392 results about "Carriageway" patented technology

The invention belongs to the technical field of expressway passenger and truck lanes, and particularly relates to an expressway passenger and truck lane dynamic allocation optimization method which comprises the following specific steps: S1, data acquisition and preprocessing: integrating ETC and radar multi-source data, cleaning and fusing original data, and accurately distinguishing small passenger cars and trucks by using an advanced vehicle type recognition technology, so as to obtain a new vehicle type recognition technology; a reliable data basis is provided for subsequent analysis; s2, traffic state perception and prediction: calculating key traffic parameters of traffic flow, speed and density based on the preprocessed data; and a scientific prediction model is called. According to the method, the future short-time passenger and freight predicted flow provided by traffic state perception and prediction is received, and the lane allocation decision is made based on the dynamic data, so that the number of the passenger car special lanes can be increased in time when the flow of the passenger cars is sharply increased in the morning peak, the delay of the passenger cars on the passenger and freight mixed lanes is reduced, and the overall traffic efficiency is effectively improved.

The invention relates to the technical field of lane line detection, in particular to a 3D-BEV lane line sensing method based on polar coordinates, and the method comprises the steps: carrying out the 3D lane line detection through employing the polar coordinates to replace a conventional method based on a rectangular coordinate system, employing a sliding long-strip-shaped window convolution to adapt to a polar coordinate system, and carrying out the 3D lane line detection through employing the sliding long-strip-shaped window convolution. According to the method, the lane line in the picture can be accurately identified in the polar coordinate system, accurate positioning is carried out in the 3D space, the false detection rate is greatly reduced, the radial distribution characteristic of perspective projection can be more naturally matched, and the modeling precision of the geometric structure of the lane line is remarkably improved.

The invention discloses a passable lane identification method and a vehicle, and belongs to the technical field of automatic driving. According to the technical scheme provided by the embodiment of the invention, under the condition that the target vehicle is in the automatic driving state, the target lane mask is determined by utilizing the environmental perception information, the local map information and the navigation instruction set in the driving direction of the target vehicle, namely, the position identification of the passable lane is realized by utilizing the multi-source and multi-mode data. The passable lane is determined by using the environmental perception information and the target lane mask subsequently, and the accuracy of the obtained passable lane is high.

The invention discloses an automatic driving method based on an improved Astar algorithm and a hierarchical state machine, and the method consists of three parts: road topology abstract graph construction, a lane adaptive Astar algorithm and the hierarchical state machine, and forms an efficient and cooperative'planning-decision 'integrated frame. Firstly, a road topology abstract graph is constructed. Secondly, path planning is carried out by adopting an improved lane adaptive Astar algorithm, and two stages of search strategies are included: in the first stage, road section-level global planning is carried out on a road topology abstract graph, and a macroscopic feasible path is rapidly determined; in the second stage, lane-level refined path generation is carried out in the selected road section, road curvature, width and static and dynamic obstacle distribution are comprehensively considered, and a smooth and safe accurate driving track meeting vehicle dynamics constraints is generated. Finally, a hierarchical state machine is designed, and the top layer is responsible for managing a driving mode and ensuring the global consistency of decisions; and the bottom layer executes fine speed control and lane changing control according to the top layer instruction.

The invention relates to the technical field of intelligent traffic, in particular to a vehicle driving area generation method and system, and the method comprises the steps: obtaining a vehicle position coordinate in a traffic image, and constructing a time-space point cloud; carrying out outlier filtering on the space-time point clouds, carrying out lane clustering based on the filtered space-time point clouds, and separating out point cloud clusters representing different lanes; calculating an alpha value based on the space density of the space-time point cloud; inputting the alpha value into an Alpha Shape algorithm to generate a road network level driving area, and independently applying the alpha value to each point cloud cluster to generate a lane level driving area; and generating a dynamic obstacle area based on the filtered static vehicle position points, integrating the generated road network level area and the lane level driving area, and outputting a vehicle multi-level driving area. According to the invention, the problems that a real dynamic driving area is difficult to accurately extract and abnormal tracks and outliers are insufficient to process in the prior art are effectively solved, and the accuracy of driving area generation and the adaptability to a complex traffic environment are improved.

