基于环境复杂度的危险品运输车辆碰撞预警策略

为研究传统车辆碰撞预警系统误警率较高、不完全适用于危险品运输车的问题,文中在碰撞预警策略中引入环境复杂度的概念。对静态复杂度和动态复杂度影响因子进行具体分析,提出了危险判别指标的计算方法和预警策略的具体框架,在此基础上结合国内公开的运输载货车事故报告及真实运输路段的环境信息,利用神经网络对非线性关系的高拟合度建立环境复杂度量化模型,收集真实路段下带有普通预警系统的危险品运输车辆实车数据并进行策略验证。研究结果表明:相较于市面上普通的碰撞预警系统,所提出的预警策略误警率从41％降低至8％,分类后的正确率达92％,可以较好地适应驾驶人对危险的判别和制动习惯,提高危险品运输车辆运行的安全性。 

Collision Warning Strategy of Dangerous Goods Transportation Vehicles Based on Environmental Complexity

In order to solve the problem that the traditional vehicle collision warning system has a high false alarm rate and is not completely suitable for dangerous goods transport vehicles, the concept of environmental complexity is introduced into the collision warning strategy. By analyzing the influence factors of static complexity and dynamic complexity, the calculation method of risk identification index and the specific framework of early warning strategy were proposed. On this basis, combined with the domestic open truck accident reports and the environmental information of the real transportation section, the quantitative model of environmental complexity was established by using the high fitting degree of neural network to the nonlinear relationship. The real vehicle data of dangerous goods transportation vehicles with common warning system in real road section were collected and the strategy was verified. Compared with the common collision warning system on the market, the false alarm rate of the proposed warning strategy is reduced from 41% to 8%. The classification mechanism can better adapt to the driver"s judgment of danger and braking habits, and improve the safety of dangerous goods transport vehicles.