Last week the discussion progressed from a discourse regarding modern vehicular safety and the plethora of new electronic systems dubbed, “crash avoidance technologies”, to discussing the acknowledged potential for autonomous automobiles by car designers and manufacturers to improve transport efficiency and road safety, and redefine the relationship between commercial entities and their fleets (for example, car rental companies in Sydney) and private owners with their own cars. The focus for this weeks’ discussion will be on the new electronic systems developed by BMW to achieve the “autonomous experience”.
BMW has developed an integrated combined systems suite that they have called “ConnectedDrive Connect”. Like Googles’ rendition of the autonomous vehicle, the car can drive with only minimal human input, but unlike Googles’, the vehicle also visually looks no different from any other vehicles on the road, thanks to the range of scanners, sensors, radars and cameras built right into the body of the car itself.
BMW says that its biggest challenge was to develop algorithms that can anticipate spontaneous events that might occur in traffic - "Our main challenge was to develop algorithms that can handle entirely new situations. In principle, the system works on all freeways that we have mapped out beforehand with accuracy," Nico Kaempchen, project manager of Highly Automated Driving at BMW Group Research and Technology says in a demonstration video. So essentially as long as the roads have been mapped previously by the automaker, BMW states that their vehicles can operate autonomously.
With typical German efficiency the vehicle, in BMWs own words, “...is capable of obeying all traffic rules, it keeps going back to the right lane, never overtakes on the right and sticks to every speed limit – even if they are indicated on the sign gantries”. In order to accomplish this rather significant milestone the technology was designed and developed based on how people drive in the real world, the information gathered by driving simulations. The next challenge was incorporating that data into test scenarios; the cars finally being taken into real traffic once they had proved reliable on the test sites. So far BMW says that their cars have so far driven some 5,000km on freeways using the technology. Still, whilst the technology is impressive BMW was keen to stress the driver is still in control and responsible.
BMW believes that the system could be available for production in cars within 10 to 15 years, but given the way this type of technology has been progressing in the last few years it’s more than likely it could be even earlier than that. There are also legal issues of responsibility and the fact that in both examples of autonomous vehicles discussed thus far; Google and BMW, the car is unable to navigate unless the street has been mapped in detail, as any error could lead to accidents or in worst case scenarios, death. In conclusion, while still a way off from being commercially viable, there is little doubt that BMW has developed and produced a software/hardware suite capable of navigating through the complex situations that arise in freeway traffic with only occasional human intervention. The crash avoidance technology and the potential for autonomous vehicles will play an important role in the future makeup of large commercial fleets, such as car rental companies in Sydney, or other particularly chaotic cities in which the majority of accidents occur.