These days we can control the key functions of a smartphone, TV or even automobile by voice. Clearly, voice control is a cool feature that will soon be commonplace on a broad range of consumer devices. For disabled people, however, assistive technologies like voice control are not just cool but an essential way to help make their lives better.
Researchers at the Massachusetts Institute of Technology argue that voice control could improve the independence, quality of life and safety of tens of thousands of wheelchair users.
The goal of the MIT Intelligent Wheelchair Project is to enhance a powered wheelchair using sensors to detect the wheelchair’s surroundings, a speech interface to interpret commands, a wireless device for room-level location determination, and motor-control software. The result is a robotic wheelchair that can learn the layout of its environment — such as hospital or care home — through a narrated, guided tour given by the user or the user’s caregivers.
Once programmed, the wheelchair can move to any previously-named location under voice command (e.g., “Take me to the cafeteria”).
The MIT researchers argue their intelligent wheelchair is ideal for people who have lost mobility due to brain injury or the loss of limbs, but who retain speech.
The technology can also enhance safety for users of ordinary joystick-controlled powered wheelchairs, by preventing collisions with walls and other fixed objects.
I suppose that unless you have been in this situation, it is easy to underestimate the learning curve required to master “driving” a powered wheelchair — and the possible risk of collision-induced injuries such as wounds and broken limbs.
But I’m surprised that the researcher haven’t gone further and taken advantage of the powerful natural language capabilities of commercial voice recognition platforms such as that offered by Artificial Solutions.
Instead of forcing the user to interact using a limited set of pre-programmed commands, a wheelchair equipped with NLI technology would be able to understand the user’s intent. For example, there are many euphemisms for “I need to go to the bathroom”.
An NLI-equipped wheelchair would quickly recognize the phrase or phrases its user most often uses, in much the same way that a smartphone learns the favorite queries of its owner.
Perhaps more importantly, it should understand that when the user utters this particular phrase, they are probably in a hurry to make use of the bathroom! The autopilot built in to the wheelchair would thus direct the wheelchair to take the most direct route and at the fastest comfortable speed — which could depend on factors such as terrain, for example.
The electric wheelchair was invented by George J. Klein, a Canadian nuclear scientist, to benefit disabled veterans of World War II. Seventy years later, the vast majority of wheelchair users remain confined to traditional self-propelled models as national healthcare schemes balk at reimbursing the high cost of an electric wheelchair.
That’s a big pity, as there are plenty of wheelchair users who, even if they do have full control of their arms, nevertheless are limited in how far they can travel under their own steam.
A motorized wheelchair with built-in intelligence to understand where the user wants to go and how to get there would provide a great boost to their autonomy — and self-confidence as well.
