Intelligent behavior generation architecture for autonomous wheeled mobile robots

Wheeled Mobile Robots (WMRs) have been an active area of research and development for many years. Improvising on the way a mobile robot acts (termed as behavior) in its dynamic environment has been one of the interesting spheres in the fields of robotics and artificial intelligence. Most proposed architectures, in this sphere, rely on the information that is gathered by the robot from its environment. This is usually achieved by piling the robot with sensors like SONAR transducers, scanning Lasers, Cameras, and Infrareds that perceive its environment and collect relevant information. The gathered information is then modeled to produce knowledge that could influence the robot's future decisions. Intelligent decisions make the system autonomous. Such systems will accept high-level descriptions of tasks and will execute them without further human intervention. The input descriptions will specify what the user wants done rather than how to do it.; Autonomy cannot be achieved by making a single module in the system ingenious. All the modules of the robot's decision-making body need to play equal roles for true autonomy. The coordination between modules has to be supervised and controlled by a higher central authority which manages and synchronizes the retrieval and updates of the shared knowledge. The algorithms that make the intelligent decisions constitute the behavior generation module of the robot's architecture. Execution of behaviors is triggered either by the controller (a human or another machine) or by the events happening in the robot's environment.; This thesis presents an architecture for autonomous wheeled mobile robots that, incorporates the capabilities of making intelligent decisions and reliably executing them. The architecture is implemented for a class of omni-directional autonomous WMRS developed at Utah State University (USU).