91 results about "Time headway" patented technology

A deep reinforcement learning-based vehicle platoon control method and system. The method comprises: on the basis of a vehicle dynamics model, constructing a motion simulation model for vehicle platoon operation; on the basis of the motion simulation model for vehicle platoon operation and vehicle operating states, by using a constant time headway strategy, obtaining a state space and an action space for vehicle platoon control; and establishing a DDPG network on the basis of the state space and the action space for vehicle platoon control, using a reward function mechanism to train the DDPG network, stopping training when the total reward obtained per episode during the training process no longer increases, so as to obtain a trained control strategy network, and by using the trained control strategy network, performing platoon control.

The invention discloses an automobile braking energy recovery method considering a driving style. The method comprises the steps that the driving style of a driver and the comfortable time headway of the corresponding style are determined; wherein the driving styles comprise aggressive styles, ordinary styles and prudent styles; when the real-time headway is larger than the safe headway, the current braking intention of the driver is recognized according to the current accelerator pedal position, the accelerator pedal change rate and the difference value of the real-time headway relative to the comfortable headway; the types of the braking intention comprise light degree, light-medium degree, medium strength and emergency; if the current braking intention of the driver is mild or mild, a regenerative braking strategy is adopted; if the current braking intention of the driver is moderate, a regenerative braking and mechanical braking mixed strategy is adopted; and if the current braking intention of the driver is emergency, a mechanical braking strategy is adopted. The driving intention recognition accuracy can be improved, so that the driving comfort and the braking energy recovery efficiency are improved.

The invention discloses a traffic signal adaptive control method based on a thunder-vision integrated vehicle detector, and relates to the technical field of intelligent traffic, and the method comprises the steps: obtaining the time headway of vehicles in an entrance road section, the number of vehicles entering a target intersection, and the total number of vehicles entering the entrance road section; detecting the number of detained vehicles in each traffic flow direction of the entrance road section, and determining the saturation degree of each traffic flow direction; according to the saturation degree, the effective green light duration of the corresponding phase of each passing flow direction of the target intersection is adjusted, and when it is detected that a congestion flow direction exists, the additionally prolonged effective green light duration is added to the corresponding phase of the congestion flow direction, so that an updated signal timing scheme is obtained; and controlling the traffic lights of the target intersection according to the updated signal timing scheme in the next control period. The signal timing scheme is adjusted in real time according to the change condition of the road traffic flow state, adaptive control of the traffic signal lamp is achieved, and the road congestion problem is effectively relieved.

The invention discloses a trunk line intelligent network connection bus speed and signal timing collaborative optimization method, which comprises the following steps: acquiring basic information of bus lines, such as departure time interval, line length, station distribution and intersection signal phase; an intelligent networked bus speed control model is constructed, and dynamic speed guidance is implemented for different deviation types by calculating the deviation of the time headways of adjacent vehicles in real time; based on a vehicle speed control result, intersection signal lamp timing is optimized, and the negative influence of traffic signals on the driving speed of the public transport vehicle is reduced. By implementing accurate speed and signal lamp cooperative control at each bus station and intersection, the method can improve the overall operation efficiency of the trunk bus system to the maximum extent and reduce intersection delay while ensuring that the buses arrive on time, and compared with a traditional whole-line adjustment strategy, the method has the advantages that the overall operation efficiency of the trunk bus system is improved to the maximum extent, and the cost is reduced. According to the control strategy, the problem that the time headway deviation of a local road section is large can be solved more accurately.

