Document presents candidate amendments for ADS Regulation Phase 2 addressing Safety Management System and in-service monitoring and reporting. Proposes:

• An emergency recovery plan for fleet-level failures,
• Harmonization of SMS audit provisions and certificate mutual recognition,
• Split of operational and design responsibilities between operators and manufacturers,
• Refinement of occurrence classes with objective measurable criteria,
• Delta-V thresholds for all vehicle categories in scope,
• Extended reporting templates,
• Leading safety performance indicators,
• Confidentiality rules for ISMR data sharing,
• Explicit feedback loops from findings to corrective action and approval review.

Agenda for the 132nd session of the Working Party on General Safety Provisions, scheduled 28 September to 2 October. The session will address amendments to UN R107, UN R43, UN R46, UN R67, UN R110, UN R116, UN R161, UN R162, UN R163, UN R151, UN R158, UN R159, UN R166, UN R121, UN R122, UN R125, UN R0, UN R160, UN R169, UN R58, UN R105, UN R73, UN R39, and UN R58. The session will also cover vehicle automation, periodical technical inspections, driver distraction and drowsiness warning systems, and door open warning systems.

Status report on the Informal Working Group on Event Data Recorder and Data Storage System for Automated Driving. The IWG has developed requirements for UN R157, UN R185/GTR, UN R160, and UN R169. DSSAD activities will be removed from the IWG on EDR/DSSAD and assumed by a new IWG on ADS2, pending WP.29 decision. EDR Step 2 work includes proposals for a 03 Series of Amendments to UN R160 and Supplement 3 to UN R169. EDR Step 3 will consider data survivability, triggering sensitivity, data elements, and EDR for category L vehicles, with a proposal planned for April 2027.

Proposal to:

• Amend para. 1.4. to clarify recording obligations when sensors cannot provide data in specified formats,
• Amend para. 2.20. to align terminology with throttle data element,
• Amend para. 2.73. to remove reference to “first” trigger activation,
• Amend para. 5.4.3. to clarify crash test requirements and reference UN Regulations 95 and either 94 or 137,
• Insert new transitional provisions paras. 12.12. through 12.17. establishing acceptance dates for type-approvals with entry into force immediately for 03 series and mandatory acceptance ceasing 1 September 2035 for preceding series,
• Amend Annex 4 Table 1 to add data elements for frontal air bag deployment times (driver and front passenger), pretensioner deployment, occupant size classification, lateral and longitudinal post-crash acceleration, trigger activated, advanced driver distraction

Minutes of the 48th SG-EDR conference call meeting held 15-16 September 2026. The group conducted final review of official documents for UN R160 and UN R169, reaching agreement on multiple technical specifications: DDAW recording duration of 60 minutes (bracketed for GRSG consideration), DCAS duration of 1000 seconds and 10 warnings, LDWS and data element frequencies of 4 Hz and 10 Hz respectively, yaw rate frequency increased to 10 Hz, and airbag warning lamp adjustments. Transitional provisions set for 2030 and 2035. EDR Step 3 workstream planning identified topics including triggering sensitivity, data elements, L-category vehicles, and data survivability. Governance discussion noted DSSAD separation under IWG ADS2, with EDR work continuing under potential new taskforce structure.

Final draft proposal to:

• Amend para. 1.4. to clarify recording obligations for data elements when sensors lack design capability to meet format requirements,
• Amend para. 2.20. to define engine throttle as driver-requested acceleration torque demand,
• Amend para. 2.73. to define trigger activated,
• Amend para. 5.4.3. to clarify crash test requirements and specify data retrievability after impacts of severity levels set by UN Regulations Nos. 94, 95 or 137,
• Insert new paras. 12.12. through 12.17. establishing transitional provisions with dates from entry into force through 1 September 2035,
• Amend Annex 4 Table 1 to add data elements including occupant size classification, vulnerable road user secondary safety system warning indicator status, and advanced driver distraction warning time since warning issued, with modified recording intervals, sample rates, and ranges for existing data elements.

Proposal to introduce a unified Position Code (X-Y) for all seating positions, add seven new mandatory data elements for driver operation and seat-track position to Annex 4 Table 1, and harmonize ADAS status enumerations into one unified set eliminating fragmented and inconsistent wording across system entries.

Document presents Korean research on event data recorder (EDR) data survivability in electric vehicles following post-crash fires. Analysis of nine Korean EV fire cases shows EDR data retrieval failed in three cases due to ACU damage in fully burned vehicles, while six cases achieved data retrieval despite water immersion from firefighting. The document proposes water immersion testing to assess EDR data survivability for inclusion in EDR Step 3, referencing international automotive standards for test method development.

