Technology Development and Testing for

Enhanced Mars Rover Sample Return Operations

Richard Volpe
Mailstop 198-219
818-354-6328

Eric Baumgartner
Mailstop 82-105
818-354-4831

Paul Schenker
Mailstop 125-224
818-354-2681

Samad Hayati
Mailstop 180-603
818-354-8273

Jet Propulsion Laboratory

California Institute of Technology
Pasadena, California 91109
Email: firstname.lastname@jpl.nasa.gov
AbstractThis paper describes several Jet Propulsion Laboratory research efforts being conducted to support Mars sample return in the coming decade. After describing the 2003/05
mission scenario, we provide an overview of new technologies emerging from three complementary research efforts:
Long Range Science Rover, Sample Return Rover, and FIDO
Rover. The results show improvements in planning, navigation, estimation, sensing, and operations for small rovers operating in Mars-like environments.
TABLE
1.
2.
3.
4.
5.
6.
7.

OF

C ONTENTS

I NTRODUCTION
M ARS S AMPLE R ETURN M ISSION S CENARIO
L ONG R ANGE S CIENCE ROVER T ECHNOLOGIES
S AMPLE R ETURN ROVER T ECHNOLOGIES
FIDO ROVER T ECHNOLOGIES
S UMMARY
ACKNOWLEDGMENTS

1. I NTRODUCTION
Even before Sojourner made its first wheel tracks on Mars in
1997, it was anticipated that this rover would be only the first
in a series of surface exploration spacecraft targeted for the
planet. While it will be Sojourners flight spare that drives
on Mars in 2002, the next leap in technical capability exhibited by rovers will be in the 2003/05 mission set, where much
larger rovers will perform rock and soil sample collection for
return to Earth. These rovers will have greater innate capabilities, opening the door for the insertion of new robotics
technologies that have been in development since the inception of the Pathfinder mission five years ago. Among these
are on-board stereo vision processing, autonomous landerless operations, manipulation and instrument positioning by
arms, precision navigation for rover/lander rendezvous, and
distributed ground operations.
Of fundamental importance to the incorporation of the new
capabilities on next-generation rovers is the use of a more
capable electronics, sensing, and instrumentation infrastructure located on-board the rover. For instance, as Sojourner
was being prepared for flight, JPL was constructing a new
prototype, Rocky 7 [16]. Several key features were added to
support long trips away from the lander: a deployable mast to

0-7803-5846-5/00/$10.00 c 2000 IEEE

raise cameras and take panoramic images of the surrounding

terrain, a shorter arm for sample acquisition and instrument
placement, and a sun sensor to accurately determine heading
while driving. These features were demonstrated during field
tests in the Mojave Desert in 1997, which led directly to the
acceptance of the 03/05 missions [15].
The selected 03/05 mission concept, however, requires an enlarged rover that has the added functionality of carrying a drill
for rock sampling, larger wheels for enhanced mobility, and
a significantly upgraded science instrument suite (as opposed
the Sojourner rover). Therefore, to support continued field
tests with the selected science team for the 03/05 mission,
a new Field Integrated Design and Operations (FIDO) rover
was conceived, designed, integrated during a 12-month period, and demonstrated in desert tests in April 1999 [14]. The
FIDO rover reflects the current engineering sensors and science instrument suite that are planned for the 03/05 mission.
While this rover will continue to act as an operations testbed
for mission scientists, it has a second function as an integration testbed for new technologies that continue to be developed by ongoing research efforts.
The JPL core robotics technology program has been supporting these research efforts, which include the Rocky 7
rover as well as another platform, the Sample Return Rover
(SRR) [13]. Each rover has been dedicated to increasing
autonomy in two respective halves of the exploration problem: autonomous motion away from and back to the lander.
New techniques used include estimation and visual localization, on-board path and sequence re-planning, and natural and
man-made target recognition and tracking.
This paper describes these techniques, as well as the details
of the mission scenario in which they will be used. Section 2 describes the Mars Sample Returns Athena rover mission. The technology developments associated with the Long
Range Science rover task are discussed in Section 3, while
the Sample Return Rover and the development of rendezvous
techniques are described in Section 4. The Athena terrestrial
prototype rover, FIDO, is discussed in Section 5, and summary remarks are provided in Section 6.

During the mission, communication with the rover will nominally take place twice per day, relayed by the lander. Each
communication window will allow a limited set of images
and data to be transmitted to operators on earth, while new
instructions are provided to the rover based on the previous
communication cycle. Typically the rover will receive instructions for the day in the morning, and transmit the results
and status at the end of the day.
After obtaining its first set of rock samples, the rover will return to the lander and deposit them in the MAV. This must be
accomplished by successfully aligning with the base of the
ramp, driving up its two narrow rails, and accurately detecting the proper position for sample transfer. At this location,
the MAV payload door will be opened and the sample transfer
will be completed robustly and autonomously; thermal considerations require that the payload door be open less than
that of the typical communications cycle.

Figure 1. Preliminary drawing of the Mars Sample Return

Athena-Rover.

2. M ARS S AMPLE R ETURN M ISSION

S CENARIO
Launching in 2003 and again in 2005, NASAs Mars Sample
Return (MSR) spacecraft will place two science rovers on the
surface of the planet. These rovers will be carried to the surface on the top deck of a three-legged lander that is roughly
1 m tall and 3 m in diameter. This size, substantially larger
than previous landers of the 1996, 1998, and 2001 missions,
makes it possible to carry a sample-return Mars Ascent Vehicle (MAV) as part of the lander payload. The lander size
also enables the use of a rover that is approximately twice the
size of the Pathfinder mission rover, Sojourner, in every dimension. (Sojourner was 60 x 40 x 35 cm.) A preliminary
drawing of this Athena-class rover is shown in Figure 1.
The larger size of this MSR rover is needed to support rock
sampling operations and to carry the seven science instruments that make up its payload. Rock sampling is accomplished using a coring drill, which relies on the rover mass to
provide the force behind it in its vertical operating configuration. The science instruments are used to select the target
rocks for sampling, to determine the composition of the rock
samples obtained, and to study rocks in the surrounding area.
Four of the instruments, located on a five degrees-of-freedom
arm, require close proximity to the sample to be measured.
Two others, located on a 1 m mast, require only line of sight
to the target. The drill, itself, is the seventh instrument.
The rover will begin its mission on the lander top deck by
obtaining a panorama of the surrounding terrain for science,
engineering, and public outreach purposes. From these images, a ramp deployment direction will be selected, as well as
initial travel routes and goals for the first rover traverse. After ramp deployment, the rover will drive to the surface and
begin navigating the terrain. The maximum distance driven
each day will be 100 m, and often much less, especially when
the rover is positioning itself for science operations.

Since mission constraints limit surface operations to less than

90 days, the sample acquisition and return to MAV cycle can
be performed only three times at most. It is likely, however,
that each cycle will see the rover venturing farther from the
lander.
After the last sample return operation, the rover will move off
the lander deck and far enough away from the lander to prevent its being damaged during the MAV lift-off. This launch
is expected to damage the lander communication system and
prevent it from acting as a relay for the rover. Therefore,
the rover will use auxiliary communication to an orbiter, to
enable it to perform the extended mission of exploring the
surface.
Obviously, the complexity and accuracy of the autonomous
operations described above directly influence the amount of
science operations that will be performed. For this reason,
JPL research projects aim to introduce new functionality and
features into Mars rovers to enable greater science return
from all upcoming Mars rover missions. Techniques for more
autonomous and robust exploration and return to the lander
have been developed and implemented in field tests. Each of
these efforts is described next.