The invention discloses an entrance ramp fleet confluence method based on reinforcement learning, and the method comprises the steps: carrying out the scene modeling according to an actual scene, and designing a plurality of fleet scales and vehicle speeds; vehicle acceleration and deceleration are designed as action selection, and the running speeds of main road vehicles and ramp vehicles are changed, so that the ramp vehicles select proper confluence opportunities; the change of the total travel time reflects the motorcade afflux efficiency, a reward function is established according to the change of the vehicle speed, the change of the time headway and the queuing waiting time of a ramp vehicle on an acceleration lane, an action mask is designed to improve the training efficiency, an intelligent agent is trained by using a PPO algorithm, and a trained deep reinforcement model is taken as a vehicle afflux model. The method gives consideration to the better passing efficiency of the main line and the shorter queuing waiting time of the vehicles on the ramp, enables the passing of the main line to be smoother, reduces the waiting time of the vehicles on the ramp as much as possible, and reduces the interference on the main line caused by the convergence of the vehicles on the ramp.

The invention belongs to the technical field of intelligent traffic, and particularly relates to an expressway traffic state judgment method based on vehicle-level macro and micro evaluation. The method comprises the following steps: firstly, acquiring millimeter-wave radar data of a target road section in a target time period, and preprocessing the millimeter-wave radar data; then, according to millimeter wave radar data, evaluation indexes are calculated, and the evaluation indexes comprise a microcosmic index and a macroscopic index; the correlation between the indexes is checked, part of indexes with high correlation are removed, and the obtained microcosmic indexes comprise the vehicle drift angle variance, the lane changing behavior frequency, the time headway variance, the acceleration variance and the aggressive driving behavior variation index variance; the macroscopic indexes comprise average flow, a speed variation coefficient, a cart mixing rate and a congestion index; and finally, based on the macro and micro evaluation indexes, calculating the traffic state entropy of each road section in the target road section in the target time period, and classifying the traffic state according to the traffic state entropy. According to the method, the aggressive driving behavior variation index is provided for measuring the vehicle danger degree caused by sudden acceleration, deceleration or too high driving speed, so that the traffic state is reflected more finely from the microcosmic level, and meanwhile, the traffic state is accurately quantified by considering the mutual influence among the road sections.

The invention discloses a trunk bus dynamic speed optimization method based on time headway equalization, and a dynamic optimization model is constructed by taking the minimum total deviation of the time headway on a bus line as a target. According to the method, factors such as departure time, driving speed and signalized intersection delay of intelligent networked buses are comprehensively considered, and the departure time of the buses at stations and the driving speed of the buses when the signalized intersection delay is not considered are optimized by establishing a calculation method for the bus delay at the signalized intersection. Meanwhile, according to a time headway balance principle, a calculation method based on the waiting time of passengers is provided so as to coordinate the operation efficiency of the motorcade and the experience of the passengers. According to the method, the fluctuation of the time headway is effectively reduced, the bunching effect is avoided, the overall operation efficiency and the service level of the bus route are remarkably improved, and theoretical support and technical guarantee are provided for design and implementation of an intelligent network connection bus system.

The invention discloses an automatic driving vehicle formation control method and system based on working condition identification, and relates to the technical field of automatic driving. Deploying a sensor; the leading vehicle is used for sensing the surrounding environment; judging the driving condition of the current motorcade; obtaining a relative position relationship between adjacent vehicles in the motorcade; setting a time headway to obtain an expected distance between adjacent vehicles; arbitrating the expected distance; calculating a distance error between adjacent vehicles; whether the speed of the rear vehicle needs to be adjusted is judged, and if yes, the speed of the rear vehicle is adjusted; and if not, continuously driving according to the current vehicle speed. Complete automatic driving modification is carried out on a small part of vehicles in a motorcade, the current environment is perceived through the leading vehicle, the perceived information is issued to all following vehicles in the motorcade in a broadcast mode, the following vehicles in the motorcade are assisted to carry out planning control, and a motorcade control task can be completed.

