Driving away your freedom
Where the Autonomous Vehicle begins, your free, secure driving days end
"Modern automobiles are pervasively computerized, and hence potentially vulnerable to attack. However, while previous research has shown that the internal networks within some modern cars are insecure, the associated threat model—requiring prior physical access—has justifiably been viewed as unrealistic. Thus, it remains an open question if automobiles can also be susceptible to remote compromise. Our work seeks to put this question to rest by systematically analyzing the external attack surface of a modern automobile. We discover that remote exploitation is feasible via a broad range of attack vectors (including mechanics tools, CD players, Bluetooth and cellular radio), and further, that wireless communications channels allow long distance vehicle control, location tracking, in-cabin audio exfiltration and theft. Finally, we discuss the structural characteristics of the automotive ecosystem that give rise to such problems and highlight the practical challenges in mitigating them." Comprehensive Experimental Analyses of Automotive Attack Surfaces (Stephen Checkoway et al.)
Electronics systems have become critical to the functioning of the modern automobile. Enabled by advances in sensors, microprocessors, software, and networking capabilities, these systems are providing a rich and expanding array of vehicle features and applications for comfort, convenience, efficiency, operating performance, and safety. Almost all functions in today’s automobile are mediated by computer-based electronics systems. Some of these systems have improved on capabilities once provided by mechanical, electromechanical, and hydraulic systems. In many other cases, electronics systems are enabling the introduction of new capabilities, including a growing number of applications intended to assist the driver in avoiding and surviving crashes. The next logical step envisioned a totally self-driven, or autonomous, vehicle, which may completely replace the human controller. Although a very old concept, the more advanced spectrum of artificial-intelligent circuitry provided the impetus current developers needed to convince governments and manufacturers of the technology's promise. However, the interest stems more from curiosity than potential support.
Autonomous systems are best deployed in a highly controlled environment, which demands smart coordination of all vehicles on the streets to traffic lights to even traffic signs. Each vehicle, traffic light, and traffic sign must share their status to one another to effect violation/accident-free safe zones. All vehicles must be uniformly controlled and monitored by a central traffic intelligence system. Essentially, a smart traffic light/sign must initiate start/stop processes via the vehicle's artificial intelligence control. Speed limitations should be governed by monitoring the imbedded GPS speedometer subsystem.
Interconnectivity with devices and networks external to the vehicle will provide the desired functions. System interconnectivity and complexity are destined to grow as the capabilities and performance of electronics hardware, software, and networking continue to expand along with consumer demands for the benefits these interconnected systems confer. Networked electronics systems and software will continue to be the foundation for much of the innovation in automobiles and may lead to fundamental changes in how the responsibilities for driving tasks and vehicle control are shared among the driver, the vehicle, and the infrastructure.
Increased interconnected systems are creating opportunities to improve vehicle safety and reliability as well as demands for addressing new system safety and cybersecurity risks. As systems share sensors and exchange data to expand functionality, an emerging safety assurance challenge is to prevent (a) the unintended coupling of systems that can lead to incorrect information being shared and (b) unauthorized access to or modifications of vehicle control systems, both of which could lead to unintended and unsafe vehicle behaviors. A critical aspect of this challenge is to ensure that the complex software programs managing and integrating these electronics systems perform as expected and avoid unsafe interactions. Another is to ensure that the electronics hardware being embedded throughout the vehicle is compatible with the demanding automotive operating environment, including the electromagnetic environment, which may be changing as electronics devices and accessories are added to automobiles. Inasmuch as many problems in software and electromagnetic interference may leave no physical trace behind, detection and diagnosis of them can be more difficult.
By enabling the introduction of many new vehicle capabilities and changes in familiar driver interfaces, electronics systems are presenting new human factors challenges for system design and vehicle-level integration. Although automotive manufacturers spend much time and effort in designing and testing their systems with users in mind, the creation of new vehicle capabilities may lead to responses by drivers that are not predicted and that may not become evident until a system is in widespread use. Drivers unfamiliar with the new system capabilities and interfaces may respond to or use them in unexpected and potentially unsafe ways. Thus, human factors expertise, which has always been important in vehicle design and development, is likely to become even more so in designing electronics systems that perform and are used safely.
