How Self-Driving Cars Can Figure Out Parking

Robocars are very close to being able to drive the roads at a safety level to match humans. But what about dropping you off, picking you up, and parking or waiting?

For most teams, driving on the roads will be done with detailed maps of the road. They will know where to go on any road, and also how to tell if the road differs from the map and what to do in that situation. (All cars will have at least rudimentary ability to drive without maps for this situation.)

Those maps will include the locations of parking spaces and standing spaces on the public roads. Standing spaces can include the entrances to driveways and spaces in front of fire hydrants, suitable for pick-up an drop-off. Cities have long allocated “taxi stands” for waiting, as well as pick-up/drop-off areas, and recently I have observed such areas being allocated for use with Uber/Lyft style services.

Drop-off points and parking spots are very commonly on private land, though. How will cars know where to go if you are heading to a private destination?

For important private venues — schools, public buildings, and large buildings, the mapping companies will map them just as they do roads. Most of them would allow a human-driven mapping car to pay a visit. In addition, privately driven cars with sensors will also frequently travel to them and can gather mapping data.

Generally, such maps need a basic level of human review, and as such, buildings below a certain threshold of traffic will not get mapped as quickly as high traffic buildings. This can be solved by allowing the owners, and even the public, to map these spaces. Somebody motivated will generally be able to start with an aerial image of the private area to mark out the geometry of the space — where can cars drive, what is sidewalk, where any pick-up/drop-off points are and, as described below, waiting/parking areas. If there are not high-quality aerial image of a location (ie. it’s non-urban) owners could, at a pretty low price, request a drone overflight to help.

The cars don’t use the aerial imagery to navigate — this is just to get the rough shape. A visit by a robocar (or human-driven car) with the right sensors is needed. That gets easier and easier as more robocars are available. Such cars can have a remote human supervisor when they enter the unknown space to map it. The supervisor almost never “drives” it like a human with a wheel, rather they give strategic commands, like pointing to a spot on an image and saying to drive to that spot. The car worries about not hit anything along the way, or noticing things the remote operator might not see.

With remote operators, such mapping can be quite cheap, because we need no human for the journey to the parking lot, just while the car is navigating strange territory.

Even companies that claim they will drive without maps, like Tesla, will need maps of private territory. Public streets are required by law to follow certain patterns and rules. Things may differ from town to town but generally there is a system to figuring out what’s road and where the parking is and what places might be suitable for pulling over. Private land follows only social conventions, not laws. It’s built by whoever owns it, however they like. Street signs on private land often have no force of law. You may see complex rules saying which parking spaces are reserved at different times or anything else the lot wants to explain in human language. You want a human to make sense of it, and encode it in a way the machines can reliably use.

My expectation is that robocar companies (not just taxi companies, because people want their private cars to be able to come and pick them up on private land) will offer private lot owners the ability to choose from various systems to manage how cars come to their land. These systems will go beyond what even humans understand today. They will be allowed to designate waiting areas, something like the taxi stands of old, and instructions on how to get in and out and in what order.

It might be common for a busy building (at the extreme, imagine a stadium after a game lets out) to want a fairly involved management system. In such a system, cars would be told that to go to the stadium, they must first get permission and instructions from a computer managing the stadium lot. (They won’t communicate with that computer, their HQ computer will do that, to avoid the risk of direct communication.)

The stadium computer will manage traffic coming in, and only let in as many cars as it can handle. It may arrange priority based on all sorts of rules, such as priority for people on the curb and ready to go, for VIPs and in particular for those carpooling or vanpooling. Each arriving car will probably be told to move to a line and then to a pickup spot. The car can then tell its rider what pickup spot it will be coming to.

For cases as extreme as emptying a stadium, at first they might allow only buses to come, and those buses might move to other, more distant lots where private cars can go to receive passengers who came on the shuttle.

Office buildings will need more simple versions of this if they don’t have a parking lot. If the office building has a parking lot (as most do from the legacy days of private driving) the cars can go pick up people in the lot by going to stalls where their master awaits.

All of this needs maps and protocols, even for a Tesla. In addition, there needs to be some management of the public street at the entrances to the stadium or office building. That gets naturally managed if the stadium is only allowing in a fixed number of vehicles, and there is enforcement of rules against stopping or driving slowly by the entrance. This will make cars doing pick-ups go to less antisocial places to wait, so that they can get a signal to come get their master when it’s a reserved straight shot to where the people are waiting.

Even things as simple as grocery store lots will have management if they wish it, to control how cars go to parking spots, how delivery vehicles arrive, and how cars move to pick up customers at the door. As the number of people using robocars rises, the pick-up areas will increase in size to handle it, but that’s easy because the need for parking drops.

Good pick-up areas with limited space can be handled in one of two ways. If there is space, the best is perpendicular. You have a long series of perpendicular spots, and cars arriving just go to a spot to load or unload. With much less space it is also possible to just have a single lane. Cars would come into the lane in a group and all stop right in the lane to do load/unload together, then all would leave at once, and be replaced by the next batch of accepted cars. Even a small single lane should be able to do a nice, high volume of pick-up and drop-off compared to the way it works today.

It surprises me more places don’t use the perpendicular approach today. That’s how Singapore Changi airport has done it for years, and it has a deserved reputation as one of the best-managed airports in the world. When I see the long, snaking taxi lines at some airports that load just a few customers at a time, it baffles me.

Valet style

One thing that will

be very popular is for lots to designate areas for robocars parking in “valet” mode, which is to say, packing themselves in tight for best use of the space. This may be as simple as laying out a space in the map of the lot and telling cars to arrange themselves, or it may involve a server which manages the lot dynamically, changing the size of the valet area based on demand, and even having the cars report their size and expected duration in the lot so the system can decide where to put each car if it wants.

While a central computer managing the lot can do a lot — parking lots won’t have to run their own, there will be companies offering this as a low-cost service – the cars can do quite a bit on their own, including managing exit from the valet area. For example, you will know that in the valet area, you can park in a way that blocks in other cars. Later, when a car wants to move, it doesn’t have to communicate with anybody. It just starts to slowly move, just an inch, closing the gap with the car that blocked it. That car will then move, and that will cause the next car to move and so on, until the whole column exits, letting that first car out. The column then re-parks itself in the empty space. This can also be done on the sides of roads, with cars turning on a turn signal to say they want to leave. There, as long as one empty space is kept per row, the cars can “move the space” to put it next to the exiting car, which pulls out, almost like the other cars are not there.

This post first appeared on Forbes.com. To see the original, click here.