The Design of an Immunity-Based Search and Rescue System for Humanitarian Logistics

This dissertation, "The Design of an Immunity-based Search and Rescue System for Humanitarian Logistics" by W Y, Albert, Ko, 高永賢, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of thesis entitled "The Design of an Immunity-based Search and Rescue System for Humanitarian Logistics" Submitted by Albert Wing-yin Ko for the Degree of Doctor of Philosophy at The University of Hong Kong on July 2006 The increasing number of armed conflicts and large-scales of natural disasters in various parts of the world in recent years have challenged the competency of traditional logistics systems in emergency relief operations. These challenges have revealed deficiencies in traditional logistics systems and have obliged the humanitarian relief sector to redefine and re-establish a logistics system that can meet the needs of humanitarian relief operations. This thesis focuses on the search and rescue aspect of humanitarian logistics systems. The scope of humanitarian logistics is defined in a wider context of logistics engineering. A review of traditional search and rescue equipment plus existing search and rescue robots is presented to illustrate how robots might develop in the future for humanitarian logistics systems. This thesis proposes the use of modular robots as the next generation search and rescue robots and has developed an immunity-based framework for controlling this class of robots. Modular robots are module-based systems that can be configured to accommodate different operation requirements. The ability to change shape dynamically during an operation enables modular robots to navigate across larger gaps, creep through smaller voids and to climb over higher obstacles than conventional robots. The inherent decentralized nature of modular robots requires a flexible decentralized control system that can adapt to topological changes to take full advantage of modular robots' versatility. Immunity-based control systems inspired by the biological immune system exhibit essential characteristics for controlling modular robots. The decentralized nature plus the 6robustness and adaptability of an immunity-based control system enable modular robots to operate effectively in several operation modes. The operation modes of modular robots are defined into four different classes. They are Homogeneous-connected, Homogeneous-distributed, Heterogeneous-connected, and Heterogeneous-distributed. Homogeneous modular systems contain only one type of modules, whereas heterogeneous modular systems contain two or more types of modules. Connected modular systems are multiple modules operating together as a single robot. Distributed modular systems are collections of disconnected modules navigating on their own in the same mission. Modular robots can change between the four operation modes under specific conditions. Four experiments using computer simulation and physical prototype are used to demonstrate the functioning of the immunity-based framework in each of the four operation modes. The Homogeneous-connected modular system uses a planar modular manipulator constructed in MATLAB to demonstrate the performance of the control framework. A physical prototype is also developed to verify the experiment. The Homogeneous-distributed system uses a fireguard simulation constructed in NetLogo(R) to demonstrate the ability of the control system to control large scaled distributed systems using simple stimulation signals. The Heterogeneous-connected modular system uses a self-balancing robot to show that stand-alone robots can be controlled as a heterogeneous modular system. The Heterog