RACE (Robustness by Autonomous Competence Enhancement) is a project funded by the European Commission’s 7th Framework Programme under contract/grant number 287752. It started on Dec 1, 2011, running for three years. The Commission’s support is gratefully acknowledged.
The overall aim of this project is to develop an artificial cognitive system, embodied by a service robot, able to build a high-level understanding of the world it inhabits by storing and exploiting appropriate memories of its experiences. For further details on the entire project, please see the the official project page and the project website for RACE of the European Commission.
Within the project consortium, the Knowledge-Based Systems Group of the University of Osnabrück is dealing mainly with the topic of planning and executing of robot activities by exploiting acquired competences. This involves the following parts:
Generating expectations about the robot’s actions.
Methods for calculating expectations about results of actions executed by the robot are developed. The calculation will be based on the available set of action representations, on the most recent situation interpretation, on predictions of independent events, on the reasoning services and on available conceptualisations.
Guiding attention based on predictions and expectations.
Methods for exploiting predictions and expectations as parts in an agent control that serve for focusing attention on relevant sensor data or reasoning steps. This helps the agent to act in real time with finite resources.
The execution of tasks will be monitored to determine whether some task still runs within its nominal conditions or whether it is done with or without success. This will be based on the situation model, a HTN plan and the expectations. In case of a detected failure, possible alternative high-level task expansions or alternative objects are used. This approach will be extended by exploiting a semantic environment representation.
Semantic temporal planning and execution
Planning and temporal reasoning methods, formalisms and implementations will evaluated and possibly modified, changed, or re-implemented.
Additionally, KBS takes part in the development of tools for hybrid knowledge representation and reasoning capabilities.