The Argo AI Approach to Deploying Self-Driving Technology: Street-by-Street, Block-by-Block
By Bryan Salesky, CEO, Argo AI
The ‘race’ narrative is often-used when defining the deployment approach of the autonomous vehicle industry. In truth, that depiction hurts the reality of how some of us are actually operating. Self-driving technology is not a winner-take-all sprint to the finish line — and even if it was, that’s not how Argo is approaching its development and deployment.
In a 2017 blog, we pointed out the state of autonomous vehicle technology, highlighting that fully self-driving vehicles aren’t likely to be common on city streets for months, or even years. The comment was overlooked at the time due to the hype surrounding the industry, but the statement and perspective remain true today.
The fact is, the development of self-driving vehicles leaves plenty of room for multiple players and many different services. That’s why the future excites me so much. The opportunity ahead is bigger than any of us can imagine. Its future will arrive gradually, and safely, if we do it right.
At Argo, we have a “street-by-street, block-by-block” mindset underlying our approach to developing and deploying technology. It’s not a race. Our goal remains to make self-driving vehicles that are safe, provide a compelling experience, and are accepted by society.
Developing a safe, reliable self-driving system
Developing a safe and reliable self-driving system starts with understanding the realities of the hardware and software used to design the vehicles, as well as the environment in which they will operate.
Certainly there will be limitations to the first-generation system. These limitations, an often-overlooked reality, are not an indictment of self-driving technology. In fact, they can help us refine the initial use-cases and define where we can operate. This facilitates safety, forcing us to be precise in creating what’s called the operating design domain — the parameters that define when and where our self-driving vehicles will operate.
These limitations, an often-overlooked reality, are not an indictment of self-driving technology.
We’re developing an SAE Level 4-capable self-driving system — a system capable of operating without human intervention under specific conditions, or within its operating design domain. And that’s what we’re designing for — a service that will only operate within specific areas of a city. The system currently under development is not intended to be applied to vehicles that will be purchased and owned by individuals or to travel outside of their operational areas. For instance, the first self-driving vehicles that Argo and Ford deploy in a service will be limited in top speed based on how far the sensors can see for safe operation. As a result, the vehicles will only be able to travel on roads that have speed limits aligned with the capability.
We believe a clear understanding of the operating domain is also a critical factor in evaluating safety performance. It’s important that there be clear rules of the road and safety metrics so we’re working with other industry stakeholders, as well as with federal, state, and city regulators and lawmakers to develop these rules and metrics to ensure the safe deployment of self-driving vehicles.
Other external factors that will limit capability of our first-generation system include heavy rain and snow. That doesn’t mean that self-driving vehicles will never be able to work in those conditions; it just means that such advanced capabilities require further innovation. To work on these longer tail issues, we’ve set up the Carnegie Mellon University Argo AI Center for Autonomous Vehicle Research.
These parameters help us focus and create opportunities to go to market. We’re thankful to be working with great partners like Ford, and soon, Volkswagen AG, who have a realistic understanding and healthy respect for the technology as they design mobility services utilizing our system integrated into their vehicles.
Creating compelling consumer experiences
Another often-overlooked factor — creating an engaging consumer experience — is key to the future of self-driving vehicles. At the same time, the services these vehicles provide need to be part of a viable business case in order to thrive in the long run.
That’s why our collaborative partnerships with automakers are critical. Our role at Argo is to develop a self-driving system that Ford, for instance, can employ when creating various mobility services.
We are currently building our system to deliver a driving experience that mimics not only the positive aspects of human driving, but also the driving behaviors that are unique to specific cities. We recognize that driving behavior differs from city to city, as does other road-use behaviors such as pedestrian, bicycle, and scooter traffic. Our self-driving system will be customized to address those specific environments.
At the end of the day, driving is local. Everyone has different expectations as to how someone should drive in a city. While our system cannot realistically accommodate every preference, we know that a self-driving car with only one set of driving characteristics will not fit every city. It’s crucial that we have a deep understanding of each community in order to build a product that everyone loves, trusts, and wants to use. When a consumer takes a self-driving car to work, they don’t need to park it, maintain it, insure it, fuel it, or worry about scratches and dents. But they will still demand that the ride experience be safe and comfortable, and that the car drives naturally without halting, stopping, or overreacting to every unexpected move made by fellow road users.
These qualities are foundational and necessary for folks to embrace the products we are building. As a result, we are spending a significant amount of time testing in the cities in which we plan to initially commercialize our technology so we can tune our system to behave appropriately.
Earning trust and acceptance
The products and services enabled by our self-driving technology need to be accepted and welcomed by society. For this reason, we are passionate about explaining our company’s mission and the purpose behind our technology to the cities and communities in which we are testing and planning to deploy.
Again, our “street-by-street, block-by-block” mentality kicks in. We have taken initial steps to educate the cities in which we operate by developing relationships with city officials and leaders, and are keeping an open dialogue with them. We’ve also engaged emergency responders who are crucial to the safety and security of the streets where our testing is conducted.
In addition, we’ve connected with local businesses, community groups, neighborhood leaders, hyperlocal media and NGOs — but have more work to do. If anyone reading this wants to reach out, feel free to contact us at firstname.lastname@example.org.
As our development continues, we will only partner with those who share our beliefs around being honest about the deployment of self-driving cars, and who are reasonable in their expectations of what the technology can and cannot do. We will work with communities to understand their transportation pain points and find ways to plug these gaps.
Our mission is not to replace the personal freedoms that driving provides, but rather to build technology that can help create compelling mobility products. We stand behind services that allow people to make the choice to take a self-driving vehicle when and where they want — in instances in which they would rather leave the driving to someone else.
Our mission is not to replace the personal freedoms that driving provides, but rather to build technology that can help create compelling mobility products.
With a potential market of three trillion miles driven per year in the United States, self-driving vehicles will only capture a small fraction in the beginning. But opportunities will be endless for companies to offer unique, and in some cases competing, services such as autonomous trucking, low-speed shuttles, middle-mile delivery, suburban and urban ride-hailing, ride-pooling, and many others.
Over time, as self-driving technology is able to operate at higher speeds and in more diverse conditions, the services will expand in their scope and scale. And they will be judged against their human operated equivalents for safety, convenience and other relevant product attributes. It’s this market environment that will determine which portion of those three trillion miles will become automated, and when.
While Argo intends to play a prominent role in the development of self-driving technology, it is worth repeating: this is not a race. We endeavor to be true to our values at all times and be honest about how we develop and deploy our technology. Our street-by-street, block-by-block approach ensures that we will engage communities every step of the way.