Using Abstraction for Interpretable Robot Programs in Stochastic Domains

TL;DR

This paper introduces an abstraction-based approach to simplify the interpretation of robot programs operating in stochastic environments by mapping high-level deterministic models to detailed stochastic models, enhancing understandability.

Contribution

It proposes a method to use high-level nonstochastic models as abstractions for complex stochastic robot programs, reducing complexity and improving interpretability.

Findings

Simplified robot programs with fewer loops and belief operators.

Shorter and clearer action traces in the high-level models.

Enhanced understandability of stochastic robot behaviors.

Abstract

A robot's actions are inherently stochastic, as its sensors are noisy and its actions do not always have the intended effects. For this reason, the agent language Golog has been extended to models with degrees of belief and stochastic actions. While this allows more precise robot models, the resulting programs are much harder to comprehend, because they need to deal with the noise, e.g., by looping until some desired state has been reached with certainty, and because the resulting action traces consist of a large number of actions cluttered with sensor noise. To alleviate these issues, we propose to use abstraction. We define a high-level and nonstochastic model of the robot and then map the high-level model into the lower-level stochastic model. The resulting programs are much easier to understand, often do not require belief operators or loops, and produce much shorter action traces.