交通信息与安全

Abstract:
In order to improve the accuracy of node localization for intelligent vehicles,an intelligent vehicles localization method based on three-dimensional point clouds semantic map representation is proposed. The method is divided into three parts. Semantic segmentation based on 3D laser point clouds includes ground segmentation,traffic signs segmentation and pole-shaped target segmentation. Semantic map representation for intelligent vehicles uses segmented targets to project. Finally directional projections with weight,semantic roads and semantic codeing are generated. The codeing and global location from high-precision GPS make up representation model. Localization based on semantic representation model includes coarse localization from GPS matching and node localization from semantic coding matching. The experiments are carried out in three road scenes with different length and complexity,and the localization accuracy is 98.5%,97.6% and 97.8%,respectively. The results show that proposed method has high accuracy and strong robustness, which is suitable for different road scenes.

Companion Relationship Discovering Algorithm for Passengers in the Cruise Based on UWB Positioning

YAN Sixun, WU Bing, SHANG Lei, LYU Jieyin, WANG Yang

Abstract(6189) PDF(2430)

Abstract:
To accurately discover the companion relationship among passengers in the interior space of a cruise, UWB positioning is employed in the cruise to carry out on-board personnel location experiment. An improved Haussdorff-DBSCAN based scheme combined with indoor positional semantics is proposed to realize the trajectory clustering of the passenger trajectories, considering the characteristics of the UWB location data. Afterwards, the LSTM neural network is applied to predict the changing similarity of the suspected companions. Traditional Hausdorff algorithm does not consider the consistency of trajectory timing while calculating the trajectory similarity, and the introduction of positional semantic sequence can solve this problem well. In the first phase, the passenger trajectory data set is input to the improved Hausdorff-DBSCAN algorithm, and the clustering radius is determined according to the overall similarity threshold of trajectories. The outputs are the emerging clusters of passenger trajectories in the same companion group. In the second phase, the LSTM neural network takes the point similarity sequence with a fixed time window as the input to predict the point similarity value at the next time. The sequential change of passengers companion relationship is analyzed by the similarity threshold and prediction results. The validity of the presented algorithm is demonstrated by the trajectory data obtained from the passengers simulation on one deck of the cruise under study, which is modeled in Anylogic. The results indicate that the precision of the proposed algorithm reaches 0.92, the recall value reaches 0.95 and the F1 value is 0.934, which are at least 5.7 percent, 8.0 percent and 6.7 percent higher than the comparing algorithm. The LSTM neural network also shows a promising effect in predicting the changing trend of the similarity, for the loss is at a stable level of 3 to 4 percent.

Data Association Method Based on Descriptor Assisted Optical flow Tracking Matching

XIA Huajia, ZHANG Hongping, CHEN Dezhong, LI Tuan

Abstract(3216) PDF(793)

Abstract:
in the view of the problem that the positioning accuracy of visual inertial odometer using multi-state constrained Kalman filter(MSCKF) is easily affected by the abnormal value of feature point matching, a data association method based on descriptor assisted optical flow tracking matching is proposed. This method uses pyramid LK optical flow to track and match the feature points in the sequence image, then calculates the rbrief descriptor of each pair of matching points, judges the similarity of the descriptor according to the Hamming distance,and eliminates the abnormal matching points. In the experiment, the effectiveness of the proposed method is evaluated from two aspects:the subjective effect of feature point matching and positioning accuracy. The results show that the proposed method can effectively filter the abnormal values of image feature matching in dynamic scene. The image processed by this method is used for msckf motion solution,and the drift rate of position result is less than 0.38%, compared with the result of msckf algorithm without eliminating abnormal matching values,The improvement is 54.7%, and the single frame image processing time is about 39 ms.

