| The primary contribution of this research is the design, implementation, and evaluation of a Cognitive Resource Management Framework for Autonomous Ground Vehicle Sensing; a novel and formalized approach for modeling and managing a distributed collection of heterogeneous resources to provide new levels of autonomous capabilities management, currently unaddressed by unmanned vehicle systems standards. The thesis behind this research is that a well-formulated framework providing real-time modeling, appraisal, and brokering functionalities utilizing abstract representations of mission, jobs, and capabilities provides new levels of intelligent resource utilization to the autonomous ground vehicle community.;This research was performed using the resources of the University of Florida's Center for Intelligent Machines and Robotics. This environment provided a real world setting that enabled the creation of a resource management framework designed to address the challenges of resource utilization on high-capability autonomous ground vehicle systems. The experimental software and test plans, developed to validate the thesis of this work, would ultimately culminate into the framework Reference Implementation.;The Cognitive Resource Management Framework has been shown to be both a viable method for modeling and managing a distributed collection of heterogeneous resources and resource-requests and a valuable contribution to the researchers whose algorithms depend on the reliable availability of resources to fulfill mission operating requirements. The framework's viability and value were demonstrated in both the Reference Implementation and during CIMAR's Environmental Mapping and Change Detection technology demonstration for the Air Force Research Laboratory.;This research presents a new approach to intelligent resource management onboard autonomous ground vehicles. The framework formalizes the discovery, modeling, monitoring and configuration of AGV resources. This work provides new levels of autonomy not currently supported by predominant unmanned systems technology standards. This research provides the foundation upon which new sensing, planning, and scheduling methodologies pertaining to intelligent resource utilization can be implemented and evaluated with the hope of promoting the advancement of highly autonomous platforms capable of sophisticated interactions with unknown, dynamic environments. |
