How It Works

The University of Michigan Transportation Research Institute (UMTRI) is conducting the Safety Pilot Model Deployment in Ann Arbor, MI, using innovative technology equipment in everyday vehicles in a real-world environment. The equipment uses Dedicated Short Range Communications (DSRC) – a technology similar to Wi-Fi – which is fast, secure, and reliable. This technology will allow vehicles to “talk” to each other and to roadway infrastructure to improve safety, mobility and the environment.

UMTRI is collecting data for one year from the test vehicles and roadside equipment. This data will be evaluated to support the 2013 NHTSA agency decision on vehicle communications for safety.

Click on the image above to see a FULL map of the Ann Arbor site plan for the safety pilot model deployment.

The model deployment involves installing devices in approximately 2,800 cars, trucks and buses; deploying devices along 73 lane-miles of roadway; equipping facilities to process the resulting data used to evaluate the safety benefits, and educating drivers about the use of the equipment.

The Roads

The model deployment includes providing equipment to send and receive data along 73 lane-miles of roadway in the northeast Ann Arbor area. The primary routes covered include U.S. 23, M-14, Plymouth Road, Washtenaw Avenue, and the Fuller/Geddes corridor. These routes were selected to capture the majority of test participant drivers during their daily commutes. The mix of freeways and city streets will allow researchers to evaluate the effectiveness of the technology on different road types and driving conditions.

Dedicated short-range communications (DSRC) equipment is installed along these routes at:

  • 21 signalized intersections,
  • 3 curve locations, and
  • 5 freeway sites.

All sites include a secure communication link for collection of the study data. A typical intersection installation is shown above.

The Vehicles

Each Safety Pilot vehicle is outfitted with one of four type of devices – integrated safety devices (ISD), aftermarket safety devices (ASD), retrofit safety devices (RSD) and basic communications devices, called vehicle awareness devices or VADs. Each of these devices broadcasts a basic safety message (BSM) ten times per second. The BSM alerts other nearby, similarly equipped vehicles to the presence of the broadcasting vehicle through vehicle-to-vehicle communications. This occurs even when a driver cannot see another vehicle, for example, at an obstructed intersection or in the vehicle’s “blind spot”.

Safety Pilot includes a rich mix of vehicle and device types to support evaluation needs. The breakdown is shown below:

Total number of vehicles 2,836
Fully integrated cars 64
Fully integrated trucks 3
Cars with VAD 2,200
Trucks with VAD (local vehicle fleets) 50
Trucks with VAD (university fleet) 100
Transit Vehicles (buses) with VAD 100
Cars with ASD 300
Trucks with Retrofit Safety Devices 16
Transit vehicles with Retrofit Safety Devices 3

The four types of devices installed in Safety Pilot vehicles are described below:

Vehicle Awareness Devices (VAD)

VADs are the most simple of the four types of devices used on Safety Pilot vehicles. The majority of model deployment vehicles are equipped with VADs. These devices focus only on emitting the basic safety message. They do not receive messages from other vehicles. Instead, the large number of VAD-equipped vehicles provides the penetration rate necessary to generate vast amounts of data for evaluation of the technology.

VAD 02 small

Aftermarket Safety Devices (ASD)

This is an aftermarket electronic device, installed in a vehicle, and capable of sending and receiving the safety messages over a DSRC wireless communications link. The device has a driver interface, runs V2V and V2I safety applications, and issues audible or visual warnings and/or alerts to the driver of the vehicle. This option is being examined as a means of increasing user adoption, and hence benefits, especially in the existing fleet of over 250 million vehicles.

ASD 02 small

The following graphic depicts a typical ASD installation for the Safety Pilot Model Deployment. As shown, many of the ASD-equipped vehicles will have additional data collection capabilities to monitor driver behavior.


Installed by the manufacturer, these devices integrate directly with the vehicle’s computers, thus providing the ability to draw on a wide range of data. In addition to emitting and receiving the basic safety message, vehicles with integrated devices can further communicate data on speed, acceleration and deceleration, yaw rate, turning, wiper activity, and braking, among others. Cooperative, crash avoidance safety applications rely on this enhanced data stream to provide highly accurate, real-time alerts, advisories, and warnings to the driver. In the case of buses, trucks, and motor coaches, these devices are integrated after production or are retrofitted.


Retrofit Safety Devices are being installed on a limited number of trucks and buses for the model deployment. Unlike ASDs, this type of device is connected to the vehicle’s databus and can provide highly accurate information from vehicle sensors. An RSD includes a driver interface, both broad¬casts and receives BSMs, and can process the content of received messages to provide warnings and/or alerts to the driver.

The Applications

Various safety applications are being evaluated as part of the Safety Pilot Model Deployment. Most of the applications focus on safety, but others will also generate mobility and environmental benefits. Additionally, many of the applications employ vehicle-to-vehicle (V2V) communications, with a few that require vehicle-to-infrastructure (V2I) communications.