3. L ONG R ANGE S CIENCE ROVER

T ECHNOLOGIES
To improve rover navigation, exploration, and autonomy, the
Long Range Science Rover (LRSR) research task has been
improving the rovers ability to navigate through the environments, while maintaining an accurate sense of its position.
This section describes advances in four pertinent areas: dynamic sequence generation, autonomous path planning, visual localization, and state estimation. All research was conducted with the prototype rover, Rocky 7, shown in Figure 2.

Dynamic Sequence Planning

On-board planning with dynamic sequence generation allows
ground controllers to provide much higher level commands,
while increasing the optimality and robustness of rover opera-

Original planned path

Re-planned path

Figure 3. After encountering a previously unknown obstacle

shown in light orange, CASPER replans the sequence
of targets. Blue crosses are the goal locations, and dark
orange shapes are obstacles known a priori.
Figure 2. The Rocky 7 research prototype.
tions on the surface. For instance, during the Pathfinder Mission, the Sojourner rover [10] was provided with extremely
detailed sequences daily, fatiguing operators while also disallowing contingency operations when the flow of execution
was non-nominal. Contrary to this, we have been experimenting with on-board replanning that can change the execution
of daily activities based on unanticipated variations in quantities such as position, terrain, power, and time. To accomplish
this, we have used a dynamic on-board planing system called
CASPER (Continuous Activity Scheduling, Planning, Execution, and Replanning) [5], [6].
Figure 3 shows an example scenario in map form, where dark
orange shapes represent obstacles known a priori (e.g. from
lander descent imagery). In this case, the initial plan for the
traverse will bring the rover to an unexpected obstacle near
the first goal, represented as a light orange shaded shape. Circumnavigation around this obstacle will move the rover closer
to other goals, triggering CASPER to recognize the situation
and re-plan to visit the closest goal first. We are currently
evaluating this and similar scenarios experimentally.
Autonomous Path Planning
For the longer traverses required of upcoming missions, autonomous path planning is desirable since operators will not
be able to see three-dimensional terrain features out to the
more distant goal locations. Whereas Sojourner drove a total
of 84 m during its entire mission, the MSR Athena rover will
be capable of driving this distance in a single day. However,
stereo imagery of the terrain provided to operators will only
have an envelope of 20 m at best resolution. Therefore, the
path planning advances described here will allow the rover to
be its own operator. It can image the terrain from periscopic

cameras, select a path through the terrain to the edge of the

effective stereo range, and repeat the process until the goal
is achieved. A representative example of a partial panorama
and the resulting elevation map of the terrain are shown in
Figure 4.
Rocky 7 uses a local sensor-based path planner called RoverBug [8], [9]. This algorithm was developed for vehicles that
have limits on sensor range, field of view, and processing.
The two main modes of operation are motion-to-goal and
boundary-following, which are used to provide global convergence.
Rover-Bug works by using the local elevation map to construct a map of convex hulls around all obstacles in the sensing range. These hulls are then merged and grown to provide
a configuration space representation of the sensed terrain. A
tangent graph is constructed to determine if there is an unobstructed path to the envelope of the sensed region in the
direction of the goal. If one exists, it is followed, and the process is repeated at the end of the path segment to the sensory
envelope.
If a free path does not exist, stereo images are obtained to
the most promising side of the current view, and the process
is repeated. In some cases, the free path places the rover at
the edge of the sensory envelope but still obstructed from the
goal by an obstacle. In this case, the rover will begin to use
its body-mounted cameras to reactively boundary follow until
there is a clear path to the next goal.
Figure 5 shows experimental data obtained from Rocky 7
while using this algorithm to traverse our MarsYard test area.
The start position is in the lower left corner, 21 m from the
goal in the upper right. Four sets of sensory data are shown as

Figure 4. Steps of on-board terrain sensing: panoramic mosaic view from rover mast stereo imager, composite range map
extracted from stereo views, and elevation map created from range data.

Goal position

Localization target

Figure 5. Experimental results from a multi-step run using

Rover-Bug in the JPL MarsYard.
orange wedges along the path. Projected on to these data sets
are the pink convex hull representations of the sensed obstacles. The green and blue lines passing through the obstacle
field are the planned and executed paths, respectively. The
gap between data sets is due to a lack of merging of data from
both the mast and body-mounted cameras, and is currently
being corrected. The sudden changes in the path direction at
the far side of each wedge does not indicated actual rover motion; rather it is an artifact of the estimated rover position that
is updated by localization at this point in the traverse. This
localization is discussed next.
Visual Localization
Visual localization uses the same terrain imagery as path
planning, but for the purpose of monitoring the apparent motion of three-dimensional ground features after the rover has
completed a move. In this way, the on-board position estimate of the rover can be updated to compensate for errors
caused by wheel slippage or rock bumping. On Pathfinder,
this localization functionality was performed manually by operators viewing Sojourner from the fixed position lander cameras, restricting the update to once a day and permitting operations only within the stereo envelope of the lander. In contrast, the terrain-based localization described here has application to many forms of landerless operations: incremental
long traverses, local operations within range of a prior stereo
panorama, localization in descent imagery, and closed chain
rover moves with estimate smoothing.

Figure 6. Example of automatic target selection for localization.

The technique relies on obtaining a stereo elevation map from
the initial rover position, as shown in Figure 4. This imagery
is automatically analyzed based on the range data quality, and
the quality expected to be seen at the goal point specified by
path planning [11]. Typically, there will be a prominent rock
between the two locations. From the analysis, it will be selected as the localization target, and viewed by the rover after
reaching the end of the local path segment. Figure 6 shows
an example of the automatic target selection, with four images
concatenated to provide a reasonable number of potential targets.
To match the two stereo views of the terrain, a multiresolution search technique is used to provide the best estimate of the displacement of the original elevation map from
the final one [12]. Typical results provide position error estimates that are 1% of the distance traveled. However, the
search uses the on-board estimate of the rover position as its
starting point, so more reliable results are obtained if the initial estimate is more accurate. Therefore, to obtain better continuous position estimates, we have been developing a new
estimation technique, discussed next.

10

20

0
5
0

10

10

10

EST
SIM

10

EST
SIM

5
0
5
0

t (s)

10

4 (deg)
5

10

15

20

25

20

10

EST
SIM

10

t (s)

10

15

20

25

Figure 7. Simulation results showing true values and estimates for wheel contact positions, where the left/right
three wheels angles are shown by the left/right side
graphs.

10

15

20

25

10

15

20

25

10

15

20

25

10
0

10
10

1
0

20

10

10

6 (deg)

3 (deg)

10

10

2 (deg)

EST
SIM

10

5 (deg)

2 (deg)

10

0
5

3 (deg)

10

10

10

5 (deg)

EST
SIM

6 (deg)

10

1 (deg)

10

EST
SIM

4 (deg)

1 (deg)

20

10

10

15

t (s)

20

25

20

t (s)

Figure 8. Experimental results from Rocky 7 showing the

estimates for wheel contact angles.

