考虑CPS传输可靠性的车车协同稳定性分析

发布时间：2018-04-24 03:11

本文选题：信息物理系统 + 稳定性　；参考：《重庆大学》2015年博士论文

【摘要】：交通阻塞和拥堵问题其本质是系统失稳导致的。缓解交通拥堵就是要探索如何改进交通系统的稳定性和提高抗干扰能力。因此,交通流稳定问题在科学认知交通拥堵事件生成机理,准确预测其特征和演化路径方面占有核心基础地位。从系统的观点,交通拥堵导致的根本原因是由于交通系统中各要素未能实现充分的协调与优化,车流运行产生扰动且不能在扰动下保持自稳定能力所致。因此要从根本上解决交通拥堵问题,其关键是要寻求一种能实现信息系统和物理系统深度融合与协作的理论。信息物理融合系统(cyber physical system,CPS)的出现为上述问题的解决提供了新途径。信息物理融合引入到交通系统后,在信息空间中由传输网络通信带来的传输时延和丢包等不可靠性因素将不可避免和忽略。由于信息空间和物理空间深度融合的原因,交通信息空间的这些不可靠因素必然会映射到物理世界,作用于实际物理对象或过程,导致包括车车协同在内的车流产生不稳定和振荡。由此可见,对于车车协同运行这类对控制品质和稳定性要求高的系统来说,不能忽视上述影响。然而,遗憾的是现有车车关系刻画模型建立的多是物理模型,尚未深入对等考虑物理及信息二元之间的密切关系。这导致现有车车协同模型及其稳定性研究成果难以刻画信息空间传输不可靠因素带来的冲击效应。基于此背景,本文在借鉴现有交通流理论研究成果的基础上,围绕车车协同的稳定性问题,从信息物理融合的角度,重点研究信息空间传输不可靠性因素(时延和丢包)对车车协同物理过程稳定性的影响,并采用理论分析和数值仿真手段,探索伴随稳定性变化的各种非线性现象。论文的主要工作如下:①考虑多前车平均速度信息,构建了CPS下车车协同行驶模型。在此基础上,获得了多前车信息可靠性对车车协同稳定性的影响规律。基于CPS信息的应用,考虑多前车的平均速度信息对跟随车操控稳定性的影响,提出了新的车车协同行驶模型。运用线性稳定性分析获得了模型的线性稳定性判据。运用非线性分析手段,导出了刻画临界点附近交通阻塞传播规律的m Kd V方程,数值仿真和理论分析结果一致表明:多前车平均速度信息对车流稳定性具有显著改善作用。在此基础上,引入多前车信息可靠性度量参数,研究了多前车信息可靠度对车车协同的稳定效应,结果表明:在同一CPS传输可靠程度下,车流的稳定性随着多前车数量的增加而逐渐增加;当多前车数量固定时,交通系统的平均速度随着CPS传输可靠程度的增加而逐渐提高。②针对带超车情形的车车协同行驶场景,考虑CPS中网络时延对其稳定性的影影响,提出带时滞的流量差格子模型,获得模型的线性稳定性判据及不稳定区域交通阻塞的非线性特征。针对具有超车情形的Nagatani格子流体力学模型没有涉及实际交通中的流量差效应,提出有超车情形的流量差格子模型。研究结果表明:流量差效应对车流稳定性及其非线性特性具有不可忽视的影响。在此基础上,进一步考虑CPS中网络时延对车车协同行驶车流稳定性的影响,提出了有超车情形下的时滞流量差格子模型。获得了模型的稳定性判据和临界点附近交通阻塞传播演化规律。研究结果表明,车流稳定性随着CPS网络时延的增加而逐渐降低。对于不稳定区域车流,当超车通过率较小时,不稳定车流以m Kd V方程描述的扭结-反扭结波形式传播;当超车通过率较大时,相空间的不稳定区域可进一步划分为密度波特征区域和混沌特征区域。在密度波特征区域,失稳车流以周期行为演化,而混沌特征区域失稳车流展现出无序震荡的混沌行为。③针对两车道系统的车车协同场景,提出了考虑最优流量差估计信息的车车协同行驶模型,在此基础上,揭示了CPS中传输可靠性对宏观车流稳定性的影响规律。基于Nagatani格子流体力学模型思想,考虑最优流量差估计效应对宏观车流的影响,构建了新的两车道车车协同行驶模型。在此基础上,获得了两车道系统车流的稳定条件及描述交通阻塞演化规律的密度波方程。研究结果表明:两车道间的换道效应和最优流量差估计信息协同作用可有效增强车流的稳定性。引入概率参数表征了CPS的传输可靠性,并引入各路段车流密度分布均方差统计量,作为CPS传输可靠性下车流稳定性的衡量指标。研究结果表明,无论是否允许车辆换道,两车道系统的车流稳定性均随着CPS传输可靠性的增加而提高。④基于CPS信息的应用,提出了计及驾驶员预估效应的前后车协同模型,在此基础上,揭示了前后车协同稳定性与车流能耗间的关联关系。引入CPS系统采集的驾驶员预估信息,基于全速度差跟驰模型,提出了考虑驾驶员预估效应的前后车协同模型。运用线性稳定性定理获得了模型的稳定性判据,结果表明,驾驶员预估效应下的前后车协同能显著提高车流稳定性,降低交通系统能耗。在此基础上,通过耦合车辆能耗公式,系统研究了前后车协同稳定性与单车能量耗散和道路整体能量耗散的关联关系。研究结果表明,车流稳定性与能耗在时空演化过程中均展现出复杂临界相变现象,且他们的相变过程具有显著映射关系。综上所述,本文从信息物理融合的角度建立了面向CPS的车车协同模型。在此基础上,揭示了信息空间中传输不可靠性问题对车车协同物理过程稳定性的影响规律。研究成果可望为CPS环境下车车协同行驶稳定性的系统认知提供支撑,为自动驾驶控制策略的设计提供一定的借鉴作用。
[Abstract]:The problem of traffic congestion and congestion is caused by system instability. To alleviate traffic congestion is to explore how to improve the stability of traffic system and improve the ability of anti-jamming. Therefore, the problem of traffic flow stability plays a key role in scientific cognition of the formation mechanism of traffic congestion events and the accurate prediction of its characteristics and evolutionary paths. From the point of view of the system, the root cause of traffic congestion is due to the failure of the full coordination and optimization of the elements in the traffic system, the disturbance of the traffic flow and the ability to maintain self stability under the disturbance. Therefore, to solve the traffic congestion problem fundamentally, the key is to seek a kind of information system and physics system. The emergence of cyber physical system (CPS) provides a new way to solve the problems mentioned above. After the introduction of Information Physics fusion to the traffic system, the unreliability factors such as transmission delay and packet loss in the information space will inevitably and overlook the unreliability factors. Because of the deep fusion of information space and physical space, these unreliable factors of traffic information space must be mapped to the