Article by : Wayne D. Cottrell, Ph.D. California State University, Long Beach.
Excitement, discussions, and progress continue with and around the rapidly developing technology of autonomous and self-driving cars. It is generally accepted that the notion of the automated car has been around since the 1920s, trials were conducted as early as the 1950s, and the first truly automated cars appeared during the 1980s. In the 30-odd years since the first automated cars were introduced, the idea has expanded greatly, to include experimental operation in mixed traffic, and the potential for mass production. In 2012, a Google driverless car successfully completed an urban trip chain, carrying a passenger who described himself as legally blind. Google’s cars had logged over one million miles on U.S. roads as of this writing. Technological improvements were still needed in inclement weather operation, obeyance of temporary traffic signals, complex intersections, object recognition, road surface imperfections, and safety officer signals. Experts within the Institute of Electrical and Electronics Engineers have forecasted that 75% of the world’s vehicles will be autonomous by 2040.
A legally blind man prepares to take a spin in a Google self-driving car
While great strides and giant leaps are being made with driverless cars, a brief review of automation in transportation may be informative and useful. Automated processing in transportation is now commonplace, of course. Trip planning, ticket purchasing, travel reservations, seat selection, baggage payment, and other transportation components can all be managed using automated systems. Automation in propulsion and guidance, and the driverless transport of cargo and people, was demonstrated nearly a century ago in some modes, but is relatively novel in others. Elevators, for example, were manually positioned at stops (i.e., floors in buildings) until well into the 20th century. Although elevators have not been “piloted” since the days of hand-powered mechanisms, human intervention for precise door operations was needed until electromechanical circuits with relay logic were developed to better control positioning. These circuits first appeared in elevator systems in the 1930s. Today, most elevators are on-call and operator-free. Elevator operators are still seen in some department stores, subway stations, and amusement settings in Japan, the U.S., and elsewhere, however.
Automation in railroads – driverless trains – was first implemented in London in 1967, on the London Underground’s Victoria Line. The level of automation of the line is “Grade 2,” with automated operation between stations, but with an in-train driver who is responsible for controlling the doors at stops, detecting obstacles along the tracks, and emergency situations. Grade 4 automation, in which on-board personnel are for customer service only, was first implemented in an airport setting during the late 1960s (Tampa, Florida), in a university campus setting during the 1970s (Morgantown, West Virginia), in an urban downtown during the 1980s (Miami and Jacksonville, Florida; later Detroit, Michigan), and in a metro-subway system during the 1990s (Paris, France). Regarding the latter, European and Asian systems have taken the lead in using Grade 4 driverless trains in high-capacity operations. In the U.S., no Grade 4 high-capacity trains are in operation, although San Francisco’s BART system has had built-in Grade 4 capability since the 1970s. Further, some airport automated people-movers, such as Hartsfield International Airport’s (Atlanta) “Plane Train”, which carries over 60 million riders annually, are nearly high-capacity systems.
San Francisco’s BART trains have Grade 4 automation capability, but nonetheless have drivers
Predating elevators and trains, aviation may have been the first mode of transportation to implement automation widely. In 1929, James Doolittle, with his cockpit covered with a canvas to blind his vision, took off, flew and landed using instruments only. It was a 15-minute flight in the New York area, and into history. Further developments led to ground-based instrumentation, with the first fully automated landing of an aircraft occurring in 1964. Today, aviation regulations pertain to two types of operation: IFR (instrument flight rules) and VFR (visual flight rules). The VFR apply when an aircraft can be operated by visual cues only, such as during the daytime, under cloudless skies. The IFR apply otherwise, and for commercial aircraft, apply at all times, even if the plane can be operated safely under VFR.
James Doolittle, under the canvas, and his copilot, get ready for the first “blind” flight (1929)
In this brief examination of automation in elevators, railroads, and aviation, there are plenty of lessons in development, policy and progress which may help guide the automation of other modes, such as motor vehicles.