IVI New Developments
Demonstration in Washington DC for IDS Technologies
A demonstration was held at the Turner Fairbank Highway Research Center (FHWA) in Washington DC in June 2003.
Intelligent intersection was shown as a promising, near-term deployable IDS system to aid drivers at signals in identifying when it is unsafe to make a permitted left turn in the face of an oncoming vehicle. Using multiple, detection and sensing devices (including Lidar, radar, loop detectors and in-vehicle GPS), the system can identify and track vehicles approaching the intersection in real time. Combined with the vehicle motion data, the central processing unit (CPU) uses signal timing and phasing data sent from the traffic controller to run a decision-making algorithm. When conditions are unsafe for making a permitted left turn, a dynamic no left turn sign that pulses (or looms) display a warning to the driver. In order to illustrate an alternate path to deployment, IEEE 802.11a wireless communication was also incorporated to allow direct communication between the CPU and approaching vehicles, thus creating a smart intersection that can provide information directly to in-vehicle devices.
The following technologies were demonstrated:
-
A viable, near-term infrastructure supported intersection decision support system for the LTAP/OD crash type
-
A robust system architecture that gathers data from the infrastructure and fuses the data to form a digital State Map
-
Transmission of the State Map via 802.11a wireless communications to a vehicle, showing how the infrastructure information can be provided to future in-vehicle warning systems.
The demonstration, while important, was essentially an off ramp to the California IDS research program, and as such, used emerging equipment, warning algorithms and driving behavior data. In the longer term, and what has been subsequently accomplished, emerging results of continuing investigation of driver behavior that leads to left turn across path / opposite direction (LTAP/OD) crashes.
To the maximum extent possible, off-the-shelf and potentially applicable enabling technologies were used. These included multiple, detection and sensing devices (including lidar, radar, inductive loop detectors and in-vehicle GPS), LTAP/OD system identified and tracked vehicles approaching the intersection in real time. Combined with the vehicle motion data, the central processing unit (CPU) uses signal timing and phasing data sent from the traffic controller to run a decision-making algorithm.
There are essentially three components to the IDS demonstration system: warning sensors, Controller/processor (including algorithm) and driver interface.
Warning Sensors and Controller
The LTAP/OD IDS system used a combination of three warning sensors. Two pairs of remote sensors (Denso lidars and Eaton Vorad radars) and seven embedded loops are used to detect the traffic downstream from the intersection and the presence of a subject vehicle (SV, or the left-turning vehicle).
Extracting useful information from the loops meant that we chose to use National Transportation Communications for ITS Protocol (NTCIP) for the 2070-advanced traffic controller. From our review, we were the first group to implement NTCIP within a 2070 advanced traffic controller. Besides using loops, radars, and lidars, differential global positioning satellite (DGPS) was also used within the vehicles, SV and POV, to relay the approximate position of the vehicles to the intersection. With the combination of the three sensors and DGPS, a rich and robust amount of information can be extracted from each approaching vehicle including range, range rate, and trajectory. The DGPS data was relayed back to the PC 104 computer by 802.11a wireless communications, while the sensor data was relayed to the PC104 computer via hardwire of Ethernet cable. Once the PC 104 computer received the data, the information was processed into a database, which implements a warning algorithm and fusion of all four sensors. This warning algorithm and senor fusion processed the data to estimate the Principal Other Vehicles (POVs, or the threat vehicle) time to intersection and any possible conflicts with an approaching SV. If a possible conflict might occur, the algorithm implemented a warning signal to be sent from the computer to the DII on the roadway infrastructure via hardwired CAT V Ethernet cables. Once the signal arrives, the DII actuates and turns on, warning the driver to not make a left turn and thus preventing a collision.
In addition to the infrastructure based solution, a vehicle-based solution was also examined with what we characterize as a State Map information broadcast from the infrastructure to the vehicle. The purpose behind this research was to show the richness and robustness of real-time data that can be presented to the vehicles using the 802.11a wireless communications, which is the basis for the emerging generation of dedicated short-range communications (DSRC). The State Map Visualizer within the car does not suggest an in-vehicle warning system; rather, we illustrated how one could notionally work, where data can be sent from the infrastructure to the vehicle in real time.
The main point behind the warning sensor portion of LTAP/OD demonstration is that system can implement the fusion of the four sensors for a more rich view of the traffic or it can function in either of two modes:
-
Loop Only Mode This mode uses only the existing infrastructure of loops within the pavement and 2070 NTCIP interactions.
-
Remote Sensor Mode This mode uses only the remote sensors (lidars, radars, and DGPS) to detect the traffic conditions.
Both of these modes are cheaper and more cost effective in field implementation and a field operations test down the future. All hardware components for our demonstration plan were commercial off the shelf equipment except for the infrastructure-based interface.
Warning algorithm
Inputs to the warning algorithm are SV sensor measurements, POV sensor measurements. The output of the algorithm is a warning signal that triggers the DII. For SV and POV sensor measurements, sensor data processing is necessary to refine the measurements by means of filtering and tracking, to form a system level description of vehicle states by means of multiple sensor data fusion, and to derive vehicle motion parameters such as range and speed by means of statistical parameter estimation. Vehicle motion parameters are further processed in a warning algorithm, based on a decision-making criterion to decide whether or not to turn on the DII. The criterion is that if the POVs time-to-intersection (T2I) falls in a critical time gap, which is a safe time gap required by the SV to make a left turn, trigger a warning, otherwise, no warning should be given. T2I is defined as the time that SV or POV needs to get into the intersection (more precisely T2I is the time to reach the stop line of SV lane.)
Driver Infrastructure Interface
The dynamic LED sign we used is based on the MUTCD R3-2 left-turn prohibition regulatory sign. It is the display part of an active infrastructure-based LTAP/OD intersection collision warning, where state information of the intersection, collected at the roadside or in the future even with other cars, is transformed to a warning signal an activated, looming LED sign and informs the subject vehicle of an impending opposite direction threat. The sign was placed just above eye level at opposite corner of the intersection. The sign is self-luminous when active (using LEDS) and thus to be neutral and icon-free when not active. The circle/slash under our design will, periodically (at 1-4 Hz), increase in scale from the standard size shown in geometrical specifications for the R3-2 to a 50% increase in the thickness of the elements. This latter activity will make the sign especially visible amongst the distracters that can be found at any intersection for the reason that the motion inherent in its elements and the looming nature of that motion should be especially suited to signaling the faster and more sensitive pathways in the visual nervous system.
Special thanks goes to Mr. Jim Misener of California PATH Program for providing this report.
Helpful Links: