Presentation
DH5 - Assistive Robots for Home Healthcare: Exploring Proxemic Factors for Older Adults
SessionPoster Session 1
DescriptionAs the global population ages, we see older people outnumber the younger generation, and there is a shortage of formal and informal caregivers for older adults. Technologies have the potential to support older adults' needs and enhance their independence and quality of life. One promising solution is the use of assistive robots, which can offer various forms of support in home environments. These robots often operate in close proximity to users, making it crucial to understand the appropriate distance robots should maintain when assisting older adults while promoting user comfort, fostering trust, and ensuring predictable robot behavior. Designing robots to adapt to the appropriate distance in different contexts can enhance the effectiveness of the robot's assistance and the overall user experience.
Older adults may have unique preferences and comfort levels related to various aspects of technology interaction, such as unfamiliarity with advanced technologies like robots. Studying proxemics in the context of older adults’ interaction with robots is crucial to understanding their preferences, comfort boundaries, and perceptions of safety. Proxemics is the study of personal space and the physical distances that people maintain in social interactions, which can be categorized into intimate (within 18” of the person’s space), personal (18” to 48”), social (48” to 144”), and public (144” and beyond) spaces. In the context of assistive robots, applying these principles may allow for more personalized interactions that respect the user’s boundaries and preferences. Proxemics not only affects the initial interaction but also plays a role in long-term acceptance and the development of trust between humans and robots. By incorporating proxemics into robotic design, robots can be designed to adapt their positioning and movements to suit the needs and comfort levels of older adults, enhancing their sense of security and promoting smoother, more natural interactions.
We conducted an initial study of proxemics with five older adults (four female; one male), aged 67-77 (M = 73.4, SD = 4.62), some of whom had mobility or cognitive impairments. The most common medical conditions among participants included arthritis, heart conditions, and high blood pressure. They interacted with an assistive robot in a simulated home environment, performing tasks related to daily living activities. We focused on two key tasks: water bottle delivery and video calls. These tasks are relevant to older adults’ daily routines. The water bottle delivery task simulates assistance with fetching and carrying objects, whereas the video call task reflects the use of technology to facilitate social connections. By recreating these activities in a controlled, simulated home environment, we could assess how proxemic variables, such as the robot’s distance and positioning, impacted user comfort and task performance, providing insights into optimal human-robot interactions.
The robot used was Stretch, a mobile manipulator designed by Hello Robot to assist with everyday tasks. It was teleoperated throughout the study. For the water bottle delivery task, the interviewer instructed the robot, through a voice command, to approach the water bottle, pick it up, and then deliver it to the participant by approaching from their right-hand side. In the video call task, the robot approached the participant from the left-hand side after successfully connecting a call via a tablet mounted on the robot. Once the call was initiated, the interviewer received the call and instructed the robot to bring the tablet to the participant.
To analyze the impact of the assistive robot on human behavior with respect to proxemic space, we developed a coding scheme for video analysis. The coding focused on identifying human behaviors such as four limbs positioning, body orientation, facial expressions, head movement, and eye gaze. The coding also included the robot’s actions such as arm and gripper actions, orientation, approach direction, and the distance between the robot’s base, gripper, and the participant. After multiple iterations and refinements of the coding scheme, we employed it for the video analysis to capture subtle behavioral changes in participants as they responded to the robot’s proximity and actions, offering insights into how proxemics influences comfort and effectiveness in older adults and robot interactions.
We explored human behaviors in different proxemic zones (Social Space, Personal Space, and Intimate Space) for the five participants. When the robot is farther away, in the Social Space, participants moved to face towards the robot, some point to it, some just watch it approach (head movement, eye gaze, and body orientation are mostly focused o on the robot). The participant would orient their body, head, and eye gaze to focus on the robot at the start of the task. After the robot picked up the water bottle, still within the Social Space, the robot proceeded toward the participant from the right-hand side. During this transition, participants would share their observations with the interviewer during the think-aloud session. Once the robot entered the intimate space, participants became more focused on the robot, paying close attention to how the robot manipulated its gripper and delivered the water bottle. In sum, the humans’ reactions to the robot varied as it moved from Social to Personal to Intimate Space. This general pattern was observed for all the participants although some participants exhibited fewer behavior changes.
The video analysis of the interactions revealed that proximity affects older adults’ behavior, within social space and intimate space, prompting more frequent and dynamic responses as the robot approached them. This understanding contributes to establishing a framework for measuring and evaluating the impact of proxemics in human-robot interactions. It also provides more details on proxemic considerations to address in the design of assistive robots that can engage in ways that will foster trust and acceptance. Future work will further explore how proxemic variables relate to the actions of assistive robots for a broader sample of older adults. This will help to identify ways the robot should provide assistance when operating in close proximity to older adults and to fine-tune robot actions for greater comfort, efficiency, and trust for the interactions.
