INSECT SENSORY SYSTEMS INSPIRED COMMUNICATIONS

AND COMPUTING (II): AN ENGINEERING PERSPECTIVE

Zhanshan (Sam) Ma, Axel W. Krings and Robert E. Hiromoto

Computer Science Department, University of Idaho, Moscow, ID 83844, USA

Key

words: Insect Sensory System, Micro Aerial Vehicle (MAV), Wireless Sensor Network, Non-cooperative Social

Behaviour, Insect-Inspired Robot, Cellular Computing, Agent-based Computing, Biosensor, Dendritic

Neuronal Computing, Molecular Network.

Abstract: This is the second article in a two-part series in which we briefly review state-of-the-art research in

communications and computing inspired by insect sensory systems. While the previous article focuses on

the biological systems, the present one briefly reviews the status of insect-inspired communications and

computing from the engineering perspective. We discuss three major application areas: wireless sensor

network, robot and micro aerial vehicle (MAV), and non-cooperative behaviours in social insects and their

conflict resolution. Despite the enormous advances in insect vision and mechanosensory inspired robot and

MAV, micro-flight emulation, motion detection and neuromorphic engineering, etc., the potential

inspiration from insect sensory system is far from being fully explored. We suggest the following promising

research topics: (1) A new grid computing architecture emulating the neuronal population such as the visual

neurons that support the compound eyes, the PN (Projection Neurons) in AL (Antennal Lobe) or the ORN

(Olfactory Receptor Neurons) from insect sensory organs (sensilla). This may be further integrated and

enhanced with the dendritic neuronal computing. (2) New generation of multimodal wireless sensor and ad-

hoc networks that emulates insect chemosensory communication. The inspiration of multimodalities in

insect sensory systems also implies that there are multiple parallel networks operating concurrently.

Furthermore, the insect chemosensory is significantly robust and dependable with built-in anti-interference

mechanisms. (3) Non-cooperative behaviours in social insects may offer insights to complement swarm

intelligence (inspired by cooperative behaviours) or to devise new optimization algorithms. It may also

provide inspiration for proposing survival selection schemes in evolutionary computing. We suggest using

evolutionary game theory to model conflict resolution in social insects, given its success in modelling

conflict resolution of other animals.

1 INTRODUCTION

Organisms interact with each other and with their

environments through sensory and motor systems;

so do the engineered systems. Their stability and

control depend on continuous sensing and actuation

(Miesenbock and Kevrekidis, 2005). This argument

shows the universal significance of sensory systems

to both biological and engineered systems, which is

particularly true to insects given insect sensory

systems are one of the top four reasons contributing

to their status as the most abundant organisms on

earth (Ma and Krings, 2007).

Two terms often appear in bio-inspired

com

puting: biomimetic and biomorphic. The former

is more common and emphasizes the mimic or

emulation of nature and the latter is more of a

metaphor (Lodding, 2004). The applications we

survey below largely fall into one of the two

categories, but in reality, the distinction is rarely

clear-cut. In some occasions, a biologically inspired

approach is recursively applied to solve biological

problems (e.g., biosensoring in section 3).

To harness the biological inspirations from insect

sen

sory systems, being familiar with the biological

aspects is necessary. We refer to the following

excellent monographs (Christensen 2005,

Drosopoulos and Claridge 2006), both of which are

dedicated to insect sensory systems. An excellent

and up-to-date monograph, rightly acclaimed by

reviewers as providing “a remarkably holistic yet

detailed view” on insect physiological systems

including the sensory systems, should be an ideal

reference for studying insect sensory system in more

comprehensive context (Klowden, 2007). General

knowledge on insect sensory systems can be found

in an entomology textbook such as Gullan and

Cranston (2005). Numerous proceedings from

292

(Sam) Ma Z., W. Krings A. and E. Hiromoto R. (2008).

INSECT SENSORY SYSTEMS INSPIRED COMMUNICATIONS AND COMPUTING (II): AN ENGINEERING PERSPECTIVE.

