The gap above the cane

A white cane tells you what is on the ground. It cannot reach the sign at head height, the open truck door, or the person coming from behind. The tool is from the 1930s, and the gap has been there ever since.

54% of cane users hit something at head height at least once a year, and 43% change the way they walk for good afterwards. Most newer aids answer with sound, which competes with the sense blind people rely on most.

So we gave ourselves one constraint: no sound, no screen. Space has to arrive as touch.

What we could learn, and what we could not

We tried to reach the blind community in Tallinn early. A contact offered to introduce us, but nobody had time in the first weeks, so we started from papers, podcasts and existing devices like BlindSight. What blind people wrote and said kept landing on one word: freedom. Walking alone, without a constant calculation running in your head.

Later we talked the concept through with a blind man. He liked it, described what his cane keeps missing, and said he would want something like it for diving, his hobby. That pointed to uses we had not thought about.

One conversation is not a study. Testing with blind people earlier, and more than once, is the first thing I would change.

Two walks with a blindfold

We picked two everyday routes: from home to a bus stop, and past a half-blocked sidewalk. Then we walked them ourselves, blindfolded and with a cane, and marked every point where the surroundings stopped giving a clear answer.

Both walks showed the same thing. The hard part is not one obstacle. It is that every step is a decision and the alertness never switches off. From the outside, none of that is visible.

Testing on our own bodies

Before any electronics, we built four cardboard forms: a belt, a necklace, a shoulder strap and an insole. Then three tests.

First, two Android phones on chest and back, vibrating on their own. Everything became a warning, and it caused more anxiety than orientation. Second, a Wizard of Oz test: one of us steered the vibration by hand while another walked blindfolded. Far less signal was enough, and silence was trusted as safe.

Third, just tapping. The waist is ticklish and every tap felt like a shock. Shoulders and neck could carry a signal for a long time without being annoying. That decided the form.

What we built

A necklace with motors at the chest, upper back and both shoulders, worn over a coat. Faux fur hides cables and makes it read as an accessory, not a medical device. Where it buzzes says the direction. How fast it pulses says how urgent. Silence means clear.

Inside, a Raspberry Pi runs object detection on two cameras, front and back, and an Arduino drives six motors so the rhythm stays steady even when detection is slow. At the exhibition on 30 May it ran live, with a manual mode as backup.

We planned four distance sensors and built two, so the sides are still blind. Whether silence alone is enough to trust that the device is on was never tested with blind users.