Whichever company can draw the best digital atlas, update it at lightning speeds and control who can use it has an edge on future industries from robot taxi and package delivery fleets to personalized shopping apps.
Sourced through Scoop.it from: www.govtech.com
No surprises HERE and pun intended. We learned the hard way at GeoSmart that existing Government maps were not suitable for car navigation and definitely not for more advanced ITS like autonomous cars. The company was forced to build a mapping car to drive all of New Zealand, setting back our plans for the introduction of high quality car navigation in NZ significantly by a couple of years.
There is an irony now that TomTom bought the company that they are the only ones with access to that fully driven data-set of every road in the country to 15cm accuracy including inclinometer, camber and other data. That’s a gold mine now. I know how long it took to drive every road and in the case of major roads, in both directions. Data is certainly king.
Great that HERE now has an owner in the industry which will no doubt provide the funding needed to delivery the quality required for autonomous or driverless cars. Not only do you need all roads to be driven with LiDAR and a host of other technologies (and that’s just the data collection) It takes huge amounts of data processing as well once it’s mapped; and those maps need to be current all the time because there are so many factors that impact on information a vehicle needs in order to make the right decisions.
These include name changes, speed zone changes, minor and major improvements, such as road alignment (removing corners from roads for example. It also needs to know about temporary road works, events, signals and right of way; and much more.
Unless they have similar expensive data collection technologies on board, to those used by map data collection cars, which costs more than the average car by itself, they will only be able to safely drive on roads already certified for them, or the risks will be very high if passengers aren’t ready or able to intervene.
Say for example a slip, or flooding where it can’t be established how deep the water is; conditions like the black ice we are currently experiencing on many roads this winter which is frequently invisible, the ambient light conditions we have in New Zealand, especially in winter, which make streets signs hard to machine read, even with a lot of processing in a lab!
These are just a few reasons why the purchase of HERE was so important. What we are now going to have is a global race to get the best possible data. Whilst HERE as a Nokia subsidiary that didn’t make cars, wanted to collect and sell data to all and sundry, would you expect a consortium of car manufacturers who now own the company to be equally magnanimous in sharing map data with their competitors?
Traditionally it is the unique features and benefits that car and truck brands focus on in order to win market share in a highly competitive market. If BMW, Audi and Daimler (Mercedes Benz) now own the company and their map data, do you think they will be keen to offer that same crucial information to competitors wanting to build and run autonomous cars in Europe, North and South America, companies like Toyota, Mitsubishi and Honda?
If you spent billions of dollars on this data and technology, I’m sure you would want a competitive advantage. It is certainly an exciting time and we will start to see where they have the best data, because those will be the cities and countries with least risk where we will see the first volumes of autonomous cars emerge.
Whilst Governments are clambering to enjoy the benefits of platooning and other travel demand and capacity flow efficiencies from driverless cars, they will not want to be known for the first major crashes that occur as a consequence of the use of poor quality map data.
I’ve driven in a lot of countries with car navigation and had to interpret many a map, which as a human is not too difficult. As recently as a week ago I drove to a subdivision in a rural Auckland area which was not on the Google map even though the house I was visiting had been occupied 4 years ago and of course the road was there well before that. Fortunately the road and the street address were was on my trusty old TomTom.
One of my frustrations in map brand comparisons over the years for consumer reports was they always tested them on old urban roads that have been there for ever and the predictable outcomes were that they felt and stated that there was little difference between one map company/product and another. However owners of car navigation systems know different, even though there is a known law in the software industry that purchasers will support a bad software decision once they have made it, which further blurs the lines.
In Fleet Management some companies profit from the difference. For example in New Zealand commercial freight companies pay road user charges based on distance traveled on public roads, but not on private property. The RUC data their taxes are calculated against are based on electronically collecting the driven data, based on matching it to the map used by the Fleet Management company. It is known in the industry that it can be advantageous to use a map data-set that is less than accurate because the likelihood is that it will report more on-road driving interpreted as off-road, because it has a computed rather than a driven road centreline. Those map data-sets are also cheaper to obtain and license.
Ask one of those truck operators using one of those systems if they would be happy to have the truck drive itself, based on the road width and centreline information they have been using for their road user charge tax rebates. What do you think the answer will be? I know what mine would be, and I know how many more truck rollovers we would be seeing if they tried to use the cheaper road data-sets that consumer researchers say are fine. I wonder who takes the liability for that?
It all sounds so easy; and it is on certain highways and it will be for future highways that are designed to defined specifications including those needed for Intelligent Transport Systems. There aren’t a lot of those around the world right now.
Next time you go out for a drive in the country and the nav unit tells you to go to the nearest road, even though you are on one, or can’t find your destination, imagine you are in a car or truck with no steering wheel or controls and ask yourself if near enough is good enough.