Generating Explanations of Device Behavior Using Compositional Modeling and Causal Ordering

Generating explanations of device behavior is a long-standing goal of AI research in reasoning about physical systems . Much of the relevant work has concentrated on new methods for modeling and simulation, such as qualitative physics, or on sophisticated natural language generation, in which the device models are specially crafted for explanatory purposes . We show how two techniques from the modeling research-compositional modeling and causal ordering-can be effectively combined to generate natural language explanations of device behavior from engineering models. The explanations offer three advances over the data displays produced by conventional simulation software: (1) causal interpretations of the data, (2) summaries at appropriate levels of abstraction (physical mechanisms and component operating modes), and (3) query-driven, natural language summaries . Furthermore, combining the compositional modeling and causal ordering techniques allows models that are more scalable and less brittle than models designed solely for explanation . However, these techniques produce models with detail that can be distracting in explanations and would be removed in hand-crafted models (e .g ., intermediate variables) . We present domain-independent filtering and aggregation techniques that overcome these problems . 1 . Introduction Generating Explanations of Device Behavior Using Compositional Modeling and Causal Ordering This paper presents a method for generating explanations of device behavior characterized by systems of mathematical constraints over continuous-valued quantities . Such models are widely used in engineering for dynamical systems, such as electromechanical and thermodynamic control systems . Given such a model and initial conditions, conventional simulation software can predict and plot the values of these quantities over time . However, the data can be difficult to interpret because conventional simulators do not explain how the predicted behavior arises from the structure of the modeled system and physical laws . What we call explanations are presentations of information about the modeled system that satisfy three requirements . First, an explanation offers a meaningful interpretation of the simulation data, explaining how and why and not just what happened . For engineering tasks Funding was provided by NASA Grant NCC2-537 and NASA Grant NAG 2-581 (under ARPA Order 6822) . Patrice O. Gautier and Thomas R. Gruber Knowledge Systems Laboratory