Kinematic State Estimation

In addition to periodic visual localization of the rover, we
have developed real-time position and heading estimation using all other sensors on the vehicle: angular rate, accelerometer, sun sensor, wheel angle rates, and mobility system linkage (rocker-bogey) configuration. This technique moves far
beyond the simple dead reckoning of Sojourner and improves
upon our previous advances in position estimation with sun
sensing [15]. The results aid navigation during path execution, provide better input to localization, and replace both in
visually featureless terrain (e.g., sand dunes) or in the case of
visual sensing failure.
The estimator uses a Kalman filter framework, with the process model chosen so that the inertial sensor data are used as
an input to drive the process equation. The procedure avoids
the difficulty of modeling the detailed process dynamics [1]
by exploiting the ability of the Kalman filter to perform the
appropriate least-squares averaging of the action of each kinematic chain in the rover. These forward kinematic chains have
velocity components defined by the sequence of links joining
the rover frame to each estimated wheel contact point, and
a component given by the slip between the wheel and the
ground. The deterministic component of the slip is used to
capture the effects of a known steering action or a known average slip rate over different kinds of terrain.
Both simulation and experimental results with this estimation
technique have been conducted. Figures 7 shows the ability of
the estimator to correctly track the wheel contact angle over
an undulating terrain. The simulation is particularly valuable
since ground truth can be known exactly. For instance, using
simple integration of the wheel velocity results in a 2% error
in measured position, while the estimator reduces the error by
an order of magnitude.
Experimental results further support these conclusions. Figure 8 shows the results from Rocky 7 driving over an obstacle on the right side of the vehicle only. While the left side

Figure 9. The Sample Return Rover.

wheel contact points change only slightly, the much larger
changes in the right side are correctly determined. Note, in
the final configuration, the bogey wheels were still on the obstacle, while the steering wheel had completely traversed the
obstacle. The estimated wheel contact angles were within
5 degrees of the true values measured independently. Further
experiments are in progress, as well as integration of visual
position estimation, as described in the next section.

4. S AMPLE R ETURN ROVER T ECHNOLOGIES

In 1997, JPL began development of the Sample Return Rover
(SRR), a small, lightweight rover that investigated focused
technology advances in the areas of rover-to-rover and roverto-lander rendezvous. SRR, shown in Figure 9, is a 7 kg
rover with a 4-wheel rocker mobility system that is capable of traversing over obstacles up to 15 cm in height. The
rover includes 4-wheel steering and carries a three degreesof-freedom manipulator arm with a 1 degree-of-freedom gripper. This robot arm is used for panoramic imaging along
with sample pick-up and transfer. The rover also includes
a posable rocker joint for variable ground clearance and rover
reconfiguration in difficult terrain. SRR carries a PC104+
derived electronics system, including a 300 MHz AMD K6
processor, motion control I/O boards (D/A and encoder readers with closed-loop control realized in software), A/D board,
PCI color framegrabbers, and a wireless Ethernet.
The original concept for SRR derived from the previous version of the MSR mission where an Athena-class rover tra-

versed long distances, stopped at interesting science sites, and

acquired and stored rock and soil samples. During a second
mission, a small, lightweight rover would land on the Martian
surface and rendezvous with the science rover to retrieve the
sample collected during the primary science mission. Such
a rover-to-rover rendezvous was demonstrated by the SRR
technology team during the summer of 1998 [13]. This rendezvous was accomplished using an RF beacon transmitter
and receiver pair for non-line-of-site navigation to the science rover, visual tracking of the static science rover during
line-of-sight navigation, determination of rover-to-rover pose
using man-made features located on the science rover, and,
terminal guidance to the sample cache container and pickup
of this container using the on-board manipulator arm.


In 1999, the SRR task turned its attention to the rover-tolander rendezvous problem in support of the MSR mission
and the requirement that the Athena rover return its cached
samples to the lander and the awaiting MAV, as described
in Section 2. Some of the technology developments established during the rover-to-rover rendezvous application were
transferred to the rover-to-lander rendezvous problem. The
requirements of the Athena rover mission, however, necessitated the development of new techniques for the robust and
accurate navigation of SRR with respect to the lander such
that autonomous lander acquisition, lander rendezvous, and
ramp climbing are possible with minimal, if any, ground support. Figure 10 depicts the scenario associated with the roverto-lander rendezvous problem in terms of the multi-phase operations and technologies used during the return to the lander.
The rover-to-lander rendezvous problem is divided into the
following phases:


Long distance visual tracking using lander texture features

derived in the wavelet space
Multi-point tracking of lander features for heading and
range estimation of the rover relative to the lander
Ramp location determination using lander features
Ramp recognition using cooperative ramp features for
heading and range estimation of the rover relative to the
bottom of the ramps

The long-distance navigation of SRR relative to the lander is

accomplished with a novel wavelet-based detection algorithm
for the long-range visual acquisition of a lander from greater
than 100 m using a single, black-and-white rover imaging
system (20 degree field of view) [4]. This information enables autonomous correction the rover heading with respect
to the lander, and to guidance of SRR to within 25 m of the
lander. Such a navigation sequence is shown in Figure 11.
Within 25 meters of the lander, a visual technique that takes
advantage of the known geometry of the lander structure is
used to track multiple features (e.g., lander leg struts, lander
deck, etc.) to determine the pose of the rover relative to the
lander. Preliminary results indicate that precision on the order
of 50 cm in range and 1 to 2 degrees in orientation is possible
using this approach.
Likewise, the known ramp geometry allows for the visual acquisition of the lander ramps and the relative positioning of
the rover relative to the ramps. These techniques combine to
produce a navigation strategy for the autonomous guidance

of the rover from 25 m from the lander to 5 m in front of

the ramps. Finally, visual acquisition of the lander ramps is
accomplished using cooperative markings from which roverto-ramp position and heading information is obtained. The
successive visual acquisition of the lander ramps brings the
rover from 5 meters to within 5 cm of the bottom of the lander ramps. Many successful experiments have been accomplished in both laboratory and outdoor settings. These results
indicate that the rover can reliable and robustly navigate to the
bottom of the ramps with an absolute precision of 1 cm in lateral and longitudinal offset and less than 1 degree orientation
error with respect to the ramps. As such, the combination of
these rover navigation techniques leads to a single-command
autonomous sequence associated with the return to the lander
and the regress of the rover up the lander ramps to deposit the
sample cache in the MAV.
Finally, the SRR task has, over the past two years, developed
an alternative form of rover state estimation for the accurate
and reliable determination of rover position and orientation
relative to a fixed reference frame. This technique uses an extended Kalman filter framework based on the work described
in [3]. Within the SRR development, the registration of successive range maps generated by the rovers forward-looking
hazard avoidance cameras are utilized to determine the frameto-frame translation and rotation of the rover. This information is combined with dead reckoned estimates of the rovers
translation and rotation to produce an optimized determination of the rover pose. This work is described in [7] and [2]
and illustrated in Figure 12 for a 6+ m traverse within a softsoil, rock-filled indoor sand pit.