In Proceedings of the First International Conference on Bio-inspired Systems and Signal Processing, pages 292-297

DOI: 10.5220/0001069602920297

 SciTePress

symposiums and conferences on bio-inspired

computing have been published since the 1990s and

a significant amount of research is inspired by

insects (e.g., Detrain et al. 1999, Dressler and

Carreras 2007). A possible starting point, which

provides an article-level review of the insect sensory

systems from the perspectives of inspiring

communications and computing, could be the Ma

and Krings (2007). Given the extensive existing

literature, which continues to accumulate faster than

ever, we choose a significantly narrow scope in this

article to focus on insect sensory system related

topics. Even with the narrowed-down scope, it still

seems impossible for us to present a comprehensive

review in such a short article. Therefore, we choose

to focus on three research areas and exclude the

others. In addition, priority was given to the state-of-

the-art review papers, monographs, and research

papers representing a major category of studies

(often limited to one per topic). Consequently, we

have to regrettably omit a number of excellent

research papers. As a minor remedy to the excluded

fields, in section 3, the other topics, we mention five

areas and a few review references about them.

2 INSECT SENSORY SYSTEMS

INSPIRED COMMUNICATIONS

AND COMPUTING

2.1 Wireless Sensor Networks

It seems that insect sensory systems may inspire the

design of wireless sensor networking on both the

node (sensor node vs. individual) level and network

level (sensor network vs. insect population).

The inspiration at the individual sensor node level

is the most obvious. Essentially, a robot emulation

of insect vision and navigation provides a typical

example for this kind of research, where each

individual insect is mapped to an engineered sensor.

Many of the neural sensory mechanisms in insects

can be emulated in individual sensor design. In

particular, multimodality capability is very desirable

in sensor networks (Ma and Krings, 2007).

From the population perspective, potentially two

types of “mappings” can be construed. The first type

is the neuron population or the group of neurons

behind a sensory organ such as antenna or

compound eyes. This type of neuron population

forms a grid computing infrastructure (similar to the

cellular computing paradigm). The populations of

ONRs (olfactory neural receptors) and PNs

(projection neurons) in the olfactory system are

examples of this type (Ma and Krings, 2007).

The other type of population organization

mapping can be the population of insect individuals

vs. population of wireless sensor nodes, i.e. wireless

sensor network. An insect population that distributes

over habitat space forms an information network.

This network may depend on infochemicals (in

chemosensory system) or vibrations (in audition) as

"packets" communicating via air, water, or other

types of substrate medium. Indeed, the

infochemicals-based wireless communication

network is probably more complex than electron-

based networks. Several categories of infochemicals

are involved, e.g., pheromones are utilized in intra-

specific communications and allelochemicals

(allomones, kairomones, and synomones) in inter-

specific communications (Ma and Krings, 2007).

What may be even more inspiring is that there are

several parallel communications networks—visual,

olfactory, auditory, etc.—in an insect population, or

the so-called multimodality sensory. All of the

sensory networks are wireless except for the taste

sensory network. This is essentially the

demonstration of multiple modalities at the

population level.

In terms of sophistication and functionalities, no

other organisms may match insects in the

chemosensory systems. The differences between the

insect chemosensory wireless network and the

engineered wireless network lie in message

encoding (infochemicals vs. radio frequencies) and

computing node (insect brain vs. microchip). The

research of insect sensory systems may inspire the

engineering of reliable and secure wireless sensor

networks. Obviously, the insect sensory wireless

network is operated under heterogeneous and

unstable natural environments. The network has to

deal with possible exploitations by other species,

which may be their competitors or natural enemies.

For example, the natural enemies may try to find

their prey by following the infochemicals, and the

insects may release interference infochemicals to

confuse their competitors. This is similar to

malicious intrusions in computer networks.

2.2 Insect-Inspired Robots and Micro

Aerial Vehicle (MAV)

The study of the aerodynamics of insect flight was

conducted as early as the 1950s. Grasshoppers and

flies seem to be the most common model insects.

Both walking (including crawling) and flying robots

based on insects have been developed. Insect

sensory systems, mainly vision and

mechanosensory, have offered inspiration for those

INSECT SENSORY SYSTEMS INSPIRED COMMUNICATIONS AND COMPUTING (II): AN ENGINEERING

PERSPECTIVE

293

designs. It can be said that the research of insect-

inspired flight has been the most intensive and

extensive field studied among all insect-related

engineering studies.

