This is the last day of the design challenge, so time to reflect on the project.
Going into the project I had 3 hardware objectives two software objectives and I also wanted to get a new member participating in design challenges.
All of these objectives were accomplished with varying degrees of adherence to the original vision for the system.
The big accomplishment of working out a successful collaboration was very satisfying. There was so much potential for disaster and there were any number of potential logistical and communication issues, but Glenn was very helpful and made major contributions to the project. It was great to have both of us be able to get the same hardware working in completely different environments. Designing and building a complicated voice-controlled electro-mechanical machine was something new for both of us and a significant challenge by itself. Overlaying the running shoe/automobile networks and transportation logistics definitely made it harder.
I found designing the electro-mechanical system to be quite a learning experience, especially working out how to make 3-D printed parts that would perform well enough to make the system work reliably. The system has 15 3-D printed parts without counting all the parts that didn't make the final cut, all of which adds up to a huge amount of printing time. All these parts had to fit with fairly tight tolerance. One of the major accomplishments was to come up with a design that was simple enough to keep costs affordable. There were many explorations into alternative mechanisms. Even with just 2 servo motors, the cost was pretty high, the 25 jars plus lots of mechanical components - gears, belts, bearings, axles, etc. We also had to buy 2 Raspberry Pi's and Google AIY Kits (since we weren't sponsored) and touch screens. I spent about $90 on plastic for 3-D printing, not to mention I had to procure a 3D printer that was large enough to print the platter.
On the software side, I had a lot of fun with Google assistant once I got it working. Although the process of installing Google Assistant is fairly well documented, it actually involves a lot of steps that must be followed precisely - it isn't like just downloading and running an app.
I am a bit amazed (and happy) that we could actually get a voice-controlled machine working. A year ago I wouldn't have even proposed it as the scope would have been too difficult.
Together we produced 20 videos and 22 blogs on this project - I would like to make more but time has run out.
I enjoyed reading all the interesting work posted by other challengers and hope they got as much out of the experience as I did. This challenge has been quite different from most previous design challenges, but a lot of fun and a great learning experience.
Design Challenge Links: