Driverless Radio

On Sunday, March 18th, at approximately 10pm, an Uber was operating in self-driving mode when the vehicle struck and killed 49-year-old Elaine Herzberg. Initial reports say that Elaine was walking her bicycle across the street when the Uber struck her. At this time, facts are limited, but at a press conference in Tempe, AZ, officials said that Elaine was crossing the street outside of a crosswalk area when the Uber struck her while traveling at approximately 40 mph.  I suspect that the car's sensors did not identify Elain or her bicycle as an obstacle or pedestrian.

In this episode, Robert Solano explores this incident and discussed the potential impacts to the development of self-driving car technology.

Phoenix has been a great test environment for companies like Waymo and Uber. The moderate climate makes it ideal for early-stage development.  In addition to the low cost of living, the "business friendly and low regulatory environment" provides an attractive alternative to the very expensive and dense Silicon Valley and San Francisco area where many tech companies are headquartered.

Uber had been testing their self-driving fleet in Arizona in addition to Pittsburg and other areas. In total, the fleet of Uber self-driving vehicles had accumulated approximately 3 million miles at the time of this accident. In other words, the current accident fatality rate of the Uber fleet is about 1 fatality in 3 million miles driven.

In contrast, the national fatality rate for human drivers is about 1.18 deaths per one hundred million miles driven. MAny advocates of self-driving cars claim that they will have better safety rates than humans, but this fatality puts self-driving car technology at a steep uphill battle to match the human fatality rate.

This accident illustrates three key elements of self-driving car technology and development.

1) There are weaknesses to single driver fleets.

Waymo CEO John Krafcik boasted that Waymo was trying to develop the most experienced driver ever. He claimed that all of the vehicles in the Waymo fleet were essentially the same driver and shared their lessons learned and experiences with each other. At first, that idea seems great, but there are serious implications if that hardware or software has a flaw.

As a result of this accident, the entire fleet of Uber self-driving vehicles is suspended from further road testing.  Since Uber has a small fleet of these vehicles, suspending their services is not a big deal. But in the future, if a company has a much larger fleet, then suspending a service could potentially leave customers without a ride-hailing service

Today, there are tens, if not hundreds, of thousands of Uber cars in operation around the world. Many people depend on them for transportation. There would be a severe econmic impact if all of the Uber (or Lyft, or Taxis) all suddenly stopped operating simultaneously.

The one-driver model sounds good for machine learning, but it severely reduces the redundancy that is inherently built into our current system where each human is unique and we all operate thousands of different are makes and models.

2) This accident will delay the deployment of self-driving technology by 12 to 18 months.

Opponents of self-driving technology will use this incident as a perfect example to push for stricter regulations, testing, and monitoring of self-driving car development. Additionally, companies will become more cautious. The result is that the deployment of self-driving car fleets will be delayed 12-18 months.

3) Company ethics are crucial to self-driving car technology.

Too many academics are focused on the ethics of machine decision making.  They talk about topics like

In this episode, Robert and Zaira travel to the South by Southwest Interactive Festival (#SXSW) in Austin, TX.  It was a very busy week. Together they saw over 30 different presentations and 20 different musical performances. The highlights include keynote speeches by Elon Musk (Tesla CEO), John Krafcik (Waymo CEO), Melinda Gates (Gates Foundation), and Ira Glass (This American Life and Serial podcasts). In this episode, we share a small part of the keynotes from Elon Musk and John Krafcik.

As first time SXSW festival attendees, Robert and Zaira share their experiences and provide tips for others that may be interested in attending the festival in the future.

In summary, tips for SXSW include:

#1. If your primary purpose is to attend the keynotes and seminars, get an 'interactive' badge instead of the 'platinum.' The interactive badge will get you primary access to the interactive sessions and secondary access to most of the music and movies.  This tip can save you a couple of hundred dollars. alternatively, you could: a) get together four of your industry partners and submit a proposal for a panel, or b) volunteer for part of the week.

#2. Book your hotel about 9 months in advance. If you wait until January, like us, the cheapest hotels downtown are $800-$1200 for a basic hotel like the  Marriott Courtyard. We got lucky and got an AirBnB, but even those are risky. We had our original AirBnB cancel on us two months before the conference.

#3. Go with a strategic plan. Get your rest, drink your caffeine, and prepare for a lot of walking and waiting in lines.

Please like and share. Mahalo.

