student profile: Mr Amir Valadkhani


Map

Thesis work

Thesis title: Dynamic Control and Multimodal Modelling for Searching Behavior and Pick up Patterns of Point to Point Transport in Large Urban Cities

Supervisors: Mohsen RAMEZANI

Thesis abstract:

The popularity of point-to-point transport and their searching process for finding an unserved passenger have a significant effect on the network congestion. Also, the network congestion influences the rate of meeting unserved passengers and point-to-point vacant vehicles. Recent researches have revealed the effect of taxis, ridesourcing, or ridesharing on network congestion independently. Although, the two-sided effect of point-to-point transport in a multimodal framework (i.e. considering taxis, ridesourcing, and ridesharing simultaneously) on network congestion is still an open problem. This proposal will develop theory and practice to model multimodal searching behaviors and pick up patterns of different modes of point-to-point transport while considering the effect of them on network congestion and vice versa. In addition, how to dispatch vacant point-to-point transport vehicles and how to match passengers (.e.g. shortest path, minimum travel time) such that network traffic delay of a city becomes better will be added to the literature to fill the gap.
To investigate the effect of congestion on point-to-point transport vehicles and vice versa, network speed is a crucial state. The network speed is not a given state of the system and must be estimated through noisy sensor measurements. Estimation approaches are grouped into two categories, namely, model-driven and data-driven. The model-driven approach uses a physical model describing traffic dynamics. The data-driven approach extensively relies on historical and/or streaming data instead of physical models. Data-driven solutions are more general and can be developed easily for different scenarios and networks, but lack of data is the main concern in utilizing them. Fusion of probe (e.g. GPS) and fixed data (e.g. SCATS) is a method in this research to overcome the inherent shortcomings of data. Probe data comes from probe vehicles. In the probe vehicle concept, vehicles themselves are acting as roving traffic detectors, which are not bound to specific and fixed locations along the road infrastructure. Probe data includes data generated by vehicles about their current position, motion, and time stamp. The challenge would be how to combine various sources of data with different sample rates and how to use them for estimation of aggregate network speed.

Selected publications

Download citations: PDF RTF Endnote

Conferences

  • Valadkhani, A., Hong, Y., Ramezani, M. (2017). Integration of loop and probe data for traffic state estimation on freeway and signalized arterial links. IEEE 20th International Conference on Intelligent Transportation Systems (IEEE ITSC 2017), Piscataway, NJ: Institute of Electrical and Electronics Engineers (IEEE). [More Information]

2017

  • Valadkhani, A., Hong, Y., Ramezani, M. (2017). Integration of loop and probe data for traffic state estimation on freeway and signalized arterial links. IEEE 20th International Conference on Intelligent Transportation Systems (IEEE ITSC 2017), Piscataway, NJ: Institute of Electrical and Electronics Engineers (IEEE). [More Information]

Note: This profile is for a student at the University of Sydney. Views presented here are not necessarily those of the University.