5. FIDO ROVER T ECHNOLOGIES

As described in Section 2, the 2003/05 MSRs Athena rover
mission represents a significant increase in complexity over
the recent Sojourner rover mission. The Athena rover carries
seven science payload elements, including a multi-spectral
imaging system, a microimager, a sample acquisition system,
and four different spectrometers as compared to Sojourners
single science instrument, the Alpha Proton X-ray spectrometer. In addition, the Athena rover has over eight times the
volume and six times the mass of the Sojourner rover. As
such, the Athena rover represents a true sciencecraft and fulfills the need for a robotic field geologist on Mars.
To facilitate the successful operation of the Athena rover during the 2003/05 Mars Sample Return missions, a terrestrial
prototype of the Athena rover is being used by the Athena science team for science and engineering testing associated with
flight mission operations. The development of this robotic
vehicle, known as FIDO (for Field Integrated, Design, and
Operations) rover began in early 1998 with the conceptual
design of the vehicle and culminated nine months later with
the full-scale integration of the rover and its science payload
[14]. In April 1999 (less than 14 months since the initial paper
designs), the FIDO rover was successfully operated at the Silver Lake field test site outside Baker in Californias Mojave
desert. Figure 13 shows FIDO operating during this desert
field test.
FIDOs mobility subsystem consists of a 6-wheel rocker-

Figure 10. The operations scenario associated with the return to the lander.

Figure 11. Wavelet-based lander detection and navigation.

Position Est. Error wrt. Ground Truth, K.F. vs. D.R.

3
Kalman Filtered
Dead Reckoned
2.5

Position Error (m)

1.5

0.5

0
0

10

20

30
40
50
Time (Navigation Iterations)

60

70

80

Figure 12. State estimation for rover pose determination.

Figure 14. The miniature core drill acquiring a core sample

from a carbonate rock at Silver Lake

Figure 13. The FIDO rover during the Silver Lake field trial.
bogie suspension system that has been scaled up by a factor
of 20/13 from the Sojourner design. This suspension system allows for the safe traverse over obstacles up to 30 cm in
height. Each wheel is independently driven and steered using a Sojourner-derived actuation and encoder system. The
top ground speed of the vehicle is 9 cm/sec. Approximate
rover dimensions are 1 m in length, 0.8 m in width, 0.5 m in
height, and 0.23 m ground clearance. The rover carries a four
degrees-of-freedom deployable mast that stands 1.94 m off
the ground surface at full extent. This mast provides the necessary pan-and-tilt control for panoramic imaging and point
spectroscopy. FIDO also carries a four degrees-of-freedom
instrument arm that is used to place the in situ suite of instruments on rock and soil targets.
The FIDO electronics are similar in nature to the SRR electronics with the CPU being a 80586 AMD processor running
at a 133 MHz clock speed. The rover uses a PC104-based
platform for all I/O functions, including a motion control
system (D/A and encoder readers with closed-loop control
in software) that can control up to 30 actuators simultaneously, two monochromatic and one color framegrabbers, digital I/O boards, A/D boards, low-pass filter and analog multiplexer boards, and a wireless Ethernet. Engineering sensors
include front and rear stereo hazard avoidance camera systems, an inertial navigation system, and a sun sensor for absolute heading determination. A differential GPS unit is also integrated within the rover electronics for ground-truthing purposes only.

The science payload on FIDO is analogous to the Athena payload. In particular, the remote sensing suite located on the
FIDO mast includes a multi-spectral, narrow field-of-view
Pancam stereo imaging system; a monochromatic, wider
field-of-view Navcam stereo imaging system; and the optics
for a near-IR point spectrometer that operates in the 1200 to
2500 nm wavelength region. The multi-spectral capability
associated with the Pancam system is realized using a Liquid Crystal Tunable Filter (LCTF) that is tuned to the three
near-IR wavelengths of 650, 750 and 850 nm. The in situ
instrument suite attached to the end-effector of the FIDO instrument arm consists of a color microimager and a Moessbauer spectrometer that are used to determine the iron content
of target rocks. A miniature core drill system, body-mounted
to the rover, provides the capability to acquire and cache rock
and soil samples. All of these instruments, with the exception
of the near-IR point spectrometer (the flight mission uses a
miniature thermal emission spectrometer in the mid-IR wavelength region), are breadboards of the Athena flight instruments.
In total, the science instrument and engineering sensing suites
and the resulting FIDO rover system represent a terrestrial
analog of the Athena rover that can be used to test and validate the Athena mission scenario and associated engineering
functions. As such, the Athena science team led by Professor
Steven Squyres, Athena PI, and Professor Raymond Arvidson, Athena Co-I, has worked with the FIDO engineering
team since March 1999 to perform rover operations testing in
support of the Athena rover mission. In particular, the desert
field trial at the Silver Lake test site represented the first ever
demonstration and validation of the sample acquisition phase
of the Athena rover mission through the identification of target rocks, approach to the target rock, placement of the miniature core drill over the target rock, successful acquisition of
a core sample using the miniature core drill, and return of
these samples to a simulated landing site. Figure 14 shows
the FIDO rover and associated core drill during the successful acquisition of a core sample from a carbonate rock. Future
field trials in 2000 and 2001 will focus on flight-like rover operations, with the rover and science teams being sequestered
at JPL while the rover is located at a remote test site. During
such blind field trials, the full Athena mission scenario will
be the further validated, including sample acquisition and return to the lander as well as long-range science exploration
and discovery.

6. S UMMARY
This paper has reviewed several recent research efforts that
support upcoming MSR mission scenarios. Three research
tasks, LRSR, SRR, and FIDO, have been developing and testing new capabilities in prototype rover platforms. LRSR research with Rocky 7 has developed four new techniques to enhance the functionality of autonomous long-range Mars rover
navigation: intelligent sequencing, sensor constrained path
planning, natural terrain visual localization, and Kalman Filter state estimation. SRR has developed a return-to-lander
capability with visual tracking at various ranges, using techniques suitable to those ranges. It has also displayed visual
odometry techniques. Finally, FIDO represents the culmination of the two core robotic technology programs since its development is based on the experiences gained within LRSR
and SRR. As a result, it has become a standard integration and
test system for the validation of rover navigation and control
strategies.

7. ACKNOWLEDGMENTS
Large systems like these require the support of many people
beyond the authors. We would like to recognize the following individuals for their invaluable contribution to the work
described in this paper: the LRSR team includes J. (Bob)
Balaram, Clark Olson, Sharon Laubach, Tara Estlin, Richard
Petras, Darren Mutz, Greg Rabideau, and Mark Maimone;
the SRR team includes Hrand Aghazarian, Terry Huntsberger,
Yang Cheng, Mike Garrett, and Lee Magnone; the FIDO team
includes Hrand Aghazarian, Terry Huntsberger, Anthony Lai,
Randy Lindemann, Paul Backes, Brett Kennedy, Lisa Reid,
Mike Garrett, and Lee Magnone.
The research described in this paper was carried out by the
Jet Propulsion Laboratory, California Institute of Technology,
under a contract with the National Aeronautics and Space Administration. Reference herein to any specific commercial
product, process, or service by trade name, trademark, manufacturer, or otherwise, does not constitute or imply its endorsement by the United States Government or the Jet Propulsion Laboratory, California Institute of Technology.

