Conference PaperPDF Available

Concept for coring from a low-mass rover

Authors:

Abstract and Figures

Future Mars missions such as the Mars Sample Return (MSR) mission may benefit from core sample acquisition from a low-mass rover where the rover cannot be assumed to be stationary during a coring operation. Manipulation from Mars rovers is currently done under the assumption that the rover acts as a stationary, stable platform for the arm. An MSR mission scenario with a low-mass rover has been developed and the technology needs have been investigated. Models for alternative types of coring tools and tool-environment interaction have been developed and input along with wheel-soil interaction models into the Stanford Simulation & Active Interfaces (SAI) simulation environment to enable simulation of coring operations from a rover. Coring tests using commercial coring tools indicate that the quality of the core is a critical criterion in the system design. Current results of the models, simulation, and coring tests are provided
Content may be subject to copyright.
Concept for Coring from a Low-mass Rover
Paul Backes*, Oussama Khatib§, Antonio Diaz-Calderon*, James Warren§, Curtis Collins*, Zensheu Chang*
*Jet Propulsion Laboratory
California Institute of Technology
Paul.G.Backes@jpl.nasa.gov
§Department of Computer Science
Stanford University
ok@robotics.stanford.edu
Abstract— Future Mars missions, such as the Mars Sample
Return (MSR) mission, may benefit from core sample
acquisition from a low-mass rover where the rover cannot
be assumed to be stationary during a coring
operation. Manipulation from Mars rovers is currently done
under the assumption that the rover acts as a stationary,
stable platform for the arm. An MSR mission scenario with
a low-mass rover has been developed and the technology
needs have been investigated. Models for alternative types
of coring tools and tool-environment interaction have been
developed and input along with wheel-soil interaction
models into the Stanford Simulation & Active Interfaces
(SAI) simulation environment to enable simulation of
coring operations from a rover. Coring tests using
commercial coring tools indicate that the quality of the core
is a critical criterion in the system design. Current results of
the models, simulation, and coring tests are provided.
1
1
TABLE OF CONTENTS
1. INTRODUCTION......................................................1
2. ROVER-TOOL CONCEPT .......................................1
3. CORING TOOL MODELS........................................2
4. SIMULATION ENVIRONMENT ................................5
5. WHEEL-SOIL INTERACTION .................................7
6. CORING TESTBED..................................................8
7. RESULTS ................................................................8
8. CONCLUSIONS .......................................................9
ACKNOWLEDGEMENTS .............................................9
REFERENCES .............................................................9
BIOGRAPHY.............................................................10
1. INTRODUCTION
Core sample acquisition from a planetary rover currently
requires that the rover be a stationary platform for a
manipulation system on which a coring tool is mounted.
Future Mars rover missions, such as a Mars Sample Return
(MSR) mission, may need to minimize mission cost by
reducing the rover mass. With a low-mass rover, the
interaction forces between the tool and terrain may cause
the rover to slip during a coring operation. A research
activity is underway to investigate how to enable coring
from a low-mass rover. A rover-tool concept has been
developed to act as the context for the work. Alternative
core sampling tool concepts are being investigated and
0-7803-9546-8/06/$20.00© 2006 IEEE
modeled to determine what type of tool would be preferable
for the low-mass rover system. A simulation system has
been developed for simulating coring operations. A coring
testbed has been developed to test core sample acquisition.
2. ROVER-TOOL CONCEPT
Figure 1. Rover-tool concept
The baseline operations scenario is to be able to acquire
cores with the rover on slopes up to 30° with the tool at
various angles relative to the rover. A vertical orientation
of the tool will cause the lowest slip force while a horizontal
orientation provides the highest slip force opposing the
wheel traction.
A baseline rover-tool concept was developed to enable core
sample acquisition for an MSR-type mission. The rover-
tool system is shown in Figure 0. A first step was to decide
how many degrees of freedom (DOFs) to have in the
manipulation system that holds the coring tool. The earlier
Athena rover concept for sample acquisition used a two
DOF pitch-translate mechanism [1]. This activity chose a
five DOF arm because it allows arm motion to
accommodate rover slip in any direction. A two DOF arm
requires that the rover act as a stationary platform or
provide a mobility DOF during compensation for slip. But
it is expected that the resolution of motion available from
2
rover mobility will be too coarse to compensate for rover
slip during coring.
Use of a three DOF arm, implemented by freezing two
DOFs of the five DOF arm, will be investigated. If it is
assumed that slippage will occur along the gravity vector
projected onto a ground plane, then accommodation in a
plane is all that will be required. Means for
accommodating slippage in other directions will need to be
included also.
Another consideration for the arm DOF is transfer of a
sample to the sample container. A five DOF arm reduces
the constraints on mounting of the sample container.
Sample contamination is a significant concern for a sample
return mission. One approach for minimizing sample
contamination is to acquire the sample directly into a
storage sleeve. A sampling and containerization approach
using a sample sleeve was developed, as shown in Figure 0.
Bits with internal sleeves are stored in a bit station. One bit
is used per core sample. Coring holes are started by
percussion-only mode with the coring tool and then rotary
percussion is used to acquire the core. The core is broken
by rotating off-axis concentric tubes or by rotating cutting
fingers at the end of the sleeve. Fingers at the end of the
sleeve contain the sample in the sleeve and the sleeve is
extracted from the bit. A tube cap is removed from the
sample container using another tool on the arm turret and
then the sleeve is inserted in the tube and seated in the
shield groove.
