Table of Content

25 July 2021, Volume 43 Issue 7 Previous Issue    Next Issue

Select

Research on Multiple Objective Coordinated Control of Speed Planning for Intelligent Connected Hybrid Electric Vehicles Shaobo Xie,Huiran Luo,Qiankun Zhang,Kangkang Zhang 2021, 43 (7):  953-961.  doi: 10.19562/j.chinasae.qcgc.2021.07.001 Abstract ( 468 )   HTML ( 36 )   PDF (2931KB) ( 603 )   Save Considering the multiple objectives of vehicle safety， mobility， energy consumption economy， comfortability as well as battery aging， this paper conducts real?time speed planning for intelligent connected hybrid electric buses in the curve road scenario. Firstly， the objective function aims for minimizing the total weighted cost associated with energy consumption， battery aging， mobility and comfortability where the speed and battery state?of?charge are chosen as state variables， and the acceleration and engine?generator?unit output power are chosen as control variables. Then， the multi?objective coordinated control based on model prediction is implemented while satisfying the constraints of curve driving safety and the physical characteristics of powertrain and battery system. Moreover， the dynamic programming algorithm is applied to solve the multiple optimization problem over the preview horizon to realize real?time speed planning and energy allocation. At the same time， different weights of mobility and comfortability on performance are discussed. The results show that （1） the control strategy considering the battery aging can lower the aging cost and total cost by 25.8% and 2.3% respectively without affecting the vehicle power and mobility； （2） Improving the weight of mobility cost shortens the driving time， but raises the total cost； （3） Improving the weight of comfortability can constrain the speed fluctuation， and reduce the total cost. Figures and Tables | References | Related Articles | Metrics

Select

Two⁃dimensional Tracking Control Algorithm for Vehicle Platoon Based on Reference Vector Field Yang Liu,Changfu Zong,Hongyu Zheng,Xiaojian Han,Dong Zhang,Kaku Chuyo 2021, 43 (7):  962-970.  doi: 10.19562/j.chinasae.qcgc.2021.07.002 Abstract ( 210 )   HTML ( 10 )   PDF (1784KB) ( 324 )   Save When the operation scene of the vehicle platoon extends from 1D to 2D， a significant coupling effect will appear between the longitudinal spacing maintenance and lateral tracking control. In high speed conditions， the ignorance of the coupling effect will lead to large tracking error and even instability. To solve this problem， a longitudinal and lateral coupling tracking control algorithm for vehicle platoon is proposed in this paper. By constructing a reference vector field based on the longitudinal spacing and lateral reference trajectory， the desired velocity vector can be obtained， and the Hamiltonian function is used to calculate the desired total forces and total moment the upper?layer motion controller of vehicle required； Meanwhile， a pseudo?inverse matrix?based control allocation algorithm is put forward to distribute the desired total forces and total moment to each wheel under constrained physical environment， enhancing the real time performance while ensuring allocation accuracy. The results of simulation and experiment show that the longitudinal and lateral coupling tracking control algorithm proposed can effectively fulfill the movement control of vehicle platoon under 2D scenes， achieving the safe， high efficient cooperated driving of vehicle platoon. Figures and Tables | References | Related Articles | Metrics

Select

Study on Vehicle Cut⁃in Intention Prediction Based on Residual BiLSTM Network Jinghua Guo,Baoping Xiao,Jingyao Wang,Yugong Luo,Tao Chen,Keqiang Li 2021, 43 (7):  971-977.  doi: 10.19562/j.chinasae.qcgc.2021.07.003 Abstract ( 290 )   HTML ( 9 )   PDF (2527KB) ( 473 )   Save A vehicle cut?in intention prediction model based on Residual BiLSTM network is proposed in this paper according to the real road features in China. The cut?in features are extracted from the trajectory information of the cut?in vehicle and its interaction information with ego vehicle， and the softmax function is used to calculate the cut?in intention， e.i. the probability of left lane keeping， left lane cut?in， right lane cut?in or right lane keeping respectively. Finally， the prediction model is trained and tested with the naturalistic driving data set on the complex roads in China. The results show that the Residual BiLSTM model proposed has obvious advantages in cut?in intention prediction， with an accuracy 8.2 percentage points higher than that of LSTM model， and can predict the vehicle cut?in intention earlier， being of great significance in enhancing the decision?making and planning ability and safety of autonomous vehicles. Figures and Tables | References | Related Articles | Metrics

