Port wharf shore crane congestion condition sensing method and device, computer equipment and medium

Abstract

This application relates to a method, apparatus, computer equipment, and medium for sensing the congestion status of quay cranes at port terminals. The method includes: detecting whether sensing conditions for quay crane congestion status are met; the sensing conditions are that the quay crane congestion route change function is enabled and the vehicle has trailer information; when the sensing conditions are met, determining the nearest left-turn and right-turn segments to the vehicle; detecting collision risk areas in the left-turn and right-turn segments, denoted as a first risk area and a second risk area; detecting whether quay cranes exist in the first and second risk areas; when quay cranes exist in both the first and second risk areas, constructing a route change request message. This application can sense the quay crane congestion status in the terminal environment in real time, accurately identify the quay crane congestion risk in turning segments, and promptly initiate route change requests, improving driving safety and route planning flexibility.

Description

Technical Field

[0001] This application relates to the fields of autonomous driving and route planning technology, and in particular to a method, device, computer equipment and medium for sensing the congestion status of quay cranes at port terminals. Background Technology

[0002] In port and terminal environments, autonomous container transport vehicles typically tow trailers and frequently traverse curves equipped with quay cranes (terminal cranes). When a quay crane obstructs a vehicle's path, failure to promptly avoid or adjust the route can lead to traffic congestion or safety accidents. Existing route planning methods struggle to detect quay crane obstruction in real time and automatically adjust routes, lacking specific handling capabilities tailored to the terminal environment. Summary of the Invention

[0003] In response to the above-mentioned deficiencies or disadvantages, this application provides a method, apparatus, computer equipment, and medium for sensing the congestion status of quay cranes at port terminals.

[0004] This application provides a method for sensing the congestion status of quay cranes at port terminals according to a first aspect, the method comprising: The detection checks whether the sensing conditions for the quay crane blockage are met; the sensing conditions are that the quay crane blockage path change function is enabled and the vehicle has trailer information. When the perception conditions are met, determine the nearest left-turn and right-turn sections to the vehicle; Collision risk areas in left-turn and right-turn sections are identified and designated as the first risk area and the second risk area. Check whether there are quay bridges in the first and second risk areas; When there are quay cranes in both the first and second risk areas, construct a route change request message.

[0005] According to a second aspect, this application provides a port terminal quay crane congestion sensing device, the device comprising: The perception condition detection module is used to detect whether the perception conditions for the quay crane blockage are met; the perception conditions are that the quay crane blockage path change function is enabled and the vehicle has trailer information. The turning segment determination module is used to determine the nearest left-turn and right-turn segments to the vehicle when the perception conditions are met. The risk area detection module is used to detect collision risk areas in left-turn and right-turn sections, denoted as the first risk area and the second risk area. The quay crane has a detection module used to detect whether a quay crane exists in the first risk area and the second risk area; The route change triggering module is used to construct and send a route change request message when there are quay bridges in both the first risk area and the second risk area.

[0006] According to a third aspect, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the port terminal quay crane congestion sensing methods described in the above embodiments.

[0007] According to a fourth aspect, this application provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executed, implements any of the port terminal quay crane congestion sensing methods described in the above embodiments.

[0008] This application enables real-time perception of quay crane congestion in the terminal environment, accurately identifies the risk of quay crane congestion on turning sections, and promptly initiates route change requests, improving the driving safety and route planning flexibility of autonomous vehicles in complex terminal environments. First, by detecting perception conditions (quay crane congestion route change function enabled and vehicle possessing trailer information), it ensures that this solution is only activated in terminal scenarios with trailers, avoiding false triggering. When the perception conditions are met, the nearest left and right turning sections are determined, ensuring that the perceived congestion is directly related to the vehicle's current driving decision and avoiding invalid calculations. Subsequently, collision risk areas in the left and right turning sections are detected separately, and it is determined whether a quay crane exists in each risk area. Only when a quay crane exists in both risk areas is a route change request message constructed. The above technical logic can accurately identify emergency situations where the vehicle cannot safely pass through any turning direction. Therefore, the vehicle can promptly trigger a route change request, avoiding traffic congestion or safety accidents caused by congestion. Attached Figure Description

[0009] Figure 1 This is a flowchart of a port terminal quay crane congestion sensing method according to one or more embodiments of this application; Figure 2 This is a schematic diagram of the port terminal quay crane blockage sensing device in one or more embodiments of this application; Figure 3 This is a schematic diagram of the internal structure of a computer device according to one or more embodiments of this application. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings. It should be understood that the described embodiments are merely some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0011] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0012] In the description of this application, it should be understood that the terms "first," "second," "third," etc., are used only to distinguish similar objects and are not necessarily used to describe a specific order or sequence, nor should they be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0013] To address the shortcomings or defects of related technologies, this application provides a method for perceiving the congestion status of quay cranes at port terminals. This method can perceive the congestion status of quay cranes in the terminal environment in real time, accurately identify the risk of quay crane congestion in turning sections, and promptly initiate route change requests, thereby improving the driving safety and route planning flexibility of autonomous vehicles in complex terminal environments.

