Dear all,

I am happy to provide the winner and runners up of the 2018 PCBWay/ATMakers Assistive Technology Award.  This year we had over a hundred submissions for consideration and several very strong candidates for the award.  I am delighted to see this amount of interest in Assistive Technology among PCB designers.

Judging Approach & Criteria

When judging submissions I reviewed the submissions for the following criteria:

Eligible as an AT Project: Is the project in any way directed towards Assistive Technology or useful to people who need AT?  A large number of submissions were eliminated on this subjective consideration because the simply weren’t possibly AT-related (for example WiFi Sniffers and Power Converters).

Clear Purpose & Description: How easy is it to understand the goal of the project?  Could an AT professional see the benefits of this? Have the developers thought through the problem they are trying to solve?

Useful to AT: How useful is this to AT users.  Can it be used to solve a real problem in the AT space (regardless of whether that is its intention).

Intended for AT: Are the developers intending this to be used in AT?  We wanted to reward those who thought about AT Users when starting their project.  Some projects, while very useful to AT, may be targetted to a broader market.

Originality: How similar is this to other products of projects on the market?  

Affordability: AT is notoriously over-priced.  Is this project as affordable as possible given the goal?  Are there more cost-effective ways to approach the problem?

Ease of SMD Assembly: How complex is the design?  Will an assembly house be able to complete the project easily?  Can this be assembled by Makers? Will there be costs/time spent that are unnecessary?

Ease of Device Assembly:  Once SMD components are onboard, how hard is the final assembly.  For PTH-only designs, this is the entire assembly, but the real question is, when the device is delivered, how hard will it be to incorporate into an AT project.

Ease of Use: Have the developers done everything they can to make the experience of the end-user (who is already facing AT challenges) as simple as possible?  Have they taken real AT users into consideration (and ideally consultation) on their design to make their device easy to use?

Documentation:  This rapidly became one of the most important considerations.  For many designs, other criteria simply couldn’t be judged because of the difficulty in finding basic documentation about this overall goals, approach and status of the project.  In reality, none of the design submissions were adequately documented.  Those that had made a significant effort to document their work were rewarded in the judging.

“Other”/Subjective Assessment: All of the factors above were considered and as you will see in the Appendix, they were rated so that I could have some clear metrics to look at.  However, this Award was also based on a subjective assessment of how likely I felt the project was to be completed, to succeed in its goal, and to be available to AT Users in the near future.

Results

Winner: NeuroLab

The NeuroLab project is the winner of the 2018 PCBWay/ATMakers Assistive Technology Award.  The PCB design is targetted towards one goal: read EEG signals from the forehead area of a user and route them to a processing device (computer, tablet, etc.) which can perform actions based on the sensor data.

While there are other EEG sensors in the Open Souce world, this one aims to be simple to use and low-cost while still producing results that are useful in assistive technology settings such as switch activation, device control, etc.

The project is relatively well documented (among submissions) and is supported by the FOSSAsia group on GitHub making it likely that it will proceed further towards completion.  In addition, it’s goals are reachable: it simply wants to extract EEG data and make it available for processing.  This approach, in comparison to other submissions that were looking to be all-in-one solutions to a problem is refreshing.

The fact that an existing desktop app and a fledgling Android app are available moves this beyond the concept phase and helped the project considerably.

Suggestions/Improvements: This team would benefit greatly from a central repository of information about this project.  The information is split among four GitHub repositories (which don’t link to each other) and there is no central point to find “the big picture”.  Additionally, I could not find any documentation of the SPI protocol used to expose the data from this board (that may be my failure, but it was not readily available).

Runner Up: Cyborg Hand

The Cyborg Hand project is the runner-up of the Best AT PCB Design Award.  This project combines known working technologies and techniques with a focus on real assistive technology goals.  

While its documentation is sparse and its team’s process could be more transparent, it is notable because it recognises a real AT problem: currently available EMG sensors fall into two categories—expensive medical devices with excellent features and hobbiest trinkets that aren’t suited for long-term use.  This project hopes to create a device between those extremes and contains the features necessary to succeed.

In the end, this project is simply too immature to win this contest.  With the limited documentation, it is hard to determine whether this design will solve this problem.  In addition, the lack of an overall goal and vision for the project anywhere I could find made it difficult to judge (let alone for an end-user to actually use).

