B.1 Approach

An important first step in determining the cost and benefit of various VDR implementations and applications was to profile the capabilities and costs of current commercially available VDRs and supporting technologies. This work provided key inputs and basic building blocks for the cost-benefit analysis in Chapter 4.

The primary objective of this process was the identification-specific VDR features and capabilities, and the estimation of the costs associated with those features to develop an understanding of cost drivers for VDR design. There is a considerable amount of information available regarding recommended data elements, sampling rates, and measurement accuracy needed to support various levels of crash causation and safety analyses (e.g., NHTSA's Summary of Findings of the EDR Working Group, Volume 2, Supplemental Findings for Trucks, Motorcoaches, and Buses). However, there is virtually no information available related to the costs associated with incorporating various data-gathering capabilities into a particular VDR design.

For example, it will be important to have high-resolution recording of numerous vehicle operating parameters just seconds before and after the event (e.g., What was the deceleration rate when the accident occurred?) to support crash reconstruction. However, a long-term histogram record of data will be needed (e.g., What percent of brake applications last week or month were above some preset deceleration rate?) to support various operational efficiency improvement initiatives (such as driver training or maintenance prognostic tools). The data storage requirements, resolution, and accuracy are very different for these two applications. In this example, the goal of this task would be to understand the cost implications for incorporating each or both of these types of capabilities into a VDR design. Key VDR cost drivers examined included:

• Number of sensors (or inputs) monitored
• Type of inputs monitored (e.g., analog, digital, on/off)
• Data collection rates (sampling frequency)
• Required accuracy (this is more likely a function of the sensor or transducer input than of the VDR, but is nevertheless a cost driver for the VDR system)
• Memory capacity and/or memory features
• "Virtual" VDR systems (distributed processing and storage) versus centralized (or conventional) VDR designs
• Level of "hardening" (i.e., vulnerability to damage from a crash)
• Data extraction methods (hand-held diagnostic tool versus various types of wireless communication protocols)
• Various supporting VDR capabilities (e.g., GPS, forward-looking radar)

In addition, it was important to understand the impacts of incorporating various combinations of capabilities and features since there are design "economies" associated with grouping selected features together.

To accomplish these objectives, the team interviewed technical staff from several stakeholder groups:

• Vehicle OEMs (Class 3 through 8)
• Trailer OEMs
• Commercial vehicle component suppliers focusing on electronic, electrical, communications, and safety specific components and systems
• Manufacturers of VDRs

The team also contacted manufacturers and suppliers of so-called "data loggers" that traditionally are used for vehicle development and research purposes (e.g., Dearborn Group, National Instruments, Link, IOtech Inc.). These manufacturers had valuable insight regarding cost and capabilities of VDR type devices. The team also interviewed selected government officials to gain a better knowledge of the experiences associated with implementing various forms of VDRs.

The intent in this task was to focus primarily on commercially available technologies, since accurate cost data for systems or components still in the research and development stages was difficult to obtain and even more difficult to extrapolate to the realm or "market ready" devices. However, to the extent possible, these developmental technologies were researched and industry stakeholders were asked to estimate the relative cost impacts of emerging technologies and capabilities. Where specific cost data was not available, costs assumptions were based on best professional judgment.

The team developed a detailed "features and performance" matrix that was used to profile each commercially available and prototype system. This profile includes:

• The parameters or data elements stored
• Sensor inputs and vehicle network communications
• Data collection frequency, accuracy, memory capacity, and storage algorithms (e.g., event triggers, histogram data)
• Data extraction methodology
• The level of "hardening" (temperature and environment ratings)
• Intended application and cost

It should be noted that while a large number of systems were profiled in this task, and every effort was made to sample as many systems as possible, there are likely other systems currently available or that have recently been introduced that were likely not sampled. As the market for VDRs is continuing to change and develop, new and redesigned systems are continuing to be introduced into the marketplace. This profiling task was intended to be a snapshot of what was available at the time.