Uber was in the news a lot this past week. First, they made positive headlines when they announced that they were expanding their partnerships with health care providers to offer patients rides to doctors' offices (at the doctor office expense). Then they made negative headlines when MIT reported that most ride-sharing drivers make less than minimum wage when you deduct their cost of insurance, maintenance, and fuel.  Finally, Uber announced that they were testing autonomous semi trucks in Arizona. This trucks may give Tesla a run for their money.  in all, a very busy week for the Uber public relations office. And of course, we can't forget the guy in West Virginia who called an Uber after a night of partying. He asked the Uber to take him home and then passed out. Too bad his home was 300 miles away in New Jersey. I'm sure he was surprised when he woke up to a $1,600 bill.

All of these Uber stories offer some very interesting insight into human sociology. However, the doctor's story is probably the most interesting. For decades, passengers have been paying for taxis and more recent ride-sharing services. Autonomous cars may flip this business model. In the future, mobility will probably become more like the Google search engine model.

Most of us do not pay to use Google search engines. It is a free service for consumers. Google makes their money by charging businesses for advertising. The businesses that pay the most for advertising get premier access to customer attention; they show up first in the search results.

Driverless cars will significantly reduce the cost of rideshare services. Once that happens, more businesses will begin to pay the Ubers and Lyfts for business. As an example, downtown restaurants will pay Uber and Lyft rides for customers who go to their restaurants for brunch.

As a customer, when you decide to leave your house, you will be given a few different free options for brunch destinations. Most of us will be more included to select a free ride to a brunch spot than pay a premium price to go to the restaurant of our choosing. This is the future of mobility and it has serious implications for small businesses and our society.

For almost 100 years, scholars have been debating the Trolley Problem. The scenario is simple.  In its original form, a trolley or train was speeding down a set of railroad tracks uncontrollable and unable to break. Ahead of the trolley is a Y-intersection. On one fork of the intersection, 5 innocent people are tied up on the tracks. On the other side of the intersection, a child is tied up on the tracks. At the intersection, a man could allow the trolley to continue its course, undoubtedly killing 5 innocent people, or the man can divert the train and kill only the one child.... a child that happens to be his own flesh and blood. Should the man sacrifice his own child to save the lives of 5?

This scenario, or others like it, have been used in philosophy and ethics classes for decades. Most recently, institutions like MIT have updated the problem to include driverless car scenarios.

The MIT Moral Machine lets users make decisions as if they were an autonomous car. Over the course of many different binary, one-or-the-other scenarios, users can select if they would rather kill a grandma or a dog, a man or a group of cats, a bank robber or a doctor, etc.

I absolutely hate the trolley problem and all its incarnations. 

The trolley problem was originally designed to provoke ethical discussions in human decision making. It should be left in ethics classes and left out of engineering discussions.

In its original form, there is nothing natural or ethical about the scenario of an out of control trolley barreling down the track towards a group of five people tied up.  It would be impossible to predict how anyone would act in that situation. The real focus should not be on the decision to direct the train, but instead, the focus should be on the sadistic bastard that cut the train's brakes, tied up six people to the track, and coerced a person to choose between the life or death of their child. That is a seriously screwed up criminal. Let's talk about criminal ethics instead of autonomous ethics.

Just how it is impossible for the trolley problem to every happen without a criminal intent, it is also impossible for a car to be traveling down the street and need to decide if it wants to kill a group of five old ladies or a group of five bank robbers.

I hate the trolley problem for three main reasons.

#1. Attempting to apply the trolley problem to teach or evaluate autonomous cars demonstrates a fundamental lack of understanding of machine learning and artificial intelligence. These cars will have many different sensors. They are looking for obstacles. They will never be programmed to rank order the value of old ladies versus children. To a car, they are both obstacles that will be sensed by a variety of cameras, radar, lidar, etc. This data feeds into an artificial intelligence machine that uses complex algorithms and probabilistic models to select the best course of action. They will almost never have to make simple binary either-or, life-or-death decisions as postulated in the trolley problem.

#2. The chances of a car getting into a situation where it needs to decide in a split second which of two obstacles it will hit are nil without criminal intent.

#3. In almost all scenarios, the best, most ethical, simplest, safest course of action for a driverless car will be to brake. It is that simple. If a car cannot navigate around an obstacle, it should brake. These cars will have reaction times and sensors that exceed human capability (I saw this first hand in Las Vegas). They will be able to sense obstacles long before humans and will almost always have the ability to brake before hitting one.

