Hardware

Determining the distance

Two distance-measuring devices were tested:

  • HC-SR04, an Arduino-compatible ultrasound module
  • Car parking sensor

Both devices have a similar operating principle. They are different in directional diagrams, range of obstacle detection, and the way they are constructed.

Table 1 – Comparison of characteristics

Characteristic HC-SR04 Parking sensor
Detection distance (m) 4 2,5
Voltage (V) 5 5
Number of sensors in one device 1 4
Information output analog digital

While testing we found that the HC-SR04 doesn’t detect obstacles as well works worse in challenging weather conditions (low temperatures)

Information processing and control

We chose the Arduino platform as a control board. Smaller versions like Arduino mini, Arduino Nano, or Arduino miniPro are the most appropriate choice for this task. In principle, any controller with similar characteristics can be used.

Power

Lithium-Ion (Li-Ion) or Nickel-Metal hydride (Ni-Mh) batteries can be used to power the device.

Li-Ion batteries have the following advantages compared to Ni-Mh for optimized performance in normal climate conditions:

  • The charging circuit is easy to make
  • Assembled charging modules are available
  • High output voltage
  • Versatile sizes and capacities

Ni-Mh batteries are more suitable for use at low temperatures

The output voltage is not sufficient for all the device components so it has to be amplified. We use DC-to-DC up-converters. The input voltage is 0.9-6V, the output is 5V.

The module starts working from 0.9 V so to get 5V we need just one 1.2V Ni-Mh element. The smaller the input, however, the less stable the device due to a low loading capacity, so it’s best to supply at least 2.4V (2 Ni-Mh elements) or 3.7V (Li-ion). Also, some DC-to-DC converters only work with voltages above 3V, which should also be considered.

Charging the batteries

There are many inexpensive ready-made modules with indicators of remaining charge. Some batteries have a built-in charge controller with 4.2-5V being sufficient as an input voltage.

Ni-Mh batteries are more complicated. There are no built-in solutions available on the market. A specialized external charging device can be used or an in-house charging circuit has to be developed.

One of the ways to charge a Ni-Mh element is by using two LM317 (or similar) linear stabilizers in series. The first operates in limiting current mode, and the second in limiting voltage mode.

A fully-charged Ni-Mh element has a voltage of approximately 1.45V. The charging current is set at 100-200mA. If there is no radiator then the charging current should be no higher than 100mA otherwise the circuits will overheat and go out of order.

The advantage of such circuit is that it is not necessary to control the charging state: once the needed voltage is reached the current will drop to a safe minimum.

The input voltage of this circuit should be at least 7.5V. If the stabilizers are uncooled it is not recommended to go beyond this voltage.

Obstacle warning.

Depending on the choice of the warning channel (audio or tactile) the device is either a buzzer or a vibrator motor. Both can be combined to give the user a choice of switching between them.

 

The working principle of the devices

The device works on ultrasound reflected from obstacles.

A picture illustrating the working principle

A picture illustrating the working principle

An ultrasound impulse is sent in a certain direction. If the signal meets an object on its way it is fully or partially reflected as an echo and can be detected by a receiver. One can learn the distance to the object by measuring the time difference between the time the signal is sent and detected. Information about the distance is conveyed by vibration: the closer the object, the more intensive the vibration.

Prototypes of the ultrasound devices

Here you can see prototypes of various ultrasound devices. They are designed to inform visually impaired people of obstacles on their way. They are not supposed to substitute a usual cane but they enhance it, increasing the safety of movement for the visually impaired.

Minimizing the price of such devices in order to make them easily available for visually impaired people is one of the main objectives of this project. We are aiming at about 30 Eur, which is several times cheaper than existing devices on the market, produced both in Russia and outside of it.

Once the project is finished, all the documentation and the code will be made freely available on the web so that people around the world can reproduce these devices to improve the quality of life for visually impaired people. Having the access to the original code will allow other people to build upon and improve our design, thus developing the project further.

While testing, we discovered that the most convenient way to tell a person that there’s an obstacle nearby is by vibration as it doesn’t interfere with their hearing – an indispensible channel for a visually impaired person. This is why all the devices you see here vibrate and they do so more intensively as you get closer to an obstacle.

The ultrasound cane cap can be mounted on a normal cane for the visually impaired people to warn them about the obstacles located on a higher level then is accessible for the cane, e.g. tall cars, fences, gates, or barriers.