IT is enabling nearly all vehicles to be equipped with EDRs that store information on collision-related parameters, as well as enabling other embedded systems that monitor the status of safety-critical electronics, identify and diagnose abnormalities and defects, and activate predefined corrective responses when a hazardous condition is detected. Access to data logged in EDRs can aid crash investigators, while diagnostics systems can facilitate vehicle repair and servicing and inform automotive manufacturers about possible system design, engineering, and production issues. Continued advances in electronics technology and their proliferation in vehicles can be expected both to necessitate and to enable more applications for monitoring state of health, performing self-diagnostics, implementing fail-safe strategies, and logging critical data in the event of crashes and unusual system and vehicle behaviors.
Even with the increasing resources provided by local artificial intelligence, the driver remains the critical determinant of safe performance. Driver actions and inactions contribute to the majority of crashes and are most often labeled as the proximate causes. The label of driver error, however, can obscure the role that vehicle designs can play in crash causation if insufficient consideration is given to human capabilities and limits. The new capabilities of vehicle electronics promise to eliminate or mitigate some driver errors, but they risk introducing new ones if drivers are not properly considered as integral to the vehicle system. The field of human factors engineering provides various standards, guidelines, and test procedures to aid in the design of systems that are less likely to induce driver errors. These practices apply to the physical layout of the vehicle to ensure that drivers can see, reach, and operate vehicle controls. For example, human factors practices guide the placement, width, and length of the brake and accelerator pedals to minimize pedal misapplication. Human factors practices also apply to the design of dashboard warning lights and control levers and buttons to ensure that drivers can easily interpret information and control critical vehicle systems. Traditional safety analysis tools such as failure mode and effects analyses (discussed below) help ensure that design choices are consistent with driver expectations and response tendencies. Increasingly, automotive manufacturers apply techniques that have been developed to make other consumer products user-friendly, such as user-centered requirements generation and usability testing. Their applicability is growing as vehicle electronics assume greater control of the vehicle through such features as adaptive cruise control, collision warning systems, lane-keeping aids, and automated braking systems. These and similar “mixed initiative” systems could cause the driver to misunderstand and be startled by the electronics even when the system is operating as designed.
Low-level sensors used in autonomous cars cannot distinguish a heavy snowfall from a snow-covered vehicle. Therefore, the vehicle's computer only 'sees' a blank white background.
A major challenge for system designers is in understanding the long-term adaptation of the driver to the electronics and the degree to which the driver will assume that the vehicle is capable of certain control functions. For example, drivers might begin to believe that the vehicle carries out some control functions in a way that is inconsistent with the designers’ intent. Advances in driving simulators and instrumented vehicles are thus being developed to give human factors engineers new tools to assess and model how the driver and automotive electronics will interact. In this sense, automotive vehicles exemplify the mass adoption of the assisting or operating “robot,” partnering with humans to ease or even take over the human workload.
Although onboard artificial intelligence systems are vulnerable to cyber attack, self-driving cars will be irresistible to hackers. Researchers have already hacked a normal car, so how what are the risks once self-driving cars reach our roads? [Interior of a steering wheelless Rolls Royce concept. The Guadian-2016]
Speaking exclusively to the Guardian, Eddie Schwartz, the vice president of global security solutions for Verizon's enterprise subsidiary, said that the cyber-security industry is still 40 years from maturity, and that the first half of the 21st century will see the number of targets increase exponentially.
"All of the major automobile manufacturers are working on self-driving cars," Schwartz explained. "For cars to be able to self-drive, they have to be able to negotiate with each other. You can't negotiate something like that without having some security principles behind it. So cars have to do basic things that we do with each other, like recognise each other – authentication.
"OK, I authenticate to you, that means there has to be an underlying artefact, a certificate or something like it that says 'you're an authorised car, and I'm an authorized car, therefore we can exchange this information really fast.' And you stop and I turn.
Schwartz described "a million applications" in the car industry alone designed for machine-to-machine communications with potentially a million underlying security issues.
Even normal cars are susceptible to hacking attacks. In August, a pair of researchers demonstrated attacks on a Ford SUV and Toyota Prius which enabled them to slam on the brakes, jerk the steering wheel, or accelerate the car using a laptop plugged into the the diagnostics port.
