Cruise Buys Voyage To Up Its Robotaxi Play

Aside from Volkswagen’s massive Power Day and battery announcements earlier today (or yesterday if you are in Europe or on the US East Coast and being technical about it), but there was another major news item on the automotive vehicle world today — Cruise bought Voyage.

Cruise, if you’re not familiar with it, is a top autonomous driving startup. You can learn more here or here. CleanTechnica will also soon be publishing a podcast interview with Cruise Senior Vice President of Government Affairs and Social Impact Robert Grant.

But today’s focus is on Voyage, the smaller autonomous driving startup that Cruise just bought.

GIF courtesy of Cruise.

I thought we had never covered Voyage before, but we actually did write about it twice in 2017 when it launched, spinning out of Udacity. At the time, it was focused on transporting members of a San Jose retirement community around the community, and that’s been a large part of its focus since then (with the service expanded to The Villages in Florida) . However, that will all change now that it’s been scooped up by Cruise. Well, as some have put it, this acquisition is just a talent scoop for Cruise — about getting the people at Voyage, not the tech itself.

GIF courtesy of Cruise.

Voyage co-founder and CEO Oliver Cameron seems to say as much in his post about the acquisition: “The self-driving industry is consolidating, and the leaders of a trillion-dollar market are fast emerging. After being intimately involved with the AV industry for the last five years, I can say with certainty that Cruise — with its advanced self-driving technology, unique auto-maker partnerships, and all-electric purpose-built vehicle with no human controls — is poised to be the clear leader. Now, with the addition of the customer-obsessed Voyage team, Cruise is well-positioned to deliver the best self-driving product in the world.

“From Cruise’s perspective, there are only so many people out there with intimate knowledge and experience working on self-driving vehicle technology for several years, and as it expands, it is probably worth pulling more of those players in and seeing what they learned or can contribute to ongoing development toward robotaxi-level tech. Cameron certainly seems to be excited to be joining Cruise, which he claims is #1 in the industry (which would mean ahead of both Waymo/Google and Tesla). “Over the last seven years Cruise has developed the most advanced self-driving capabilities in the world. Today, their vehicles are adept at handling the most complex driving challenges that San Francisco can throw at them, uniquely positioning Cruise to rapidly expand into other complex cities across the world while others go back to the drawing board. Just recently, Cruise reported that in the second half of 2020, their self-driving technology improved to more than 60,000 miles between reportable disengagements. And in the final three months of 2020, Cruise had zero reportable disengagements. Wow!

“Voyage’s experience and development of Commander (our self-driving A.I.), Shield (our collision mitigation system), and Telessist (our novel remote assistance solution) will only supercharge Cruise’s goal of superhuman driving performance.”

Kyle Vogt (Cruise President & CTO), Oliver Cameron (Voyage CEO). Image courtesy of Cruise/Voyage.

If that looked like a bunch of bold statements, you may want to sit down for this one: “Cruise is the only autonomous vehicle company with the ability to produce fully integrated self-driving vehicles at global manufacturing scale.”

That’s quite a statement. It’s basically throwing all other autonomous driving startups (including Tesla?) under the electric bus, while also indicating that Cruise is getting close to mass production.

“Not only is Cruise’s self-driving software uniquely positioned to rapidly scale from San Francisco to dense, urban cities around the world, but Cruise’s self-driving hardware is, too. Cruise’s deep partnerships with GM and Honda deliver the crucial engineering, manufacturing, and safety expertise required to scale a handful of self-driving vehicles to millions and do so with unrivaled cost efficiencies.”

Cameron indicates that the Voyage team’s role will be helping Cruise to turn all of that tech leadership into a business that makes money — which is harder than it sounds.

Some members of the Voyage development team will also seemingly help to refine the Cruise Origin.

Cruise Origin autonomous shuttle
Image courtesy of Cruise.

I was curious about Voyage’s previous development, so went ahead and read through its previous 5 blog posts, which were published from July 2020 to December 2020. Below are some of the highlights from those.

Voyage’s Commander self-driving AI in action.
Voyage’s Commander self-driving AI in action.

In November, Cameron wrote about Voyage’s new “brain” — the 3rd generation of its self-driving A.I. Here’s a key section of that piece: “The first input to each driving decision is our cutting-edge perception module, responsible for identifying dynamic objects and static obstacles to pay close attention to. Commander’s perception module combines a deep learning-powered computer vision system—trained on millions of data points—with multiple reliable classical computer vision algorithms as fallbacks. This approach brings to bear the benefits of deep learning’s intelligence while retaining the robustness of tried-and-tested robotics, ensuring we detect every dynamic and static object.