Select

Intelligent Vehicle Lane Changing Trajectory Planning Based on Double Quintic Polynomials Guochen Niu,Wenshuai Li,Hongxu Wei 2021, 43 (7):  978-986.  doi: 10.19562/j.chinasae.qcgc.2021.07.004 Abstract ( 574 )   HTML ( 27 )   PDF (3065KB) ( 673 )   Save In order to meet the requirements of safety and comfort during lane changing of the intelligent vehicle， an intelligent vehicle lane changing trajectory planning algorithm based on double quintic polynomials is proposed. The quintic polynomial programming algorithm is improved with the condition of dynamic programming of lane changing time and increased comfort constraints. On this basis， the transit state is calculated by combining the current environment and the beginning and end states of the lane changing， and the twice improved quintic polynomial algorithm is used to avoid collision with the vehicle in front. The simulation and experiment results of trajectory planning and trajectory tracking show that the proposed double quintic polynomials lane changing trajectory planning algorithm has advantages in lateral velocity， acceleration， acceleration rate of change and running time of the algorithm under different working conditions. In addition， the trajectory obtained can also meet the requirements of vehicle lane changing under the actual situation， with improved safety and good handling stability， which proves that the algorithm has certain practical application value. Figures and Tables | References | Related Articles | Metrics

Select

Research on Vehicle Vertical Parking Path Planning and Path Tracking Qiang Gao,Zhou Lu,Chendong Duan,Ting Xu 2021, 43 (7):  987-994.  doi: 10.19562/j.chinasae.qcgc.2021.07.005 Abstract ( 511 )   HTML ( 18 )   PDF (1794KB) ( 571 )   Save Automatic Parking System is an important part of intelligent vehicle research. Most of the current vertical automatic parking algorithms fail to consider the influence of the heading angle error of the initial postion of the vehicle， which leads to unsatisfactory parking or scratch as a result. For this problem， the vertical parking path planning and path tracking algorithms， when the heading angle of the initial position of the vehicle has error （i.e. the initial position heading angle is not zero）， are studied in this paper. Firstly， the quadric spline function is applied to plan the parking path to improve the path characteristics. Then， a path tracking control algorithm combining preview error feedforward and heading angle feedback is designed in this paper， in order to improve the performance of the current commonly used preview error feedforward tracking algorithm， which shows big error and obvious lag in variable curvature path tracking. The proposed parking algorithm is verified by Simulink/Carsim co?simulation and real vehicle test. Both the simulation and real vehicle test results indicate that when the initial position heading angle of the vehicle is not zero， the designed algorithm is able to plan a feasible parking path， while the path tracking control algorithm combining preview error feedforward and heading angle feedback has smaller tracking error and better final parking position， comparing with that just using preview error feedforward. Figures and Tables | References | Related Articles | Metrics

Select

Trajectory Tracking Coordinated Control for Autonomous Vehicle in High⁃speed Overtaking Zhiyong Zhang,Kai Long,Ronghua Du,Caixia Huang 2021, 43 (7):  995-1004.  doi: 10.19562/j.chinasae.qcgc.2021.07.006 Abstract ( 478 )   HTML ( 20 )   PDF (1961KB) ( 479 )   Save In high?speed overtaking of autonomous vehicles， it is necessary not only to plan a reasonable path for vehicle safety， but also to ensure the lateral stability and ride comfort of the vehicle in high?speed turning on curve. Firstly， the longitudinal speed and the lateral overtaking path are planned respectively for the three phases of overtaking， i.e. lane change， uniform speed and lane change. Then a calculation method of desired yaw rate with consideration of path curvature， lane change time， and longitudinal speed is put forward. Finally， with minimizing lateral position error， yaw rate tracking deviation and control increment as optimization objectives， and by using the correlation function of extension set to dynamically assign the weighting factors of trajectory tracking accuracy and lateral stability， an extension set?based multi?objective model predictive coordinated control strategy for autonomous vehicle trajectory tracking is established. Numerical simulation results show that the path planning method proposed can ensure the safe overtaking of vehicle， while the trajectory tracking control strategy set up can accurately track the planned path， with better lateral stability and ride comfort. Figures and Tables | References | Related Articles | Metrics

