Improved Haptic Feedback Realism using a Multimodal Haptic
Interface and Model
N.G.Tsagarakis, Y. Sarakoglou, and Darwin G.Caldwell
Dept. of Electronic and Electrical Eng.,
Telford Building, University of Salford,
Manchester,M5 4WT, UK
Tel: +44-(0)161-2954010
Email: n.tsagarakis@salford.ac.uk
Abstract - In the real world people can interact with
real objects and be able to identify them by using the
haptic information that the objects transmit. This
implies that, virtual objects should also have haptic
features such as stiffness, friction, texture and
temperature just as the real objects do. Vital
requirements for producing multimodal realistic
haptic cues are the ability to co-ordinate kinaesthetic
and cutaneous cues within the virtual environments as
well as the existence of interfaces capable of
generating a wide range of sensations and the
modelling of these sensations. Most of the existing
haptic systems have limited capacity to generate
multimodal feedback and are more dedicated to
simulating some aspects of the human touch
sensations. In this work the rendering of these
multimodal cues is performed using an integrated
multimodal haptic interface. The hardware
components of the multimodal haptic interface,
employing a full arm exoskeleton kinaesthetic device,
a finger grasping force feedback exoskeleton and a
wireless input/cutaneous feedback glove interface are
presented.
I. INTRODUCTION
Advances in VR and telepresence have highlighted the
need for ever more effective human-computer interfaces
and this has produced astonishing improvements in the
capacity to produce compelling visual and audio effects.
However, as this ability to view this virtual world has
improved it becomes increasingly clear that the capacity
for physical interaction forms a vital thread in moving to
the next level of virtual communication.
The ability to physically interact with virtual objects as
well as observing them forms the key advancement
offered by Haptic (touch/force) technology and the lack of
effective input/feedback system has been identified as a
key gap area in the generic technology for the exploitation
of VR/telepresence. This has led to research activity in all
aspects of input and feedback technology related to
cutaneous/tactile (skin based sensation) and
kinaesthetic/force (deeper muscle and joint based
sensation) modelling and reflection [Linderman et al.,
1999]. With the core field of haptics there is a number of
key technologies that have sought to address different
aspects of the overall haptic need from the finger tips to
the large motions of the arm.
Among the most important hand motion sensing systems
are the VLP DataGlove [Zimmerman and Lanier, 1987],
the original hand tracking system based on detection of
finger movements using a fibre-optic system, and more
recently the CyberGlove [Virtex Co., 1999] a strain
gauge based system, which also has additional cutaneous
tactile feedback (vibration) functionality. Unfortunately,
most of the high quality data glove systems still have only
limited cutaneous capacity, lack true portability and
flexibility and costs are often prohibitively high.
Nonetheless the ability to track finger motions is now
achieving a level were accurate monitoring is becoming
increasingly reliable.
Systems aimed at simulating some aspects of the human
tactile (skin) sensing have also been produced including:
Air jet or bladder displays, piezoelectric disks, micro-pin,
arrays, voice coil displays, electrotactile [Shimoga.,
1993][Kaczmarec et al., 1991][Kontarinis and Howe
1993][Hasser and Weisenberger, 1993][ Kammermeier,
2002] and thermal displays [Benali-Khoudja 2003].
Depending on the actuation technology used these tactile
feedback systems can simulate limited aspects of human
cutaneous/tactile sensation. However, most of them are
relatively or very bulky, heavy desk mounted displays,
which cannot move freely with the operator’s hand and
explore the environment in a natural and realistic manner.
Further they are unable to provide the truly generic
cutaneous feedback needed to create a haptic illusion that
is compelling under a wide range of applications since
each device can only simulate specific and relatively
limited sensations.
Devices to measure and feedback forces derived from the
motion of the upper limbs, hand or arm form another
discrete area of development with a number of force
feedback devices having been designed and tested.
Generally these are classified as desk-top/joysticks
[Burdea, 1996] and arm/hand masters/exoskeletons
systems depending on the mounting structure.
Considering in particular arm exoskeleton systems, these
devices have some important advantages compared to the
.....