Tesla patent describes autonomous inductive charging depot for robotaxi fleets with price-signal scheduler
A patent application published this week describes a charging depot in which robotaxi vehicles dock autonomously onto floor-integrated inductive rails without human intervention. The claims cover a coil array embedded in the pavement, a short-range handshake between vehicle and depot, and a scheduler that sequences fleet charging against electricity price signals and demand charges. The application states an alignment tolerance of plus or minus 100 millimeters and describes depots intended to serve 'dozens of vehicles in continuous rotation.' Three inventors are listed, all with prior filings on wireless charging pads. The application was filed twenty months before publication. Tesla has not announced a charging facility of this design, and the claims remain pending.
Agent Analysis
This is the first autonomy artifact in the corpus that improves unit economics rather than raising risk: unattended docking, embedded coils, and a scheduler that sequences charging against price signals, all designed for vehicles rotating through a depot without staff.
Autonomous docking with plus or minus 100 millimeter alignment tolerance
OperationalScheduler sequences fleet charging against price signals and demand charges
StrategicFiled twenty months before publication with no announced facility
Strategic
→ Fleet economics require depots that need no driver
→ Charging cost becomes a managed input
The fun part of autonomy is the driving. The patent is about the parking and the power bill, which is where the margin actually lives and where the announcements never go.
Scheduler claims reference demand charges
Insider SignalNobody engineers around demand charges unless they have modeled the fleet's power bill at scale.
Filed twenty months earlier, published with no fanfare
Behavioral TellTwo years of quiet patent work with no press release is a program that is either early or private.
Depot language anticipates rotation, not a pilot
Hidden AgendaThe claims describe dozens of vehicles in continuous rotation while the disclosed fleet remains under 400.
Every autonomy press event is about the car. Every margin deck is about the depot.
They are patenting the moat, not the car.
Agent Debate
This is the only autonomy artifact in the corpus that improves unit economics rather than adding exposure. Autonomous inductive docking at plus or minus 100 millimeter tolerance with no human intervention removes depot labor from a fleet's cost, and the scheduler claims sequence charging against electricity price signals and demand charges, treating power as a managed input. The weakness is timing: filed twenty months ago, published this week, unannounced, claims still pending.
- Autonomous inductive docking with plus or minus 100 millimeter alignment tolerance and no human intervention
- Scheduler sequences fleet charging against electricity price signals and demand charges
- Depots intended to serve dozens of vehicles in continuous rotation, against a disclosed fleet under 400
- Application filed twenty months before publication, with no announced facility and claims pending
The interesting part of autonomy is the driving, and nobody files a patent about the driving. This one is about the parking space and the power bill, which is where the margin actually lives and where the announcements never go. Twenty months of quiet work surfacing with no announcement attached is either a program that is early or one somebody wants to keep off the podium.
- Claims cover embedded pavement coil arrays and a vehicle-to-depot handshake rather than any driving behavior
- Demand-charge optimization only pays at fleet scale, which is a margin model and not a demonstration
- Filed twenty months before publication and published without a facility announcement
- Depot language anticipates dozens of vehicles rotating continuously while the deployed fleet remains under 400
- The design targets an unattended depot operating a continuous rotation rather than a supervised pilot site
- Electricity cost is being engineered as a managed operating input rather than absorbed
- Nothing in the filing indicates an announced or committed deployment
Applications publish on a statutory schedule independent of company communication plans. A twenty-month filing lag is routine and carries no information about readiness.
Nothing about patent timing is accidental in how it lands. An invention filed twenty months ago surfacing this week, unannounced, is a program being held below the waterline until the fleet justifies putting it above.
- Application filed twenty months before publication
- Tesla has not announced a charging facility of this design and the claims remain pending
Claims are drafted to a scope that survives examination, so dozens of vehicles in continuous rotation describes the design intent of the invention, not a deployment schedule. It should not be read against the current fleet count.
Engineers size depots for the fleet they intend to run. A depot specified for dozens of rotating vehicles is a sizing decision made against a fleet that does not exist yet, which is the same gap the under-400 disclosure opened.
- Depots intended to serve dozens of vehicles in continuous rotation
- Deployed robotaxi fleet disclosed at fewer than 400 vehicles with safety drivers in all jurisdictions
Neither agent disputes that the design is operationally serious; the disagreement is whether it is timely. The analyst is right that patent scope is drafted for survival rather than for forecasting, and the gossip read is right that the sizing language betrays an intended fleet the public record has not reached. The unresolved item is where the tolerance came from, because plus or minus 100 millimeters is tight enough to imply physical fleet engineering rather than a schematic, and a published depot specification with a committed vehicle count would tell us whether this is an early program or a private one.