Active Tactile Sensing
Although the sensory organs of arthropods are
morphologically distinct from their vertebrate equivalents,
they serve the same purpose and are often based on similar
neural mechanisms. Whereas the visual, auditory and olfactory
sensory systems of insects are very well studied, the tactile
sense has received much less attention. This is in harsh
contrast to the fact that many insect species make intense
tactile use of their antennae. Behavioural contexts, in which
antennal mechanoreception are relevant, range from simple
orientation and escape behaviours (e.g. in cockroaches), to
active exploration (e.g. in stick insects), pattern recognition
(e.g. in honey bees) and communication (e.g. in crickets).
Whereas there is reasonably good knowledge on insect
mechanoreceptors and their transduction mechanisms, their
distribution and the interaction (or specific function) of
different types of afferent mechanosensory information is
hardly studied. Particularly the impact of active movements on
tactile sensing has been neglected. The projects outlined below
are representative for our approach to shed more light on the
workings of active tactile sensors - both in insects and in
technical application.
A) The role of the antennae in insectlocomotion
Stick insects continuously move their antennae during walking.
Also, they coordinate the movement of their antennae with the
stepping cycle of the legs (Dürr et al., 2001). Nevertheless,
the exact coordination pattern also depends on the behavioural
context. For example, it changes when searching for foothold
(Dürr, 2001).
It s possible to show that the typical movement pattern is
suitable for early detection of obstacles in a height range
that requires a change in locomotor strategy. Indeed, stick
insects use antennal tactile information for early body axis
inclination in a climbing paradigm (Dürr et al., 2003).
The photograph shows a stick insect walking to
the left. Stroboscopic illumination during a 1 s exposure
reveals that antennae are continuously moved during
locomotion.
References:
Dürr V (2001) Stereotypic leg searching-movements in
the stick insect: Kinematic analysis, behavioural context and
simulation. J. Exp. Biol. 204 (9): 1589-1604.
Dürr V, König Y, Kittmann, R (2001) The antennal
motor system of the stick insect Carausius morosus:
Anatomy and antennal movements during walking. J. Comp.
Physiol. A 187 (2): 131-144.
Dürr V, Krause A, Schmitz J, Cruse H (2003)
Neuroethological concepts and their transfer to walking
machines. Int. J. Robotics Res. 22 (3), 151-167.
Student project:
Tactile use of antennae in obstacle climbing (Tamina Pinent).
B) Theoretical analysis of active tactile sensors
Why are the joint axes of the stick insect antenna oblique and
non-orthogonal to each other, whereas those of the cricket
antenna are aligned orthogonally with respect to the
horizontal?
Does the movement pattern of a stick insect antenna make sense,
i.e. can we identify a possible reason why it could pay off for
the insect to move its antennae in the way it does?
Answering such questions means understanding active tactile
sensing in general. One principal problem here is that both the
construction and the movement strategy of the feeler have great
impact on its sampling efficiency. Also, sampling efficiency of
a given strategy depends on external, i.e. environmental
features. Accordingly, aspects of morphology, movement
physiology and environment of an antenna can have conflicting
impact on its performance. Thus, there exist trade-offs between
various parameters. General information about how a given
parameter affects the workspace or tactile acuity of an
antenna, and particularly the trade-offs between two or more
parameters, are best investigated by modelling studies. This is
because modelling allows complete control of any experimental
parameter. Theoretical studies are necessary to allow us to
interpret the shape and movement strategy of insect antennae,
but also to design biomimetic sensors, e.g. for use on a legged
robot.
An example video of an tactile efficiency evaluation of a
simple antenna with two hinge joints
can be downloaded by clicking on this link (5.2 MB)
Reference:
Dürr V, Krause A (2002) Design of a biomimetic active
tactile sensor for legged locomotion. In: Bidaud P, Ben
Amar F (eds) Proc. 5th Int.Conf. Climbing and Walking Robots
(CLAWAR 2002). Professional Engineering Publishing, London,
255-262.
PhD student project:
Tactile efficiency of insect antennae with two hinge joints.
(Andre Krause)
C) Biomimetic tactile sensors for a walking
robot
This section is currently being updated.
Current project:
Tactile antennae on hexapod walking robot TARRY
Did you know ?
An important aspect to bear in mind when studying antennae as
active tactile sensors is the following interpretation:
Insect antennae are in fact transformed legs. In the
course of evolution, leg-like limbs of the head region are
thought to have lost their propulsive function and have become
specialised sensory probes instead. The following pictures
illustrate one reason, why biologists think so.
Cuénot (1921) demonstrated that amputation of the
antenna within a narrow band on the distal pedicel (the 2nd
antennal segment) can cause ectopic regeneration of a leg
instead of an antenna. Thus, antennal tissue seems to have not
only the information but also the 'ability' to develop into a
leg.
This shows the result of an experiment as done by Cuénot.
Left: Top view of an intact adult stick insect (Carausius
morosus). Right: Adult stick insect after antennectomy
(probably in 3rd instar) with regenerated tibiae and
tarsi.
|
Universität | 
International | 
Benutzer | 