B.2 System Descriptions

Several different types of systems that incorporate varying VDR functionality were researched, including:

• EDRs
• Video event data recorders
• Vehicle data loggers
• Trip activity report systems built into engine ECUs (e.g., Cummins, DDC, Caterpillar, Mack)
• ABS-based recording systems

Each of these types of systems are described in the following subsections.

B.2.1 EDRs

Several systems are available that are intended to record vehicle data triggered on an event or accident. These systems are what are typically referred to as EDRs. Generally, EDRs are programmed with a set of conditions or triggers that when reached cause the recording and storage of high-resolution vehicle data. Exhibit B.1 summarizes the EDRs that were profiled during this task. Appendix A includes more detailed references and sources for information about these systems.

Exhibit B.1 - Event Data Recorders

AP+ Series Data Recorders - Accident Prevention Plus

The AP+ Series data recorders are a series of programmable data recording systems for both event data logging and logging of operational data. They are custom programmable and have provisions for an optional driver smart card. The optional smart card can store driver identification information along with summary and histogram data on driving behavior (e.g., maximum speed, hours of service, vehicle and engine speed histograms, braking intensity and occurrence histograms). This system can use the SAE J1939 network as the source for much of this information, or it can use discrete analog or digital inputs.

UDS Accident Data Recorder - Mannesmann VDO

The UDS accident data logger is designed to record lateral and longitudinal acceleration, vehicle speed, vehicle direction, and 10 status inputs (e.g., ignition on/off, left and right turn signal status, brake light status) for 30 seconds before an incident and 15 seconds after. The system has the ability to record up to 12 events (9 triggered events and 3 manual events). The system is marketed through VDO for truck, bus, and taxi fleet markets. The USD recorder has a number of analog and digital discrete inputs to record these various parameters.

Loss Management Services (LMS) - MACBox

The Mobile Accident Camera Box (MACBox) developed by LMS records both video and accelerometer data before, during, and after an accident event. The unit can then either be returned to the manufacturer for downloading and analysis, or it can be wirelessly transmitted to the manufacturer via a cellular transceiver. LMS charges a fee for an accident report along with an annual service fee. The MACBox has a digital discrete input.

Incident Data Recorder - Bowmonk, Ltd.

The Incident Data Recorder is an event triggered data recorder that records data 45 seconds prior to and 20 seconds after an incident. The system incorporates a bi-directional accelerometer for measuring longitudinal and lateral acceleration (an optional second accelerometer is available to record vertical acceleration). In addition, the system stores histogram data on average speed, distance, and time on one-minute intervals. There are eight additional inputs available for recording both analog or digital signals. The Incident Data Recorder is capable of recording these data elements over both analog and digital discrete inputs.

Accident Data Recorder (ADR2) - Delphi Automotive

The Delphi ARD2 is a second-generation accident EDR designed for motorsport applications. It has an internal three-axis accelerometer and can record data from external sources with seven analog and three timer inputs. It also features four communications links (three serial, one CAN). It is designed to withstand the high g-force (500 g) impacts of race cars and has an uninterruptible power supply to record data in the event of data loss. It is currently being used in all CART racing series vehicles. The ADR2 has CAN inputs for ISO 11898 and SAE J1939, along with analog and digital discrete inputs.

The Witness - Independent Witness

The Witness is a incident data recorder that monitors the acceleration profile of a vehicle during an accident. The system measures and records the acceleration of the vehicle in three directions along with the date, time, and direction of the accident using all internal sensors. The system uses a portable reader to download the accident data. In 2002, the National Association for Stock Car Auto Racing (NASCAR) began installing The Witness in all of its race vehicles. There are no connections to vehicle databuses with this system.

B.2.2 Video Event Data Recorders

Several systems are currently available in the marketplace that record video data during an event or accident. As a subset of EDRs focused on video recording, they have been termed Video Event Data Recorders. These systems often record multiple views (e.g., frontal, rear, side) of video along with a minimal set of data parameters triggered on various events or accidents. Exhibit B.2 summarizes the systems that were profiled in this task.

