Robotics, Temporal Logic and Stream Reasoning

The area of AI robotics offers a set of fundamentally challenging problems when attempting to integrate logical reasoning functionality in such systems. The problems arise in part from the high degree of complexity in such architectures which include realtime behaviour, distribution, concurrency, various data latencies in operation and several levels of abstraction. For logic to work practically in such systems, traditional theorem proving, although important, is often not feasible for many of the functions of reasoning in such systems. In this article, we present a number of novel approaches to such reasoning functionality based on the use of temporal logic. The functionalities covered include automated planning, stream-based reasoning and execution monitoring.