NASA RoboticConjunction Assessment Process:

Overview and Operational Experiences

Lauri Kraft NewmanConjunction Assessment Manager

NASA Goddard Space Flight CenterSpace Systems Protection Mission Support Office/Code 590.1

301-286-3155Lauri.K.Newman@nasa.gov

April 15, 2008

GSFC Conjunction Assessment Team:D. McKinley, R. Frigm

a.i. solutions, Inc.

AIAA Space Operations & Support Technical Committee 14th Annual Improving Space Operations WorkshopSpacecraft Collision Avoidance & Co-location Track

2

Agenda

• NASA Policy for Conjunction Assessment• NASA Robotic Conjunction Assessment

Process• Operations Process• Tool Suite Description• Statistics

• Terra vs. 14222 Case Study• Lessons Learned

3

NASA CA Policy

• Policy for Limiting Orbital Debris (NPR 8715.6) requires routine CA for all NASA assets with maneuvering capability (signed 8/17/07).– FDAB has the capability to offer this service to any mission

3.4 Conjunction Assessments during Mission Operations (for Earth-Orbiting Spacecraft)

3.4.1 The NASA Program/Project Manager shall have conjunction assessment analyses performed routinely for all maneuverable Earth-orbiting spacecraft with a perigee height of less than 2000 km in altitude or within 200 km of GEO (Requirement 56891).

3.4.2 Conjunction assessment analyses shall be performed using the USSTRATCOM high-accuracy catalog as a minimum (Requirement 56892).

3.4.3 The NASA Program/Project Manager shall have a collision risk assessment and risk mitigation process in place for all maneuverable Earth-orbiting spacecraft that are performing routine conjunction assessment analyses (Requirement 56893).

4

Existing NASA CA Process

• NASA Performs CA for its Human space assets (Shuttle and Station) as well as for various unmanned assets.– Requirements dictated by NPR 8715.6– NASA/JSC performs CA for Human Space Flight assets 3 times a

day– NASA/GSFC manages the CA for all unmanned NASA assets

• GSFC has performed CA routinely since January 2005 for:– 11 Earth Science Constellation members, including USGS-owned

Landsat missions– 9 TDRS satellites

• GSFC program is expanding to encompass all NASA unmanned missions per the new requirement. – Plan to add 12 more missions over the next year.

5

Missions Supported [1 of 2]Earth Science Constellation

• Combination of NASA and Foreign assets

• Each mission makes its own risk mitigation decisions

• Each mission subject to own maneuverability, comm, and ops concept constraints

6

Missions Supported [2 of 2]Tracking Data & Relay Satellite System

• Relay system composed of ground systems and nine spacecraft in geosynchronous orbit positioned at various longitudinal slots about the Earth.

TDRS-7150°

Stored

TDRS-8174.3°WStored

TDRS-9062°WStored

TDRS-10041°W

GRGT

WSC

TDRS-5171°W

TDRS-1049°W

TDRS-4046°W

TDRS-3275°W

TDRS-6173.7°W

7

FY09 CA Customer Base

Earth Science Constellation

GEO Regime Other NASA Missions

14780 - Landsat -5

25682 - Landsat-7

25994 - Terra

26619 - EO-1

26620 - SAC-C

27424 - Aqua

27642 - ICESat

28376 - Aura

28498 - PARASOL

29107 - CloudSat

29108 - CALIPSO

TBD – OCO (launch 12/15/08)

TBD – Glory (launch 3/1/09)

13969 - TDRS 1 

19548 - TDRS 3 

19883 - TDRS 4 

21639 - TDRS 5

22314 - TDRS 6

23613 - TDRS 7

26388 - TDRS 8

27389 - TDRS 9

27566 - TDRS 10

TBD - GOES-8 (launch 11/08 – L&EO only)

TBD - SDO (launch ~ 4/09)

26997 - JASON

22076 - TOPEX

25063 - TRMM

25789 - QuickSCAT

27391 - GRACE-1

27392 - GRACE-2

TBD - GLAST (launch5/16/08)

TBD - IBEX (launch 7/15/08)

TBD - OSTM (launch 6/15/08)

8

Current GSFC CA Effort

• Pertinent National Space Policy excerpt– National Security Space Guidelines: To achieve the

goals of this policy, the Secretary of Defense (SECDEF) shall: Have responsibility for space situational awareness; in this capacity, the SECDEF shall support the space situational awareness requirements of the Director of National Intelligence and conduct space situational awareness for: …. civil space capabilities and operations, particularly human space flight activities.

