基于模糊着色Petri网的车辆行驶状态估计

车辆行驶状态的有效估计是改善综合交通运输效能的有效途径,针对车辆行驶过程数学模型难以准确建立,依据车辆的行驶状态流跟Petri离散并行系统的相似性,提出利用具有良好层次化和时序性特点的Petri网建立车辆行驶状态估计模型。分析影响车辆状态变化的影响因素,将车速、车辆质心侧偏角以及车辆横摆角速度作为描述车辆状态属性指标,并确定输入的论域以及隶属度函数,依据车辆运动状态的可控性和驾驶舒适性建立相应的模糊规则。最后,在CPNtool中构建车辆行驶状态估计模型,采集路车试验数据对模型进行训练与测试,结果表明,Petri网模型不仅能够以可视化方式充分展现车辆状态变化过程,且能够通过着色的库所变迁确定影响车辆状态发生的关键性因素,模型的估计结果逼近真实值。

Estimation of Vehicle States Based on Fuzzy Colored Petri Net

The state effective estimation of vehicle is an effective way to improve traffic transportation efficiency.Because it is difficult accurately set up mathematical model for vehicle motion process.According to the similarity between driving state flow and a discrete parallel system(petri), Petri net which has good hierarchical and sequential characteristics is proposed to establish state estimation model for vehicle.After analysis on influence factors of vehicle state, the speed, side-slip angle of vehicle centroid and yawing angular velocity are taken as the attribute indexs of vehicle state.Then we determine the domain of discourse of input and membership functions.The corresponding fuzzy rules can be established through vehicle motion controllability and driving comfort. Finally, vehicle state estimation model is built in CPNtool and the datas of road vehicle are used to train and test the model.The results show that petri net model can not only show vehicle state change process in visual way,and also the key factors affect vehicle state change can be determined by the colored places and transitions.In addition,the model estimation results are closer to the real value.