In 2011, a different team of researchers managed to penetrate similar systems through bluetooth, mobile data and even a malicious audio file burned onto a CD played in the car's media player.
But self-driving cars have many more avenues of communication with the outside world, and – definitionally – less oversight from a driver to correct any errors.
A ransom for your medical data? As well as self-driving cars, Eddie Schwartz cautioned that the entire field of machine-to-machine communications, also known as "the internet of things", presents an enticing target to hackers.
"How many IP-based [internet connected] devices does the average person have in their home today? Most people can't even count them. If you ask them, they would probably say 'oh, I have two computers and a whatever', but the reality is it's probably more like 20 to 30 if they start thinking about it… You're going to see a spill from 4 or 5bn IP devices to hundreds of billions over the next 10 years."
Schwartz cautioned that with the growth of new devices and services in the health space the potential for malicious hacks will grow exponentially, including devices that gather intimate personal medical data.
"These are going to be embedded solutions. It's going to be wireless communications or NFC. These are machine-to-machine communications, and for critical care, they are going to have telemetry going on 24/7.
The fully connected autonomous car will likely require a service provider for internet and GPS connectivity. By not paying your internet/cellphone bill in an IoT environment, it is likely that your IP can deny your driving priviledges.
"There's an underlying security and privacy issue: imagine ransom-ware [software such as Cryptolocker that breaks devices and demands a fee to fix them] in that world." Experiments have been conducted by researchers at the University of Washington and Mr. Checkoway and associates at the University of California, San Diego, to examine cybersecurity vulnerabilities in modern automobiles. They have demonstrated how individuals with sufficient skill and malicious intent could access and compromise in-vehicle networks and computer control units, including those controlling safety-critical capabilities such as braking, exterior lighting, and engine operations. In the laboratory and in road tests, the researchers first demonstrated the ability to bridge internal networks and bypass what the researchers described as “rudimentary” network security protections to gain control over a number of automotive functions and ignore or override driver input, including disabling the brakes, shutting off the engine, and turning off all lights. To do so, they extracted and reverse-engineered vehicle firmware to create messages that could be sent on the CAN through the OBD port to take control of these systems. This included the insertion of code in the control units to bridge across multiple CAN buses. In follow-up experiments, the researchers examined all external attack surfaces in the vehicle to demonstrate and assess the possibility of remote access to cause similar outcomes. The experiments indicated that such exploitation can occur through multiple avenues, including those requiring physical access to the vehicle (e.g., mechanics’ tools, CD players) and those using remote means such as cell phones, other shortrange wireless devices, and tire pressure monitoring systems.
Your IP can determine the mileage allowable by your autonomous vehicle according to your contract. The higher the payment, the greater mileage the vehicle will be granted.
The committee was briefed by the researchers, who described in more detail the many possible means by which an adversary could attack a vehicle in the manner outlined above and the implications for the safe operation of a vehicle. In the briefing and published papers cited, the researchers surmise that automotive manufacturers have designed their networks without giving sufficient attention to such cybersecurity vulnerabilities because automobiles have not faced adversarial pressures (unlike PCs connected to the Internet) and because of the incremental nature by which these networks have been expanded, interconnected, and opened to external communication channels. Recognizing that high levels of interconnectedness among vehicle control units are necessary for desired functionality, the researchers did not propose the creation of physically isolated networks. Instead, they proposed the hardening of remote interfaces and the underlying code platform, greater use of antiexploitation mitigations used elsewhere, and the use of secure (authenticated and reliable) software updates as part of automotive component design. The committee notes that although the researchers did not give specific examples of a vehicle having been compromised by such an external attack, cyberattacks in the field have been reported. One such incident, in early 2010, involved a former employee of an automotive dealership alleged to have remotely hacked into systems that had been installed in purchased vehicles to track their whereabouts and gain access to them in the event of a bank repossession. About 100 private vehicles were targeted; their starters and GPS were deactivated and their horns were triggered. Many of the owners were stranded and incurred towing expenses, according to media reports. Obviously, had such an attack compromised a vehicle’s power train, braking, and other operating systems while being driven, the consequences could have been much more severe.
···Ebook update: Learn more about IoT, artificial intelligence, and cloud computing and their collective impact on society
https://www.amazon.com/-/e/B01HA82EA8?ref_=pe_1724030_132998070
Now available for download