Self-Driving Car Trains

As self-driving cars (SDCs) become available and more popular, it will become more obvious that these vehicles have the ability to navigate with superhuman precision. With many SDCs on the road, they will also be able to coordinate with each other for navigation. Though SDCs can utilize all conventional forms of energy for transportation, for this idea, it will be assumed that they are electric.

IDEA: self-driving cars should be able to physically connect to each other to form ‘trains’ that reduces wind resistance and enables the sharing of high-bandwidth sensory information and energy.

When passengers in different SDCs are travelling on the same road in the same direction for more than a certain amount of time (let’s say 5 minutes for simplicity), these SDCs should be able to physically link front-to-back and form an SDC train.  By linking, each SDC will minimize the amount of wind resistance it encounters and thereby substantially increase its efficiency, reaching a higher mileage. In addition, linking will allow for the exchange of high-bandwidth information so that a more accurate environment map could be produced for all vehicles in the train, making the trip potentially safer.

I shared this idea with a friend and he asked “what happens if a deer jumps on the road?”. This is a fantastic question! The answer is: the same thing as what the SDC train would normally do if it was a single SDC. Adding more SDCs does not limit braking power. Though something jumping in front of a conventional train poses an impossible maneuver to avoid crashing into the obstacle, the reason for this inevitable crash isn’t that the train-cars are linked; the train crashes because normal trains are bad at stopping quickly. I will repeat, adding more SDCs to the SDC train will not limit braking power. And, because sensory information is intimately shared among the SDCs within the train, they could all begin braking simultaneously. Even better, the SDC train is capable of detecting the deer more accurately and begin braking even sooner than a loner SDC. Because the sensory information and computation power are shared within the SDC train, there would be more sensors searching for this impending deer attack while the combined processing power allows the driving algorithms to handle the computations faster. Finally, if the best method of crash avoidance is to swerve, [with proper design] the SDC train could maintain ‘the swerve’ throughout the entire length of a train, similar to the way a millipede travels.

It is mutually beneficial for the SDCs in a train to share processing power, wireless bandwidth, sensory information, and therefore this feature should be free and ubiquitous. However, the most transformative transaction which an SDC train can offer would likely not be free: exchanging energy. The concepts of energy transfer and storage can be transformed if each SDC would be able to decide if it wants to buy or sell energy, and the price at which it is willing to buy or sell that energy. By exchanging energy during travel, SDC trains would be able to homogenize energy distribution, acting like a portable power-storing micro-grid.

In this manner, an SDC would be able to operate continuously without the need to ever stop and charge, as long as the owner of that SDC is willing to buy energy from other SDCs in trains. SDCs that decide to stop and charge will likely benefit from lower energy prices while SDCs that choose to charge during travel will likely pay a premium fee. This premium fee can potentially create a previously unforeseen industry: mid-travel energy delivery to trains. Therefore, for those that decide to pay for it, the distance that could be travelled by an SDC would no longer be limited by a single charge.

Self-driving car trains have the potential to make travel more efficient, safer, and to transform the concept of charge. Let’s be Thoughtful about how we roll them out.