Passable area analysis method based on high-precision map and fusion perception

Abstract

The present application relates to the technical field of driving environment perception, in particular to a passable area analysis method based on high-precision map and fusion perception, comprising: obtaining the associated road node corresponding to the road node to which the current position of the target vehicle belongs; and combining the passable state of the associated road node in the current time obtained from the road monitoring image, predicting the passable congestion risk area of the road node to which the current position of the target vehicle belongs. The present application not only realizes the perception identification of the position of the target vehicle in the high-precision map, but also combines the obstacle information around the vehicle perceived by multiple sensors to realize the calibration of the high-precision map to which the target vehicle belongs, ensuring the accuracy of the generated passable area; at the same time, the influence of the flow of vehicles in the road and the congestion risk of different roads in different time periods on the passable area is also considered, ensuring the accuracy of the constructed passable area.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of driving environment perception, in particular to a passable area analysis method based on a high-precision map and fusion perception. BACKGROUND

[0002] As an indispensable important component of an autonomous vehicle, a high-precision map provides favorable support for vehicle positioning, safe driving, path planning, and vehicle energy saving. However, in a real scene of autonomous driving, in order to ensure the safety of autonomous driving as much as possible, the environment must be perceived under multi-source data fusion. Therefore, obtaining a passable area of a vehicle based on a high-precision map and fusion perception technology has gradually become a focus of research in the industry.

[0003] The existing passable area analysis method based on a high-precision map and fusion perception only realizes the identification of the position of a target vehicle in a high-precision map, combines multi-sensor perception of obstacle information around the vehicle, and realizes the extraction of the passable area of the target vehicle. However, this method has a large drawback, as it does not consider the flow of vehicles on the road and the congestion risk of different roads at different time periods, and thus the precision of the passable area extracted by the existing technology has a large deviation. SUMMARY

[0004] The purpose of the present application is to provide a passable area analysis method based on a high-precision map and fusion perception to solve the problems raised in the background.

[0005] In order to solve the above technical problems, the present application provides the following technical solution: a passable area analysis method based on a high-precision map and fusion perception, the method comprising the following steps:

[0006] S1, obtaining a task planning map of a target vehicle; when the target vehicle is located in the task planning map, real-time obtaining vehicle monitoring information corresponding to each position point of the target vehicle and vehicle monitoring information corresponding to associated vehicles of the target vehicle through fusion perception technology;

[0007] S2, optimizing the task planning map of the target vehicle in combination with the vehicle monitoring information of the target vehicle and its associated vehicles, and locking the passable area within a preset first radius area around the target vehicle;

[0008] S3, combining historical road traffic information within a preset second radius area around the current position of the target vehicle, analyzing the congestion state of each road node within the preset second radius area around the current position of the target vehicle at different time points, and the congestion correlation influence between different road nodes;

[0009] S4, acquire the associated road node corresponding to the road node to which the current position of the target vehicle belongs; and in combination with the passing state of the associated road node obtained from the road monitoring image at the current time, predict the passing congestion risk area of the road node to which the current position of the target vehicle belongs;

[0010] S5, based on the passable area within the preset first radius area around the current target vehicle and the passing congestion risk area of the road node to which the current position of the target vehicle belongs, generate the passable area corresponding to the target vehicle at the current time.

[0011] Further, the task planning map of the target vehicle is a planning path map between the initial point corresponding to the target vehicle and the destination;

[0012] The vehicle-mounted monitoring information includes obstacle information, path boundary information and road surface flatness information of the surrounding obstacles other than vehicles identified based on image recognition technology;

[0013] The obstacle information includes the obstacle contour and the position vector of the corresponding obstacle based on the vehicle-mounted monitoring information of the vehicle; the path boundary information includes the position vector of each path boundary point identified based on the vehicle-mounted monitoring information of the vehicle; and the road surface flatness information includes each road surface position point corresponding to the flatness not belonging to the preset interval and the position vector of the corresponding road surface position point based on the vehicle-mounted monitoring information of the vehicle;

[0014] The associated vehicle of the target vehicle is the same as the vehicle-to-vehicle network connected to the target vehicle, and the distance between the vehicle positions uploaded by the target vehicle and its associated vehicle within the preset upload time deviation is less than the preset distance;

[0015] The preset upload time deviation represents the maximum interval time length between the data upload times corresponding to the target vehicle and its associated vehicle in the database.