Indoor Sign-based Visual Localization Method

HUANG Gang, CAI Hao, DENG Chao, HE Zhi, XU Ningbo

Abstract(7614) PDF(1041)

Abstract:
To solve the problem of localization calculation of intelligent vehicles and the mobile robot in the indoor traffic environment, by exploiting kinds of signs which existed in the indoor environment, a visual localization method is proposed through using BEBLID (Boosted Efficient Binary Local Image Descriptor) algorithm. The proposed method enforces the ability to characterize the whole image by improving the classic BEBLID. In this paper, the localization method consists of an offline stage and an online stage. In the offline stage, a scene sign map is created. In the online stage, the calculation progress is divided into 3 parts, which include holistic and local BEBLID method from current image and image in the scene sign map, closet sign site and closet image calculation by using KNN method, metric calculation by using coordinate information which is stored in the scene sign map. The experiment is conducted in three kinds of indoor scenes, including a teaching building, an office building, and an indoor parking lot. The experiment shows the scene sign recognition rate reached 90%, and the average localization error is less than 1 meter. Compared with the traditional method, the proposed method improves about 10% relative recognition rate with the same test set, which verified the effectiveness of the proposed method.

A Cooperative Map Matching Algorithm Applied in Intelligent and Connected Vehicle Positioning

CHEN Wei, DU Luyao, KONG Haiyang, FU Shuaizhi, ZHENG Hongjiang

Abstract(7687) PDF(769)

Abstract:
In order to achieve low-cost and high-precision vehicle positioning in the intelligent and connected environment,a cooperative map matching algorithm based on adaptive genetic Rao-Blackwellized particle filter is studied in this paper,improving the accuracy of vehicle positioning by using the real-time location data and road constraints of other connected vehicles. The adaptive genetic algorithm is introduced into the re-sampling process of the particle filter to increase the diversity of particles,so as to solve the problems of "particle degradation" and "particle exhaustion" that are prone to appear in traditional particle filter algorithms. Model of the algorithm is established and simulated. The positioning results under the traditional particle filter and Kalman smooth particle filter are compared,and the influence of the number of different connected vehicles on the positioning accuracy is analyzed. The experiment is completed in real-world and the performance of the algorithm is verified. The experimental results show that taking a typical intersection with four connected vehicles as an example,the range of position error of cooperative map matching is 1.67 m. It is only 41.03% and 56.80% of the traditional GNSS and the single map matching positioning results. At the same time,the circular error probable(CEP) of this algorithm is 1.06 m, which is 2.52 m higher than raw GNSS positioning result.

MORE>

Select All

Display Method:

2023, 41(6): .

Abstract(0) HTML(0) PDF(0)

Abstract:

An Analysis and Prospects of Hot Topics on Maritime Autonomous Surface Ship Safety Research

ZHANG Di, LI Zhihong, WAN Chengpeng

2023, 41(6): 1-11. doi: 10.3963/j.jssn.1674-4861.2023.06.001

Abstract(0) HTML(0) PDF(0)

Abstract:
In recent years, the maturation of technologies such as autonomous navigation, sensors, communication, and networking has spurred rapid advancements in Maritime Autonomous Surface Ship (MASS) research. In September 2023, the 33rd European Safety and Reliability Conference was successfully held in Southampton, UK. The conference them centered on building a safe future in an interconnected world, with a particular emphasis on the safety of MASS. Based on a comprehensive analysis of 514 conference papers (including 19 papers related to intelligent ship safety topics), combined with the previous two conference proceedings and relevant research from the past decade both domestically and internationally, four hot topics in the field of MASS safety research are summarized: ①Autonomy level and related regulations: As the autonomy of MASS increases, the current legal frameworks need updating to accommodate new technologies, with research focusing on defining the autonomy levels of MASS and exploring the corresponding legal and regulatory frameworks. ② Human factors in remote operations: Remote operation of MASS introduces new challenges related to human factors. Research is oriented towards designing remote operation systems to reduce the psychological burden on operators, enhance communication efficiency, and provide effective decision support to ensure safety. ③ Risk assessment of MASS: This field aims to use advanced technologies for more accurate safety and risk evaluations, incorporating the use of multi-dimensional sensor data, real-time monitoring, and diversified data analysis models. ④ Applications of artificial intelligence and machine learning in MASS: Both technologies are regarded as innovative directions in the field of MASS safety, with research primarily focusing on their application in fault prediction, route optimization, and automated safety monitoring. Through a survey of existing literature, future research directions for MASS safety are prospectively discussed from four critical perspectives. ① By adopting Model-based Systems Engineering approach for ship safety analysis, potential safety threats can be identified and eliminated from the design phase, promoting interdisciplinary collaboration, and enhancing the accuracy of safety and reliability analysis. ② In terms of human factor risk analysis, the Functional Resonance Analysis Method is considered more suitable for complex systems like MASS. By evaluating the interactions between system functions, failures can be identified, and preventive measures can be formulated. ③ To improve efficiency in emergency situations, research needs to develop support systems that assist operators in making rapid and accurate decisions, considering the psychological and physiological states of operators. ④ The application of artificial intelligence and machine learning to deepen theoretical research involves developing autonomous decision-making models capable of making accurate decisions in complex maritime environments and advanced algorithms that integrate multiple data sources for accurate weather forecasting and route optimization.