3. CORING TOOL MODELS
Tube
caps
Shield
Groove
Inserted
Sleeves
with
Samples
Figure 2. Sample container
The type of coring tool used to acquire the sample is very
important. Different types of coring tools require different
control relative to their environment and produce different
quality cores. There are three basic types of coring tools for
Mars sampling missions: rotary friction, ultrasonic, and
rotary hammer.
Rotary-Friction Coring Tool Model
Rotary friction coring tools impart normal and tangential
forces into the material. Large normal forces cause the bit
teeth to catch and compress the rock surface and the rotary
action causes tension or shear stress buildup that is relieved
by the formation of tension or shear fractures along the
direction of tooth motion [2]. There are two primary
drawbacks of rotary friction coring tools, the relatively large
normal force, or preload, required between the bit and rock
and the need for a centering bit to start a coring hole. For
coring from a low-mass rover, the large normal force
required affects the rover mass by requiring the tool
deployment device to apply this force against the
environment. An example of this type of coring tool is the
Mini-corer from Honeybee Robotics [3].
Figure 3. The planar model of the rotary-friction
drill system. 1: Drive with constant velocity v; 2: tip
of the bit with mass m1; 3: main spring with stiffness
k1; 4: dashpot with viscous coefficient c2; 5: pin,
attached with a lever with mass m2; 6: secondary
spring with stiffness k2.
A model for rotary friction coring tools was proposed by
Batako [5] to simulate the motion of a rotary corer with and
without percussive actions. The model focuses on the tip of
the corer bit. The movement of the tip of the corer bit is
modeled as a stick-slip motion. The tip of the bit is
connected to the shaft of the motor through the corer stem
which is represented by a spring-damper system, as shown
in the planar model in Figure 0. Note that the velocity v is
the linear speed of the bit rotation; the lever connected to
the drive at pin 5 is free to rotate; and there is dry friction
between the bit and the surface of rock. The shaft of the
motor is assumed to be rotating at a constant velocity. A
preload is applied to the corer system, and the corer bit is
pressed against the rock. When the motor just starts
working, the tip of the bit sticks to the rock due to friction;
and then when the maximum static friction is reached, the
tip of the bit slips until it sticks again. The stick-slip cycle
repeats it self and such motion is known as “stick-slip
motion”.
The governing equations of the motion of the tip of the
corer bit are:
()
()
122
.
22
..
22
.
21211
..
11
xxkxcxm
xFxxkxkxm r
=
= (1)
3
where is the friction force,
.
xFr
()
[]
()()
=++
+
=
0forsgn,min
0forsgn
3
1
.
2121121211
..
3
3
.
.
.
xxxkxkfxxkxk
xx
v
x
v
x
f
xF
st
crcr
st
r
(2)
where fst is the maximum static friction, and x = x1 + vt; x1
and x
2 are the coordinates of the bit and the lever,
respectively, relative to drive 1.
A computer program was developed in Matlab/Simulink to
solve the governing equations. To verify the validity of the
computer program, the example problem shown in the paper
[5] was solved by the program. It is found that the
computer program we developed successfully repeated the
results shown in the paper. We then modified the variables,
m1, m2, k1, k2, c2, and v, in the program to simulate the
rotary-friction corer under investigation, the Mini-Corer
developed by Honeybee. Some of the parameters in the
model are not available, for example, the spring constants,
equivalent mass of the bit-head (not the whole bit), etc. So
we chose some values for these parameters so the results are
compatible with the data we received from Honeybee
including the rotational speed (about 200 RPM), the preload
(155 N), and the average torque (2 N-m), etc. The preload
(155 N) is also felt by the robotic arm, and it is taken as a
constant in the model.
The model depicted in Figure 0 gives us the friction force
between the drill bit and the surface of the rock, as well as
the torque that exists in the drill stem. However, what we
need is the torque felt by the robotic arm that holds the drill.
So we modified the model to predict also the rotational
speed of the rotor of the motor which drives the drill stem,
and through the torque-rotational speed relationship we are
able to derive the torque that was delivered from the stator
to the rotor. The torque delivered across the stator-rotor
interface is actually the torque that the robotic arm needs to
provide to hold the drill.
Figure 0 shows the dynamic torque that transmitted from the
corer to the robotic arm. It is predicted by the model that
the torque reaches the steady state after a short period of
transient state. It should be noted that the solution is not
unique, since there are more than one combination of
variables in the model which would satisfy the
specifications of the corer. To be able to predict the
reaction forces more precisely, we need to know the spring
constants, k1 and k2, the mass, m1 and m2, and the damping
coefficient c2, that characterize the corer bit.
Figure 4. Torque produced by a rotary friction
corin
g
tool
Rotary-Percussive Coring Tool Model
The model of a self-excited rotary-percussive corer
proposed by Batako [5] is shown in Figure 0. The impact
loading is provided by the striker with mass m2. At impact
the friction force increases sharply, and it confines the bit to
decelerate and to come to rest. The parameters of the
system strongly affect each other, therefore a proper
correlation of masses and stiffness in the system is needed
for the process to converge to a stable solution.
The governing equations, from Batako, of the motion are:
()
()
=
+
=
.
22122
.
22
..
22
.
22
.
21211
..
11
,
,
xxΓxxkxcxm
xxΦxFxxkxkxm r
(3)
where is the friction force,
.
xFr
4
()
[]
()()
=++
+
=
0forsgn,min
0forsgn
3
1
.
2121121211
..
3
3
.
.
.
xxxkxkfxxkxk
xx
v
x
v
x
f
xF
st
crcr
st
r
(4)
and fst = μm1g is the static friction force, and μ is the
coefficient of static friction.
.