Exhibit B.2 - Video Event Data Recorders

Road Recorder 6000 - Safety Vision

The Safety Vision Road Recorder 6000 is a digital video recording system designed for mobile applications. The system records up to five video sources (cameras or other image capturing devices like infrared), and has a built-in monitor output. The video is stored in a removable hard drive and is stamped with the date and time. The system also includes a GPS interface that allows it to record longitude, latitude, speed, and heading. It also has multiple discrete analog sensor inputs that are configurable and a J1807/J1587 interface. The system is focused primarily on the school bus market.

DriveCam - DriveCam Video System

DriveCam is a palm-sized video recorder that mounts behind the vehicle's rearview mirror. The DriveCam records both video and audio 10 seconds before, during, and after a trigger. Additionally, the DriveCam includes an accelerometer to record g-forces. DriveCam is designed to record erratic driving behaviors (e.g., hard braking and acceleration, and hard cornering). The DriveCam also records crash events, and can be manually triggered.

Vehicle Accident Video Recorder (Prototype) - Personal Eyewitness

The Vehicle Accident Video Recorder continuously records video into Flash memory, overwriting old video every 30 seconds until an accident triggers the system to permanently store the video (using airbag deployment or crash sensors). The system includes a video camera, built-in accelerometer, and backup battery. The video is stored for 30 seconds-20 seconds before and 10 seconds after an incident occurs. The system can then retain the video for up to one week on the internal battery. The data/video can be extracted using a serial or infrared connection. The system is designed initially for the truck, bus, and taxi fleet markets.

B.2.3 Vehicle Data Loggers

Three systems were profiled that were designed to monitor and record vehicle information for fleet management, maintenance, and operations monitoring purposes. This type of system, referred to as Vehicle Data Loggers, is intended to continuously monitor and record, in a histogram-type format, various vehicle parameters while in operation. Vehicle data loggers typically are not focused on recording event or accident data based on triggers, but occasionally do have such capabilities. Exhibit B.3 summarizes the vehicle data loggers profiled.

Exhibit B.3 - Vehicle Data Loggers

DataLogger - ArvinMeritor

The ArvinMeritor Data Logger, referred to as a Data Logging Unit (DLU), is an onboard data recorder designed specifically for heavy-duty commercial vehicles. The DLU uses the vehicle's onboard J1708/J1587 network to continuously monitor and record information from the engine, instruments, ABS, collision warning system, and any other ECU that is connected to the serial databus. The DLU can be programmed to be triggered by up to 14 events, including air bag deployment, hard deceleration rate, J1587 Diagnostic Message (#194), out-of-range engine parameters, and out-of-range battery voltage. The driver can also manually activate a trigger to record data. The system is mainly marketed for maintenance monitoring and fault reporting applications, and is available as a factory-installed option on select Freightliner heavy-duty vehicles.

Mobius TTS - Cadec

The Mobius TTS is an onboard logistics management system. The system provides fleet asset tracking, paperless driver logs to support DOT regulations (e.g., hours of service, CDL information, manifest data), driver performance coaching (e.g., fleet speed standards, idling guidelines, handling practices), maintenance monitoring, and fault code warnings. The system consists of a touch-screen display panel, onboard computer, GPS receiver, J1708/J1939 network connections, 18 digital or analog I/O channels, and multi-mode wireless communication (e.g., 802.11, cellular, GPRS, iDEN, 1xRTT).

Tacholink Millennium - Bowmonk, Ltd.

The Tacholink Millennium is an onboard computer system that monitors and records various vehicle information for maintenance and fleet efficiency purposes. The system performs several functions including recording a histogram data of the vehicle speed, excessive RPM events, heavy braking events, fast acceleration events, trip summaries, fuel tank level, driver identification, GPS position, idling time, engine hours, and status of other inputs. This system is designed primarily for heavy-duty commercial vehicles, buses and coaches, emergency vehicles, car fleets, and construction equipment.

