Yesterday’s science fiction is becoming reality, with multiple corporations working on their take for what a self-driving car could be. From tech giant Google to the ‘Apple’ of car companies, Tesla, everyone wants a piece of the pie. But how long until these self-driving cars go mainstream?
Some autonomous vehicles have already been sent into the wild, though this first breed of self-drivers isn’t quite as advanced as people would hope. Most people when they hear of self-driving cars, envision a hop-in and go situation where the passengers all sit back and relax during the drive. However, with most current consumer technology, it is more of a small driving enhancement than a self-driving machine. With Tesla’s Autopilot, certain actions can be performed by the vehicle such as accelerating and braking on the highway, and with high-end models, even highway lane changes are enabled. But, this is still nowhere near the promise of what self-driving was supposed to be.
That’s where Google comes in, with Waymo, a self-driving car company based in Phoenix Arizona. They have already brought fully autonomous cars to the streets with their ride-share service and mobile App. The App allows users to call a ride on their phones and have a fully automated driverless vehicle pull up and take them to their destination. The technology created by Waymo will no doubt spread from Phoenix to other places either through licensing the technology to other companies or Waymos own expansion. But for now, it is held back by certain challenges that exist with expanding to new areas.
Self-driving cars use a complex vision system with a mix of cameras and radar technologies to get a clear view of the entire world around them. However, this system relies on certain environmental factors to function properly. For example, in colder areas where the road could be obstructed by snow, it can be very difficult for these systems to differentiate between what is and what is not a road. The vehicles may also have a difficult time accounting for ice that may cause them to slip or move in a way that is not predictable. This is why manufacturers are opting to test full autonomous tech in areas with good visibility, consistently dry weather, and clear paved roads, such as in Phoenix. One key advantage of self-driving cars is their ability to improve through their shared experiences. When one of the autonomous vehicles gets in an accident or even nearly does, the manufacturer will use the data from that event to improve the car’s awareness and help it to make better decisions the next time. This is important because, when you have hundreds of thousands of individual cars learning from each other it creates an exponential improvement curve to where they should, in theory, surpass human drivers who only have their experiences to work off of.
So, given enough time and some improvement to the overall systems to all vision in snowy areas self-driving cars will not only become mainstream but will also allow for a much safer experience. Having only robot drivers would eliminate the potential for human error and traffic violations. With the potential of car-to-car communication, speed limits could be increased as the robot drivers would know exactly what other vehicles are around a corner and share information about construction, accidents, and nearby pedestrian crossings. A world with self-driving cars would be a world where getting from one place to another is easier and more comfortable than ever before.