22 ,xxΓ is the impact force which is defined as
follows:
()
<
<
>
+
=
Δx
xxΓΔx
xxΓΔxxcΔxk
xxΓ
2
.
222
..
2222020
.
22
for0
0,andfor0
0,andfor
,
(5)
where k0 is constant stiffness of the bit and c0 is constant
damping, and Δ is the initial gap between bit and striker.
.
22 ,xxΦ is the added friction force due to
impact:
=
.
22
.
22 ,, xxΓxxΦ
μ
(6)
Figure 5. The planar model of the rotary-
percussive drill system. 1: Drive with constant
velocity v; 2: tip of the bit with mass m1; 3: main
spring with stiffness k1; 4: striker with mass m2;
5: lever; 6: secondary spring with stiffness k2; 7:
pin; 8: dashpot with viscous coefficient c2.
Since the rotary-percussive corer we want to model is
actually a “forced” impact system instead of a self-excited
one, we have modified the model shown in Figure 0
accordingly. The impact is created mechanically by a cam-
spring system. The cam is fixed onto the drill stem, and
while the stem rotates, the cam strains the spring and
releases it periodically to create the impact. It is assumed
that the cam-spring system creates three impacts with each
rotation of the drill stem, and since the rotation speed of the
stem is about 800 RPM, the impact rate is about 40 impacts
per second. Equation (5), the impact force is rewritten as,
() ()
<<
<<<<
=
otherwise
t
tttfa
tΓ
0
,0501.005.0
,0251.0025.0,0001.00for2sin .
0
L
π
(7)
where the impact force is assumed to be sinusoidal with a
frequency f, and f is assumed to be 5000 Hz. So the
duration of the impact is 1000 micro-seconds. The a0 is the
amplitude of the impact, and it is assumed to be 200 N.
And Equation (6), the added friction force, is rewritten as,
(
)
(
)
tΓtΦ
μ
=
(8)
Upon careful inspection of the Equation (3), it is found
inadequate to add the “added friction force” directly to the
governing equation. Instead, it should be treated as an
increase of normal force between the drill bit and the
surface of the rock, and be added to the static friction force,
fst, as shown in Equation (4). So both Equations (3) and (4)
are rewritten as,
()
()
=
=
.
22122
.
22
..
22
.
21211
..
11
,xxΓxxkxcxm
xFxxkxkxm r
(9)
and
()()
() ()()
[]
()()
=+Φ++
+Φ+
=
0forsgn,min
0forsgn
3
1
.
2121121211
..
3
3
.
.
.
xxxkxktfxxkxk
xx
v
x
v
x
tf
xF
st
crcr
st
r
(10)
Figure 0 shows the time history of the torque generated by a
rotary-percussive corer with 155 N preload, and transmitted
to the robotic arm which holds the corer. All the conditions
and parameters used in this model are the same as the ones
used for the rotary-friction model, except that periodical
impact is imposed upon the drill bit here. Compared with
the torque created by a rotary-friction corer, shown in
Figure 0, the average torque shown in Figure 0 is higher.
Also, the peaks of the torque in Figure 0 oscillate due to the
periodical impact.
5
Figure 0 shows the time history of the torque generated by a
rotary-percussive corer with 50 N preload. Compare to
Figure 0 where the preload is 155 N, the average torque
here is lower. The phenomenon is expected since lower
preload creates lower friction which results in lower torque.
Additionally, when the preload is higher, the dynamic
torque reaches its steady state faster.
Ultrasonic Coring Tool Model
An ultrasonic drill/coring tool (USDC) consists of three
main parts: an ultrasonic transducer (piezoelectric stack, a
backing element, and a horn), free-mass and a drill stem.
The ultrasonic transducer vibrates at a frequency of about
20 kHz. These vibrations of the horn tip excite the free-
mass, causing it to hop between the horn tip and the top of
the drill stem at frequencies around 1000 Hz. The free-
mass transfers energy from the ultrasonic transducer to the
drill stem. In order to determine the reaction force
transmitted from the USDC to the robotic arm, a computer
program was developed to simulate the interaction between
the ultrasonic horn and the free mass, and between the free
mass and the drill bit [4]. Time history of the location of
the ultrasonic horn was predicted by the program, and it is
assumed that the ultrasonic horn is connected to the robotic
arm through a spring. Thus the reaction force can be
calculated with the location of the ultrasonic horn and the
spring constant available. Specifications for a USDC
prototype are shown in Table 0. Figure 0 shows the time
history of the reaction force.
Figure 6. Torque generated by a rotary-
percussive coring tool with 155N preload
Table 1. Specifications of Prototype USDC
Mass 0.3 kg
Envelope 4 cm dia. × 25 cm
Power 40 watt
Resonance
Frequency 22500 Hz
Free mass 2 g
Figure 8. USDC Reaction Force
Figure 7. Torque generated by a rotary-
percussive coring tool with 50N preload
4. SIMULATION ENVIRONMENT
The cost of experimentation for rover design and control
can be significantly reduced by developing a high fidelity
virtual environment. Collaboration with Stanford
University has led to the development of SAI [8]
(Simulation & Active Interfaces), a simulation tool to assist
with the redesign, reconfiguration and control of low-mass
rovers. SAI is unique in that it is a real-time interactive
environment that allows the user to apply and sense forces
within the virtual world via haptic devices. For the
purposes of rover simulations, SAI features multi-contact
resolution for multi-body systems, efficient algorithms for
articulated body dynamics, and simulated friction and
ground reaction forces.