• GSFC has an agreement with the Joint Space Operations Center (JSpOC)/Space Superiority Cell to obtain conjunction assessment screening predictions for NASA robotic assets

9

CA Functional Overview

Curvilinear Collision ProbabilityCA ScreeningCA Screening Risk Assessment

Risk Assessment

Risk Mitigation

Risk Mitigation

JSpOC

• Generate Close Approach Predictions

• Consists of:• Performing Orbit Determination

for catalog objects• Computing separation distance

between objects in high accuracy space object catalog

• Summarizing and reporting results

• Analyze data to determine threat• Consists of:

• Trending miss distance and specific orbit determination related parameters

• Computing collision probability

• Performing collision probability sensitivity analysis

• CA analysts assist the Flight Operations Team

• Consists of:• Maneuver planning• Maneuver execution

GSFC CA Team Mission Owner/Operator

10

GSFC CA Operational Summary

• Goddard-dedicated Orbital Safety Analysts at the JSpOC generate CA data M-F (weekends as needed) using the high-accuracy (SP) catalog– Close approach predictions are made 7 days into the future

for LEO missions– Close approach predictions are made 10 days into the future

for GEO missions• Any planned maneuvers are modeled in the

ephemerides provided by the mission.• The CA data is provided to Goddard via secure FTP

and e-mail.• The GSFC CA team processes the data and provides

risk assessment analysis results to the mission stakeholders

11

LEO Safety Volumes

• Three different mission safety volumes define data product delivery from JSpOC/SCC and data processing by GSFC CA team

• The safety volumes are expressed in the primary UVW coordinate frame: U (radial), V (in-track) and W (cross-track)

• Monitor Volume (ellipsoid) +/- 2 x 25 x 25 km– Largest filter used to initially identify and report potential close

approaches

• Tasking Volume (box) +/- 0.5 x 5 x 5 km – Serves as a second warning and an elevated level of concern– Tasking level on the secondary object is increased (if

necessary)

• Watch Volume (standoff distance) 1 km

12

GEO Safety Volumes

• Monitor Volume (standoff distance) 40 km– Largest filter used to initially identify and report potential close

approaches

• Alert/Tasking Volume (standoff distance) 15 km– Serves as a second warning and an elevated level of concern– Tasking level on the secondary object is increased (if

necessary)

• Watch Volume (standoff distance) 2 km– Risk mitigation maneuver planning options examined

13

Data Products from JSpOC

• All Monitor Volume violations are summarized in a Conjunction Screening Summary and delivered to the CA SFTP server. The Screening Summary contains:– Time of Closest Approach (TCA)– Total Miss Distance– Miss Distance Position and Velocity Components in RIC frame– An additional email is sent documenting tracking data and tasking level

• An Orbital Conjunction Message (OCM) is provided for Tasking/Alert Volume violations. The OCM contains:– TCA– Asset State/Covariance at TCA– Object State/Covariance at TCA– Other orbit determination information helpful in performing collision risk

assessment.• Vector Covariance Messages (VCMs) for both objects are provided

for Watch Volume violations.– VCMs contain epoch state and covariance information– Used for maneuver planning

Risk Assessment Process

15

Risk Assessment by CA Team

• Screening data is analyzed in two distinct ways– Routine Operations

• Daily activity to assess most recent delivery of close approach predictions

• Disposition conjunctions as a “threat”, “not a threat”, or “monitor” event based on analysis

– High Interest Events• Events that have significant potential to be a threat or

provide a unique analysis opportunity• Trending of orbit determination parameters and conjunction

geometry• Probability of Collision Sensitivity analysis• Risk Mitigation Maneuver planning

16

Routine Operations

• At the end of the day, the OSA provides data products to the GSFC CA team– JSpOC OSA posts data products to the SFTP site.– CAS automatically parses the data and puts it into the database for

trending and use with other tools– The CAM Tool Suite is run each time new data is received

• Data is processed by the automated CAS utility and generates various reports– A summary report is generated containing all pertinent information and

delivered to the stake-holders– An ‘OCM Analysis Report’ is generated for each event and posted to a

secure website• The following morning the CA Analyst

– Verifies all data delivered to CA team and mission stakeholders– Reviews Screening Summary– Reviews all OCM Analysis Reports– Creates Watch List detailing all conjunctions in the Screening Summary

and the action to be taken– Performs additional analysis on conjunctions using the Collision Risk

Assessment Tool Suite

17

High Interest Event Risk Assessment

• Pc and miss distance data alone cannot be used to fully assess the threat

• Additional analyses to help establish and quantify risk include:– Orbit determination (OD) consistency from solution to solution

