Cooperative Optimization Of Vehicle Trajectories And Signal Timings For Connected And Automated Arterials

With the rapid development of connected and automated vehicle(CAV)technologies,the penetration rate of connected and automated vehicles will continue to increase in future road traffic flow,until fully connected and automated driving system is achieved.When all vehicles in the road system are CAVs,they may be able to strictly follow real-time driving commands sent from control centers.Combining innovative traffic organization strategies and optimal control theories,the entire driving trajectories of CAVs can be optimized to improve the traffic flow efficiency at intersections and along arterials.During the large-scale application of CAVs,for a considerable period of time,mixed traffic flow consisting of both human-driven vehicles(HVs)and CAVs will be the norm.The relationship among elements within the mixed traffic system will be extremely complex.Studying the dynamic interaction mechanism of different vehicles and performing cooperative optimization of vehicle trajectories and signal timings may still have large potential to leverage the advantages of CAV technologies,so as to improve the overall system performance.Under connected and automated environment,this thesis focuses on the cooperative optimizations of vehicle trajectories and signal timings in the following two scenarios:one is the pure CAV environment where all vehicles in the road system are CAVs;the other is the mixed traffic environment where vehicles in the road system consist of both HVs and CAVs.Under the pure CAV environment,in order to improve the traffic flow efficiency,a cooperative optimization method for vehicle trajectories and signal timings is proposed,based on the concept of tandem intersection.For any road segment between two intersections,different algorithms for controling the lateral and longitudinal trajectories of CAV groups are proposed,to achieve tandem arrangement of different turning CAVs.Then,a mixed-integer linear programming model is established to optimize the CAV longitudinal speeds and signal timings along arterials,enabling CAVs to pass through all intersections on the arterials without stops as soon as possible.Comparisons and analyses are performed on various control schemes derived with different control algorithms and different target tandem arrangements,which show the effectiveness of the proposed cooperative optimization method for vehicle trajectories and signal timings along pure connected and automated arterials.Under mixed traffic environment,a two-stage control methodology is proposed to achieve the cooperative optimization of vehicle trajectories and signal timings.In the first stage,a cooperative lane-changing method for CAVs is proposed to ensure that all CAVs change lane to their corresponding turning lanes.In the second stage,by solving the established cooperative optimization model of vehicle trajectories and signal timings,the signal timing scheme is updated and the speeds of vehicle platoons are adjusted,to ensure that all vehicle platoons can pass through the intersections without stops.By applying the twostage control methodology to each entrance along arterials,a rolling optimization framework for the cooperative optimization of vehicle trajectories and signal timings is established.A connected and automated simulation testbed is built in SUMO simulation platform,which is capable to achieve real-time optimization and updating of control variables,to achieve cooperative rolling optimization of vehicle trajectories and signal timings for arterials.Considering various performance measures,massive simulation experiments are conducted under different control schemes,different CAV penetration rates,and different demand levels,to validate the effectiveness of the cooperative optimization of vehicle trajectories and signal timings for arterials with mixed-autonomy traffic.