Select

Research on Ground Segmentation Algorithm Based on Adaptive Thresholds for 3D Laser Point Clouds Kai Zhang,Chunlei Yu,Yali Zhao,Yifei Chen,Mengmeng Yang,Kun Jiang 2021, 43 (7):  1005-1012.  doi: 10.19562/j.chinasae.qcgc.2021.07.007 Abstract ( 382 )   HTML ( 6 )   PDF (3750KB) ( 403 )   Save To address the problem of false background segmentation on 3D Lidar point clouds in the perception module of autonomous vehicles， an adaptive threshold segmentation method based on fluctuation range of road surface is proposed. At first， the original point cloud is divided into grids and the corresponding height threshold segmentation algorithm and the ground plane model segmentation algorithm are designed according to the number of points in a certain grid cell. Specifically， the ground plane model is fitted to a local point cloud subset inside the grid cell. Afterwards， an equation of road surface fluctuation is constructed for the problem of false segmentation in ground point cloud segmentation. Based on these point set distribution characteristics， an adaptive threshold method is used to realize an initial segmentation. Finally， the segmented point cloud subset is used to optimize the ground plane model and the segmentation threshold in an iterative manner. The paper proposes a unified benchmark dataset Semantic?Nova based on the open dataset Semantic?KITTI and performance evaluation indicators. Meanwhile， the performance test is conducted based on the actual scenes collected by the self?developed autopilot vehicle platform. The test results show that the adaptive threshold ground segmentation algorithm proposed in this paper can achieve high accuracy in benchmark dataset. Furthermore， it can meet the requirements of robustness and real?time applications in actual scenes， which has high engineering application value. Figures and Tables | References | Related Articles | Metrics

Select

3D Real⁃Time Vehicle Tracking Based on Lidar Hai Wang,Yang Li,Yingfeng Cai,Kai Sun,Long Chen 2021, 43 (7):  1013-1021.  doi: 10.19562/j.chinasae.qcgc.2021.07.008 Abstract ( 399 )   HTML ( 10 )   PDF (3716KB) ( 316 )   Save The 3D multi?object tracking algorithm is an essential part of the intelligent vehicle perception algorithm. The existing tracking algorithm is mostly coupled with the detection algorithm to improve the accuracy， resulting in insufficient real?time performance. To solve this problem， a 3D real?time vehicle tracking algorithm based on lidar is proposed. Firstly， for the working conditions with less clutter in the detection results of lidar， a double?validation gate GNN algorithm with a simple structure is proposed to effectively improve its correlation speed and accuracy； secondly， the correlation vector and correlation distance are optimized， which improves the tracking accuracy while ensuring the generality of the algorithm. Finally， the 3D IMM?KF algorithm is used to solve the tracking problem of 3D object with changing dynamics. The proposed algorithm achieves a MOTA accuracy of 81.55% at a tracking speed of 266.1 FPS according to the public data set KITTI. Based on the self?developed unmanned vehicle platform， the verification of facing occlusion conditions shows that the algorithm has good object tracking and correlation performance. Figures and Tables | References | Related Articles | Metrics

Select

Multi⁃target Detection and Tracking with Fusion of Millimeter⁃wave Radar and Deep Vision Yaodong Gan,Ling Zheng,Zhida Zhang,Yinong Li 2021, 43 (7):  1022-1029.  doi: 10.19562/j.chinasae.qcgc.2021.07.009 Abstract ( 405 )   PDF (3609KB) ( 506 )   Save Aiming at the problems of low accuracy and poor real?time performance of the existing vehicle detection algorithm fusing millimeter wave radar and traditional machine vision， multi?target detection and tracking are studied in this paper. Firstly， the millimeter?wave radar data is preprocessed by using threshold screening and the data association of adjacent frames， and an adaptive extended Kalman filter algorithm is proposed for millimeter?wave radar data tracking. Then， for enhancing the accuracy and speed of target detection， the convolutional neural network is trained based on the real?vehicle data set collected to complete multi?vehicle detection with deep vision. Finally， a decision?level fusion strategy is adopted to fuse the information of millimeter?wave radar and deep vision， and a framework for the multi?target detection and tracking of front vehicles is designed. Real vehicle tests under different traffic environment are carried out to verify the designed framework. The results show that the method adopted can detect and track the front vehicles in real time with higher reliability and robustness， compared with the vehicle detection method fusing millimeter wave radar and traditional machine vision. References | Related Articles | Metrics