[0016] In the present application, the task planning map of the target vehicle is obtained based on the preset high-precision map, that is, the task route of the target vehicle is planned on the preset high-precision map, and the task route planning result of the target vehicle is intercepted in the area corresponding to the high-precision map belonging to the target vehicle as the task planning map of the target vehicle;

[0017] Further, when optimizing the task planning map of the target vehicle, the mapping position points of the current position of the target vehicle and the position of its associated vehicle in the task planning map of the target vehicle are acquired respectively;

[0018] Based on the mapping position points in the task planning map of the target vehicle, a boundary position mapping area corresponding to the path boundary information in the vehicle monitoring information corresponding to each mapping position point is generated in the task planning map of the target vehicle, boundary feature information in the obtained boundary position mapping area is extracted, and the associated boundary feature information of the obtained boundary feature information in the task planning map of the target vehicle is obtained; according to the boundary feature information and the associated boundary feature information corresponding thereto, the forward mapping calibration coefficient and the bias mapping calibration coefficient corresponding to the vehicle monitoring information corresponding to the target vehicle position and the associated vehicle position at the current time respectively are obtained;

[0019] The straight line in which the respective advancing directions of the target vehicle and the associated vehicle are located is taken as the first reference pointing line of the corresponding vehicle; and the straight line perpendicular to the respective advancing directions of the target vehicle and the associated vehicle is taken as the second reference pointing line of the corresponding vehicle.

[0020] The vehicle monitoring information calibration result corresponding to the target vehicle and the associated vehicle at the current time is calculated, the calibration result corresponding to the position vector in each element in the vehicle monitoring information calibration result is equal to the vector sum of the first calibration subvector and the second calibration subvector corresponding to the corresponding position vector; the first calibration subvector corresponding to the corresponding position vector in the vehicle monitoring information calibration result is equal to the subvector of the corresponding position vector on the first reference pointing line multiplied by the number of the forward mapping calibration coefficient corresponding to the corresponding vehicle monitoring information; the second calibration subvector corresponding to the corresponding position vector in the vehicle monitoring information calibration result is equal to the subvector of the corresponding position vector on the second reference pointing line multiplied by the number of the bias mapping calibration coefficient corresponding to the corresponding vehicle monitoring information.

[0021] According to the vehicle monitoring information calibration result corresponding to the target vehicle position and the associated vehicle position at the current time; based on the mapping position points in the task planning map of the target vehicle, the position mapping area corresponding to the obstacle information, the path boundary information and the road flatness information in the vehicle monitoring information calibration result corresponding to each mapping position point is generated in the task planning map of the target vehicle and is marked, and the optimization result of the task planning map of the target vehicle is obtained.

[0022] The passable area in the preset first radius area around the current target vehicle that is locked indicates the area in the optimization result of the task planning map of the target vehicle that is not marked and belongs to the road area in the preset first radius area around the current target vehicle; the area in the preset first radius area around the current target vehicle that does not belong to the road area is also marked.

[0023] The passable area in the first radius area around the current target vehicle in the application is obtained based on high-precision map and sensor perception data analysis, and belongs to the analysis result of the passable area at the perception level; in the process, the high-precision map and the sensor perception result complement each other, the high-precision map can dynamically select the accurate map model around the target vehicle according to the position of the target vehicle, and the sensor perception technology can perceive and obtain the obstacle information, path boundary information and road flatness information around the vehicle, and can also calibrate the position of the target vehicle in the extracted high-precision map.

[0024] Further, the boundary feature information includes the positions of each guardrail post in the road boundary and the curvature of the road corresponding to the position of each guardrail post; when obtaining the associated boundary feature information corresponding to the boundary feature information of the vehicle in the task planning map of the target vehicle, the associated boundary feature information is obtained by querying the boundary feature information in the task planning map of the target vehicle that is closest to the mapping position point of the corresponding vehicle and has the same boundary feature information as the corresponding vehicle.

[0025] The forward mapping calibration coefficient represents the mapping calibration coefficient corresponding to the component vector of each position vector in the obtained vehicle monitoring information on the first reference pointing line corresponding to the corresponding vehicle;

[0026] The forward mapping calibration coefficients of the target vehicle and its associated vehicles are equal to the average value of the mapping calibration coefficients corresponding to the component vectors of each position vector in the path boundary information of the corresponding vehicle on the first reference pointing line, and the mapping calibration coefficient corresponding to the component vector of each position vector in the path boundary information of the vehicle on the first reference pointing line is equal to the quotient of the scalar of the component vector of each position vector in the path boundary information of the vehicle on the first reference pointing line and the scalar of the component vector of the corresponding position vector in the corresponding associated boundary feature information on the first reference pointing line, and when the scalar of the component vector of the corresponding position vector in the corresponding associated boundary feature information on the first reference pointing line is 0, it is determined that the mapping calibration coefficient corresponding to the component vector of the corresponding position vector in the path boundary information on the first reference pointing line is 1;

[0027] The bias mapping calibration coefficient represents the mapping calibration coefficient corresponding to the component vector of each position vector in the obtained vehicle monitoring information on the second reference pointing line corresponding to the corresponding vehicle;

[0028] The mapping calibration coefficient corresponding to the partial vector of each position vector in the path boundary information of the target vehicle and the corresponding associated boundary feature information on the second reference direction line is equal to the quotient of the scalar of the partial vector of each position vector in the path boundary information of the vehicle on the second reference direction line and the scalar of the partial vector of the corresponding position vector in the corresponding associated boundary feature information on the second reference direction line, and when the scalar of the partial vector of the corresponding position vector in the corresponding associated boundary feature information on the second reference direction line is 0, the mapping calibration coefficient corresponding to the partial vector of the corresponding position vector in the path boundary information of the vehicle on the second reference direction line is determined to be 1.

