The parts...

Below is a picture of all the parts that we will need for our product (minus the walking cane).

In the middle of the picture is the Arduino microcontroller.  In the bottom-right hand of the picture is our ultrasonic range finder -- the LV-MaxSonar-EZ4.  To the left of that is a cell phone vibrator within a plastic housing that we printed out using Penn's 3D printer.

Our Design

Below is concept art for "The Speed Stick."

What the art shows is a blind person being alerted that there is an object in front of them before the end of their stick makes contact with the object.  This is being performed with a microcontroller interfacing with an ultrasonic range finder and a cell phone vibrator.

There is, however, one additional functionality to "The Speed Stick."  Our device can also determine if there is an object moving toward it at a speed above a certain threshold.  It then communicates this information to the user with a unique vibration pattern so that they are aware that there is an object moving toward them.

Competing Technology Search and Reverse Engineering

Competing Technologies:

• The white cane
• Simple stick that is made of light-weight aluminum
• Color is universally identifiable as being for the blind
• Most-used walking assistant
• Requires up to 100 hours of training¹
• GuideCane
• Walking aid that physically guides the user
• Uses 10 ultrasound sensors
• The Mygo
• Rolling cane (1 wheel)
• Detects with a “smart sensor” and a live camera
• Gives audio feedback to the user
• Battery life up to 6 hours
• The guide dog
• Dog specially trained for physically guiding the user
• Are expensive: Cost between $12,000 and $20,000¹
• Only useful for about 5 years¹
• Only 1% of the blind population use this method¹
• C-5 Laser Cane²
• Cane that is useful for detecting curbs and edges using infrared light
• Feedback is via auditory measures
• 3 hours of continual use
• Mowat Sensor
• No longer being developed
• Nottingham Obstacle Detector
• Ultrasonic hand-held detector
• Sonicguide
• Eyeglass-type sensors that emit and receive sonar signals
• NavBelt
• Uses ultrasonic sensors placed on the belt to detect objects
• Uses auditory feedback (image below)

Reverse-Engineering the GuideCane:

Pros:

• Decreases work and conscious effort
• No active scanning
• Is on wheels
• Only 1 piece of information given – direction
• Minimal training required
• No sound cues necessary – all haptic feedback
• Passive wheels use less batteries – but still puts a time limit on the device’s use
• Handle angle is adjustable to adapt to users of various heights
• User’s can walk effectively at speeds up to 1 m/s

Cons:

• Drops when going down stairs
• No easy way to get up stairs
• Detects minor disturbances on ground (false positives)
• Square-shaped handle is not ergonomically designed
• Tabletops and other objects higher than the detecting range of the sensors will be ignored by the device
• Relatively heavy (4 kg)
• Only 8 directions that the user can choose between

Components:

• Consists of: housing, wheelbase, handle
• Housing: made of acrylic, contains most of the electronic components
• 8 ultrasonic sensors are placed on the front of the housing with an angular spacing of 15^o; 2 ultrasonic sensors are on the side to detect walls
• Sensors are close to the ground
• Housing/wheelbase: 43 cm wide, 25 cm high, 23 cm deep; weighs 4 kg
• Lightweight quadrature encoders to move the wheels; 2000 pulses per revolution, 5 pulses for a wheel movement of 1 mm
• Handle that changes angles to adapt to various user heights
• Electronics: PC/104 computer equipped with 486 microprocessors clocked at 33 MHz; 125 MB hard drive; custom built interface between PC and the sensors and actuators
• Interface sends signals for constant US rays and readings
• “The interface consists mainly of three MC68HC11E2 microcontrollers, two quadrature decoders, a FIFO buffer, and a decoder.”
• Powered by recharcheable NiMH batteries; lasts several hours
• No global navigation – just local
• Ultrasonic sensors controlled by the Error Eliminating Rapid Ultrasonic Firing (EERUF) method; allows firing 5-10 faster than conventional methods; EERUF is necessary to eliminate cross-talk between sensors
• Ultrasounds fire at 10 Hz
• Uses Histogramic in-motion mapping (HIMM) for map-building

Works Cited:

¹ Ulrich, Iwan. "The GuideCane -- Applying Mobile Robot Technologies to Assist the Visually Impaired." IEEE Transactions on Systems, Man, and Cybernetics. 31.2 (2001): 131-136. Print.

² Benjamin, J. Malverin. "The Laser Cane." State of Effort. Print.