One of the goals of research in intelligent systems is the development of autonomous mobile robots that can perform various tasks in industrial, military, household and hazardous environments. In industrial applications, we need flexible robots that can perform various tasks at different locations. In military applications, autonomous land and sea vehicles are being developed. In our home environment, an intelligent robot can do many housework for us, including mowing the lawn and cleaning the carpet. In a hazardous environment, an autonomous mobile robot will be able to perform jobs that human should not perform, such as management of nuclear and fire hazards.
Two basic problems in autonomous mobile robots research are planning and navigation. "Planning and Reasoning for Autonomous Vehicle Control", by Joseph S. B. Mitchell, David W. Payton and David M. Keirsey, deals with a reasoning system that supports the planning and control requirements of an autonomous land vehicle. This article provides a very nice description of a part of the "Autonomous Land Vehicle" project of US Army. This system is designed to handle diverse terrain with maximal speed, efficacy and versatility. The hierarchical architecture is presented along with the detailed algorithms, heuristics, and planning methodologies for the component modules. The article of Mitchell, Payton and Keirsey is also concerned with navigation in the second part of this rather long document.
A. Meystel's "Theoretical Foundations of Planning and Navigation for Autonomous Robots" is based on methods in the team theory of decentralized stochastic control that are applicable to autonomous mobile robots. A class of autonomous control systems is defined and a problem of information representation is investigated for this class. Meystel and his colleagues, including G.N. Saridis, are developing the field of intelligent control which lies at the intersection of control theory, artificial intelligence, and operations research. An important concept of Meystel is hierarchical nested knowledge-based controller.
The navigation and control of autonomous mobile robots require a real-time perceptual feedback system that receives sensory data from the environment. Such a sensing system may consist of visual, sonar, radar, thermal and other sensors. Usually, a heterogeneous sensor array provides different sensory data for