Interprofessional and Agile Development of Mobirobot: A Socially Assistive Robot for Pediatric Therapy Across Clinical and Therapeutic Settings

TL;DR

Mobirobot presents a multiprofessional, agilely developed socially assistive robot designed to support pediatric mobilization in both inpatient and outpatient settings for psychiatric and somatic therapy. Through an intensive co-development process with clinicians, therapists, and patients, the study demonstrates how a NAO6-based robot can deliver adaptable HIIT-like exercise regimes, motivational dialogue, and no-code system feedback while integrating into real-world clinical workflows. Early findings highlight improvements in engagement and the importance of context-aware design, robustness, and minimal intrusion, though challenges in sensor capability, long-term engagement, and recruitment remain. The work offers practical design principles and a scalable platform for future health-tech interventions in paediatric rehabilitation and mental health care.

Abstract

Introduction: Socially assistive robots hold promise for enhancing therapeutic engagement in paediatric clinical settings. However, their successful implementation requires not only technical robustness but also context-sensitive, co-designed solutions. This paper presents Mobirobot, a socially assistive robot developed to support mobilisation in children recovering from trauma, fractures, or depressive disorders through personalised exercise programmes. Methods: An agile, human-centred development approach guided the iterative design of Mobirobot. Multidisciplinary clinical teams and end users were involved throughout the co-development process, which focused on early integration into real-world paediatric surgical and psychiatric settings. The robot, based on the NAO platform, features a simple setup, adaptable exercise routines with interactive guidance, motivational dialogue, and a graphical user interface (GUI) for monitoring and no-code system feedback. Results: Deployment in hospital environments enabled the identification of key design requirements and usability constraints. Stakeholder feedback led to refinements in interaction design, movement capabilities, and technical configuration. A feasibility study is currently underway to assess acceptance, usability, and perceived therapeutic benefit, with data collection including questionnaires, behavioural observations, and staff-patient interviews. Discussion: Mobirobot demonstrates how multiprofessional, stakeholder-led development can yield a socially assistive system suited for dynamic inpatient settings. Early-stage findings underscore the importance of contextual integration, robustness, and minimal-intrusion design. While challenges such as sensor limitations and patient recruitment remain, the platform offers a promising foundation for further research and clinical application.

Figures (6)

• Figure 1: Mobirobot during physiotherapy with a patient in the InPT Setting.
• Figure 2: Schematic depiction of the setups in the three settings a)-c).
• Figure 3: Schematic representation of the iterative, co-development process, illustrating multilevel feedback loops at different scales. Small purple loops depict rapid exchanges on the exercise program via online meetings, email, and file sharing. Medium orange loops represent sprint meetings within the development team, used to prioritize and implement changes. Large loops correspond to full-team meetings featuring system demonstrations and structured feedback sessions as well as engagement activities with patients and clinical staff, including introductory workshops and deployment site visits.
• Figure 4: Overview of communication between the different system components. The connections represent local Wi-Fi links initiated by the laptop.
• Figure 5: Schematic depiction of the GUI with regimen editor, controls and displays for system feedback.