Review of Artificial Intelligence and Mobil Robotics: Case Studies of Successful Robot Systems

Article excerpt

In the 1970s, a schism developed between (symbolic) Al and robotics (including computer vision).

Today, mobile robotics is an increasingly important bridge between the two areas. It is advancing the theory and practice of cooperative cognition, perception, and action and serving to reunite planning techniques with sensing and real-world performance. Further, developments in mobile robotics can have important practical economic and military consequences.

For some time now, amateurs, hobbyists, students, and researchers have had access to how-to books on the low-level mechanical and electronic aspects of mobile-robot construction (Everett 1995; McComb 1987). The famous Massachusetts Institute of Technology (MIT) 6.270 robot-building course has contributed course notes and hardware kits that are now available commercially and in the form of an influential book (Jones 1998; Jones and Flynn 1993). The MIT-inspired materials concentrate mainly on Rodney Brooks's subsumption architecture. This approach to robot control relies on simple reactive behaviors, notionally running in parallel, with only simple inhibitory connections between them. Actions are determined by the sensed state of the external world, but internal representation of the world is held to be unnecessary. With these tenets, it is possible to produce robots with reliable, robust, compelling, but arguably limited, repertoires of behavior.

Indeed, pure subsumption robots have not yet demonstrated the ability to perform complex real tasks. Practical autonomous systems tend to use more complex, multilevel organizations of robot capabilities. Common components are a deliberative, strategic, or cognitive level (in the domain of symbolic Al); a tactical actionsequencing level; and a real-time control level (which often can be implemented as reactive behaviors).

Papers in the robotics literature often concern specific technical aspects of robot research and development. At the same time, several robot competitions (run by the American Association for Artificial Intelligence [AAAI], the International Joint Conferences on Artificial Intelligence [IJCAI], and the Department of Transportation [DOT], for example) have emphasized integrated, practical approaches to increasingly general versions of real-world problems such as driving, inspection, and environmental cleanup. However, with a few exceptions (for example, Dorigo and Colombetti 1998), it is rare to find thorough, holistic descriptions of the techniques and software organization used in these practical robots.

Enter Artificial Intelligence and Mobile Robots: Case Studies of Successful Robot Systems, which, as the title says, is about AI algorithms, representations, and software for perception, planning, and robot control. As the subtitle says, the chapters often describe practical, successful, and competition-winning robots. The book concentrates on the organization of high-level functions in mobile robots, not mechanical or electronic design: Most of the robots mentioned are commercial products, and robot construction is not usually addressed. The book is likewise not concerned with traditional control theory for low-level systems such as direction or speed control. Its strength is its description of software, which is at a level of detail and breadth that one does not find in conference papers or usually in the archival literature. Coupled with an existing text on mobile robot construction and programming, this book would be an excellent primary or supplementary text for the next classroom stage beyond a successful subsumption robot. It is also a source book for professionals who must build software architectures for robot perception and control because it contains detailed, current best-practice solutions from proven authorities. The book begins with an editorial introduction and intellectual history of robotics, a brief description of robot competitions, and an overview of the volume. …