Micro Aerial Vehicles (MAV), also known as

Mini Aerial flight Vehicles, have been studied for

over a decade. An MAV is based on UAV

(Unmanned Aerial Vehicle) technology, but there

are significant differences. According to DARPA's

definition, an MAV has a wingspan of less than 15

cm. It turned out that the 15 cm is an interesting

threshold to separate two types of flights: flapping

flight (micro-flight, used by insects) vs. fixed wing

soaring flight (Pornsin-sirirak et al. 2001).

At least seven laboratories started insect-inspired

robots research in approximately the same period

about a decade ago. The Biomimetic Millisystem at

U.C. Berkeley has been developing the so-called

minimally-invasive flying robots, weighing 0.1g,

using insect-inspired wing kinematics (Wood et al.

2005, Steltz 2005). The group at CalTech’s

(Pasadena, CA) Micromachining Lab focused on the

design and manufacturing of flight wings for MAV.

For example, they developed the first MEMS-based

(Micro Electro Mechanic Systems) wing technology

with titanium-alloy metal as wingframe and

parylene-C as wing membranes (Pornsin-sirirak et

al., 2001). The "Entomopter" is a multimode

(flying/crawling) robot designed by the joint team of

Georgia Tech Research Institute (GTRI) and

Cambridge University. The effort has been made to

develop an Entomopter-based Mars surveyor

(Michelson, 2002). The Biorobotic Vision

Laboratory at the Australia National University has

focused on the insect vision-driven behaviors and

their inspiration for machine vision, as well as

visually guided robots (Srinivasan et al., 2001,

2003). Their researchers, in cooperation with the Jet

Propulsion Laboratory at Cal-Tech and NASA, have

developed robots for Mars exploration based on the

study of ocelli of dragonflies. The design of Mars

exploration robots has taken inspiration from the

unique skills of dragonflies in navigation, hazard

avoidance, altitude hold, stable-flight, terrain-

following and smooth deployment of payload

(Thakoor, 2003). The Center for Intelligent

Mechatronics at Vanderbilt University studied

Mesoscale Crawling Robots based on insect model

(Lobontui et al., 1999). CAVIAR is a European

Commission funded project to develop a multi-chip

vision system based on Address-Event

Representation (AER) communication of spike

events (http://www.ini.uzh.ch/~tobi/caviar/). This

system emulated biological visual pathways (

Liu et

al. 2002).

Fly-by-Sight-Microrobots is a project

headed by Nicolas Franceschini in France. His team

developed neuromimetic robots by emulating the

fly's compound eyes (www2.cnrs.fr/en/582.htm).

Besides the previous groups' comprehensive

research projects, quite a few researchers have

conducted relatively ad-hoc studies in the field.

Motamed & Yan (2005) is a review of insect-

inspired micro-flight. Ma and Krings (2007)

reviewed more case studies in insect-inspired robots

and MAV.

2.3 Non-Cooperative Behaviours in

Social Insects — Conflicts

Resolution

Non-cooperative behaviors in social insects are

contrary to the cooperative ones that have inspired

swarm intelligence and similar algorithms, also

referred to as ants colony optimization algorithms.

The reason we single out this type of insect behavior

is an intuitive argument: If the solution for the

opposite side of a problem is inspiring, one may be

able to get the solution by conducting inverse

transformation. This is often true in optimization.

The society of social insects, like any society, is

never a perfect world. The dominant organization of

the insect societies such as bees, ants and termites is

the caste system, and individual rights are often not

fully protected. Two major conflicts exist in social

insects: (1) direct reproduction rights and (2) the

manipulation of fellow colony members. Ratnieks

and Foster et al (2006) reviewed five major

reproductive conflicts in insect societies, including:

(1) sex allocation, (2) queen rearing, (3) male

rearing, (4) queen-worker caste fate, and (5)

breeding conflicts among totipotent adults. These

reproductive conflicts exist widely in the colonies

and sometimes have dramatic effects on the

colonies. Three essential mechanisms: kinship,

coercion, and constraint typically jointly limit the

effects of conflicts and often the reproductive

conflict is resolved totally. The inclusive fitness

theory has been proposed to explain both

cooperation and conflict. Essentially some

individuals of a colony relinquish their direct

reproductive rights to help rear and defend the

offspring of other colony members. A major factor

in conflict resolution is the kinship, since the great

relatedness suppresses the incentive to be selfish.