Acknowledgment:
This research is funded by the National Institute on Aging (National Institutes of Health) Phase II Small Business Innovation Research Grant #2R44AG072982-02.
Older adults may have unique preferences and comfort levels related to various aspects of technology interaction, such as unfamiliarity with advanced technologies like robots. Studying proxemics in the context of older adults’ interaction with robots is crucial to understanding their preferences, comfort boundaries, and perceptions of safety. Proxemics is the study of personal space and the physical distances that people maintain in social interactions, which can be categorized into intimate (within 18” of the person’s space), personal (18” to 48”), social (48” to 144”), and public (144” and beyond) spaces. In the context of assistive robots, applying these principles may allow for more personalized interactions that respect the user’s boundaries and preferences. Proxemics not only affects the initial interaction but also plays a role in long-term acceptance and the development of trust between humans and robots. By incorporating proxemics into robotic design, robots can be designed to adapt their positioning and movements to suit the needs and comfort levels of older adults, enhancing their sense of security and promoting smoother, more natural interactions.
We conducted an initial study of proxemics with five older adults (four female; one male), aged 67-77 (M = 73.4, SD = 4.62), some of whom had mobility or cognitive impairments. The most common medical conditions among participants included arthritis, heart conditions, and high blood pressure. They interacted with an assistive robot in a simulated home environment, performing tasks related to daily living activities. We focused on two key tasks: water bottle delivery and video calls. These tasks are relevant to older adults’ daily routines. The water bottle delivery task simulates assistance with fetching and carrying objects, whereas the video call task reflects the use of technology to facilitate social connections. By recreating these activities in a controlled, simulated home environment, we could assess how proxemic variables, such as the robot’s distance and positioning, impacted user comfort and task performance, providing insights into optimal human-robot interactions.
The robot used was Stretch, a mobile manipulator designed by Hello Robot to assist with everyday tasks. It was teleoperated throughout the study. For the water bottle delivery task, the interviewer instructed the robot, through a voice command, to approach the water bottle, pick it up, and then deliver it to the participant by approaching from their right-hand side. In the video call task, the robot approached the participant from the left-hand side after successfully connecting a call via a tablet mounted on the robot. Once the call was initiated, the interviewer received the call and instructed the robot to bring the tablet to the participant.
To analyze the impact of the assistive robot on human behavior with respect to proxemic space, we developed a coding scheme for video analysis. The coding focused on identifying human behaviors such as four limbs positioning, body orientation, facial expressions, head movement, and eye gaze. The coding also included the robot’s actions such as arm and gripper actions, orientation, approach direction, and the distance between the robot’s base, gripper, and the participant. After multiple iterations and refinements of the coding scheme, we employed it for the video analysis to capture subtle behavioral changes in participants as they responded to the robot’s proximity and actions, offering insights into how proxemics influences comfort and effectiveness in older adults and robot interactions.
We explored human behaviors in different proxemic zones (Social Space, Personal Space, and Intimate Space) for the five participants. When the robot is farther away, in the Social Space, participants moved to face towards the robot, some point to it, some just watch it approach (head movement, eye gaze, and body orientation are mostly focused o on the robot). The participant would orient their body, head, and eye gaze to focus on the robot at the start of the task. After the robot picked up the water bottle, still within the Social Space, the robot proceeded toward the participant from the right-hand side. During this transition, participants would share their observations with the interviewer during the think-aloud session. Once the robot entered the intimate space, participants became more focused on the robot, paying close attention to how the robot manipulated its gripper and delivered the water bottle. In sum, the humans’ reactions to the robot varied as it moved from Social to Personal to Intimate Space. This general pattern was observed for all the participants although some participants exhibited fewer behavior changes.
The video analysis of the interactions revealed that proximity affects older adults’ behavior, within social space and intimate space, prompting more frequent and dynamic responses as the robot approached them. This understanding contributes to establishing a framework for measuring and evaluating the impact of proxemics in human-robot interactions. It also provides more details on proxemic considerations to address in the design of assistive robots that can engage in ways that will foster trust and acceptance. Future work will further explore how proxemic variables relate to the actions of assistive robots for a broader sample of older adults. This will help to identify ways the robot should provide assistance when operating in close proximity to older adults and to fine-tune robot actions for greater comfort, efficiency, and trust for the interactions.
Acknowledgment:
This research is funded by the National Institute on Aging (National Institutes of Health) Phase II Small Business Innovation Research Grant #2R44AG072982-02.
Event Type
Poster Presentation
TimeMonday, March 314:45pm - 6:15pm EDT
LocationFrontenac Foyer
Digital Health (DH)
Simulation and Education (SE)
Hospital Environments (HE)
Medical and Drug Delivery Devices (MDD)
Patient Safety and Research Initiatives (PS)