Navigant Research Blog

Time for Automakers to Get Real on Vehicle Security

— August 21, 2014

Recently, the annual Black Hat and DefCon computer security conferences took place in Las Vegas, and this week the National Highway Traffic Safety Administration (NHTSA) announced a notice of proposed rulemaking regarding vehicle-to-vehicle (V2V) communications.  Hacking cars was once again one of the hot topics at the two security conferences this year, in part because automakers don’t appear to have done much to improve the security of the vehicles we drive.  Each year researchers announce some newly discovered vulnerability that gets blown out of proportion by the mainstream media.

Fortunately for drivers everywhere, none of the issues discovered so far have actually amounted to anything worthy of concern.  However, as vehicles continue to get increasingly advanced in the coming years, the potential for attackable flaws will only increase.  Automakers are notoriously quiet when it comes to publicly discussing anything that might potentially be deemed a flaw in any of their products, but it’s time to change that attitude when it comes to electronic security.

Calling All Cars

Over the past half-decade, advanced driver assist systems such as adaptive cruise control, automatic parking systems, and lane departure warning and prevention have rapidly migrated down-market from expensive European luxury models to mainstream, high-volume family cars, such as the Toyota Camry and Ford Fusion.  With the addition of just a few extra sensors and a lot more software, these are the building blocks for tomorrow’s fully autonomous vehicles.

One other piece of that puzzle is the V2V communications that the NHTSA would like to mandate.  Along with vehicle-to-infrastructure  communications, cars will be able to send and receive messages that can influence the behavior of the vehicle.  Initially, the plan is to send these alerts only to drivers.  However, it’s only a matter of time before that expands to include autonomous vehicle capabilities like automatic braking or steering to avoid a collision.

Anyone who’s ever worked on software will acknowledge that it’s virtually impossible to write absolutely perfect and bug-free code, and the task gets exponentially more difficult as systems get more complex.  Automakers often like to brag about how many millions of lines of code are in the latest and greatest new vehicle and how many gigabytes of data are processed every second.  They neglect to mention how every additional byte of code means more potential for mistakes or security flaws.

No Such Thing as Bug-Free

Companies with vast software engineering expertise, including Google, Facebook, and Microsoft, have acknowledged that they cannot possibly find every potential issue in their products.  The impact of a Facebook or Google breach can be annoying, and potentially expensive, but not life threatening.

It’s time for automakers to follow suit and acknowledge that despite their best efforts to secure vehicles, the potential does indeed exist for security vulnerabilities.  Tesla Motors started on the right track this year with the hiring of security expert Kristin Paget away from Apple.  The company also sent a team of recruiters to the Black Hat and DefCon conferences to find more talent.

Each automaker should also set up a bounty program similar to those established by the big tech firms, which pay researchers cash rewards for disclosing security vulnerabilities to the companies.  The corporate lawyers might not be crazy about the idea, but with the recent flood of vehicle recalls from General Motors and other manufacturers, the increased focus on safety and quality might actually make this an ideal time to do this.

 

Connected Cars Get Closer

— April 10, 2012

The United States has been working on various safety-related “V2X” applications since 1999, when the FCC set aside spectrum (the 5.9 GHz band) for short-range vehicle to vehicle (V2V) and vehicle to infrastructure (V2I) communications (“V2X” simply includes both of those categories).  After more than a decade of R&D in this area, developers can finally see a pathway to commercial viability, but the finish line is still several years away.  The time it’s taking to get this technology to market a was a cause of some concern at the recent “V2X for Auto Safety and Mobility” conference, in Detroit, in part because Congress has recently considered opening up the 5.9 GHz band to other, unlicensed uses.  It appears that this is partly a reaction to a sense that safety V2V applications haven’t moved as quickly as had been hoped in 1999, so Congress is now eyeing the valuable bandwidth.

Auto OEMs, Tier 1 suppliers, communications companies and the usual smattering of consultants and associations gathered in Detroit to talk about connected vehicles, machine to machine (M2M) and vehicle to infrastructure connectivity.  Or, as one attendee put it, “We are building the Internet for the road.”

Since this event was U.S.-focused, much of the discussion was on safety applications.  This focus separates the United States, to some extent, from the other ITS markets in Europe and in Asia where sustainability and mobility are big drivers.

The US Department of Transportation is helping to drive the technology forward with its Safety Pilot Driver Clinics, which are helping industry test out some of the safety applications.  These are designed to get driver feedback on several automated safety applications in anticipation of a ruling by the National Highway Traffic Safety Administration (NHTSA) in 2013 on passenger car V2V.  The applications being tested include blind spot warnings; systems that tell the driver to slow down in response to a slowing vehicle in front of her; and notification when a car ahead that the driver cannot see is slowing down.  (As a not overly tall woman who has trouble seeing over all the SUVs on the road, this last one strikes me as especially useful.)  While these systems may simply warn the driver, the end goal is clearly to have the vehicle respond without driver intervention.

The trial is designed to find out how drivers respond –do they find the safety measures helpful or off-putting, and do the V2V applications work correctly (for example, eliminating false warnings)?  The results have not yet been announced, but the DOT representative at the V2X show indicated that, so far, drivers have responded favorably and even indicated they might be willing to pay for these applications.  That is a key point, as the general sense among the OEMs in attendance is that consumers won’t pay for safety features.  Indeed the whole question of how to get anyone to pay for V2X ran through the entire show.  A speaker from Battelle discussed one way to get public agencies to pay for V2X: mileage-based user fee programs.  I’ve talked about these in an earlier blog, on ideas to replace the gas tax.  The speaker described Minnesota’s “test drive” of this system, using smart phones installed in the vehicle.  Participating drivers receive invoices based on the number of miles driven and a varying fee schedule.  While this could help spur public agencies to invest in V2X technology, a smartphone cannot be used for most safety-related apps, which require much faster response time than cellular networks can provide.  Ultimately, the safety applications may simply end up as standard equipment if NHTSA decides to require them in its 2013 decision.

 

Blog Articles

Most Recent

By Date

Tags

Clean Transportation, Electric Vehicles, Policy & Regulation, Renewable Energy, Smart Energy Practice, Smart Energy Program, Smart Grid Practice, Smart Transportation Practice, Smart Transportation Program, Utility Innovations

By Author


{"userID":"","pageName":"Vehicle Telematics","path":"\/tag\/vehicle-telematics","date":"11\/1\/2014"}