Whether or not the pheromones, which play crucial

roles in cooperative behaviors, are involved in

conflict resolution is still unknown, and neither are

the genes affecting conflict resolution (Ratnieks and

Foster et al. 2006). There has been no modeling

research of the conflict resolution in social insects.

BIOSIGNALS 2008 - International Conference on Bio-inspired Systems and Signal Processing

294

Whether or not pheromones are involved in the

conflict resolution is really not important for their

potential inspiring in devising new computation

strategies or extending the existing swarm

intelligence. (The latter is based on pheromone-

regulated cooperative behaviors.) We see three

potentially rewarding explorations. (1) Extending

swarm intelligence. In real world populations, both

cooperative and non-cooperative (conflict

resolution) mechanisms exist simultaneously and the

successful resolution of conflict may enhance

cooperation. Therefore, introducing conflict

resolution into swarm intelligence algorithms should

make the algorithms match biological mechanisms

more consistently. Cooperative behavior is

essentially a positive feedback mechanism, and non-

cooperative behavior often acts as the negative

feedback mechanism. A system should become more

stable with both types of feedback regulations.

Certainly, what we suggest here is just a conjecture.

(2) The mechanisms of conflict resolution may be

inspirational for designing survival selection

mechanisms in evolutionary computation, or

extending the existing survival selection schemes.

(3) Mathematical modeling of the conflict resolution

in social insects has not yet been explored. Given the

dominant role of evolutionary game theory in

modeling conflicts resolution in other animals

(Maynard-Smith 1982), it makes great sense to

apply evolutionary game theory first. Obviously, the

studies of (2) and (3) should be compared to inspire

each other, since the topic of (2) is essentially an

evolutionary computation issue and that of (3)

belongs to evolutionary biology.

3 THE OTHER TOPICS

Insect Vision Inspired Motion Detection and

Neuromorphic Engineering.

This topic was

addressed in the first article of this two-part series

(Ma and Krings 2007), since it was more convenient

to discuss it in the context of insect vision sensory

systems. Given its extreme importance, we include

the following brief summary.

One field that has made enormous progress in

recent years is the motion detection of insect eyes

and their applications to bio-inspired robot sensors.

This is one area of neuromorphic engineering.

Parallel and analog are two trademark properties of

insect neural systems. It is now possible to design

and manufacture a fully integrated neuromorphic

olfaction chip (Liu et al. 2002, Stocker 2006,

Koickal et al. 2007). A possible reason for the

advancement is that motion-detection neurons are

some of the largest in insect vision systems and easy

to observe (Rind, 2005). Rind (2005) summarized

three types of contributions where man-made vision

systems are based on insect vision system: (1) Bio-

inspired circuits embedded in the control structure of

mobile robots. Examples are the Lobula Giant

Movement Detector (LGMD) for collision detection

based on locust eyes (Blanchard et al., 2000) and

flying motion detectors. (2) Neuromorphic chips

based on fly eyes (Harrison, 2000) and VLSI retinal

circuits (Liu and Kramer et al., 2002), and (3) Bio-

inspired behavioral strategies (Srinivasan et al

2001). In these insect vision-inspired designs, the

goal has been to make fast, robust, lightweight and

low-power vision systems. Another feature is that

analog-VLSI has been the dominant choice in insect-

vision-based chips. Ruffer and Franceschini (2004)

have designed neuromorphic eyes for a mini-UAV

with eye weights of only 0.8g and a weight of only

100g for the entire rotorcraft. Tests reveal that these

artificial vision chips (even the most flexible analog-

VLSI fly eye) still have significant gaps with real

insect visions systems upon which the chips are

based. This indicates that a better understanding of

insect eye motion detection has to be gained to make

further breakthroughs (Rind, 2005). More recently,

Fife and Archibald (2007) applied FPGA approach

to support real-time vision processing for the small

UAV.

