部分车辆联网条件下多车协同避撞算法研究

发布时间：2018-01-23 04:56

本文关键词： 部分车辆联网车车通信协同避撞智能车辆　出处：《清华大学》2015年硕士论文　论文类型：学位论文

【摘要】：作为智能交通领域的研究热点,车联网技术近年来得到了越来越多的关注。与美国、日本和欧洲不同,中国车联网技术的研究起步较晚,近十年才在政府支持下开展较为系统的研究,但是车联网技术在中国的研究进程很快,许多汽车企业已经参与到车联网技术的研究中。然而,车联网技术的全面应用必然会经历过渡阶段,即联网车辆与非联网车辆共存的交通环境。在这种部分车辆联网条件下,许多基于车联网的应用将受到极大限制甚至完全失效,这无疑不利于车辆网技术的发展与推广。为充分利用车联网技术提供的信息,针对部分车辆联网环境开发安全应用将是面对的新问题。同时,在部分车辆联网条件下,联网车辆对非联网车辆乃至整个车辆队列安全到底会产生怎样的影响、产生多大的影响,也是车联网技术在实际应用中值得关注的问题。针对这两个问题,本文开展了如下研究:(1)提出了相对动能密度的概念,在车联网环境下设计了总相对动能密度最小的多车协同避撞算法,并采用模型预测控制对该算法进行了实现,针对10辆车组成的车辆队列的首车紧急制动工况,进行了1000组对比仿真,验证了所提算法的有效性。(2)分析了总相对动能密度最小的多车协同避撞算法应用于部分车辆联网环境的可行性,对算法进行了修正,结合改进的驾驶员心理跟驰模型,进行了部分车辆联网环境下多车协同避撞仿真研究,探明了部分车辆联网环境下联网车辆对车辆队列整体安全的影响规律。(3)针对包括10辆车的车辆队列的首车紧急制动工况,进行了大量不同联网车辆比例下协同避撞仿真研究,通过对仿真结果的统计分析,得到了联网车辆比例与车辆队列安全性的定量关系,确定了安全性提高效率最高和连环追尾抑制效率最大的联网车辆比例。此外,分析了联网车辆对非联网车辆制动减速度的影响规律。(4)搭建了车车协同控制试验平台,基于该试验平台开展了3辆车组成的车辆队列首车紧急制动实车试验,定性验证了联网车辆对队列安全性的积极作用。研究结果表明,在部分车辆联网条件下,通过开发车车协同安全应用,联网车辆能够提升车辆队列整体安全性,并且安全性的提升随联网车辆比例的提高而增加,因此,车联网技术具备进一步研究并推广应用的价值。
[Abstract]:As a research hotspot in the field of intelligent transportation, vehicle networking technology has received more and more attention in recent years. Different from the United States, Japan and Europe, the research of vehicle networking technology in China started late. In the past ten years, more systematic research has been carried out under the support of the government, but the research process of the vehicle networking technology in China is very fast, many automobile enterprises have already participated in the research of the vehicle networking technology. The full application of vehicle networking technology will inevitably go through a transitional stage, that is, the traffic environment in which networked vehicles and non-networked vehicles coexist. Many applications based on vehicle networking will be greatly restricted or even completely invalid, which is no doubt not conducive to the development and promotion of vehicle network technology, and provides information for making full use of vehicle networking technology. It will be a new problem to develop the safety application of partial vehicle networking environment. At the same time, under the condition of partial vehicle networking, what kind of impact will the networked vehicle have on the safety of non-connected vehicles and even the whole vehicle queue. It is also a problem to pay attention to in the practical application of the vehicle networking technology. In view of these two problems, this paper has carried out the following research: 1) put forward the concept of relative kinetic energy density (RKD). A multi-vehicle cooperative collision avoidance algorithm with minimum total relative kinetic energy density is designed in the networked vehicle environment, and the model predictive control is used to implement the algorithm. According to the emergency braking condition of the first vehicle in the vehicle queue composed of 10 vehicles, 1000 sets of comparative simulation are carried out. The validity of the proposed algorithm is verified. (2) the feasibility of applying the multi-vehicle cooperative collision avoidance algorithm with the minimum total relative kinetic energy density to the partial vehicle network environment is analyzed, and the algorithm is modified. Combined with the improved driver's psychological car-following model, the simulation of multi-vehicle cooperative collision avoidance under partial vehicle networking environment is carried out. The influence of networked vehicles on the overall safety of the vehicle queue under partial vehicle networking environment is investigated. (3) the emergency braking condition of the first vehicle which includes 10 vehicles in the queue is investigated. Through the statistical analysis of the simulation results, the quantitative relationship between the proportion of networked vehicles and the safety of vehicle queue is obtained. The proportion of networked vehicles with the highest efficiency of safety improvement and maximum efficiency of rear-end suppression is determined. The influence of networked vehicles on braking and deceleration of non-networked vehicles is analyzed. (4) A vehicle collaborative control test platform is built. Based on the test platform, a three-vehicle queue first vehicle emergency brake real vehicle test is carried out, which qualitatively verifies the positive effect of networked vehicles on queue safety. The research results show that under the condition of partial vehicle networking, the experimental results are as follows. Through the development of vehicle-vehicle cooperative safety applications, networked vehicles can enhance the overall safety of the vehicle queue, and the increase of safety with the increase of the proportion of networked vehicles, so. Vehicle networking technology has the value of further research and application.
【学位授予单位】：清华大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495;U463.6

【相似文献】