Application platform: Humanoid robot battery pack qualification Last verified: June 9, 2026 Pack-level governing standard: None.
No governing standard exists for humanoid robot battery pack qualification. The landscape does not yet exist. Every entry in the matrix below is borrowed from an adjacent domain or defined from first principles. The sourcing tier is labeled on every entry so you know whether you are working from a published standard, a published analysis, or editorial inference.
Borrowed baselines cited in this card:
| Standard | Edition | Scope (as written) | Borrowing status for humanoid robotics |
|---|---|---|---|
| IEC 62619 | 2022 (Ed. 2.0) | Secondary lithium cells/batteries for industrial applications, including stationary | Borrowed baseline; structurally mismatched on six parameter classes (documented below) |
| ISO/FDIS 13482 | FDIS registered 2025-07-24; still in approval phase as of June 2026 (see note below) | Safety requirements for service robots, personal and professional/commercial | Robot-system-level hazard analysis only; no battery pack thresholds |
| ISO 10218 | 2025 | Industrial robot safety (integrates former ISO/TS 15066) | Defers battery thermal management to manufacturer guidelines; no pack-level test protocol |
| UN 38.3 | Rev. 8 (2024) | Transport of lithium batteries | Applies directly — the only Tier 1 standard on this card |
| IEC 60068-2-64 / -2-27 | Current editions | Environmental testing: vibration (random) / shock | Test methodology applicable; reference profiles not representative of bipedal locomotion |
The FDIS was registered July 24, 2025. As of June 9, 2026, the ISO catalogue still designates this document as an FDIS in the approval phase, nearly eleven months after registration. Cite as "ISO/FDIS 13482 (approval phase — verify publication status before use)" and do not treat it as a finalized requirement document. Neither the 2014 edition nor the FDIS addresses battery pack qualification at any parameter-class level.
The classification dependency
The reason no pack-level standard exists is upstream: no regulatory body has settled what a humanoid robot is.
Classification determines which existing frameworks get borrowed as baselines. If a humanoid robot is industrial equipment, ISO 10218:2025 provides the system safety framework and IEC 62619:2022 is the natural battery baseline. If it is a personal care or service robot, ISO/FDIS 13482 applies at the system level, but 13482 sets no pack-level thresholds. If it is a consumer product, the regulatory apparatus shifts entirely and the battery baseline moves toward IEC 62133-2, which is a worse fit than 62619 for the power and energy requirements involved. Three classification paths, three different borrowed baselines, three different assumptions about what failure modes matter.
The IEEE Humanoid Study Group identified this dependency as foundational in its September 2025 framework document, "A Pathway Study For Future Humanoid Standards." The classification section was, by the study group's own account, the most contested portion of the work. The framework is a strategic roadmap, not a standard. No IEEE Project Authorization Requests for humanoid battery systems have been filed as of this writing. The study group chair estimated 18–36 months to first ratified standards, placing the earliest output at Q1 2027.
Until classification is resolved, every DVP for a humanoid robot battery pack is constructing its own qualification framework from borrowed parts. The matrix below makes the borrowing explicit and documents the structural mismatches.
Requirements matrix
Energy density and capacity
What the DVP needs: Gravimetric energy density target at pack level (Wh/kg), derived from platform mass budget and mission endurance. Humanoid platforms are weight-constrained for balance and locomotion efficiency. Supplier marketing materials cite 250–350 Wh/kg at cell level (Tier 3: supplier marketing materials from stdbattery.com and others; independent verification required). Pack-level density is materially lower after structural, thermal management, and BMS mass integration. Reported runtimes in current programs land around 2–4 hours per charge (Tier 3: supplier marketing materials; manlybattery.com).
What governs it: No standard sets energy density or capacity requirements for humanoid robot battery packs.
What's missing: Targets are entirely application-derived from platform mass allocation and endurance specifications. DVP defines its own threshold.
Power capability
What the DVP needs: Continuous and peak discharge C-rates by operating mode. Published analyses describe humanoid duty cycles as peak-heavy: 3C–5C continuous for locomotion and manipulation, with peaks exceeding 10C for high-dynamic events such as rapid acceleration, stair climbing, and load lifting (Tier 3: supplier marketing materials and market analyses from stdbattery.com, manlybattery.com, and similar sources; not independently verified). Some supplier marketing materials cite 50C peak for sub-10-second bursts; this figure derives from battery manufacturer product pages, not independent test data, and should not be used as a DVP reference without platform-level validation.
What governs it: No standard addresses this. IEC 62619:2022 tests at industrial discharge rates (typically 1–2C) that do not represent humanoid duty cycles.
