By Frank Palermo
Is the automobile as we know it dying? The auto industry is certainly in major transition. While the last several years have seen record auto sales peaking at 17.5 million last year, the heyday of human driven cars may be over. U.S. auto sales were expected to continue to rise this year, but with increasing dealer inventory and rising discounts and incentives, it’s clear the industry is starting to cool off.
So the auto industry needs a new attraction. Will the emergence of driverless cars be enough to save the auto industry?
Why Do I Need a Car Anyway?
Car sharing services, favored by the millennial generation due to the ability to make a reservation at the tap of a mobile device, are expected to grow significantly in the next two years, further eroding cars sales.
And with baby boomers accounting for fewer new car sales, millennials and the subsequent Generation Z become all that much more vital to a century-old industry facing its first true challenge in the form of shared and autonomous vehicles.
Recently released Census data shows that from 2010 to 2015, the number of American households without a car increased. With more people moving into cities with rapidly expanding transit systems, it’s becoming more and more possible to shrug off a car.
The Game is Changing
Companies like GM, who invested $500 million in Lyft last year and partnered with Uber for their ride sharing services clearly understands the industry is changing. Others like Ford are investing over $1 billion in technology like Argo AI and $150 million in LIDAR sensor manufacturer Velodyne in hopes it will help establish them as a leader in the autonomous space. It’s a new arms race to autonomy.
And there are new players emerging. Nvidia, a company that pioneered high performance video cards during the PC craze has silently had their sights set on a new capability. The company has been focused on expanding into the artificial intelligence (AI) realm, deep learning and automotive technology, specifically driverless cars. Nvidia’s speedy GPUs and machine learning software have unquestionably become the gold standard for building AI applications.
Toyota just announced that it will use Nvidia’s platform to help it bring self-driving cars to people in just a few years. The automaker will use Nvidia’s Drive PX supercomputer, a platform with a robust new processor called Xavier, to power the autonomous driving systems inside its future cars.
It may be that in the future, auto manufacturers really become software companies.
Does Driverless Really Mean No Driver?
It’s hard to imagine sitting in the back seat of a car with no driver. This is the stuff of science fiction, but it is an inevitable reality of our future. To create a consistent definition of what an autonomous car actually is, the Society of Automotive Engineers created a self-driving classification system in 2014.
The system is grouped into 5 “levels” with Level 0 being no self-driving capabilities like most of the cars on the road today. Level 1 provides basic driver assistance functions like cruise control, automatic braking, and active lane control. Level 2 builds on Level 1 by offering simultaneous control of 2 or more systems like steering and speed. The best example is probably Tesla’s Autopilot feature that measures drive attentiveness and intervenes where appropriate. Level 3 introduces a concept of conditional autonomy which while it can control the car, still relies on handing control back to a human should a situation arise where the auto pilot doesn’t work. These levels are all very assistive technology and is nowhere close to handing over the controls.
Things get interesting at Level 4. This is the first truly autonomous driving level that requires no driver interaction. Unlike Level 3, if a failure occurs, the car will stop itself. Level 5 defines a completely autonomous vehicle that is not designed to be driven by humans.
Most claims by automakers focus on Level 4 autonomy. Not many automakers have committed to timeline for Level 5, fully autonomous cars.
When Can I Stop Driving?
Analysts predict the number of cars with various levels of autonomy will grow to a total of 150 million vehicles by 2025.
There’s a lot of uncertainty in the exact numbers and timing, but one thing is for certain, every major car manufacturer has gone on record on the expected availability of autonomous cars.
Scott Keogh, head of Audi America, announced at the CES 2017 that an Audi that really would drive itself would be available by 2020. Mark Fields, Ford’s CEO, announced that the company plans to offer fully self-driving vehicles by 2021. The vehicles, which will come without a steering wheel and pedals, will be targeted to fleets which provide autonomous mobility services. At their annual shareholder meeting, BMW CEO Harald Krueger said that BMW will launch a self-driving electric vehicle, the BMW iNext, in 2021.
Probably the most aggressive outlook came at the Bosch Connected World 2017 in Berlin where Nvidia’s CEO, Jen-Hsun Huang, announced that Nvidia will provide technology enabling Level 4 autonomous driving capabilities by the end of 2018. Level 4 autonomy is where a car can drive on its own without any human intervention.
The reality is that our first experiences may end up coming from ride sharing companies like Uber and Lyft. These platforms will introduce self-driving cars a lot faster reducing the dependency on waiting on consumers to adopt and purchase self-driving vehicles.
It took approximately 50 years for electricity to be adopted by 60 percent of U.S. households, it took cell phones only about ten years and, and smartphones only about five years to reach the same penetration. How long will it take driverless cars? No one’s certain, but once your friend has that latest gadget, you’ll want one too!
Is the Car the Next Mobile Phone?
Much like the evolution of the mobile phone where the most uninteresting thing you do on your phone today is make a call, soon the most uninteresting thing you will do with your car is drive it. The car of the future is fully connected. It will be intelligent enough to monitor its own parts and road safety conditions. Cars will become a central integration point for various technologies, data, and services. They have the potential to become a roaming office, a roaming entertainment center, and even better—a new software platform.
With this evolution, the complexity of the platform increases. Research and development shifts from traditional automotive engineering and manufacturing to software driven innovation. It’s already started as today’s average high-end car has roughly seven times more code than a Boeing 787.
Impact on Infrastructure
As autonomous vehicles hit the roadway, many factors and implications need to be considered. The most obvious is how do autonomous vehicles and human operated co-exist on the roadway? Our current infrastructure is designed with human drivers in mind and while autonomous vehicles are being designed to mimic human activity, maybe that’s the wrong approach. Is it a better approach to designate certain lanes on the highway to isolate autonomous vehicles?
And there are still limitations of autonomous vehicles. For instance, if a construction worker uses hand gestures to tell a car to either go or to stop, no autonomous car today can reliably make the right decision. What about when the sun obscures the color of a traffic light, most car cameras won’t be able to recognize the color of the signal through the glare. What if children are in the street playing and are distracted, we know to slow down, but maybe the computer speeds up. Today’s computers aren’t nearly as skilled at interpreting complex situations like these. Our infrastructure needs to be significantly transformed to support autonomous vehicles.
This isn’t the first time we’ve had to make an infrastructure transition to accommodate a new transportation paradigm. In the 19th century the rise of the railroad gave citizens a faster and safer mode of transportation than the horse. This absolutely transformed society but we had to learn how to behave around trains. Train travel today is far safer than horseback travel. In the future, computer driven cars will be far safer than human driven cars. The future is clear, the auto industry is transforming and as citizens we need to adapt. There’s no room for horseback riders on the roads. SW
Frank Palermo is the EVP, global digital solutions, Virtusa, a global provider of information technology consulting and outsourcing services.
June2017, Software Magazine