Standardization News Search

Magazines & Newsletters / ASTM Standardization News


November/December 2009

Automobile Black Box Data

Determining What Happened

A new book published by ASTM International details how to obtain and use automotive “black box” data for today’s engineers and investigators who seek information about a passenger vehicle impact-event.

William Rosenbluth, Automotive Systems Analysis Inc. in Reston, Va., a fellow of the American Academy of Forensic Sciences, shares his expertise in Monograph 5, Black Box Data from Accident Vehicles: Methods of Retrieval, Translation and Interpretation. Rosenbluth, an ASTM International member, focuses on electronic data analysis and data mining from his extensive background in the field and a career that includes automotive on-board computer technology, electronic diagnosis and restraints component testing.

His earlier 2001 work on the subject, Investigation and Interpretation of Black Box Data in Automobiles (Monograph 4), guides users in how to obtain and interpret vehicle black box data and parameters that the investigator can integrate into a vehicle impact-event condition overview. That overview assists investigators in learning more about vehicle conditions before and during an accident.

In the years since the publication of Monograph 4, automobiles have become more sophisticated, as have the techniques of analyzing information from the various automobile electronic data recorders. In current passenger vehicles, EDR data can typically be found in electronic air bag, engine, braking and stability controllers. Many of these controllers retain data even when the car battery is disconnected because that data is saved in non-volatile memory.

The overall purpose of EDR data continues to be the same: to find out what happened at and before an impact event — for purposes of safety system improvement and accident analysis. However, current versions of this data can include improved versions of impact-event magnitude (Delta-V), vehicle speed, restraint usage, restraints timing, braking, acceleration and stability data. That data is often retrieved and available only in raw hexadecimal form. Often, common devices only translate a subset of that data while the raw hexadecimal data define the complete information available.

The new book, Monograph 5, focuses on methods for translating and interpreting raw hexadecimal data retrieved from various EDRs. This is illustrated by using case examples that provide practical illustrations for many different analysis processes. “After the retrieval, there’s a lot of thought that goes into the post processing methods, the translation of that data into engineering units,” says Rosenbluth.

The work covers both inductive and deductive analysis methods and shows how to provide answers in common engineering units. The examples show how to parse, format, translate and interpret data, including the creation of data charts that illustrate the various derived data tables.

In the course of those examples, Rosenbluth discusses vehicle data retrieval, including two familiar forensic engineering methods (J1962 serial download and direct EDR/electronic control unit umbilical downloads) and introduces a third method, direct-EEPROM-read umbilical. This third method, introduced in Monograph 5, requires engineering skill and care, however, it allows data retrieval where more common methods are not available. For each of these retrieval processes, he notes how they can be accomplished in a “forensically neutral” manner, compliant with ASTM E2493, Guide for the Collection of Non-Volatile Memory Data in Evidentiary Vehicle Electronic Control Units, so that retrieving the data does not change the evidentiary information and future investigators can re-retrieve the same evidentiary information.

Rosenbluth summarizes Monograph 5: “It’s a collection of examples showing its readers various methods to derive and create engineering data in common engineering units from data sources that are not so intuitively obvious, and it shows how to validate such methods with independent test monitors.”