3D model of the ultrasound cane top

Picture of a 3D model of the ultrasound cane top

Picture of a 3D model of the ultrasound cane top

Picture of a 3D model of the ultrasound cane top

Picture of a 3D model of the ultrasound cane top

 

Picture of a 3D model of the ultrasound cane top

Picture of a 3D model of the ultrasound cane top

Picture of a 3D model of the ultrasound cane top

Picture of a 3D model of the ultrasound cane top

 

We printed the sensor case on a 3D printer and assembled a prototype of an ultrasound cane top

Фото корпуса, напечатанного на 3D-принтере

Фото корпуса, напечатанного на 3D-принтере

Фото корпуса, напечатанного на 3D-принтере

Фото корпуса, напечатанного на 3D-принтере

Фото корпуса, напечатанного на 3D-принтере

Фото корпуса, напечатанного на 3D-принтере

 

The ultrasound badge will transmit the information about knee- to head-high obstacles. The information is transmitted through a belt on the neck, which supports the device.

Ultrasound badge

Picture of a 3D model of the ultrasound badge

Picture of a 3D model of the ultrasound badge

Picture of a 3D model of the ultrasound badge

Picture of a 3D model of the ultrasound badge

 

 

Picture of a 3D model of the ultrasound badge

Picture of a 3D model of the ultrasound badge

 

We printed the sensor case on a 3D printer and assembled a prototype of an ultrasound badge

Фото корпуса, напечатанного на 3D-принтере

Фото корпуса, напечатанного на 3D-принтере

Фото корпуса, напечатанного на 3D-принтере

Фото корпуса, напечатанного на 3D-принтере

 

 

All tests were performed by two visually impaired volunteers, Valeriy Shalintsev and Pavel Bortnikov, and we express our sincere gratitude to them. Without them we would not know which direction we should be going how the visually impaired people would actually use our devices. With their help, we know what the final product should look like and are now testing final prototypes.

Ultrasound cane design concepts

A sketch of various ultrasound devices

Picture of a sketch

Picture of a sketch

 

The workpieces have arrived

A picture of a mold for device cases

A picture of a mold for device cases

 

 

Various ultrasound devices

Photo of assembled devices

Photo of assembled devices

Photo of assembled devices

Photo of assembled devices

 

The photo shows product concepts: (left to right) ultrasound flashlight, ultrasound glasses, ultrasound cane top, ultrasound badge.

 

Ultrasound cane development starts

We decided to make a cane for the visually impaired equipped with sensors to detect waist- and head-high obstacles.
The objective was to create an openhardware version of existing canes priced under 30 euro and then make the design freely available on the Internet.
I wanted to take an easily available device to build upon, so I decided to use a parking sensor.
I ordered a $15 sensor from aliexpress.com to try the idea out and put together the first prototype which started to vibrate as you got closer to an obstacle and the vibrations became more frequent. This was based on 1 channel processing.
Picture of prototype 1

Picture of prototype 1

Picture of prototype 1

Picture of prototype 1

  Video of prototype 1 testing

Video of prototype 1 testing

I was happy with the results of testing and decided to continue development in this direction. Now I needed to determine the position of the ultrasound sensors.
As the next step I decided to assemble a device with 2 sensors in a special case to be mounted on a cane. At this stage, Timur Gazizov joined the project and since then we’ve continued work together.
Picture of prototype 2

Picture of prototype 2

Picture of prototype 2

Picture of prototype 2

The tests have shown that the size and weight of the setup has to be reduced :) The third prototype was created
Picture of prototype 3

Picture of prototype 3

Picture of prototype 3

Picture of prototype 3

As we were testing the 3rd prorotype we discovered that it didn’t work at temperatures below -20 ºС. We decided to build the 4th version of the device based on Arduino ultrasound sensors and to add new functionality. We added a colour sensing function, but it can’t tell colours apart too well yet :) Testing prototype 4 showed that the beam pattern was too narrow and the cane didn’t “notice” some of the obstacles.
Picture of prototype 4

Picture of prototype 4

Picture of prototype 4

Picture of prototype 4

The 5th version as a result :)
Picture of prototype 5

Picture of prototype 5

Picture of prototype 5

Picture of prototype 5

Ok, prototype 5 failed because it was inconvenient. 6th version in process!
So far, all our prototypes are tested by Valeriy Shalintsev, we are very grateful to him!
One more thing. After talking to Pavel Vladymirovich Bornikov, a teacher at Correction School 127 in Chelyabinsk (the only school for blind and visually impaired children), we’ve decided to create a talking version of the Braille alphabet to help the children learn the Braille script.
Picture of prototype 6

Picture of prototype 6

Picture of prototype 6

Picture of prototype 6

 

Picture of prototype 7

Picture of prototype 7

Picture of prototype 7

Picture of prototype 7

Picture of prototype 7

Picture of prototype 7

So I’m still not satisfied with the results.
:)

The search goes on…