6
SAI is based upon a general framework [7] for the
resolution of multi-contact between articulated multi-body
systems in the context of operational space control for
robots [9]. Using this framework, the dynamic relationships
between all existing contact points can be described. These
relationships are characterized by the masses as perceived at
the contact points. A force exerted at a contact point,
whether from a collision with another object or from
interaction with a user, can be translated into forces at all
related contact points. The necessary computations can be
performed with an efficient recursive algorithm.
The contact space representation allows interaction between
groups of dynamic systems to be described easily without
having to examine the complex equations of motion of each
individual system. A collision model can be developed with
the same ease as if one were considering interaction only
between simple bodies. Impact and contact forces between
interacting bodies can then be solved efficiently.
The simulation of articulated body dynamics moving in free
space has led to a recursive algorithm for computing the
operational space dynamics and control of an n-joint
branching, redundant, articulated robotic mechanism with m
operational points [6]. An operational point is a point on the
robot at which a certain behavior is controlled; in most
cases it represents an end effector. The computational
complexity of this algorithm is O(nm + m3); existing
symbolic methods require a computational effort of O(n3 +
m3). Since m can be considered as a small constant in
practice, the algorithm attains a linear time O(n) as the
number of links increases. SAI integrates this framework
with a haptic rendering system to provide a general
environment for interactive haptic dynamic simulation.
Since SAI was initially designed as a robotic research tool,
certain modifications were made to customize the software
for rover simulations. Most of these changes were required
to refine the interaction between the wheels and the
surrounding terrain. The translational motion of wheeled
robots was previously represented by two planar prismatic
joints; now the translational effect of the rotation of each
wheel is individually modeled. This change in the modeling
of translational motion necessitated refinements in the
friction models between the wheels and the soil. Grip and
dynamic friction were added to static and viscous friction
models to provide a means of representing different soil
types the rover will encounter on Mars. Furthermore, since
SAI currently models all objects as rigid bodies, the effect
of wheel sinkage was approximated by artificially
increasing the contact area of the wheel in proportion to the
expected sinkage depth.
Figure 9. SAI simulation of the Rocky8 rover
performing a coring task. The ground reaction
forces are denoted by blue vectors whose length is
proportional to the force magnitude.
One of the major applications of SAI is to simulate the
extraction of core samples from the surface of Mars. With
the aforementioned modifications, SAI is well suited to test
different rover designs and develop appropriate control
strategies for this task. The user may analyze different
design options in a multitude of coring scenarios (different
terrain slopes and friction profiles). Moreover, different
coring tools may be evaluated in SAI by specifying the
expected forces and moments applied by the tool (Figure 0).
Using SAI to resolve the resulting ground reaction forces
allows the user to determine the effects of using a certain
tool for a specific task. The ability to analyze these options
in simulation will reduce the time and cost for rover
development.
To increase the likelihood of success, the rover should
remain stationary while extracting cores. The capability of
SAI to efficiently determine ground reaction forces and
simulate friction forces provides the ability to determine the
region of coring tool contact forces that can be applied
without causing slippage at the wheels. This analysis can
assist in evaluating different coring tools as well as
determine the minimum required mass of a rover to
successfully core using a given tool.
7
The scenario for this discussion involves the Rocky8 rover
on flat terrain. A 5 DOF arm is mounted at the front left of
the rover and is fully extended in the horizontal direction.
The coring tool is percussive (zero contact moment) and
applies a horizontal coring force. There is no assumed
wheel sinkage, and gravitational forces consistent with Mars
gravity are applied to the rover. During the simulations, the
actuated joints are controlled to maintain the given
configuration. The applied coring force is incrementally
increased until movement is observed at the rover wheels.
By repeating this simulation for different friction profiles,
the slippage boundary depicted in Figure 0 is determined. It
should be noted that the computed slippage boundary is an
approximation, but accuracy will be significantly improved
with the use of more representative terrain models
Similar to the previous example, SAI will be used as an
evaluative tool for many other design decisions. By
determining the optimal choices of coring tool type, 4 or 6
wheel base, arm mounting location and other design criteria,
SAI addresses the overall goal of minimizing the required
mass required to successfully perform coring tasks. SAI can
also be used to evaluate different control strategies. In
particular, SAI can help determine the optimal rover
placement, wheel and arm configuration, and coring
direction to increase the likelihood for successful core
extraction. Moreover, should wheel slippage be detected,
SAI can assist in designing control strategies to use the
rover degrees of freedom to assist in stabilizing the rover.
5. WHEEL-SOIL INTERACTION
A wheel-soil interaction model was developed to estimate
the traction available at the rover wheels to react coring
operation forces. The baseline design consists of a 62 kg 6
wheel rocker-bogie rover (Rocky 8), a 5 DOF arm, and a
conceptual coring tool. During the coring operation, the
coring tool requires forces and torques to be applied along
and about the coring tool axis. These forces will be
transmitted to the rover and to the ground through
wheel/soil interaction. If the coring forces exceed the
traction forces, the rover will slip.
Figure 10. Simulated wheel slippage force
boundary determined via SAI simulations. Yellow
region denotes stable forces, red region predicts
wheel slippage.
Three elements are required to predict the available traction
force. The first is a wheel/soil interaction model, the second
is a set of mass and dimensional parameters of the rover,
and third is a set of soil parameters.
Wheel/Soil Interaction Model
The fundamental equation for wheel/soil interaction is
Coulomb’s equation:
φ
σ
τ
tan
mm c
+
=
where
τ
m and
σ
mare the maximum shear stress and
normal stress at the wheel/soil interface, is the coefficient
of soil cohesion and c
φ
is the internal friction angle (of the
soil).
