A CAN test system for accident reconstruction & road testing
A test system for use in operational road tests and accident reconstructions created by automotive forensic analysis experts at Advanced Analysis Associates, Inc.and Automotive Systems Analysis, Inc., is detailed in a paper entitled “A Multi-Purpose Vehicle Test Instrument with MCC DASYLab Graphics and Kvaser CAN Interfaces” that can be found here.
Explains co-author, Bill Rosenbluth: “Our goal was to create a cost-effective, real-time test data recorder that, in operational road tests and accident reconstructions, recorded individual controller performance, communications between controllers and external factors, while presenting an easily readable, scaled graphical feedback interface to the test operator.”
The instrument system detailed in the paper uses a Kvaser Leaf CAN to USB interface, a DASYLab™ graphics display and USB A/D modules from Measurement Computing Corp., plus an American Engine Management GPS module, 8-channel K-thermocouple module and A/D module. Two custom Python scripts were created to combine the test interfaces. The resulting data acquisition and recording system logs both raw CAN and scaled parameter data, and has a logged CAN playback feature, which also accepts CAN logs from other popular formats and translates them identically to the system configuration in an original instrument.
Rosenbluth explains why the research team chose Kvaser’s CAN to USB and PCI CAN interfaces: “All Kvaser interfaces handled 11-bit and 29-bit CAN modes, including J1939 messages. Additionally, Kvaser offered a publicly available and straightforward *.dbc database editor and free basic CAN data monitor(CanKing), which allowed quick and straightforward testing of incremental *.dbc databases. The Kvaser *.dbc database editor is an easy to use and intuitive tool, allowing straightforward merging of our many incremental and separately developed *.dbc databases.”
As this paper describes work accomplished as part of a continuing development project, the authors are keen to hear feedback on how to improve their system. To contact them directly, please email Bill Rosenbluth.