R EFERENCES
[1] J. Balaram. Kinematic State Estimation for a Mars
Rover. Robotica, Special Issue on Intelligent Autonomous Vehicles, Accepted for publication, 1999.
[2] E. T. Baumgartner, P. C. Leger, P. S. Schenker, and T. L.
Huntsberger. Sensor Fused Navigation and Manipulation
from a Planetary Rover. In Sensor Fusion and Decentralized Control in Robotic Systems, SPIE Proceedings
3523, Boston, November, 1998.
[3] E. T. Baumgartner and S. B. Skaar. An Autonomous
Vision-Based Mobile Robot. IEEE Transactions on Automatic Control, 39(3):493502, March 1994.
[4] J-K. Chang and T. L. Huntsberger. Dynamic Motion
Analysis using Wavelet Flow Surface Images. Pattern
Recognition Letters, 20(4):383393, 1999.
[5] S. Chien, R. Knight, R. Sherwood, and G. Rabideau. Integrated Planning and Execution for Autonomous Space-

craft. In IEEE Aerospace Conference, Aspen CO, March

1999.
[6] T. Estlin, G. Rabideau, D. Mutz, and S. Chien. Using
Continuous Planning Techniques to Coordinate Multiple
Rovers. In IJCAI Workshop on Scheduling and Planning,
Stockholm, Sweden, August 1999.
[7] B. D. Hoffman, E. T. Baumgartner, T. L. Huntsberger,
and P. S. Schenker. Improved Rover State Estimation in
Challenging Terrain. Autonomous Robots, 6(2), 1999.
[8] S. Laubach. Theory and Experiments in Autonomous
Sensor-Based Motion Planning with Applications for
Flight Planetary Microrovers. PhD thesis, California Institute of Technology, May 1999.
[9] S. Laubach and J. Burdick. An Autonomous SensorBased Path-Planner for Planetary Microrovers. In IEEE
International Conference on Robotics and Automation,
Detroit MI, 1999.
[10] J. Matijevic et al. Characterization of the Martian Surface Deposits by the Mars Pathfinder Rover, Sojourner.
Science, 278:17651768, December 5 1997.
[11] C. Olson. Subpixel Localization and Uncertainty Estimation Using Occupancy Grids. In IEEE International
Conference on Robotics and Automation, pages 1987
1992, Detroit, Michigan, May 1999.
[12] C. Olson and L. Matthies. Maximum-likelihood Rover
Localization by Matching Range Maps. In IEEE International Conference on Robotics and Automation, pages
272277, Leuven, Belgium, May 1998.
[13] P. S. Schenker et al. New Planetary Rovers for Long
Range Mars Science and Sample Return. In Intelligent
Robots and Computer Vision XVII, SPIE Proceedings
3522, Boston, November, 1998.
[14] P. S. Schenker et al. FIDO Rover and Long-Range
Autonomous Mars Science. In Intelligent Robots and
Computer Vision XVIII, SPIE Proceedings 3837, Boston,
September, 1999.
[15] R. Volpe. Navigation Results from Desert Field Tests of
the Rocky 7 Mars Rover Prototype. International Journal of Robotics Research, 18(7), 1999.
[16] R. Volpe et al. Rocky 7: A Next Generation Mars Rover
Prototype. Journal of Advanced Robotics, 11(4):341
358, 1997.

Richard Volpe, Ph.D., is the Principal Investigator for the

Long Range Science Rover Research Project, and the System
Technologist for the Athena-Rover sample collection robot
for the 2003 Mars Sample Return Project. His research interests include real-time sensor-based control, robot design,
software architectures, path planning, and computer vision.
Richard received his M.S. (1986) and Ph.D. (1990) in Applied Physics from Carnegie Mellon University, where he
was a US Air Force Laboratory Graduate Fellow. His thesis research concentrated on real-time force and impact control of robotic manipulators. Since December 1990, he has
been at the Jet Propulsion Laboratory, California Institute of
Technology, where he is a Senior Member of the Technical
Staff. Until 1994, he was a member of the Remote Surface Inspection Project, investigating sensor-based control technology for telerobotic inspection of the International Space Station. Starting in 1994, he led the development of Rocky 7, a
next generation mobile robot prototype for extended-traverse
sampling missions on Mars. In 1997, he received a NASA
Exceptional Achievement Award for this work, which has led
to the design concepts for the 2003 Mars rover mission.

Eric T. Baumgartner, Ph.D., is a group leader in the Mechanical and Robotics Technology Group and a senior member of engineering staff in the Science and Technology Development Section at NASAs Jet Propulsion Laboratory in
Pasadena, CA. At JPL, he serves in a systems engineering capacity for the development of advanced planetary rovers and
also contributes to technology developments in the areas of
robotic sensing and control. Prior to his tenure at JPL, he
was an Assistant Professor in the Mechanical Engineering Engineering Mechanics Department at Michigan Technological University in Houghton, MI. He has published over 30
articles in the area of robotic, controls, and state estimation
and is active in the SPIE and ASME. He received his B.S. degree in Aerospace Engineering from the University of Notre
Dame, the M.S. degree in Aerospace Engineering from the
University of Cincinnati in 1990, and the Ph.D. in Mechanical Engineering from the University of Notre Dame in 1993.

Paul S. Schenker, Ph.D., is Supervisor of the Mechanical and Robotics Technologies Group, Mechanical Systems
Engineering and Research Division, Jet Propulsion Laboratory. His current work emphasizes planetary rover development and robotic sampling technologies; he is task manager for NASA/JPLs Sample Return Rover (SRR) and Exploration Technology Rover Design, Integration, and Field
Test R&D efforts (ET Rover/FIDO a Field Integrated
Design & Operations rover prototype supporting the NASA
Mars 03-05 sample return missions, and related terrestrial mission simulations), as well as new NASA tasks on
use of cooperating robotic assets/rovers for Mars exploration
and future human habitation. Schenker also recently led
JPLs Planetary Dexterous Manipulators R&D under NASA
funding, work that proto- typed a robotic sampling concept which flies on the NASA Mars Polar Lander mission
now in route to the red planet. Schenkers other recent
robotics R&D activities include a role as founding co-PI for
NASA/MicroDexterity Systems Inc. development of a Robot
Assisted Microsurgery high-dexterity tele-operative workstation and a longer standing involvement in various tele-robotic
technology and system developments for orbital servicing
and autonomous robotic exploration. Schenker is a member
of AAAI, IEEE, and SPIE; he is a Fellow and 1999 President of the last. Schenker is widely active in external technical meetings, publications, and university collaborations in
the areas of robotics and machine perception, having contributed about 100 archival articles to same. Dr. Schenker
received his B.S. in Engineering Physics from Cornell University, and completed his M.S., Ph. D. and postdoctoral studies in Electrical Engineering at Purdue University. Prior to
joining JPL/Caltech in 1984, Schenker was with the Electrical Sciences faculty, Brown University, and later the Research
Section Chief for Signal & Image Processing, Honeywell Inc.

