Nuro and Kroger are deploying self-driving cars for grocery delivery in Arizona today

Self-driving car startup Nuro is ready to put autonomous vehicles on the road in partnership with Kroger to deliver groceries in Scottsdale, Arizona. This comes a couple of months after Nuro and Kroger announced their partnership to offer same-day deliveries. This pilot will serve a single Fry’s Food and Drug location in Scottsdale starting today. Customers […]

Self-driving car startup Nuro is ready to put autonomous vehicles on the road in partnership with Kroger to deliver groceries in Scottsdale, Arizona. This comes a couple of months after Nuro and Kroger announced their partnership to offer same-day deliveries.

This pilot will serve a single Fry’s Food and Drug location in Scottsdale starting today. Customers can shop for groceries and place either same- or next-day delivery orders via the grocer’s website or mobile app. There’s no minimum order but there is a flat delivery fee of $5.95.

“We’re proud to contribute and turn our vision for local commerce into a real, accessible service that residents of Scottsdale can use immediately,” Nuro CEO Dave Ferguson said in a statement. “Our goal is to save people time, while operating safely and learning how we can further improve the experience.”

Nuro’s intent is to use its self-driving technology in the last mile for the delivery of local goods and services. That could be things like groceries, dry cleaning, an item you left at a friend’s house or really anything within city limits that can fit inside one of Nuro’s vehicles. Nuro has two compartments that can fit up to six grocery bags each. 

In Scottsdale, however, Nuro will initially use Toyota Prius cars before introducing its custom self-driving vehicles. That’s because the main purpose of this pilot is to learn, and using the Prius self-driving fleet can help to accelerate those learnings, a Nuro spokesperson told TechCrunch.

“The Priuses share many software and hardware systems with the R1 custom vehicle, so while we compete final certification and testing of the R1, the Prius will begin delivering groceries and help us improve the overall service and customer experience,” the spokesperson said.

When it came to going to market, Ferguson previously told me groceries were most exciting to him. And Kroger particularly stood out because of its smart shelf technology and partnership with Ocado around automated fulfillment centers.

“With the pilot, we’re excited about getting more experience interacting with real customers and understanding exactly what they want,” Ferguson told me. “The things they love about it, the things they don’t love as much. As an organization for us, it’s also very valuable for us to have to exercise our operational muscle.”

Throughout the pilot program, Nuro will be looking to see how accurate its estimated delivery times are, how the public reacts to the vehicles and how regular, basic cars interact with self-driving ones.

Drive.ai’s self-driving vehicle service is now live in Texas

The bedroom community of Frisco, Texas might seem like an unusual place to find a self-driving vehicle. But here in this city of nearly 175,000 people, there are seven. And as of Monday, they’re available for the public to use within a specific sector of the city that has a concentration of retail, entertainment venues […]

The bedroom community of Frisco, Texas might seem like an unusual place to find a self-driving vehicle. But here in this city of nearly 175,000 people, there are seven.

And as of Monday, they’re available for the public to use within a specific sector of the city that has a concentration of retail, entertainment venues and office space.

Drive.ai, an autonomous vehicle startup, launched the self-driving on-demand service Monday that will cover a two-mile route. The service will be operated in conjunction with Frisco TMA, a public-private partnership focused on “last-mile” transportation options. People within this geographic zone can hail a ride using a smartphone app.

Even in their small numbers, the modified Nissan NV200s will be hard to miss. The self-driving vehicles are painted a bright orange with two swooping blue lines — with the words “self-driving vehicle” and “Drive.ai” set in white.

The vehicles, which have been given distinctly human names like Anna, Emma, Bob, Fred and Carl, are equipped with LED screens on the hood and rear, and above the front tires, which will display messages as well as the vehicle’s name to pedestrians.

This isn’t a business enterprise just yet. The service, which is considered a pilot project, is free and will be operational for six months. The program will begin with fixed pick-up and drop-off locations around HALL Park and The Star and then will expand into Frisco Station.