Select

An Extraction Method of Scenario Elements for Autonomous Driving Simulation Xuesong Bai,Weiwen Deng,Bingtao Ren,Peng Liu,Jiangkun Li,Juan Ding 2021, 43 (7):  1030-1036.  doi: 10.19562/j.chinasae.qcgc.2021.07.010 Abstract ( 476 )   HTML ( 39 )   PDF (1123KB) ( 689 )   Save In view of the problems of the unclear definition of scenes and their content， the extraction of basic scene elements is limited to subjective analysis and the selection and design of scene element of different test subjects are inexplainable etc.， an extraction method of the key scene elements for vehicle autonomous driving simulation is proposed. The method analyzes the effects of scene elements on the modules of perception， decision?making and motion?control from the perspective of autonomous driving system， and establishes a mapping equation of element?structure?function plane according to the effects of different sub?modules of autonomous driving. Then， an extraction model of scene elements is set up based on the plane node discriminant matrix， to quantify the importance of scene elements with discrimination and screening performed. Finally， through analyzing the element composition of the pedestrian collision avoidance scenes in the 4th World Intelligent Driving Simulation Challenge， the feasibility and effectiveness of the method proposed are verified. Figures and Tables | References | Related Articles | Metrics

Select

Eco⁃driving Control for Hybrid Electric Vehicle Platoon with Consideration of Driver Operation Error Lijun Qian,Jian Chen,Bing Wu,Liang Xuan,Chen Chen,Liangliang Chen 2021, 43 (7):  1037-1045.  doi: 10.19562/j.chinasae.qcgc.2021.07.011 Abstract ( 189 )   HTML ( 6 )   PDF (2826KB) ( 258 )   Save In view of that the most existing studies on eco?driving control are based on fully intelligent network connected environment， which may not be suitable for the mixed traffic scenes covering traditional human?driven vehicles （HDVs） and network connected vehicles （CVs）， a hierarchical eco?driving control method for hybrid electric vehicles platoon consisting of HDVs and CVs is proposed in this paper with consideration of the operation error of drivers. The upper layer controller is designed by using stochastic model predictive control algorithm to fulfill the multi?objective optimization of the mobility， fuel economy and comfort of the platoon， while the adaptive equivalent consumption minimization strategy is adopted to design the lower layer controller for optimizing the power assignment between vehicle engine and battery. Simulations results show that the method proposed can effectively reduce the speed trajectory deviation of hybrid electric vehicles in the platoon caused by the driver operation error， with its average fuel consumption lowering by 2.82%. Figures and Tables | References | Related Articles | Metrics

Select

Research on Efficiency Optimization Based Energy Management Strategy for a Hybrid Electric Vehicle with Reinforcement Learning Ningkang Yang,Lijin Han,Hui Liu,Xin Zhang 2021, 43 (7):  1046-1056.  doi: 10.19562/j.chinasae.qcgc.2021.07.012 Abstract ( 274 )   HTML ( 25 )   PDF (4490KB) ( 446 )   Save Taking the power split hybrid electric vehicle as the object ，this paper establishes the model for calculating the system comprehensive efficiency and proposes an efficiency optimization based energy management strategy with reinforcement learning. Firstly， the efficiency model of key components and the efficiency model of coupling mechanism are established. Based on the general structure of stepless speed regulation of composite transmission， the influence law of power splitting coefficient on the efficiency is analyzed， and the system comprehensive efficiency model is further constructed. Then with efficiency optimization as the goal， an energy management strategy based on reinforcement learning is proposed. Simulation comparisons are implemented， and the results show that the proposed strategy can achieve excellent fuel economy while maintaining battery SOC within a smaller fluctuation range. Finally， a test bench is built and the test results prove the correctness of the established efficiency model and the effectiveness of the proposed energy management strategy. Figures and Tables | References | Related Articles | Metrics