Coulomb’s equation can be modified to account for other
aspects of the wheel/soil interface geometry. Following
Iagnemma [10], the shear stress can be written as
()
+=
k
r
mm ec 2
1tan
θ
φστ
where
r
is the radius of the wheel,
θ
is the angle of the
contact region, and
k
is a constant. This equation follows
from equations (19) and (20) in Iagnemma by setting i
=
1.
Multiplying Coulomb’s equation by the area of contact
gives an estimate of the traction force
H
in terms of the
wheel/soil interaction geometry, soil parameters, and weight
W acting on the wheel.
.
()
+=
k
r
eWAcH 2
1tan
θ
φ
The contact area for each wheel can be expressed as
θ
brA
=
where b is the width of the wheel.
Rover Parameters
The Rocky 8 rover with a 5 DOF arm and Mast weighs 62
kg. The wheel diameter is 0.200 m and the wheel width is
0.123 m.
Soil Parameters
Soil parameters for a small range of Earth terrain can be
found in Table 1 of Iagnemma. For Sandy Loam, the
maximum rover traction is estimated to be 262 N (59 lb).
This agrees surprisingly well with some rough force
measurements of dragging the Rocky 8 rover over sand
(average of about 55 lb).
8
Soil parameters for Mars are estimated from measurements
taken from the Viking, Pathfinder, and MER missions.
Representative corresponding values for c and
φ
are as
follows (we assume a value of k = 0.025):
c=1.7 kPa and
φ
=20° (VK1)
c=1.1 kPa and
φ
=35° (VK2)
c=0.25 kPa and
φ
=35° (PF)
c=5 kPa and
φ
=20° (MER)
The only other change to estimate traction forces on Mars is
to use Mars gravity
2
693.3 s
m
g=
6. CORING TESTBED
A testbed was developed for testing coring from a
manipulator, as shown in Figure 0. A four DOF arm was
equipped with a 6-axis force-torque sensor and a
commercial coring tool. A compliant mount was used
between the tool and force-torque sensor to minimize
vibration from the tool to the arm. Cores that were acquired
were generally broken up which indicated that quality of
core is an important issue. Alternative coring tools and
mounting approaches are being considered.
7. RESULTS
Prediction of wheel-soil slippage is difficult due to the wide
variety of soil types and wheel configurations relative to the
soil. MER mission operators purposely wiggle the wheels
before performing a Rock Abrasion Tool (RAT) operation.
The RAT tool on the arm turret requires a preload against a
rock surface and the wheels are wiggled in order to sink the
wheels into the soil to provide a more stable platform for
the RAT operation. To get an idea of the scope of the
problem for coring from a low-mass rover, parameters of
the Rocky8 research rover where put into the wheel-soil
interaction model [10] to determine whether that rover
would slip in expected Mars sampling conditions. The
rover is of a similar scale as expected for a small sampling
rover. It has a six-wheel rocker-bogie configuration and
weighs 62 kg, (see Figure 0). The traction of the rover on
horizontal and 25 degree slopes were computed using the
wheel-soil interaction models. The results are shown in
Table 0.
Cores acquired using the commercial rotary percussive
coring tool were of low quality, generally broken in many
pieces. We have therefore procured a new commercial
coring tool that has less impact energy. Also, we are
planning on testing other coring tools. Work in the next
year will include determination of the need for bracing a
coring tool against the environment to provide needed tool
stability and determining what type of coring tool to use for
the coring operation from a low-mass rover. Also
coordinated rover-arm control approaches to enable
continued coring after rover slippage will be investigated.
COT coring tool
Mount/compliance
Force/Torque
sensor
Galvatron (4-DOF)
COT coring tool
Mount/compliance
Force/Torque
sensor
Galvatron (4-DOF)
Figure 11. Coring testbed
9
8. CONCLUSIONS
Coring from a low-mass rover provides new technical
challenges where the rover cannot be assumed to be a
stationary platform during the coring operation. Coring
tool models have been developed to help in the selection of
the best type of coring tool to use when coring from a low-
mass rover. A simulation environment has been developed
to simulate coring from a low-mass rover. A testbed has
been developed to test coring from a manipulator. Current
results show that stability of the coring tool relative to the
environment during the coring operations is important in
obtaining an intact core sample. On-going work will focus
on providing a stable tool relative to the environment.
ACKNOWLEDGEMENTS
Figure 12. Rocky8 rover in JPL marsyard
The research described in this paper was carried out at the
Jet Propulsion Laboratory, California Institute of
Technology, under a contract with the National Aeronautics
and Space Administration as part of the Mars Technology
Program.
REFERENCES
[1]. A. F. C. Haldemann, E. T. Baumgartner, G. H. Bearman,
D. L. Blaney, D. I. Brown, B. P. Dolgin, L. I. Dorsky, T.
L. Huntsberger, A. Ksendzov, J. C. Mahoney, M. J.
McKelvey, B. E. Pavri, G. A. Post, E. F. Tubbs, R. E.
Arvidson, N. O. Snider, S. W. Squyres, S. Gorevan, G.
Klingelhofer, B. Bernhardt, and R. Gellert, “FIDO
science payload simulating the Athena Payload,” Journal
of Geophysical Research, Vol. 107, No. E11, 8006, doi:
10.1029/2001JE001738, 2002
[2]. M. Bruno, G. Han, and C. Honeger, “Advanced
Simulation Technology for Combined Percussion and
Rotary Drilling and Cuttings Transport,” GasTIPS,
Winter, 2005.
[3]. T. Myrick, , S. Gorevan, C. Batting, S. Stoescu, M.