Conway Chen, Drive.ai’s vice president of business strategy, emphasized to TechCrunch that this is designed as an on-demand service, and not a shuttle. When the vehicles are not being used they won’t just keep circling the route, which could cause more traffic congestion, Chen said. Instead they will be able to park along the route.

In the weeks since announcing plans to launch in Frisco, Drive.ai has been tweaking the service, its schedule as well as racking up miles on the road and in simulation. The company said it has logged 1 million simulated miles on its Frisco route. In its simulation, Drive.ai replicates scenarios — taken from its driving logs — the vehicles encountered while driving the route, as well as creating its own scenarios.

As Drive.ai explains in a post on Medium: “It’s like a high tech version of SimCity, where we design the world, and can then replay events and modify their components to explore how our technology responds in unique scenarios. This is a good place to start for the more common things that people do on the roads: navigating tricky intersections, right-of-way decisions, and observing the behaviors of cyclists and pedestrians.”

Drive.ai simulation.

The service, which will operate weekdays from 10 a.m. to 7 p.m., will initially have a safety driver behind the wheel. That person will eventually move to a passenger seat and take on a chaperone role, whose primary responsibility will be to answer questions and make riders comfortable. At some point, Drive.ai will remove the employee from the vehicle completely.

The company also has a remote monitoring feature, called “telechoice,” that allows a human operator to see everything in real-time that the self-driving vehicle can see using HD cameras.

Telechoice is not like the full remote control teleoperation that startup Phantom Auto provides. The telechoice operator can control basic functions like braking, but it cannot take full control of the vehicle or make it accelerate. With Drive.ai’s feature, if “Bob” the self-driving vehicle struggles with a certain situation on the road, the telechoice operator can help it make the right decision.

Uber self-driving cars back on public roads, but in manual mode

Uber is putting its autonomous vehicles back on Pittsburgh’s city streets, four months after a fatal accident involving one of its self-driving cars prompted the ride-hailing company to halt testing on public roads. But for now, Uber’s modified self-driving Volvo XC90 vehicles will only be driven manually by humans and under a new set of safety […]

Uber is putting its autonomous vehicles back on Pittsburgh’s city streets, four months after a fatal accident involving one of its self-driving cars prompted the ride-hailing company to halt testing on public roads. But for now, Uber’s modified self-driving Volvo XC90 vehicles will only be driven manually by humans and under a new set of safety standards that includes real-time monitoring of its test drivers and efforts to beef up simulation. 

The sensors, including light detection and ranging radar known as LiDAR, will be operational on these self-driving vehicles. They won’t be operated in autonomous mode, however. Uber will use these manually operated self-driving vehicles to update its HD maps of Pittsburgh.

This manual-first rollout is a step toward Uber’s ultimate goal to relaunch its autonomous vehicle testing program in Pittsburgh, according to Eric Meyhofer, head of Uber Advanced Technologies Group, who published a post Tuesday on Medium.

Uber halted all of its autonomous vehicle operations March 19, the day after one of its vehicles struck and killed pedestrian Elaine Herzberg in the Phoenix suburb of Tempe. Uber was testing its self-driving vehicles on public roads in Tempe, Ariz., where the accident occurred, as well as in Pittsburgh, San Francisco and Toronto.

In the days and weeks following the fatal accident, it appeared the company’s self-driving vehicle program might end for good. Arizona Governor Doug Ducey, a proponent of autonomous-vehicle technology who invited Uber to the state, suspended the company from testing its self-driving cars following the accident. Last month, Uber let go all 100 of its self-driving car operators in Pittsburgh and San Francisco.

Those drivers affected by the layoffs, most of whom were in Pittsburgh, are being encouraged to apply for Uber’s new mission specialist positions. Uber is holding off on making these positions public until the laid-off drivers have a chance to apply and go through the interview process.

Even now, with the company beefing up its safety protocols and taking a slower approach to autonomous vehicle testing, the program’s future is still uncertain. Another accident would likely derail it for good.

