[Image above] Michael Raitor, Ph.D. student in mechanical engineering at Stanford University, tests a new lightweight, affordable smart cane. Credit: Stanford, YouTube
Though there are many things I enjoy about living in Columbus, driving is certainly not one of them. Columbus consistently ranks near the top of cities with the worst drivers, and I personally was almost hit on multiple occasions when cars ran a red light—both when I was driving and walking!
The times I narrowly avoided being hit when walking across the road have heightened my awareness of the limitations of our built environment. Ever since cars became a main mode of transportation in the 20th century, urban planners designed cities for private vehicles rather than pedestrians, cyclists, or even public transport like buses. This prioritization was on full display last year, when the Governors Highway Safety Association announced a surge in the rate at which drivers struck and killed people on foot. They attributed this increase to more instances of dangerous driving during the pandemic combined with the private vehicle prioritization in infrastructure.
In recent years, more city planners are embracing a sustainable urban mobility plan, which emphasizes evaluating the whole functional urban area when designing rather than just accounting for cars. However, for these plans to be successful, city planners need to account for people who face additional limits to their mobility, such as people with visual impairments.
Pedestrian infrastructure such as tactile paving and sound signals can play a positive role in promoting mobility among people with visual impairments, but relying on governments to install this infrastructure can be a long and frustrating wait. Thus, it is desirable to develop technologies that help people take mobility into their own hands.
In June, we covered the development of an affordable braille display that can help people with visual impairments better use the capabilities of their smartphone. Then in August, we covered the development of an in-shoe navigation device to help people with visual impairments navigate a city on their own. Today, we cover the development of an affordable smart cane that relies on the same technology used in many self-driving cars.
LIDAR-based smart cane for indoor and outdoor navigation
A white cane, or long cane, is one of the most common mobility aids for people with visual impairments. However, white canes are static devices that cannot detect obstacles beyond the length of the cane and so are somewhat limited in the aid they can provide.
Researchers have experimented with creating augmented canes containing sensors capable of detecting approaching obstacles, but many of these previous attempts were heavy and expensive. The new smart cane by researchers at Stanford University weighs only 3 pounds, however, and it can be built at home using off-the-shelf parts and free, open-source software.
In the Stanford press release, graduate research assistant Patrick Slade says they wanted something more user-friendly than just a white cane with sensors. “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,” he says.
To achieve this functionality, Slade, graduate student Arjun Tambe, and associate professor Mykel Kochenderfer equipped the cane with a light imaging, detection, and ranging (LIDAR) sensor, which is the technology used in many self-driving cars. They also included GPS, accelerometers, magnetometers, and gyroscopes, like those on a smartphone, to monitor the user’s position, speed, and direction.
The cane uses artificial intelligence-based way finding and robotics algorithms like simultaneous localization and mapping (SLAM) and visual servoing to make decisions about navigation. Once a decision is made, the cane directs a motorized, omnidirectional wheel attached to its tip to gently tug left or right around an impediment.
See the smart cane in use in the video below.
Tests of the smart cane found that it increased the walking speed of participants with visual impairments by roughly 20% over a regular white cane. In addition, for people without visual impairments wearing blindfolds, walking speeds increased by more than a third.
While the researchers plan to refine their prototype and develop a model that uses an everyday smartphone as the processor, they are currently offering their current design for free. “We wanted to optimize this project for ease of replication and cost. Anyone can go and download all the code, bill of materials, and electronic schematics, all for free,” Kochenderfer says in the press release.
The code for the smart cane is available at this link. The paper explaining their research, published in Science Robotics, is “Multimodal sensing and intuitive steering assistance improve navigation and mobility for people with impaired vision” (DOI: 10.1126/scirobotics.abg6594).