What Happens When Robotaxi Technologies Race? We Just Found Out in San Francisco

In a recent video, The Kilowatts partnered with Whole Mars Blog (Omar Qazi) and Arash Malek to see what happens when Tesla’s FSD Beta, Waymo, and Cruise all compete to see who can get around San Francisco faster and with fewer problems. We’ve embedded the video below, but there’s more commentary and a recap further down!

Tesla FSD Beta, Waymo, & Cruise Self-Driving in San Francisco – Race Recap
w/ @WholeMarsBlog & @MinimalDuck

Watch in 4k: https://t.co/4m3Eim0XZ4 pic.twitter.com/5kWqjuwyUF

— The Kilowatts 🚗⚡️ (@klwtts) June 21, 2023

TL;DR

Tesla finished the route in 20 minutes, followed by Waymo at 25 minutes, and Cruise at 29 minutes. 

How It Went

Before we get into what happened, let’s do a little bit of background. Readers who don’t follow these companies might wonder why they chose do run this race in San Francisco. I mean, hotels (and basically everything else) are expensive in San Francisco, and a longer race across California or some other state would be even better, right? But, when it comes to Waymo and Cruise, San Francisco is the only place where you can get a ride from both companies. Tesla’s FSD Beta works anywhere, but isn’t yet approved for driverless operation.

The cars all started at McKinley Park, with a goal to get to a spot near the Golden Gate bridge. The cars all started off at the same place, but their computers chose different routes. Most notably, Cruise refused to plan a trip down Lombard (San Francisco’s famously twisty and steep street). As the vehicles started going, the different test riders/driver talked about the ups and downs to their platforms, including things like hours of availability, seating capacity, and ride comfort.

Tesla’s FSD Beta car completed the route in 20 minutes, followed by Waymo at 25 minutes, and Cruise at 29 minutes. For ride quality, Waymo waited for a garbage collection worker, but then squoze by him, making him mad. Cruise had one hiccup in the form of a 10-15 second stop for no apparent reason. FSD Beta went over some speedbumps a little too quickly, and Omar had to give it an extra push of the accelerator at one point for safety.

One big advantage Tesla’s FSD had over the others was that it navigated onto the freeway for part of the drive, while Cruise is limited to 25 MPH and Waymo chose to stay on the surface streets (as a matter of policy at present, too). Traffic appeared to be relatively light in the video, so that gave Tesla the 5-10 minutes of edge. 

This is something the three of them say they’ll be doing more in the future as the technology by these companies (and possibly others) improves. 

Want to see the full unedited videos from each car? Omar posted that for us all to see. I’ll go ahead and skip recapping that, but here’s the tweet with that video:

The Great Self-Driving Race: @klwtts, @MinimalDuck and I made Cruise, Waymo, and Tesla Full Self-Driving Beta race each other across San Francisco. Which car do you think will win?

Watch in 4K: https://t.co/7HVMKcGy9w pic.twitter.com/ktz6QLmn4P

— Whole Mars Catalog (@WholeMarsBlog) June 20, 2023

The State Of Robotaxi Tech

While this is a race, the race isn’t for who’s the fastest, but who gets a passenger safely there with minimal problems. In that regard, it’s more of a toss-up. Tesla’s technology impressed them with its ability to do real-time rendering of the environment based on its camera systems, while Waymo and Cruise use lidar, pre-mapped routes, and have several backup/failsafe mechanisms at play. At the same time, though, Tesla’s FSD Beta would, given enough tries in different conditions, likely show more problems like disengagements, even if it can run most anywhere.

Why This Video Matters

Once again, the concept of a race is a little silly on the surface for robotaxis. There are self-driving car races, and those are important ways to test technology, but nobody’s going to load their families and their belongings into an experimental self-driving racecar. The idea with robotaxis includes efficiency, but it also includes safety as the more important consideration. People would rather get there just a little slower if it means arriving outside of a pine box, right?

So, the real important test here isn’t about numbers (quantitative) as much as the whole experience and the totality of the circumstances (qualitative). Who gets there first is important, but who gets there with the least problems is also highly important. So, these videos (this one and future ones like it) should be viewed in that light. The overall data collection and experience is key.

It’s also pretty much the only way to get anything resembling a fair comparison. FSD works anywhere, at least when it’s working (there are still plenty of disengagements and errors being worked out). Waymo and Cruise only work in places that have been mapped for the vehicles and have things like backup drivers who can remotely help the vehicle get through a tough spot or even show up in person to retrieve a malfunctioning vehicle.

But, that’s just how things are in 2023. Things have changed so much over the last few years, even if full self driving and widespread robotaxis didn’t happen as quickly as many of us had hoped. This is a changing environment, and the capabilities of each of these contenders in the space will change over time, too.

There’s also the wider benefits that come as all of these approaches mature. Robotaxis are poised to play a significant role in the future of transportation, as they offer numerous benefits that can transform the way people travel. One of the key advantages of robotaxis is their potential to reduce traffic congestion in urban areas. By optimizing routes and navigating traffic efficiently, these self-driving taxis can contribute to smoother commutes and decreased travel times.

Another important aspect of robotaxis is their ability to improve road safety. Autonomous driving technology can greatly minimize human errors, which often cause accidents. By analyzing traffic situations in real-time and making safer decisions, robotaxis can help lower the number of collisions on the roads.

Moreover, robotaxis are likely to be electric vehicles, which means they can aid in reducing greenhouse gas emissions and air pollution. This shift towards cleaner transportation options can have a positive impact on the environment and help combat climate change.

Robotaxis also hold the potential to increase accessibility for individuals who face challenges in driving, such as the elderly, disabled, or visually impaired. By providing convenient transportation options, robotaxis can grant greater independence and mobility to these individuals.

So, with all of this in mind, it’s good to see not only that robotaxis are increasing in capability, but that there’s an effort underway to get a good look at how it’s going from time to time.

Featured image by CleanTechnica.