The invention provides an automatic driving vehicle collaborative lane changing strategy considering dynamic risk information, and the method is mainly divided into three parts: firstly, constructing a queue collaborative lane changing control architecture, taking a queue collaborative lane changing decision making method considering the dynamic risk information as a decision making layer, and taking queue collaborative lane changing control based on a four-element method as a control layer; and the two are cooperated to complete decision making and control of queue vehicles and lane changing vehicles. Then, by analyzing the motion states of the lane changing vehicles and the queue vehicles, the relative positions and the relative speeds serve as main influence factors, and an optimal cut-in position determining method and lane changing allowing conditions of the lane changing vehicles are given; a risk field with road traffic participants as a field source is designed based on the field theory, the risk field is used as a decision factor of a transverse lane changing track of a lane changing vehicle, and a transverse lane changing track dynamic planning model is constructed. Finally, on the basis of the basic structure of the four-element model, the linear two-degree-of-freedom dynamic model serves as transverse dynamics of vehicle nodes, the characteristics of the queue vehicles compared with single vehicles are considered, and the mass-spring-damper model serves as longitudinal dynamics of the vehicle nodes; the queue expected configuration selects a constant time headway strategy; taking a PLF structure as an information flow topological structure of a queue; and respectively designing a transverse controller based on MPC and a longitudinal controller based on linear control.

The invention discloses an intelligent signal transmission system based on intelligent traffic, and the system comprises a data collection module which collects traffic flow parameters of each lane, and the traffic flow parameters comprise vehicle speed distribution, lane occupancy, time headway and vehicle position; the state coefficient acquisition module is used for constructing a # imgabs0 # model through the traffic flow parameters, introducing a dynamic congestion propagation factor into the # imgabs1 # model to obtain an improved # imgabs2 # model, inputting the traffic flow parameters into the improved # imgabs3 # model, and acquiring a state coefficient of driving out of each lane; the priority acquisition module is used for extracting a priority sequence based on the state coefficient, adjusting signal timing of different intersections according to the priority sequence and dynamically adjusting the signal timing based on the state coefficient and the priority sequence, can flexibly allocate green light duration according to the real-time traffic state of each lane, reduces the phenomena of green light idle running and vehicle backlog, and improves the traffic safety. The overall traffic efficiency of the intersection is improved.

The invention discloses a CAV dedicated lane deployment strategy double-layer optimization design method, and relates to the technical field of traffic network computers in an intelligent network connection environment, and the method comprises the steps: constructing an intelligent network connection expansion network; constructing a road section impedance function considering a CAV proportion and a time headway parameter, and constructing a travel cost function of CAV users and HV users; constructing a double-layer optimization model of the traffic bearing capacity of the intelligent network connection expansion network; forming a double-layer optimization model of a CAV special lane deployment strategy; converting the nonlinear user equalization UE traffic distribution model into a linearized user equalization UE traffic distribution model; performing linearization processing on the nonlinear road section impedance function to obtain a CAV special lane deployment strategy double-layer optimization model in a mixed integer linear programming form; and solving the CAV special lane deployment strategy double-layer optimization model in the mixed integer linear programming form to obtain a design result of the urban road network CAV special lane deployment strategy. According to the method, the optimal deployment scheme can be found.

The invention discloses a data-driven network traffic modeling and simulation method based on dynamic modulus decomposition. The method comprises the following steps: collecting traffic data through sensors arranged in urban roads; filling missing values in the traffic data by adopting a data reconstruction method based on equilibrium state hypothesis; reconstructing and predicting the nonlinear dynamics of the space average speed in the traffic data based on the Koopman operator theory; calculating an expected time headway corresponding to a space average speed predicted by dynamic modulus decomposition KMD based on time headway distribution estimated offline, and calculating an outflow rate, namely traffic capacity, according to the expected time headway; and predicting the traffic density change based on a cellular transmission model and the traffic capacity calculated in the previous step. According to the method, the time headway (which not only represents the dynamic characteristics of traffic capacity in a congestion state, but also is a core parameter of a CTM model) is introduced, and a dynamic outflow flow rate estimation method based on speed constraint is constructed, so that accurate modeling of a flow density nonlinear relation is realized.

