Font Size: a A A

Non-Motorized Vehicles Riding Safety Analysis And Decision Model Construction Under The Interference Of Bus Entry And Exit Process

Posted on:2024-08-01Degree:MasterType:Thesis
Country:ChinaCandidate:Y LiuFull Text:PDF
GTID:2542307157466664Subject:Traffic and Transportation Engineering
Abstract/Summary:PDF Full Text Request
The bus entrying and exiting the linear bus station has a serious impact on the riding safety of passing non-motorized vehicles.This article addresses the current lack of research on the evasive behavior and decision-making of non-motorized vehicles under the interference of bus entry and exit processes.Based on the analysis of the movement characteristics of non-motorized vehicle evasion behavior,the study focuses on the analysis of non-motorized bicycle safety under the interference of bus entry and exit and decision-making modeling.This research is supported by the National Natural Science Foundation of China’s Youth Science Fund project study on guidance mechanism and combinatorial optimization research of pedestrian crossing induction facilities without signal control in cities(51908060).The main body of this paper can be summarized into four parts:the study of the movement characteristics of non-motor vehicle evasion behaviors,the modeling of non-motor vehicle traffic conflicts,the analysis of factors influencing non-motorized bicycle safety,and the study of decision-making modeling for non-motorized bicycle riding.The research contents are as follows:(1)The analysis method for non-motorized vehicle operation characteristics under the effect of bus entering and leaving the station is proposed.Videos of non-motor vehicle avoidance behavior during the process of bus entering and leaving the station are obtained via the drone aerial photography technology.Based on the video analysis platform George 2.1,trajectory data of the research object are extracted,and the obtained data is statistically analyzed as the basis for subsequent research.According to the path selection and motion state changes of non-motorized vehicles in avoiding collision risks,riding behaviors are divided into three types:rushing,detouring,and giving.From two aspects of spatial distribution and motion characteristics,the decision-making and conflict characteristics of non-motor vehicles under different riding behaviors are analyzed.(2)The traffic conflict model based on the conflict severity measurement indicators is proposed forward.Based on the conflict characteristics of non-motorized vehicles under the interference of buses,the post-encroachment time(PET),deceleration to safety time(DST),and bus conflict speed(V_c)are selected as indicators for measuring the severity of conflicts.Based on the results of the indicator calculation,the avoidance behavior characteristics of non-motor vehicles under different riding behaviors are analyzed.The K-means clustering algorithm is used to reasonably cluster conflict indicators,and the conflict severity level is divided according to the clustering results,interpreting the distribution level of conflict severity under different avoidance riding behaviors.(3)The riding safety impact factor analysis model based on ordered multi-class logistic regression is proposed.Based on the analysis of non-motorized vehicle operation characteristics and behavior,individual characteristics of traffic participants,decision-making status characteristics,traffic operation characteristics and other factors are selected as independent variables,and the severity level of conflicts is taken as the dependent variable.A riding safety impact factor analysis model based on ordered multinomial logistic regression is constructed.The significance and goodness-of-fit analysis methods are used to test the validity of the model and quantitatively analyze the influence mechanism of significant factors on conflict severity.(4)The riding risk avoidance decision-making model based on non-cooperative dynamic games is proposed forward.Select conflict risk and expected delay factors to formulate riding decision utility functions,consider the strategy selection process of riders,construct riding strategy sets and utility matrices,respectively establish non-cooperative dynamic game models when rider make decision firstly and rider make decision secondly,solve the equilibrium solution of the model,reveal the dynamic decision-making rules of non-motorized vehicles under disturbance,and provide reference for non-motorized vehicle riders’behavior decisions.
Keywords/Search Tags:riding safety of non-motorized vehicle, k-means clustering algorithm, non-cooperative dynamic game, linear bus station
PDF Full Text Request
Related items