This is a really exciting episode.  I recorded this episode from the inside of the Aptiv and Lyft self-driving car at the Consumer Electronics Show in Las Vegas. Aptiv and Lyft partnered up to offer limited demonstrations of their self-driving car during the massive tech conference.

When I first checked into the Lyft tent at the Gold Lot at CES, I received instructions on how to request a self-driving Lyft ride from a staff member. At first, I needed to update my app. Once the app was updated, I opened it up and then had to continuously close and refresh it for about 10 minutes until a pop-up window showed up that asked me if I wanted to take a demo ride in a Lyft self-driving car. I promptly agreed but was disheartened to find out that there were no available spots. After about fifteen minutes of staring at my Lyft app, the button finally changed to an option that I could reserve a spot on the waitlist. Needless to say, the self-driving Lyft rides were in very high demand and the process to request one was not the easiest. Considering that there were hundreds of people in the parking lot that were all requesting a Lyft ride, I was very fortunate to get added to the waitlist.

Once I got added to the wait list, the app showed that my Lyft ride was 21 minutes away. An hour later, the wait time finally changed to 20 minutes.  A half-hour later, the wait time decreased to 16 minutes.  As I waited, there was a race between the decreasing life of my iPhone battery and the arrival of my ride.

While I waited at the Lyft tent, I took photos of the car on display and compared its sensors to the tech that we discussed in episode 11.  The BMW was much sexier than the Keolis shuttle.

Finally, a Lyft or Aptiv engineer came into the tent to inform Britney that her ride had arrived. Clearly, Brittney was not in it to win it, because she was nowhere to be found. The engineer asked me if I wanted to take her spot and if her predestined drop off point at the Bellagio Hotel was acceptable to me.  The Bellagio was 20 minutes away from my hotel at the Venetian, but after waiting for two hours I would take that car anywhere its artificial intelligence system would take me.

As I was waiting around, I met a friendly guy named Brent Hankins, who happened to be a Design Engineer in the self-driving car industry. We agreed to share the ride to the Bellagio and the rest is history.

Listen to the podcast for the play-by-play.

Of course, since this was a demo, there was a safety driver and technician in the front seats. Odin and Murun were both great hosts and represented the Aptiv and Lyft companies exceptionally well. I can't thank them enough for their great insight and patience with us.

Overall, I was very impressed with the Aptiv and Lyft self-driving car. Odin had to take control three times during our 30-minute drive.  He navigated out of the convention center parking lot, took control momentarily while a police officer was controlling traffic, and had to take control at the end when the car was headed to a location other than the Bellagio. 

Traffic was the worst that the Las Vegas Strip had to offer, and the Aptiv and Lyft self-driving car handled it with ease.  I was very impressed.

For some reason, the self-driving experience continued to freeze my Lyft app for almost an hour after I got dropped off. I guess it didn't like the fact that I took Britney's ride, so it was still showing that my Lyft was 15 minutes away. So in a strange irony,

In this episode, Robert and Zaira discuss the technology behind self-driving cars in preparation for the 2018 Consumer Electronics Show (#CES2018). Listen in to learn more about the basic functions of LIDAR, radar, ultrasonic sensors, and cameras. The most noticeable piece of tech on self-driving cars is usually the LIDAR sensor (Light Detection and Ranging).  Early prototypes for self-driving cars usually included a dome-shaped LIDAR sensor mounted to the roof. These sensors are great at detecting detailed objects at long distances. For example, they can 'see' a dime size object up to a football field away. However, traditional LIDAR sensors are poor performers in bad weather such as rain or fog. And they are also very expensive, although companies like Velodyne are making advances every day to reduce the cost of LIDAR sensors.

Next up is radar. These object may not be able to discern an object the size of a dime, but they can sense larger obstacles at longer distances. Most modern cars that have cruise control systems that speed up or slow down with the flow of traffic use some type of radar sensor.  Future autonomous cars will likely continue to use this reliable tech.

Ultrasonic and cameras are already widely used in cars today. Ultrasonic sensors help with blind-spot protection and cameras prevent us from backing up over our child's bicycles in the driveway. Future self-driving cars will probably have even more sensors like this than our standard equipped cars of today.

With all this self-driving car tech, some people may be wondering, "What is the optimal mix of sensors?" That is a very difficult question to answer and there are varying opinions, but the human is a great example to discuss this question. There are people with less than 20/20 vision that are still valuable members of society, and deaf people can still drive. Likewise, not all sensors may be required for future autonomous and self-driving cars.

