Temporal Reasoning: the LaTeR project (LAyered TEmporal Reasoning)

SCIENTIFIC CONTEXT:

Artificial Intelligence,

Constraint Satisfaction Problems

Temporal Reasoning

Temporal Databases

PARTICIPANTS:

Vittorio Brusoni

Luca Console

Paolo Terenziani

COOPERATIONS

Politecnico di Milano, Dip. di Elettronica on Temporal databases

MAIN GOALS:

Analysis of the computational complexity of temporal reasoning

Development of an efficient temporal reasoner

Loosely coupled approach to temporal reasoning: temporal reasoner as an agent that can cooperate with other systems (problem solvers)

Use of the temporal reasoner in other tasks

Diagnosis (see MBD project)

Extension of relational databases to deal with different types of temporal information

METHODOLOGY:

Use of constraint frameworks based on linear inequalities (specifically ˆbounds on differences‰)

Analysis of complexity results for such constraint frameworks

New approach for the complexity of querying such constraint frameworks and checking whether new constraints are consistent with existing ones

Development of a general purpose temporal reasoner;

Two-layered architecture

• low level based on constraint framework
• high level interface language

Loosely coupled integration of the temporal reasoner with other problem solving tasks

• temporal relational databases
• model-based diagnosis (see MBD project)

MAIN RESULTS

New complexity results for a class of constraint frameworks based on linear inequalities

Development of a system for efficient temporal reasoning

• high level interface language (manipulation and query)
• efficient query processing (complexity independent of the dimension of the knowledge base)

Extension of the relational database model to deal with time

• time referent associated with each tuple
• various type of temporal information on tuples (temporal location, duration, delays, qualitative relations)
• temporal relational algebra

A new approach temporal model-based diagnosis (see MBD project)

PROTOTYPES AND SYSTEMS:

LaTeR: A general purpose temporal reasoner

LDL - LQL Extensions of Oracle to create and query temporal relational databases

APPLICATIONS

Temporal relational databases

Model-based diagnosis (see MBD project)

RELATIONS WITH OTHER PROJECTS

MBD on temporal model-based diagnosis (see MBD project)

REFERENCES

V. Brusoni, L. Console, P. Terenziani: On the computational complexity of querying bounds on differences constraints, Artificial Intelligence, vol. 74, no. 2, pp. 367-379, 1995

V. Brusoni, L. Console, B. Pernici, P. Terenziani: LaTeR, a general purpose manager of temporal information, LNCS 869 Springer Verlag 1994

V. Brusoni, L. Console, B. Pernici, P. Terenziani: Extending Temporal Relational Databases to Deal with Imprecise and Qualitative Temporal Information; Proc. International Workshop on temporal databases, Zurich, September 1995.

L. Console, B. Pernici, P. Terenziani: A general temporal manager for temporal databases, Proc. Int. Works on an infrastructure for temporal databases, (R. Snodgrass ed.), 1993