PROACTIVE ANOMALY DETECTION CRUCIAL FOR SOFTWARE SECURITY IN AUTOMOTIVE INDUSTRY

• More than one-fifth of industry experts expect software sales to account for at least ten percent of carmakers' sales as early as 2027
• More than half (53%) of respondents expect setback for electric offensive due to semiconductor shortage
• 45% of respondents expect each connected vehicle to receive up to six OTA updates per year from 2025 onwards

Aurora Labs, provider of self-healing software for connected, autonomous, and electric vehicles announces the results of its annual Automotive Software Survey conducted together with the leading market research and consulting firm Strategy Analytics. For the survey, 140 international experts from the automotive and supplier industry as well as the software sector were surveyed. The participants work for European, American and Asian organizations.

Registration of electric vehicles is on the rise. However, the mass adoption of electric vehicles is threatened with a setback. 53% of the industry experts surveyed by Strategy Analytics and Aurora Labs expect the current semiconductor shortage to delay electric vehicle programs by at least six months. 15% of respondents even expect delays of one to two years.

INDUSTRY IN TRANSITION: CARMAKERS BECOME SOFTWARE-BASED ORGANIZATIONS

The Automotive Software Survey 2021 also shows how business models in the automotive industry are changing. 56% of the respondents assume that car manufacturers will already generate more than five percent of their turnover in 2027 through the sale of software that is rolled-out to vehicles via OTA updates. 21% of the experts even assume that software sales will already account for more than ten percent of carmakers' turnover in 2027. Car manufacturers must therefore transform themselves into software-based organizations considering the increasing revenue potential in the sale of software functions.

45% of respondents believe that each connected vehicle will receive two to six OTA updates in 2025. 15% of the participants even expect up to 24 OTA updates per year. The experts thus expect a significant increase in annual wireless software updates. This illustrates the increasing demand for cost-effective solutions for OTA updates in the automotive industry. As a recent analysis by Aurora Labs and the consultancy Guidehouse shows, automotive manufacturers can save nearly $100 million annually in data transmission costs by using smaller delta files for updates. In addition to the cost of OTA updates, security and redundancy of the solution are most important to 34% of the experts surveyed, as well as a seamless user experience by performing updates without downtime for 26%.

PROACTIVE ANOMALY DETECTION ENSURES SOFTWARE QUALITY

66% of the industry experts surveyed expect that by 2027 at the latest, more than one million vehicles per year will be produced worldwide that have powerful E/E architectures. This shows how relevant the UNECE's WP.29 requirements are for regulating software updates to ensure vehicle safety. The catalogue of regulations stipulates in which cases vehicles need a new type approval after receiving a software update. According to the survey by Strategy Analytics and Aurora Labs, only 14% of experts currently assume that the automotive industry will be able to meet the WP.29 requirements as early as 2022. 37% are in favor of postponing the implementation of the regulations until 2024.

Aurora Labs and Strategy Analytics also wanted to find out in their survey how software quality management is faring in the automotive industry. 47% of the experts surveyed said that it is difficult to understand the mutual relationships between vehicle control units. This can be remedied by solutions that use artificial intelligence and machine learning to understand the interdependencies between software systems and thus proactively detect software anomalies. Industry experts agree that these technologies are urgently needed: 93% of respondents believe that it is important or very important to proactively fix software anomalies after vehicle delivery, rather than just reacting to them to avoid recalls.

To read the full 2021 Automotive Software Survey report, please click here

To read the 2020 Automotive Software Survey report, please click here

The 2020 Automotive Software Survey Report, written by Ian Riches from Strategy Analytics and Roger Ordman from Aurora Labs, is now online!

Our 2020 Automotive Software Survey shows more evidence of the transformation happening in the automotive industry. Unique to this report, however, is the focus on two main vectors of transformation: more centralised vehicle architectures and more software developed in-house by the automotive OEM.

There are significant concerns as to how quickly the industry is moving and whether car manufacturers own the required skills. The function of OTA updates is crucial. Cost control and complexity will be essential to ensure that customers stay satisfied with their vehicles throughout their lifecycle.

The key conclusions of the 2020 Automotive Software Survey are as follows:

• There was wide agreement that OEMs will develop more software in-house, but only a lukewarm affirmation that they possessed the required skills to do so.
• This widely agreed trend is at odds with the current status quo, with the most popular answer to how many software suppliers there are to a current vehicle being 'Over 50'.
• Domain-based architectures are coming - but most saw volume deployment in MY2027 or later.

