A Screw Propulsion Design for Mobility in High

46th Lunar and Planetary Science Conference (2015)
B. Settle2, D. Battel2, A. Carsello2, A. Cooperberg2, D. Gebreselasse2, J. Gross2, M. Kelley2, P. Kharel2, S.
Lazechko2, H. McConnell2. 1Washington University in St. Louis, Dept. of Earth and Planetary Science, St. Louis,
MO ([email protected]); 2Washington University in St. Louis, School of Engineering And Applied Science, St. Louis, MO.
Introduction: The Screw-Propelled Automated
Regolith Collecting (SPARC) robot is Washington
University in St. Louis’ (WUSTL) entry into NASA’s
annual Robotic Mining Competition, in which robots
must navigate and mine in a field of BP-1 lunar regolith simulant. The simulant, which is characterized by
relatively low cohesion and a high shear deformation
modulus, demands novel mobility approaches to minimize sinkage and slip. SPARC’s mobility platform is
based on an Archimedes screw pontoon system that
minimizes sinkage and provides additional mechanical
advantage in low cohesion materials relative to more
traditional wheel designs. Screw-propelled vehicles
have long been used terrestrially to navigate through
snow and swamps and, more recently, in tailings management. This abstract describes the design and testing
of the SPARC screw-propelled mobility system in BP1 regolith simulant.
Screw Propulsion: Screw-propelled vehicles use
one or more cylinders fitted with helical flanges to navigate terrains typically characterized by high sinkage.
Two-cylinder screw-propelled vehicles such as SPARC
are capable of forward and backward (longitudinal)
motion during which the augers are rotated inward and
outward, respectively, and side-to-side (lateral) motion
during which the augers are simultaneously rotated in
the drive direction. During lateral drives, the screws
can be treated as long, non-deformable wheels with
motion dominated by shear-stresses between the cylinder surface and the soil. Ignoring flight-soil interactions, lateral motion is controlled by motion resistance
(R) and traction (T), which are related to normal and
shear stresses as T = τcosθ and R = σsinθ where τ and
σ are shear and normal stress, respectively, and θ is the
angle between the side of the screw and the terrain at
the exit point (Fig. 1). In deformable surfaces such as
lunar regolith, τ and σ are a function of soil properties
including cohesion, internal friction angle, and shear
modulus [2]. Traction and motion resistance are functions of wheel sinkage. Due to the large area of the
screw-terrain interface, which reduces surface pressure,
screw-propelled vehicles are characterized by relatively low sinkage that minimizes motion resistance and
maximizes traction.
Fig. 1: Cutaway view of pontoon. Lateral motion is
dominated by shear and normal stresses between the
pontoon and surface.
During longitudinal drives, the vehicle is propelled
by interactions between the flanges and the surrounding medium with an efficiency proportional to the helix
angle (Fig. 2). Variations in helix angle affect multiple
parameters including ground deformation, drawbar
pull, and slip, although no single angle optimizes every
parameter [3].
Fig. 2: Side view of SPARC pontoon.
SPARC Mobility Design: SPARC consists of two
91.45 cm long, 15.25 cm wide carbon-steel tubes with
oppositely threaded, 3.8 cm high continuous helix
grousers made from 12-gaugue carbon steel. Each isdriven by an 88:1 geared 12 V 6.35 cm brushed DC
motor. A hollow axle carries power from an external
power source to the internal motors, which are shielded
from dust by PVC endcaps that serve the additional
purpose of bulldozing material during longitudinal
drives. A helix angle of ~25⁰ was selected to balance
motor efficiency with slip and drawbar pull (Fig. 3).
The screws propel the 1.2 m long, 0.75 m wide, 0.75 m
high, ~175 kg robot (Fig. 3).
46th Lunar and Planetary Science Conference (2015)
with the surrounding media as the primary form of
propulsion, outperforms lateral driving in materials
with high shear deformation modulus (Fig. 6).
Future Work: Several changes to the screwpropulsion design will be pursued in order to further
examine drive performance in BP-1 and similar media
including weight reduction through the use of carbon
fiber tubing, increasing drive actuator torque, and
steepening flight angle to maximize drawbar pull.
Fig. 3: View of fully assembled SPARC robot.
Performance: Assuming limited skid, the maximum longitudinal travel distance of the robot per rotation d of the screws is πdtanϕ for flight angle ϕ, allowing lateral propulsion to drive the vehicle d/(πh) times
farther per auger rotation than longitudinal driving.
Consequently, lateral locomotion is preferred in terrains characterized by relatively low cohesion and
shear deformation modulus (Fig. 4). In media with a
high (>30 mm) shear deformation modulus, longitudinal drives approach and exceed lateral drives in displacement per auger rotation.
Fig. 5: Measured wheel slip as a function of drive distance for lateral and longitudinal drives in beach sand
and BP-1.
Fig. 4: Displacement as a function of pontoon rotations
for longitudinal and lateral drive modes.
In order to compare pontoon slip as a function of
drie distance, SPARC mobility was tested in two media: well-sorted beach sand with small, rounded grains,
and BP-1. Slip was estimated by comparing the distance between flight marks with commanded displacement. During low sinkage (< 4cm) traverses, SPARC
incurred lateral slip of less than 10% in both BP-1 and
sand, compared to longitudinal slip of ~15-20% in both
materials (Fig. 5). Comparing slip as a function of lateral and longitudinal shear deformation modulus indicates that longitudinal driving, which engage flights
Fig. 6: Slip as a function of shear deformation modulus
during lateral and longitudinal drives.
References: [1] Cooling D. J. (2009). Austalian
Centre for Geomechanics. ISBN 0-9756756-7-2.
[2] Wong J. (2001). Theory of Ground Vehicles, 3rd ed.
9902-4. [3] Cole N. (1961). Proc. Inst. Mech. Eng.