What's missing: Mission-phase-specific C-rate definitions. DVP must specify continuous and peak discharge rates by operating mode (walking, running, manipulation, stair climbing, idle), derived from power traces on the target platform. No Tier 1 or Tier 2 source provides these.
Cycle life and calendar life
What the DVP needs: Cycle life at representative DOD, temperature, and C-rate conditions. Calendar life under expected storage conditions.
What governs it: IEC 62619:2022 includes cycle life test methodology (per publicly accessible summary: §7 series, environmental and endurance tests), but its reference discharge profiles use moderate C-rates under near-steady-state conditions. NMC cells at humanoid-representative discharge rates show 800–1,200 cycles per supplier marketing materials (Tier 3: supplier marketing materials; independent verification required), significantly below the same chemistry tested at IEC 62619's industrial baseline rates. The gap between those two numbers is the mismatch. (Tier 2: IEC 62619:2022 methodology borrowed. Tier 3: cycle life estimates at humanoid C-rates from supplier sources.)
What's missing: A cycle life test protocol at humanoid-representative C-rate profiles. DVP should borrow IEC 62619 test methodology but substitute a mission-representative pulse-discharge profile including peak events. Document the mismatch explicitly: the borrowed methodology and the substituted profile are not the same test, and the cycle life number produced is not comparable to IEC 62619 reference data.
Operating envelope
What the DVP needs: Temperature, humidity, and altitude ranges for the target deployment environment. Current humanoid deployments span warehouse, factory, and outdoor environments with ambient temperature ranges typically from −10°C to +45°C.
What governs it: IEC 62619:2022 specifies thermal cycling from −20°C to +60°C (per publicly accessible summary: §7 series, environmental tests; cell-level test). This range is broadly applicable but was not validated against humanoid thermal management architectures, which must dissipate heat from high-C discharge in a mass-constrained, orientation-varying enclosure. (Tier 2: IEC 62619:2022 thermal cycling range as borrowed baseline.)
What's missing: Two things. First, altitude and humidity requirements specific to humanoid deployment environments.
Second, and more consequential: operating orientation. IEC 62619 assumes fixed or limited-angle installation. A bipedal robot operates at continuously varying pitch and roll during gait, crouching, and stair climbing. Electrolyte distribution in liquid-electrolyte cells, thermal contact resistance at cell-to-busbar interfaces, and BMS sensor accuracy can all be orientation-dependent. The effects accumulate over thousands of hours of operation at orientations the cell was never tested in. No standard addresses this. (Tier 3: editorial inference; no governing standard.)
Mechanical: vibration, shock, crush
What the DVP needs: Vibration profile representative of bipedal locomotion. Shock profile representative of fall events. Crush resistance.
What governs it: IEC 62619:2022 references IEC 60068-2-64 (random vibration) and IEC 60068-2-27 (shock) (per publicly accessible summary of IEC 62619:2022, §7 series mechanical tests; pack-level). The vibration PSD profiles in these references derive from industrial equipment and ground vehicle transport environments. Crush test (cell-level): 13 kN, no fire or leakage at 30% deformation. Drop test (pack-level): 1.2m in six fixed orientations, >90% voltage retention. (Tier 2: IEC 62619:2022 mechanical test clauses as borrowed baseline.)
What's missing, vibration: Bipedal gait generates cyclical impact loading at stride frequency (~1–4 Hz fundamental, higher harmonics during running) with high-amplitude vertical footstrike impulses. This combines low-frequency cyclical stress accumulation over millions of stride cycles with high-frequency impulse content not characterized in the IEC 60068-2-64 industrial PSD. A pack that passes IEC 62619's random vibration test has not been tested against a representative gait vibration spectrum. DVP must define a gait-derived PSD from accelerometer measurements on the target platform, borrowing IEC 60068-2-64 test methodology but substituting the reference curve. (Tier 3: editorial inference.)
What's missing, shock/fall: A humanoid robot fall involves dynamic multi-body collapse from standing height (center of mass ~0.8–1.0m) at variable orientation, with rotational energy contributing to impact severity. The kinetic energy and loading mode exceed what a 1.2m vertical drop in fixed orientations captures. DVP must define fall-event acceptance criteria using IEC 60068-2-27 shock methodology (half-sine pulse, g-level, duration) with pulse parameters derived from fall dynamics modeling on the target platform. (Tier 3: editorial inference.)
Safety: thermal runaway propagation, abuse tolerance
What the DVP needs: Thermal runaway propagation test (pack-level: cell-to-cell within pack). Overcharge, external short circuit, internal short circuit, forced discharge (cell-level tests). Nail penetration or equivalent internal short circuit simulation (cell-level).
