This book derives techniques which allow reliable plans to be automatically selected by intelligent machines. It concentrates on the uncertainty analysis of candidate plans so that a highly reliable candidate may be identified and used. For robotic components, such as a particular vision algorithm for pose estimation or a joint controller, methods are explained for directly calculating the reliability. However, these methods become excessively complex when several components are used together to complete a plan. Consequently, entropy minimization techniques are used to estimate which complex...

This work presents the result of 30 years' work on the original material related to "thinking machines", a subject initiated by the author and his colleagues. It is based on the ability of the computer to represent the hierarchical procedure of task conception and execution found in human beings. It is arranged in three levels representing the structure of organizational systems: organization, co-ordination and execution. The volume can serve as a guide to modern intelligent robots.