Episode Summary:
NASCAR is America’s favorite motorsport, but it has been historically resistant to change. With the launch of their Next Gen car, the sanctioning body has moved in a single leap from steel bodied, solid axle 1980s technology to full modern race car technology.
Baidu’s Apollo AI business has launched a robotaxi system in Beijing, using factory equipped luxury cars in one of the future venues for the 2022 Winter Olympics. Combined with a larger, integrated traffic control system, Baidu hopes to take the lead in self driving technology.
With millions of lines of code in most systems, making websites and databases secure is a major headache for the Pentagon. A new initiative from the Department of Defense aims to use a different approach: pay hackers to attack systems to find weaknesses.
Access all episodes of This Week in Engineering on engineering.com TV along with all of our other series.
Transcript of this week's show:
Segment 1: Data security in military software has been a major concern for decades, but how vulnerable are US defense industrial base companies to cyberattack? That’s a major concern for the Defense Department and to address it, the DoD has turned to an unusual resource: hackers. The origin of the strategy was a memorandum of agreement signed in 2019 between the Department of Defense Cyber Crime Center (DC3) and the Defense Counterintelligence and Security Agency (DCSA) which outlined security information sharing between the agencies. Called the Vulnerability Disclosure Program, it’s a formal expansion of the 2016 “Hack the Pentagon” initiative, where outside ethical hackers were invited to attempt to breach DoD online assets, and were paid a bounty for success. Over 16,000 vulnerabilities have so far been revealed, and following a Carnegie Mellon study published in July of last year, the program is now expanding to include non-DoD sites in industry covering frequency-based communication, Internet of Things, industrial control systems, and other critical industry data weaknesses. As 5G connectivity and the Internet of Things exponentially increases the amount of data moving through the cloud, the new bounty on vulnerabilities appears to neatly solve two problems: it avoids the need to survey millions of lines of code in the quality assurance process of software development looking for points of entry, and it allows thousand of DoD contractor and supplier companies to test their systems without the cost and bureaucracy of a Pentagon-mandated review process. And it could be a way for hackers to come over from the dark side and make some money, legitimately.
Segment 2: NASCAR is America’s favorite form of motorsport, and from an engineering standpoint, it’s famous for one thing: resistance to change. Technology is heavily regulated and the technologies that are common in every other form of motorsports have been banned by the sanctioning body, often for decades. That is changing, however and the newly introduced Next Gen car was launched with fanfare by NASCAR at The Park Expo in Charlotte, NC last week. The rule changes are the most sweeping in the history of the sport and they essentially drag the technology of the cars from the 1980s into 2021, all at once. Bodies will be symmetrical and made of composites instead of steel. Three models are approved, a Chevrolet Camaro, Ford Mustang and Toyota Camry and all will use independent rear suspension, a 5 speed sequentially shifted transmission and larger, 18 inch tires on aluminum alloy wheels. The traditional 5 lug nuts are gone, replaced by an Indy-style single central nut. Brakes are bigger and the cars will be flat bottomed, race with an extra 2 inches of ride height and are equipped with rear diffusers to reduce turbulence in the draft. Tires will be 2 inches wider and steering will be by rack and pinion. The traditional engine mounted gearbox has been replaced by a transaxle and in a first for NASCAR, will incorporate a power take off that will allow future use of electric motors in a hybrid configuration. Chassis will still be steel space frames, but will be sole sourced like the transmissions and bodies, from NASCAR approved suppliers. Multiple in car cameras will be included and with limited rearward vision, rear view cameras will be used by the drivers for the first time. The changes are many, but according to NASCAR, they will result in closer, more aggressive racing.
Segment 3: The race for the self-driving car, in the form of the robotaxi, has been hotly contested for years, with different technology developers and manufacturers claiming leadership. Tesla has an advanced driver assist system on the road now and GM’s Cruise Automation and Alphabet’s Waymo are operating with regular passengers in test cities like Phoenix and San Francisco. But who will be first with a truly automated system for public use? Well, the lead may have shifted to Beijing, with the approval of Baidu Apollo’s SAE Level 4 robotaxi for public use in Beijing. According to the South China Morning Post, passenger operations have already begun in the Shougang Industrial Park, one of the venues for the 2022 Winter Olympics. The system allows riders to summon the driverless taxis with a smartphone app, and Baidu will charge a flat rate of 30 yuan, about $4.60, for rides within the complex on the western outskirts of Beijing. Although driverless, the cars can be remotely controlled in an emergency, using a 5G data link. Both Alphabet’s Waymo and GM/Honda’s Cruise partnership are doing something similar in the US, but Baidu hopes to lead the race with factory integration of the sensor suite and automation hardware through a partnership with the FAW Group, makers of the Hongqi luxury auto brand. Waymo currently uses converted gasoline passenger cars and Cruise also adapts Chevrolet Bolt sedans, but plans to launch a GM-designed pod with no steering wheel or driver controls in the future.
The Baidu Apollo system uses a similar set of sensors to most self-driving projects, LIDAR, GPS, forward, side and rear cameras, plus ultrasound ranging and millimetre wave radar. One possible advantage of the Apollo system is their plan to integrate self-driving with overall city traffic control systems, altering traffic light timing and feeding information into what the company calls an Intelligent Vehicle Infrastructure Cooperation Solution. The idea is to optimize traffic flow and assist self-driving algorithms by controlling the vehicles around the self-driving taxi to eliminate possible hazards. It’s an interesting approach: reduce the possibility of unexpected events to reduce the load on the on-board computers. Will it work? No one really knows who will be first with a universal solution, but if it requires whole city traffic system integration, China may be able to adopt a regulatory system very quickly.