Maksymuk, K. R. Davis, M. A. Ummy, and the Athena
Science Team, “The Athena miniature rock coring and
rock core acquisition and transfer system (Mini-Corer),”
in Workshop on Concepts and Approaches for Mars
Exploration, abstract 6105, Lunar and Planet. Inst.,
Houston, Tex., 2000.
Table 2. Predicted traction on Mars terrains
Location Horizontal
Traction (N) 25o Slope
Traction (N)
Viking 1 97 Negative, slips
Viking 2 132 25
Pathfinder 110 4
MER 117 15
[4]. X. Bao, Y. Bar-Cohen, Z. Chang, B. P. Dolgin, S. Sherrit,
D. S. Pal, S. Du, and T. Peterson, "Modeling and
Computer Simulation of Ultrasonic/Sonic Driller/Corer
(USDC)" IEEE Transaction on Ultrasonics,
Ferroelectrics, and Frequency Control (UFFC), Vol.
50(9), p. p. 1147-1160, 2003.
[5]. A. Batako, V. Babitsky, and N. Halliwell, "A Self-
Excited System for Percussive-Rotary Drilling", Journal
of Sound and Vibration, Vol. 259(1), p.p. 97-118, 2003.
[6]. K.-S. Chang and O. Khatib, “Operation Space Dynamics:
Efficient Algorithms for Modeling and Control of
Branching Mechanisms”, Proceedings of the
International Conference on Robotics and Automation,
850--856, 2000
[7]. O. Khatib, “A Unified Approach to Motion and Force
Control of Robot Manipulators: The Operational Space
Formulation”, Int. J. Robot. Automa., 3(1), 43--53, 1987.
[8]. O. Khatib, O. Brock, K. Chang, F. Conti, D. Ruspini, and
L. Sentis, “Robotics and Interactive Simulation”,
Communications of the ACM, 45(3), 46--51, March
2002.
[9]. D. Ruspini and O. Khatib, “Collision / Contact Models
for Dynamic Simulations and Haptic Interaction”,
Proceedings of the International Symposium on Robotics
Research, 185--195, 1999.
[10]. K. Iagnemma, S. Kang, H. Shibly, and S. Dubowsky,
“Online terrain parameter estimation for wheeled mobile
robots with application to planetary rovers,” IEEE
Transactions on Robotics, 20(5): 921-927, October 2004.
BIOGRAPHY
Paul Backes, Ph.D. is the Group
Supervisor of the Mobility and
Manipulation group at Jet Propulsion
Laboratory, California Institute of
Technology, where he has been since
1987. He received the BSME degree
from U.C. Berkeley in 1982 and
Ph.D. in 1987 in Mechanical
Engineering from Purdue University.
Dr. Backes received the 1993 NASA
Exceptional Engineering Achievement Medal for his
contributions to space telerobotics, 1998 JPL Award for
Excellence, 1998 NASA Software of the Year Award Sole
Runner-up, and 2004 NASA Software of the Year Award.
He has served as an Associate Editor of the IEEE Robotics
and Automation Society Magazine.
10
Oussama Khatib, Ph.D. is a Professor of
Computer Science at Stanford University.
He received his Ph.D. in 1980 from
Sup’Aero, Toulouse, France. His current
research is in human-centered robotics,
human motion synthesis, humanoid
robotics, dynamic simulation, haptic
interfaces, and human-friendly robot
design. This builds upon a large body of work pursued over
the past 25 years and published in over 200 contributions in
the field. Professor Khatib was Program Chair of
ICRA2000 and editor of The Robotics Review. He served as
director of the Stanford Computer Forum and is currently
president of the International Foundation of Robotics
Research and editor of Springer Tracts in Advanced
Robotics. Professor Khatib is an IEEE Fellow, a
Distinguished Lecturer of IEEE, and a recipient of the
JARA Award.
Dr. Antonio Diaz-Calderon Antonio
Diaz-Calderon is a member of
Information Systems and Computer
Science Staff with the Mobility and Manipulation Group at
the Jet Propulsion Laboratory, Pasadena, CA. His research
work at JPL involves manipulation kinematics, optimal
estimation and motion control. He received the BSEE in
1987 and the M.S. in Computer Science in 1990 from
Monterrey Institute of Technology (Mexico City, Mexico),
and M.S. in 1993 in Engineering and Ph.D. in 2000 in
Electrical and Computer Engineering from Carnegie
Mellon University. He held a Post-Doctoral Fellowship
appointment with The Robotics Institute at Carnegie Mellon
University (2000-2001), and a faculty appointment as
Commercialization Specialist with The Robotics Institute
(2001-2003). While at the Robotics Institute Dr. Diaz-
Calderon developed an estimation-based tipover sensing
system to enhance dynamic stability of heavy articulated
machines, robust position estimation algorithms for the
PerceptOR program and vehicle dynamic models used for
mobility analyses on prototype vehicles for Future Combat
Systems’ Unmanned Ground Combat Systems (UGCV), and
Gladiator. His research interests include: sensor fusion and
state estimation, and rover navigation and mobility.
James Warren is a Ph.D. candidate in
the Artificial Intelligence Laboratory
at Stanford university. He received
his B.S. degree in Mathematics from
Texas A&M University in 1996 and
his M.S. degree in Scientific
Computing and Computational
Mathematics in 2000. His research is
in identifying human motion behaviors
from motion capture data and applying these strategies for
humanoid robotic control.
Curtis Collins, Ph.D. received the
B.S. Degree in Bioengineering, from
the University of California, San
Diego, and the M.S. and Ph.D.
