Neural Mechanisms of Leg Motor Control in Crayfish - Advancing Artificial Intelligence Through Biological Process Applications - page 22

INTRODUCTION

(see Cattaert & Le Ray, 2001). Here, we describe,
the encoding signals generated by a specific sen-
sor that monitors both upward and downward leg
movements, as well as the integrative mechanisms
used to process these signals within the central
nervous system (CNS). The many ways of tun-
ing sensory-motor processes are then presented,
which allow the system to adjust perfectly the
motor command and, consequently, to generate
fully adapted behaviors (Clarac et al., 2000).
Some possible applications and future research
directions will conclude this chapter.

In the second half of the last century, diverse sets
of computer or robotic models have appeared in
an attempt either to explain or reproduce various
features of behavior. Although the majority of
models first arose from mathematical approaches,
the use of ideas originating from biological stud-
ies are now pre-eminent in the construction of
artificial organisms designed to achieve a given
task. However, although such robots perform quite
well ( e.g. , the salamander robot by Ijspeert et al. ,
2007), the actual animal's performance always
appears much more efficient and harmonious. This
probably results from the continuous interaction
between motor activity and sensory information
arising from the surrounding environment (Ros-
signol et al. , 2006), in association with the perfect
use of the biomechanical apparatus. It is thus pri-
mordial to understand fully the neural mechanisms
involved in such a dynamic interaction to be able
to implement realistic integrative algorithms in
the design of computational models.
In this context, knowledge in invertebrate neu-
roethology has demonstrated unique advantages
for engineering biologically-based autonomous
systems ( e.g. , Schmitz et al. , 2001; Webb, 2002).
Although invertebrates are able to generate com-
plex adaptive behaviors, the underlying neuronal
circuitry appears quite simple compared to ver-
tebrates. This chapter aims at presenting some
basic neuronal mechanisms involved in crayfish
walking and postural control involving a single
key joint of the leg. Due to its relative simplicity,
the neuronal network responsible for these motor
functions is a suitable model for understanding
how sensory and motor components interact in
the elaboration of appropriate movement and,
therefore, for providing basic principles essential
to the design of autonomous embodied systems.
In walking legs for example, sensory information
is provided by simple sensory organs associated
with each joint, and integrated by relatively small
populations of neurons within the central ganglia

MULTI-SENSORy CODING OF LEG
MOvEMENTS

Crayfish possess an external skeleton that allows
movements only at the various joints, the move-
ment of each joint being coded by simple sensory
organs. Among these, the leg coxopodite-basipo-
dite chordotonal organ (CBCO) plays a pivotal role
in the control of locomotion and posture, since
it monitors vertical leg movements (Figure 1).
This proprioceptor consists of an elastic strand
of connective tissue in which sensory cells are
embedded and whose function is comparable
to that of joint receptors in mammals (Clarac et
al. , 2000). The CBCO strand is stretched during
opening of the second, unidirectional joint and
released during closure, which corresponds re-
spectively to downward and upward movements
of the leg. The sensing apparatus of the CBCO is
composed of 40 neurons that are equally divided
into 20 stretch-sensitive and 20 release-sensitive
neurons that code depression and levation of the
leg, respectively (see Cattaert & Le Ray, 2001).

Coding Movement Parameters

Joint movement can be monitored either as a
displacement of one segment relative to the other
(dynamic parameter) or in terms of the relative
position of both segments (static parameter). In