The invention discloses a real-time traffic jam prediction method and system based on multi-integration deep learning, relates to the technical field of traffic safety, and aims to solve the problem of low accuracy of existing traffic jam prediction, comprehensively considers a vehicle speed variance, an interval variance and a time interval variance, and avoids the problem that a clustering result is easily influenced by original data. And the prediction accuracy of the traffic jam is improved. The network model combines multi-head attention, online learning and an LSTM network, can effectively predict the traffic flow in real time by capturing short-term and long-term dependency relationships, dynamically adjusts new data through online learning, and improves the accuracy of congestion prediction by using an integration method, thereby avoiding errors in specific driving path planning, and improving the traffic congestion prediction accuracy. And the accuracy of driving path planning is not influenced.

The invention discloses an automatic driving scene risk assessment method fusing a diffusion model and a random forest, and belongs to the technical field of automatic driving control. The problems of early recognition and early warning of potential risks of automatic driving are solved. A diffusion model is improved on the basis of a traffic scene state, a condition variable is introduced to obtain a condition generation distribution model, and a traffic scene sample data set is generated; setting the minimum collision time and the minimum time headway as key risk indexes, and generating a risk label for each generated traffic scene sample; constructing a continuous risk quantitative index meeting a functional safety standard as exposure degree and severity, and constructing a training data set; a random forest model is adopted to construct a risk assessment model, the training data set is utilized to train the risk assessment model, the training target is to learn the mapping relation between the exposure degree and severity and the risk label, and the trained risk assessment model is obtained; and evaluating the performance of the risk evaluation model by adopting a multi-dimensional evaluation index system to realize automatic driving scene risk evaluation.

The invention discloses an intelligent network connection bus operation optimization control method based on a heterogeneous line scene, and relates to an intelligent network connection bus operation optimization control method. The invention aims to solve the technical problems of non-uniformity of bus time intervals and congestion of stations and intersections of intelligent networked buses running on bus lanes due to coexistence of a manual driving bus route and an intelligent networked bus route at present. According to the intelligent network connection bus operation hierarchical optimization control model established by the invention, the optimal speed control scheme and the signal priority scheme are output under the condition of considering the uncertainty of manual driving bus operation, the uniformity of the time headway is improved, and the number of times of entering a node for queuing is reduced. The method provides a new thought for efficient operation of an intelligent network connection bus company, and is of great significance in promoting large-scale operation of the intelligent network connection bus from a specific simple road to a complex road.

The application belongs to the technical field of intelligent transportation, and particularly relates to a highway traffic state discrimination method based on vehicle-level macro and micro evaluation. First, millimeter wave radar data of a target section in a target period is acquired, and the millimeter wave radar data is preprocessed. Then, evaluation indexes are calculated according to the millimeter wave radar data, including two types of indexes, i.e., micro and macro indexes. The correlation between the indexes is tested, and some indexes with strong correlation are removed, so as to obtain micro indexes including vehicle yaw angle variance, lane changing behavior frequency, vehicle head time interval variance, acceleration variance and aggressive driving behavior variation index variance. Macro indexes include average flow, speed variation coefficient, large vehicle mixing rate and congestion index. Finally, based on the macro and micro evaluation indexes, traffic state entropy values of each section in the target period of the target section are calculated, and the traffic state is classified according to the traffic state entropy values. The method proposes the aggressive driving behavior variation index for measuring the risk degree of vehicles caused by sudden acceleration, deceleration or too fast driving speed, so as to more carefully reflect the traffic state from the micro aspect, and simultaneously consider the mutual influence between the sections to accurately quantify the traffic state.