[0014] In some exemplary embodiments of this application, such as Figure 1 As shown, the method includes steps S110 to S150. The following detailed explanation of each step is based on an example of this method applied to the control system (or autonomous driving controller) of an autonomous vehicle (such as an unmanned container truck). The control system is typically integrated into the vehicle and includes a perception module (such as radar, cameras, etc.), a positioning module, a map module, a path planning module, and a message management module.

[0015] S110: Detect whether the sensing conditions for the quay crane blockage are met.

[0016] Quay crane congestion refers to the state in which a quay crane (shoreside container crane) occupies or blocks the travel path of a vehicle (referring to an autonomous container truck). The perception conditions are that the quay crane congestion path changing function is enabled and the vehicle has trailer information. The quay crane congestion path changing function refers to the autonomous driving lateral function configuration, used to enable or disable the quay crane congestion perception logic of this method.

[0017] In this step, the control system reads the on / off status of the quay crane blocking route change function from the route planning data, and checks whether the vehicle currently has trailer information. If the quay crane blocking route change function is enabled and the vehicle has trailer information (indicating that the vehicle is currently towing a trailer), then the perception condition is deemed met, and the control system will continue to execute subsequent steps; if the quay crane blocking route change function is disabled, or the vehicle does not have trailer information (indicating that the vehicle is not currently towing a trailer), then the perception condition is deemed not met, the control system will not execute subsequent steps, and will continue to monitor whether the perception condition is met.

[0018] S120: When the perception conditions are met, determine the nearest left-turn and right-turn sections to the vehicle.

[0019] In this step, the control system can identify the nearest left-turn and right-turn segments along the driving direction to the vehicle's current position based on the vehicle's planned path and map annotation information. A left-turn segment consists of at least one continuous lane segment with left-turn attributes, and a right-turn segment consists of at least one continuous lane segment with right-turn attributes.

[0020] S130: Detect collision risk areas in left-turn and right-turn sections, and label them as the first risk area and the second risk area.

[0021] A collision risk zone refers to a potentially hazardous area where a vehicle and trailer may collide during a turn. This area can be represented by a two-dimensional geometric shape. The first risk zone refers to the collision risk zone in a left-turn section, and the second risk zone refers to the collision risk zone in a right-turn section.

[0022] In this step, the control system calculates the corresponding collision risk zones for the identified left-turn and right-turn sections. The collision risk zone can be calculated based on the size parameters of the vehicle and trailer, the preset front and rear buffer distances, and the geometry of the turning section. Specifically, it is a two-dimensional area that represents the risk range in which the vehicle may collide with the quay crane during the turning process.

[0023] S140: Detect whether there is a quay bridge in the first risk area and the second risk area.

[0024] In this step, the control system can obtain the geometry of all quay cranes in the vehicle's current environment through the perception module, and determine whether the geometry of the first risk area overlaps with that of each quay crane, and whether the geometry of the second risk area overlaps with that of each quay crane. If there is at least one quay crane in the first risk area and at least one quay crane in the second risk area, it is determined that there is a blockage on both the left and right sides.

[0025] S150: When there are quay cranes in both the first and second risk areas, construct a route change request message.

[0026] In this step, if quay cranes are detected in the collision risk areas of both the left and right turning sections, the control system generates a route change request message. This message includes at least a list of blocked lane segments (i.e., the lane segments included in the aforementioned left and right turning sections) and a route change type identifier (such as quay crane blockage). Subsequently, the message is written into a preset message container for the route planning module to read and execute route replanning.

[0027] This embodiment first detects the perception conditions (i.e., the function is enabled and trailer information is available) to ensure that the action is triggered only in scenarios with trailers at the dock, avoiding false alarms. By separately detecting the risk areas in the two nearest left and right turning directions to the vehicle, and only initiating a path change when both directions are blocked by quay cranes, it can accurately identify emergency situations where the vehicle cannot safely pass through any turning direction. Therefore, this embodiment can effectively prevent vehicles from blindly entering blocked turning areas, reducing the risk of traffic congestion and collisions caused by quay crane blockages in port dock environments, and improving the driving safety and path planning flexibility of autonomous vehicles in complex dock environments.

