Why did I fail? A Causal-based Method to Find Explanations for Robot Failures

TL;DR

This paper introduces a causal-based method for robots to generate explanations for failures by learning a causal model from simulation data and identifying the closest successful state to the failure, enhancing transparency.

Contribution

The paper presents a novel approach combining causal Bayesian networks and contrastive explanations for robot failure analysis, scalable across multiple tasks.

Findings

Causal models achieved 70-72% accuracy in sim2real scenarios.

The method enables robots to generate human-understandable failure explanations.

Scalable to multiple tasks and applicable to real robots.

Abstract

Robot failures in human-centered environments are inevitable. Therefore, the ability of robots to explain such failures is paramount for interacting with humans to increase trust and transparency. To achieve this skill, the main challenges addressed in this paper are I) acquiring enough data to learn a cause-effect model of the environment and II) generating causal explanations based on that model. We address I) by learning a causal Bayesian network from simulation data. Concerning II), we propose a novel method that enables robots to generate contrastive explanations upon task failures. The explanation is based on setting the failure state in contrast with the closest state that would have allowed for a successful execution. This state is found through breadth-first search and is based on success predictions from the learned causal model. We assessed our method in two different scenarios I) stacking cubes and II) dropping spheres into a container. The obtained causal models reach a sim2real accuracy of 70% and 72%, respectively. We finally show that our novel method scales over multiple tasks and allows real robots to give failure explanations like 'the upper cube was stacked too high and too far to the right of the lower cube.'