The application discloses a kind of control methods of multiple vehicles into CAV special lane under network environment, comprising:1, calculate influence interval and judge whether the headway of adjacent target CAV car of second lane is less than influence interval;2, obtain the speed of adjacent target CAV car at initial time, calculate the speed of adjacent target CAV car at lane-changing stable time;3, calculate the speed index of CAV car into special lane;4, calculate the headway of adjacent target CAV car and its preceding vehicle at initial time and lane-changing stable time respectively;5, calculate the safety index of CAV car into special lane;6, calculate the comprehensive index of CAV car into special lane;7, take the maximum comprehensive index strategy, output lane-changing decision.The application can provide optimal merging decision through information interaction between network vehicles and network vehicles under network environment when multiple adjacent CAV cars simultaneously issue merging demand into special lane, thereby improving the efficiency of CAV car merging into special lane and improving road traffic capacity.

The application discloses a mountain road automatic driving car rear vehicle safety detection method and system, and belongs to the technical field of vehicle active safety detection. The application collects the position, speed and acceleration of a driver following an automatic driving car at each moment in a no-lane-changing overtaking scene, and then extracts eight features including maximum speed, minimum speed, speed average, speed standard deviation, acceleration, deceleration, distance and headway. The driver is divided into three types of aggressive, conservative and intermediate by using principal component analysis and K-means clustering algorithm. A following safety threshold model is constructed for each type of driving style. The following data of the rear vehicle is collected in real time, the features are extracted and the driving style is matched, the safety threshold of the corresponding style is called to complete safety detection, and a warning is given if the threshold is exceeded. The application can effectively avoid the collision risk of the automatic driving car rear vehicle at the mountain curve, and provides technical support for the driving safety of the mountain road automatic driving scene.

The present invention relates to the field of urban road traffic technology, and discloses a generalized estimation method and system for saturated headway at urban road signal intersections, comprising obtaining license plate recognition data and extracting an intersection signal cycle; matching the license plate recognition data with floating vehicle data, extracting saturated headway samples of vehicles queuing at the intersection under all intersection signal cycles; screening out factors influencing the saturated headway; using the saturated headway samples as input, and constructing a probability estimation model of the saturated headway and various saturated headway influencing factors based on the NGBoost model according to the saturated headway influencing factors; using each saturated headway influencing factor as input to the probability estimation model, and outputting a generalized estimation result of the saturated headway. The beneficial effect of the present invention is that it can generalize the saturated headway based on existing and easily obtainable license plate recognition data and floating vehicle data.

The present invention discloses a method for analyzing the reliability of path travel time considering the headway time characteristics, and the steps include: Step S01. Obtain the moments when vehicles in each lane of the target section pass through the upstream and downstream detection points to obtain the passing time data of each vehicle; Step S02. Calculate the headway time of each vehicle passing through the upstream and downstream detection points in the section according to the passing time data of each vehicle; Step S03. According to the headway time of each vehicle passing through the upstream and downstream detection points in the section, use the pre-constructed path travel time budget model to calculate the path travel time budget value. In the path travel time budget model, the mean and variance of the path travel time are calculated respectively using the headway time of each vehicle passing through the upstream and downstream detection points in the section; Step S04. Obtain the reliability condition of the path travel time according to the path travel time budget value. The present invention has the advantages of simple implementation method, small calculation amount, high analysis efficiency and high accuracy.

The invention discloses a mixed traffic flow regulation and control method based on communication delay adaptation, and aims to overcome the defect that a traffic flow control method aiming at communication delay adaptation is lacked in the prior art. The method comprises the following steps: introducing communication delay to a cooperative adaptive cruise control (CACC) system, and constructing a mixed traffic flow model in combination with a vehicle type and a communication mode; constructing a critical equation, dividing a control compensation dominant area and a communication delay dominant area according to the mixed traffic flow model, and determining critical communication delay; and obtaining a safe time headway function according to the mixed traffic flow model, defining a ratio of a control compensation item to a communication delay item based on the function, performing safe time headway judgment according to the ratio, and realizing communication delay adaptive mixed traffic flow regulation and control. The blank of current research is filled, and a more refined optimization scheme is provided for actual traffic management.

