Sub-system
Satellite positioning is accurate to a few metres, and a lane is about three and a half metres wide. Closing that gap — knowing which lane you are in, and where within it — is the localisation problem, and how much prior map a system needs to solve it is one of the field's genuine strategic divides.
In one sentence
A high-definition map records road geometry, lane boundaries, signs and landmarks to centimetre accuracy; localisation is the process by which a vehicle determines its precise position and orientation relative to that map or to its own observations.
A vehicle localises by matching what it currently sees against a stored representation. Lidar systems match point clouds against a recorded three-dimensional map; camera systems match visual features. Either way the vehicle finds the position and orientation that best explains its observations, which is far more precise than satellite positioning alone.
The strategic question is how much of that stored representation should exist in advance. A prior map lets the system know what is beyond the sensors' range, which is genuinely useful. It also has to be built and kept current for every road the vehicle will drive, which is a large and permanent operating cost.
Lane boundaries and centrelines with curvature, the position and meaning of signs and signals, stop lines, crossings, speed limits and the connectivity between lanes at junctions. It is a machine-readable description of the road's rules and geometry, not a picture, and its value is that it is known before the vehicle can see it.
Roads change constantly — resurfacing, roadworks, new markings, altered junctions. A map that is out of date in a way the vehicle trusts is worse than no map. Keeping one current at national scale requires either a dedicated survey fleet or continuous crowdsourced updates from production vehicles, both of which are ongoing costs rather than one-time builds.
The argument against prior maps is scalability: a system that requires a current map can only operate where one exists, which caps deployment at the mapping operation's coverage. Building a system that drives from what it sees removes that cap, at the price of solving a much harder perception problem. Both approaches are in commercial use, in different products.
1 of these is marked as a chokepoint: a handful of qualified suppliers, a multi-year lead time, or a single geography.
Survey fleets or crowdsourced pipelines producing and validating map data continuously; a permanent operating cost.
Correction services improve satellite accuracy substantially, providing a strong prior even when map matching does the fine work.
Satellite constellationsLocalisation ultimately compares live sensor data against a stored representation; its accuracy depends on both.
Perception sensorsOther pages in this map that name Maps and localisation as something they cannot do without.
What each company supplies at this step, and — where a public figure exists — its share of this specific market — with what that share measures, the period it covers and who published it. Some rows also show the company’s own reported revenue for the segment covering this step, which is a different thing: it says how much this business matters to that company, not how much of the market it holds. Not a ranking and not a recommendation.
Builds crowdsourced map data from production vehicle fleets.
Operates mapping and automated driving programmes.
Supplies the vehicle systems and software integration around localisation.
Pursues an approach that minimises dependence on prior high-definition maps.
Supplies the map layer several carmakers use rather than build, including the lane geometry assisted driving needs.
The carmaker-owned mapping company, and the reason the European industry has a map it does not buy from a platform.
Supplies the high-definition maps most Chinese assisted-driving systems localise against.
Supplies the correction services that turn satellite positioning from metres into centimetres.
Whether crowdsourced map maintenance proves cheap enough to scale nationally.
Whether map-light systems reach the reliability of mapped ones in complex environments.
Whether map data becomes a shared industry utility or stays a competitive asset.
Each page explains one technology in plain language, states what it depends on, and names companies by what they supply at that step. Company roles are described qualitatively and deliberately carry no market shares, revenue figures or rankings — those change faster than an explainer can, and a stale number is worse than none. Ticker links point at company pages on this site and are provided for reference only.
Nothing here is investment advice, a recommendation, or a forecast. A company named on a page about a technology is not thereby a good investment, and the chokepoints described are structural facts about supply chains rather than predictions about prices. Technology moves; where a page describes something as unresolved or in development, that was true when it was written.
Plutux is not an investment adviser. Market data and AI-generated analysis are for information and education only, not investment advice. Disclaimer