Suggestions/Improvements

· Separate the sensor and activation portions of the design on to two (possibly connected) boards.  PWM Servo output is likely to interfere with delicate EMG measurement and the sensor portion would be useful in other applications

· Upgrade the microcontroller – today’s modern MCUs (ATMel’s SAMD51, ESP32, NRF52, etc.) provide significant advantages over the older ATMEGA328p used in this design without significant impacts on cost

· Open your development process—move to Github or other platforms that will get more eyes on your project earlier.

· Document your goal & approach somewhere and provide links to that documentation.

Honorable Mentions

Blind Walking Device  and Talking Smart Glasses for the Blind

These two projects attempt to solve the same problem: provide feedback to individuals with low vision or blindness by using sensors to detect nearby objects.  The projects differed in their sensor choices, their board design style (one looks like a pair of glasses), and their approach to notification.

I would recommend that these two groups merge or at least compare notes—this is where Open Source shines.  Both should also work directly with some potential users who are actually blind to see how their approaches would work in terms of usability, social interaction etc.

4 ch. oled wi-fi relay

This project deserves mention because it is the design that is not intended to be an AT project that might have the most significant impact on AT users.

The WiFi Relay board provides 4 power relays controlled by an ESP8266 WiFi breakout board.  This is hardly original (and can be found commercially in “smart power strips”), but the design is remarkable for a few reasons

1. The device can swap out an AC/DC converter (from 120/240V) or a DC/DC step-down module (from 12/24V for car/boat/solar use).  That makes it equally useful in a home-automation setup or on a 24V power-chair.

2. It includes an OLED display for tracking the state of the device

3. It is sized perfectly to fit in a standard waterproof enclosure (including room for cable glans)

In short, this is a very nice PCB design that solves many of the problems of smart relays.  It is not targetted to AT, but will be very useful in some AT setups.

In all honestly, the biggest challenge with this project was the lack of English documentation.  (And that can be seen as a failing of the judge).  I would suggest that this developer consider selling this design with a partner or posting its design to public tools like GitHub where a community could support it.

Appendix: Submission Notes

Project Name: Neurolab

URL: https://www.pcbway.com/project/shareproject/NeuroLab___Open_Source_Brain_Wave_Analyzing_Tool.html 

Scoring

Comments

This could be better documented – it’s not clear if there is a relationship with the commercial NeuroLab product – I am reviewing with the assumption that there is none.

However, this could be used as a type of switch or environmental control by those w/SMA etc.

This is a great project, and I think it’s worth watching long-term, it’s just so far out that it’s hard to see how effective this particular PCB design is going to be.

Project Name: Cyborg Hand

URL: https://www.pcbway.com/project/shareproject/cyborg_hand_ver_0_1.html 

Scoring

Comments

This is an excellent idea that needs some refinement.  It includes an EMG sensor, temp sensor, servo drivers, etc. to try to create the electronics needed to make a hand controled by EMG.

More likely we could turn this into an effective EMG sensor for other more traditional AT tools (in place of switches, etc.).  However, their original goal is excellent.  

They say “more info coming”… I would prefer that existed to help us see how possible this is.

Project Name: Blind walking device

URL: https://www.pcbway.com/project/shareproject/W123843ASZ15_Gerber_Contest.html 

Scoring

Comments

Documentation is non-existant

Looks like a tool to notify blind folks when they’re close to objects.  Has near and far sensors, small microcontroller and a vibration motor for notification.

Great idea, but not fleshed out well enough – no idea if it will work.

Project Name: Talking Smart Glass For Blind

URL: https://www.pcbway.com/project/shareproject/Talking_Smart_Glass_For_Blind.html 

Scoring

Comments

Great idea – I have some questions – like what the buttons do, why there are three sensors (are they angled?), how is it powered?

Also, these choice of tech is a bit out of date.  A more modern processor could eliminate many of the components (ItsyBitsy M4  which is < $15) would have room for the sound files, no need for FTDI, etc.

3D Printed housing would likely be a better choice than the sunglass-shaped PCB (although they do look neat)

Project Name: 4 ch. oled wi-fi relay

URL: https://www.pcbway.com/project/shareproject/4_ch__oled_wi_fi_relay.html 

Scoring

Comments

Might be very useful as a WiFi environmental control device.  Lack of documentation (even in Russian) is a big problem.

Not original (many of these exist or are HATs for Pis etc.)

Watching the video shows many nice features (AC/DC use, OLED display, etc.)

Not having ANY docs in English hurts the submission (might be my fault… did we tell them the judge only reads English & is relying on Google Translate for others?)