[0029] The present application not only has the basis guarantee of high-precision map (HDMAP), but also plays the respective advantages of sensors to improve the reliability of environmental perception information. The construction of the mapping calibration coefficient can calibrate the vehicle-mounted monitoring information collected by each vehicle, and the high-precision map obtained by the vehicle-mounted monitoring information calibration result can be updated, which lays a data foundation for subsequent accurate acquisition of passable areas.

[0030] Further, when analyzing the congestion states corresponding to each road node in a preset second radius area around the current position of the target vehicle in different preset time intervals,

[0031] The historical road traffic information in the preset second radius area around the current position of the target vehicle is obtained, and a set of each road node in the preset second radius area around the current position of the target vehicle is constructed, which is denoted as a road node set;

[0032] The historical road traffic information includes traffic feature information corresponding to each road node in the corresponding area in different preset time intervals in a day, and the traffic feature information includes a traffic state corresponding to each preset position area in the corresponding road node. The traffic state includes a normal traffic state and a traffic congestion state. The state of a construction area or an obstacle occupied area existing in the road node is also recorded as a traffic congestion state.

[0033] The road node corresponds to a preset road area in the database, different road nodes correspond to different preset road areas, and the areas corresponding to each road area are equal. The preset time intervals in a day are multiple, and the interval lengths corresponding to different preset time intervals are equal.

[0034] The traffic feature information corresponding to each day and each preset time interval in the historical road traffic information corresponding to the kth element in the road node set is obtained,

[0035] A set of traffic feature information corresponding to the jth preset time interval in different dates in the historical road traffic information corresponding to the kth element in the statistical road node set is denoted as a congestion state analysis set YDF of the kth element in the statistical road node set in the jth preset time interval (k，j) ;

[0036] The frequency of the traffic state being a traffic congestion state in the elements of YDF (k，j) of each preset position region in the kth element in the statistical road node set is obtained, and all preset position regions with a ratio of the obtained frequency to the total number of elements in YDF (k，j) greater than a preset frequency are marked; each position point in the marked preset position region is also marked;

[0037] In the congestion state corresponding to the kth element of the road node set at the jth preset time interval in a day, the traffic state of all marked preset position regions is a traffic congestion state, and the traffic state of all unmarked preset position regions is a normal traffic state.

[0038] Further, when analyzing the congestion correlation influence between different road nodes,

[0039] The historical road traffic information in a preset second radius region around the current position of the target vehicle is obtained, and any element in the road node set corresponding to the road node is denoted as a reference node;

[0040] The congestion time interval corresponding to each occurrence of the congestion state of the reference node in the obtained historical road traffic information is denoted as a first time interval, and a set of all non-reference nodes with a non-empty intersection of the corresponding congestion time interval and the first time interval is denoted as a congestion correlation subset corresponding to the congestion state of the reference node;

[0041] The set of road nodes in the road node set that have a congestion correlation influence with the reference node is the union set of each congestion correlation subset corresponding to the congestion state of the reference node.

[0042] Further, the associated road node corresponding to the road node to which the current position of the target vehicle belongs is all road nodes that have a congestion correlation influence with the road node to which the current position of the target vehicle belongs;

[0043] When predicting the traffic congestion risk area of the road node to which the target vehicle currently belongs, the traffic state of the obtained associated road node in the current time in the road monitoring image is obtained,

[0044] If the traffic state of any one of the obtained associated road nodes is a traffic jam state, it is determined that all the obtained associated road nodes are congestion risk nodes; if the traffic state of the obtained associated road nodes is all normal traffic state, it is determined that all the obtained associated road nodes are not congestion risk nodes.

[0045] A time interval to which the current time belongs is recorded as T.

[0046] Obtained each congestion state analysis set corresponding to each congestion risk node in the time interval T, and a set of each preset location area marked in each obtained congestion state analysis set is taken as a traffic congestion risk area of a road node to which a current location of the target vehicle belongs.

[0047] Further, when the passable area corresponding to the target vehicle at the current time is generated,

[0048] A passable area in a preset first radius area around the current target vehicle is obtained, which is recorded as a first passable area.

[0049] A traffic congestion risk area of a road node to which a current location of the target vehicle belongs is obtained, which is recorded as a second passable area.

[0050] The passable area corresponding to the target vehicle at the current time is a set of position points in the intersection area of the first passable area and the second passable area, which are not marked in the first passable area and the second passable area.