Neural Network Modelling and Dendritic

Neuronal Computing. It is interesting to note that

recent advances in neural biology may change our

thinking about modeling neural networks, perhaps

including the ANN (Artificial Neural Network).

Vogels and Rajan et al. (2005) present an excellent

critical review on neural network dynamics, and

they call for the models that go beyond describing

and adapting to the input-output dynamics. The

mathematical modeling has to address the

fundamental property of the brain, that is, the neural

circuits perpetually generate complex activity

patterns of extraordinarily rich spatial-temporal

structure, yet they remain highly sensitive to sensory

inputs. London and H¨ausser (2005) offered the

perspective from the computation capability of

single neuron, the so-called dendritic computation.

They argue that the computing "tool kit" of dendrites

may play roles well beyond currently acknowledged

properties. Ma and Krings (2007) suggested that the

neuronal population in insect sensory system such as

the visual neurons that support the compound eyes,

the PN (Projection Neurons) in AL (Antennal Lobe),

or the ORN (Olfactory Receptor Neurons may be

emulated to develop a new grid computing

INSECT SENSORY SYSTEMS INSPIRED COMMUNICATIONS AND COMPUTING (II): AN ENGINEERING

PERSPECTIVE

295

architecture. It seems that the integrated model of

grid computing with dendritic computing may offer

new insights. That is, a grid-computing

infrastructure is supported by tool kits from dendritic

nodes.

Biosensoring. A biosensor is an integrated device

that combines a biological component with a

physicochemical detector component that converts a

biological response to specific substances being

monitored into an electrical signal. An annual

review has been published by Rich and Myszka

since 1999 (Rich and Myszka, 2005).

Cellular Computing, Agent-based Computing

and Swarm Intelligence. Amos et al. (2004) and

Dogaru (2003) presented two of the latest review on

cellular computing. Ma and Krings (2007)

contrasted cellular grids in cellular computing vs.

neuron populations in insect sensory systems. The

latter can be the visual neurons that support the

compound eyes, the PN (Projection Neurons) in AL

(Antennal Lobe), or the ORN (Olfactory Receptor

Neurons) from insect sensory organs (sensilla). This

neuronal population can be emulated to develop a

new grid computing architecture.

Agent computing is another field where insect

model plays a significant inspirational role. The

most well- known paradigm should be Swarm

intelligence (Bonabeau et al. 1999, Dorigo and

Stützle 2004), which is inspired by ants pheromone-

modulated cooperative behaviors. This is in contrast

with the non-cooperative behavior we discussed in

subsection 2.3. Lodding’s (2004) biomorphic

software and the design patterns of Babaoglu and

Canright et al. (2006), and Dobson and Massacci's

(2006) are more examples of agent-based adaptive

computing.

Molecular Networks and System Biology. A

biological cell is a complex "network of networks"

from the information processing perspective.

Physiologically, it is an integrated device consisting

of several thousands of types of interacting proteins.

Molecular network is often used as a generic term to

refer to the networks involved in cell biology (Alon,

2007). The gene regulatory networks and various

molecular networks in cells involved extremely

complex yet robust networks, which is one of the

focuses of the newly emerged system biology. There

are enormous opportunities for computer scientists

to contribute and to learn from the fields. The

following are a few review references: on genome

project by Ideker and Galitski et al. 2001, the bio-

mimetic nano-scale reactors and networks by

Karlsson et al. (2004), molecular networks by

Galitski (2004), gene regulatory networks by

Davidson (2006).

4 PERSPECTIVE

In the following, we mention some promising

research topics that seem not yet being explored. In

various previous sections and Ma and Krings (2007),

we briefly discussed them in corresponding context;

the following is simply a list of summary statements.

