IEEE Robotics & Automation Magazine - September 2022 - 60

research effort is always devoted to fluid dampers with fixed
and variable rheology. Between the two, variable rheology has
the lion's share, an aspect that partially correlates [also see Figure
5(b)] with their intrinsically controllable nature and the
predilection for semiactive solutions observed in Figure 4(a).
Indeed, in the design of semiactive systems, variable-rheology
dampers are easy to control in the absence of additional movers.
Papers about electromagnetic dampers are scarce and
concentrated between 2006 and 2013. We suspect that their
poor success could be explained by the observation that the
size and mass of an electromagnetic device that should dampen
a torque at a given speed closely resembles the size and
weight of the electric motor used to move the actuator in the
first place. This makes it more difficult to find applications
where performing dampening with active solutions on the
motor side is not more convenient. Finally, from 2010 to the
present, we observe a substantial shift toward friction-based
dampers. Once again, we presume that their simplicity and
lack of fluids (and seals and leaks), together with their effectiveness,
motivates this approach now that the technology is
nearing real applications.
Figure 4(c) compares the effort devoted to developing different
topologies through the years. As anticipated in the
" Topology " section, the SEDA architecture is the oldest and
most investigated in the referenced papers. A possible reason
for this could be the topology's reproduction of the traditional
mass-spring-damper conceptual architecture, which the
SEAwEPD began altering to offer a different dynamic characteristic
(recall Table 4) that may be more suitable for some
applications. Nevertheless, one thing keeping the SEDA in use
is its suitability to passive and semiactive solutions, whereas the
SEAwEPD is almost always employed in semiactive scenarios.
Finally, note that the rSEDA topology appeared recently in the
literature, and only a few works about it exist. Although we
acknowledge its originality, which made it worthy of being
included in this article, it is difficult to locate trends and correlations
between it and other categories, due to its novelty.
Therefore, in Figure 5(a) and (b), we grouped rSEDA and
SEDA configurations, as they share many properties.
Figure 4(d) examines how damping systems for robotic
actuation are spread across the fields presented in the " Applications "
section. We observe that a substantial number of the
earliest works (all until 1997, most until 2005) do not focus on
one application but remain general purpose, a trend that is
pretty natural for the earliest development stages of a technology
characterized by a low readiness level [112]. To confirm
Not Specified
Electromagnetism-Based Dampers
Fiction-Based Dampers
Semiactive
100
80
60
40
20
1
4 33
1
1
1
4
10
11
6
6
Passive
100
80
60
40
20
Fluid Dynamic-Based Dampers
5
1
1
1
2
5
3
3
1
2
1
8
6
1
5
2
(a)
(b)
SEDA SEAwEPD
100
70
90
80
50
60
30
40
20
10
1
2
4
1
1
4
8
3
4
6
2
1
rSEDA
100
70
90
80
50
60
30
40
20
10
1
1
1
4
MHW HAL
ICR
2
1
2
6
General Purpose
2
5
4
4
1
4
4
1
3
1
4
(c)
(d)
Figure 4. The trends for (a) damping types, (b) technologies, (c) topologies, and (d) applications.
60 * IEEE ROBOTICS & AUTOMATION MAGAZINE * SEPTEMBER 2022
(%)
(%)
(%)
(%)
1990-1993
1994-1997
1998-2001
2002-2005
2006-2009
2010-2013
2014-2017
1990-1993
2018-2021
1998-2001
2002-2005
1990-1993
2006-2009
1994-1997
2010-2013
1998-2001
2014-2017
2018-2021
2002-2005
2006-2009
2010-2013
2014-2017
1990-1993
2018-2021
1994-1997
1998-2001
2002-2005
2006-2009
2010-2013
2014-2017
2018-2021

IEEE Robotics & Automation Magazine - September 2022

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - September 2022