Autonomous Vehicles Meet Inhospitable Roadways

It is difficult enough to design, develop and implement safe and secure software, equipment and services when the direction of technologies is well-known but, for autonomous road vehicles, it so much harder with an ever-changing context for self-driving vehicles and the rapid evolution of onboard vehicle technologies. Until recently, automakers and companies building software and equipment for robotic vehicles were anticipating the early rollout of fully-automated driverless cars, trucks and buses, but we are seeing much more conservative projections and substantial downsizing of proposals and plans. It is beginning to dawn on developers of robotic technologies that their systems, as currently being designed, may never become completely reliable and the need for some form of human intervention, either onboard or remote, may continue indefinitely. In this paper, we take a fresh look at how technologies for both vehicles and infrastructure are changing and what their longer-term prospects might be. We suggest that both vehicles and roadway technologies need to evolve in tandem in order to arrive at an optimal mix. We also look at the economics of driverless vehicles and consider whether the original claims for reductions in accidents, manpower and insurance costs and the changes in vehicle ownership will be realized.