(1) The new grid computing architecture that

emulates neuronal population such as vision neurons

for insect compound eyes. This neuronal population

architecture may be further integrated and enhanced

with dendritic computing (2) Wireless sensor

networks that emulate the insect chemosensory

networks and the multimodal architecture that has

several parallel networks concurrently in operation

(such as audition, vision chemosensory, etc.). In

addition, the bio-robustness mechanism in these

insect sensory networks should be captured. (3) The

implications of non-cooperative behaviors in social

insects to swarm intelligence, evolutionary

computing, and to devising new optimization

algorithms. (4) Insect audition, which was

considered as less developed in insects until

recently, is recognized now as underestimated in

entomology (Drosopoulos and Claridge, 2006). Still

the field of insect audition has received little

attention from the bio-inspired perspective. It is

interesting to note that insects audition truly

resembles the engineered wireless communications.

(5) The integration of technologies that are

developed for sensors, robots, MAVs, and

neuromorphic technologies, in particular, the multi-

modality integration may provide better solutions for

the sophisticated MAV flights control, especially in

unstable and hostile military operations

environments.

In the recent report on "Computing and Biology"

from the US National Academy of Sciences

(National Research Council, 2005), the ants colony

optimization and neural-inspired sensors, together

with hundreds of other research topics, were

recommended as fields of strategic scientific and

technological significances. However, the majority

of topics on insect sensory systems, such as those

discussed in this series of articles, were omitted.

There was significant coverage (nearly 4 pages) on

ants colony optimization in the National Academies'

report. This coverage may also indicate the

significance of the areas omitted in the report,

which, in our opinion, should prove to be as

promising as the ants colony optimizations, if not

BIOSIGNALS 2008 - International Conference on Bio-inspired Systems and Signal Processing

296

more.

ACKNOWLEDGEMENTS

This research was partially supported by a UAV and

Ad hoc Networking research grant from the US DOE

INL. We wish to thank the two anonymous

reviewers for their insightful comments.

REFERENCES

Alon, U. 2007. An Introduction to Systems Biology:

Design Principles of Biological Circuits. Chapman.

Amos, M. 2004 (ed.). Cellular Computing. Oxford

University Press. 217pp.

Babaoglu, O. and G. Canright et al. 2006. ACM Trans. on

Auton. & Adap. Syst. 1(1):26–66.

Blanchard, M. and F. C. Rind, P. F. M. Vershure. 2000. J.

Robot. Auto. Syst. 30:16-38.

Bonabeau, Eric. M. Dorigo, G. Theraulaz. 1999. Swarm

Intelligence. Oxford University Press. 320pp.

Christensen, T. A. (ed). 2005. Methods in Insect Sensory

Neuroscience. CRC Press. 435pp.

Davidson, E. H. 2006. The regulatory genome: gene

regulatory networks in development and evolution.

Academic Press. 304pp.

Detrain, C., J. L. Deneubourg, and J. M. Pasteels.(eds).

1999. Information Processing in Social Insects.

Birkhauser Verlag. 415pp.

Dobson, S., F. Massacci, et al. 2006. ACM Transactions

on Autonomous and Adaptive Systems, 1(2): 223–259

Dogaru, R. 2003. Universality and Emergent Computation

in Cellular Neural Networks. World Scientific Press.

Dorigo, M., T. Stützle. 2004. Ants Colony Optimization.

The MIT Press.

Fife, W. S. and J. K. Archibald. 2007. EURASIP J. on

Embedded Systems. (1) 33

Dressler, F. I. Carreras. 2007. Advances in Biologically

Inspired Information Systems: Models, Methods, and

Tools. Springer, 302pp.

Drosopoulos, S. and M. F. Claridge. (eds). 2006. Insect

Sounds and Communication. Taylor & Francis. 532pp.

Galitski, T. 2004. Molecular networks in model systems.

Annu. Rev. Genomics Hum. Genet. 2004. 5:177–87

Gullan, P. J. and P. S. Cranston. 2005. The insects: an

outline of entomology. 3rd Ed. Blackwell Publishing.

Harrison, R. R. 2000. An analog VLSI motion sensor

based on the fly visual system. Ph.D. Dissertation,

CalTech, CA.

Ideker, T., T. Galitski1, and L. Hood. 2001. Annu. Rev.

