I like my job. The university where I work on developing software used for brain research isn't exactly a sweatshop. While pay isn't that great, they match two dollars for retirement for every one you put in. They provided my partner with domestic partner benefits, which we made use of until we could get married. They provide two days vacation per month from day one on the job, as well as true flex hours. My boss, while sometimes demanding, is a good person. If you had asked me three months ago whether I'd be leaving my job, I'd have laughed at the concept. But one thing was to change all of that:
Electric cars.
Read on for more and for a simple thing you can do to help us succeed in our goal of aiding the EV revolution.
Anyone who's read my comments or diaries could tell that I've long been into electric vehicles. For example, I'm early on the waiting list for an Aptera 2e (pictured above). A while back, after encountering lots of fearmongering about "range anxiety", I got to thinking: if one could accurately tell how much energy it was going to take to drive a given route, there would be no such thing as range anxiety.
The problem is that just because a vehicle manufacturer says "100 miles range" doesn't mean it'll always go that far; all sorts of things can affect your range. Just to pick a handful: what route you take, what sort of elevation changes there are, what the temperature is, how windy it is, what direction the wind comes from, what accessories you're running, how fast you're driving, and so forth. In unusually bad conditions, that 100 miles could be as low as 50 miles. Or as high as 150 miles in unusually good conditions. It's not something one can generally have a lot of comfort about.
The way I looked at it, though, it's just a giant physics problem. And the key details, such as altitude, weather forecasts, and so forth, are all available. So I set out to code an electric vehicle state of charge simulator that could solve the problem. It took a long time, but I finally achieved my goal, and produced a prototype application. Recent testing with a Tesla Roadster found that out of seven or so driving cases, all but one were forecast within just a few percent of the actual power consumption of the vehicle. The one that wasn't (about 12% error) was due to Google Maps poorly approximating how much traffic there will be, and stating that it would take 40 minutes to drive a distance that only took 25; the solution to this, obviously, is to make use of actual commercial traffic data rather than Google Maps' approximations.
The prototype is more complicated than the final version that we're planning, but to sum up basic usage: enter a reasonable-length route on the lower left, press enter, load a vehicle in the top center, and click simulate. You'll see the vehicle drive along your route, leaving a color-coded indication of its charge state along the way, while graphs on the right hand side and animated street views give you all the information you could possibly want on the trip. In the final version, there will also be a map of charge stations overlaid on the trip, which allows you to add your own and comment on existing stations.
The prototype isn't perfect, and still has some known bugs (such as running slowly on long routes due to a map altitude loading routine issue), and of course there are still plenty more features we can incorporate to give the user better results (better pack degradation information and so forth; I've been talking some with Martin Eberhard, founder of Tesla Motors). And before we release, we want a simpler, cleaner interface (for versions embedded into a specific vehicle, 90% of the blanks will go away in lieu of sensor data and driver profiling). But it gives you an idea of what we're looking at.
This was initially a labor of love; I've been writing free applications all my life, and this was just the latest of them. But when I introduced it on the Aptera forum, the feedback was overwhelming -- people volunteering how much they'd pay for it (without prompting), people talking about how their grandmother would buy an electric vehicle with something like this available, etc. And I came to the realization that commercialization would allow me to pay for the servers and bandwidth needed to take this large scale, to hire people to add all the new features everybody wants, and to ultimately expand my efforts into other fields to help bring EVs and clean power to the masses. All of the sudden, I found myself contemplating starting one of those abstract "green businesses" that everyone has been talking about.
It didn't remain abstract for long. Market forecasts were all over the place for how many EVs will be hitting the road, but they generally ranged from a couple million to over 30 million worldwide shipped by 2015. That's a huge market, and there's very little competition for my product in this new market. And not only can we help electric vehicles, but the product can also help plug-in hybrids by telling them when to start and stop the generator in order to waste as little gasoline as possible.
By starting a business, I realized I'd be able to pay and treat my employees well, and to provide jobs in this down economy. I formed a company (Celadon Applications, LLC), retained an attorney and filed for a patent, and entered in a business pitch competition, where we placed second out of 27 companies. We're also a finalist in a business plan competition. I have several advisers, including one who has sold multimillion dollar products to all of the Big Three. I raised some money via a Friends and Family round, and was able to start cutting back my hours at the university to focus more on commercialization.
Which brings us to the reason for posting this: a simple way that you can help. Are you considering buying an electric vehicle at some point in the future? One thing that will greatly help us win investors and contracts with OEMs is to be able to show interest in our product. And to that end, we're looking for Letters of Intent from prospective customers. This is not a contract; it simply states that if we can release the product with X specs at Y price for Z vehicle once that vehicle hits the road, you would be likely to purchase it. We're looking at three primary products: a smartphone/web application, a vehicle-embedded version that makes use of sensor data, and a standalone device you can put on your dash, like a Garmin. We're considering a variety of pricing options, ranging from a $35 upfront fee to a $3-5 monthly fee, to a hybrid upfront/monthly cost structure. In the letter, you can indicate what you feel you'd be willing to pay in exchange for the comfort of being able to tell how far your EV will be able to go. It also can equally apply to other powertrains -- hydrogen, natural gas, even gasoline and diesel.
Such letters of intent will go a long way toward helping us raise funds and get our product to by default come with upcoming electric vehicles (which is our primary goal). And we strongly feel that having our product out there can make electric vehicles a more "mainstream" solution. If you're interested, drop a line to my partner (elaine@celadonapps.com); we can either send you a physical copy with a SASE, or email you a PDF which you can print, fill out, scan in, and email back to us. Also feel to comment on and rec this diary so more people can hear about it. If you have any questions or comments, by all means, let me know!