Traffic hell is alive and well in Los Angeles. In 2017, Angelenos were stuck on the road for 102 hours each (more than four full days), costing the city $19.2 billion, according to INRIX's annual global traffic scorecard. Traffic is almost as bad--and costly--in Moscow, Sao Paulo, and London. But this is the 21st century! Can't AI fix these problems by optimizing traffic flow?
Motor Trend will never publish a First Drive review on the Volkswagen SEDRIC. That's because the SEDRIC, which Volkswagen Group is currently evaluating for launch sometime after 2023, is a completely autonomous vehicle. It has no steering wheel. You just call it up on your smartphone and tell it where you are and where you want to go. When it arrives, get in, sit back, relax, and enjoy the ride.
Mr. Musk's frustration is fueled in part by widespread media attention to accidents involving Tesla's self-driving "autopilot" feature. On the company's most recent earnings call, he said that "there's over one million…automotive deaths per year. And how many do you read about? So they write inflammatory headlines that are fundamentally misleading to the readers. But not all accidents are created equal.
Computers may be poised to take control of driving in the future, but humans will be backing them for some time yet. Tech giants Waymo and Uber Technologies Inc., auto makers General Motors Co. GM -0.85% and Nissan Motor Co. NSANY -1.16%, and upstarts like Phantom Auto are all developing ways for people to remotely assist their autonomous vehicles during complicated driving situations. "You're going to want as many different backup systems as possible, and human beings performing remote operations is one of those," said Anthony Foxx, former U.S. Transportation Secretary and adviser to venture-capital firm Autotech Ventures. Having human backup will likely help alleviate concerns that regulators and insurance companies have about the new technology, he added. Driverless cars, using sensors, cameras and digital maps, tend to navigate the world based on road markings and rules of the road.
AirSim is a simulator for drones, cars and more built on Unreal Engine. It is open-source, cross platform and supports hardware-in-loop with popular flight controllers such as PX4 for physically and visually realistic simulations. It is developed as an Unreal plugin that can simply be dropped in to any Unreal environment you want. Our goal is to develop AirSim as a platform for AI research to experiment with deep learning, computer vision and reinforcement learning algorithms for autonomous vehicles. For this purpose, AirSim also exposes APIs to retrieve data and control vehicles in a platform independent way.
Artificial intelligence, or AI: what exactly is it? How much should we trust it? There are numerous questions we could ask regarding this issue, and many of these concerns were brought into focus several weeks ago when an Uber self-driving car struck and killed a pedestrian in Tempe, Arizona. The accident resulted in the first fatality from a self-driving car accident in history, and as a result, Uber and other developers of self-driving vehicles temporarily halted their testing. Self-driving cars are just one of thousands of different ways that AI and machine learning are used, and the way that we view this incident has implications for the ways in which we think about AI in the future.
According to Warren Buffet, the CEO of Berkshire Hathaway, any technology that reduces auto accidents is remarkable, but auto insurance companies should not be holding a party, yet. The truth is, 90% of car accidents are a result of human error. Thanks to autonomous vehicle technology, that percentage will soon be lower than 5%. There are so many technological advancements taking place now. The future of car insurance is here and it's amazing.
"The computer is fallible, so it's the human who is supposed to be perfect," one former Uber test driver said. "It's kind of the reverse of what you think about computers." The fatal crash last week in Tempe, Ariz., involving an Uber autonomous vehicle is bringing new scrutiny to both the quality of Uber's technology for avoiding collision and the efficacy of its backup system of so-called safety drivers. The accident, in which a woman was struck and killed as she walked a bicycle across a road at night, is believed to be the first involving a death from a self-driving car. In much of the autonomous-vehicle testing done on public roads, there are two safety drivers: one in the driver's seat; and one in the front passenger seat who is assigned the task of logging incidents onto a computer, but, drivers say, also helps by keeping a second set of eyes on the road.
Self-driving technology is advancing rapidly, largely due to recent developments in deep learning algorithms. To date, however, there has been no systematic comparison of how different deep learning architectures perform at such tasks, or an attempt to determine a correlation between classification performance and performance in an actual vehicle. Here, we introduce the first controlled comparison of seven contemporary deep-learning architectures in an end-to-end autonomous driving task. We use a simple and affordable platform consisting of of an off-the-shelf, remotely operated vehicle, a GPU equipped computer and an indoor foam-rubber racetrack. We compare a fully-connected network, a 2-layer CNN, AlexNet, VGG-16, Inception-V3, ResNet-26, and LSTM and report the number of laps they are able to successfully complete without crashing while traversing an indoor racetrack under identical testing conditions. Based on these tests, AlexNet completed the most laps without crashing out of all networks, and ResNet-26 is the most 'efficient' architecture examined, with respect to the number of laps completed relative to the number of parameters. We also observe whether spatial, color, or temporal features - or some combination - are more important for such tasks. Finally, we show that validation loss/accuracy is not sufficiently indicative of the model's performance even when employed in a real vehicle with a simple task, emphasizing the need for greater accessibility to research platforms within the self-driving community.
In response, Uber on Monday temporarily pulled its self-driving cars off the roads where it has been testing them in four cities. An Uber spokeswoman said the company is investigating the incident and cooperating with authorities. Police in Tempe, Ariz., said the Uber vehicle was in autonomous mode with a human safety operator at the wheel when it hit 49-year-old Elaine Herzberg on Sunday night while she was walking her bicycle outside of a crosswalk. The woman later died from her injuries, according to a police statement. While it isn't clear yet whether Uber's vehicle was at fault in the accident, the fatality confirmed the fears of those who have warned for several years that someone would eventually die from driverless cars.