It is clear that the importance of software is growing for the automotive industry and evolving from an enabler for the hardware, to become key differentiating features. The vehicle manufacturers are being challenged to meet the resource requirements for software to become a strategic component of the vehicle and the automotive industry.

• Multiple aspects of ensuring software quality are seen as difficult and/or getting more difficult, with a strong expressed preference for the ability to have insight into the behavior of the software functions during the development process as well as to be able to predict software anomalies.
• There was overwhelming agreement that OTA updates bring far more than just the ability to roll-out bug fixes. This opinion was strengthened by the fact that 83% of respondents envisaged at least two OTA updates per vehicle per year, with one-in-six (17%) predicted more than 24 updates a year - and so we're clearly looking to see more than bug fixes implemented.
• It is clear that the OTA business will be on a very steep growth curve in the coming years. Market needs (a vehicle which continues to improve post-factory), OEM requirements (controlling the spiraling cost of physical recalls) and legislation frameworks are all now aligning.

The industry is acknowledging that software quality is no longer a single stage of the vehicle development process. Software quality is expected to be maintained throughout the vehicle's lifecycle and the vehicle manufacturers are expected to take a more proactive approach towards maintaining vehicle software quality.

To read the full 2020 Automotive Software Survey report with over 20 engaging graphs, please click here

'One Plus One Equals Three' is the title of a 1927 German silent film as well as a book by Dave Trott on creative thinking and creative writing. 'One Plus One Equals Three' is also a phrase that illustrates what is happening within the automotive ecosystem.

Over the last few years and moving forward, automotive companies are partnering with businesses outside of the automotive industry to deliver offerings that bring two disparate technologies together to create a third solution that reinvents business models, generates additional revenue streams and cuts costs.

According to BearingPoint/Beyond, a European, independent management and technology consultancy:

If automotive and transport companies want to thrive in a digital, mobility-enamored world, partner ecosystems must become a vital component of their reinvented business models. This will allow them to put a greater emphasis on customer relationships as the motor for innovation to create compelling new digital services formed through partner ecosystems which not only meet customer needs but drive them forward - while also being harder for competitors to simply copy.

Unexpected partnerships are bringing the automotive industry together with insurance, entertainment and hospitality companies, just to name few. For example:

• Ford has a deal with Liberty Mutual Insurance offering insurance discounts to customers who drive a Ford connected vehicle. On a consumer opt-in basis, Liberty Mutual is able to access data from Ford's connected cars to assess driving habits and decrease auto insurance premiums based on good driving.
• At CES this year, the electric vehicle manufacturer, BYTON, announced a wide range of partnerships bringing fresh content to an in-vehicle 48-inch display. ViacomCBS and Accuweather are two entertainment/information content companies working with BYTON to deliver an offering that enhances the driving experience.
• Uber has a deal with Hilton Worldwide enabling guests to set ride reminders, request vehicles to and from nearby locations as well as explore local scenes via a digital guide powered by Uber within the HHonors loyalty app.

Credit: Center for Automotive Research

For more than 100 years the auto industry was based on one company building a vehicle, one person buying it and one person driving it. Those days are over. One, single company cannot provide the new types of services and solutions desired by today's consumer.

While collaboration is key to the successful disruption of the industry, it is not without challenges. Once companies agree to partner, the complexities to success range from go-to-market strategies, branding, revenue splits, and team coordination. Core to the success of the partnership is its fluidity through planning and execution as well as ensuring a common lexicon, common measures, and common processes to help companies from vastly different industries integrate, communicate and prosper.

The majority of these services and joint mobility offerings require integration of software components into complex, multi-supplier systems. The mindset of fluidity, dynamic environments and consistent measurement and processes is also true for software integration and the dependencies and compatibility challenges that arise with software integration.

From a pure business perspective, the stronger the dependency between colleagues, supply channels, and infrastructure the higher the probability change will cause challenges to the integrated service. Having insight into how these elements interact will help predict and prepare for change assuring a successful launch of the unified offering.

Similarly, the stronger the dependency between functional components and software lines-of-code, the higher the probability change in one component will have a carry-on effect on other functional components in the system. Having insight into how these elements interact will help predict and prepare for change, again, assuring a successful launch of the unified solution.