Agenda for the 48th session of the Subgroup on EDR, a hybrid meeting scheduled for 15-16 September 2026 in Berlin. The agenda includes a final review of and potential amendments to official documents to GRSG regarding UN R160 and UN R169, discussion of the need for Step 3 and inventory of potential items for Step 3 workstream with presentations by IWG delegations expected, EDR for ADS, and a list of action items and schedule of future conference call meetings.

Proposal to increase yaw rate recording frequency to 10 Hz. Yaw-rate data recorded at 2 Hz corresponds to one measurement every 500 ms; increasing to 10 Hz provides a measurement every 100 ms, resulting in a fivefold improvement in temporal resolution. This enhancement significantly improves ability to reconstruct pre-crash vehicle dynamics, identify vehicle instability onset and progression, and evaluate Electronic Stability Control functionality. A yaw-rate recording frequency of at least 10 Hz reduces temporal uncertainty from 500 ms to 100 ms and enables more accurate reconstruction of loss-of-control events, vehicle rotation, ESC interventions, and driving dynamical issues.

Proposal to:

• insert new paragraphs 2.71. to 2.78. defining
  • Position Code,
  • Turn Signal Switch Status,
  • Hazard Warning Signal Switch Status,
  • Hazard Warning Signal Lamp Status,
  • Turn Signal Lamp Status,
  • Gear Positions,
  • Seat-Track Position-1, and
  • Seat-Track Position-2,
• Delete existing data elements for Safety belt status by driver position, Frontal air bag deployment by driver and front passenger, Side air bag deployment by driver and front passenger, and Pretensioner deployment by driver and front passenger,
• Amend Table 1 to replace position-specific safety belt, frontal air bag, side air bag, knee airbag, far-side airbag, pretensioner, and occupant size classification entries with X-Y Position Code entries,
• Add Turn Signal Lamp status, Hazard Warning Signal Lamp Status, Turn Signal Switch Status, Hazard Warning Signal Switch Status, Gear Positions, and Seat-Track Position entries with recording intervals from -5.0 to 0 seconds or -1.0 seconds and sample rates of 2 samples per second.

A study of 24 professional truck drivers investigated how drowsiness develops during real-road driving to test a DDAW system. At night, 17 of 24 drivers reached a KSS level of 8, with average time to reach KSS 8 being 56 minutes. Ten drivers chose to stop or were requested to stop in the sleep-deprived condition. A DDAW recording window of 600 s would fail to capture information relevant to the EDR, whereas time windows of up to one hour could provide important indications of increased driver fatigue.

Proposals for EDR Step 3:

• Improved triggering to increase detection rate of vehicle-to-vulnerable road user collisions and low-severity vehicle-to-vehicle collisions by considering delta-v and jerk-based trigger thresholds and developing updated triggering thresholds,
• Overwriting and multi-event issues by reviewing crash reconstruction experiences and developing updated requirements on overlapping events and overwriting,
• Data elements and formats to improve meaningfulness and close gaps by considering new data elements, modifications to data element format specifications, deviating recording duration, and focus on functions compared to specific systems,
• EDR for vehicles of category L₁–L₇ to establish relevant EDR requirements by understanding specificities of vehicles of category L₁–L₇ and developing triggering conditions, events, data elements, and survivability requirements.

For vehicles equipped with an Automated Driving System, it shall be possible to retrieve DSSAD time-stamp data elements specified under paragraph 5.2. of Annex 8 of UN Regulation No. 185 recorded within at least the last 30 seconds prior to triggering of the respective EDR event, regardless of whether the DSSAD and the EDR are implemented as a single system or as separate systems. The time-stamp data elements retrievable from the DSSAD shall include the record for the last activation or deactivation of the feature used at the time of the event within the same power cycle. The data elements shall be retrievable in the format specified in Annex 8 of UN Regulation No. 185, though DSSAD location data may be omitted and dates and times recorded may be replaced with information representing the time elapsed between the recording of each DSSAD data element and the triggering of the EDR.

Proposal to amend footnote 25 to reduce the maximum number of recorded DCAS warnings from 15 to 10 per current ignition cycle. The justification presented cites data burden reduction while retaining information about habitual driver disengagement patterns. Supporting documentation references two autopilot collision cases—Florida 2016 and California 2018—where recorded warnings within existing data retention periods would have revealed repeated disengagement that current recording parameters miss, demonstrating that 10 warnings combined with extended recording duration enables identification of ineffective warning strategies.