[0028] In some embodiments, determining the nearest left-turn and right-turn segments to the vehicle includes: (1) Obtain the planned route of the vehicle, which includes lane segments arranged in the order of travel.

[0029] A planned path is a collection of lanes constructed by the navigation module based on information such as the map, vehicle origin, and task destination. This includes lane segments arranged in the order of vehicle travel. A lane segment is the basic path unit in the map, and each lane segment has a start point, end point, length, curvature, and map label attributes. Map label attributes refer to the pre-labeled turning attributes for each lane segment in the map data, such as "left turn," "right turn," and "straight ahead." A continuous lane segment group is a collection of multiple lane segments with the same turning attribute (e.g., all left turns or all right turns) and geographically connected (i.e., the end point of one lane segment coincides with the start point of the next).

[0030] In this step, the control system can read the planned route of the current vehicle from the navigation module. This planned route is stored as an ordered list, with elements arranged in the order the vehicle travels from its current location to the destination. Each element corresponds to a lane segment. The data structure for each lane segment includes its identifier, geometric information, length, and map annotation attributes.

[0031] (2) Traverse each lane segment in the planned path, and based on the map label attributes of the lane segments, select the lane segments with left turn attributes as the first lane segment set and the lane segments with right turn attributes as the second lane segment set.

[0032] In this step, the control system sequentially traverses each lane segment in the planned path list, checks the map label attributes of each lane segment, and for any lane segment, if the label attribute of the lane segment is "left turn", then the lane segment is added to the first lane segment set (i.e., the left turn lane segment set); if the label attribute is "right turn", then the lane segment is added to the second lane segment set (i.e., the right turn lane segment set); if the label attribute is other types (such as straight), then the lane segment is ignored and not added to any set.

[0033] After traversing the path, you will obtain the first set of lane segments (including all lane segments with left-turn attributes in the planned path) and the second set of lane segments (including all lane segments with right-turn attributes). It is important to note that the lane segments in these two sets retain their original order within the planned path.

[0034] (3) Select the group of consecutive lane segments closest to the vehicle’s current position along the driving direction from the first lane segment set as the first lane segment group, and determine the left turn segment closest to the vehicle based on the first lane segment group; Since the lane segments in the first set of lane segments may be discontinuous on the planned path (for example, there may be a straight lane segment between two left-turn lane segments), it is necessary to extract continuous lane segment groups from them. Specifically, from the first set of lane segments, the left-turn lane segment closest to the vehicle's current position can be located according to the vehicle's travel direction (i.e., the original order of the elements in the set). Then, starting from this lane segment, check whether the subsequent lane segments also belong to the first set of lane segments and are geographically connected to the current lane segment. If so, these continuous lane segments are merged into a continuous lane segment group. Finally, this continuous lane segment group is taken as the first lane segment group, and the actual road segment corresponding to the first lane segment group is the left-turn road segment closest to the vehicle.

[0035] For example, if the planned path passes through lane segments A (straight), B (left turn), C (left turn), D (straight), and E (left turn) in sequence, then the first set of lane segments includes B, C, and E. The left-turn lane segment closest to the vehicle is B. Checking its successor lane segment C, it is also a left turn and continuous. Therefore, the first set of lane segments is {B, C}, and the corresponding left-turn segment is composed of B and C. Although E is also a left turn, it is not continuous with B and C, so it does not belong to the left-turn segment closest to the vehicle.

[0036] (4) Select the group of continuous lane segments that are closest to the current position of the vehicle along the direction of travel from the set of second lane segments as the second lane segment group, and determine the right turn segment closest to the vehicle based on the second lane segment group.

[0037] The operation process in this step is the same as in step (3), that is, locate the right-turn lane segment closest to the vehicle's current position along the driving direction, and then extract the right-turn lane segments that are continuous with this lane segment to form a continuous lane segment group, which is the second lane segment group. The road segment corresponding to this second lane segment group is the right-turn road segment closest to the vehicle.

