Term of Award

Spring 2013

Degree Name

Master of Science in Applied Engineering (M.S.A.E.)

Document Type and Release Option

Thesis (restricted to Georgia Southern)

Copyright Statement / License for Reuse

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.


Department of Mathematical Sciences

Committee Chair

Biswanath Samanta

Committee Member 1

Jordan Shropshire

Committee Member 2

Yong Zhu


The neuromodulatory systems in the brain of a vertebrate help regulate its decision making in response to the sensory signals from the environment. These neuromodulators influence a vertebrate’s behaviors like focusing attention, cautious risk-aversion and curiosity-seeking exploration. The main objective of this thesis was to study and implement a neuronal model inspired by the neuromodulatory systems on a relatively simple autonomous robot demonstrating its ability to adapt to changes in the environment. This work presents a control approach based on vertebrate neuromodulation and its implementation on an iRobot Create platform. Three of the built-in sensors, namely, bump, dock beam, and battery state of Create were used in this study. In addition to these, five ultrasonic ping sensors were mounted in the front of the robot for measuring the robot distance from a nearby object like obstacle or wall. The data from ping sensors were transmitted via an Arduino board to Matlab workspace of a laptop through serial interface. The neural network incorporated three types of neurons- cholinergic and noradrenergic (ACh/NE) neurons for attention focusing and action selection, dopaminergic (DA) neurons for curiosity-seeking, and serotonergic (5-HT) neurons for risk aversion behavior. The robot was run through Matlab under three different behavioral modes, namely, risk aversive, risk-taking, and distracted, all inside the clutter environment of a lab space. However, unlike the traditional neural networks, the neuromodulatory model did not need any pre-training. Instead, the system learnt from its sensory inputs and followed the behavioral facets of a living organism. The robot showed very interesting behavioral patterns which were analyzed and fully justified. The details of algorithm development, the experimental setup and implementation results under different conditions are presented along with a discussion on the scope of further work.

Research Data and Supplementary Material