≡ Menu

Fiat Chrysler Automobiles is recalling 1.4 million vehicles to update software to prevent some models from being hacked by two expert hackers with experience finding holes in the security of a variety of systems.

Charlie Miller is a security expert working for Twitter. He’s famous for hacking iPhones and Apple products, finding holes in iOS and Android OS. Chris Valasek is the Director of Vehicle Security Research at IOActive. The two of them teamed up and were able to take control of much of the car’s systems over the internet, thanks to the connectivity provided in the infotainment head unit.

In the video from Wired, the two take over the Chrysler SUV’s brakes, climate controls, display, wipers, more and even shut down the engine. They can also track the vehicle via GPS. Without ever being near the car, all over the internet.

Fiat Chrysler Automobiles is issuing a security update patch, though they note that they have not seen any instances of hacking beyond this one. They also claim that unauthorized remote manipulation of a vehicle is a criminal act. The vehicles affected are:

• 2013-2015 MY Dodge Viper specialty vehicles
• 2013-2015 Ram 1500, 2500 and 3500 pickups
• 2013-2015 Ram 3500, 4500, 5500 Chassis Cabs
• 2014-2015 Jeep Grand Cherokee and Cherokee SUVs
• 2014-2015 Dodge Durango SUVs
• 2015 MY Chrysler 200, Chrysler 300 and Dodge Charger sedans
• 2015 Dodge Challenger sports coupes

Chrysler used the Sprint cellular network for the connectivity, and the companies have been working together to take network-level security measures to block attacks like this. Valasek tweeted that he had tested the attack again and that it had effectively blocked his attacks:

Screen Shot 2015-07-24 at 10.44.23 AM

Fiat Chrysler Automobiles is taking this seriously. And the rest of the industry should, as well. V2V communication is becoming necessary and will need to be standardized. Congress is working on the rising car hacking threat, with Senators Edward J. Markey (D-Mass.) and Richard Blumenthal (D-Conn.), members of the Commerce, Science and Transportation Committee introduced legislation that would direct the National Highway Traffic Safety Administration (NHTSA) and the Federal Trade Commission (FTC) to establish federal standards to secure our cars and protect drivers’ privacy. The Security and Privacy in Your Car (SPY Car) Act also establishes a rating system — or “cyber dashboard”— that informs consumers about how well the vehicle protects drivers’ security and privacy beyond those minimum standards.


Google Autonomous Car Involved in First Injury Accident

google self-driving lexus suvSo far, the Google driverless test cars have been involved in just over a dozen accidents, none involved an injury, until now.

On July 1, 2015, a Google test vehicle, a specially-equipped Lexus LX450H hybrid crossover vehicle was traveling at 15 MPH toward an intersection in traffic when the vehicle in front came to a stop to not block the intersection. The Google vehicle also stopped. The vehicle behind it, going 17 mph, did not stop. In fact, the sensors indicate that the brakes were never applied.

Human error was the cause, and there were three Google employees in the car. One in the driver’s seat as required by law for testing, one in the passenger seat entering information into the laptop, and a passenger in the back seat. All three on board were treated at a hospital for minor whiplash and released. The driver of the other vehicle reported minor neck and back pain. The Google vehicle’s rear bumper had minor damage, the vehicle that cause the accident lost its front bumper.

Mountain View police responded to the incident, but no report was filed.


How A Driverless Car Sees The Road

Chris Urmson, head of the Google Autonomous Vehicle program, gives a great TED talk about how driverless cars actually see the world around them and respond better than humans. It’s a really fascinating look at how far Google’s program has come and how safe the technology is.



Google Autonomous Car Accidents: 100% Human Error

autonomous-car-googleGoogle has begun releasing monthly reports about the progress of their driverless cars. In the May, 2015 report they note that in the six years they’ve been on the streets testing, their vehicles have been involved in 12 minor accidents. Spread across their fleet, and in more than 1.8 million miles of driving, that’s a good record. Even better? Not one of those accidents was caused by the self-driving car.