“Our computer vision approach is predicated on both measuring and estimating depth. Our 3D depth sensors feed physics-accurate depth measurements into our deep learning algorithms, while our 2D sensors—cameras—feed pixels from which estimate the depth of objects. Both the measurements and estimations are then fused into a single representation, resulting in a view of the world that is both accurate and information-rich.

“With precise perception input, Commander then predicts what the identified objects are going to do next. Commander’s prediction engine uses a combination of advanced probabilistic models, behavioral models, and high-definition maps to predict what’s going to happen around our robotaxis. For example, if a pedestrian is walking in the direction of the roadway ahead of our robotaxi, our prediction engine outputs a range of possible futures for where that pedestrian could be located a few moments from now. Commander’s prediction engine then weighs thousands of potential combinations of scenarios, selecting the future it has the highest confidence will come true.

The art of prediction is one of the great problems in self-driving, and we are really proud with just how superhuman our prediction engine has become over time.”

In a followup article in December, Cameron explained how Voyage developed a behavior planner that was smooth, safe, and more human-like. Here’s a key segment of that:

“Voyage has pioneered a new form of decision making that combines the verifiability and reliability of the classical approach with the modern approach’s intelligence. The result is a technique we call High-Quality Decision Making.

“High-Quality Decision Making is fueled by two models, one optimization-based (i.e., reliable) and one machine-learned (i.e., intelligent), with each serving different responsibilities. The optimization-based model is responsible for ensuring our vehicle always adheres to the rules of the road (e.g., preventing the running of stop-lines, or getting too close to pedestrians), while the machine-learned model—trained on rich, historical driving data—is responsible for tapping into its vast history of experience to select the most human-like decision to make from a refined list of safe options.

“Combining these models—optimization-based and machine-learned—in the way we have results in deterministic decisions (crucial for a measurable and validated safety case), while delivering smooth, human-like decision-making. What’s more, our decisions only improve over time with the addition of rich data.”

Our self-driving A.I. slows down to understand if the overtake can be made without interfering with the pedestrians. GIF courtesy of Voyage/Cruise.


“In the above instruction, a member of our data team took a look at the recorded driving data and codified—explicitly—the actions they would have taken if they were in the driver’s seat. In this case, the author says we should have navigated the vehicles and the pedestrian on the right, and that we should have done so slower than usual.

“What’s special here—and what makes this data rich—is that other machine learning approaches (e.g., end-to-end) may feed recorded driving events with just subtle signals (i.e., driver take-over) into their model, without also feeding explicit detail on was right or wrong in the event. It is then up to the machine-learned model itself to infer exactly what the self-driving A.I. did right or wrong, with the theory being that with enough data, it will figure this out by itself. Our approach explicitly feeds our machine-learned model detailed instructions—interpretable by both machine and human—on exactly how a human would have handled the situation. With this smaller, richer data—and by bounding what the machine-learned model is tasked to do—we have achieved great results.”

How is this different from what Cruise does or what Tesla does? Well, we don’t actually know, because we don’t have this level of nuance or explanation on what goes on behind the doors and windows of these companies as they refine their systems.

In October of 2020, Nina Qi announced a new partnership with First Transit, “the industry leader in autonomous mobility solutions.”

In August of 2020, Cameron introduced Voyage’s 3rd-generation robotaxi, and that followed the introduction of Telessist, which “combines the intelligence of a human driver with our self-driving A.I. to handle edge cases.”

Zachary Shahan

Zach is tryin' to help society help itself one word at a time. He spends most of his time here on CleanTechnica as its director, chief editor, and CEO. Zach is recognized globally as an electric vehicle, solar energy, and energy storage expert. He has presented about cleantech at conferences in India, the UAE, Ukraine, Poland, Germany, the Netherlands, the USA, Canada, and Curaçao. Zach has long-term investments in Tesla [TSLA], NIO [NIO], Xpeng [XPEV], Ford [F], ChargePoint [CHPT], Amazon [AMZN], Piedmont Lithium [PLL], Lithium Americas [LAC], Albemarle Corporation [ALB], Nouveau Monde Graphite [NMGRF], Talon Metals [TLOFF], Arclight Clean Transition Corp [ACTC], and Starbucks [SBUX]. But he does not offer (explicitly or implicitly) investment advice of any sort.

Zachary Shahan has 7126 posts and counting. See all posts by Zachary Shahan