[0051] Compared with the prior art, the present application has the beneficial effects that: the present application not only realizes the identification of the position of the target vehicle in the high-precision map, but also combines the obstacle information around the vehicle perceived by multiple sensors to realize the calibration of the high-precision map to which the target vehicle belongs, and ensure the accuracy of the generated passable area; at the same time, the influence of the flow of vehicles in the road and the congestion risk of different roads in different time periods on the passable area is also considered to ensure the accuracy of the constructed passable area. BRIEF DESCRIPTION OF DRAWINGS

[0052] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate the present application and explain the principles of the present application, and do not constitute a limitation of the present application. In the drawings:

[0053] Figure 1 is a flow diagram of the passable area analysis method of the present application based on high-precision map and fusion perception. DETAILED DESCRIPTION

[0054] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0055] Please refer to Figure 1 The present application provides a technical solution: a passable area analysis method based on a high-precision map and fusion perception, which comprises the following steps:

[0056] S1, obtaining a task planning map of a target vehicle; when the target vehicle is located in the task planning map, obtaining, in real time, vehicle-mounted monitoring information corresponding to the target vehicle at each position point and vehicle-mounted monitoring information corresponding to associated vehicles of the target vehicle through fusion perception technology

[0057] The task planning map of the target vehicle is a planning path map between an initial point corresponding to the target vehicle and a destination.

[0058] The vehicle-mounted monitoring information comprises obstacle information, path boundary information and road surface flatness information of the surrounding obstacles other than vehicles, which are identified based on image recognition technology.

[0059] The obstacle information comprises an obstacle contour and a position vector of the corresponding obstacle based on the vehicle-mounted monitoring information of the vehicle; the path boundary information comprises a position vector of each path boundary point identified based on the vehicle-mounted monitoring information of the vehicle; and the road surface flatness information comprises each road surface position point corresponding to a flatness not belonging to a preset interval and a position vector of the corresponding road surface position point based on the vehicle-mounted monitoring information of the vehicle.

[0060] The associated vehicles of the target vehicle are connected to the same Internet of Vehicles as the target vehicle, and the distance between the vehicle positions uploaded by the target vehicle and its associated vehicles within a preset upload time deviation is less than a preset distance.

[0061] The preset upload time deviation represents the maximum interval duration between the data upload times corresponding to the target vehicle and its associated vehicles in the database.

[0062] S2, optimizing the task planning map of the target vehicle in combination with the vehicle-mounted monitoring information of the target vehicle and its associated vehicles, and locking a passable area within a preset first radius area around the target vehicle.

[0063] When the task planning map of the target vehicle is optimized, the mapping position points of the target vehicle position and the associated vehicle positions of the current time in the task planning map of the target vehicle are obtained.

[0064] Based on the mapping position points in the task planning map of the target vehicle, a boundary position mapping area corresponding to the path boundary information in the vehicle monitoring information corresponding to each mapping position point in the task planning map of the target vehicle is generated, boundary feature information in the obtained boundary position mapping area is extracted, and associated boundary feature information of the boundary feature information in the obtained boundary position mapping area in the task planning map of the target vehicle is obtained; according to the boundary feature information and the associated boundary feature information corresponding thereto, a forward mapping calibration coefficient and a bias mapping calibration coefficient corresponding to the vehicle monitoring information corresponding to the target vehicle position and the associated vehicle position at the current time respectively are obtained;

[0065] A straight line in which the respective advancing directions of the target vehicle and the associated vehicle are located is taken as a first reference pointing line of the corresponding vehicle; and a straight line perpendicular to the respective advancing directions of the target vehicle and the associated vehicle is taken as a second reference pointing line of the corresponding vehicle;

[0066] A vehicle monitoring information calibration result corresponding to the target vehicle and the associated vehicle at the current time is calculated, a calibration result corresponding to a position vector in each element in the vehicle monitoring information calibration result is equal to a vector sum of a first calibration subvector and a second calibration subvector corresponding to the corresponding position vector; the first calibration subvector corresponding to the corresponding position vector in the vehicle monitoring information calibration result is equal to a subvector of the corresponding position vector on the first reference pointing line multiplied by a number of the forward mapping calibration coefficient corresponding to the corresponding vehicle monitoring information; and the second calibration subvector corresponding to the corresponding position vector in the vehicle monitoring information calibration result is equal to a subvector of the corresponding position vector on the second reference pointing line multiplied by a number of the bias mapping calibration coefficient corresponding to the corresponding vehicle monitoring information;

[0067] In the embodiment, the vehicle pointing of the target vehicle and the associated vehicle may be different, and thus the first reference pointing line and the second reference pointing line corresponding to different vehicles in the target vehicle and the associated vehicle may also be different; and the first reference pointing line and the second reference pointing line corresponding to the same vehicle at different times may also be different, and thus the forward mapping calibration coefficient and the bias mapping calibration coefficient corresponding to the same vehicle at different times may be changed.

[0068] According to the vehicle monitoring information calibration result corresponding to the target vehicle position and the associated vehicle position at the current time; based on the mapping position points in the task planning map of the target vehicle, a position mapping area corresponding to obstacle information, path boundary information and road flatness information in the vehicle monitoring information calibration result corresponding to each mapping position point in the task planning map of the target vehicle is generated and marked, and an optimization result of the task planning map of the target vehicle is obtained;

[0069] The passable area in the preset first radius area around the current target vehicle represents an area in the preset first radius area around the current target vehicle in the task planning map of the target vehicle that is not marked and belongs to a road area in an optimization result of the task planning map of the target vehicle, and an area in the preset first radius area around the current target vehicle that does not belong to the road area is also marked.

[0070] The boundary feature information includes positions of each guardrail post in the road boundary and curvatures of the road corresponding to the positions of the respective guardrail posts; when the associated boundary feature information corresponding to the boundary feature information of the vehicle in the task planning map of the target vehicle is obtained, the associated boundary feature information is obtained by querying boundary feature information in the task planning map of the target vehicle that is closest to a mapping position point of the respective vehicle and is identical to the boundary feature information corresponding to the respective vehicle.

[0071] The forward mapping calibration coefficient represents a mapping calibration coefficient corresponding to a component vector of a position vector corresponding to each element in the obtained vehicle monitoring information on a first reference pointing line corresponding to the respective vehicle;

[0072] The forward mapping calibration coefficients of the target vehicle and the associated vehicle respectively are equal to average values of mapping calibration coefficients corresponding to component vectors of respective position vectors in the path boundary information of the respective vehicle on the first reference pointing line, the mapping calibration coefficient corresponding to the component vector of each position vector in the path boundary information of the vehicle on the first reference pointing line is equal to a quotient of a scalar of the component vector of each position vector in the path boundary information of the vehicle on the first reference pointing line and a scalar of a corresponding position vector in the associated boundary feature information on the first reference pointing line, and when the scalar of the corresponding position vector in the associated boundary feature information on the first reference pointing line is 0, the mapping calibration coefficient corresponding to the component vector of the corresponding position vector in the path boundary information on the first reference pointing line is determined to be 1.

[0073] The bias mapping calibration coefficient represents a mapping calibration coefficient corresponding to a component vector of a position vector corresponding to each element in the obtained vehicle monitoring information on a second reference pointing line corresponding to the respective vehicle;

[0074] The corresponding bias mapping calibration coefficient of the target vehicle and its associated vehicles is equal to the average of the mapping calibration coefficients corresponding to the respective position vectors in the path boundary information of the corresponding vehicle on the second reference pointing line. The mapping calibration coefficient corresponding to the position vector in the path boundary information of the vehicle on the second reference pointing line is equal to the quotient of the scalar of the position vector in the second reference pointing line of each position vector in the path boundary information of the vehicle and the scalar of the corresponding position vector in the second reference pointing line of the corresponding associated boundary feature information. When the scalar of the corresponding position vector in the second reference pointing line of the corresponding associated boundary feature information is 0, the mapping calibration coefficient corresponding to the position vector in the second reference pointing line of the corresponding position vector in the path boundary information is determined to be 1.

[0075] S3, in combination with the historical road traffic information in the preset second radius area around the current position of the target vehicle, analyze the congestion states corresponding to each road node in the preset second radius area around the current position of the target vehicle at different time points, and the congestion correlation influence between different road nodes;

[0076] When analyzing the congestion states corresponding to each road node in the preset second radius area around the current position of the target vehicle in different preset time intervals,

[0077] Obtain the historical road traffic information in the preset second radius area around the current position of the target vehicle, and construct a set of each road node in the preset second radius area around the current position of the target vehicle, denoted as a road node set;

[0078] The historical road traffic information includes traffic feature information corresponding to each road node in the corresponding area in different preset time intervals in a day. The traffic feature information includes the traffic state corresponding to each preset position area in the corresponding road node. The traffic state includes normal traffic state and traffic congestion state. The state of the construction area or the obstacle occupied area existing in the road node is also recorded as the traffic congestion state;

[0079] The road node corresponds to a preset road area in the database. Different road nodes correspond to different preset road areas, and the areas corresponding to each road area are equal. The preset time intervals in a day are multiple and different, and the interval lengths corresponding to different preset time intervals are equal;

[0080] In this embodiment, a day is divided into 24 continuous and equal length preset time intervals, and the length of each preset time interval is one hour. The number of preset time interval divisions is artificially preset and can be adjusted according to human needs;

[0081] obtaining the traffic feature information corresponding to each preset time interval of each day in the historical road traffic information corresponding to the kth element in the road node set,

[0082] a set of the traffic feature information corresponding to the jth preset time interval in different dates in the historical road traffic information corresponding to the kth element in the road node set is denoted as a congestion state analysis set YDF of the kth element in the road node set in the jth preset time interval (k，j) ;

[0083] counting the frequency of the traffic state being the traffic congestion state in the elements of YDF (k，j) of the kth element in the road node set, and marking all the preset location areas whose ratio of the obtained frequency to the total number of elements in YDF (k，j) is greater than a preset frequency; each location point in the marked preset location area is also marked;

[0084] when the jth preset time interval in a day is based, the traffic state of all the marked preset location areas in the congestion state corresponding to the kth element in the road node set is the traffic congestion state, and the traffic state of all the unmarked preset location areas is the normal traffic state.