Human error was to blame each time, both from other vehicles hitting the Google vehicle, or when one of the test operators was at the wheel.

In May of 2010, a Google Prius AV was in manual mode and stopped at a stop light. It was rear-ended by another vehicle. The Google AV did sustain some damage.

In August of 2011, a Google Prius AV in autonomous mode was at a traffic light and was rear-ended by another vehicle. That Google AV also sustained some damage. The same happened in May of 2012.

In December of 2012, a Google Lexus AV in manual mode was driving past a disabled vehicle and emergency vehicles on the shoulder when it was rear-ended by another vehicle traveling at 20-25 MPH. No injuries, some damage to the rear of the Google vehicle.

It gets more interesting in March of 2013, when a vehicle adjacent to the Google Lexus AV, which was traveling at 63 MPH on the freeway, veered into the Google AV’s lane and impacted the side of the vehicle. The test driver took immediate control via the steering wheel. The Google AV was slightly damaged, but there were no injuries.

In 2014, there were two more rear-end collisions, both times another vehicle colliding with a stopped Google Lexus AV. Some damage both times.

In February of this year, a Google Lexus AV was traveling in autonomous mode when another vehicle failed to come to a stop at a stop sign. The Google AV had detected this and began applying the brakes. Just before collision, the driver disengaged autonomous mode and took manual control in response to the application of the brakes. No indication of whether the braking caused the ultimate impact, but the other vehicle struck the Google AV’s right rear quarter panel and caused some damage, but no injuries were reported.

The rest of the report is available on Google’s monthly report.

It appears that the vehicles do react properly, if not erring on the side of very conservative in these situations. It would be interesting to see what they’ve applied to what they’ve learned from the incidents involving the people rolling stop signs especially. I’m sure these will be figured into the software as they continue to learn, and it’s great that they are releasing these monthly reports to the public to see how they are doing and the good safety record they have so far.

Audi Q5 crossover

Audi Q5 crossover. Image courtesy of flickr user @realmcflier

A Delphi-powered Audi Q5 crossover self-driving vehicle was being tested on the roads near Palo Alto, California. It was about to change lanes, when suddenly a Google self-driving Lexus 450h pulled into the lane next to it and caused the Delphi test vehicle to abandon its planned lane change. It was reported by some media outlets, sensationally, as a near-miss, but that wasn’t the real story. The two vehicles did what they were supposed to, one (the Google car) driving exactly as planned and the other responding to the car’s presence accordingly. This without any vehicle-to-vehicle communication, something that would have increased the safety of all cars involved.

Google Driverless Car

Google’s Lexus RX450h crossover with roof-mounted Lidar is their main testing platform. They have 23 of these and they have clocked over 1 million fully autonomous miles in them. Photo courtesy flickr.com user @MarkDoliner

It does show that even without V2V communication, the cars do respond exactly as they should, keeping both cars and all others around them safer.


The National Aeronautics and Space Administration is known for technological advances. They got us into space, to the moon, and up until recently, were the only United States-based entity servicing the International Space Station. And they’ve made cars. The Lunar Rover, shown here:

Lunar Rover

Apollo Lunar Rover


Though hardly a sexy vehicle, it was more expensive than any luxury car on the planet, at $38,000,000 for three made. That’s in 1971 money, too.

More recently, they have made an autonomous vehicle, the Mars Rover. Are you sensing a naming theme here?

They’ve just unveiled their own Earth-based autonomous vehicle, this time a passenger car. Or cart, as it looks. It has liquid-cooled drive motors at all 4 wheels, and all four can be independently steered to make incredibly tight turns and park in spots previously only available to stunt drivers who could slide a car in. It can be autonomous, driven or even remotely controlled. With 4-wheel steering, drivers report it feels like driving on ice, yet with complete control.

The short video shows off the capabilities. They even made it a little less industrial with one shot of the cart splashing through water in slow motion. And the pseudo-drifting gives it a hint of a Gymkhana video, minus the speed and smoke.