Select

Research on Intelligent Vehicle Lateral Control Model Based on Neuro⁃Ergonomics Yingshi Guo,Hongjia Zhang,Rui Fu,Chang Wang 2021, 43 (7):  1057-1065.  doi: 10.19562/j.chinasae.qcgc.2021.07.013 Abstract ( 214 )   HTML ( 10 )   PDF (4172KB) ( 415 )   Save In the process of revealing the driving mechanism of the driver， some of the existing driver models are from the perspective of driver cognition alone and the other from the perspective of control alone， so there is a lack of a system model that organically combines the driver’s cognitive process and control principle. In order to solve the above problems， a vehicle lateral control model based on neuro?ergonomics is established by integrating model predictive control （MPC） algorithm and arm muscle model based on neuro?ergonomics cognitive architecture. The model is verified by test using the CarSim/Simulink co?simulation and dSPACE/driving simulator hardware in the loop. The results show that the trajectory tracking accuracy of the vehicle lateral control model based on neuro?ergonomics is better than that of the MPC algorithm. Additionally， the control of steering wheel angle， yaw angle and lateral acceleration fluctuation amplitude is also improved compared with MPC algorithm. Figures and Tables | References | Related Articles | Metrics

Select

Recognition of Pedestrians’ Street⁃crossing Intentions Based on Action Prediction and Environment Context Biao Yang,Fucheng Fan,Jicheng Yang,Yingfeng Cai,Hai Wang 2021, 43 (7):  1066-1076.  doi: 10.19562/j.chinasae.qcgc.2021.07.014 Abstract ( 325 )   HTML ( 14 )   PDF (4520KB) ( 481 )   Save In view of that pedestrian?vehicle collisions often happen in the process of pedestrians’ street crossing， a street?crossing intention recognition network MIFRN is proposed based on pedestrians’ action prediction and environment contexts in this paper. MIFRN encodes pedestrians’ future actions information， local traffic scenes surrounding pedestrians， vehicle speeds， and pedestrian?vehicle distance information respectively through structure?varying sub?networks， and predicts pedestrians’ intention of street crossing on the basis of information fusion. Finally， the performance of algorithm is verified based on two public databases PIE and JAAD. The results indicate that the method proposed can recognize pedestrians’ street?crossing intentions accurately and robustly. Figures and Tables | References | Related Articles | Metrics

Select

An Adaptive Cruise Control Scheme Based on Merging Behavior Recognition Yingfeng Cai,Lü Zhijun,Xiaoqiang Sun,Hai Wang,Qingchao Liu,Long Chen,Chaochun Yuan 2021, 43 (7):  1077-1087.  doi: 10.19562/j.chinasae.qcgc.2021.07.015 Abstract ( 208 )   HTML ( 9 )   PDF (3187KB) ( 206 )   Save In view of the uncertainty of braking intervention timing for the conventional adaptive cruise control system under side?car merging condition， an adaptive cruise control strategy is proposed which is optimized based on side car merging behavior. Firstly， with the historical driving data and surrounding environment as inputs and based on long short?term memory network， a driving behavior recognition model is set up to fulfill the effective recognition of the driving behavior category of side?lane vehicles. Once the merging behavior is recognized， an acceleration control is applied to the adaptive cruise system according to the motion state of merging vehicle， with a predictive control model for the system established. Then tracking performance， ride comfort and fuel consumption three performance indicators and constraint conditions are determined， and the desired acceleration is solved out by using the utopia point method， effectively avoiding the interference of manually selected weighting factors. Next， the first element of optimal control sequence is acting on the system for evaluating the system state information to achieve rolling optimization. Finally， a simulation model is established with MATLAB/Simulink to conduct a comparative simulation on three conditions of constant?speed cruising， vehicle tracking driving and merging， with real vehicle test performed for verification. The results show that the algorithm proposed can response to the change of tracked target faster in side car merging， effectively reduce the speed fluctuation and avoid the most of vehicle emergent braking， while the control model adopted with consideration of merging driving characteristics can enhance the ride comfort and reduce the safety risk of vehicle. Figures and Tables | References | Related Articles | Metrics

Select

Multi⁃material Topology Optimization of Automotive Control Arm Xiaokai Chen,Chao Li,Yingchun Bai,Zifa Yang 2021, 43 (7):  1088-1095.  doi: 10.19562/j.chinasae.qcgc.2021.07.016 Abstract ( 272 )   HTML ( 15 )   PDF (1979KB) ( 432 )   Save In order to further reduce the mass of vehicle control arm， a multi?material topology optimization method for shell/filler structure is proposed. Two groups of design variables are adopted to establish the material interpolation model， in which a two?step filtering is conducted on the first group of design variables and the normalized density gradient norms are used to separate shell region and filler region， while the second group of design variables are used for the assignment of multi?material in filler region. To avoid over aggregation of infill material， the local volume constraint is extended to the issue of multi?material filling by implicit mapping function. The results of calculation example of control arm optimization indicate that the material interpolation model can effectively identify the feature of shell and achieve local volume constraints， the mass of the main part of control arm is 13.2% less than the original steel counterpart， and the stress levels in steering and braking conditions are reduced by 55.0% and 39.7%， respectively， getting a better result of lightweighting. Figures and Tables | References | Related Articles | Metrics