In January, Robert and Zaira will head to the Consumer Electronics Show in Las Vegas.  Subscribe on your favorite podcast app in order to listen to future updates about these emerging technologies. Hopefully, Robert and Zaira will be able to get a ride on Lyft's driverless car at the show.

Listen in to the episode for more details and don't forget to leave a comment on Facebook.

In this episode, Robert gets a ride on a driverless shuttle at the Florida Automated Vehicle Summit in Tampa, FL.; and he interviews the team from the Easy Mile shuttle and May Mobility.

This episode is a little different since Robert is covering a life event. In this episode, Robert provides live commentary to Tesla's unveil of their semi-autonomous, electric, semi-truck.

This is the event that was originally promised this summer, but then delayed until September, but then delayed until November. In typical Tesla fashion, Elon announced that the unveil would occur at 11pm EST on November 16th, but then it was delayed until 11:15pm.  After the truck unveil, Elon had a special surprise!!!!

In this episode, Robert joins the live event online. While getting temporarily delayed, he explores designing his dream Model S before the live event starts. The live event starts with the trucks driving onto the stage. Elcon then highlights the exceptional acceleration and speed capabilities of the trucks. These trucks could reach 60 mph up a 5% slope while traditional trucks would struggle to reach 45 mph. Elon also discussed the great aerodynamic properties of the truck, the fantastic recharge capabilities, and some other cool stuff.

It cost a hefty $5,000 deposit to register to buy a truck. I am not sure how many people will register for a truck since many people are still waiting on the Model 3 that they registered for almost two years ago.

Overall, the truck was lackluster. It is a very cool concept, and will significantly help the environment, but until we have a cost and can validate a business case for the trucks, the verdict is out.

After some technical difficulties, the show resumed and Elon had a freaking awesome surprise that I don't think anyone was expecting. Listen to the episode for the details.

In this episode, Robert takes a trip to the Florida Automated Vehicles Summit in his hometown of Tampa, Fl.  The FAV Summit assembles industry leaders from around the world to address technologies, operations, and policy issues related to driverless and self-driving cars.  In this first part of the FAV Summit episodes, Robert interviews Florida State Senator Jeff Brandes, Richard Bishop, and Joe Waggoner from the Tampa Hillsborough Expressway Authority.

This year's exhibitors include companies such as Tesla, Nissan, Audi, Sirus XM, and COAST, who makes the P-1 Self-Driving Shuttle similar to those seen in Las Vegas. A key message from the key speakers is that the future of transportation will likely be 1) autonomous, 2) electric, 3) connected, and 4) shared.  In order to be a pioneering city, Tampa has promoted development and testing of this technology. Recently they announced plans to launch an autonomous bus in downtown Tampa sometime in 2019.

We expect a lot of great photos and videos over the next few days. Follow @DriverlessRadio on Facebook, Instagram, and Twitter for future updates and photos.

In this episode, Robert and Zaira take a virtual trip around the globe exploring self-driving car current events from various cities. Some of the current events discussed include the following:

New York: GM announced plans to begin testing its driverless Chevy Bolt in this busy city. 

The yellow cab drivers won't know what hit them.

Detroit: Waymo, a spin-off from Google, announced that it will test self-driving cars in Detroit in order to evaluate how those cars perform in the snow.

Ann Arbor: Also in Michigan, Dominoes pizza continues to test its pizza delivery service and is evaluating how customers react to getting their pizza delivered in a self-driven car. Although it may be fun during the summer time, I suspect that most people would prefer a driver that brings the pizza to your doorstep on cold days, rather the customer needing to venture out into the cold to get the pizza from the car.

Beijing: Baidu, the big tech company from China, announced that it will invest $1.5 billion (with a B) into self-driving cars this year. In comparison, Google invested $1.1 billion from 2009 to 2015 before spinning Waymo off into its own company.

Germany: Boring buses.

Singapore: Delphi announced that it would buy Boston based nuTonomy and continue testing self-driving cars in this Lion City.

Toronto: The Google subsidiary, Sidewalk Labs, announced that it will team with the city of Toronto to design a waterfront community. This initial investment of $50 million will initiate the design, but the estimated cost for the total project is approximately $1 billion. Concept sketches from Sidewalk Labs show driverless cars in mixed-use spaces and tunnels below the city where driverless delivery trucks and trash collectors shuttle supplies and waste.