The Execution Monitoring Project
Now and then, things will go wrong. This is both a fact of life and a fundamental problem in any robotic system intended to operate autonomously or semi-autonomously in the real world. Like humans, robotic systems (or more likely their designers) must be able to take this into account and recover from failures, regardless of whether those failures result from mechanical problems, incorrect assumptions about the world, or interference from other agents.
We have therefore developed a temporal logic-based task planning and execution monitoring framework and integrated it into our fully deployed rotor-based unmanned aircraft system. In the spirit of cognitive robotics, we make specific use of Temporal Action Logic (TAL), a logic for reasoning about action and change. This logic has already been used as the semantic basis for TALplanner, which is used to generate mission plans that are carried out by an execution subsystem. Building further on the use of TAL, knowledge gathered from the appropriate sensors during plan execution can be used by the knowledge processing middleware framework DyKnow to create state structures, incrementally building a partial logical model representing the actual development of the system and its environment over time. Formulas in TAL can be used to specify the desired behavior of the system and its environment over time and violations of such formulas can be detected in a timely manner using a formula progression algorithm.
The pervasive use of logic throughout the higher level deliberative layers of the system architecture provides a solid shared declarative semantics that facilitates the transfer of knowledge between different modules. Given a plan specified in TAL, for example, it is possible to automatically extract certain necessary conditions that should be monitored during execution.
|||Patrick Doherty, Jonas Kvarnström, and Fredrik Heintz. A Temporal Logic-based Planning and Execution Monitoring Framework for Unmanned Aircraft Systems. Journal of Autonomous Agents and Multi-Agent Systems, 2009. Published and available online February 2009; print publication forthcoming. [ Journal ]|
|||Jonas Kvarnström, Fredrik Heintz, and Patrick Doherty. A Temporal Logic-Based Planning and Execution Monitoring System. In Jussi Rintanen, Bernhard Nebel, J. Christopher Beck, and Eric Hansen, editors, Proceedings of the 18th International Conference on Automated Planning and Scheduling (ICAPS), Sydney, Australia, September 2008. [ Conference | .pdf ]|
Page responsible: Patrick Doherty
Last updated: 2014-04-30