Select

Motion Planning for Active Collision Avoidance of Intelligent Vehicles Based on Predictive Risk Field Anjie Wang,Ling Zheng,Yinong Li,Kan Wang 2021, 43 (7):  1096-1104.  doi: 10.19562/j.chinasae.qcgc.2021.07.017 Abstract ( 230 )   HTML ( 14 )   PDF (3221KB) ( 313 )   Save Aiming at the side and rear collision avoidance problem for autonomous vehicle， a predictive risk field fusing obstacle motion prediction and a motion planning method based on predictive risk field are proposed in this paper. In the Frenet coordinate system， the kinematics model is used to predict the information of each obstacle vehicle in the future scene， and the predictive risk field is established based on the three dimensions of longitudinal， lateral and time. Considering vehicle dynamics and velocity， acceleration and curvature constraints， the dynamic programming method is adopted to complete the behavior decision， and the polynomial curve and quadratic programming method are used to optimize the decision trajectory. The results show that the predictive risk field can accurately identify the changing trend of the potential risks of the surrounding obstacle vehicles over time， and plan the collision avoidance trajectory meeting various constraints， ensuring the safety and stability of vehicle operation. Figures and Tables | References | Related Articles | Metrics

Select

Research on Safety Strategy for Collision Avoidance by Automatic Emergency Braking on a Road with Varying Adhesion Coefficient Fengchong Lan,Yingjie Liu,Jiqin Lan Qingsheng Chen 2021, 43 (7):  1105-1112.  doi: 10.19562/j.chinasae.qcgc.2021.07.018 Abstract ( 175 )   HTML ( 7 )   PDF (1553KB) ( 326 )   Save In view of that the emergency braking system is prone to making an inaccurate decision on braking timing when road conditions change， an emergency braking strategy is proposed based on enhanced safety model with dynamic decision?making on vehicle kinematics. Firstly， a dynamic decision?making safety model is established based on vehicles’ speed and acceleration under the subdivided working conditions of the acceleration and deceleration states of target vehicle， so as to enhance the adaptability of the control strategy to the dynamic driving speed of the vehicle in extreme working conditions. Then， the road adhesion coefficient is obtained by continuous identification using unscented Kalman filter （UKF） algorithm， the relationship between road conditions and vehicle deceleration capability is established through braking performance tests on both real vehicle and model under a series of road conditions， and the model?dependent extreme deceleration parameters are updated real time according to road conditions， which further enhances the safety of control strategy and its adaptability to dynamic road conditions. Finally， the control strategy is verified through both the test on the road with continuously varying adhesion coefficient and C?NCAP test， and the results show that the UKF algorithm can get accurate identification results and the automatic emergency braking strategy can make the accurate decision on braking timing on the road with varying adhesion coefficient. Figures and Tables | References | Related Articles | Metrics

Select

New Type of DKT Combined Thin Shell Element and the Improvement of the One⁃step Collision Algorithm for Rollover Tong Wang,Yisong Chen,Yongtao Liu 2021, 43 (7):  1113-1120.  doi: 10.19562/j.chinasae.qcgc.2021.07.019 Abstract ( 121 )   HTML ( 2 )   PDF (2404KB) ( 133 )   Save Aiming at the warping problem of some elements in the finite element analysis of bus rollover， the four?node thin shell element is improved with warp correction first. Based on the improved four?node thin shell element and the total strain theory， a new type of four?node DKT combined thin shell element is constructed by combining the DKT four?node thin plate element with the traditional four?node isoparametric membrane element. Then the four?node Mindlin shell element used in the one?step collision initial algorithm for bus rollover is replaced by the new DKT element， with the initial algorithm improved. Finally， the improved algorithm is applied to the rollover simulation of the typical body section model for a 12m highway bus and its result is compared with those of initial algorithm， LS?DYNA and real bus rollover test， verifying the effectiveness of the constructed new type of four?node DKT combined thin shell element in practical engineering application. Figures and Tables | References | Related Articles | Metrics