The invention discloses a road passage control method and system based on visual detection, and relates to the technical field of road passage control methods, and the method comprises the steps: determining a release lane passage road according to a multi-phase passage road, and determining the congestion condition of each driving road, according to the congestion condition of each driving road, the plurality of traffic routes and the real-time image of the multi-phase passing road, the traffic guidance of the traffic signal to each driving road is determined, and the accuracy of the traffic guidance of the traffic signal to each driving road is improved. Determining an overall congestion combination according to the abnormal congestion event, the daily congestion event and the traffic plan project of each driving road; according to the traffic combination content of each overall congestion combination, the sub-traffic route of each driving road and the corresponding road peak driving flow, the driving load level corresponding to each driving road is determined, the accuracy of the driving load level corresponding to each driving road is improved, and intelligent traffic regulation and control of each driving road are triggered.

A method for controlling a driving mode transition of an autonomous driving vehicle may safely perform a lane change to avoid a collision accident with the other vehicle and prevent an injury to a joint, etc., while minimizing a sense of heterogeneity of a driver according to minimization of rotational steering of a steering wheel by determining whether the other vehicle in a next lane suddenly cuts in in an autonomous driving controller while the autonomous driving vehicle is driving in a manual driving mode and determining a safety distance from a rear vehicle to minimize rotational steering of the steering wheel held by the driver and at the same time perform partial braking control by autonomous driving mode transition control using an autonomous driving controller.

Provided is a travel road recognition device that can accurately recognize a travel road end of a travel road on which an own vehicle travels. A travel road recognition device of the present invention is a travel road recognition device that recognizes the travel road end of the travel road on which an own vehicle 301 travels, and includes: a laser radar 101 that emits a laser bean from the own vehicle 301 toward the travel road; a road surface determination unit 203 that obtains a cross slope of the travel road based on coordinate values of a point cloud obtained by the laser radar 101; and a road end determination unit 204 that obtains a change point at which a slope angle changes in the cross slope of the travel road obtained by the road surface determination unit 203, and obtains coordinate values of at least one travel road end of both sides in a cross direction of the travel road based on the coordinate values of the change point.

Provided in the present application are a vehicle driving control method, and a device. The method comprises: determining the distance between a vehicle and an intersection; when the distance is less than a first distance threshold, acquiring camera image data and radar point cloud data; on the basis of the camera image data and the radar point cloud data, determining perception information; on the basis of the perception information, determining whether the vehicle needs to make a lane change; and when it is determined that the vehicle needs to make a lane change, generating lane change path planning information on the basis of navigation map information, navigation positioning information and the perception information, wherein the lane change path planning information is used for controlling the vehicle to complete a lane change before the intersection. The method is used for achieving the effect of using a standard map to efficiently perform lane-level lane change navigation in an energy-saving manner.

A system and method for lane control of a fleet (18) of autonomous vehicles (16) in a road network with multi-lane roads is characterized in that it comprises a backend server (10) wirelessly connected to the fleet vehicles (16), which contains location-specific levels of automation (L1..L5) for each lane (26a, 26b, 26c) on multi-lane roads (24), wherein the fleet vehicles (16) moving on such a road (24) are designed and configured to determine environmental data of the fleet vehicle (16) for the current vehicle lane (26a, 26b, 26c) and vehicle position and to transmit this data to the backend server (10), wherein the backend server (10) uses at least this environmental data to determine a level of automation (L1...L5) determines and stores, and the backend server (10) is further trained and configured to transmit the levels of automation (L1...L5) to each fleet vehicle (16) for a given section of track (24) for the available lanes (26a, 26b, 26c), and each fleet vehicle (16) includes a trajectory computer (22) which is trained and configured to determine and follow a lane trajectory (27) from the transmitted levels of automation (L1...L5) of the lanes (26a, 26b, 26c) of the section of track (24), taking advantage of the highest possible level of automation.

A first toll advertisement message (TAM) is broadcast from a roadside unit (RSU), the first TAM indicating geographic locations of lanes of a roadway for which tolls are due and payment information for traversing the lanes of the roadway. A tolling usage message (TUM) is received from a first priority vehicle in receipt of the first TAM, the TUM including a request to reallocate one or more of the lanes of the roadway for use by the first priority vehicle. Responsive to receipt of the TUM, the RSU sends a TUM acknowledgment to the first priority vehicle indicating a lane allocation for use by the first priority vehicle and broadcasts a second TAM indicating the geographic locations of lanes of the roadway for which the tolls are due, the payment information for traversing the lanes of the roadway, and the lane allocation.

