摘要
为提高智能车辆在城市多信号灯路口道路通行时的燃油经济性,提出一种基于分层优化的节能车速规划方法。上层采用随机撒点方式划分可通行区域,基于车辆油耗模型构建加权有向无环图,利用Dijkstra算法求解加权有向无环图粗解,利用三次样条插值模型进行轨迹平滑;下层将车速规划最优控制问题转化为非线性优化问题,利用上层平滑后的轨迹作为内点法优化的初始迭代值。仿真结果表明,所提出的方法比改进的智能驾驶员模型(M-IDM)的油耗降低25%左右,与动态规划(DP)相比较,所提优化方法提高了计算速度,并兼顾节能与实时性计算要求。
To improve the fuel economy of intelligent vehicles at urban multi-signal junctions, an energy-saving speed planning method based on hierarchical optimization was proposed. In the upper layer, the feasible regions are planned by the random scattering algorithm. A weighted directed acyclic graph was constructed based on the vehicle fuel consumption model, and the Dijkstra algorithm was used to find the rough solution. The initial speed profile is smoothed via cubic spline interpolation. In the lower layer, the vehicle speed planning optimal control problem is transform into a nonlinear optimization problem, and uses the solution trajectory of the upper layer as the initial iteration value of the interior point method optimization. The simulation results show that the proposed method can reduce energy consumption by 25% compared with Modified Intelligent Driving Model(M-IDM) and the proposed method improves the calculation speed compared with Dynamic Programming(DP). The proposed method balances the requirements of energy-saving and real-time computation.
作者
韦麟
赵清海
Wei Lin;Zhao Qinghai(College of Mechanical and Electrical Engineering,Qingdao University,Qingdao,China;National Engineering Research Center for Intelligent Electrical Vehicle Power System,Qingdao University,Qingdao,China)
出处
《科学技术创新》
2023年第3期211-215,共5页
Scientific and Technological Innovation
基金
国家自然科学基金(批准号:52175236)资助。
关键词
智能交通
车速规划
DIJKSTRA算法
内点法
燃油车
多信号灯路口
intelligent transportation
speed planning
Dijkstra algorithm
interior-point
fuel vehicle
multi-signal intersections