The following applications are providing input to drivers during the model deployment:

Safety ApplicationTypeDescription
Forward Collision Warning (FCW) V2V A V2V application where alerts are presented to the driver in order to help avoid or mitigate the severity of crashes into the rear end of other vehicles on the road. Forward crash warning responds to a direct and imminent threat ahead of the host vehicle.
Emergency Electronic Brake Light (EEBL) V2V A V2V application where the driver is alerted to hard braking in the traffic stream ahead. This provides the driver with additional time to look for, and assess, situations developing ahead.
Intersection Movement Assist (IMA) V2V A V2V application where alerts are given to drivers as they begin to accelerate from rest into, or across, another road, to help the driver avoid crashes with crossing traffic.
Blind Spot Warning (BSW)/ Lane Change Warning (LCW) V2V A V2V application where alerts are displayed to the driver that indicate the presence of same-direction traffic in an adjacent lane (Blind Spot Warning), or alerts given to drivers during host vehicle lane changes (Lane Change Warning) to help the driver avoid crashes associated with potentially unsafe lane changes.
Do Not Pass Warning V2V A V2V application where alerts are given to drivers to help avoid a head-on crash resulting from passing maneuvers.
Left Turn Across Path / Opposite Direction (LTAP) V2V A V2V application that alerts the driver of a transit vehicle if another vehicle intends to make a right turn in front of it while the transit vehicle is stopped at a bus stop near an intersection.
Right Turn in Front V2V A V2V application where alerts are given to the driver as they attempt an unprotected left turn across traffic, to help them avoid crashes with opposite direction traffic.
Signal Phase and Timing (SPaT) V2I A set of V2I applications where intersection traffic signals broadcast the current state of signal phasing (red, yellow, or green) and time remaining in that phase. The SPaT data would be used by the vehicle to achieve safety, mobility and environmental benefits.
Curve Speed Warning (CSW) V2I A V2I application where alerts are provided to the driver who is approaching a curve at a speed that may be too high for comfortable or safe travel through that curve.
Railroad Crossing Warning V2I A V2I application that alerts the driver of approaching trains at railroad crossings without warning signals or gates.
Pedestrian Detection V2I A V2I application that alerts the driver of turning transit vehicles if a pedestrian has pushed the crosswalk button at an upcoming intersection, or a remote sensor system detects a pedestrian in the crosswalk at the intersection.

The Drivers

UMTRI recruited and prepared volunteers for participating in the model deployment. The genders and ages for driver recruits reflect those of the general driver population to the greatest extent possible to reflect real-world conditions. Drivers have been recruited primarily from areas east and northeast of downtown Ann Arbor. The model deployment requires high concentrations of drivers along the selected routes.

UMTRI is seeking a one-year commitment from drivers to participate in the program. Most private vehicles will be equipped with VADs or ASDs. Some participants may qualify to drive a new vehicle, equipped with integrated technology, for 6 months. Drivers are being asked to visit UMTRI periodically during the one-year test, for equipment checks and data retrieval. Other than that, participants follow their daily driving routines.

The Data

Data collected during the model deployment will be the basis of an independent evaluation of potential safety benefits and driver acceptance of connected vehicle technology. Data is being collected from the connected vehicle devices and from the drivers themselves.

Data Collected from Connected Vehicle Devices

Data is being collected from onboard equipment (OBE) and roadside equipment (RSE) during the model deployment. The following table shows the types of data collected from these devices. This data will be used to evaluate potential safety benefits.

Type of Data CollectedSelected ExamplesVADASDRSE
Host vehicle state Speed and motion variables, driver pedal and steering wheel controls, Anti-Lock Braking System (ABS) activation, wiper status, lamp status, cruise control status, temperature No Some
GPS/Mapping Latitude, longitude, time, heading, data quality indicators Yes Yes
Exposure data Trip distance, vehicle sizes, lane change counts, system/application availability Some Some
OBE data Basic safety message log, driver display state, remote device types No Yes
Alerts & remote vehicle data Alert events, remote vehicle data: position, speeds, motions, pedal states No Yes
Forward vehicle/object Position and motion of vehicles ahead No Some
Video Coverage of up to six views No Some
V2I data Intersection type, signal timing status, curve information No Yes

All vehicles in the Model Deployment will log the BSM that the vehicle broadcasts. The RSEs log BSMs received from passing vehicles, and ASDs and Integrated Vehicles are capable of logging BSMs received from other devices.

The UMTRI Safety Pilot Operations Center collects and displays some of the data. The display includes maps of vehicle locations and data collected from the vehicles.

Driver Acceptance Data

The model deployment also is collecting driver acceptance data. As opposed to the controlled demonstrations of the previous driver clinics, the model deployment is evaluating driver acceptance by exposing everyday drivers to the safety applications under natural driving and traffic conditions over multiple months.

Each participant receiving an ASD or an Integrated Vehicle will be asked to complete a questionnaire at the end of their participation. This questionnaire will ask participants about their experience with the safety applications they experienced. Among other questions, participants will specifically be asked about acceptance of the warning applications, whether they perceived there to be a safety benefit, and their willingness to purchase safety systems based on connected vehicle technology. Some participants who received the ASD’s or the Integrated Vehicles will be asked to take part in focus groups where they will have the opportunity to discuss their experiences with among other participants.