The present disclosure provides a region of interest selection method and device based on vehicle navigation, equipment, medium and product, relating to the technical field of artificial intelligence, especially in the field of autonomous driving. The specific implementation scheme is: determining the standard recognition result corresponding to the road video collected by the vehicle-mounted camera; using a region selection strategy to select a candidate region of interest for the road video to obtain at least one candidate region of interest; performing lane line recognition on the candidate region of interest to obtain a to-be-compared recognition result, and determining the selection result of the corresponding candidate region of interest according to the to-be-compared recognition result and the standard recognition result; determining the target recognition result that meets the result selection condition according to the selection result corresponding to each of the at least one candidate region of interest, and determining the candidate region of interest corresponding to the target recognition result as the target region of interest. The embodiment of the present disclosure realizes efficient and accurate selection of the target region of interest.

The embodiment of the invention provides a lane centering control method and device, a vehicle and a medium, and is applied to the technical field of vehicle automatic driving. The method comprises the steps that the current transverse deviation and the current course angle deviation of the vehicle relative to the current road track are acquired; according to the current transverse deviation and the current course angle deviation, a target steering wheel turning angle is determined, and lane centering control is conducted on the vehicle according to the target steering wheel turning angle. According to the embodiment of the invention, the target steering wheel turning angle is determined according to the current lateral deviation and the current course angle deviation, and then lane centering control is performed on the vehicle according to the target steering wheel turning angle, so that the problems that lane centering control is too complex, not fast enough, not accurate enough and poor in stability are solved, the complexity of lane centering control is reduced, and the lane centering control efficiency is improved. And the response speed, accuracy and stability are improved.

The invention discloses a traffic lane weed dynamic detection method and system fusing road surface features. The method comprises the following steps: acquiring videos of a road and a guardrail area; identifying trees, vegetation and road and guardrail ranges to obtain an identification result; calculating a three-dimensional space position of vegetation, and screening a weed candidate set invading the surface of a road; tracking and analyzing the motion trail characteristics, and evaluating the vegetation growth direction to obtain weeds to be determined; in combination with edge analysis and high-precision map data, determining whether to-be-determined weeds grow out of the cracks and the specific positions of the to-be-determined weeds so as to determine whether the to-be-determined weeds are traffic lane invasion weeds; and if yes, outputting the specific position of the candidate weed, and giving an alarm. By implementing the method, intelligent, efficient and full-scene automatic distinguishing and recognition of rigid weeds and non-rigid fallen leaves in the traffic lane can be achieved, the accuracy and efficiency of road maintenance are remarkably improved, and meanwhile manual intervention and cost are reduced.

The invention discloses a main line intersection CAV special lane setting method and a self-adaptive control system in a mixed driving environment. The method comprises the following steps: collecting information in real time; according to the current CAV permeability, the traffic capacity in the trunk line direction in three different states is calculated, the maximum net income of the traffic capacity is calculated, and then lane function switching judgment conditions are set in a mode of comparing the maximum net income of the traffic capacity; the lane function switching judgment condition is used for judging whether to open or close a CAV special lane, and determining that the CAV special lane is set as a CAV full-steering special lane or a CAV straight special lane; and iteratively optimizing the generated signal control scheme by adopting a PS-PPO-based multi-agent deep reinforcement learning algorithm, and generating an optimal signal control scheme in real time. According to the invention, the hybrid lane and CAV lane functions are dynamically adjusted according to the change of the CAV permeability, the cooperative optimization of the lane functions and the signal control scheme is realized, and the traffic efficiency of the trunk line intersection is improved to the greatest extent.

The utility model discloses a central median strip pavement drainage structure for superhigh excavation curved section, including outer, inner lane, central median strip and inner, outer drainage side ditch, the road height of outer lane to inner lane is gently reduced, be provided with two rows of anti -collision retaining wall by the crash barrier of anti -collision of central median strip, characterized in that: the lower part of crash barrier is provided with the flow guide gap, and the flow guide gap on the crash barrier in two rows of anti -collision retaining wall is aligned, and the flow guide gap of crash barrier is fixed with the groove type buckle, and the lower portion of groove type buckle forms the drainage channel, and the surface water on outer lane flows to inner lane through the drainage channel and flows into the inner drainage side ditch. The utility model discloses a central median strip pavement drainage structure, solved the existing superhigh excavation curved section and set up the problem of central median strip concrete pavement drainage, and this pavement drainage structure drainage mode is simple, direct, convenient construction, and the problem of highway drainage design has been solved.

The application relates to the technical field of intelligent driving, and discloses a lane line detection method, a model training method, a device, equipment and a storage medium, the model training method comprises the following steps: lane line labeling is performed on an original-size training sample image to obtain a first lane line coordinate label; the original-size training sample image is reduced to a preset size to obtain a preset-size training sample image; the preset-size training sample image is used for detection training of a lane line detection model, a detection result is output, and lane line detection coordinate values are determined according to the detection result; quantitative error values are obtained according to the lane line detection coordinate values and the detection result; and the lane line detection coordinate values, the quantitative error values and the first lane line coordinate label are used for parameter adjustment of the lane line detection model. The application solves the quantitative error problem existing in lane line detection based on key points at present, and improves the accuracy of lane line detection.

The invention discloses a method and a system for constructing a semantic topological map of an intersection, a medium and a vehicle. The method comprises the following steps: acquiring real-time pose data, lane-level topological map data and aerial view semantic map data of the vehicle; if the real-time pose data of the vehicle meets a triggering condition for constructing an intersection semantic topological map, encoding intersection lanes contained in the aerial view semantic map data to obtain intersection lane encoding map data; determining the current lane of the vehicle according to the real-time pose data of the vehicle and the lane-level topological map data; cutting the lane-level topological map data by referring to the real-time pose data of the vehicle and the current lane of the vehicle to obtain cut topological map data retaining the current lane of the vehicle and the available lanes of the vehicle; and fusing the cutting topology map data and the intersection lane coding map data to obtain an intersection semantic topology map.

The invention discloses a vehicle traffic state evaluation method and system based on a millimeter wave radar. The method comprises the following steps: acquiring a starting state and residual green light time of each lane; when the lane is not in the green light starting passing state, if the lane state correction condition is met, the green light starting passing state is corrected; when the lane is in the green light starting passing state, step-by-step condition judgment of the lane state is carried out, and the starting state of the lane is correspondingly corrected; all lanes are traversed, and the starting state and the remaining green light time of the lanes are corrected; and evaluating the starting state of the queuing vehicle based on the starting state of the nearest preceding vehicle of the queuing vehicle and the starting state of the corresponding lane. The vehicle starting state is evaluated by combining the lane starting state and the vehicle queuing condition in the lane, when a certain lane is in a green light starting passing state, the queuing vehicle can automatically start and advance after being not matched with the starting point cloud for a long time, and the situation that the queuing vehicle always waits in situ and is collided, rolled and deleted by other vehicles is avoided.

The invention relates to a vehicle control device, a vehicle control method, and a computer program for vehicle control. A vehicle control device is provided with: a planning unit (32) that, when a vehicle (10) is requested to perform a lane change from the own lane to another lane while traveling, determines a driving plan indicating the travel behavior of the vehicle before the lane change has been completed so as to satisfy a safety condition, and, under the safety condition, determines a driving plan indicating the travel behavior of the vehicle before the lane change has been completed; an object around a vehicle detected from a sensor signal obtained by a sensor (2) mounted on the vehicle does not collide with the vehicle. A setting unit (33) that sets a change completion condition indicating a position or time at which a lane change is requested to be completed; a determination unit (34) that determines whether or not a change completion condition is satisfied when the vehicle is caused to travel in accordance with the driving plan; and a control unit (35) that controls the vehicle so as to change the traffic lane in accordance with the driving plan when the change completion condition is satisfied, and restricts the implementation of the traffic lane change when the change completion condition is not satisfied.

A kind of suspension marshalling type highway or road indicating system and control method, by setting several suspension indicating devices on highway to form marshalling type prompting system, the indication of multiple suspension indicating devices in the direction of vehicle near the position where the road is occupied is adjusted in turn, and the indication of multiple suspension indicating devices forms a guide indicating line, the indication of the guide system is suspended in the air, and the oncoming vehicle can be observed from a long distance, enough time is provided for lane changing, and risk is avoided, in addition, the system is controlled remotely or by roadside operation, without entering the middle of the road, and the operator is not at risk during operation, and the safety is good.

The invention discloses a smart parking lot parking space detection and guidance system, which comprises a wiring pipe arranged above a parking lot traffic lane and distributed along the traffic lane, a plurality of shells are arranged on one side of the wiring pipe along the wiring pipe, a plurality of positioning ports are formed in the top surface of each shell, fixing frames are inserted into the positioning ports, and the positioning ports are provided with positioning holes. LED lamp strips are mounted on the sides, located in the shell, of the fixing frames, mounting openings communicated with an inner cavity of the shell are formed in the middle of the shell, mounting shells are arranged in the mounting openings, the two sides of each mounting shell obliquely protrude out of inclined shells facing the parking space, and sensors are mounted in the inclined shells. The parking lot is simple in structure, a complete guide line from a barrier gate to an unoccupied parking space is formed through dynamic lightening of the sectional type LED lamp strips, so that a vehicle can quickly find the unoccupied parking space, and the parking efficiency is greatly improved.

The application discloses a kind of vehicle-mounted panoramic all-around looking automatic calibration method, system, equipment and storage medium, the method includes: obtaining the lane line image under different angles that multiple vehicle-mounted cameras are shot;According to the internal parameter and external parameter of adjacent two vehicle-mounted cameras, the corresponding adjacent two lane line images are spliced, and multiple local all-around looking views are obtained;Each local all-around looking view is treated to remove distortion, so that the intersection center of local all-around looking view after removing distortion coincides with the plane center of imaging plane;Lane line projection and splicing fusion are carried out on each local all-around looking view after removing distortion, and panoramic all-around looking view is obtained;Wherein, the lane line in the panoramic all-around looking view is a smooth straight line, the present application uses lane line as calibration reference point, and the image non-smooth problem in panoramic all-around looking calibration process can be solved.

The invention belongs to the technical field of automatic driving and computer vision, and particularly discloses an intelligent lane line detection method for different illumination conditions, which comprises the following steps of: enhancing an original image by adopting multi-scale Retinex transformation in combination with G-L fractional order differential to obtain an enhanced image, and performing image segmentation on the enhanced image; the brightness, the contrast ratio and the edge details of the original image under complex illumination are improved; and performing lane line semantic segmentation on the enhanced image based on a lane line detection algorithm, and outputting a lane line segmentation result. According to the method, the lane line detection precision and robustness under different illumination and curvature conditions can be remarkably improved, and a reliable environment perception capability is provided for an automatic driving system.

Disclosed herein are system, method, and computer program product embodiments for detecting lane segments in an image for creating high-definition (HD) maps. A neural network can be used to classify a pixel of an image of a location as belonging to a driving lane. If the pixel belongs to a first boundary line of the driving lane, it is labeled accordingly. Based on the labeled pixel and one or more additional labeled pixels as part of the first boundary line of the driving lane, a first line drawing of the first boundary line of the driving lane is constructed. A lane segment based on a combination of the first line drawing of the first boundary line of the driving lane and a second line drawing of a second boundary line of the driving lane can be created.

The present application relates to the technical field of lane optimization, and more particularly to an efficient and safe dynamic lane optimization system and method, which comprises a server, a lane Internet of Things gateway, a toll card collection and processing device, a vehicle detection device and a lane display, an electric barrier is arranged in the middle of the lane, a manual toll booth is arranged on the safety island corresponding to the manual lane, a vehicle identifier is installed on the safety island corresponding to the switchable lane through a mobile device, and the vehicle identifier is used for identifying the license plate or ETC of the vehicle entering the lane. The method based on the above system is also disclosed, which optimizes the lane according to the actual toll station vehicle condition, can not only adjust the lane of the station, but also switch the ETC lane to a mixed lane, realizes the flexible and convenient switching of the ETC special lane as needed under special circumstances, solves the instantaneous non-ETC passing demand, and achieves the intelligent expansion without increasing additional lanes.

The invention discloses a guideboard shielding detection method and system based on fusion of a dynamic video and distance perception. The method comprises the following steps: acquiring a road video stream and vehicle dynamic parameters; identifying and positioning a guideboard in the road video stream, performing lane line segmentation, and determining whether the guideboard is located in a lane concerned by a driver in combination with a lane line segmentation result; if yes, determining the guideboard as an associated guideboard, performing semantic segmentation on the area of the associated guideboard to identify the type of the shelter, and determining whether the shelter is an effective shelter or not by calculating the weighted overlap ratio of the shelter and the associated guideboard area; and if yes, estimating the actual distance between the vehicle and the guideboard based on a distance measurement algorithm to determine whether the shelter is only within the effective sight distance and continuously occurs, and if yes, generating an alarm record. By implementing the method provided by the invention, not only can self-adaption to a complex driving environment be realized, but also an area which is crucial to a driver can be intelligently concerned, and accurate shielding judgment can be realized within an effective sight distance.

This specification provides a method for training a lane detection model. The method includes: determining lane line image samples and sample labels, wherein the sample labels are target lane lines in the lane line image samples; extracting features from the lane line image samples to obtain a first feature image and a second feature image; determining lane line clustering information and initial lane line key points based on the first feature image and the second feature image; performing clustering based on the lane line clustering information and the initial lane line key points to obtain predicted lane lines; and training a lane line detection model based on the predicted lane lines and the target lane lines. By integrating the clustering process of lane line key points into the entire lane line detection model training process, lane line key point clustering is performed simultaneously with network training, thereby improving the overall performance of the lane line detection model.