• Number of tracks and observations• Ballistic Coefficient• Solar Radiation Pressure Coefficient• Energy Dissipation Rate• Radar Cross Sectional Area

– Probabilistic Risk Assessment Analysis• Realistic probability calculations based on ‘realistic’ state and

covariance predictions• Pc evolution as the time to the close approach event gets shorter• Pc sensitivity analysis based on changes to inputs

– Conjunction Geometry (clock angle, approach angle)– Position of hard body radius with respect to the 3-sigma

covariance ellipse

18

Risk Mitigation

• If the threat evaluation indicates the need to plan and (possibly) execute a maneuver– CA Team notifies Mission Owner/Operator– CA Team analyzes maneuver options that will mitigate

the threat - first guess

• CA Team works with Mission Owner/Operator to plan risk mitigation options.– Maneuver must sufficiently increase the separation

distance and decrease the collision probability– Maneuver must meet orbit requirements if at all possible

• CA Team analyzes sensitivity of Pc to expected variations in burn performance

19

Typical LEO Risk Mitigation Maneuver Planning Process

• Maneuver planning begins ~ TCA-3 days– As TCA approaches, uncertainty decreases, but

avoidance options decrease

• Allows time to:– Improve the OD solution on the secondary object– Evaluate several maneuver options– Have 1st SPCS screen the options for post-maneuver

close approaches– plan the final maneuver– Upload commands to the spacecraft

20

The Collision Assessment System

• Collision Assessment System (CAS) was developed to store and analyze the large volumes of data received.

• CAS is automated and comprised of several elements:– Secure File Transfer Protocol Server– Parser / Monitor Scripts– Database– Collision Assessment and Mitigation (CAM) Tool Suite– Secure Website– Configuration Management System

21

Collision Assessment and Mitigation Tool Suite

• The CAM Tool Suite is the part of CAS that provides analysis utilities

• The CAM Tool Suite consists of 6 modules:1. Conjunction Visualization Script2. 2-D Collision Probability Utility3. Monte Carlo Simulation4. 3-D / Curvilinear Collision Probability Tool5. Time History Trending Utility6. Collision Avoidance Planning Tool

• The modules are built using FreeFlyerTM and MatlabTM

• Output from tools is formatted into a single PDF report for each OCM

22

Screening Data Processing: Conjunction Summary Report

• Overlap compare computes differences between subsequent solutions for the same close approach

• JSpOC and Owner/Operator solutions are compared• Results are posted to the Portal website

23

Screening Data Processing: CA Calendar

• A CA Calendar is produced and posted to the Portal– Contains close approach predictions of less than 1 km, events

having Pc > 1e-7, and planned maneuver dates/times.

24

Screening Data Processing: Watch List

•CA Analyst examines all data on Portal daily to produce a “watch list” of events warranting further analysis.

25

OCM Analysis: Conjunction Orientation

Conjunction Orientation: Shows the position and position covariance of thePrimary (blue) and Secondary (green) Objects in Earth Centered Inertial (ECI) coordinates.

26

OCM Analysis: Pc Sensitivity

27

OCM Analysis: Conjunction Plane

Earth Science ConstellationConjunction Statistics

29

CA Statistics - ESC

• For the Earth Science Constellation:– Each asset averages 15 unique conjunctions per week

within the Monitor Volume (780/yr)– Each asset averages 1 unique conjunction per week within

the Tasking Volume (52/yr)– Each asset averages 3 ‘high interest’ events per year - ops

team engages in maneuver planning process

• International Space Station (ISS) statistics for comparison (~340 km altitude, 51.6 deg inc):– For 2005, saw 24 Monitor Volume conjunctions, none with

Pc > 1 x 10-5

– Have seen 251 conjunctions from 7/99 – 12/05– Have executed 4 debris avoidance maneuvers– Average of 1.2 maneuvers per year predicted

30

ESC Safety Volume ViolationsPer Month

31

Average Number of Watch Volume Violations per Asset per Month

L5 L7 Terra EO-1 SAC-C Aqua ICESat Aura PARASOL CloudSat Calipso0

0.5

1

1.5

2

2.5

Nu

mb

er o

f U

niq

ue

Vio

latio

ns

ESC Watch Volume Violations - Monthly Averages

32

ESC Risk Mitigation Maneuvers Performed

Asset Secondary Maneuver Date

Terra 14222 (SCOUT G-1) 21-Oct, 2005

PARASOL 81257 (Analyst SAT) 16-Jan, 2007

SAC-C 14345 (SL-8 DEB) 16-Feb, 2007

Terra 31410 (FENGYUN 1C DEB)