Degrees in Mechanical Engineering
from the University of California,
Irvine. After completing his degree he
taught kinematic, robotics, and design
at the California Institute of
Technology, the University of California, Riverside, and the
University of California, Irvine. While at Caltech, he helped
to restructure the Mechanical Engineering Design
Laboratory course and continues to work with faculty to
improve student design experience. Dr. Collins joined JPL
in 2005. His research interests include the design and
analysis of parallel and multi-limbed mechanisms,
simulation and visualization of multi-loop mechanical
systems, and novel kinematic and mechanical designs for
advanced mobility systems.
Zensheu Chang, Ph.D. is an engineer
at Jet Propulsion Laboratory,
California Institute of Technology. Dr.
Chang received his MS degree from
National Taiwan University in 1988,
and Ph.D. degree in Mechanical and
Aerospace Engineering from
University of California, Los Angeles
in 1997. His research areas include
structural dynamics, ultrasonic NDE/NDT, piezoelectric
actuators, CAD/FEM modeling, elastic wave propagation,
and solid mechanics. He has published more than 50 papers
in related research areas.
11
12
... However, the rotary-percussive drills have been proven to operate at lower energies, penetrate with much lower interaction forces, and drill through much more material with the same drill bit before needing a replacement [85]. In the case of drilling on a small celestial body, the maximum weight on bit (WOB) that can be practically applied to the PRS is highly limited because of the low mass of rovers and landers [86]. An efficient solution to penetrate with a low WOB is to augment the rotary drill with a percussive mechanism; however, as a result, the power consumption and mechanism complexity would also increase [87]. ...
Article
Regolith penetration for placing instruments into ground and regolith-sampling for re-entry or in-situ analysis play an extremely critical role in searching for alien life and revealing the geological information of extraterrestrial bodies. The planetary regolith sampler (PRS), a type of device that can penetrate, collect, transfer, and stow regolith samples, is commonly equipped on a lander or rover in planetary exploration, and is regarded as essential to broaden the application scenarios of planetary robots. Because of their extensive application prospects in deep space exploration, scientists and engineers worldwide have shown great interest in the design and development of such multifunctional devices. However, owing to the significant environmental differences among extraterrestrial bodies, it is challenging to create a full-featured PRS. To date, a large number of PRSs have been designed and developed for terrestrial scientific research or extraterrestrial regolith exploration. The PRSs utilized in previous extraterrestrial regolith-sampling missions and the latest advancements are reviewed in detail. Next, this work classifies the current PRSs into six categories according their sampling methods, namely drilling, excavating/grabbing, projecting, ultrasonic/sonic, pneumatic, and bio-inspired samplers, and summarizes their general characteristics. The challenges and constraints in sampling extraterrestrial bodies, including terrestrial technology, planetary environment, and remote distance, are analyzed and discussed in depth. The critical technologies for PRSs to change from a conceptual stage to a practical prototype are detailed, including design and fabrication, tool–regolith interaction, terrestrial validation, autonomous control, and sample fidelity. Finally, the critical trends of PRS are presented, covering near-term robotic exploration to cover the full geography to long-term human settlements.
Conference Paper
A potential future mission to acquire Mars core and soil samples for return to Earth could utilize a rover with a sample acquisition tool and sample caching subsystem. Samples would need to be stored in individual sealed tubes in a canister that could be transferred to a Mars Ascent Vehicle and returned to Earth. A sample handling, encapsulation and containerization (SHEC) subsystem has been developed as part of an integrated system for acquiring and storing core samples for application to a potential future Mars sample caching mission. Requirements and design options for the SHEC subsystem were studied and a recommended design concept was developed. Two families of solutions were explored: 1) transfer of a raw sample from the tool to the SHEC subsystem and 2) transfer of a tube containing the sample to the SHEC subsystem. The recommended design, called the Bit Changeout Caching (BiCC) design, utilizes sample tool bit changeout as the mechanism for transferring tubes to and samples in tubes from the tool. Copyright © 2009 by the American Institute of Aeronautics and Astronautics, Inc.
Article
This paper described a spring roll-style retractable sampling arm with open cylindrical shell structure. It has small shrink volume, long expand path, low power consumption, large pushing force, and wide adaption, which can be widely used in mini-sampler in dangerous or ungetatable place to sample the rock, dust and other materials. Specifically, the design of the sampling arm is described, and finite element method is then carried out to study the relationship between geometric parameters and the largest pushing force. Finally, a simple and efficient mathematical formula on the maximum pushing force is given and it agrees well with the finite element analysis through the comparison between them. Experimental results show the feasibility of this design.
Article
Full-text available
The Jet Propulsion Laboratory's Field Integrated Development and Operations rover (FIDO) emulates and tests operational rover capabilities for advanced Mars rover missions, such as those originally planned for the Mars Surveyor 2001 Rover and currently planned for the Athena Payload on the Mars Exploration Rovers scheduled for launch in 2003. This paper describes FIDO's science instrument payload, which is fully integrated with rover hardware and software. Remote science teams visualize instrument suite data and generate FIDO commands using the Web Interface for Telescience. FIDO's instrument suite has been used in terrestrial laboratory and field tests to simulate Mars operations, to train Mars scientists, and to improve Mars rover mission science operations protocols. The payload includes a deck-mounted, stowable mast that is deployed for acquisition of stereo imaging and spectral reflectance data. The mast head houses Pancam, Navcam (the navigation camera stereo pair), and the Infrared Point Spectrometer (IPS). Pancam is a three-band, false-color infrared (0.65, 0.74, 0.855 mum) stereo imaging system. The three wavelengths were chosen to yield information on the ferric nature of observed minerals. IPS acquires spectral radiance information over the wavelengths from 1.3 to 2.5 mum (spectral resolution ~13 cm-1). A 4-degree-of-freedom arm is included on the front of FIDO. The arm end effector is the mounting point for a Color Microscopic Imager and an 57Fe Mössbauer Spectrometer. FIDO also carries a MiniCorer, which is an Athena prototype rock drill that can acquire 0.5-cm-diameter by up to 1.7-cm-long cores.