[0038] This embodiment traverses the planned path and filters left-turn and right-turn lane segments based on map annotation attributes, then extracts continuous lane segment groups. This accurately identifies the closest and complete turning segment to the vehicle's current position, avoiding misjudging distant or discontinuous turning segments as the current target for detection, thereby improving the targeting of subsequent congestion detection. Since the planned path within a port terminal may include multiple alternating left-turn, right-turn, and straight-ahead lane segments, this embodiment uses a "continuous lane segment group" merging mechanism to treat multiple consecutive turning lane segments in the same direction as a single turning segment. This avoids splitting the same continuous turning arc segment into multiple fragmented segments for processing, improving the rationality of the perception logic and computational efficiency.

[0039] In some embodiments, detecting collision risk areas in left-turn and right-turn sections includes: (1) Obtain the size parameters of the vehicle and its towed trailer, the pre-set front and rear buffer distances, and the turning directions of the left and right turning sections; Dimensional parameters refer to the physical measurement data of the vehicle and its towed trailer, including but not limited to the length and width of the vehicle, the length and width of the trailer, the distance from the front of the vehicle to the cab, the distance from the front of the trailer to the cab, and the effective length from the rear of the trailer to the cab. These parameters can be obtained through actual measurement and pre-stored as configuration information. Front and rear buffer distances are manually determined safety redundancy distances based on the daily driving speed of vehicles in a dock environment. They include a reserved safety distance in front of the vehicle (to avoid collisions between the vehicle's front and obstacles) and a buffer length behind the vehicle (to avoid collisions between the trailer's rear and obstacles). This distance can be configured and adjusted according to scenario requirements. Turning direction refers to the turning type of the target turning segment, including left and right turns. The turning direction affects the lateral offset direction of the collision risk area. Spatial geometric information is data describing the geometric characteristics of the target turning segment in space, including at least the starting position, ending position, and center point position of the target turning segment. This positional information is typically obtained by projecting the path mileage (i.e., the S-value) onto the vehicle's centerline driving path curve. Shape parameters are quantitative indicators used to define the geometry of the collision risk area, mainly including the length of the area along the path and the width of the area perpendicular to the path. Position parameters are data used to determine the spatial placement of the collision risk area, mainly referring to the coordinates of the center point of the area. As mentioned above, the collision risk area is a spatial range where a collision may occur during a vehicle's turning process, taking into account the actual contours of the vehicle and trailer and the need for cushioning.

[0040] In this step, the control system can read the size parameters of the vehicle and its towed trailer and the preset front and rear buffer distances from the configuration file, as well as obtain the turning direction attributes (i.e., left turn or right turn) of the left turn and right turn segments from the route planning information.

[0041] (2) Taking the left-turn section and the right-turn section as the target turning sections respectively, perform the following operations: a. Extract the spatial geometric information of the target turning segment based on its start and end position information; Based on the start and end position information of the target turning segment (e.g., the starting mileage value s_start and ending mileage value s_end of the segment on the path), the corresponding spatial coordinates are extracted from the centerline curve of the vehicle's planned path. Specifically, this includes the following operations: Projecting s_start onto the centerline curve yields the coordinates of the starting point of the turning segment; Projecting s_end onto the centerline curve yields the coordinates of the endpoint position; Calculate s_center=(s_start+s_end) / 2, and project s_center onto the centerline curve to obtain the coordinates of the center point of the turning section.

[0042] These three locations together constitute the spatial geometric information of the target turning section.

[0043] b. Determine the shape and location parameters of the collision risk area based on spatial geometry information, the size parameters of the vehicle and its towed trailer, the front and rear buffer distances, and the turning direction of the target turning section; In this step, the control system calculates the shape and location parameters of the collision risk area based on spatial geometric information, combined with the size parameters of the vehicle and trailer, the front and rear buffer distances, and the turning direction of the target turning section.

[0044] The operation of determining shape parameters includes: calculating the length (i.e., length dimension) and width (i.e., width dimension) of the risk area along the path direction based on the size parameters of the vehicle and trailer, as well as the front and rear buffer distances. The length dimension is typically related to the front and rear buffer distances of the vehicle and the total length of the vehicle and trailer; the width dimension is typically related to the effective length from the front of the vehicle to the tractor unit and from the rear of the trailer to the tractor unit.

[0045] The process of determining the location parameters includes: using the center point of the turning segment as a reference, and based on the turning direction (left or right turn) and shape parameters, making corresponding offsets along the path direction (i.e., the tangential direction of the turn) and perpendicular to the path direction (i.e., the normal direction) to ultimately determine the coordinates of the center point of the risk area. Under different turning directions, the offset direction of the risk area's center point in the normal direction is opposite (offset towards the inside of the curve when turning left, and offset towards the outside of the curve when turning right).

