You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In parallel/multi-scale simulation, we often need to partition the road network into several regions which are simulated in different simulation instances. To ensure simulation consistency across adjacent partitions, a plausible border control mechanism is to duplicate the vehicles on the boundary: the vehicles are running on both partitions of the two sides of the border, and the vehicles on the downstream partition is set as the primary vehicle while the replica ones in the upstream partition is shadow/ghosting vehicle. To ensure the transmission of information from downstream to upstream, the primary vehicle is controlled by the internal car-following and lane-changing models (simulationStep()), while the shadow vehicle is the replica of the primary vehicle, which means it acts as the 'shadow', adjusting its speed and position according to the primary vehicle, and not controlled by the simulationStep() command.
For example, you can refer to the paper 'Chen, Hao, et al. "QarSUMO: a parallel, congestion-optimized traffic simulator." Proceedings of the 28th International Conference on Advances in Geographic Information Systems. 2020'. In '3.3 Inter-Partition Synchronization', they introduce this method. However, they did not make the corresponding code public....
Can you give some insight into how can I achieve this function of the shadow vehicle, specifically making its movement completely a replica of another vehicle, and in not only microscopic but also SUMO's mesoscopic model?
Thank you!
The text was updated successfully, but these errors were encountered:
Hi SUMO community,
In parallel/multi-scale simulation, we often need to partition the road network into several regions which are simulated in different simulation instances. To ensure simulation consistency across adjacent partitions, a plausible border control mechanism is to duplicate the vehicles on the boundary: the vehicles are running on both partitions of the two sides of the border, and the vehicles on the downstream partition is set as the primary vehicle while the replica ones in the upstream partition is shadow/ghosting vehicle. To ensure the transmission of information from downstream to upstream, the primary vehicle is controlled by the internal car-following and lane-changing models (
simulationStep()
), while the shadow vehicle is the replica of the primary vehicle, which means it acts as the 'shadow', adjusting its speed and position according to the primary vehicle, and not controlled by thesimulationStep()
command.For example, you can refer to the paper 'Chen, Hao, et al. "QarSUMO: a parallel, congestion-optimized traffic simulator." Proceedings of the 28th International Conference on Advances in Geographic Information Systems. 2020'. In '3.3 Inter-Partition Synchronization', they introduce this method. However, they did not make the corresponding code public....
Can you give some insight into how can I achieve this function of the shadow vehicle, specifically making its movement completely a replica of another vehicle, and in not only microscopic but also SUMO's mesoscopic model?
Thank you!
The text was updated successfully, but these errors were encountered: