Dynamic Optimization Of Ecological Driving Speed For Intelligent And Connected Vehicle At Signalized Intersections

Growing transportation activities have enhanced both the mobility of people and goods significantly,but they have also increased greenhouse gas emissions and energy consumption.Stricter energy-saving and emission-reduction rules for vehicles have been developed in an effort to lessen their negative effects on the environment.Energy conservation and emission reduction technologies have become a hot issue of vehicle development,necessitating the need to continually increase the vehicle’s energy economy,whether as a result of exogenous mandated rules or endogenous product initiatives.In light of the growing marginal effect of fuel-saving internal combustion engine technology,the low energy density of batteries in electric vehicles,the deployment difficulty and the energy economy of many energy-saving technologies are inconsistent.As a prominent technology to improve vehicle energy efficiency,eco-driving has been widely studied.The primary idea of eco-driving is controlling the vehicle speed drive in an energy-efficient manner with significant energy-saving potential and good implementability.Additionally,because of the vast number of vehicles and the complex traffic environment,the level of vehicle energy consumption is high in urban traffic.To improve vehicle energy efficiency and traffic throughput,it is required to develop an eco-approach and departure strategy that uses traffic and vehicle information at the signalized intersection.Although researchers have made several theoretical and practical discoveries in the field of ecoapproach and departure control,the current research needs to be enhanced and refined:(1)How to deeply integrate intelligent vehicles and smart roads,build intelligent traffic information and vehicle dynamics coupled optimization problems and multi-objective coordinated control scheme,and improve vehicle energy efficiency and traffic throughput in intersection crossing scenarios?(2)How to design an eco-approach and departure method with excellent energysaving effects and strong scenario adaption capabilities,to expand the operational design domain for adapting to the dynamic traffic environment?For solving the aforementioned issues,research on eco-approach and departure control of the intelligent and connected vehicle(ICV)is being done to increase vehicle energy economy and traffic throughput.This study focuses on two levels: "hybrid high-dimensional vehicle speed optimization system construction" and "operational design domain expansion of eco-approach and departure method," by revealing the vehicle energy-saving potential and influence mechanism of intersection crossing scenarios,and designing eco-approach and departure method with vehicle queues,dynamic preceding vehicle,and multiple intersections.Finally,simulation and field experiments are used to investigate the validity of eco-driving strategy.Through the aforementioned study,we hope to contribute to vehicle energy conservation and emission reduction requirements in the transportation sector by providing the theoretical foundation and methodological support for safe,energy-saving,and efficient traffic control at signalized intersections.In summary,the following are the key research contents and contributions of this study.(1)To address the unknown influence mechanism of vehicle energy-saving control at the signalized intersections.The connected electric vehicle model is formulated,and an ecoapproach and departure optimization control problem is proposed,which involves energy consumption,vehicle dynamics,and traffic environment.Then,two regular intersection crossing control methods are defined,and the typical scenarios are selected to investigate vehicle energy-saving mechanisms.The aforementioned studies serve as model and theoretical foundation for following research on eco-approach and departure control.(2)The enhanced eco-approach control(EEAC)strategy is developed to solve the problem of energy-saving benefits of eco-approach and departure methods deteriorating due to vehicle queue blocks ego vehicle movement.First,a vehicle queue discharge prediction model is designed that takes into account the composition of heterogeneous vehicle queue while combining driver reaction and vehicle kinematics.Then,based on the prediction of the queue,the EEAC strategy is proposed with a hierarchical framework.The upper-stage uses dynamic programming to find the general trend of the energy-efficient speed profile,which is followed by the lower-stage model predictive controller to compute the explicit solution for a short horizon with guaranteed safe inter-vehicular distance.(3)The spatial and temporal scales of the preceding vehicle and traffic light are different,resulting in poor applicability of regular traffic scenario-oriented eco-approach and departure methods.In this context,an event-driven energy-efficient driving control(EEDC)strategy with a receding optimization scheme is proposed,which harnesses traffic lights and preceding vehicles to adapt to various traffic scenarios.First,the vehicle driving event classification rules are proposed,which classify the urban traffic scenarios into four events.Then,utilizing Pontryagin’s minimal principle and taking into account vehicle dynamics,control input,and speed limit constraints,an analytical solution for energy-efficient driving is constructed.The conditions of constraints activation are analyzed and a receding optimization framework is designed for real-time energy-efficient speed optimization.(4)To address the problem of spatial and temporal coupling between multiple intersections,which makes coordinating the goals of low energy consumption and travel time challenging.The relationship between vehicle energy-efficient operations and signalized traffic light states is investigated,then a multi-intersections-based eco-approach and departure strategy(M-EAD)is proposed to improve vehicle energy efficiency,traffic throughput,and battery life,while maintaining acceptable driving comfort.M-EAD is a two-stage control scheme that includes efficient green signal window planning as well as energy-saving speed optimization.The traffic light green signal window planning is defined as the shortest path problem in the upper stage and addressed using an A* algorithm for travel time reduction.The speed optimization problem is tackled in the lower stage by employing a receding horizon,in which energy consumption and battery capacity losses are reduced using an iterative dynamic programming algorithm.(5)To solve the problem that it is difficult to simulate the dynamic of real vehicle and traffic conditions,resulting in the inability to quantify the actual application of the eco-driving method.We developed an eco-driving test platform for ICV.The test platform is made up of two parts:a connected electric vehicle and an internet of vehicle system.Based on the experiment objectives and an appropriate test route in an open traffic environment,we developed a realistic test plan and process.The EEDC and M-EAD strategies’ effects on real-world applications have been fully validated.As a result of the extensive study completed above,a comprehensive analysis of theoretical approaches to engineering practice was conducted.In the future,based on the findings of this paper,we will develop a stochastic optimization method for eco-driving speed that incorporates dynamic traffic,investigate the multiple ICV control method that balances energy-saving and platoon stability,and further tap the potential for vehicle energy conservation and emission reduction by using vehicle-road-cloud integrated control.