[0085] when analyzing the congestion correlation influence between different road nodes,

[0086] obtaining the historical road traffic information in a preset second radius area around the current position of the target vehicle, and recording any element in the road node set as a reference node;

[0087] obtaining the congestion time interval corresponding to each occurrence of the congestion state of the reference node in the obtained historical road traffic information, denoted as a first time interval, and obtaining a set of all non-reference nodes whose corresponding congestion time interval and the first time interval have a non-empty intersection, denoted as a congestion correlation subset corresponding to the congestion state of the reference node;

[0088] the set of road nodes in the road node set that have a congestion correlation influence with the reference node is the union set of each congestion correlation subset corresponding to the congestion state of the reference node.

[0089] S4, obtaining the associated road node corresponding to the road node to which the current position of the target vehicle belongs; and combining the traffic state of the obtained associated road node in the current time in the road monitoring image to predict the traffic congestion risk area of the road node to which the current position of the target vehicle belongs;

[0090] The associated road nodes corresponding to the road node to which the current position of the target vehicle belongs are all road nodes that have congestion associated influence with the road node to which the current position of the target vehicle belongs;

[0091] When the traffic congestion risk area of the road node to which the current position of the target vehicle belongs is predicted, the traffic states of the associated road nodes obtained from the road monitoring image at the current time are acquired,

[0092] If the traffic state of any one of the associated road nodes obtained is a traffic congestion state, it is determined that all the associated road nodes obtained are congestion risk nodes; if the traffic states of the associated road nodes obtained are all normal traffic states, it is determined that all the associated road nodes obtained are not congestion risk nodes;

[0093] The time interval to which the current time belongs is recorded as T;

[0094] The congestion state analysis sets corresponding to each congestion risk node in the time interval T are acquired, and the set of each preset position area marked in each congestion state analysis set obtained is taken as the traffic congestion risk area of the road node to which the current position of the target vehicle belongs.

[0095] S5, based on the passable area in the preset first radius area around the current target vehicle and the traffic congestion risk area of the road node to which the current position of the target vehicle belongs, a passable area corresponding to the target vehicle at the current time is generated;

[0096] When the passable area corresponding to the target vehicle at the current time is generated,

[0097] The passable area in the preset first radius area around the current target vehicle is acquired and recorded as a first traffic area;

[0098] The traffic congestion risk area of the road node to which the current position of the target vehicle belongs is acquired and recorded as a second traffic area;

[0099] The passable area corresponding to the target vehicle at the current time is a set of position points in the intersection area of the first traffic area and the second traffic area, which are not marked in the first traffic area and the second traffic area.

[0100] It is to be noted that, in the present text, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0101] Finally, it should be noted that the above-mentioned only constitutes preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications, equivalent replacements, improvements and the like of the technical solutions described in the foregoing embodiments can still be made. Any modifications, equivalent replacements, improvements and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A passable area analysis method based on a high-precision map and fusion perception, characterized in that, The method comprises the following steps: S1, acquiring a task planning map of a target vehicle; when the target vehicle is located in the task planning map, acquiring vehicle-mounted monitoring information of the target vehicle at each position point and vehicle-mounted monitoring information of associated vehicles of the target vehicle in real time through fusion perception technology; S2, optimizing the task planning map of the target vehicle in combination with the vehicle-mounted monitoring information of the target vehicle and its associated vehicles, and locking a passable area in a preset first radius area around the current target vehicle; S3, in combination with historical road traffic information in a preset second radius area around the current position of the target vehicle, analyzing congestion states of each road node in the preset second radius area around the current position of the target vehicle at different time points, and analyzing congestion correlation influences between different road nodes; S4, acquiring associated road nodes corresponding to the road node to which the current position of the target vehicle belongs, and in combination with a traffic state of the associated road nodes at the current time obtained from road monitoring images, predicting a traffic congestion risk area of the road node to which the current position of the target vehicle belongs; S5, based on the passable area in the preset first radius area around the current target vehicle and the traffic congestion risk area of the road node to which the current position of the target vehicle belongs, generating a passable area corresponding to the target vehicle at the current time: acquiring the passable area in the preset first radius area around the current target vehicle, denoted as a first traffic area; acquiring the traffic congestion risk area of the road node to which the current position of the target vehicle belongs, denoted as a second traffic area; the passable area corresponding to the target vehicle at the current time is a set of position points in the intersection area of the first traffic area and the second traffic area, which are not marked in the first traffic area and the second traffic area.