NASA just announced a 5-year partnership with Nissan for driverless car technology development.  They will be sharing information and developing autonomous vehicle systems, robotics, human-machine interfaces, software analysis/verification and network-enabled applications. The first thing they’ll be working on is cooperative research and development of algorithms, concepts and self-driving car prototypes.

Nissan will be footing the bill for this, but will gain some access to NASA Ames Research Center. Nissan will gain massive benefit from their research and facilities, and NASA will benefit from Nissan’s experience building a variety of vehicles, allowing them to design autonomous and semi-autonomous vehicles to carry payloads and humans on terrestrial and extraterrestrial missions.

“The work of NASA and Nissan – with one directed to space and the other directed to earth, is connected by similar challenges,” says Carlos Ghosn, president and CEO of Nissan Motor Company. “The partnership will accelerate Nissan’s development of safe, secure and reliable autonomous drive technology that we will progressively introduce to consumers beginning in 2016 up to 2020.”

What would a NASA/Nissan autonomous car be called? The Earth Rover? NASA Leaf? The possibilities are almost endless, except for Land Rover. That one is already taken.


New Aluminum Battery Can Be Charged In One Minute!

Screen Shot 2015-04-07 at 11.45.27 AM

Stanford University Professor Hongjie Dai and his colleagues have developed an aluminum ion battery that can fully charge in under one minute. It is also pliable, meaning it can bend around other things in manufacturing or use, allowing it to be used in bendable smartphones, easily put into existing car bodies, etc.

Unlike Lithium-based batteries, the aluminum batteries won’t short or catch fire, even if you drill a hole through them while in use. They also show no decay in capacity after hundreds, even thousands of charge cycles. This has amazing implications for uses in everything from cars to phones to grid-level storage.

Currently, the cells use an aluminum anode and a graphite cathode with a salt liquid that acts as the electrolyte. The output is only 2 volts per cell, compared to lithium cells’ 3.7 to 5.2V, but with different combinations of materials they will likely improve. The Stanford team was able to connect two cells in series with a converter and power a smartphone with 5V. Currently, Lithium-ion cells carry about 100-200 watts of electricity per kilogram compared to aluminum cells’ 40 W/kg, but the chemists think that improving the chemistry could increase the energy density as well as the voltage. Aluminum and carbon are also cheaper.

Imagine being able to charge your smartphone in one minute. Or your car in just a few, like pulling into a gas station and filling up with gas. A cheaper, safer, longer life battery that charges in a minute or two? Count us in!

In their video on YouTube, you can see the researchers powering an LED while they bend and even drill through the battery.


Your Car Will Be Connected To The Internet Within 5 Years

Freeway and highway driving can be made safer and more efficient by V2V communication.

By 2020, most cars and trucks will be connected to the internet. According to a Gartner report, nearly 150 million vehicles will have internet connectivity, including 60-75% of them capable of consuming, creating and sharing web-based data. This brings with it a $141 billion market for automotive connectivity. This will allow cars to be part of the “Internet of Things”, a way for devices (including vehicles) to communicate for exponentially higher levels of efficiency and convenience.

The upcoming IEEE standard 802.11p standard is specifically designed for automotive applications for Intelligent Transportation Systems (ITS). In the same way that apps like Waze allow crowdsourced traffic conditions and rerouting for faster travel between points, vehicle-to-vehicle (V2V) communication allows vehicles to instantly know about traffic, speed changes and hazards ahead and respond more accurately, as well as safely platooning to reduce congestion and improve fuel economy.

In the United States, faster LTE networks are coming in and older technology (e.g. 2G) are being phased out, increasing speeds and connectivity. This can only help V2V and V2I communication.

Telematics and infotainment will continue to improve. Gartner predicts that 58% of the U.S. and 53% of German vehicle owners want tech firms to develop in-vehicle technology instead of the car companies. This will definitely standardize and improve the user interface (UI). Cars are increasingly adding connectivity to phones, like Android Auto and Apple CarPlay that allow an already familiar interface between drivers and vehicles. If you’ve driven a rental car that’s different from your own, just navigating an unfamiliar center stack can be frustrating because the interface was designed completely differently.


{ 1 comment }