Deployment of Heterogeneous Swarm Robotic Agents Using a Task-Oriented Utility-Based Algorithm

In a swarm robotic system, the desired collective behavior emerges from local decisions made by robots, themselves, according to their environment. Swarm robotics is an emerging area that has attracted many researchers over the last few years. It has been proven that a single robot with multiple capabilities cannot complete an intended job within the same time frame as that of multiple robotic agents. A swarm of robots, each one with its own capabilities, are more flexible, robust, and cost-effective than an individual robot. As a result of a comprehensive investigation of the current state of swarm robotic research, this dissertation demonstrates how current swarm deployment systems lack the ability to coordinate heterogeneous robotic agents. Moreover, this dissertation's objective shall define the starting point of potential algorithms that lead to the development of a new software environment interface. This interface will assign a set of collaborative tasks to the swarm system without being concerned about the underlying hardware of the heterogeneous robotic agents. The ultimate goal of this research is to develop a task-oriented software application that facilitates the rapid deployment of multiple robotic agents. The task solutions are created at run-time, and executed by the agents in a centralized or decentralized fashion. Tasks are fractioned into smaller sub-tasks which are, then, assigned to the optimal number of robots using a novel Robot Utility Based Task Assignment (RUTA) algorithm. The system deploys these robots using it's application program interfaces (API's) and uploads programs that are integrated with a small routine code. The embedded routine allows robots to configure solutions when the decentralized approach is adopted. In addition, the proposed application also offers customization of robotic platforms by simply defining the available sensing and actuation devices. Another objective of the system is to improve code and component reusability to reduce efforts in deploying tasks to swarm robotic agents. Usage of the proposed framework prevents the need to redesign or rewrite programs should any changes take place in the robot's platform