Perhaps no realm of technology has changed as quickly and as drastically in the last century as the computing industry. From the analog algorithmic devices of the early 1900s, to the semi-automatic vacuum-tube designs of the ’50s, to the potent microprocessing power of today’s hardware, computational ability has skyrocketed. Every two years, the ability of a processing chip doubles, rendering the supercomputers of our parents’ generation weaker than smartphones by a factor of millions. Now, this technological revolution has enabled the next great industrial transformation: driverless cars. The current state of the transportation market bears a striking resemblance to the information systems landscape at the start of the Cold War era. In order to get from point A to point B, a typical motor vehicle requires a series of direct instructions from a driver, conveyed efficiently through an automatic transmission, which in turn informs the engine of its responsibilities, similar to how semi-automated computer processors operated a half-century ago. However, much like artificially intelligent supercomputers, the promise of driverless cars lies within the ability to autonomously navigate the labyrinth of split-second decisions necessary to complete a journey, receiving only the destination as an instruction.
Unfortunately, the daunting level of innovation needed to perform a task completely independent of human activity has inspired pessimism in the general public. According to polling conducted by the University of Michigan, most Americans “expressed high levels of concern about riding in self-driving vehicles, security issues related to self-driving vehicles, and self-driving vehicles not performing as well as actual drivers.” Despite this skepticism, the technological potential inherent within autonomous vehicles render them safer, more efficient and — in the coming years — cheaper than manually piloted alternatives.
The future of self-driving cars involves employing the latest in GPS, motion tracking and information sharing technology to create onboard processing centers that control the basic functions of individual automobiles. These localized processing centers then feed into a comprehensive, cloud-based system that analyzes the data from each vehicle, updating the network of autonomous cars below with real-time traffic patterns. In essence, this creates an interconnected grid of mobile computers, allowing each car to know the intent of every other car in its vicinity. When one car brakes, every other car can instantly respond accordingly, modulating their speed to maximize the efficiency of the continuous flow of traffic. Without the impediment of slow reaction times weighing them down, these cars could eliminate the risk of traffic jams, and make smarter and more dispersive navigational decisions.
But perhaps more importantly, the removal of people from the decision-making process of driving vastly improves the safety of travelers. The same reaction speed that would assist these cars in communicating with each other in real time would also prevent many potential accidents. Complex combinations of sensors within the vehicles can detect sudden obstacles and engage in aversive action well before a person would detect the problem. 93 percent of the more than 5 million U.S. car crashes per year are caused by human error, as drivers tend to get distracted by vices ranging from cell phones to alcohol. A system of driverless cars neutralizes this threat and could even mitigate those few crashes not caused by an imperfect driver. Additionally, when a driver encounters a negative situation, that driver alone learns from the experience. However, since every experience of each autonomous car can be added to the collective memory of the system as a whole, self-driving cars can learn at exponential rates, with the failures and successes of one car influencing the future decision making of every other car as well.
Although the technology necessary to control such vehicles may seem futuristic, many cars on the road today have already incorporated some or all of these features into their driving experience. The Tesla Model S includes driver-independent collision-prevention software, while the latest fleet of Google X cars can operate autonomously (although by company policy they are required to host a supervising driving chaperon at all times). As the tide of automated technology engulfs more aspects of the driving experience, the American public will finally begin to overcome their fear of self-transporting vehicles.
Reuven Bank is a freshman enrolled in letters and sciences. He can be reached at rbanksdbk@gmail.com.