Aurora Labs is the company leading this new dimension of providing transparency into the relationships and behavior between functional components in complex and dynamic automotive systems. The actionable data created enable the automotive ecosystem to predict, fix and validate software behavior, the foundation of all new mobility systems and services.

A watershed describes an area of land that contains a common set of streams and rivers that all drain into a single larger body of water. History shows that for new technologies to be successfully deployed, there needs to be a watershed moment where everything from powerful chipsets, to advanced networks, standards and interoperating components need to come together at the same time to truly leverage the power of new solutions and change the way people live, work and play.

This reality was recently reported in The Wall Street Journal's Decade of Disruption supplement. In Joanna Stern's article, First the Smartphone Changed then Over a Decade, it Changed Us, she quotes AT&T's former Vice Chairman, Ralph de la Vega, saying "When one piece of technology changes, it's a big deal, but when two or three things change that are complementary at the same, it's really disruptive."

For the mobile phone industry, only when the network technology (3G), operating systems (iOS and Android) and the phones' processing power, battery life and storage capacity reached a tipping point did mobile phones become ubiquitous to our way of life with consumers watching videos, receiving mails, playing, shopping, banking and surfing online all from the computer in their hand. This watershed moment also made over-the-air (OTA) updates truly beneficial for smartphone users. At this point in time, all of the firmware and applications - the entire phone - could be updated and continuously maintained throughout the lifetime of the phone.

Once OTA solutions were ubiquitous within the mobile phone industry, the OTA vendors set their sights on the automotive market. The car became the next device to update. Today, most of the over-the-air updates are keeping the non-critical head unit, infotainment system and telematics control unit updated. With the exception of Tesla, that built their vehicle platform as a software platform from the ground up.

New technologies comprising today's watershed moment for software as the predominant technology in cars -- the technology rivers and streams that are aligning -- are electrification, 5G, computer vision and software solutions based on advanced AI and machine learning. This convergence will ensure that automotive manufacturers can update more than the head unit, infotainment system and telematics control unit. This watershed moment will result in the ability to run software diagnostics and updates for the safety of critical and non-critical components alike throughout the entire vehicle - end-to-end, bumper-to-bumper, hood-to-trunk, bonnet-to-boot.

This watershed moment also offers the opportunity for automotive manufacturers to redesign their entire E/E architecture and adopt new centralized systems that manage in-vehicle software components as micro-services, ensuring flexibility, cost savings, software quality, safety and security.

Smart Summon is a new feature Tesla has installed with an over-the-air (OTA) update as part of its version 10.0 software. Smart Summon enables the car owner to summon his or her car from a parking space within 200 feet. A few days after the update was delivered, a video surfaced of a Tesla Model 3 being summoned from a public parking lot, driving through a stop sign and getting pulled over by the police.

The Tesla owner said to the police officer "I wasn't driving it. The car ran the stop sign. So, no ticket. "The police officer was understandably perplexed. Who should he issue the ticket to? The owner wasn't driving the car when it made the traffic offence? The car? Maybe Elon Musk himself?

In this instance, no harm came to any pedestrians or other cars, however, the question of liability remains. Earlier this year, executives from Ford and Bosch came together with executives from AllState and Nationwide as the two industries - automotive and insurance - work through how to insure connected and autonomous cars and figure out who is liable for what.

While in many cases the hardware and sensors may be to blame for a vehicle malfunction, according to a report from Stout, in 2018, nearly 8 million vehicles in the US were affected by software-based defects. This is a higher total than the previous five years combined and three times more than any previous year. Both the insurance and the automotive industries - which haven't had major changes or new business models in decades - are being disrupted with the ushering in of connected cars, autonomous cars and mobility services. Who is responsible when an autonomous car crashes? The 'driver' in the car? The owner of the vehicle? The vehicle manufacturer? The supplier of the autonomous driving technology?

Too often the question of liability is framed as a 'blame game' by either the regulators or the insurers, to find the entity that can be shouldered with the cost after the fact. However, the focus of liability should not be on pinning the blame but rather on taking proactive responsibility so that crashes do not occur in the first place.

Understanding of automotive software is going to be the difference between those automotive manufacturers that maneuver through and succeed in this liability storm. Software is entering the car from several entities - the automotive manufacturer's software; suppliers' software and software created from the open-source community. The goal of the vehicle manufacturer should be to ensure that there are technologies and solutions in place to enable transparency into which software is actually running in the car and how the software functions are behaving, to be able to understand the behavior of the software, to be able to detect software malfunctions and to be able to fix them before they affect the driving experience leading to a liability situation.