physical world, acting on the actual physical objects or processes, resulting in the instability and oscillation of vehicle flow including car and car coordination. However, it is regrettable that the existing vehicle vehicle relationship depicting models are mostly physical models, and the close relationship between the two elements of physics and information has not been considered in depth. This leads to the difficulty of describing the unreliable factors in the transmission of information space by the research of the existing vehicle and car coordination model and its stability. On the basis of this background, on the basis of the existing traffic flow theory research results, this paper focuses on the influence of the unreliability factors (delay and packet loss) on the physical process stability of vehicle vehicles from the point of view of Information Physics fusion, and uses theoretical analysis and Analysis on the basis of the existing traffic flow theory research results. The main work of this paper is as follows: (1) considering the average speed information of the multi front vehicle, the cooperative driving model of CPS vehicle is constructed. On this basis, the influence of multi vehicle information reliability on the coordination stability of the vehicle is obtained. Based on the application of CPS information, the multi vehicle information reliability is considered. The influence of the average velocity information of the front car on the stability of the vehicle handling, a new model is proposed. Linear stability criterion is obtained by linear stability analysis. The m Kd V equation, numerical simulation and theoretical analysis are derived by using nonlinear analysis method. The results show that the multi front vehicle average speed information has a significant effect on the vehicle flow stability. On this basis, the multi front vehicle information reliability measurement parameters are introduced to study the stability effect of the multi front vehicle information reliability on vehicle coordination. The results show that the stability of the traffic flow is with the number of multiple front vehicles under the same CPS transmission degree. When the number of multiple front cars is fixed, the average speed of the traffic system increases with the increase of the reliability of the CPS transmission. Secondly, considering the effect of the network delay on the stability of the car and car in the case of overtaking, the delay lattice model with time lag is proposed, and the linear stability of the model is obtained. The qualitative criterion and the nonlinear characteristics of the unstable regional traffic congestion are not related to the flow difference effect in the actual traffic. The flow difference lattice model with the overtaking situation is proposed. The results show that the flow difference effect can not be used to deal with the stability and nonlinear characteristics of the Nagatani flow. On this basis, the effect of network delay on the stability of vehicle traffic flow in CPS is further considered, and a delay lattice model with time delay is proposed in the case of overtaking. The stability criterion of the model and the propagation evolution law of traffic congestion near the critical point are obtained. The results show that the stability of the vehicle flow is with the CPS net. For the unstable regional vehicle flow, when the overtaking rate is small, the unstable traffic flow is propagating in the form of the kink anti kink wave described by the m Kd V equation. When the