Genomics Hum. Genet. 2001. 2:343–72

Karlsson, M., M. Davidson et al. 2004. Annu. Rev. Phys.

Chem. 2004. 55:613–49

Koickal, T. J., A. Hamilton, S. L. Tan, J. A. et al. 2007.

IEEE Trans. Circuits and Systems.54(1): 60-73.

Klowden, M. J. 2007. Physiological Systems in Insects.

2nd ed. Academic Press.

Liu, S. C., J. Kramer, G. Indiveri, T. Delbrück, R.

Douglas2002. Analog VLSI: Circuits and Principles.

MIT Press. 472pp.

Lobontui, N, Goldfarb, M., et al. 1999. Design and

Analysis of an Elastodynamic Inch-Worm Robotic

Insect. IEEE Conference on Robotics and Automation,

pp. 2120-2125, May 1999.

Lodding, K. N. 2004. Queue, June 2004:68-75.

London, M. and M. H¨ausser 2005. Dendritic

Computation. Annu. Rev. Neurosci. 2005. 28:503–32

Michelson, R.2002. The Entomopter. "Neurotechnology

for Biomimetic Robots", The MIT Press, pp.481-50

Ma, Z., and A. W. Krings. 2007. Insect Sensory Systems

Inspired Computing and Communications. Technical

Reports, TR-CS-09-01-2007, Computer Science Dept.

University of Idaho.

Maynard Smith, J. 1982. Evolution and the Theory of

Games. Cambridge University Press.

Miesenbock, G. and I. G. Kevrekidis. 2005. Annu. Rev.

Neurosci. 2005. 28:533–63

Motamed, M., and J. Yan. 2005. A review of biological,

biomimetic and miniature force sensing for

microflight. International Conference on Intelligent

Robots and Systems (IROS). 2-6 Page(s): 3939 - 3946.

National Research Council of National Academies. 2005.

Catalyzing Inquiry at the Interface of Computing and

Biology. The National Academy Press.

Pornsin-sirirak, T. N. Y. C. Taia, et al. 2001. Sensors and

Actuators A 89. (2001): 95-103

Ratnieks, F. L. W., K. R. Foster, and T. Wenseleers.

2006. Conflict resolution in insect societies. Annu.

Rev. Entomol. 2006. 51:581–608

Ruffer, F. and N. Franceschini, 2004. "Visually guided

micro-aerial vehicle: automatic take off, terrain

following, landing and wind reaction," in

Proceedings of ICRA 2004), vol. 3, pp. 2339-2346.

Rich, R. L. and D. G. Myszka. 2005. J. Mol. Recognit. 18:

431–478

Rind, F. C. 2005. Bioinspired sensors. in. "Methods in

Insect Sensory Neuroscience". ed. by T. A.

Christensen. CRC Press. pp. 213-235.

Srinivasan. 2003. Journal of Robotic Systems 20(1), 35–

Srinivasan, M. V., S. W. Zhang. et al. 2001. Biol. Bull.

200:216-221.

Steltz, E., S. Avadhanula, R.J. Wood, and R. S. Fearing.

2005.

Characterization of the Micromechanical Flying

Insect by Optical Position Sensing,

IEEE Int. Conf. on

Robotics and Automation, Barcelona, April 2005.

Stocker, A. A. 2006. Analog VLSI Circuits for the

Perception of Visual Motion. Wiley.

Thakoor, S., and N. Cabrol, et al. 2003. J. of Rob. Syst.

20(12), 687–706.

Vogels, T. P., K. Rajan, and L. F. Abbott. 2005. Neural

Network Dynamics. Annu. Rev. Neurosci. 28:357–76

Wood, R. J., S. Avadhanula, E. Steltz, M. Seeman, et. al.

2005. Design, Fabrication and Initial Results of a 2g

Autonomous Glider. The 31st Annu. Conf., IEEE

Indust. Electron. Soc., Raleigh North Carolina.

INSECT SENSORY SYSTEMS INSPIRED COMMUNICATIONS AND COMPUTING (II): AN ENGINEERING

PERSPECTIVE

297