The invention discloses a double-layer cooperative control method for improving the traffic capacity of an uphill road section in a network connection environment, which is suitable for a road basic road section containing a ramp bottleneck, and comprises the following steps: 1, obtaining a current traffic flow state, calculating an equivalent time headway and mixed traffic flow basic diagram parameters, and predicting a traffic state of a next time period; 2, calculating the entrance sending capacity of the upslope bottleneck road section and the supply capacity and falling capacity of the downstream straight road section, and determining the upper flow limit of the upslope bottleneck road section; and 3, when the regional transfer total demand flow exceeds the regional transfer total supply and the regional transfer total supply flow is greater than the falling capacity, judging that the capacity falling is about to occur, activating the double-layer cooperative control framework, and performing rolling optimization to obtain a variable speed limit value and a special road layout scheme, and issuing and implementing the variable speed limit value and the special road layout scheme. According to the method, modeling and management and control requirements of a ramp capacity falling mechanism are considered, generation of capacity falling can be inhibited, the bottleneck queuing length of the ramp is reduced, and the traffic capacity and the traffic efficiency of the basic road section of the road are improved.

The invention discloses a traffic flow parameter real-time sensing method based on air-ground cooperative sensing, which can provide an accurate geometric alignment basis for fusion of air-based hovering video data and ground-based radar data by analyzing the tracking life cycle of a vehicle target to adaptively divide sensing areas and optimize calibration parameters. On the basis, through cross-modal feature matching and data association, the wide coverage advantage of an air-based video and the accurate distance measurement advantage of a ground-based radar are effectively converged, so that large-range vehicle accurate detection and tracking are realized, and the detection accuracy is improved. Furthermore, various key traffic flow parameters including queuing length, section traffic flow, interval vehicle number, vehicle existence and time headway are synchronously calculated and output.

An attention calling system includes a time headway (THW) acquisition unit that repeatedly acquires THW at predetermined time intervals, the THW being a value obtained by dividing an inter-vehicle distance between own vehicle and a preceding vehicle by vehicle speed of the own vehicle, an identification unit that identifies a type of the preceding vehicle, and a notification unit that outputs to an occupant of the own vehicle an approach alarm that notifies an approach to the preceding vehicle when the THW becomes a first threshold or less. The notification unit outputs the approach alarm by using the first threshold that takes different values depending on the type of the preceding vehicle, and / or by setting a saliency level that is a degree of saliency of the approach alarm to a first level that has different magnitudes depending on the type of the preceding vehicle.

The invention discloses a tunnel traffic congestion state recognition method based on deep learning, and the method comprises the steps: recognizing a vehicle in a traffic monitoring video, and obtaining the coordinate information of the vehicle in an image; according to the known lane line position and the vehicle coordinate information, determining the lane number of each vehicle; for each lane, a traffic flow statistical area is preset for each lane for traffic flow statistics, and the average traffic flow in unit time is calculated; counting the time for passing vehicles and front vehicles in each lane to arrive at the traffic flow statistical area, and calculating the average time headway in unit time; counting the time required for passing vehicles on each lane to pass through the traffic flow statistical area of each lane, and calculating the average lane passing speed in unit time; weighting coefficients are assigned to the average traffic flow, the average time headway and the average lane passing speed respectively, a tunnel traffic jam index is obtained through weighted calculation, and the tunnel traffic jam state is judged according to the index. Manual intervention is not needed, and automatic detection and recognition are achieved.

