By Jeff Sanford
Toronto, Ontario — January 29, 2017 — News continues to flow from the emerging world of autonomous vehicles (AVs). This week we look into a report by KPMG on the revenue potential of AVs, the nomadic communities of the future, how cars may become the next big weapon for black hat hackers and much, much more!
Watch for a fresh report on the latest developments in the AV world every Monday on collisionrepairmag.com.
– Consultancy KPMG put out a report this week explaining how connectivity, big data, autonomous vehicles and artificial intelligence are creating new economic models for automakers. According to the report the revenue potential around AVs is “tremendous.”
A new form of consumer value is emerging in which OEMs and tech firms will leverage driver and vehicle data to create new revenue streams.
According to a survey conducted by KPMG a single connected car will generate, “… more revenue streams than 10 conventional cars.” Expectations for data-driven revenue are so great that 71 percent of the auto industry executives polled agree with the statement that, “… measuring OEM market share based on units sold is ‘outdated’.” As the muscle of Detroit merges with the technology of Silicon Valley, “The game has changed for automakers as cars have evolved into rolling computers and consumers have been quick to embrace autonomy, connectivity and mobility-on-demand. A car is no longer defined by its utility, it is defined by the experience it provides to the driver and passenger – and that opens a tremendous pipeline for new revenue streams and business services that KPMG projects could top $1 trillion in the next decade or so …,” according to the report. The poll also found that 83 percent of those contacted believe they will, “… make money off of that data.”
Also in the survey: Consumers overwhelmingly believe they should own their vehicle data. The full report can be found at home.kpmg.com/xx/en/home/insights/2017/01/global-automotive-executive-survey-2017.html
– A project released by a San Francisco design firm has conceptualized a fascinating AV future in which some people will live a new nomadic life, existing permanently on wheels. Phones become permanent addresses. Self-driving cars hitch onto configurable moving platforms. These platforms join up to create always-moving communities. One-wheeled bots provide mobile services to the constantly moving tribe. The design firm used a phrase, Autonomics, to describe this new economy.
“Autonomics is a strategic insight into how autonomous objects will influence not just our transportation, but future businesses, services and brands,” reads a press release. “In this framework, we see the largest impact of autonomous vehicles in suburban and rural areas, where the majority of American’s daily activity takes place, not dense city centres.”
The design project goes on to describe a future in which one-wheeled LeechBots zip among vehicles as they travel along highways. The bots dock temporarily to the moving vehicles to deliver products. ZoomRooms are configurable platforms that act like mobile malls. Autonomous vehicles attach and ride along. A temporary pop-up community, DetourCity, made up of ZoomRooms and LeechBots, emerges.
“One possibility is that we’ll have moving, crawling communities along the highways,” says the founder and principal of NewDealDesign, Gadi Amit. The designer outlined the new world in an interview with Fast Company magazine. A write-up on the project notes that, “The whole concept hinges on using autonomous electric vehicles that don’t currently exist, and a willingness on the part of Americans to ditch real estate for life in transit.”
– Wired magazine ran a requisite “Big Stories for 2017” piece recently. One of the predictions, “Connected cars will become the next big weapon” for black hat hackers. According to the writer, smart cars offer a, “… grim opportunity.” Instead of locking down a target’s computer and holding the data ransom for money, talented psychotic hackers will be able to take over your car and threaten to drive you into a tree if you don’t transfer funds immediately.
The next round of the dot-com revolution, the AI-AV-Internet of Everything, is going to be a weird era according to the story. Cities will be freed from traffic jams. No more traffic lights–AVs will communicate before they get to the intersection and so will just organize to move through as efficiently as possible. But there will also be new dangers.
The Wired writer makes a prediction: “In 2017, somebody will be assassinated while driving. Hackers will seize control of the car’s critical control systems and steer it into a river, a wall or oncoming traffic. How do I know this? Because it’s now possible – and what proves possible eventually ends up happening in the dark, nasty world of cyber conflict.”