B.2.4 Trip Activity Report Systems

Several of the major heavy-duty vehicle engine manufacturers offer add-ons to their engine ECU systems that monitor, summarize, and record vehicle information for trip activity reporting purposes and fault code warnings. These systems are typically built into the engine ECU, but can have additional components (e.g., ECUs, displays, transceivers). A great deal of information on the vehicle activities can be reported by these systems purely by data available within the engine ECU. Exhibit B.4 summarizes the trip activity report systems profiled.

Exhibit B.4 - Trip Activity Report Systems

Road Relay 4 - Cummins

The Cummins RoadRelay 4 is a dashboard-mounted display unit that connects with the SAE J1587 or J1939 databus to display engine parameters, trip data, or other subsystem data. The unit reads stored engine data such as vehicle speed, engine speed boost pressure, oil temperature, coolant temperature, battery voltage, and fuel consumption. It also monitors distance traveled, travel time, and estimated time of arrival, and can be connected to a GPS receiver to monitor heading and location. The Road Relay can also read information on the vehicle databuses from other subsystems (e.g., transmission gear, gear selected, transmission fluid temperature).

Pro Driver - Detroit Diesel

Detroit Diesel's Pro Driver system records data inside the engine ECU. Pro Driver manages three types of data-activity-based data, time-based data, and event-based data. Activity-based data includes trip activity reports, life-to-date activity, and engine service intervals. Time-based data uses a built-in clock and calendar to record monthly activity summaries, daily engine usage, and engine service intervals. Event-based data records hard-braking incidents (e.g., total count and detailed information for the last two) with an adjustable event threshold, detailed last stop records, and engine fault events. Data can be extracted and displayed using a variety of methods (e.g., dashboard displays, vehicle communications networks, or PC software).

Vital Information Management System (VIMS) - Caterpillar

The Caterpillar Messenger display unit provides real-time, visual feedback on engine or truck operating conditions. Messenger uses a graphics information LCD display. Multiple communication data links (J1939, J1708) can be utilized to provide real-time performance and operating information while the vehicle is in use. The information can be displayed in either multilingual text (even graphical languages) or a graphical format.

VMac II Engine ECU/Co-Pilot/InfoMax Wireless - Mack Trucks

The Mack Truck Co-Pilot is an optional dash-mounted driver display system that can display engine and vehicle information to the driver. It can display trip summaries, fuel economy, vehicle system status, and engine sensor data. In addition, the VMac engine ECUs can record histogram and driver event data (e.g., vehicle over speed, engine over speed, hard braking). This data can then be downloaded via the J1708/J1587 connector or sent wirelessly over short-range WiFi service using Mack's InfoMax Wireless module to a terminal at a fleet's distribution center. The InfoMax Wireless system includes an onboard module, WiFi antenna, and optional GPS receiver. The terminal or depot requires a network access point (NAP) and a PC running dedicated InfoMax software from Mack.

B.2.5 ABS ECUs

In addition to the trip activity report systems available from engine suppliers, several of the manufacturers offer ABS ECUs that have some limited data recording capabilities, both from ABS data (e.g., speed sensors, brake light, fault codes) and external data via analog and digital inputs. Exhibit B.5 summarizes the ABS ECUs profiled.

Exhibit B.5 - ABS ECUs

Tractor ABS ECU - Bendix

The Bendix tractor ABS ECU can monitor ABS components (e.g., wheel speeds, ABS brake control status, ABS retarder control status, ATC (Automatic Traction Control) brake control status, ATC retarder control status, road speed, ABS faults). The Bendix tractor ABS ECU connects to the J1587 and J1939 networks to provide diagnostic information and control the engine and retarder during ATC events, respectively. Additionally, the Bendix Tractor ABS ECUs can communicate with the trailer over the PLC, SAE J2497, signal for the purposes of illuminating an in-cab trailer ABS indicator lamp. There are currently no provisions for recording other data.

Tractor ABS ECU - Meritor WABCO (MW)

The MW tractor ABS ECU can monitor ABS components (e.g., wheel speeds, ABS brake control status, ABS retarder control status, ATC brake control status, ATC retarder control status, road speed, ABS faults). The MW tractor ABS ECU connects to the J1587 and J1939 networks to provide diagnostic information and control the engine and retarder during ATC events, respectively. There are currently no provisions for recording other data.

