Systems, Man & Cybernetics - January 2016 - 23

framework. De tails of the tests
are discussed in Part II of "Supplementary."
Results of the tests have shown
that the proposed framework has
a stable performance across SPHs
and MPHs. In case 1, its update
r a t e i s sl ig ht ly slower t h a n
CHAI3D with both homogeneous
and heterogeneous device setups.
In case 2, it performed obviously
faster than CHAI3D, where the
three distinct HIPs are transmitting data through a single pipeline
and interfinger communications
are done within a single manuallevel component instance. A perfor ma nce difference between
CHAI3D and the proposed framework is
also observed to be more significant with
the increment of the complexity and heterogeneity in the MPHs.
Extendibility
T he f ra mework a nd it s u nderly i ng
abstract MPH model can potentially
increase the complexity of MPHs so as to
help establish more immersive and practical applications. Figure 4 demonstrates
the transition map among various MPH
configurations, as well as the conditions
required to fulfill the transition from one
configuration to another. All MPH configurations listed in Figure 4 are supported
by the framework.
Based on Figure 4, we discuss an
application example that is very representative in actual haptic-related training and simulation tasks but can hardly
be implemented without the framework.
Consider an emergency medical training scenario of tension pneumothorax,
where the procedure of needle thoracotomy has to be performed based on the midclavicular approach. The major tasks in
this procedure include:
◆ Finger palpation on the chest to
identify the second intercostal spaces between the second and the third
ribs. This usually involves multiple
fingers gliding on the skin while
occa siona lly pushing dow nwa rd.
This requires multisensory interactions to concurrently receive tactile
a nd k i ne s t he t ic fe e d ba ck f r om
the fingers.

◆ Once the insertion location has

been determined, sterilization
will be performed. This involves
a bimanual operation to tear
open the packing of an antiseptic pad and sterilize around the
location. Although this is a simple task for human hands, it
involves multimanual and multifinger interactions and can be
very difficult to implement on
existing MPHs and frameworks.
The thumbs and index fingers
from both hands have to constantly exert pressure on the
packaging, while the hands are
twisted to open it.

Due to the limitations
on hardware, computation power, and
cost, the development
of MPHs has been
evolving slowly in the
past decades, and
many existing solutions act in their own
ways.

Single-Point
Kinesthetic
Haptics

Increase HandLevel Interaction
Multimanual
Kinesthetic Haptics
Without Finger-Level
Interactions [17]

Tactile Array [33]

Increase FingerLevel Interaction

Single-Manual,
Multifinger Kinesthetic
Haptics [12]

Combine

Combine
Holistic Single-Point
Haptic Device [26]

Combine

Multimanual, Multifinger
Kinesthetic Haptics [21]
Single-Hand,
Multifinger
Holistic MPH

Combine
Multimanual, Multifinger
Holistic MPH

Figure 4. an overview of transitions among various mPH configurations based on the proposed abstract mPH model. Conditions that
are required to fulfil the transition between configurations are
also shown.

Ja nu a r y 2016

IEEE SyStEmS, man, & CybErnEtICS magazInE

Table of Contents for the Digital Edition of Systems, Man & Cybernetics - January 2016