2. The passable area analysis method based on a high-definition map and fusion perception according to claim 1, characterized in that: The task planning map of the target vehicle is a planning path map between an initial point corresponding to the target vehicle and a destination; The vehicle-mounted monitoring information comprises obstacle information, path boundary information and road surface flatness information of the surrounding obstacles other than vehicles, which are identified based on image recognition technology; The obstacle information comprises an obstacle contour and a position vector of the corresponding obstacle based on the vehicle-mounted monitoring information of the vehicle; the path boundary information comprises a position vector of each path boundary point based on the vehicle-mounted monitoring information of the vehicle; and the road surface flatness information comprises each road surface position point corresponding to a flatness not belonging to a preset interval and a position vector of the corresponding road surface position point based on the vehicle-mounted monitoring information of the vehicle; The associated vehicles of the target vehicle are connected to the same Internet of Vehicles as the target vehicle, and the distance between the vehicle positions uploaded by the target vehicle and its associated vehicles within a preset upload time deviation is less than a preset distance; The preset upload time deviation represents a maximum interval time length between the data upload times corresponding to the target vehicle and its associated vehicles in the database.

3. The passable area analysis method based on high-definition map and fusion perception according to claim 2, characterized in that: When the task planning map of the target vehicle is optimized, the mapping position points of the current time target vehicle position and its associated vehicle positions in the task planning map of the target vehicle are acquired respectively. The boundary position mapping area corresponding to the path boundary information in the vehicle-mounted monitoring information corresponding to each mapping position point in the task planning map of the target vehicle is generated in the task planning map of the target vehicle based on the mapping position points in the task planning map of the target vehicle, the boundary feature information in the obtained boundary position mapping area is extracted, and the associated boundary feature information of the boundary feature information in the obtained boundary position mapping area in the task planning map of the target vehicle is obtained; the forward mapping calibration coefficient and the bias mapping calibration coefficient corresponding to the vehicle-mounted monitoring information corresponding to the target vehicle position and the associated vehicle position at the current time respectively are obtained according to the boundary feature information and the associated boundary feature information corresponding thereto; The straight line in which the respective advancing directions of the target vehicle and the associated vehicle are located is taken as the first reference pointing line of the corresponding vehicle; and the straight line corresponding to the respective advancing directions of the target vehicle and the associated vehicle and being perpendicular to the advancing directions is taken as the second reference pointing line of the corresponding vehicle; The vehicle-mounted monitoring information calibration result corresponding to the target vehicle and the associated vehicle at the current time is calculated, the calibration result corresponding to the position vector in each element in the vehicle-mounted monitoring information calibration result is equal to the vector sum of the first calibration subvector and the second calibration subvector corresponding to the corresponding position vector, the first calibration subvector corresponding to the corresponding position vector in the vehicle-mounted monitoring information calibration result is equal to the number multiplication vector of the subvector of the corresponding position vector on the first reference pointing line and the forward mapping calibration coefficient corresponding to the corresponding vehicle-mounted monitoring information, and the second calibration subvector corresponding to the corresponding position vector in the vehicle-mounted monitoring information calibration result is equal to the number multiplication vector of the subvector of the corresponding position vector on the second reference pointing line and the bias mapping calibration coefficient corresponding to the corresponding vehicle-mounted monitoring information; The vehicle-mounted monitoring information calibration result corresponding to the target vehicle position and the associated vehicle position at the current time is obtained; the position mapping area corresponding to the obstacle information, the path boundary information and the road flatness information in the vehicle-mounted monitoring information calibration result corresponding to each mapping position point in the task planning map of the target vehicle is generated in the task planning map of the target vehicle based on the mapping position points in the task planning map of the target vehicle, and is marked to obtain the optimization result of the task planning map of the target vehicle; The passable area in the preset first radius area around the current target vehicle that is locked indicates the area in the optimization result of the task planning map of the target vehicle that is not marked and belongs to the road area in the preset first radius area around the current target vehicle; The area in the preset first radius area around the current target vehicle that does not belong to the road area is also marked.

