Practical Ways AI Could Make Autonomous a Reality

Artificial intelligence (AI) and machine learning are the talk of the tow, and they have a wide array of applications.

One of those is intelligent mobility and the use of simulation software to improve integrated transport planning. Robin North, CEO and lead technologist of Immense Simulations, adds that there is also a “big trend towards digitalization, being able to automate certain processes and getting data both organized as well as shareable”. After all, without data there can be no AI.

Another trend he sees is: “Around sharing and the leasing of assets – whether than be infrastructure and vehicles, then there is the sharing of services such as trips, data, mapping or digital assets.” The significance of new technology is that is has made all of these possible, in his view, to do at a lower cost.

He adds: “Mobility is being driven by four business models and technology trends. You have autonomy, connectivity, electrification and sharing that are disrupting business models of mobility service providers and it is changing the automotive technology base. Connectivity is enabling infrastructure and people to be smarter through optimization. Over time it will change the economics of the way that services are delivered once we reach Level 5 of autonomy.”

He reveals that consumer expectations are changing too: “People are expecting transport and mobility to be seamless and pay-as-you go for whatever you want to do. They are realizing that owning a depreciating and under-used asset, such as a car, may not be the best solution for their mobility.”

Business model innovation

Data-driven technologies, such as digital transformation and AI, have also led to business model innovation with platforms such as Uber. They have also led to other platforms that are, he says, “disrupting the way people do business”. He believes that the availability of data supports consumer aspirations too, opening up to different options to travel differently, while mobility providers and their partners can also exploit the data created by their customers to make money.

He comments: “People have for a long time wanted better mobility services to get from A-to-B more quickly and more cheaply. We are closer to this than we have been in the past but it’s more disruptive than the traditional ways of doing business. It’s a fertile sector and things are changing very rapidly.”

Embracing change

AI can help organizations to embrace change by automating many of the labor-intensive process.  Through automation he finds it possible for people to create a more consistent means of making decisions, which will as a result be of a better and sounder quality.

For example, AI and machine learning can help mobility and transport stakeholders to make decisions about the movement of people and goods. “They help us to build and maintain those digital worlds, and its can then be used to guide decisions by providing insights to human decision-makers,” he explains.

Simulating transport planning

By simulating transport planning an opportunity is created to look at the impact of different or specific mobility scenario models on specific days of the week during different times of the day. “It’s traditionally been hard to do this stuff and quite expensive, limited to very few conditions, but we accelerate it with AI and machine learning,” he claims before adding: “Traditionally people might look at 14-day types. You might look at a morning peak and evening peak – to replay what happened or may happen if you set up your fleets differently. It’s about enabling ‘what ifs’, about how to run the transport system. If you ask better questions, you should be able to make better decisions.”

He explains that it’s about how to manage the road space better. He asks: “Do you need more vehicles in your fleet or fewer? Where are they going to re-charge as they are likely to be all electric. If you are running as a taxi service, how many vehicles do you need for London, Milton Keynes, or San Francisco and how much do you need to learn before you incur the expense of running them?”

Capture the market

By being able to ask these questions and to model everyday scenarios, he argues that mobility service providers can capture the market more effectively, test different tiers of services, such as autonomous vehicle ride-sharing and how it operates commercially.

However, he claims: “It’s harder to make money out of mobility than people think because the public infrastructure, and how it is managed, is a key part of a mobility service and how it is run efficiently.” It is therefore crucial, in his opinion, to have a neutral platform for the public and private sectors “to engage with leads to better mobility decisions”.

He adds: “We can use the same core capability to support planning and operational decisions for infrastructure and mobility services providers. This is about looking at it from different angles and through different lenses.”