A Suspension Stiffness Optimization for Driving System in High-speed Train with the Built-in Axle Box Based on Orthogonal Test

WANG Jiaxin, ZHANG Hanchen, WU Zhiqiang, LIU Yuqing, CHEN Zaigang

2023, 41(6): 12-19. doi: 10.3963/j.jssn.1674-4861.2023.06.002

Abstract(0) HTML(0) PDF(0)

Abstract:
As the core subsystem of the power bogie of high-speed trains, the drive system is an important guarantee for the safe operation of high-speed trains. However, with the continuous increase in operating speed, the reliable and safe operation of high-speed trains is seriously challenged. To reduce the dynamic loads on the suspension nodes of the axle box built-in high-speed dynamic vehicle driving system and to reduce the vibration level of the key components of the driving system, this paper carries out an optimization study on the suspension stiffness of the driving system. To reduce the dynamic loads and the vibration levels of components in the driving system, an optimization analysis of the suspension stiffness is performed in this study. Based on the multi-body system dynamics theory, the axle box built-in high-speed locomotive dynamics model is established by comprehensively considering the effects of track random uneven excitation, traction power transmission and gear meshing. Using the orthogonal test design method, with the optimization objective of reducing the suspension load at the traction motor lifting point and the vertical load at the axle articulation point of the gearbox, the influence of the suspension stiffness of the traction motor lifting point, the gearbox boom lifting point, and the motor-gearbox connection point on the vibration acceleration of the key components of the vehicle and the dynamic load at the suspension nodes of the driving system are investigated. The influence law is also analyzed by using the extreme difference analysis method to obtain the optimal matching combination of the suspension stiffness of the driving system. The results show that the maximum longitudinal, lateral, and vertical suspension loads of the motor with optimized parameters are reduced by 22.3%, 37.9%, and 9.8%, respectively. Meanwhile, the vertical load between the gearbox and wheel axle is reduced by 9.1%. The lateral vibration accelerations of the motor, gearbox, and axle box are significantly reduced.

An Investigation on Vehicle Trajectory Characteristics at Exit and Entrance of High-density Interchanges Based on Naturalistic Driving Data

XU Jin, YANG Xuemin, ZHANG Xueyu, ZHANG Jie, KONG Fanxing, JIAO Chengwu

2023, 41(6): 20-31. doi: 10.3963/j.jssn.1674-4861.2023.06.003

Abstract(0) HTML(0) PDF(0)

Abstract:
Interchange is an important node of the road traffic network, and as the spacing between neighboring interchanges continues to shrink, a high-density cluster of interchanges is gradually formed, which is prone to traffic congestion, increasing driving load and accident risk. To clarify the operational risks and safety hazards in the entrance and exit sections of high-density interchanges, the field driving test is conducted on the Inner Ring Road of Chongqing. This test takes a cluster of high-density interchanges as the research object. Using onboard instruments, vehicle trajectory data are collected under natural driving conditions. The vehicle trajectory data includes speed, real-time driving position, and lateral distance between the vehicle center and the lane markings on both sides. By analyzing the measured data, the vehicle trajectory pattern of the interchange entrances and exits as well as the relation-ship between the behavioral characteristics of lane selection and the influence of the driver's gender on the trajectory pattern are clarified. The lane-changing behavioral characteristics and driving risk influencing factors in the process of vehicles leaving (merging into) the mainline are explored. The results reveal the following conclusions: ①The type of entrance or exit has a significant influence on lane selection and trajectory shape. Compared to parallel-type exits, direct-type exits have smoother trajectories and fewer lane changing numbers. ②When entering and exiting two adjacent interchanges with a short clearance, drivers tend to choose the auxiliary lane or the outermost lane on the mainline to reduce the number of lane changing. ③The number of mainline lanes near the entrance and exit affect drivers' lane selection behavior. ④When drivers leave the mainline, the lane-changing duration of the parallel-type exit is higher than that of the direct-type exit. The entrance type does not have a significant impact on lane-changing time. The lane-changing time for 78% of drivers is between 5 to 10 seconds. ⑤The operational risks in the exit section are higher than in the entrance section. It is recommended to use solid white lane line to prohibit crossing same-direction lane markings on the left side of the rightmost lane of the exit section. The length should cover from the beginning of the taper section to the diverging point, extending 50 meters forward from the start of the taper section.

