HiveNet
Self-driving, full self-driving, levels of automation…what does the end-user care what they are called as long as they can use them, right? What is utility to the end-user though? Automakers have been fighting with themselves, technology, data, and regulators to find a way to get their cars to perform self-driving functions with the goal of achieving full automation without human control. Simply tell the car where you want to go and it will travel there from point of origin to destination.
It should be simple enough, but it is not because of the complexities of real-world objects, weather, speed limits, speed needs, road quality, and road marking quality. So companies like Tesla and its lesser-capable competitors have been pouring billions into the development of their self-driving compute capabilities. As of this writing, Tesla is the most capable, functional, and safe of all, but Tesla has so far missed a massive market opportunity because it is focused on its own vehicles. It certainly should focus on improving its own vehicle capabilities, but for Tesla’s vision to truly work, it needs to not only enable self-driving capabilities to its own vehicles, it should begin developing the capabilities for existing competitor vehicles to join in the self-driving network.
Why? Because it is not enough for the Tesla Model S, 3, X, or Y to be connected and aware of each other. These vehicles can see other vehicles and such using their onboard cameras and processors, but what if each self-driving vehicle could share real-time information about their position, speed, velocity, and heading? The FAA mandates that each aircraft use an ADS-B transmitter so all pilots and controllers know position data of each aircraft in their areas of responsibility. This helps pilot avoid near misses (or hits) and helps controllers maintain safety in the skies.
So why don’t we take that approach with cars and trucks? Data safety and security should be a concern, but every human being in North America drives around with an Android or Apple phone in their pockets, on their dashboards, or in their cup holders to help them navigate. Many Americans connect their phones to their cars. What do the terms of use from Tesla OEM competitors tell us? They’ll use your data for their purposes and they’ll like it. Assume two adults and three children are riding in a car on their way to a weekend at Schiltterbahn in New Braunfels, Texas. Mom has a phone, Dad has a phone, and each kid has a tablet. Dad is driving so he’s using his for navigation and probably listening to Rousseau on YouTube Vanced (the Music one). Mom is probably chatting with friends on Snapchat about horses and husbands who complain about traffic too much. The oldest brother is chatting with his girlfriend who he will meet at Schiltterbahn. The middle brother is playing Asphalt 8 – actually – older brother probably is too, and the youngest brother is probably watching Blippi.
So what does this tell us? Data privacy and connectivity are not the real issues for this family. Everyone is connected and everyone is sharing data with the telecom, app maker, Google, Apple, Ford, you name it. But these are entities who provide us with things we like – entertainment and mobility to break the monotony of a daily work/school life and a long drive to their travel destination – the best waterpark in America.
But what is the purpose of self-driving capabilities? Primarily safety. There are a lot of idiot drivers – and some good ones too. But there is always a risk to the safety of that family who is traveling to have some fun. Tesla is doing remarkable work to improve the safety of the vehicles it manufactures, and by proxy the safety of those around them. But Tesla cannot control the safety of those not in Teslas.
But what if there were a simple hack for that? What if we leveraged the capabilities of an OBD-II plugin module like OpenXC, bluetooth, and radio communications so vehicles could not only communicate, but become fully aware of each other all while using the OBD-II and CAN BUS communications to enable self-driving capabilities? If we can map a communication method between our Google or Apple maps navigation path to our vehicles and provide a logic to manage the vehicle’s movement from point A to point B, and add a layer of visibility of the traffic around it, would it not create a hive mind of vehicles? APIs can be a powerful thing if used to solve simple problems. We can solve the problem of having only a portion of the vehicles on the road with intelligence by providing intelligence to the remainder of modern vehicles to create a HiveNet. A hive mind of networked vehicles. Of course, those vehicles that do not have extensive CAN BUS access would not be participants in the self-driving world, but at least we could still see them through their radio emissions, engine emissions, and thermal signatures.
This can be done.