22-Jun, 2007

CloudSat 28893 (SINAH 1) 04-Jul, 2007

Chinese ASAT Event

34

ASAT Event - Background

• On January 11th, 2007 China performed a successful test of an anti-satellite (ASAT) weapon

• The ASAT test consisted of a medium-range ballistic missile destroying a Chinese weather satellite

• Event occurred at an altitude of ~535 miles (861 km)

• First close approach with an ESC mission was predicted weeks after the event (04 Feb)

• Current number of cataloged objects ~2000

• The NASA Orbital Debris Program Office estimates >35,000 pieces larger than 1 cm

35

Tracked objects >10 cm diameter (FENGYUN 1-C Debris in red)

March 2007

Images courtesy NASA Orbital Debris Program Office

Debris Environment Growth

36

FENGYUN 1C DEB Mean Equatorial Height vs. Inclination -

0 500 1000 1500 2000 2500 30000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Mean Equatorial Height (km)

Cu

mu

lativ

e D

istr

ibu

tion

FENGYUN 1C DEB - Height Distribution

94 96 98 100 102 104 1060

500

1000

1500

2000

2500

3000Height vs. Inclination - FENGYUN 1C DEB

Inclination (deg)

Eq

uat

oria

l He

igh

t (km

)

Height Statistics (Oct 2007)

(Min, Max, Mean, Std)

195.7 2577.9 872.5 174.8

Inclination Statistics (Oct 2007)

(Min, Max, Mean, Std)

95.2 105.6 99.0 0.7

●

ESC orbit

37

ASAT Safety Volume Violations

38

ASAT - Percent of Total

39

ESC Monitor Volume Violations FENGYUN 1C Debris

L5 L7 Terra EO-1 SAC-C Aqua ICESat Aura PARASOL CloudSat Calipso0

10

20

30

40

50

60

70

80

Nu

mb

er o

f U

niq

ue

Vio

latio

ns

ESC MV Violations - ASAT

Case Study

Terra vs. 14222TCA October 23, 2005

41

Pc Trend

• On Monday Oct 17th, 1st SPCS predicted a close approach between Terra and object 14222 (SCOUT G-1 debris)

– TCA: Oct 23rd – Miss distance < 500 m– Pc ~1e-2

• 14222 characteristics at TCA:– Period: 98.66 min– Apogee Height: ~ 710.79 km– Perigee Height: ~ 679.20 km– Inclination: ~ 82.39˚

123456710

-3

10-2

10-1

Days to TCA

Co

llisi

on P

rob

abili

ty

• Throughout the week, the Pc remained high primarily due to the decreasing miss distance and the close approach geometry.

Day -6 Day -5 Day 0Day -3 Day -1

GSFC-Computed Pc ValuesFor Each OCM Solution

Days to TCA

Col

lisio

n P

roba

bilit

y

42

Avoidance Maneuver Planning

• Throughout the week, the miss distance trended downward from solution to solution

• On Thursday, October 20th the miss distance had dropped to <200 m, risk mitigation maneuver planning took place

• Four different maneuver options were generated and sent to JSpOC-Mountain for screening:

• Performing any of these maneuver options would increase the miss distance to a safe level and decrease the Pc by several orders of magnitude

Day -6 Day -5 Day 0Day -3 Day -1

Option # V (m/s) Burn Time Duration SMA (m) Separation @ TCA

1 0.0114 10/21 22:30 Z 2.0 sec +21.6 5.5 km

2 0.0157 10/21 22:30 Z 2.7 sec +29.6 7.5 km

3 0.0172 10/21 22:30 Z 3.0 sec +32.4 8.5 km

4 0.0173 10/22 14:07 Z 3.0 sec +32.6 5.5 km

43

Miss Distance Trend

• On Friday, October 21st the reported miss distance reached a local minimum and the collision probability reached a local maximum– Miss distance 50 – 60 m– Collision probability still on the

order of 1e-2

• Decision was made to perform maneuver option #2– Based on trends throughout the

week and the post-maneuver close approach screening results

– Executed at 22:30 Z– Actual miss distance = 4.6 km– Actual Pc ≈ 0

Day -6 Day -5 Day 0Day -3 Day -1Day -2

12345670

50

100

150

200

250

300

350

400

450

Days to TCA

Mis

s D

ista

nce

(m

eter

s)

44

Lessons Learned

• Automation is essential for managing the workload of routine data processing

• Personnel experienced in orbit determination are required to assess the threat using multiple criteria

• Each event appears to be sufficiently unique such that a standardized mitigation approach cannot be adapted

• Each spacecraft sees a handful of conjunctions per year for which avoidance maneuver planning is considered