4. The passable area analysis method based on a high-definition map and fusion perception according to claim 3, characterized in that: The boundary feature information includes the positions of each guardrail post in the road boundary and the curvature of the road corresponding to the position of the corresponding guardrail post; when the associated boundary feature information corresponding to the boundary feature information of the vehicle in the task planning map of the target vehicle is obtained, the boundary feature information closest to the mapping position point of the corresponding vehicle and the same as the boundary feature information corresponding to the corresponding vehicle in the task planning map of the target vehicle is queried; The forward mapping calibration coefficient represents the mapping calibration coefficient corresponding to the subvector of the position vector in the obtained vehicle-mounted monitoring information on the first reference pointing line corresponding to the corresponding vehicle. The corresponding positive mapping calibration coefficient of the target vehicle and its associated vehicles is equal to the average of the corresponding mapping calibration coefficients of the corresponding position vectors of each position vector in the path boundary information of the corresponding vehicle on the first reference direction line, the corresponding mapping calibration coefficient of the corresponding position vector of each position vector in the path boundary information of the vehicle on the first reference direction line is equal to the quotient of the scalar of the corresponding position vector of each position vector in the path boundary information of the vehicle on the first reference direction line and the scalar of the corresponding position vector in the corresponding associated boundary feature information on the first reference direction line, and when the scalar of the corresponding position vector in the corresponding associated boundary feature information on the first reference direction line is 0, it is determined that the corresponding mapping calibration coefficient of the corresponding position vector of each position vector in the path boundary information on the first reference direction line is 1; The bias mapping calibration coefficient represents the mapping calibration coefficient corresponding to the corresponding position vector of each element in the obtained vehicle monitoring information on the second reference direction line corresponding to the corresponding vehicle; The corresponding bias mapping calibration coefficient of the target vehicle and its associated vehicles is equal to the average of the corresponding mapping calibration coefficients of the corresponding position vectors of each position vector in the path boundary information of the corresponding vehicle on the second reference direction line, the corresponding mapping calibration coefficient of the corresponding position vector of each position vector in the path boundary information of the vehicle on the second reference direction line is equal to the quotient of the scalar of the corresponding position vector of each position vector in the path boundary information of the vehicle on the second reference direction line and the scalar of the corresponding position vector in the corresponding associated boundary feature information on the second reference direction line, and when the scalar of the corresponding position vector in the corresponding associated boundary feature information on the second reference direction line is 0, it is determined that the corresponding mapping calibration coefficient of the corresponding position vector of each position vector in the path boundary information on the second reference direction line is 1.

5. The high-definition map and fusion perception based passable area analysis method of claim 1, wherein: When analyzing the congestion state corresponding to each road node in a preset second radius area around the current position of the target vehicle at different time points, Obtain the historical road traffic information in a preset second radius area around the current position of the target vehicle, and construct a set of road nodes in the preset second radius area around the current position of the target vehicle, denoted as road node set; The historical road traffic information includes traffic feature information corresponding to each road node in the corresponding area at different preset time intervals in a day, the traffic feature information includes traffic states corresponding to each preset position area in the corresponding road node, and the traffic state includes normal traffic state and traffic congestion state; The state of the construction area existing in the road node or the obstacle occupied area is also recorded as the traffic congestion state; The road node corresponds to a preset road area in the database, different preset road areas corresponding to different road nodes are different, and the areas corresponding to each road area are equal; The preset time interval in a day is multiple and different, and the interval length corresponding to each different preset time interval is equal; Obtain the traffic feature information corresponding to each day and different preset time interval in the historical road traffic information corresponding to the kth element in the road node set, A set of traffic feature information corresponding to the jth preset time interval in different dates in the historical road traffic information corresponding to the kth element in the statistical road node set is denoted as a congestion state analysis set YDF of the kth element in the statistical road node set in the jth preset time interval. （k，j） ; counting the frequency of the passing jam state in the element of the YDF （k，j） corresponding to each preset location area in the kth element in the statistical road node set, and marking all preset location areas whose ratio of the obtained frequency to the total number of elements in the YDF （k，j） is greater than a preset frequency; marking each location point in the marked preset location area; If all the marked preset location areas in the congestion state corresponding to the kth element of the road node set at the jth preset time interval in a day are in traffic congestion state, and all the unmarked preset location areas are in normal traffic state, the traffic state of the road node corresponding to the kth element of the road node set is determined as traffic congestion state.

6. The passable area analysis method based on a high-definition map and fusion perception according to claim 5, characterized in that: When analyzing the congestion correlation influence between different road nodes, The historical road traffic information in a preset second radius area around the current position of the target vehicle is obtained, and any element in the road node set is recorded as a reference node. The congestion time interval corresponding to each occurrence of the congestion state of the reference node in the obtained historical road traffic information is recorded as a first time interval, and a set of all non-reference nodes whose corresponding congestion time interval has a non-empty intersection with the first time interval is recorded as a congestion correlation subset corresponding to the congestion state of the reference node. The set of road nodes in the road node set that have a congestion correlation influence with the reference node is the union set of each congestion correlation subset corresponding to the congestion state of the reference node.

7. The passable area analysis method based on a high-definition map and fusion perception according to claim 6, characterized in that: The associated road nodes corresponding to the road node to which the target vehicle's current position belongs are all road nodes that have a congestion correlation influence with the road node to which the target vehicle's current position belongs. When predicting the traffic congestion risk area of the road node to which the target vehicle's current position belongs, the traffic state of the associated road nodes in the road monitoring image at the current time is obtained. If the traffic state of any one of the obtained associated road nodes is in traffic congestion state, it is determined that all the obtained associated road nodes are congestion risk nodes; if the traffic state of the obtained associated road nodes is in normal traffic state, it is determined that all the obtained associated road nodes are not congestion risk nodes. The time interval to which the current time belongs is recorded as T. The congestion state analysis set corresponding to each congestion risk node in the time interval T is obtained, and the set of each marked preset location area in the obtained each congestion state analysis set is taken as the traffic congestion risk area of the road node to which the target vehicle's current position belongs.

Images