TECHNICAL FRAMEWORK FOR A CARE ROBOT - open-source control architecture for home care robots

Sep 29, 2026

TECHNICAL FRAMEWORK FOR A CARE ROBOT describes an open-source, platform-agnostic control architecture for home care robots tasked with reliable sequential decision-making under uncertainty in human behaviors and environments. The approach combines a lightweight ROS2-based AI planner with integrated robotics and computer-vision algorithms to generate executable task plans while sensing the environment and care recipient. The system supports care services such as medication reminders and exercise encouragement, and reports early data from ongoing field studies.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages1

Reading Time3 min

AccessFree to download

TECHNICAL FRAMEWORK FOR A CARE ROBOT

Open-source control architecture and perception-action pipeline

Lightweight ROS2-based AI planner and executable task plans

Integrated robotics/computer vision algorithms for real-time services

Care services and early field-study data

454 Innovation in Aging, 2025, Vol. 9, No. S2 the identification of movement deficiencies and tracking of progress over time. We will illustrate the process by which therapists can teach robots exercises using AI, how the exercises are encoded and delivered by the robot, a framework for tracking performance, and results of a survey of 110 licensed therapists who were asked to evaluate the potential for robot-led therapy in home settings. We will discuss the emerging potential of SAR as an in-home adjunct therapy to supplement clinical exercise programs as well as the barriers to implementation including the anatomical structure of current robots, ability to provide real-time, physical, corrective feedback, potential loss of human connection, user acceptance and cost. Abstract citation ID: igaf122.1545 TECHNICAL FRAMEWORK FOR A CARE ROBOT Momotaz Begum, Ola Ghattas, Marzan Alam, Moniruzzaman Akash, Noushad Sojib, and Mostafa Hussein, University of New Hampshire, Durham, New Hampshire, United States A robot that cares for older adults in their home is tasked with making reliable sequential decisions subjected to uncertainties of human behaviors and environments. This is highly non-trivial due to a plethora of challenges involved with real-time processing of perception-action pipeline. The state of the art in AI robotics offers efficient tools to solve individual problems that may arise along this pipeline. An integrated solution, however, is still non-existent. We will present our ongoing work on developing an open-source control architecture that implements the entire perception-action pipeline for home care robots in a platform agnostic manner. The core of this architecture is a lightweight ROS2-based AI planner that can generate task plans executable by any robot – equipped with appropriate actuators – while sensing the current state of the environment and the care recipient (ROS: Robot Operating System; AI: Artificial Intelligence). Tasks typically included care services – e.g., reminding an older adult to take the medication or encouraging him/ her to perform physical exercise – defined based on the needs of a care recipient. We integrate state of the art algorithms in robotics and computer vision – such as navigation, tracking, face recognition, natural language processing, etc.– in an innovative way with this lightweight AI planner to realize real-time planning and execution of a wide range of services that are critical for older adults to age gracefully in their own homes. We will report some data from our ongoing field studies. Abstract citation ID: igaf122.1546 ETHICAL CHALLENGES AND STRATEGIES IN IMPLEMENTING SOCIAL ASSISTIVE ROBOTS IN DEMENTIA CARE Jing Wang, Sajay Arthanat, Momotaz Begum, and Dain LaRoche, University of New Hampshire, Durham, New Hampshire, United States The integration of social assistive robots (SARs) in dementia care presents both opportunities and challenges in enhancing support for persons living with dementia (PLWD) and their care partners. From the initial stages of development through deployment, this study highlights how ethical considerations were embedded in the interdisciplinary collaboration between researchers from engineering, health sciences and nursing to ensure SARs align with the needs, values, and autonomy of PLWD. Drawing on data from baseline and follow-up interviews, participant-robot interaction observations, and interdisciplinary discussions, we examine how ethical awareness shaped SAR design, functionality, and implementation strategies. Key ethical considerations included balancing autonomy with safety in SAR interventions, addressing privacy concerns related to smart sensing technologies, and ensuring transparency in decision-making. Findings reveal that co-designing SAR care protocols with PLWD and their care partners through person-centered design fostered trust and engagement. Adaptations based on user feedback allowed for greater responsiveness, en

FAQs

What problem does the care robot architecture address?

It targets reliable sequential decision-making in home settings under uncertainty from human behaviors and the environment, which requires real-time perception-action processing.

How is the proposed system organized?

It uses a lightweight ROS2-based AI planner to generate executable task plans, integrating state-of-the-art robotics and computer-vision algorithms for navigation, tracking, face recognition, and natural language processing.

What kinds of care services can the framework support?

The tasks include reminding an older adult to take medication and encouraging physical exercise, based on the care recipient’s needs.

Sep 29, 2026

Upload by pixelkiddo