In May of 2012, the first license for self-driven cars were issued to an experimental driverless-technology and modified vehicles that belong to the Google autonomous vehicle program, after successful test drives in the streets of Carson City and Las Vegas, Nevada. Along with the license came red license plates that included the words “autonomous car” and the infinity symbol, as “the best way to represent the ‘car of the future”, according to the director of the Nevada Department of Motor Vehicles
Last September 25, California became the third state in the US to legally allow autonomous vehicles on their roads, following the states of Nevada and Florida.
Some regulatory ground rules are still in effect, though. For now, all self-driving cars in California should have an easily-activated mechanism to switch off the self-driving mode and a licensed human driver who will be in the front seat for all testing, ready to take over if the autonomous systems malfunction
Over the last few years, various cars manufacturers and computer companies have been prepping the American public for a driverless future, and the future might be sooner than we all think. But is the public ready for them?
Autonomous vehicles: What are they and how do they work?
An autonomous car, also known as “robotic”, “driverless” or “self-driving” car is capable of sensing the environment and can navigate on its own. It will know where to turn, when to apply the breaks and will know what speed it should be in. It can recognize go and stop signs. The human can simply choose the destination, but will not be required to perform any mechanical operation or control navigation of the vehicle.
Autonomous cars use technologies such as radar, lidar(or ladar), GPS and computer vision to gather information from the world outside, and myriad of processors to take in all these information. Such vehicles can identify appropriate navigation paths, obstacles and relevant road signages with the use of advanced control system.
Traditional radar sensors on the front of the vehicle and a position sensor attached to one of the rear wheels help locate the car’s position on the map.
LIDAR (Light Detection And Ranging) or laser radar, usually placed at the top of the vehicle, scans the vehicles’ surroundings in a 100 meter radius, using combinations of optical technologies such as ultraviolet, visible, near-infrared light and pulses letting the car know what objects to avoid. It can be used with a wide range of target including non-metalic objects, rocks, rain, chemical compounds, clouds and even single molecules.
Optical video cameras are used to identify road markings and traffic signals.
A global positioning system keeps the vehicle on its intended route with the accuracy of 30 centimeters.
Multiple core processors, RAM and artificial intelligence are needed to make sense of the combined input from video cameras, LIDAR and radar sensors.
Maps used by autonomous cars are updated based on sensory input, which makes them able to navigate even uncharted environment.
An emerging type of network called Vehicular Communication Systems is being developed in which vehicles and roadside units are the communicating nodes themselves, providing each other with information. This will allow all cooperating vehicles to be more effective by using obtained information from other vehicles in the vicinity, such as traffic congestion, conditions and safety hazards.
What are the advantages of autonomous cars compared to manually driven-cars?
Autonomous cars are not yet in widespread use, but their introduction could produce several advantages:
- Reduced accidents due to fewer traffic collisions because of the autonomous system’s increased reliability compared to human drivers. The tools that the car uses to drive, including radar and GPS tracking, help manage traffic flow by maintaining a constant speed.
- Increased roadway capacity and reduced traffic congestion due to less safety gaps between cars.
- Less stress and more relief of vehicle occupants because of reduced driving and navigation chores.
- It would not matter if the occupants were too young, too old or if their frame of mind were not suitable to drive a traditional car. Disabilities, mental or physical would no longer matter.
- Lesser needs for parking space as cars could just drop off passengers, park where there’s ample space, or just drive back home and return as needed to pick up passengers.
- Reduction in the need for traffic police and reduction of physical road signage. Autonomous cars could receive necessary communication electronically (although the physical signs would still be required for any human drivers).
- Autonomous cars are more convenient for car-sharing services which results to reduced fuel consumption and optimal energy usage.
Will they make the streets safer?
According to World Health Organizations (WHO), road accidents annually cause approximately 1.2 million deaths worldwide, with one fourth of all deaths caused by injury. A comprehensive study of road safety found that human error was the sole cause in 57% of all vehicular accidents and was a contributing factor in over 90% of them. A big contrast to only 2.4% caused by mechanical fault and 4.7% caused by environmental factors.
Currently, new cars have the cruise control or the autocruise feature, a system that automatically controls the speed of the vehicle and takes over the throttle of the car to maintain a steady speed as set by the driver. The system aims to improve the driver’s comfort in steady traffic conditions, long drives and sparsely populated roads, but not applicable for congested traffic conditions, where the car’s speed can have a wide varying range. Most cruise control systems do not allow the use of cruise control below a certain speed. Drivers can control the speed of the car with their fingertips, but would still need to control the steering of the wheel and the braking.
Cruise control can cause accidents if used improperly or in hazardous road conditions such as city streets, heavy traffic, hills, winding roads, and wet, slippery roads which autonomous cars can navigate.
Among the Google fleet, only two have been reported and both have been reported as caused not by mechanical or programming failure. The first accident involved a car crash near the Google Headquarters in Mountain View, California, where the car was being driven manually at the time of the accident. The second altercation involved the Google driverless car being rear-ended while stopped at a stoplight.
In August 2012, the Google Driverless Car project have reported that they completed over 480,000 autonomous-driving miles accident-free with about a dozen cars on the road at any given time, and are starting to test them with single drivers instead of in pairs.
Are we ready to trust machines to drive?
An online survey of 2,006 consumers conducted by Accenture in the US and the UK says that 49 percent of the responders said they would be comfortable using a “driverless car”. According to another survey made by J.D. Power and Associates with 17,400 vehicle owners, more than a third (37 percent) of all survey responders initially said they would be interested in purchasing a fully-autonomous car but the number has dropped to 20 percent once they learned that such technology would cost an additional $3,000.
It seems like most of us are ready to embrace what technology has to offer, it may be the high cost and widespread access to technology that remains a hurdle.