The application provides a traffic signal optimization method based on parameter calibration and improved Webster timing method, the method comprising: arranging detection coils in each lane in front of a stop line at an intersection, collecting vehicle passing information and storing locally; based on traffic flow theory analysis, calibrating saturated headway and calculating saturated flow of each lane according to the vehicle passing information, reducing the influence of oversaturated traffic flow on signal timing; for phase overlap and stage secondary release, the concept of phase residual flow ratio is proposed to solve the adaptation problem of Webster timing method under the signal timing requirements of the phase overlap and the stage secondary release; in the signal timing self-optimization link, if the signal timing meets the minimum green constraint relationship, the signal timing ends; if the signal timing does not meet the minimum green constraint relationship, the minimum green adjustment process is entered; when real-time optimization is performed, the updated signal timing is executed in the next cycle and the vehicle passing information is recorded again; when timing optimization is performed, adjustment is performed within a certain fixed time period based on historical data. The problems of unreasonable setting of basic parameters and limited use of traditional timing methods in actual application scenarios are solved, and a fast, accurate and stable traffic signal adaptive optimization system is established.

The invention belongs to the technical field of intelligent transportation, and particularly relates to an interoperation-based intelligent rail route time division multiplexing method, which comprises the following steps that: vehicle data are obtained through roadside equipment, and the vehicle data comprise position coordinates; processing the vehicle data based on a trajectory smoothing method to obtain a fitted vehicle trajectory; processing the plurality of traffic indexes based on a fuzzy evaluation algorithm, and evaluating the traffic state grades according to the weights of the plurality of traffic indexes and the membership degrees of the traffic state grades of the corresponding indexes; when the traffic is in the deadlock state level, the original state is kept for driving; and when the traffic is in a non-locking state, according to the protection time headway, the safety time headway, the controlled network connection vehicle governing range duration and the average driving speed of the vehicles with the priority right of way, the length of the vehicles changing lanes and occupying the intelligent rail and the time of occupying the intelligent rail are obtained.

The invention relates to the technical field of intelligent traffic management and planning, and discloses a lane design method and system based on mixed traffic flow lane traffic capacity, and the method comprises the steps: constructing a traffic feature basic data set of a target road scene; classifying the vehicles and the following states of the vehicles; calculating occurrence frequencies of various car-following states; calculating the average time headway of the mixed traffic flow lane; calculating the traffic capacity of the mixed traffic flow lane according to the average time headway of the mixed traffic flow lane, and deducing the traffic capacity of other lanes; obtaining a lane design scheme according to the traffic demand and the number of lanes in the target road scene; and based on the obtained lane design scheme, according to vehicle following and lane changing rules, constructing a simulation platform and a comprehensive evaluation method based on an entropy evaluation method, and determining an optimal lane design method. According to the invention, a reasonable layout scheme is provided for the design of high-truck-ratio lanes in an intelligent network connection environment, refined lane management is realized, internal traffic conflicts are reduced, and the operation efficiency of the expressway is improved.

The application provides a driver behavior analysis and prediction method based on causal inference, belongs to the field of traffic engineering human factor analysis, and mainly evaluates and understands the driver behavior by constructing four types of driving index evaluation systems, including the headway, the time to collision, the speed and brake amplitude, and the eye tracking information. Then, the causal inference model of double machine learning is adopted to deeply study the driving behavior of the driver in various scenes and the difference thereof, and accurate causal estimation of the driver behavior is provided, so that the driver behavior can be further predicted. The method comprehensively analyzes the relationship between the driving behavior and the driving environment, can be close to the actual situation, improves the driving safety and the driving efficiency, and has high practical value for the fields of intelligent driving and traffic safety research.

An operation optimization control method for an intelligent connected bus based on a heterogeneous line scenario relates to an operation optimization control method for an intelligent connected bus. The present invention aims to solve the technical problem that the coexistence of current manually driven bus lines and intelligent connected bus lines causes uneven headway time of intelligent connected buses traveling on bus lanes, as well as congestion at stations and intersections. The intelligent connected bus operation hierarchical optimization control model established by the present invention outputs the optimal speed control scheme and signal priority scheme under the condition of uncertainty in the operation of manually driven buses, improves the uniformity of headway time, and reduces the number of times of entering node queues. The present invention provides a new idea for the efficient operation of intelligent connected bus companies, and is of great significance to promoting the large-scale operation of intelligent connected buses from specific simple roads to complex roads.