The standard white cane is an essential aid for getting out and about for many visually impaired people, but to date, it hasn't offered much in the way of affordable modern updates – which is something that a group of researchers wants to change.
Borrowing technology designed for autonomous vehicles, the team has come up with a self-navigating smart cane that can identify obstacles in the surrounding environment, and nudge the user safely away from them.
In tests, the smart, augmented cane increased walking speed for visually impaired volunteers by 18 percent. For the 250 million people with sight difficulties worldwide, this assistive technology could lead to a serious increase in their quality of life.
"We wanted something more user-friendly than just a white cane with sensors," says mechanical engineer Patrick Slade from Stanford University in California. "Something that cannot only tell you there's an object in your way, but tell you what that object is and then help you navigate around it."
The smart cane relies heavily on LIDAR (Light Detection and Ranging), which uses reflecting lasers to spot objects and just distance. Sensors common to smartphones – GPS, accelerometers, magnetometers, and gyroscopes – keep track of the user's position, speed, and direction, respectively.
On the software side, the cane uses several artificial intelligence algorithms, including simultaneous localization and mapping (SLAM) – a way of building up a map of an unknown area while also keeping track of a user's location within it.
At the tip of the cane, a motorized, omnidirectional wheel can nudge walkers in one direction or another, and can even be used to direct someone to a destination like a coffee shop (much like a smartphone or car sat nav would).
"We want the humans to be in control but provide them with the right level of gentle guidance to get them where they want to go as safely and efficiently as possible," says computer scientist Mykel Kochenderfer, also from Stanford.
This isn't the first time that technology has been packed into smart canes, but current models are heavy and expensive. This new one is lightweight and could be available for a few hundred dollars – less than a tenth of the cost of some existing smart canes.
The cane was developed with the help of visually impaired people, who gave "incredible feedback" according to Slade. A lot of the decisions around design and navigation were made based on the feedback of those who would actually be using the device.
For now, though, this is still a research prototype. The team has made their design open-source, so anyone with the necessary know-how and materials (which cost about $400) can build their own version of the smart cane and assist in its development.
"We wanted to optimize this project for ease of replication and cost," says Kochenderfer. "Anyone can go and download all the code, bill of materials, and electronic schematics, all for free."
The research has been published in Science Robotics.