A typical cycle of exaggeration around “self-driving” automobiles reached its pinnacle five years ago. Automated driving systems (ADS) were expected to be widely used by 2020, according to practically every major automaker and high-tech business at the time, which would supposedly cause traditional human driving to quickly become obsolete.
Also, with the added benefit of hindsight, it is clear that the widely held belief at the time was incorrect; by last year, only a small number of advanced prototype vehicles had been driven on public roads without the need for onboard safety drivers to step in when the automation systems required human assistance.
Because the “full self-driving” automobiles’ driving aid functions cannot operate without ongoing human supervision and because trucks, buses, and shared-ride vans are now more significant for automation than cars, the phrase “self-driving” has lost its original meaning.
By 2018, the CEOs of the major corporations (Waymo, General Motors, Ford, and Aurora) that had made the largest investments in ADS were beginning to temper their earlier optimism by emphasizing that the rollout of automated driving would be incremental, starting with operations under constrained conditions in strictly restricted locations. It will take decades for them to spread to a level that even comes close to countrywide deployment at the current rate.
Costs and the organizational learning curve have been substantially higher and longer than anticipated. After investing at least a decade and billions of dollars in ADS development, businesses have found that the technical requirements to allow the widespread deployment of the technology are far more challenging than they had first anticipated.
Knowledgeable ADS Developers Are Becoming More in Agreement Regarding Several Important Characteristics of the Technology And Its Impending Use:
- In the upcoming years, automated operations will only be practical under very specific circumstances that include favorable weather, illumination, traffic, and electronically geofenced places that have been highly precisely mapped (and, in many cases, equipped with suitable physical and digital infrastructure support features).
- ADS must contain far greater levels of safety assurance than driving-assistance systems that rely on human supervision because they must avoid all traffic dangers they encounter without human driver intervention.
- The driving environment and its dangers must be learned by ADS from a variety of independent sources. These facts are gathered via cameras, radars, and lidars (light detection and ranging devices), accurate location, and incredibly thorough digital maps.
- The capacity to drive without depending on wireless communications from other cars, warnings from at-risk road users, or even the real roadside infrastructure itself is something that many ADS developers promise, however, a recent study has revealed that widespread
- ADS without such cooperative communication is likely to have negative effects on traffic flow, energy use, and environmental emissions because it is impossible to anticipate future changes in road conditions.
- The technology will first be used for specialized tasks including local package delivery, long-distance haulage on highways, fixed-route urban transport systems, and, in fewer places, automated passenger ride-hailing in cities and suburbs.
- ADS will still require remote assistance from expert drivers even after they can operate cars without an onboard human driver as a backup to address “corner case” situations that the automation cannot handle.
These conclusions are a result of a few fundamental lessons the technology’s creators have learned. The most important of them is the fact that because of the complexity of the driving task, automated systems have difficulty effectively perceiving the driving environment, anticipating the actions of other drivers, and identifying and reacting to traffic dangers.
ADS must learn to crawl before they can walk and to walk before they can run, just like toddlers learning to regulate their motions. Because of this, they must be deployed and refined in straightforward settings before being used for complex interactions with erratic road users (such as pedestrians and cyclists) or operations in inclement weather (such as heavy rain, snow, fog, and icy roads).
To ensure safe operations, the ADS must be able to observe their surroundings and determine their location utilizing technologies based on a variety of fundamental physical concepts. They must contend with unfavorable situations like electromagnetic interference from electrical storms
nearby electrical equipment, low sun angles that can cause cameras to become blind, precipitation or smoke that obstructs the light that imaging sensors need, and cyberattacks that aim at any of the vehicle’s sensor technologies.
Information on the local infrastructure as well as the relative speeds of any surrounding moving objects is also required. To correctly portray the region surrounding the vehicle and to identify any problematic inputs, data from all of these sensors must be combined.