Trailer ABS ECU - Haldex

The Haldex trailer ABS ECU has the ability to monitor up to five channels (two analog inputs, one digital input, two digital I/O). These channels can be configured to monitor a number of systems or sensors and report this information on Haldex's Info Centre or over PLC 4 Trucks, and SAE J2497. These systems or sensors could include Air-Weigh trailer axle load sensing systems, PSI Tire Inflation Systems, reservoir pressure sensors, trailer door latches, brake lining wear sensors, backup distance warning sensors, or trailer rollover warning systems.

Trailer ABS ECU - Wabash National

The Wabash National trailer ABS ECU can monitor ABS components (e.g., wheel speeds, wheel end temperature, trailer ID, brake light status, fault codes, ABS activation, trailer odometer) and has two additional analog or digital inputs that could be used to monitor other sensors or systems, such as yaw sensors, control pressure, or door latch sensors. The system can then report this information over PLC 4 Trucks, SAE J2497.

B.3 Data Parameter Profile

The systems in Exhibits B.1 through B.5 have the capability to record multiple vehicle parameters (e.g., speed, acceleration, ignition status) at multiple rates (e.g., 1 Hz, 10 Hz, 100 Hz) and during various operational states (e.g., at impact, before and after impact, throughout a specific trip, continuously). Exhibit B.6 lists the systems along with the parameters they are capable of recording. It should be noted that most of these systems are re-programmable and could be reconfigured to record more, less, or different parameters by the manufacturer (or in some cases the fleets) based upon customer requirements.

Exhibit B.6 - Commercial VDR Data Parameter Profiles

Exhibit B.6 shows that systems intended for similar applications (e.g., EDRs, trip activity report systems, data loggers) record similar sets of parameters. EDRs, for example, all record longitudinal and lateral accelerations, and most record vertical acceleration also. Data loggers and trip summary recorders record several engine and transmission parameters, odometer, VIN, and system fault codes.

Exhibit B.6 demonstrates, however, that there is considerable overlap of functionality among the devices, and, in some respect, the product categories are notional (or directional) rather than absolute clear-cut divisions.

B.4 Sensor Inputs and Vehicle Network Communication Profile

To record each of these parameters, each system requires some combination of analog or digital inputs and/or vehicle communication network connections.

Analog inputs are used to record a signal that has a continuous nature rather than a pulsed or discrete nature. An analog signal may vary in frequency, phase, or amplitude in response to changes in physical phenomena, such as sound, light, heat, position, or pressure. For example, engine speed, pedal position, temperature, or control pressure would likely use an analog signal.

Digital inputs are used to record a signal in which discrete steps are used to represent information. Digital inputs generally have steps that typically refer to voltages on the signal. Often digital inputs can be thought of as "on" or "off". For example, left turn signal status, headlight status, or wiper status would likely use a digital signal.

Pulse Width Modulation (PWM) is a way of digitally encoding analog signal levels on a digital signal. Through the use of high-resolution counters, the "on" or "off" cycle of a digital signal is modulated, creating a square wave, to encode a specific analog signal level. By varying the length of time the signal is "on" versus "off", a cycle (called the duty cycle) can be used to represent analog values. The PWM signal is still digital because, at any given instant of time, the voltage signal is either fully on or fully off. Given a sufficient bandwidth, any analog value can be encoded with PWM.

Many of the systems also have the ability to record information over a pre-existing vehicle communications network. On heavy-duty commercial vehicles, there are a number of communications networks that, if installed, could provide a centralized source for data. These protocols include:

• ISO 11898 European controller area network communications standard
• SAE J1708/J1587 North American serial communications standard
• SAE J1939 North American controller area network communications standard
• SAE J2497 North American power-line carrier communications standard

Exhibit B.7 shows the inputs and connections each system is capable of using.