European Commission priorities for EDR Step 3 include improving detection of vehicle-to-vulnerable road user collisions and low-severity vehicle-to-vehicle collisions through adjusting delta-v trigger thresholds, developing new jerk-based trigger criteria, and exploring ADS collision detection for EDR triggering in automated vehicles on a long-term timeline. Data elements and format must be ensured to remain up-to-date with new safety system developments and close gaps between light and heavy vehicles through reviewing data elements in light of newly regulated safety systems and mirroring DDAW, ADDW and DCAS data elements from UN R160 to UN R169 on short-term and ongoing timelines.

Proposal based on GRSG/2026/38 to amend para. 2.18. to define end of event time separately for planar events and all other events, amend para. 2.20. to redefine engine throttle percent full, insert new definitions for Advanced emergency braking system (AEBS) and Trigger activated, amend para. 5.4.3. to modify crash test data requirements and reference UN Regulations Nos. 95 and either 94 or 137, and amend Table 1 in Annex 4 to modify data element requirements including changing Air bag warning lamp and other elements from conditional to Mandatory, adding AEBS event recording, modifying Yaw Rate recording intervals and sample rate, clarifying Vulnerable road user secondary safety system warning indicator status recording, modifying Accident emergency call system status recording interval, and revising footnotes regarding TTC calculation and AEBS requirements, effective upon entry into force of the 03 Series of Amendments.

Proposal to amend UN R160:

• Insert new paras. 12.12. to 12.17. establishing transitional provisions for the 03 series of amendments, including that:
  • from the official date of entry into force, no Contracting Party shall refuse type approvals under the amended Regulation,
  • from 1 September 2031, Contracting Parties shall not be obliged to accept type approvals to preceding series first issued after that date,
  • until 1 September 2033, Contracting Parties shall accept type approvals to the 02 series first issued before 1 September 2031,
  • from 1 September 2033, Contracting Parties shall not be obliged to accept type approvals to preceding series,
  • Contracting Parties may grant type approvals according to any preceding series,
  • and Contracting Parties shall continue to grant extensions of existing approvals to any preceding series,
• Amend Annex 4 Table 1 and footnotes 28, 29, and 30 regarding trigger activated recording, powertrain torque demand signals, AEBS target classification, and TTC calculation requirements.

Proposal to amend UN R160 definitions and event recording requirements:

• Amend definitions 2.8., 2.10., and 2.11. regarding Automatically Commanded Steering Function categories A, B1, and C,
• Amend definition 2.18. on “End of event time”,
• Amend definition 2.20. on “Engine throttle, percent full”,
• Delete definition 2.25.,
• Amend definition 2.35. on “Non-volatile memory”,
• Amend definition 2.45. on “Service brake status”,
• Amend definition 2.51. on “Time interval between current event and previous event”,
• Amend definition 2.65. on “VRU secondary safety system”,
• Insert new definitions 2.71. through 2.83.,
• Renumber para. 2.2. as 2.25.,
• Amend para. 5.3.1. to establish trigger conditions for planar, VRU, and rollover events separately from AEBS events with provision for events occurring within 300 ms overlap,
• Amend para. 5.3.1.1. through 5.3.1.5. to specify deployment triggers,
• Add new para. 5.3.1.2. for AEBS emergency braking demand trigger,
• Amend para. 5.3.2. and insert new para. 5.3.2.4. on rollover protection system deployment locking,
• Amend para. 5.3.3. to establish time zero for planar, VRU, and rollover events separately from AEBS events,
• Amend para. 5.4.3. to clarify crash test requirements as UN Regulations 95 and either 94 or 137,
• Amend para. 6.1. on acceleration measurement verification,
• Insert transitional provisions in new paras. 12.12. through 12.16. establishing entry into force of 03 series amendments, non-obligation to accept earlier series approvals as from 1 September 2031, and obligation to continue accepting earlier series approvals until 1 September 2033,
• Amend Annex 4 Table 1 data elements for recording intervals, sample rates, ranges, and new elements including Pre-Event Synchronization Timer, Trigger activated, ACPE, ADDW, DDAW, blind spot monitoring, turn signal, DCAS, emergency stop signal, and TTC.

Document SG-EDR-47-02 and SG-EDR-47-03 were discussed. The group agreed to include definitions 2.8, 2.10, 2.11, and 2.20 with edits. Proposed definitions will be circulated by June 15 for review with comments due June 26. For data elements, disagreements on frequency rates between contracting parties and industry remain in square brackets. EDR for ADS will be addressed under GRVA and the ADS IWG. An October in-person meeting in Berlin is planned, pending WP.29 direction on integrating DSSAD into the ADS IWG.