Skip to main content
Thesis defences

PhD Oral Exam - Mohammad Karimi, Civil Engineering

Cooperative Traffic Control Framework for Mixed Vehicular Flows

Wednesday, January 22, 2020
10 a.m. – 1 p.m.

This event is free


School of Graduate Studies


Engineering, Computer Science and Visual Arts Integrated Complex
1515 St. Catherine W.
Room EV 11.119



When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.


A prompt revolution is foreseen in the transportation sector, when the current conventional human-driven vehicles will be replaced by fully connected and automated vehicles. As a result, there will be a transition period where both types will coexist until the later type is fully adopted in the traffic networks. This new mix of traffic flow on the existing transportation network will require developing a new ecosystem able to accommodate both types of vehicles in traffic network environments of the future. A major challenging issue related to the emerging mixed transportation ecosystem is the lack of an adequate model and control framework. This is especially important for modeling traffic safety and operations at network bottlenecks such as highway merging areas. Therefore, the main goal of this thesis is to develop a microscopic modeling and hierarchical cooperative control framework specifically for mixed traffic at highway on-ramps. In this thesis, a two-level hierarchical traffic control framework is proposed for mixed traffic at highway merging areas. In this regard, for the lower level of the proposed traffic control framework, this thesis establishes a set of fundamental trajectory-based cooperative control algorithms for different merging scenarios under mixed traffic conditions. We identify six scenarios, consisting of triplets of vehicles, defined based on the different combinations of CAVs and conventional vehicles. For each triplet, different consecutive movement phases along with corresponding desired distance and velocity set-points are defined. Via the movement phases, the CAVs engaged in each triplet cooperate to calculate their optimal-smooth trajectories—using Model-based Predictive Control (MPC) scheme—aiming at facilitating the merging maneuver (i.e., to avoid stopping at the end of the acceleration lane) while complying with the realistic constraints related to the safety and comfort of vehicle occupants. The vehicles in each triplet are modeled by a distinct system, and a Model Predictive Control scheme is employed to calculate the cooperative optimal control inputs (acceleration/deceleration values) for CAVs, accounting for human-driven vehicles’ uncertainties such as drivers’ reaction time and free-flow desired speed tracing error.

In the next step of the thesis, for the higher level of the proposed traffic control framework, a merging sequence determination and triplets’ formation methodology is developed based on predicting the arrival of vehicles into the merging area and according to the priority in choosing different triplet types. To be able to model the merging maneuvers when two consecutive triplets share a vehicle, the interactions between triplets of vehicles are also investigated. In order to develop a microscopic traffic simulator, we analytically formulate different vehicles’ driving behaviors under cooperative (i.e., the proposed traffic control framework) and non-cooperative (i.e., normal) operation modes and discuss the switching conditions between these driving modes.

To evaluate the effectiveness of the proposed traffic control framework, first, each triplet is simulated in MATLAB and evaluated for different sets of system initial values. Without a need for readjusting the algorithm for different initial values, the simulation results show that the proposed cooperative merging algorithms ensure smooth merging maneuvers while satisfying all the prescribed constraints, e.g., speed limits, safe distances, comfortable acceleration and jerk values. Moreover, a simulator is developed in MATLAB for the entire framework (including both the higher level and the lower level of the traffic control framework) to evaluate the impact of all the triplets on continuous mixed traffic flow. Different penetration rates of CAVs under different traffic flow conditions are evaluated through the developed simulator. The simulation results show that the proposed cooperative methodology, in comparison to the non-cooperative operation, can improve the average travel time of merging vehicles while not disturbing the mainstream flow, provide safer merging maneuvers by avoiding the merging vehicles to stop at the end of the acceleration lane, and guarantee smooth motion trajectories for CAVs (i.e., derivable position and speed along with limited changes in acceleration values).

Generally, the results of this thesis emphasize that the proposed cooperative traffic control framework can improve the traffic conditions of the mixed traffic environment in terms of both traffic safety and traffic operations. Moreover, the developed simulator provides a tool for the transportation community to evaluate their existing infrastructures under different penetration rates of CAVs and examine different traffic control plans for a mixed traffic environment. As the merging maneuver is only one application of gap-acceptance models, other types of maneuvers from the same category (e.g., lane changing, vehicle turning, etc.) can be similarly modelled. Thus, we can extend the proposed framework to the multi-lane highways and urban area intersections. Furthermore, the arrival time prediction of the vehicles can be improved to elevate the performance of the proposed framework during the very congested traffic conditions.

Back to top Back to top

© Concordia University