Site mapV2X communication can prevent tram crash
V2X communication can prevent tram crashes
On 13. 12. 2019 our company, in cooperation with Czech Technical University in Prague and Public Transport Company of Ostrava (DPO), conducted a test of a tram collision-warning system (forward collistion warning). The collision warning system is based on a new V2X communication technology. Using such communication, the vehicles exchange their movement state – their position, velocity, acceleration etc. The collision warning system was implemented in our OBU (on-board unit) UCU 5.0V, which is already mounted in every vehicle of the fleet of DPO.
The test was done in depot in Poruba, Ostrava, using tram Vario LF2, operated by DPO. The assessment, whether there is an imminent collision risk, was completely implemented in communication units UCU-5.0V.
- Fig. 1: Test track in Poruba depot. The red curve is a GPS-measured trajectory, the blue curve is the estimated and projected track, as a result of sensor fusion. The system localized the estimated position on a digital map of the rail network.
The goal of the experiment was to assess the performance of the algorithm, which should warn the driver, that there is a risk of crashing a stationary tram ahead. After the warning, the driver shall immediatelly brake. The systems warns the driver by an accoustic signal, triggeered when a possible collision is detected by the algorithm, running in UCU. The situation ofone experimental run is shown in Fig. 2.
- Fig. 2: Analysis of one of the experimental runs. The key points are: 1 = point where the warnigng was triggered by the application, 2 = point where the driver started to brake (after some reaction time, expected to be 1 s), 3 = final vehicle position – the position where it stopped, 4 = position of the simulated vehicle.
V2X communication is a key element in the driver warning system, as it is a mean to share data about vehicles, mainly their movement state – position and speed. Both vehicles communicate and share their estimated state. To improve the accuracy, a sensor fusion at every vehicle is done. Each vehicle combines in an intelligent form data from GNSS (positioning system), wheel speed from a tachograph unit and vehicle acceleration – measured by an internal accelerometer in UCU 5.0V. Moreover, the estimated vehicle position is projected on a digital map, therefore all vehicles are located on a track. This algorithm thus fuses all advanced functions, available in our OBU unit.
- Fig. 3: Speed plot using different sources – before and after the sensor fusion. Blue corve is the fused speed (combination of data from GNSS, wheel speed and acceleration). It is apparent that thanks to the acceleration sensor, the fused speed at time 80s is closer to the true speed, as there is no delay caused by GNSS sensor or tachograph.
The results of the experiments were very promissing. After the acoustic warning, the driver started to brake and the final position was within 5 meters from the simulated vehicle (see Fig. 4). No safety distance was used, so the the goal was to make the driver stop close to the simulated vehicle, yet still in front of it. The simulated vehicle had its position artificialy placed on the rail.
The tests were done with speeds 10 km/h, 30 km/h, 40 km/h, 50 km/h a 60 km/h. With all these speeds, the vehicle stopped safely.
- Fig. 4: The result of experiment with steady-state speed of 30 km/h. The tram really stopped right at the simulated position of the vehicle ahead – marked by warning cones.
This entry was posted in Press releases. Bookmark the permalink.
