a passport with an unmeasured vital sign

Problem statement

Under Regulation (EU) 2023/1542, every electric-vehicle and light-means-of-transport battery, and every industrial battery above 2 kWh, placed on the EU market from 18 February 2027 must carry a digital battery passport, whose contents include "performance and durability data" and information supporting repair, reuse and recycling; the Commission's July 2026 implementing decision adopted six of the eight CEN-CENELEC JTC 24 harmonised standards for the passport's data architecture. Yet the single number that second-life buyers, insurers, recyclers and used-car purchasers most want from that passport — the battery's state of health (SoH) — has no agreed measurement basis. A 2025 analysis from TU Munich's automotive-technology institute states that "neither industry nor academia has a consensus on a vehicle-level SOH definition or a standard procedure for its measurement," that the regulation's SoH articles "lack technical detail and merely state that it must be given as an energy-based value for electric vehicles, named the state of certified energy (SOCE)," and that "the regulation misses a definition of standard measurement and reference for obtaining the SOH or SOCE, which substantially impacts the values obtained." The number a vehicle reports today "is supplied by the original equipment manufacturer (OEM), which can have an incentive to manipulation so that it does not necessarily coincide with actual battery aging." The unsolved problem is a reproducible, scalable, manipulation-resistant vehicle-level SoH measurement that can be written into a passport before the passport becomes mandatory.

Why this matters

SoH is the value on which an EV's residual value, warranty disputes, insurance premiums, and its routing into reuse versus recycling all turn — the paper notes it "lays the foundation for assessing the residual value of secondhand vehicles" and that stakeholders need it "to operate the battery within ideal parameters and make informed decisions regarding maintenance, use in second-life applications, or recycling." A passport that carries an OEM-estimated SoH with no common reference invites exactly the cross-manufacturer incomparability that the passport was created to remove: two cars reporting "90 percent" may have degraded by different amounts, through different mechanisms, measured by different algorithms. The regulatory clock makes this a window rather than a research curiosity: the passport date is fixed, the JTC 24 standards are being adopted now, and the paper reports that the Battery Pass consortium — the publicly funded German initiative — has already criticised the lack of regulatory clarity.

What’s been tried and why it hasn’t worked

The regulatory and standards landscape has produced pieces that do not fit together. UN Global Technical Regulation No. 22 (incorporated in Euro 7) is, per the authors, the only instrument that "addresses the necessity and challenges for vehicle-level health metrics," but it verifies durability through the WLTC driving cycle, "through which the SOCE and state of certified range (SOCR) are determined, but not the remaining capacity, which is, however, necessary for the previously mentioned EU regulation"; ISO 12405-4 "vaguely proposes standard (dis)charging cycles for battery packs of EVs relying on the battery 'supplier's recommendations,' which are not publicly available." Technically, model-based SoH estimation (electrochemical or equivalent-circuit models running in the BMS) "inherit[s] model inaccuracies and [is] either unsuitable for online application or imprecise" and is OEM-controlled; measurement-based alternatives each fail a criterion — chassis-dynamometer driving cycles are reproducible "but is cumbersome, time-consuming, and expensive, impairing scalability," on-road profiles are "not reproducible due to fluctuating external influences," continuous discharge through auxiliaries wastes energy, and bidirectional charging "has yet to be widely adopted." A practical obstacle the authors flag: pack voltage, the quantity a reproducible protocol should be defined on, "is less accessible and currently must be reverse-engineered from controller area network (CAN) data so that simple access to the voltage must be granted, e.g., through the onboard diagnosis (OBD)-II interface." So the field has a mandated metric, competing partial procedures, and no public, vehicle-agnostic reference.

What would unlock progress

The authors' proposal is a concrete candidate for the standard: use ordinary onboard charging (which happens anyway and follows standardised protocols such as ISO 15118 / IEC 61851) as the controlled stress, at a constant power low enough to be reproducible (they derive a maximum of about 8 kW for a 120 kWh pack, below a typical 11 kW EU wallbox), over a fixed pack-voltage window "declared in the vehicle's registration documents," after a rest period (a 30-minute rest sufficed in their study), at room temperature, with differential-voltage analysis to attribute the loss to aging mechanisms — measurable charger-side on DC chargers via ISO 15118 data. What would unlock adoption is independent replication across many vehicle makes, an agreed public voltage-window and rest-period specification, regulatory access rights to pack voltage via OBD-II, and a mapping from the resulting capacity- and energy-based SoH to the regulation's SOCE so that JTC 24 or a delegated act can reference it. The adjacent precedent is the emissions world's shift from manufacturer-declared figures to standardised in-use test procedures after the diesel scandal.

Entry points for student teams

A team with access to one or more EVs and a wallbox could prototype the charging-based SoH measurement: log CAN/OBD-II pack voltage and current during controlled constant-power charges over a fixed voltage window, compute capacity- and energy-based SoH with differential-voltage analysis, repeat to quantify reproducibility, and compare against the BMS-reported SoH — publishing the protocol and data openly. A standards-oriented team could draft the specification text (voltage window declaration, rest, temperature, charge power, data fields) in the JTC 24 / Annex XIII data-model format and test it against second-life buyers' and insurers' needs. A data team could analyse existing open EV telemetry to estimate how far OEM-reported SoH diverges across brands for similar mileage and age. Relevant skills: electrical/battery engineering, vehicle CAN diagnostics, standards writing, data analysis.