Article
Full-text available
As applications of robots extend into everyday human life, new approaches to simulating interactions between them and their environments are emerging at the intersection of the physical and virtual worlds.
Conference Paper
Full-text available
This paper discusses intuitive and efficient ways to model and control the dynamics of highly redundant branching mechanisms using the operational space formulation. As the complexity of mechanisms increases, their modeling and control become increasingly difficult. The operational space formulation provides a natural framework for these problems since its basic structure provides dynamic decoupling among multiple tasks and posture behaviors. Efficient recursive algorithms are presented for the computation of the operational space dynamics of branching mechanisms with multiple operational points. The application of these algorithms results in a significant increase in the interactivity and usability of dynamic control of complex branching mechanisms. The experimental results are presented using real-time dynamic simulation
Article
Full-text available
Simulation and analytical models for the ultrasonic/sonic drill/corer (USDC) are described in this paper. The USDC was developed as a tool for in-situ rock sampling and analysis in support of the NASA planetary exploration program. The USDC uses a novel drive mechanism, which transfers ultrasonic vibrations of a piezoelectric actuator into larger oscillations of a free-flying mass (free-mass). The free-mass impact on the drill bit creates a stress pulse at the drill tip/rock interface causing fracture in the rock. The main parts of the device (transducer, free-mass, bit, and rock) and the interactions between them were analyzed and numerically modeled to explore the drive mechanism. Each of these interactions is normally described by a time-dependent 2- or 3-D model involving slowly converging solutions, which makes the conventional approach unsuitable for USDC optimization studies. A simplified integrated model using tabulated data was developed to simulate the operation of the USDC on desktop PC and successfully predicted the characteristics of the device under a variety of conditions. The simulated results of the model and the experimental data used to verify the model are presented.
Article
A dynamic model for a new principle of percussive-rotary drilling is presented. This is a non-linear mechanical system with two degrees of freedom, in which friction-induced vibration is used for excitation of impacts, which influence the parameters of stick–slip motion. The model incorporates the friction force as a function of sliding velocity, which allows for the self-excitation of the coupled vibration of the rotating bit and striker, which tends to a steady state periodic cycle. The dynamic coupling of vibro-impact action with the stick–slip process provides an entirely new adaptive feature in the drilling process. The dynamic behaviour of the system with and without impact is studied numerically. Special attention is given to analysis of the relationship between the sticking and impacting phase of the process in order to achieve an optimal drilling performance. This paper provides an understanding of the mechanics of percussive -rotary drilling and design of new drilling tools with advanced characteristics. Conventional percussive-rotary drilling requires two independent actuators and special control for the synchronization of impact and rotation. In the approach presented, a combined complex interaction of drill bit and striker is synchronized by a single rotating drive.
Article
A framework for the analysis and control of manipulator systems with respect to the dynamic behavior of their end-effectors is developed. First, issues related to the description of end-effector tasks that involve constrained motion and active force control are discussed. The fundamentals of the operational space formulation are then presented, and the unified approach for motion and force control is developed. The extension of this formulation to redundant manipulator systems is also presented, constructing the end-effector equations of motion and describing their behavior with respect to joint forces. These results are used in the development of a new and systematic approach for dealing with the problems arising at kinematic singularities. At a singular configuration, the manipulator is treated as a mechanism that is redundant with respect to the motion of the end-effector in the subspace of operational space orthogonal to the singular direction.
Article
In the article, the author says that as applications of robots extend into everyday human life, new approaches to simulating interactions between them and their environments are emerging at the intersection of the physical and virtual worlds. Robots today are being moved toward applications beyond the structured environments of manufacturing plants, making their way into the everyday world of social, entertainment, and industrial environments inhabited by people. The successful introduction of robotics into human environments will depend on the development of competent and practical systems that are dependable, safe, and easy to use. As advances are made in the methodologies and techniques needed to address these challenges, it has become increasingly apparent that their effect goes beyond the physical robot. For robots with human-like structures, tasks are not limited to the specification of the position and orientation of a single effector, or operational point. A robotic system must be capable of a sufficient level of competence to avoid obstacles during motion.
Article
Future planetary exploration missions will require wheeled mobile robots ("rovers") to traverse very rough terrain with limited human supervision. Wheel-terrain interaction plays a critical role in rough-terrain mobility. In this paper, an online estimation method that identifies key terrain parameters using on-board robot sensors is presented. These parameters can be used for traversability prediction or in a traction control algorithm to improve robot mobility and to plan safe action plans for autonomous systems. Terrain parameters are also valuable indicators of planetary surface soil composition. The algorithm relies on a simplified form of classical terramechanics equations and uses a linear-least squares method to compute terrain parameters in real time. Simulation and experimental results show that the terrain estimation algorithm can accurately and efficiently identify key terrain parameters for various soil types.
The Athena miniature rock coring and
  • S T Myrick
  • C Gorevan
  • S Batting
  • M Stoescu
  • K R Maksymuk
  • Davis
. T. Myrick,, S. Gorevan, C. Batting, S. Stoescu, M. Maksymuk, K. R. Davis, M. A. Ummy, and the Athena Science Team, " The Athena miniature rock coring and