Smart Pill Dispenser Hardware

A Critical Look At Smart Wheelchairs

Benjamin Narin1, Makenzie Brian1, and William D. Smart1 Abstract— Research into smart wheelchairs has been con- ducted for decades, but we have yet to see the widespread use of this technology among full-time wheelchair users. We argue that the main reason for this is that there is a mismatch between research and the actualities of using a powered mobility device in the real world. Based on our own research experiences, we enumerate some of these disparities, and offer some suggestions for where work in smart wheelchairs might focus in the coming years.

I. Introduction

Research into smart wheelchairs has been carried out since the 1980s , . However, after almost forty years of effort, we have yet to see a widespread adoption of this technology. Although some self-driving wheelchairs are actually operating in the real world , they are designed to carry people who are not normally powered wheelchair users. We believe this is a key point that greatly simplifies the problem. Wheelchairs with sensors are simply not the same as robots.

Despite the fact that there is a large body of excellent work in this area, we claim that there is still a long way to go before we can realize the full potential of smart wheelchairs.

Critically, we must take a hard look at the realities of powered mobility devices as they are used in everyday life. We begin with a brief survey of related work, offer thoughts on the complications of designing a practical smart wheelchair system for everyday use, and finish by offering some suggestions for future directions in this field.

Ii. Related Work

Smart wheelchairs incorporate sensors that allow the sys- tem to perceive its environment and react to it. Leaman and La give a comprehensive survey of smart wheelchair research. While there are many such systems described in the survey, we have noticed some commonalities.

Many of the wheelchairs that they describe are used in systems, 48 of the 155 cited papers, focus on a navigation task where a human user is not required to be considered.

While navigation is important, it also fails to consider complexities of the real world. Additionally, a common theme in smart wheelchair work is to not address long term applications. This assumption is indicative of the lack of real-world focus currently in the area of smart wheelchairs.

*This work was supported, in part, by the National Science Foundation, under award 1541251. 1The authors are with the Collaborative Robotics and Intelligent Systems

Iii. Complications

In this section, we discuss the complications of imple- menting a smart wheelchair system for practical, daily use.

A. Physical Structure And Environment

It is tempting to analogize a smart wheelchair to a robot or a self-driving car , since these share the same basic components: sensing, computation, and mobility. But this analogy breaks down in a subtle way: for a robot or a car, we can create a physical model, used by the navigation system, that does not change over time. We know, with great accuracy, where the physical boundaries of the robot are.

While we can do this for the wheelchair itself, it will be different when there is someone using it. People will move the seat, hang things on the side of the chair, and lean outside the defined boundary. All of these things change the “shape” of the wheelchair, and the volume that we must check for collisions.

B. Environment And Social Conventions

By simply keeping the wheelchair further away from obstacles, we can address these un-modeled elements. This works in a building that complies with the Americans with Disabilities Act (ADA). However, due to cost, many full-time wheelchair users live in ADA non-compliant spaces, with narrow hallways and doors. Padding the navigation system with a safety buffer can cause problems in these places, since doorways might be too confined for the system to safely plan a path through.

Self-driving cars work, in part, because of an assumption that there is a highly detailed map in which the car is well-localized . While we can build maps for our smart wheelchairs, they are subject to drastic change over time, as rooms are rearranged, and people wander through. Localiza- tion, too, is problematic in cluttered environments. A more serious inconsistency in the analogy with self-driving cars is the assumption of a set of fixed social rules. Cars drive on a particular side of the road and usually follow well- defined protocols. This is not true of wheelchairs, which must operate in a more fluid social environment. How close they can come to people and objects varies by context and cultural setting. While the repercussions of navigation mistakes in a smart wheelchair are less likely to be fatal than in an automobile, they are far easier to make, and can lead to confusion, embarrassment, and physical harm.

C. Sensors And Control

Individuals modify their wheelchairs, both to personalize them and to install medically-necessary equipment on them.

Related Journal Articles & DOI Links

Selected peer-reviewed publications relevant to 12 Lead ECG Acquisition. Click the DOI to access the full paper (may require institutional access).

Why Choose Us?

Bangalore guidance for robotics, Spectre and autonomous systems projects.