A week or so ago, I did a piece regarding the headlong attempt to rollout a ‘new’ auto-drive product built by a small group of tech wonks in San Francisco. In my view, the product was not ready for a host of reasons, but mostly based on the potential of unplanned public safety concerns, since the new product would have to be operated as part of a mix of traditional vehicles on highways, in addition to its unrefined characteristics operating within one or more highly-urban environments.
Now, while I am not attempting to beat a dead horse, it appears that I’m not the only person who is looking at the larger downside impacts of a too-early emergence of auto-drive, while at the same time considering resolutions to the problem. In this case, I specifically refer to urban areas, where mixes of pedestrians, hordes of traditionally driven cars and the emergence of auto-drive vehicles could come together to create a perfect roadside storm unless planned for properly.
As I said, however, others are beginning to take a look at the problem, including Mercedes-Benz and Nokia, who have taken upon themselves to begin to work on a series of early technical strategies while looking down the road toward the ultimate emergence of auto-drive as a fait accompli. “While autonomous vehicles may not hit the streets commercially for several years,” said a Nokia spokesman recently, “(…) automakers and tech innovators alike must already think about the infrastructure and technological requirements needed to support this technology.”
Essentially, the two companies are working to develop and integrate big data sensor arrays within auto-drive systems to, as Nokia’s Head Researcher for the company’s HERE project Dr. Jane Mcfarlane suggested, leverage the “wisdom of crowds.” In this system’s model, real-time vehicle-to-cloud-to-vehicle protocols essentially watch, index, catalog and subjectively weigh threats to the vehicle’s environment including things like people moving on the street, the level of traffic congestion and overall climatic conditions in order to produce an active 360 degree arc of ‘smartness.’ These components in turn ‘act’ on the corresponding data, thereby producing an appropriate dynamic response.
Again, slow is better, since we have lots of things to learn yet, before auto-drive can be safely referred to as a primetime product, but we are learning in the larger sense.