Exhibit B.7 - Commercial VDR Sensor Inputs and Vehicle Network Connections

Exhibit B.7 shows that many of the systems use analog and digital inputs as well as connections to the vehicle networks. The vehicle communications networks provide a centralized source of data to these systems, minimizing the need for discrete wiring and numerous extra analog or digital inputs (which require analog-to-digital converters and sampling hardware that increases the cost and complexity of the systems). A number of the systems use general purpose discrete analog and digital inputs to record data directly from the source as opposed to reading it off of a vehicle databus. This is largely advantageous because these systems could be fitted or retrofitted onto many different vehicles that may or may not have a databus.

B.5 Data Storage Medium Profile

The systems listed in Exhibits B.1 though B.5 were designed to perform several different functions, from event-triggered recording to trip summary recording, which require different data storage and triggering algorithms. In general, data is collected in two ways in these systems:

• Summarized data on a per trip interval (e.g., maximums, minimums, averages, histograms)
• Event-triggered continuous data (i.e., data channels are recorded at a specific frequency for a given interval, usually before and after an event trigger)

Exhibit B.8 describes how these systems store data and their available memory capacity.

Exhibit B.8 - Commercial VDR Data Collection and Storage Methodology

N/A - Data not available or proprietary

It is interesting to note that for VDRs that record event-triggered data, no clear standard is being followed in terms of how many seconds before and after the trigger data is stored. Also, these systems vary in memory capacity from a few hundred KB to almost one hundred MB, and even many GB for the video recorder systems.

B.6 Data Extraction Method Profile

These systems use a variety of methods for data extraction and downloading for analysis. There are five main approaches for data extraction:

• The manufacturer provides a proprietary reader that attaches directly to the system
• The system or a portion of the system must be returned to the manufacturer for download and/or analysis
• A removable storage device (e.g., CompactFlash card, hard disk)
• Wired downloading via various onboard vehicle networks or communications standards
• Wireless downloading of the data to a land-based hub

Wireless downloading of data can also be classified into two general categories:

• Local short-range communication (e.g., infrared, WiFi 802.11, Bluetooth)
• Long-range communication (e.g. , cellular, satellite, QualComm)

Exhibit B.9 shows each system's methods for data extraction.

Exhibit B.9 - Commercial VDR Data Extraction Methods

Exhibit B.9 shows that a number of methods are currently being used by manufacturers of VDR systems to extract data. In general, most of the EDR and VDR systems have removable media options and some form of wired downloading capability. There are a number of systems that offer infrared wireless downloading, and a couple use WiFi.

B.7 Application and Cost Profile

Exhibit B.10 shows various design aspects of each system along with their cost and intended application, and includes:

• The level of "hardening" or resistance to both the environment and vehicle impacts
• Additional or optional capabilities of the system
• Cost of the system
• Additional costs required (e.g., infrastructure costs, proprietary reader costs, software costs)
• Intended application of each system

Exhibit B.10 - Commercial VDR Intended Application and Cost

N/A - Data not available or proprietary

In Exhibit B.10, most systems are designed to be either mounted in the cab of the vehicle or designed to meet the "hardening" environmental standards of the SAE J1455 standard for heavy-duty vehicles.

The costs in Exhibit B.10 also vary, with many of the systems having additional or optional components that can significantly add to the cost.

Foot Notes

MAF Mass Air Flow, AFM Air Flow Meter, MAP Manifold Absolute Pressure

Unless otherwise noted, costs are for individual quantities, aftermarket, not installed. Cost data is intended to provide an approximate cost for each system, and was provided by each vendor through interviews and/or published pricing. It should be noted that, while this data is intended to provide a general cost feel for each system/category of VDR, the actual price of each system could vary depending upon numerous factors, including: options/accessories selected, quantity discounts, customization, or installation and sensor hardware costs.

SAE J1455 "Joint SAE/TMC Recommended Environmental Practices for Electronic Equipment Design (Heavy-duty trucks)" - This SAE Recommended Practice includes test methods used to simulate climatic, dynamic, electrical environmental conditions from natural and vehicle-induced sources that influence the performance and reliability of vehicle and tractor/trailer electronic components. This guideline is intended to aid the designer of automotive electronic systems and components by providing material that may be used to develop environmental design goals.