IEEE Robotics & Automation Magazine - December 2013 - 79

PackBot, a Typical Tracked
Vehicle with Tracked "Flippers"

0.14

0.16

0.12

0.14

0.1
0.08

Wheels
Tracks
Legs
Combination
Modular

0.06
0.04
0.02

101

102
103
Complexity

104

Figure 5. The upper bound of versatility versus complexity
categorized by locomotion mode.

Tracked Versus Wheeled Robots
For a given complexity, tracked systems have higher versatility
than wheeled or legged systems. This is because of increased
mobility on deformable terrains and use of skid steering,
which helps robots maneuver in tight environments. The fact
that tracked systems are generally considered more mobile
than wheeled systems in the vehicle design community serves
as a check for the approach proposed in this article. Tracked
systems have other design concerns (e.g., low efficiency);
however, the approach presented here focuses solely on a
system's versatility.
Legged Versus Wheeled Robots
For a given complexity, legged systems are more versatile than
wheeled systems. We originally hypothesized that wheeled
systems would have greater versatility, as legged systems are
generally more mechanically complicated. However, most
of the legged robots shown in Figure 5 use rotary legs
(e.g., RHex [52], Loper [11], Asguard [33], and Whegs II
[78]), which greatly simplify the mechanical design. Thus,
they combine the mechanical simplicity of a wheel with the
vehicle/terrain interaction benefits of legs. Other legged systems tend to have both poor maneuverability and obstacle
mobility, which lead to low versatility scores.

Versatility

0.16

0.12
0.1
0.08
0.06

Talon, a Typical Tracked
Vehicle
0

RHex
0.12 Asguard
0.11 Whegs II
Loper
0.1 1
10

Wheeled Systems with
Active Suspensions
ATHLETE

102
Complexity

103

Figure 6. A comparison of rotary leg and active-suspension
wheeled robots.

Robots That Use Multiple Locomotion Modes
Although only one combination system (i.e., it uses both
wheels and tracks) was studied, it did not offer any significant advantage. However, robots that combine multiple technologies do have advantages. Two examples are the
aforementioned rotary legs and active-suspension wheeled

0.12

Vortex

ACM-R5

MMALV

0.1

Whegs II

0.08
0.06

Crusher

Modular Robots
Figure 5 shows that, for a given complexity, modular systems
do not demonstrate high versatility. This is most likely
because most modular systems are able to surpass a large
number of obstacles or exhibit locomotion on a wide variety
of terrains, but they are not particularly proficient on any of
them. Thus, the systems typically received more scores of
1 than of 2 in their versatility assessment.

Versatility

0.13

The dichotomic design to rotary legs is wheeled systems
with articulated active suspensions, or legs with wheel
end effectors. Examples include Crusher [20] and NASA's
ATHLETE [13]. A comparison of these two classifications of
robots is shown in Figure 6. Both systems offer good versatility; however, rotary leg robots do so with less complexity.
Nonetheless, as seen in in Figure 4, adding an articulated suspension greatly increases the versatility when comparing only
wheeled systems.

0.14
Rotary Leg Systems

30
40
Complexity

Figure 7. Versatility versus complexity for tracked robots.

0.16
0.14

Chaos, Two
Tracked Flippers

0.04
101

Amphibot II
102

103

104

Complexity
Figure 8. Versatility versus complexity for systems explicitly
designed to operate in multiple environmental domains.

DECEMBER 2013

IEEE ROBOTICS & AUTOMATION MAGAZINE

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - December 2013