What governs it: IEC 62619:2022 (per publicly accessible summary, §7.2–7.3 series) includes: thermal runaway propagation testing (pack-level), overcharge at 1.5× rated voltage for 7 hours with no fire or explosion (cell-level), external and internal short circuit (cell-level), and nail penetration (cell-level). These are applicable as a baseline. (Tier 2: IEC 62619:2022 safety test clauses as borrowed baseline.)
What's missing: Two gaps.
First: post-fall sequential abuse tolerance. No standard addresses whether a pack that has experienced a bipedal fall event retains its safety margins against thermal runaway. The sequence matters. A pack that passes cell-level abuse testing in pristine condition may not pass after mechanical damage from a fall. DVP should consider a sequential test protocol: fall event (pack-level) followed by abuse test (cell-level), with acceptance criteria defined by the engineering team. No governing standard requires this. (Tier 3: editorial inference.)
Second: BMS response time under high-peak-C conditions. IEC 62619's BMS functional safety framework (per publicly accessible summary, §7 series functional safety assessment) specifies fault detection and current interruption, with response intervals on the order of ~200ms appropriate for industrial equipment. At sustained discharge rates exceeding 10C, thermal and voltage dynamics evolve faster than at the 1–2C rates the standard was designed around. The higher the peak C-rate, the less protection a 200ms window provides. No standard addresses BMS response timing for high-peak-C applications. (Tier 3: editorial inference.)
Environmental protection
What the DVP needs: IP rating for the battery enclosure. Salt fog, dust, and water ingress resistance as required by deployment environment.
What governs it: IEC 60529 (IP rating) applies to the enclosure. Available test methodologies for specific environmental exposures include IEC 60068-2-52 (salt fog), IEC 60068-2-68 (dust), and IEC 60068-2-18 (water), though none specify humanoid-specific requirements. (Tier 2: IEC 60529 and IEC 60068-2 series as borrowed baselines for test methodology.)
What's missing: IP rating selection criteria for humanoid deployment environments. The hot-swap interface introduces an additional ingress pathway not present in sealed industrial packs. DVP must define IP requirements based on deployment environment and hot-swap connector design, selecting from the available IEC 60068-2 test methods as appropriate.
EMC
What the DVP needs: Electromagnetic compatibility testing for the battery pack and BMS.
What governs it: IEC 62619:2022 (Ed. 2.0) added EMC requirements (per publicly accessible summary). Broadly applicable. (Tier 2: IEC 62619:2022 EMC clauses as borrowed baseline.)
What's missing: No humanoid-specific EMC profile exists. The proximity of the battery pack to motor controllers, actuators, and sensor arrays in a humanoid chassis produces EMI coupling paths not represented in IEC 62619's industrial equipment assumptions. DVP should validate EMC under representative installed conditions, not on the bench alone.
Transport
What the DVP needs: Compliance with transport regulations for lithium batteries.
What governs it: UN 38.3 Rev. 8 (2024). (Tier 1.) Applies directly, without structural mismatch.
What's missing: Nothing for transport.
Hot-swap
This parameter class has no home in any existing standard. It cuts across safety, mechanical, and electrical domains.
What the DVP needs: Test protocol for battery pack removal and insertion under partial load or live bus conditions, covering inrush current on connection, voltage mismatch between fresh and partially depleted packs, connector arcing, and transient thermal events.
What governs it: Nothing. IEC 62619:2022 does not contemplate hot-swap. ISO 10218:2025 defers hot-swap thermal management to manufacturer guidelines.
What's missing: The entire test protocol. DVP defines hot-swap acceptance criteria from first principles. Adjacent reference points exist in aerospace battery disconnection procedures, but these were not written for rapid field-swap with a live robot platform. (Tier 3: no governing standard; protocol must be defined by the engineering team.)
Gaps summary
Nine of ten parameter classes on this card have no Tier 1 governing standard. Transport is the exception.