Samad Hayati received his M.S. and Ph.D. in Mechanical

Engineering with a specialty in controls from the University
of California at Berkeley in 1972, and 1976, respectively. He
joined the Jet Propulsion Laboratory in 1979 and worked in
the guidance and control of the Jupiter orbiter Galileo spacecraft, currently orbiting Jupiter. From 1983-1999 he performed research in robotics at JPL. He was also a lecturer of
robotics at the California Institute of Technology from 1986
to 1988. He has published numerous conference and journal
papers and holds two US patents related to robotics control.
His pioneering work in robot calibration was used to develop
techniques to utilize manipulators as an aid in brain neurosurgery at the Long Beach Memorial Hospital in California
in 1986. His most recent research efforts were in the development of long range science rover and rendezvous and sample
retrieval technologies for NASAs planned missions to Mars.
Currently, Samad is the manager of Robotics and Mars Exploration Technology Programs at JPL.

NASA Facts
National Aeronautics and
Space Administration
Jet Propulsion Laboratory
California Institute of Technology
Pasadena, CA 91109

Mars Exploration Rover

In April 2004, two mobile robots named Spirit
and Opportunity successfully completed their primary
three-month missions on opposite sides of Mars and
went into bonus overtime work. These twin vehicles
of NASAs Mars Exploration Rover project continued
their pursuit of
geological clues
about whether
parts of Mars formerly had environments wet
enough to be hospitable to life.

As Opportunitys primary mission ran out and an

extended mission began, the rover was headed for
thicker layers of exposed bedrock that might bear evidence about how long or how often water covered the
region.
Spirit, during
its primary mission, explored a
plain strewn with
volcanic rocks
and pocked with
impact craters. It
found indications
that small
amounts of water
may have gotten
into cracks in the
rocks and may
also have affected some of the
rocks surfaces.
This did not indicate a particularly
favorable past
environment for
life.

Opportunity
hit the jackpot
early. It landed
close to a thin
outcrop of rocks.
Within two
months, its versatile science instruments found evidence in those
rocks that a body
of salty water
deep enough to
splash in once
Spirits
flowed gently
extended mission
over the area.
Shadow of rover Opportunity in Endurance Crater, July 26, 2004.
began with the
Preliminary interrover starting a
pretations point to a past environment that could have
long trek toward a range of hills on the horizon whose
been hospitable to life and also could have preserved
rocks might have come from an earlier and wetter era
fossil evidence of it, though these rovers are not
of the regions past.
equipped to detect life or to be fossil hunters.

Second Extension as Adventure Continues

In late September 2004, NASA approved a second
extension of the rovers missions. The solar-powered
machines were still in good health, though beginning
to show signs of aging. They had come through the
worst days of the martian year from a solar-energy
standpoint. Also, they had resumed full operations
after about two weeks of not driving in midSeptember while communications were unreliable
because Mars was passing nearly behind the Sun.

yards) deep. The rover entered this crater in June after

careful analysis of its ability to climb back out.
Inside, Opportunity examined layer upon layer of
bedrock with characteristics similar to those of the
outcrop inside the smaller crater where it landed. This
indicated a much longer duration for the watery portion of the regions ancient past. The rover also found
some features unlike any it had seen before, evidence
of changes in the environment over time.
Whether the rovers unpredictable life spans
would extend only a few more days or several more
months, they had already racked up successes beyond
the high expectations set for them when the Mars
Exploration Rover project began.

Spirit had driven 3.6 kilometers (2.25 miles), six

times the goal set in advance as a criterion for a successful mission. It was climbing hills where its examinations of exposed bedrock found more extensive
alteration by water than what the rover had seen in
rocks on the younger plain. During the long trek,
Spirits right front wheel developed excessive friction.
Controllers found a way to press on with the exploration by sometimes driving the rover in reverse with
the balky wheel dragging.

Favorable Time to Build on Experience

Mars came closer to Earth in August 2003 than it
had in thousands of years. NASA decided in the summer of 2000 to take advantage of this favorable planetary geometry to send two rovers to Mars.
The design began with some basics from
Sojourner, the rover on NASAs 1997 Mars Pathfinder
mission. Some of the carried-over design elements are

Opportunity had driven about 1.6 kilometers (1

mile). It was studying rocks and soils inside a crater
about 130 meters (142 yards) wide and 22 meters (24

Artists simulation of a Mars Exploration Rover at work on Mars.

six wheels and a rocker-bogie suspension for driving

over rough terrain, a shell of airbags for cushioning
the landing, solar panels and rechargeable batteries
for power, and radioisotope heater units for protecting
batteries through extremely cold martian nights.
However, at 174 kilograms (384 pounds), each Mars
Exploration Rover is more than 17 times as heavy as
Pathfinder. It is also more than more than twice as
long (at 1.6 meters or 5.2 feet) and tall (1.5 meters or
4.9 feet). Pathfinders lander, not the Sojourner rover,
housed that missions main communications, camera
and computer functions. The Mars Exploration
Rovers carry equipment for those functions onboard.
Their landers enfolded them in flight and performed
crucial roles on arrival, but after Spirit and
Opportunity rolled off their unfolded landers onto
martian soil, the landers jobs was finished.
NASAs Jet Propulsion Laboratory, Pasadena,
Calif., designed and built the two new rovers plus the
lander and the cruise stage for each. The cruise stage
provided capabilities needed during the journey from
Earth to Mars. In early 2003, the hardware arrived at
NASAs Kennedy Space Station in Florida for final
assembly, testing and integration with Boeing Delta II
launch vehicles.
While the twin spacecraft were being built, scientists and engineers winnowed a list of 155 candidate
landing sites to a final pair best suited to the missions goals and safety. More than 100 Mars experts
participated in evaluating the sites. They made heavy
use of images and other data from NASAs Mars
Global Surveyor and Mars Odyssey orbiters.

size exposure of gray hematite, a mineral that usually

forms in the presence of liquid water.
Getting to Mars
Both rovers were launched from Cape Canaveral
Air Force Station on central Floridas Space Coast.
Spirit ascended in daylight on June 10, 2003.
Opportunity followed with a nighttime launch on July
7 after several days of delays for repairing cork insulation.
During the cruise to Mars, Spirit made four trajectory correction maneuvers. Opportunity performed
three. The two spacecraft survived blasts of highenergy particles from some of the most intense solar
flares on record. To prevent possible problems from
the flares effects on computer memory, mission controllers commanded rebooting of the rovers computers, a capability originally planned for use on Mars
but not during the cruise.
Each rover made the trip tightly tucked inside its
folded-up lander, which was encased in a protective
aeroshell and attached to a disc-shaped cruise stage
about 2.6 meters (8.5 feet) in diameter. The cruise
stage was jettisoned about 15 minutes before the
spacecraft reached the top of Mars atmosphere.

