Research Challenges in Intelligent Robotic Systems: From Evolutionary Methods to KASER's and KANSEI

摘要

Research in the area of robotics continues to be a rich and exciting field. Further, the need for robotic systems with more capabilities for the growing number of complex applications drive this research. For example, autonomous intelligent robot colonies may be used in reconnaissance missions or seek-and-capture scenarios involving a complex set of interactions between machines as well as between machines and humans and may cover long distances to remote sites. Because of the nature of the tasks, new classes of robotic systems will be required that have a high level of specification for efficiency and reliability. This, we believe, can only be accomplished through sophisticated intelligent control and efficient sensor integration as an integral part of the design of the robot and the robot's supporting systems. In this seminar, a brief historical perspective of robotic systems, particularly some projects developed by the author, will be presented. Some key research issues such as parametric versus non-parametric system models as well as types of controllers are discussed. As robotic systems evolve to more sophisticated architectures to address the needs for various applications, the requirements for enhanced intelligence as well as integration of larger sets of different sensors also grows. We will discuss the characteristics of intelligence and offer some approaches to implement this feature, including some evolutionary methods developed by the author, the Knowledge Amplification by Structural Expert Randomization (KASER) and KANSEI Engineering. Finally, we will consider some applications which offer challenges that continue to drive some of the research in the exciting area of intelligent robotic systems.