A Personalized Lane Change Decision Method Based on Driver's Psychological Risk Field Model

LI Ke, HAN Tongqun, YANG Zhengcai, GAO Fei, WU Haoran

2023, 41(6): 32-41. doi: 10.3963/j.jssn.1674-4861.2023.06.004

Abstract(0) HTML(0) PDF(0)

Abstract:
In driving environments, the motion behavior of interacting vehicles can stimulate the psychological and mental state of drivers, subsequently influencing their lane-changing decision behavior. In response to this, a personalized lane-changing decision method based on a driver's psychological risk field model is investigated. Focusing on a three-lane expressway traffic scenario, the vehicles' lateral speed and lateral offset are analyzed by interacting multiple models. Variable lateral speed-related transition probability matrices are introduced to predict the target lane selection of interacting vehicles. A model is established to quantify the impact of the driving environment and interacting vehicles' motion behavior on drivers' psychological risk. The experiment is conducted by establishing mixed traffic scenarios in a SUMO-based driving simulator, and 287 cases of lane-change datasets are collected. Two characteristic parameters, average collision time and driver psychological risk factor, are selected. The K-means algorithm is used for driver style clustering, categorizing drivers into conservative, normal, and aggressive styles. Furthermore, different thresholds for psychological risk at the initial moment of lane-changing are determined for drivers with different styles. Then personalized safe lane-changing decisions for vehicles are implemented. Experimental results show that, for conservative, normal, and aggressive drivers, the actual minimum lane-changing decision times are 3.48, 6.29, and 11.33 s, respectively. The actual maximum lane-changing decision times are 4.65, 7.45, and 12.52 s, respectively. The theoretical lane-changing decision times are 4.09, 6.83, and 11.95 s, respectively. The prediction errors of the personalized lane-changing decision model are all less than 0.62 seconds. This approach accurately assesses the psychological risk of drivers with different styles and achieves personalized lane-changing decisions.

Coupling Failure Mode and Risk Modeling of Typical Aircrafts Runway Excursion

WANG Feiyin, YUAN Jintong, WANG Lei

2023, 41(6): 42-50. doi: 10.3963/j.jssn.1674-4861.2023.06.005

Abstract(0) HTML(0) PDF(0)

Abstract:
Runway excursion is identified as high-risk event by the International Air Transport Association. To explore the pattern of runway excursion incidents in global civil aviation and to explore the influencing factors and their coupling characteristics, the investigation reports of 57 runway excursion incidents of typical aircraft types from 2007 to 2018 are analyzed from the perspectives of number of casualties, aircraft types and causes of the incidents. The HFACS model and SHELL model are used to compensate for the limitations of using a single method considering the diversity and complexity of influencing factors of runway excursion incidents. Specifically, the HFACS model is optimized and adopted to vertically analyze the influence of human factors in the runway excursion event, change the traditional method of the SHELL model to analyze the coupling influence of multiple factors in the runway excursion event systematically and comprehensively and use the FMEA method to explore the coupling effect of multiple influencing factors in the runway excursion event and find 18 multifactor coupling failure modes that induce the runway excursion. The results showed that the risk priority of the failure modes is quantified by identifying the occurrence, severity, and detection of the failure modes. The results showed that 91.2% of the runway excursion events occurred in the landing phase, and 87.7% of the runway excursion events were related to the crew human influence, among which insufficient control of the aircraft occurred most frequently, accounting for 31.1%. Multi-factor coupling caused 78.9% of the events, and the risk priority value of failure mode F2-1 crew factors and meteorological factors in multi-factor coupling failure mode is the highest at 364.8, with an occurrence rate of 21.05%, which is the object that needs to be focused on prevention and control, indicating that pilots need to strengthen the simulation training of runway excursion under complex weather conditions.

