What’s the point of low-speed automated shuttles?

We’re going to get a spectrum of driverless road vehicles: from pod-vehicles for one person with groceries or luggage, mid-sized buses carrying 30-40 people, to even longer buses on dedicated lanes. The largest portion of this future, autonomous fleet, by vehicle count, would be on the smaller side (1 to 5 passengers). Of course, we can expect every other workable size: smaller vehicles to carry 10-15 kg of goods are already being piloted; articulated buses for 200 passengers are under consideration.

Absolutely critical in this spectrum is the 10-passenger shuttle that is a current darling of any press that focuses on transit-lite applications but is, to date, a failure for viable deployments.

Let’s look at this 10-passenger shuttle, its original purpose, its current failings, and its critical place for the future.

Carlos Holguin, leader of the CityMobil2 (2012-2016) team that first specified the 10-passenger shuttle writes here:

“We requested several robotics companies to propose a fleet of at least 6 vehicles and at least 60 seats. The companies simply divided 60 seats by 6 vehicles and designed 10-passenger shuttles. That's where today's unofficial "10-passenger shuttle" standard comes [from].”

The Holguin-CityMobil2 specification required the modest size and low-speed because theirs was an experiment. These vehicles needed to run on calm, local streets to test the viability of short, common routes for passengers going modest distances and perhaps taking children or carrying small loads on routine trips. This was not for highway driving and not for replacing municipal buses— and not even remotely able to replace an intraurban rail system. This was solely a proof-of-concept for a local short-distance, capillary system — nothing operational, nothing permanent, nothing fast or massive.

The CityMobil2 experiments needed regulatory permission — or at least a temporary exemption — to operate on the local, non-arterial streets of the several, forward-thinking European cities hosting the experiment. Extreme caution was necessary to “avoid the crash of a prototype that would destroy public trust” (Holguin). So low-speed, dedicated lanes, hyper-cautious braking algorithms, and onboard attendants were required.

All of this, in addition to the fact that these experimental machines would be hand-crafted in single-digit numbers, meant that each would cost on the order of a quarter-million euros, and operating them would cost twice more per annum for a city trial. All of this was deemed acceptable for an experiment funded by the European Commission.

In his article, Holguin identified that futurists and the press eagerly touted this CityMobil2-specified experimental, low-speed shuttle as a go-to component of the transportation future to be adopted. This led to many start-ups (besides the CityMobil2 winner, EasyMile), such as Navya and the older 2GetThere, to start building look-alike shuttles. All with similar features, none (so far) with the ability to be viable, driverless vehicles, without safety attendants, or without protected or isolated traffic lanes.

This lack of viability is three-fold. The low-speed shuttle systems so far available are missing the safety, security, speed, all-weather viability, and braking-comfort elements necessary for a driverless, public system (sudden breaking can throw people to the floor, as of course happens on existing bus systems, as well — but those vehicles have an operator that can address any injury). Secondly, the local, municipal transportation environments that would benefit from such technology are not suitable without street-intrusive preparations, expensive oversight, and at least temporary regulatory relief. Thirdly, the current cost of the still-inadequate technology is far out of line for the dubious benefit of repeating the Citymobil2 experiment for the umpteenth time.

As a transportation system, these shuttles are not yet a financially sustainable solution. Not today.

In a recent article, Colin Barnden quotes Missy Cummings:

“The entire driverless car community took what was basic research and made an assumption that it was ready to deploy. They were just wrong and they didn’t do their homework. They made the wrong decision that the technology was much closer to deployment than it actually is.”

We’ll pay for this blunder in more ways than one.