The article goes on to say that an, “… estimated 1.3 million people die every year in automobile accidents … After a fatal automobile accident, there is great haste to clean up, tow away the car and bag the body. It happens more than a million times a year.” Who is going conduct an assassination by hacking a vehicle? “If it’s been done already we don’t know about it – and it was likely to be the work of one of the world’s elite intelligence services. But come 2017, the capabilities once reserved for the world’s most powerful state-based cyber-warriors will have spread to black-hat mercenaries. It usually takes about two years for the best weaponized code to move from government to any entity with enough zeros in their bank account. That means that 2017 is the year when anybody with the motivation, money and connections has a new way to kill. And if they’re sloppy with their work, then we might just learn that it happened,” according to the author. The writer is Alec Ross, an adviser to former US Secretary of State Hillary Clinton.
– Self-driving cars could eventually, “… save the U.S. hundreds of billions of dollars because of a dramatic reduction in traffic crashes,” according to a new study reported on by The Hill. A GPS tracking company, Global Positioning Specialists, found that the US loses more money on, “… car crashes than any other country in the world.” The nation could save $306 billion a year with AVs. The study calculated the percentage of gross domestic product (GDP) lost to, “… all types of road crashes each year and factored in how much of that would likely be reduced by driverless car technology.
Self-driving cars could cut the 90 percent of crashes caused by human error. If all vehicles in the U.S. were driverless, the country would save a majority of the $340 billion in GDP lost to traffic accidents each year,” the report finds. “The findings underscore the potential economic and safety benefits of autonomous vehicle technology, which is emerging at a time when traffic fatalities are climbing. Road deaths increased to 35,092 fatalities in 2015 — the largest spike in 50 years. But self-driving cars can create their own set of safety issues, and any bumps in the road could deliver a major setback for their widespread adoption,” according to the story.
– The US federal government has introduced regulations for AVs. According to one of the lawmakers involved, “… the core of the law is the legal equality of human driver and computer.” The question of who is liable in an accident is central to the success of autonomous driving – and still blurred. “It is not the driver but the manufacturer who has to be held responsible for accidents caused by the autopilot,” according to the report.
– New Jersey will become the third state after Nevada and Florida to require owners of autonomous vehicles to maintain liability insurance coverage. According to a story on Property Casualty 360, “New Jersey lawmakers have introduced legislation that would, if enacted, require owners of driverless cars to maintain liability insurance coverage…New Jersey could become the third state in the nation, after Nevada and Florida, to require [such coverage].”
– It could take up to 25 seconds for the average driver to take back control of the wheel according to a pair of engineers from the University of Southampton. An experiment saw, “… drivers climb into a driving simulator and measured how long it took them to relinquish and reclaim control from an autonomous AI system. While other research has looked at the transition, the researchers think this is the first study to look at how long it takes for a person to give an autonomous system control. They … found a much wider range in reaction times than other studies on resuming control from an autonomous system …” According to one of the engineers, “Up to 25 seconds is absolutely insane. We kind of had our suspicions, but we didn’t think it would be that significant.”
– The Foretellix Blog discusses the “Synthetic Sensor Input” (SSI) problem, which is an issue for those trying to program AVs. The problem is this: Programmers are finding it hard to synthesize the “… realistic, synchronized streams of sensor inputs,” that come into a computer through the video, LiDAR and radar devices. AVs will use these various sensors and radar systems to pilot the vehicle.”
An issue however, is this: How to stream the data from sensors into the computing process without interrupting the clock process that is at the heart of the computing function? Computers will have to insert data from the sensors into the process of code computing. But interrupting a computer program to insert a stream from multiple sensors is tough. The insertion of the stream necessarily interrupts the clock and verification process that is intrinsic to software. How can a computer verify what is being seen if the clock function is interrupted?
As the story notes, “Even if you have a good random-scenario-creation system which can produce scenarios like, ‘at time X there is a car going in direction Y at velocity Z, and a person crossing, and so on’, it is very hard to transform that into realistic, synthetic, synchronized input streams (e.g. Video+LiDAR+Radar).” Tying together the streams arriving from all the various devices that the car uses to pilot itself is one of the deep issues in AV programming.
