自动驾驶对沥青面层寿命及养护成本影响分析

自动驾驶车辆具有鲁棒的横向控制稳定性与居中车道行驶偏好，这将显著影响车辆轮迹横向分布。集中的轮迹分布将加速沥青路面车辙发展，应合理利用自动驾驶车辆横向控制，量化轮迹横向分布对路面车辙发展以及建设养护成本的影响。以浙江上三高速为对象，利用视频观测现阶段车道横断面车辆横向分布状态并测量路面车辙深度，提取了车辆横向分布基准参数；据此设计了几种轮迹分布函数并构建轮迹横向分布实施流程；通过室内试验采集材料参数，建立上三高速路面结构有限元仿真模型，分析了不同自动驾驶车辆比例以及横向分布函数下沥青罩面使用年限以及年均建设养护成本。研究结果表明：按照现阶段自动驾驶车辆小于10 cm的横向分布标准差，当车辆全部自动化后，路面车辙达到15 mm的年限将由11年多提前至8年，上三高速沥青罩面年均建设养护成本将增加37.4%(达到49.9万元/km)。随着横向分布标准差的逐渐增长，自动驾驶车辆的不利作用将会逐渐减弱；当其值大于25 cm时，自动驾驶车辆将有利于减缓路面车辙发展进程，且此时自动驾驶车辆的比例越高，减缓效果越明显；均匀分布状态下，路面车辙年限将推迟约3年(升至14年)，年均建设养护成本将减少约24.2%(降至27.6万元/km)。因此，合理控制自动驾驶车辆轮迹横向分布将有效降低沥青面层建设养护成本。

Influence of Autonomous Vehicles on Service Life and Maintenance Cost of Asphalt Pavements

Autonomous vehicles tend to drive along the center of a lane using a robust lateral controller, which significantly influences the entire lateral distribution of the wheel path. The centralization of the wheel path can facilitate the development of asphalt-pavement rutting; thus, it is necessary to appropriately utilize the lateral controller of autonomous vehicles and quantify the influence of the lateral distribution on the development of pavement rutting and maintenance costs. Considering the Zhejiang Shangsan Expressway as an example, the lateral distribution of human-driven vehicles in the cross-section of the lane was collected with a camera, and the rutting depth across the lane was measured. The data were used as a reference for the lateral distribution of autonomous vehicles. Several functions with different standard deviations were proposed, and a scheme for realizing the lateral distribution was developed. A finite-element model of the Shangsan Expressway was constructed using specific material parameters collected from laboratory tests. Thereafter, the service life and maintenance costs of asphalt pavements were calculated considering different ratios of autonomous vehicles and lateral distribution functions. The results indicate that, if the lateral standard deviation is less than 10 cm (the current performance of autonomous vehicles) and all vehicles are autonomous, the time required for the rutting depth to reach 15 mm will decrease from 11 years to 8 years, and the annual cost of maintaining the asphalt pavement of Shangsan Highway will increase by 37.4% (at 499 000 yuan per kilometer). As the standard deviation increases, the disadvantages of using autonomous vehicles become severe. When the standard deviation is greater than 25 cm, autonomous vehicles will slow down the development of pavement rutting. The higher the proportion of autonomous vehicles, the more evident the slowing effect. When a uniform distribution is used, the time when the rutting depth reaches 15 mm will increase by approximately three years (to 14 years), and the annual maintenance cost will decrease by approximately 24.2% (at 276 000 yuan per kilometer). Therefore, automatic driving will effectively reduce the cost of maintaining the asphalt pavement.