Take your first breath.
As you get acclimated to the diy community for autonomous cars. we figured some quick advice would be helpful.
Point #1: This is greenfield tech. Understand that people just make stuff up.
This is the first, and perhaps, most critical piece of information we can offer anyone trying to get more familiar with the field of autonomous driving. Autonomous driving, like any greenfield area of research, can get messy. Nothing is standardized or bench-marked. Because of this, if you understand that breakthroughs in technology happen in iterative steps and rarely, if ever, in paradigm shifts, you’ll be able to suss out what’s real and what’s not. The key here is to be flexible on receiving new information and take time to research their claims in the context of what other players and researchers in the field are doing. If they make sense and somewhat align, there may be something there. If not, don’t waste your time.
Point#2: The best technology may not win the day.
This is something most people tend to forget when looking at developing technologies. Unlike in sports, the best tech may not always win on the field. Cost, complexity, speed to market, and audience adoption are typically the biggest drivers of successful technology. If something seems very clumsy in operation, even if its the best technology out there, it will lose to a more streamlined counterpart. The audience (or adopters) are typically the biggest factor in determining which technology prevails. If they find something hard to use, they will abandon it for the more streamlined solution — especially if it comes out earlier.
The best advice we can offer is that you take all of the competing technologies and look for the one that offers solid technology in a more streamlined and thought-out package. The key here is simple and efficient technology. It doesn’t have to be the best, most robust solution available. It just needs to work well, be consistent, and not break the bank. If it can hit on all three, it will have a greater chance at success. Don’t be the guy clinging to HD DVD when everyone else has already moved on to Bluray. If you are serious about getting involved in this field, the last thing you want to do is hamstring yourself with obscure tech no one else in the field uses.
Point#3: This $%^& is hard.
New tech is always the most challenging, however the benefit to this is that its breakthroughs are often the most rewarding. As some of you are experiencing now, there are limits to what information is available, leaving many of you scrambling to figure things out on your own. As difficult as this may seem, this is really the key to you understanding how the technology works. Most people learn by doing — sometimes the harder the lesson, the more it sticks.
Some members of the community will already have technical expertise that better adapts to the technology, so they are able to offer meaningful contributions much faster. Don’t pay attention to them, you are not in competition, you did not join up to “SDC Survivor”, relax. If you find someone that is killing it, ask questions. If you don’t understand the information they are giving you, do not, for the love of God and all that is holy, keep asking them to better explain it. Get what they tell you down on paper (or in a Word Doc) and research. Google every single term they use if you are completely at a loss. Do this until you understand what their response means. After you do this and you are confident in knowing what concept they were trying to get across, feel free to reach back out for follow up information if needed. Just keep in mind that this follow up should be in line with furthering the conversation, not restating or reconfirming what was already said.
What you will find is that a lot of technical guys are good at making statements and conclusions, but terrible at teaching. It has to do with how they are wired. They are great with “IF/THEN”, not so much “WHY”. “WHY” is an annoyance. They don’t have time to tell you why — they have better things to do, like making DBCs and going after bounties. This is not a knock on them, they just get very excited and involved in the technical details of what they are doing, that pulling them out to answer a question in layman’s terms completely kills their vibe. This is why people say they are dickish. They aren’t, sometimes they are just fully consumed in what they are doing.
Point#4: This is your proving ground.
When it comes to tech, the brightest stars are usually the ones that burnout the quickest. The key to surviving is to take small, deliberate steps. You don’t need to learn and solve everything all at once. Reflect on your end goal, take stock on what you have currently, and create a series of small milestones to get you there.
Let’s take on one of the trickier bits for people wanting to get involved in the self-driving car community — can bus hacking. What we typically see is a lot of newcomers arrive eager and ready to learn everything and immediately contribute. While this is great, the problem with this is that they haven’t done the appropriate education to prepare them for that effort. They buy all the necessary equipment, they read a few forum or slack posts, and they ask a ton of questions to try and get started. Ultimately, they grow frustrated with their lack of success in decoding the canbus and they quit. This is unfortunate, because even if they only contribute a small amount to the community decoding efforts, it would be a boon for getting more cars on board.
The key here is to do the education first, before you do anything else. In the case of canbus hacking, this would be learning everything you can on canbus messages, their sequence,bytes, bit rates, etc. This is important because when it comes time for you to actually dig into your car, you will have an idea of how to decipher the messages in front of you. Better yet, it will prevent you from over-purchasing on equipment because you will be better educated on what you actually need to get stuff done. Take the time to learn. The car will be there when you are ready. If you take that approach, you may find yourself better equipped to handle more complex projects while also helping others. This is how you move the community further uphill.
For those of you that are still struggling, or would like to learn how to do canbus hacking, stay tuned — tutorials and resources are coming.
Stick with us as we discuss the levels of autonomy. Your next dose is up soon.
Happy Hacking,
-Jay(Latimer)