Connected cars, mobility services and self-driving vehicles are driving a dramatic rise in the amount of software in the car and promise a shift in vehicle control away from the driver and toward the vehicle. The same shift may occur in liability for harm caused by vehicle crashes and related events. Automotive manufacturers need to adopt new in-vehicle software management technologies and develop new business models to be prepared to take on the liability responsibility.

Everyone knows someone with a disability, especially a driving disability that makes it unsafe to get behind the wheel. It may be an elderly parent, a friend with special needs, or a family member that has experienced a serious injury.

For years, the automotive industry has promised that connected and autonomous vehicles, as well as new mobility services, can provide more efficient transportation to those without a disability. But in the wake of the increasing adoption of these technologies, the industry also an obligation: To make transport more accessible and realize more independence and chances for the disabled with better access to basic services like healthcare and better access to employment.

The disabled market also is substantial. In the United States, for example, around 53 million adults have a disability, which is around 22 percent of the adult population. About 13 percent of adults have mobility problems and 4.6 percent have vision impairments.

In Germany alone, 7.8 million people live with a severe disability - many of them unable to drive a car. At the same time, 7.5 billion Euros have been spent on patient transport services - more than ever. Safe and secure autonomous vehicles in the form of configurable pods are a chance to enable those 7.8 million people better access to mobility and at the same time reduce the costs for patient transport.

However, there still remains one problem: The new autonomous technology needs to be safeguarded by a secure back-up plan in case the technology fails. Normally, this security back-up would be the driver, who can intervene and take over controls. But people with disability might not be capable of maneuvering the vehicle themselves.

Self-Healing, In-Vehicle Software Management capabilities of connected and autonomous cars that enable the vehicle to self-correct when a system malfunctions are crucial for all passengers - both those with disabilities and those without. Only if the vehicle is able to automatically detect problems and "heal" them, can everyone benefit from the new world of mobility. This is still a few years off however it is our obligation to build systems that are inclusive, and only then will consumers gain the level of trust for this changing world of transportation to become a reality.

With all the excitement around autonomous vehicles lately, you would expect that consumers are ready to buckle up and embrace full autonomy. On the contrary, there is a surprising amount of caution amongst drivers when it comes to self-driving cars.

According to a study conducted by AAA last year, 73 percent of American drivers report that they would be too afraid to ride in a fully autonomous vehicle - and that number is up from 63 percent in 2017. Furthermore, that same study found that 63 percent of U.S. adults reported that they would feel less safe sharing the road with an autonomous vehicle while walking or riding a bicycle.

How can we expect the entire world to adopt autonomous vehicle technology if there are high levels of distrust within the U.S. alone? Trust is a crucial component - and the first step - when it comes to the widespread use of autonomous vehicles. Autonomous driving will only become mainstream if drivers and pedestrians feel that they can fully trust the vehicle to function correctly.

What we can learn from the past - Trust and Technology

It is interesting to look at how elevators transformed the way people get from the first floor to the penthouse and how autonomous vehicles will transform how people get from point A to point B on the highways and back roads.

There are similar "trust" factors regarding both elevator and autonomous vehicle adoption. The elevator technology itself was not the deciding factor for people to adopt elevator usage. The first elevators were installed in England in the 1830s. However, early elevators used rope- and belt-driven systems that would often snap resulting in injury and death.

It wasn't until 1854 - 24 years after the technology was introduced - when Elisha G. Otis introduced a safety brake using a spring action system. At the 1854 World's Fair in New York City's Crystal Palace, Otis rode the elevator intentionally severing the cable. The safety brake stopped the elevator. The brake is considered the key in gaining public confidence in elevators.

We are in a similar situation today with autonomous vehicles. The major automotive manufacturers, major technology companies and hundreds of start-ups are working on technology to make autonomous vehicles a reality. Key to the adoption of the autonomous vehicle will be the technology that keeps the software safe and guarantees a driving experience that people trust.

Self-Healing Software Delivers Trust

This trust in autonomous vehicles starts with drivers understanding and believing that the software in the car works correctly, is secure, and is safe from cyber-attacks. Only then will the masses adopt self-driving vehicles. That means that the software must be functioning to the best of its ability.

