LEADOR: A Method for End-To-End Participatory Design of Autonomous Social Robots.

Séverin Lemaignan,Katie Winkle,Emmanuel Senft

doi:10.3389/frobt.2021.704119

Abstract

Participatory design (PD) has been used to good success in human-robot interaction (HRI) but typically remains limited to the early phases of development, with subsequent robot behaviours then being hardcoded by engineers or utilised in Wizard-of-Oz (WoZ) systems that rarely achieve autonomy. In this article, we present LEADOR (Led-by-Experts Automation and Design Of Robots), an end-to-end PD methodology for domain expert co-design, automation, and evaluation of social robot behaviour. This method starts with typical PD, working with the domain expert(s) to co-design the interaction specifications and state and action space of the robot. It then replaces the traditional offline programming or WoZ phase by an in situ and online teaching phase where the domain expert can live-program or teach the robot how to behave whilst being embedded in the interaction context. We point out that this live teaching phase can be best achieved by adding a learning component to a WoZ setup, which captures implicit knowledge of experts, as they intuitively respond to the dynamics of the situation. The robot then progressively learns an appropriate, expert-approved policy, ultimately leading to full autonomy, even in sensitive and/or ill-defined environments. However, LEADOR is agnostic to the exact technical approach used to facilitate this learning process. The extensive inclusion of the domain expert(s) in robot design represents established responsible innovation practice, lending credibility to the system both during the teaching phase and when operating autonomously. The combination of this expert inclusion with the focus on in situ development also means that LEADOR supports a mutual shaping approach to social robotics. We draw on two previously published, foundational works from which this (generalisable) methodology has been derived to demonstrate the feasibility and worth of this approach, provide concrete examples in its application, and identify limitations and opportunities when applying this framework in new environments.

Highlights

In the context of robotics research, participatory design (PD) attempts to empower non-roboticists such that they can shape the direction of robotics research and actively collaborate in robot design (Lee et al, 2017)
Output may include potential use case scenarios (Jenkins and Draper, 2015), design guidelines/recommendations (Winkle et al, 2018), and/or prototype robot behaviours (Azenkot et al, 2016). Šabanović identified such methods as appropriate for the pursuit of a mutual shaping approach in robot design that is one that recognises the dynamic interactions between social robots and their context of use (Šabanović, 2010), an approach that we find compelling for designing effective and acceptable social robots efficiently
LEADOR is agnostic with regard to the specific computational approach to facilitating the teaching interaction, but we find Interactive Machine Learning (IML) to be a compelling solution, in line with the overall aims of the method, as it makes for an intuitive bi-directional teaching interaction for the domain expert

Summary

Introduction

In the context of robotics research, participatory design (PD) attempts to empower non-roboticists such that they can shape the direction of robotics research and actively collaborate in robot design (Lee et al, 2017). Common methods for the automation of social robot behaviour include utilising models based on human psychology (e.g., Theory of Mind, Lemaignan et al, 2017) or animal behaviour (Arkin et al, 2001) or attempting to observe and replicate human-human interaction behaviours (e.g., Sussenbach et al, 2014) This limits the potential for direct input from domain experts (teachers, therapists, etc.) who are skilled in the use of social interaction in complex scenarios. Previous work with such experts has demonstrated that a lot of the related expertise is intuitive and intangible, making it difficult to access in a way that can inform robot automation (Winkle et al, 2018). There is no opportunity for experts to create an adequate, situated mental model of the capabilities of the robot, limiting the guarantee of appropriate behaviour when the robot is eventually deployed to interact with users autonomously

Objectives

Methods

Discussion

Conclusion