| Parameter class | Borrowed baseline (test level) | Written for | Structural mismatch | Standards body activity |
|---|---|---|---|---|
| Energy density / capacity | None | — | — | None identified |
| Power capability | None | — | — | None identified |
| Cycle life | IEC 62619:2022, methodology (cell + pack) | Industrial stationary | Test C-rates (1–2C) vs. humanoid duty cycles (3–5C continuous, >10C peak) | None identified; no IEC TC21 work item for mobile robot batteries |
| Operating envelope | IEC 62619:2022, thermal range (cell-level) | Industrial stationary | Fixed-orientation assumption invalid for bipedal locomotion | None identified |
| Vibration | IEC 62619 → IEC 60068-2-64 (pack-level) | Industrial equipment / ground transport | PSD profiles do not capture gait-induced cyclical loading or footstrike impulses | ISO 25785-1 (WD stage; industrial bipedal robot safety); publication not expected before 2027; no battery clauses anticipated |
| Shock / fall | IEC 62619 → IEC 60068-2-27 (pack-level) | Handling drops | 1.2m fixed-orientation drop does not represent dynamic multi-body fall | Same as above |
| Safety (abuse) | IEC 62619:2022 (cell-level abuse; pack-level propagation) | Industrial stationary | No sequential test (fall → abuse); no BMS response time spec for high-peak-C | None identified |
| Hot-swap | None | — | No standard contemplates this use case | None identified |
| Environmental protection | IEC 60529; IEC 60068-2 series (enclosure-level) | General industrial | No humanoid-specific IP; hot-swap connector ingress unaddressed | None identified |
| EMC | IEC 62619:2022 Ed. 2.0 (pack-level) | Industrial equipment | Installed EMI coupling in humanoid chassis not characterized | None identified |
Standards development timeline
IEEE Humanoid Study Group: Published framework document September 2025 ("A Pathway Study For Future Humanoid Standards"). No PARs filed. Earliest ratified standards Q1 2027–Q3 2028. No battery-specific working group identified within the effort. (Tier 2.)
ISO 25785-1 (safety requirements for dynamically stable industrial mobile robots): Working Draft stage, comment period closed June 2025. Led by representatives from Boston Dynamics and Agility Robotics under ISO TC299. Publication not expected before 2027. Industrial scope only. No battery pack clauses anticipated. (Tier 2.)
ISO/FDIS 13482 (service robot safety): FDIS registered July 2025; still in approval phase as of June 9, 2026. Does not address battery pack qualification in any edition. (Tier 1 for robot safety scope; no battery content.)
IEC TC21 (secondary cells and batteries, responsible for IEC 62619): No publicly identified work item addressing mobile robot battery applications. (Tier 2: absence confirmed via publicly accessible committee information.)
HEIS 2026 is described as the most comprehensive humanoid-specific standard framework globally (robottoday.com, March 2026). Full text not available in publicly accessible English sources. If it contains pack-level battery provisions, they may represent the only jurisdiction-specific requirements anywhere. Becomes directly relevant if a program targets Chinese market deployment or if provisions are adopted by reference elsewhere. Not citable as a governing standard for Western-market programs as of this date. (Tier 3.)
DVP construction posture
Every DVP entry for a humanoid robot battery pack, except UN 38.3 transport compliance, must document the borrowed baseline, the originating application domain, and the specific mismatch between the borrowed standard's assumptions and the humanoid operating profile. The entry format:
"IEC 62619:2022 as baseline, with the following known gaps relative to actual operating conditions: [list]"
is more defensible than "IEC 62619:2022 as baseline" alone, because the latter implies coverage that does not exist. The mismatch documentation is the mechanism by which the engineering team's judgment becomes traceable when the standards landscape eventually shifts. Treat it accordingly.
That shift will not arrive as a single clean standard. The IEEE framework's classification debate, the parallel ISO working drafts with different scopes (25785-1 industrial, 13482 service/personal), and the absence of any IEC TC21 activity on mobile robots all point toward the same trajectory: borrowed frameworks will accrete, layer, and partially converge without producing a unified document. The DVP entries you write today will need revision as that accretion proceeds. Document the borrowing provenance now so the revision is traceable later.
All standards citations verified June 9, 2026. IEC 62619:2022 clause-level references (§7 series) derive from publicly accessible certification guides and summaries; the full standard is paywalled. Pull the full IEC 62619:2022 text to confirm specific clause numbers before populating DVP entries.
- ISO/FDIS 13482 publication status: The FDIS was registered in July 2025 and the ISO catalogue still showed approval phase eleven months later, so check the ISO catalogue page before citing it in any DVP entry.
- China's HEIS 2026 battery provisions: The most comprehensive humanoid-specific standard framework globally may contain pack-level battery requirements, but the full text is not yet available in English per robottoday.com's March 2026 analysis.
- IEEE PAR filings for humanoid standards: No Project Authorization Requests have been filed since the September 2025 framework publication, and the study group chair's 18–36 month timeline to ratified standards means the earliest output lands Q1 2027.
- IEC TC21 mobile robot activity: No work item for mobile robot battery applications has been publicly identified under the committee responsible for IEC 62619, a gap worth monitoring via IEC TC21 committee pages as humanoid deployment volumes increase.