The well-documented growth in the amount of automotive software eliminates any question about the importance of automotive software management. Using new technology, including artificial intelligence and machine learning, is required to deliver a Self-Healing Software solution that can detect a software problem, automatically roll-back to a safe software version and efficiently update the software.

Technology changes and advances - the safety mechanisms required for people to adopt technology - are timeless. Learning lessons from 1854, we know that trust is a key element for technology adoption and Self-Healing Software is to the vehicle what the brake was to the elevator.

Earlier this month, SAE International announced that it is joining forces with Ford, General Motors, and Toyota to create the Automated Vehicle Safety Consortium (AVSC). This consortium will look to create industry-wide safety standards for autonomous cars - something the industry certainly needs as the technology within it expands exponentially. The AVSC will also work with other organizations around the globe to develop these industry standards.

This consortium is excellent news for the automotive industry. Assuming that the AVSC recognizes the importance of maintaining the quality of the AV software throughout its lifetime and mandates that OTA updating be part of any AV system, our In-Vehicle Software Management solution could help OEMs in meeting those requirements. The possibilities are endless in terms of how our software can play an important role in what the AVSC is trying to accomplish.

And when we got to thinking about it, we realized that there are two common goals between the AVSC and Aurora Labs that will be crucial in moving this industry forward: safety and trust.

Safety is at the heart of this consortium - just as it is at the heart of our technology. Similar to how AI is at the core of the AV technology, here at Aurora Labs we believe that it is only natural that AI will also be used to autonomously cure vehicles from malfunctioning software, whether malicious or incidental. Self-Healing Software can enable vehicle software to detect anomalies in the software behavior and health and either on command or autonomously recover to its last known secure, certified and functional version without any downtime. This ensures that the AV stays functional at all times, creates a seamless experience for the driver/passengers and ensures the car is safe from any software anomalies.

Not only that, but trust largely underlies the goal of this consortium. In order for the innovation of autonomous vehicles to move forward, our society needs to trust that they will always work as advertised, safely and securely. Having industry-wide safety standards will help build this trust in consumers. Similarly, a goal of our Self-Healing Software is to build up driver trust in self-driving cars. Drivers must have faith that the technology in cars is continuously operating safely and securely before these vehicles will become widely adopted and commonplace in our society.

These shared goals ultimately come down to one focus: the people on the roads, whether they are behind the self-driving wheel, passengers or pedestrians. The AVSC is hoping to set industry-wide standards to make sure drivers have the most optimal, convenient and safest experience possible with self-driving cars - and this naturally also translates to pedestrian safety.

We are in the midst of an extremely innovative and exciting time for self-driving cars. Here at Aurora Labs, we're looking forward to seeing what will come from this consortium and the positive impact it will have to make truly autonomous mobility a reality.

Editor's Note: Anyone born after 1970 should take the time to click on the links pretty funny stuff.

Oscar Wilde said life imitates art. In our industry, technology imitates the imaginations of TV producers from the 60s, 70s and 80s. Maxwell Smart of Get Smart was the first to use a mobile phone - so mobile it was embedded in his shoe. The Jetsons were the first to have a "smart house" and KITT from Knight Rider was the first intelligent car based on the "Knight Industries Two Thousand" (KITT - acronym and all) processor.

Here we are heading into 2019 and expectations are that the number of mobile phones around the globe will pass the 5 billion mark; robots are being used in homes and enterprises, and based on the automotive industry's transition from hardware to software-centric offerings - we are on the cusp of the intelligent car.

AI and Machine Learning will be cornerstone technologies used to develop an intelligent and independent car, allowing autonomous vehicles to conduct and continuously learn how to do human-like activities. The car will soon be driving itself allowing people to do other things during the time they are being transported. Even imagine a pedestrian walking down the street and the autonomous car knows enough to flash the headlights - or blink his/her eyes - to let the person know it is safe to walk across the street.

Also like the human body, imagine this car with software so sophisticated that the software itself detects anomalies, stops bad processes from causing the car to malfunction and seamlessly updates with new features and functionality - fixing itself like the human body heals itself.

Similar to the realization on Get Smart's mobile phone and the Jetson's smart home, Aurora Labs believes that advances in AI and Machine Learning weaved into our Self-Healing Software solution will help make the vision of the intelligent self-healing car a reality.