The relevance of causation in robotics: A review, categorization, and analysis

Thomas Hellström

doi:10.1515/pjbr-2021-0017

Abstract

Abstract In this article, we investigate the role of causal reasoning in robotics research. Inspired by a categorization of human causal cognition, we propose a categorization of robot causal cognition. For each category, we identify related earlier work in robotics and also connect to research in other sciences. While the proposed categories mainly cover the sense–plan–act level of robotics, we also identify a number of higher-level aspects and areas of robotics research where causation plays an important role, for example, understandability, machine ethics, and robotics research methodology. Overall, we conclude that causation underlies several problem formulations in robotics, but it is still surprisingly absent in published research, in particular when it comes to explicit mentioning and using of causal concepts and terms. We discuss the reasons for, and consequences of, this and hope that this article clarifies the broad and deep connections between causal reasoning and robotics and also by pointing at the close connections to other research areas. At best, this will also contribute to a “causal revolution” in robotics.

Highlights

Causal reasoning is generally considered to be a crucial human cognitive competence necessary for a range of tasks such as predictions, diagnoses, categorization, action planning, decision-making, and problem solving [1]
Inspired by a categorization of human causal cognition [5], we propose a categorization of robot causal cognition
Randomized Controlled Trials (RCTs) have already been mentioned several times in this article, as a technique that a robot can use to deal with confounding variables in causal inference

Summary

Introduction

Causal reasoning is generally considered to be a crucial human cognitive competence necessary for a range of tasks such as predictions, diagnoses, categorization, action planning, decision-making, and problem solving [1]. The. The purpose of this article is to provide a structured analysis of how the broad area of causality relates to the broad area of robotics. Inspired by a categorization of human causal cognition [5], we propose a categorization of robot causal cognition. We identify several additional aspects of robotics research where causation plays, or may play, an important role. We review related earlier work in robotics and identify connections to earlier work on causation in other sciences. We find very little earlier work in robotics where causation is explicitly mentioned. We identify several interesting connections between large sub-areas of robotics and causal concepts. Task planning and several common learning paradigms rely on, often implicit, causal assumptions and principles

Objectives

Discussion

Conclusion