Hey all, I’m Seth Welday. I’ve been with Seeed for over 2.5 years and I’ve decided I should give blogging a try. I’ve rounded up some interesting articles that I’ve read in the last week, mostly about tech. Some I’ll provide my opinion on while others I’ll just give a brief blurb.
We’ll start with a sexy topic: autonomous vehicles with a side of 3d printing!
I’m sure some of you recall Olli. Olli is a small self-driving shuttle from Local Motors that is being aimed towards college campuses as well as downtown areas with little/no traffic. It’s all-electric, and it’s partially 3D printed to boot. To be honest, I’d forgotten about Olli for the past few years until I saw the article above on TechCrunch.
Well, I guess they have hit a milestone and now have Olli 2.0. What makes Olli different is that it will operate only at low speeds and is essentially ready now. It features the ability for passengers to interact via voice interaction, and being 3D-printed, it can be customized easily for the customer, allowing Olli to fill even the most niche roles.
I loved the quote the CEO gave regarding when we will see autonomous transportation everywhere, “The future is here; it’s just not evenly distributed.”
So what about vehicles which can transport us around the city? The kind companies such as Uber are hoping for? Well, I recently read this article from TechCrunch where Ford is predicting life spans of 4 years for their autonomous vehicles. Oh, and they don’t think electric is the way to go either. Before we grab the pitchforks and storm their social media you should read what the chief of their autonomous vehicle group had to say about it. Essentially he sees the vehicles running near 24/7, and being hybrids. He argues that autonomous vehicles will be fleet-based, largely not owned by individuals.
In this scenario you would want the vehicles running nearly nonstop to generate revenue, as a sitting car is just depreciating value. Also he points out how constantly doing quick charges for the vehicles would degrade the battery life quickly. These points seem valid, though I’d imagine we could still have some electric vehicles that come out during rush hours to help with surplus demand.
The idea of fleets of autonomous vehicles reducing the need and demand for individuals to own a vehicle has long been touted about, but I’ve never connected the dots on how this could negatively impact battery life. I’m fascinated to see how Americans, such as myself, will react to this. I love to drive so it may be hard for me to let go of the idea of driving myself to the store or work.
Also in the autonomous vehicles theme is this neat idea for canal-based transport. Worth a look! If anyone can find out what hardware these things are packing I’d appreciate it.
Here’s the shill: if you want to work with some autonomous navigation systems check out the RPLidar series! We carry many of them on our Bazaar. One is even $99. I’m always astounded by how much the price of lidar has dropped since I graduated from university.
Back into it, I also saw that Alasdair Allan has thrown together a big benchmark for ML on the edge. Definitely worth a look if you’re interested in this area. I would post the test results here; however, that would be a disservice to you as you should read his post to have a better understanding. I’ll let this abridged preface sell it:
Over the last six months, I’ve been looking at machine learning on the edge, publishing a series of articles trying to answer some of the questions that people have been asking… it’s all bit of a sprawling mess and the overall picture is of what’s really happening is rather confusing.
So here’s a great big benchmarking roundup!
Essentially he has been testing and gathering the results himself and posting about them individually in detail. Now he has compiled them together making it significantly easier to digest. If you have any interest in ML on the edge this is well worth a read. Plus Alasdair’s writing style is entertaining – always a plus.
Found this today by Zach Shelby. For those who don’t know Zach was formerly a VP with ARM and helped found and lead the Micro:bit foundation prior to that. Here he essentially provides the state of ML on the edge and where he sees it going. He also pitches his new company, but it largely takes a back seat.
On a non-technical note, I’ve found this article about a woman’s only written language in China, now almost extinct. It’s a fascinating read. Also if you are unfamiliar with the website Atlas Obscura you should give it a look. I rediscover it several times each year, each time resulting in hours of neglected house chores…
That’s all for this week. I’ll try to get these going either each week or biweekly. Let me know what you all think.