About Vehicular Automation

Automation in vehicles

Vehicular Automation covers support systems for the vehicle operator (smart) or systems to work without any operator (autonomous). Generally, vehicles are categorized as ground vehicles, aircraft, watercraft and submersibles. I’ll focus on ground vehicles like trains and especially cars.

Car in the dark, in the snow
If I’d hit Autopilot, would the car keep me on the road at safe speed? At all time?

The many support systems that my car has built-in awakens the dream of a fully autonomous car. It feels like all sensors and actuators are already in place. What is missing is the software that ties everything together. Especially when we see that software updates can improve the performance of different sensors.

Autonomous cars

Already now with its adaptive cruise control that supports stop-and-go and its steering support, driving in heavy traffic has become a much less stressful experience than it usually is. Imagine then having a car that drives itself all the way from home to work. Such a relaxing experience that would be.

Take that further to a situation where all cars on the road would be autonomous. Would driverless taxis still be as expensive as today? If taxi or car sharing will be cheap enough, is it worth it to own your own car? What will happen to bus services if everyone would use an affordable car sharing service? After all, the advantages of door-to-door service at the time that suits us would be a huge benefit.

Consequences of autonomous traffic

When cars use their distance sensors and don’t have to deal with manual drivers anymore, the much shorter response times would allow to drive much more closely together. This would lead to better use of the roads. If I could just call a car just like a taxi, I wouldn’t have to worry parking anymore. Fewer cars  on the road would be utilized better for the same level of mobility.

All this raises questions about other users of the road: motorbikes, cyclists and pedestrians. Already now we see lots of conflict situations between cars and cyclists (at least in Sweden, compared to Holland). We should separate the infrastructure more for the different kinds of traffic. Essentially the lanes for autonomous cars could be considered as train tracks but with a layer of asphalt instead of rails. These ‘asphalt’ track should then be off-limit for non-autonomous vehicles.

Higher requirements on infrastructure

Thinking of trains, I noticed that the maximum speed for a track is always low enough so that every train can pass safely. That makes sense since a train cannot slow down easily. On the road, the assumption seems to be that at a speed reduction, drivers will just release the pedal and possibly brake a bit. And if an obstacle like a pothole or a way too tight curve appears, the driver quickly adjusts the speed or the course of the car.

One can wonder how much logic we’d need to build into the car so that software can really replace the driver. Do we expect that the car will pick up all visual signals and deal with missed or contradicting signals like human drivers do?

Maybe we should put higher demands on the infrastructure instead. Then we can lower the demand on intelligence in the cars? I’m sure a lot could be gained by proper infrastructure to vehicle communication (i2v) instead of relying on purely visual communication. Also vehicles (vehicle to vehicle communication, v2v) could communicate with each other, like Volvo already started implementing this for some of their cars and specific conditions.

I think this is a very exciting development we are witnessing and I spend plenty of time in my car reflecting on this. Let me share my reflections with you in this blog. See all blog posts categorised as “Vehicular Automation.”

Find all posts about vehicular automation here. And leave your email address to stay updated below.