4 Autonomous Vehicles With a rapid increase in the types, production and sales of battery electric vehicles, the transporta- tion sector has gone through a major transformation over the past decade. This technical growth in vehicle electrification coincided with a radical transformation of society’s need and understand- ing of mobility. The tremendous increase in the use of shared vehicles signifies a trend to move away from personal car ownership to an era where transportation is seen as a service. Self-driving vehicles appear to be on the horizon with companies testing driverless vehicles in geofenced areas.
The future of mobility is likely to involve the convergence of these three technological areas: elec- tric drivetrains, self-driving vehicles and ride sharing services. While significant benefits can be realized from each of these technologies independently, the convergence of these technologies has the potential to improve our current transportation by being cheaper, more convenient, safer, allow higher productivity for workers, be better for the environment and make better use of our infrastructure.
According to an analysis by former GM vice president Lawrence Burns and Columbia University economist Jefferey Sachs, a shared, electric, autonomous transportation service would be more than 70% cheaper for a citizen of a city like Ann Arbor, MI, and about 80% fewer vehicles would be needed than personally owned vehicles to provide the same mobility [1]. Traditional owned vehicles are overbuilt, able to travel at speeds more than 120 mph and carry five or more passengers. With ride sharing services, vehicles could be tailored for a specific trip allowing for most vehicles to be significantly reduced in size and weight.
Not only would such a service provide lower personal costs, but the environmental costs would be significantly reduced as well. Emissions from cars and trucks amount to one fifth of the greenhouse gases released in the United States [2]. With electric drivetrains, the energy required to propel the vehicle can be produced from cleaner sources such as solar, wind, hydro-electric and nuclear power plants. The combination of self-driving taxis and electric vehicles allows for the vehicles to self-charge, a safer task than the autonomous refueling of a gasoline-powered vehicle.
Self-driving vehicles would also eliminate the need for vast expanses of valuable real estate to be covered with asphalt for parking. In addition to the environmental and cost savings of self-driving, electric, shared vehicles, they will also be safer than traditional vehicles. According to the National Highway Traffic Safety Administration, 36,560 people were killed in motor vehicle traffic crashes on US roadways in 2018 [3]. It is expected that the development of self- driving vehicles will greatly reduce this number.
While autonomous vehicles have been discussed almost since the invention of the automobile itself, it was not until the convergence of computational processing, sensor technology and the advances in probabilistic robotics that the possibility of self-driving cars was technically possible. In 2004, the Defense Advanced Research Projects Agency (DARPA) held a competition that would award the first team to autonomously complete a 150-mile course in the Mojave desert a one million dollar prize. None of the 15 teams that qualified for the final race finished the course and no one claimed the prize. The Carnegie Melon Red Team went the farthest completing 7.32 miles.
DARPA held the competition again in 2005 this time awarding a $2 million prize to the Sebastian Thrun-led team with their entry “Stanley” from Stanford University [4]. In total, five vehicles completed the 130- mile race. The methods used in the 2005 race were published here [5]. The successful autonomous navigation of 130 miles of desert road opened a new error of autonomous mobility. The DARPA challenge was again held in 2007 in an urban setting to see if the vehicles could learn to navigate city streets dealing with other cars and objects. A team from Carnegie Mellon and GM known as Tartan Racing claimed the $2 million prize with their entry named “Boss” [6].
The methods used were published here [7]. The DARPA Grand Challenges were significant in that it opened a new era 87