volume integral can be replaced by a surface
  integral
• only valid for a constant density
• the iso-potential curves follow closely the shape
  of the nucleus
• no divergence problem
• Comet density 0.75 g/cm3
• Gravitational constant 45.28 m3/s2

10
3. Cometary Environment Modeling 3.5.
Aerodynamic Forces (1/2)

• Aerodynamic forces are not negligible because
  of the outgassing that increases close to the Sun
  (3 AU)
• An outgassing model was used to analyze these
  aerodynamic forces
• Sun in the equatorial plane
• lander characteristics
• sphere of 48 cm of radius
• mass of 100 kg

11
3. Cometary Environment Modeling 3.5.
Aerodynamic Forces (2/2)

• Aerodynamic forces in the same order of magnitude
  of gravitational forces in the vicinity of a
  small comet
• comet radius 600 m
• So the aerodynamic forces can dominate the
  gravitational forces
• where ratio gt 1

12
4. Landing Phase 4.1. Nominal Landing Scenario

• 1- The spacecraft moves on the delivery orbit
• 2- The lander is released with the help of a
  separation maneuver
• 3- At least 1 ADS maneuver is planned during the
  descent (autonomous mode)
• 4- Comet landing according to the terminal
  constraints
• no relative horizontal velocity
• a minimal vertical velocity

13
4. Landing Phase 4.2. Backup Landing Scenarios

• No human intervention are possible during the
  descent because the lander will behave
  autonomously
• The occurence of a failure of one of the ADS
  maneuvers must be taken into account earlier at
  the trajectory planning phaseI
• If a failure occurs during the release maneuver
  we can still intervene with the help of the
  mechanical backup release mechanism
• If the umbilical cord between the lander and the
  orbiter is disconnected, we must launch another
  landing attempt in less than 4 hours
• no delivery orbit modification allowed
• but we can change the attitude of the orbiter
• If the lander is still connected to the orbiter,
  we can wait up to 48 hours to analyze the
  situation before a new attempt

14
5. Descent Maneuvers 5.1. Computation Method

• CNES objectives are to validate a robust landing
  scenario despite the uncertainties about the
  comets characteristics
• Instead of trying to solve the problem globally,
  we opted for a progressive approach
• Many sensitivity analyses have been made to land
  on simple ellipsoidal shapes, assuming various
  comet sizes and densities, with and without
  outgassing
• More realistic comet shapes (Muinonen) have been
  tested afterwards
• The next step is to optimize the landing
  trajectories according to the terminal velocity
  constraints
• Finally, the optimal landing trajectories will be
  validated with the help of thorough Monte Carlo
  analyses

15
5. Descent Maneuvers 5.2. Trajectory Optimization

• We must solve a complex nonlinear control problem
  with constraints
• Performance index
• terminal velocity components
• descent duration
• Control variables
• delivery orbit
• separation maneuver
• ADS maneuvers
• Constraints included in the objective function
  (penalty functions)
• Direct optimization algorithms have been used
  (Nelder-Mead simplex)

16
5. Descent Maneuvers 5.3. Monte Carlo Analysis

• Because of the multiple uncertainties in the
  system, the optimal control maneuvers cannot be
  taken for granted without any further
  verification
• Every optimal landing trajectory will be
  validated with a Monte Carlo computation campaign
• Stochastic variables varied according to their
  probability distribution
• A statistical analysis of resulting landing
  trajectories will give the expectations of
  success of a specific scenario
• Finally, we will compare the landing scenarios
  and select the best one

17
6. Numerical Results 6.1. Landing on a Small
Spherical Comet
18
6. Numerical Results 6.2. Importance of
Aerodynamic Forces

• Aerodynamic forces can sometimes dominate
  gravitational forces (for a small comet)
• The resulting force on the lander will tend to
  throw it away from the comet
• Ignoring aerodynamic forces, an expected perfect
  landing can become a catastrophic miss
• We need precise outgassing data to evaluate that
  risk

19
7. Conclusion

• Aerodynamic forces should not be neglected in a
  comet landing
• they can throw the probe away from the comet
• but they can diminish the impact velocity
• Simple models are sufficient for preliminary
  mission analyses, but a more precise modeling is
  necessary to simulate a realistic trajectory
• The landing scenario have to be robust enough to
  ensure a safe landing despite the lack of
  knowledge about the force fields
• A failure can always occur and we have to be
  prepared to handle it
• ideally, the occurrence of a failure should be
  managed at the maneuver planning phase
• or a backup scenario must be implemented