With the heat-shield portion of the aeroshell

pointed forward, the spacecraft slammed into the
atmosphere at about 5.4 kilometers per second
(12,000 miles per hour). Atmospheric friction in the
next four minutes cut that speed by 90 percent, then a
parachute fastened to the backshell portion of the
aeroshell opened about two minutes before landing.
The rover projects science goal has been to assess About 20 seconds later, the spacecraft jettisoned the
the history of environmental conditions at sites that
heat shield. The lander descended on a bridle that
may once have been wet and favorable to life. Each
unspooled from the backshell. A downward-pointing
of the two selected landing sites showed evidence
camera on the lander took three pictures during the
detectable from orbit that it may have once been wet. final half-minute of the flight. An onboard computer
For Spirit, NASA chose Gusev Crater, a Connecticut- instantly analyzed the pictures to estimate horizontal
size basin that appears to have once held a lake, judg- motion. In the final eight seconds before impact, gas
ing from the shapes of the landscape. A wide channel, generators inflated the landers airbags, retro rockets
now dry, runs downhill for hundreds of kilometers or on the backshell fired to halt descent speed, and transmiles to the crater and appears to have been carved by verse rockets fired (on Spirits lander) to reduce horiwater flowing into the crater. For Opportunity, NASA zontal speed. The bridle was cut to release the lander
chose part of a broad plain named Meridiani Planum
from the backshell and parachute. Then the airbagbased on a different type of evidence for a possibly
encased lander dropped in free fall.
watery past. A mineral-mapping instrument on Mars
Spirit landed on Jan. 4, Universal Time (at 8:35
Global Surveyor had identified there an Oklahomap.m. Jan. 3, Pacific Standard Time). It bounced
3

Spirits landing site on a plain inside Gusev Crater, viewed with the rovers panoramic camera before leaving the lander.

about 8.4 meters (27.6 feet) high. After 27 more

bounces and then rolling, it came to a stop about 250
to 300 meters (270 to 330 yards) from its first impact.
Spirit had journeyed 487 million kilometers (303 million miles). JPL navigators and engineers successfully
put it only about 10 kilometers (6 miles) from the
center of its target area. Coordinates of Spirits landing site are 14.57 degrees south latitude and 175.47
degrees east longitude.

site crater, later informally named Eagle Crater, is

about 22 meters (72 feet) in diameter, 3 meters (10
feet) deep. Its coordinates are 1.95 degrees south,
354.47 degrees east.
Science Instruments: A Geology Toolkit
Like a human field geologist, each Mars
Exploration Rover has the capabilities to scout its surroundings for interesting rocks and soils, to move to
those targets and to examine their composition and
structure.

Opportunity landed on Jan. 25, Universal Time (at

9:05 p.m. Jan. 24, Pacific Standard Time). It traveled
about 200 meters (220 yards) while bouncing 26
times and rolling after the impact, with a 90-degree
turn northward during that period. It came to rest
inside a small crater. One scientist called the landing
an interplanetary hole in one. Opportunity had
flown 456 million kilometers (283 million miles)
from Earth and landed only about 25 kilometers (16
miles) from the center of the target area. The landing-

Spirit and Opportunity have identical suites of

five scientific instruments: a panoramic camera provided by JPL; a miniature thermal emission spectrometer from Arizona State University, Tempe; a
Moessbauer spectrometer from the Johannes
Gutenberg University, Mainz, Germany; an alpha particle X-ray spectrometer from Max Planck Institute
for Chemistry, also in Mainz, Germany; and a micro-

Opportunitys landing site inside Eagle Crater, looking back at the empty lander after leaving the crater.

scopic imager from JPL. These are augmented by a

rock abrasion tool from Honeybee Robotics, New
York, N.Y., for removing the weathered surfaces of
rocks to expose fresh interiors for examination. The
payload also includes magnetic targets provided by
Niels Bohr Institute in Copenhagen, Denmark, to
catch samples of martian dust for examination. The
spectrometers, microscopic imager and abrasion tool
share a turret at the end of a robotic arm provided by
Alliance Spacesystems Inc., Pasadena, Calif.

Panoramic Camera Providing the
geologic context: This high-resolution stereo camera
reveals the surrounding terrain at each new location
that the rover reaches. Its two eyes sit 30 centimeters
(12 inches) apart, atop a mast about 1.5 meters (5
feet) above the ground. The instrument carries 14 different types of filters, allowing not only full-color
images but also spectral analysis of minerals and the
atmosphere. Its images are used to help select rock
and soil targets for more intensive study and to pick
new regions for the rover to explore.

Picture from Opportunitys microscopic imager showing an

iron-rich spherule embedded in layered rock. The area
covered in this image is 3 centimeters (1.2 inch) wide.

Miniature Thermal Emission Spectrometer

Identifying minerals at the site: This instrument
views the surrounding scene in infrared wavelengths,
determining types and amounts of many different
kinds of minerals. A particular goal is to search for
distinctive minerals that are formed by the action of
water. The spectrometer scans to build up an image.
Data from it and from the panoramic camera are used
in choosing science targets and new areas to explore.
Scientists also use it in studies of Mars atmosphere.

small amounts of curium-244 in measuring the concentrations of most major elements in rocks and soil.
Learning the elemental ingredients in rocks and soils
helps scientists understand the samples origins and
how they have been altered over time.

Microscopic Imager Looking at fine-scale

features: The fine-scale appearance of rocks and soils
can provide essential clues to how those rocks and
soils were formed. For instance, the size and angularity of grains in water-lain sediments can reveal how
they were transported and deposited. This imager provides the close-up data needed for such studies.

Moessbauer Spectrometer Identifying

iron-bearing minerals: Mounted on the rover arm,
this instrument is placed against rock and soil targets.
It identifies minerals that contain iron, which helps
scientists evaluate what role water played in the formation of the targets and discern the extent to which
rocks have been weathered. The instrument uses two
cobalt-57 sources, each about the size of a pencil
eraser, in calibrating its measurements. It is a miniaturized version of spectrometers used by geologists to
study rocks and soils on Earth.

Supplemental Instruments Engineering

tools aid science: Each rover also has other tools
that, while primarily designed for engineering use in
the operation of the rover, can also provide geological
information. The navigation camera is a wider-angle
stereo instrument on the same mast as the panoramic
camera. Hazard-avoidance cameras ride low on the
front and rear of the rover in stereo pairs to produce
three-dimensional information about the nearby terrain. The front pair provides information to aid positioning of the tools mounted on the rovers arm.
Rover wheels, in addition to allowing mobility, are
used to dig shallow trenches to evaluate soil properties.

Alpha Particle X-Ray Spectrometer

Determining the composition of rocks: An
improved version of an instrument used by the
Sojourner rover, this spectrometer is also similar to
instruments used in geology labs on Earth. It uses
5

order to avoid confusion when talking about plans

and results related to those features. The named features range from stadium-size craters to coin-size
spectrometer targets on rocks.
Persistent Spirit
Spirits first photos looking around its landing site
revealed a rock-strewn plain. A few shallow, dusty
hollows lay nearby and a few hills and crater rims
interrupted the flat horizon. Even before the rover had
rolled off its lander platform, scientists chose
Bonneville Crater, about 300 meters (328 yards) to
the northeast, as a destination that might offer access
to underlying rock layers. They eyed the Columbia
Hills, about 2.6 kilometers (1.6 miles) to the southeast, as a tempting but probably unreachable goal for
later.
An airbag that was not fully retracted under the
lander presented an obstacle to simply driving Spirit
forward off the lander. Engineers had practiced many
scenarios for getting the rover off. They chose to tell
Spirit to turn in place about 120 degrees before driving down a side ramp. The rover rolled onto martian
soil on Jan. 15. The next day, it extended its robotic
arm to a patch of soil and took humankinds first
microscopic image of the surface of another planet.
Scientists chose a nearby, football-size stone dubbed
Adirondack as the first rock to get full research
treatment with all four tools on the arm. Spirit
reached out to the rock on Jan. 20, but the examination was interrupted by a computer and communication crisis.