passing rate is large, the unstable region of the phase space can be further divided into the density wave characteristic region and the chaotic characteristic region. In the region, the unstable car flow is evolutional with periodic behavior, and the chaotic characteristic region instability car flows out of chaotic behavior. Thirdly, based on the vehicle coordination scene of two lane system, a vehicle coordinated driving model considering the optimal flow difference estimation information is proposed. On this basis, the stability of the transmission reliability in the CPS is revealed to the stability of the macro traffic. Based on the Nagatani lattice fluid mechanics model thought, considering the effect of the optimal flow difference estimation effect on the macro traffic flow, a new two lane vehicle coordinated driving model was constructed. On this basis, the stability conditions of the two lane system and the density wave equation describing the evolution of the traffic stopper were obtained. The results showed that the density wave equation was used to describe the traffic stopper evolution law. The interaction between the two lanes and the optimal flow difference estimation information synergy can effectively enhance the stability of the vehicle flow. The probability parameters are introduced to characterize the transmission reliability of the CPS, and the mean variance statistics of the traffic density distribution in various sections are introduced as a measure of the stability of the traffic flow reliability of the CPS transmission. The traffic stability of the two lane system is increased with the increase of the reliability of the CPS transmission. (4) based on the application of CPS information, a model of front and rear vehicle coordination with the driver prediction effect is proposed. On this basis, the relationship between the coordination stability of the front and rear vehicles and the energy consumption of the traffic flow is revealed. The driving of the CPS system is introduced. Based on the full velocity difference model, a front and rear vehicle synergistic model considering driver prediction effect is proposed. The stability criterion of the model is obtained by using the linear stability theorem. The results show that the front and rear vehicle coordination under the pilot prediction effect can significantly improve the stability of the vehicle flow and reduce the energy consumption of the traffic system. Through the coupling vehicle energy consumption formula, the relationship between the coordination stability of the front and rear vehicles, the energy dissipation of the single vehicle and the overall energy dissipation of the road is systematically studied. The results show that the stability and energy consumption of the vehicle flow show a complex critical phase transition in the process of time and space evolution, and the phase transformation process has a significant mapping relationship. From the point of view of information physical fusion, this paper establishes a CPS oriented vehicle coordination model. On this basis, it reveals the effect of transmission unreliability in the information space on the stability of the vehicle's coordination physical process. The research results are expected to provide support for the system cognition of the coordination stability of the vehicle in the CPS environment, and for automatic driving control. The design of system strategy can be used for reference.

【学位授予单位】：重庆大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：U495

【相似文献】