A Method of Ship Trajectory Prediction Based on a C-Informer Model

CHEN Lijia, ZHOU Naiqi, LI Shigang, LIU Kezhong, WANG Kai, ZHOU Yang

2023, 41(6): 51-60. doi: 10.3963/j.jssn.1674-4861.2023.06.006

Abstract(0) HTML(0) PDF(0)

Abstract:
The navigation of ships in complex environments is influenced by various uncertain factors, such as wind, waves, water depth, and ship performance, etc. It is challenging to precisely define and reflect the dynamic patterns of ship trajectories simply using mathematical models. To address this issue, a multi-step prediction method for ship trajectories based on feature engineering and neural networks is studied. The task of trajectory prediction is divided into two parts: data processing and model prediction. The data processing module preprocesses AIS trajecto-ry data using feature engineering methods. It starts by cleaning the raw AIS data, then uses the maximal information coefficient to select features highly correlated with the position prediction task. Additionally, a variable time interval information is introduced to address the problem of existing models only being able to select fixed time interval data for training and prediction. This module ultimately reconstructs high-quality ship trajectory sequences. The model prediction module constructs a ship trajectory prediction model based on C-Informer. It utilizes the multi-head Prob-Sparse self-attention mechanism of the Informer model to reduce the time complexity of the network model. Simul-taneously, it enhances prediction speed by generative decoding. By introducing a causal convolution module, the sensitivity of the model to neighboring time trajectory features is increased to compensate for the deficiencies of the Informer model in extracting local information. This adaption makes the model more suitable for ship trajectory prediction tasks. The experimental results based on Automatic Identification System (AIS) data near Nanjing port show that the C-Informer model for trajectory prediction has an overall mean square error (MSE) of 1.72×10－7 and a mean absolute error (MAE) of 2.43×10－4. Compared to the original Informer model, this represents a reduction of 28.6% and 31.9%, respectively. When training the C-Informer model with the selected feature combinations, the MSE and MAE are decreased by 57.7% and 42.1%, respectively, compared to using only latitude and longitude fea-tures. In predicting trajectories at different time steps, the C-Informer model reduces prediction time by up to 69.6% compared to the long short-term memory network model, with a maximum loss reduction of 75.8%.

Path Tracking and Lateral Stability Control for Distributed Drive Vehicles with Low Adhesion

YANG Wei, TAN Liang, DU Yafeng, SUN Xue, ZHANG Yujie

2023, 41(6): 61-70. doi: 10.3963/j.jssn.1674-4861.2023.06.007

Abstract(0) HTML(0) PDF(0)

Abstract:
Due to the coupling relationship between tracking and lateral stability of vehicles under low adhesion conditions (such as snow and moisture), it is difficult to control both tracking accuracy and good stability simultaneously. Therefore, a joint control model of path tracking and lateral stability is proposed based on distributed independent drive electric vehicle platform. The transverse and longitudinal decoupling control is adopted for the path tracking problem. Besides, the model predictive control (MPC) method based on Frenet coordinate system is adopted for the horizontal tracking control problem, and angle compensation strategy is introduced to improve the accuracy of path tracking. For the longitudinal speed control problem, the model uses MPC to solve the expected acceleration, and determines the motor torque output according to the driving balance equation and the maximum utilization rate of road adhesion, so as to achieve the longitudinal speed control. For lateral stability control, a yaw torque control model based on stability augmentation system (STA) is proposed. After additional torque is obtained, it is effectively distributed to each wheel by quadratic programming method, thus enhancing the lateral stability of the vehicle. Moreover, the CarSim/Simulink co-simulation platform is used to simulate and verify the double-shift road conditions. The results show that under the condition of snow-covered pavement, the maximum lateral deflection angle of the improved model is reduced by 83.1% compared with the traditional MPC under the condition that the lateral error is close. Under wet road conditions, the maximum lateral error and the maximum lateral deflection angle of the improved model are reduced by 52.2% and 83.3%, respectively, compared with the traditional MPC model. Compared with the traditional synovial model, this model can effectively suppress the oscillation phenomenon when the tracking error and the side deflection angle of the center of mass are dominant. Through the joint control, the stability and safety of the vehicle on the low adhesion road surface can be enhanced.