Spirits robotic arm reaching to a rock informally named

Adirondack for examining the rock with tools on the arm.

Names of Rovers and Features

The names of the rovers, Spirit and Opportunity,
were selected in a student essay contest that drew
nearly 10,000 entries.
After the spacecraft reached Mars, NASA dedicated the landers as memorials to astronauts who perished in space shuttle accidents. Spirits lander
became Columbia Memorial Station. Opportunitys
became Challenger Memorial Station.
A committee of the International Astronomical
Union designates official place names on Mars, such
as the names Gusev Crater and Meridiani Planum.
NASA and members of the rover science team have
put unofficial names on many natural features seen by
the rovers.

Spirit stopped communicating on Jan. 21. For two

A range of hills that Spirit saw on the eastern

horizon from the rovers landing site is unofficially
called the Columbia Hills, with seven individual
hilltops named for members of the Space Shuttle
Columbias last crew: Anderson, Brown, Chawla,
Clark, Husband, McCool and Ramon. Spirit drove
more than three kilometers (about two miles) to reach
those hills and begin climbing them.
Map of Spirits travels through Sept. 3, 2004, from landing
site at left northeastward to rim of Bonneville Crater then
southeastward into the Columbia Hills. Scale bar is 500
meters (1,640 feet).

As in earlier Mars surface missions Viking and

Pathfinder scientists assign informal names to
smaller features, such as rocks and patches of soil in
6

Spirits view after climbing into Columbia Hills, part of a full-circle panorama taken between Aug. 9 and Aug. 19, 2004.

worry-filled days, it sometimes failed to send signals

at all and other times sent signals without meaningful
data. Engineers began to coax some helpful data from
Spirit on Jan. 23. They learned the onboard computer
had rebooted itself more than 60 times in three days.
They developed a strategy to stabilize the rover while
continuing to diagnose and remedy the problem over
the next several days. The diagnosis was a flight-software glitch that obstructs proper management of the
onboard computers flash memory when the memory
is too full. The main remedy was to clear Spirits
flash memory and, from that point on, to avoid getting the memory too full on either rover.

portion, during which the rover watched ahead for

hazards and chose its own path to avoid them.
Spirit reached the base of the hills on June 11.
There, it examined an oddly knobby rock dubbed
Pot of Gold and other eroded features before
ascending a ridge called West Spur. Climbing that
ridge in early August, Spirit reached exposed bedrock
for the first time, seven months after landing.
Well-Placed Opportunity
Opportunity drove up to exposed, layered bedrock
in Eagle Crater on Feb. 7, just two weeks after
landing. It spent most of the next six weeks examining this outcrop, which arcs about halfway around the
inner slope of the crater but stands only about as high
as a street curb.

Spirit finished with Adirondack, where the rock

abrasion tool provided the first-ever look inside a
rock on Mars. Then the rover set out toward
Bonneville Crater, examining Humphrey and
other rocks on the way. It reached the crater rim on
March 11 and looked inside. No bedrock layers were
visible to tempt the team into sending Spirit down
into the bowl. On March 31, the rover completed an
eight-day inspection of a wind-scalloped boulder
called Mazatzal just outside the crater. That rock,
like all others examined on the plain Spirit was crossing, came from a volcanic eruption. However, thin
coatings on the rock and veins inside it suggest that it
might have been affected by water at some point.

The rover discovered BB-size gray spheres

embedded in the rock like blueberries in a muffin.
These spherules are also plentiful in the soil of the
area, apparently set loose when erosion wore away
softer rock material around them. They contain
hematite, the mineral whose detection from orbit had
made Meridiani a compelling landing site.
Spectrometers on the rovers found that the outcrop is rich in sulfate-salt minerals, evidence that the

The rover spent 10 weeks driving from near

Bonneville to the edge of Columbia Hills while
surveying soils, rocks and smaller craters along the
route. Its longest single-day advance was 123.7
meters (135 yards) on May 10, about 20 percent farther than Sojourner drove during its entire three
months of operations on Mars in 1997. As became
typical for long-drive days, the feat combined a blinddrive portion, in which Spirit followed a route that
rover planners at JPL had determined in advance
using stereo images, and an autonomous navigation

Map of Opportunitys travels through Aug. 21, 2004, from

landing site on left eastward to Endurance Crater, then
into that crater. Scale bar is 100 meters (328 feet).

Portion of the outcrop in Eagle Crater, where Opportunity landed. The rocks are about 10 centimeters (4 inches) tall.

Opportunitys view northeastward into Endurance Crater, combining frames taken with the panoramic camera between
May 23 and May 29, 2004. The crater is about 130 meters (about 425 feet) in diameter.

rock had been drenched with salty water. The

spherules are distributed throughout the rocks, rather
than only in particular layers. This observation contributed to a conclusion that they are concretions,
another sign of mineral-rich water soaking through
the rocks. The microscopic imager revealed rippled
bedding patterns in some of the finely layered rocks,
indicating that the rocks not only were exposed to
water after they formed, but actually formed from
sediment particles laid down in flowing water.

Project/Program Management
The Mars Exploration Rover program is managed
for NASA by JPL, a division of the California
Institute of Technology, Pasadena, Calif.
At NASA Headquarters, David Lavery is the program executive and Dr. Curt Niebur is the program
scientist. Dr. Catherine Weitz was the program scientist through August 2004. At JPL, Peter Theisinger
was project manager until February 2004, followed
by Richard Cook and, currently, Jim Erickson. JPLs
Dr. Joy Crisp is the project scientist. The principal
investigator for the science payload is Dr. Steve
Squyres from Cornell University, Ithaca, N.Y. Deputy
principal investigator is Dr. Ray Arvidson from
Washington University, St. Louis.

Opportunity climbed out of Eagle Crater on

March 22. It examined some rocks and soil on the
dark surrounding plain, then headed east toward a stadium-size crater called Endurance. It set a one-day
martian driving record of 140.9 meters (462 feet) on
April 17 and reached the rim of the crater on April
30.

On the Internet
Additional information and images are available
on Web sites for the Mars Exploration Rover Mission
at http://marsrovers.jpl.nasa.gov and for the suite of
science instruments at http://athena.cornell.edu .

The rovers panoramic camera and miniature thermal emission spectrometer surveyed the interior of
Endurance from two overlook points about a third
of the way around the rim from each other. That
information helped the rover team plot the safest
route to the most interesting targets accessible. The
rover drove into Endurance Crater on June 8. It
found that as far down as outcrops extended, they
bore evidence of extensive exposure to water.

10-04