These new safeguards aim to avoid such a scenario. Uber said Tuesday that all its self-driving vehicles, whether they’re driven manually or eventually in autonomous mode, will have two Uber employees inside. These “mission specialists” — a new name Uber has given to its test drivers — will have specific jobs. The person behind the wheel will be responsible for maintaining the vehicle safely, while the second “mission specialist” will ride shotgun and document events.

Uber is also equipping every self-driving vehicle with a driver monitoring system that will remain active whenever the vehicle is in use. The system will track driver behavior in real time. If it detects inattentiveness an audio alert will cue the driver. An alert is also sent to a remote monitor who will take appropriate action once they’ve assessed the situation, Uber said. 

The driver monitoring system, which an Uber spokesperson declined to share details about, is an off-the-shelf aftermarket product.

Investigators determined that Rafaela Vasquez, who was operating the Uber self-driving vehicle involved in the fatal crash, looked down at a phone that was streaming The Voice 204 times during a 43-minute test drive that ended when Herzberg was struck and killed, according to a 318-page police report released by the Tempe Police Department.

Based on the data, police reported that Vasquez could have avoided hitting Herzberg if her eyes were on the road. The case has been submitted to the Maricopa County Attorney’s office for review against Vasquez, who could face charges of vehicular manslaughter.

Uber has always had a policy prohibiting mobile device usage for anyone operating its self-driving vehicles, according to a spokesperson. However, without a proper driver monitoring system or another passenger in the vehicle, it was impossible for Uber to really know if that rule was being followed.

Now, the driver monitoring system can spot the behavior immediately. If the system detects the driver is looking at a phone, the remote monitor will call the team immediately back, a spokesperson said, adding it was grounds for dismissal.

Other safeguards include a defensive and distracted driving course conducted on a test track and fatigue management program that requires the two mission specialists in each vehicle to periodically switch between driver and data logger roles, according to Uber.

The National Transportation Safety Board is also investigating the accident. A preliminary report by the NTSB found Uber’s modified Volvo XC90’s LiDAR and radar first spotted an object in its path about six seconds before the crash. The self-driving system first classified the pedestrian as an unknown object, then as a vehicle and then as a bicycle. At 1.3 seconds before impact, the self-driving system determined that an emergency braking maneuver was needed to mitigate a collision, according to the NTSB. But to reduce the potential for “erratic behavior,” Uber had disabled Volvo’s emergency braking system so it didn’t work when the vehicle was under computer control.

Uber said it will keep Volvo’s emergency braking and vehicle collision warning systems enabled while the vehicle is in manual mode. Engineers are examining whether the Volvo’s safety system can work in conjunction with its own self-driving technology while the vehicle is in autonomous mode.

A $225 GPS spoofer can send sat-nav-guided vehicles into oncoming traffic *

* Some restrictions apply.

Enlarge (credit: Zeng et al.)

Billions of people—and a growing number of autonomous vehicles—rely on mobile navigation services from Google, Uber, and others to provide real-time driving directions. A new proof-of-concept attack demonstrates how hackers could inconspicuously steer a targeted automobile to the wrong destination or, worse, endanger passengers by sending them down the wrong way of a one-way road.

The attack starts with a $225 piece of hardware that’s planted in or underneath the targeted vehicle that spoofs the radio signals used by civilian GPS services. It then uses algorithms to plot a fake “ghost route” that mimics the turn-by-turn navigation directions contained in the original route. Depending on the hackers’ ultimate motivations, the attack can be used to divert an emergency vehicle or a specific passenger to an unintended location or to follow an unsafe route. The attack works best in urban areas the driver doesn’t know well, and it assumes hackers have a general idea of the vehicle’s intended destination.

“Our study demonstrated the initial feasibility of manipulating the road navigation system through targeted GPS spoofing,” the researchers, from Virginia Tech, China’s University of Electronic Sciences and Technology, and Microsoft Research, wrote in an 18-page paper. “The threat becomes more realistic as car makers are adding autopilot features so that human drivers can be less involved (or completely disengaged).”

Read 10 remaining paragraphs | Comments