The first is that mayors of cities eager to check off environmental, FOMO, or smart-city tickboxes will invest public subsidies in the misguided belief that they are preparing their city, their public transportation system, and their transit users for a pending technology. Again, Holguin:

“[we see] low-speed shuttle "pilots" pop up in all the wrong places, as a recent study confirmed: parks, pedestrian areas, dense city centres, business and technology parks... What's wrong with that? Competing for space with pedestrians, the shuttles' speed gets even lower, and a distance that a human driver would normally cover in 3 to 5 minutes, can take three times as much in a low-speed shuttle: people will use it once or twice for curiosity, but it will never make them leave their car. In dense city centres, where mass transit is and will always be the most efficient transport mode, low speed shuttles not only take up as much space as cars, but they create more congestion. Basically, in those places and conditions, low-speed shuttles provide little to no service at all. Therefore, cities end up concluding that, with this configuration, low-speed shuttles are useless, and end up rolling back their experiments.”

This misapplication will give these machines a black-eye that will take years to outgrow, delaying the appropriate solution and viable deployments.

But we also want to be cautious that this shuttle not be demonized. While so many things are wrong with its current state, status, and adoption, three of its aspects are critical for workability going forward: their size, their ease of boarding, and their (eventual) suitability for suburban and small-town transit.

What that means is that we need to stop trialing expensive experimental vehicles that are not ready for the applications we need them for and signal readiness to trial — for the appropriate circumstances — a vehicle that is indeed ready for us.

Here is what we need to have addressed before any more orders are placed for autonomous, low-speed shuttles:

Safety

The crashworthiness of the proof-of-concept shuttles currently available is not suitable for head-on or side-impact crashes. Even rear crashes above a few km per hour could result in unacceptable injuries from whiplash for passengers seated forward or being thrown to the floor for unseat-belted passengers facing rear or standing.

If such vehicles will require seat belts, how will liability be assigned if a passenger not wearing a belt is thrown against (and injures) another passenger who is wearing a belt?

Locations

Find locations where:

1. passengers will switch from private vehicles to shuttles on a permanent basis (not just to take an amusement ride and revert to personal vehicle use) — you may include incentives programs;
2. there is an expected ridership such that a service, when fully operational, would not require more than the current average per-boarding subsidy;
3. intersection crossings (that provide an elevated risk of crashes) are such that the vehicles selected are sufficiently crash-worthy — in other words, transit passenger injuries per passenger kilometre cannot be higher or more injurious than on current bus systems; the injury spectrum may be different but it cannot be worse from a medical harm perspective. One way to measure this is to compare the crash-safety features of the shuttles proposed to the crash-safety features of human-driven shuttles of the same size, weight, and occupancy.

Security

What measures can be provided for passenger safety in the event of human mischief (assault, robbery, harassment)? This is especially difficult because we have little comparable experience of vehicles without proximate human oversight, i.e., without the authority to intervene.

If this problem is not addressed, how can it make sense to trial 10-passenger driverless vehicles, if they cannot be used without a human operator or attendant? Doing so is an example of not considering the full system aspects of these new vehicles. Rather it is a case of simply being enamored with driverlessness and its hyped promises, which is insufficient for an operational public transit system.

Speed

While highway speeds are unnecessary, and perhaps not even appropriate for an intra-urban short-haul vehicle that may have standing or un-belted passengers, it is the case that to attract choice users to these systems, the vehicles must be able to take its passengers to their destinations in times roughly equivalent to the use of a private vehicle.

Such time comparisons should include waiting and boarding times for the shuttle and circling and parking times for the private vehicle. Hence, to level the playing field, and make this comparison work, such shuttles may need to travel for short periods at speeds modestly higher than 40 km/h, given appropriate circumstances.

Weather resilience

To be viable, a transportation service vehicle must be able to operate in most weather circumstances. Conditions that are a barrier to using light vehicles or buses today would remain acceptable exceptions, but heavy rain and non-blizzard snow conditions cannot be a reason to suspend service. This constraint may be addressed by infrastructure considerations that would benefit all vehicles, but it would not be viable to require special infrastructure just for driverless shuttles.

Braking-management

Stopping comfort is necessary for a driverless, public system (sudden breaking has thrown people to the floor, as happens on existing bus systems, as well).

Cost

Clearly, if all these problems were solved, the first such vehicles would be made in very small runs. Even if not handcrafted, they will still be expensive. If municipal governments want to “try out” this technology for its transit purposes, then they should specify exactly what this technology must do before they make any further investments.