The car is ready; the wire isn't. That has been the shape of the electrification problem for a decade — the vehicle can drive farther than the charging network can reliably support, and the failure mode that scares people most is the one with no infrastructure at all: a flat battery somewhere without a charger in reach. A patent application published on July 2, 2026 and assigned to Hyundai Motor Company takes that failure mode directly and answers it with a piece of logistics rather than a new connector. Carried at US20260184225A1, titled "System for Charge Management and Method Thereof," the application describes turning ordinary electric vehicles into a dispatchable pool of roving chargers, coordinated by a server the way a ride-hail platform coordinates cars.

The disclosed system has three actors. A second electric vehicle is the one that needs power and requests a charging service. A first electric vehicle is the one that can supply power. And a charging-management server sits between them, monitoring both and managing the exchange. The abstract states the core loop plainly.

The charging management server may be configured to dispatch the first electric vehicle corresponding to a request of the second electric vehicle based on status information of the first electric vehicle.— System for Charge Management and Method Thereof, US20260184225A1

Read at an engineering level, the interesting work is in the messages, not the metal. The application describes the server holding status information for each candidate donor vehicle — an identifier, charger information, position, price information, communication-protocol information, and customer information. When a stranded vehicle asks for help, the server builds request information from that vehicle's data — a user identifier, battery information, position, and a request message — and uses the two together to pick a donor. In one described implementation, the selection rule is simply proximity: the server chooses the first electric vehicle positioned closest to the second, weighing both cars' locations. The chosen donor then generates response information — a response message, dispatch information, an expected charging fee, and an expected charging time — before it ever moves. It is, in effect, the quote-and-ETA screen of a service app expressed as a vehicle-to-vehicle message set.

The protocol layer is borrowed, not invented

The part a charging engineer will notice first is that Hyundai is not proposing a new physical-layer standard. The application specifies that the communication protocol supports at least one of OCPP (the Open Charge Point Protocol that fleets and networks already use to talk to chargers), ISO 15118 (the vehicle-to-grid communication standard that underpins Plug and Charge), or IEC 61850 (a substation and distributed-energy communication standard). That choice matters. By carrying the donor/requester exchange over protocols built for station-to-vehicle and grid-to-device sessions, the system treats a donor EV as if it were a mobile charge point — a station that happens to have wheels. The security handshake comes along for free: a sibling filing in the same cluster, US20260184224A1, adds certificate verification for the communication protocol and payment support, so the server checks the validity of a protocol certificate before it trusts a donor's status and settles the transaction afterward.

What actually flows through the cable is addressed separately. A second sibling, US20260184216A1, describes a vehicle charging-control apparatus that pauses charging, generates a discharge pulse using the vehicle's own battery, and then resumes — a technique aimed at conditioning the pack during a session rather than pushing current straight through. Taken together, the trio sketches the whole transaction: match and authorize it (the hero), secure and bill it (the certificate filing), and manage the current profile at the pack (the discharge-pulse filing).

Where the power electronics sit

A roving charger is only useful if the donor vehicle can move energy out of its pack efficiently, and the same July 2 Hyundai drop carries the supporting silicon. US20260189136A1, a controller for a converter with multiple power-factor-correction circuits, is classified in part under B60L 53/20 — the on-board charging branch — and describes coordinating two PFC cores so their pulse-width-modulation registers stay synchronized across control periods. That is the kind of multi-core converter control a bidirectional on-board charger needs when it has to run cleanly in both directions. Alongside it, US20260184196A1 pairs a battery with a supercapacitor to run a "virtual gear shift" torque map, and US20260190288A1 discloses a direct-cooling manifold for a power module whose flow channel narrows from inlet to outlet to hold cooling efficiency up — both signs of a company working the thermal and energy-buffer problems that a donor vehicle would lean on hard during an unplanned roadside transfer.

The reliability question a charging writer always asks is what happens when the exchange completes or fails. The application answers the completion case: when the charging module finishes supplying power, the donor's control module generates charging service information — a supply identifier, charging position, power-supply information, protocol information, charging rate, and charging time — and transmits it back to the server. That closing report is what turns a one-off good deed into a logged, auditable transaction the server can price and reconcile. The fleet-health instrumentation that would keep such a service honest is visible nearby too: US20260186060A1 describes detecting an abnormality in a battery pack by comparing a modeled voltage against the measured voltage, and US20260188777A1 details a multilayer cooling structure for a pack's power-relay assembly — the state-of-health and thermal monitors a server would want to read before it dispatches a car to give away energy it may need itself.

None of this is a deployed feature. The record is a published application, and the independent claim is broad in the way early filings are — a first EV, a second EV, and a server that dispatches one to the other on status. But the disclosed message flow is coherent and standards-anchored, and it names the real hard part: not the physics of moving electrons car-to-car, which is well understood, but the dispatch, authorization, and settlement logic that would let a fleet do it at scale. On paper, the connector is the easy part; the protocol stack around it is the invention.