[0046] c. Generate the collision risk zone corresponding to the target turning segment based on the shape and position parameters.

[0047] In this step, a geometric shape is constructed in a spatial coordinate system based on shape and position parameters to represent the collision risk area. This geometric shape can be a two-dimensional rectangle, with the center of the rectangle being the aforementioned center point coordinates. The longer side of the rectangle follows the path direction (length is the length dimension), and the shorter side follows a direction perpendicular to the path (width is the width dimension). The direction of the rectangle is consistent with the direction of the turning segment. This two-dimensional rectangle represents the collision risk area corresponding to the target turning segment.

[0048] Through the above operations, the control system can calculate the first risk area for the corresponding left-turn section and the second risk area for the corresponding right-turn section.

[0049] This embodiment, by introducing the actual size parameters of the vehicle and trailer and a buffer distance pre-set according to the dock conditions, can dynamically generate a collision risk area that matches the current vehicle configuration (especially with a trailer). This avoids the problems of overly lenient detection (leading to missed detections) or overly strict detection (leading to false detections) caused by using fixed-size areas. Furthermore, when determining the location parameters of the risk area, this embodiment clearly distinguishes between left and right turns and adopts different offset strategies in the normal direction accordingly. This ensures that the generated risk area accurately reflects the actual risk distribution caused by the difference between the inner and outer wheels of the trailer when the vehicle turns left or right, improving the geometric rationality of the blockage detection.

[0050] In some embodiments, the spatial geometric information of the target turning segment includes the center point, start point, and end point of the target turning segment, determined based on the start and end point information of the target turning segment. The center point, start point, and end point are three spatial coordinate points projected onto the vehicle's centerline travel path curve based on the starting and ending path mileage (s-value) of the turning segment. Specifically, the center point corresponds to the midpoint of the turning segment, the start point corresponds to the beginning of the turning segment, and the end point corresponds to the end of the turning segment.

[0051] Based on this, the shape and location parameters of the collision risk area are determined according to spatial geometric information, the size parameters of the vehicle and its towed trailer, the front and rear buffer distances, and the turning direction of the target turning section, including: (1) Determine the lateral direction vector and normal direction vector of the target turning section based on the center point position, the starting point position and the ending point position.

[0052] The lateral direction vector is a unit vector pointing from the center point of the turn to the end point of the turn. This vector represents the vehicle's direction of travel on the turning segment (i.e., the tangential direction of the turning path) and is used to determine the orientation of the collision risk area. The normal direction vector is a unit vector obtained by rotating the lateral direction vector 90 degrees clockwise. This vector is perpendicular to the lateral direction vector and represents the vehicle's width direction (i.e., the lateral direction inside or outside the turning segment). It is used to determine the offset of the collision risk area in the normal direction.

[0053] In this step, the control system first obtains the center point position (denoted as turn_center_pos), the starting point position (denoted as turn_start_pos), and the ending point position (denoted as turn_end_pos) of the target turning segment, and then performs the following operations: a. Calculate the vector (denoted as temp_vector) pointing from the center point to the endpoint using the following formula: temp_vector=turn_end_pos-turn_center_pos.

[0054] b. Normalize the vector temp_vector to obtain the lateral direction vector (denoted as cross_dir). The length of the lateral direction vector cross_dir is 1, and its direction is consistent with the tangent direction of the turning segment at the center point (pointing in the direction of vehicle movement).

[0055] c. Rotate the lateral direction vector `cross_dir` 90 degrees clockwise to obtain the normal direction vector (denoted as `cross_dir_norm`). The specific steps for rotating 90 degrees clockwise are: if the lateral direction vector is (x, y), the rotated vector is (y, -x). Then, normalize this vector to obtain the unit normal direction vector. At this point, the normal direction vector is perpendicular to the lateral direction and points to the right of the vehicle (in the standard coordinate system, a 90-degree clockwise rotation is to the right).

[0056] (2) Determine the length and width of the collision risk zone based on the size parameters of the vehicle and its towed trailer and the front and rear buffer distances; The length dimension refers to the extension length of the collision risk area along the path direction (i.e., the direction indicated by the lateral direction vector), which is determined by the safety distance reserved in front of the vehicle and the buffer length behind the vehicle. The width dimension refers to the extension width of the collision risk area along the normal direction (i.e., the direction indicated by the normal direction vector), which is determined by the distance from the front of the vehicle to the tractor and the effective length from the rear of the trailer to the tractor.

[0057] In this step, the control system reads the pre-stored dimensions of the vehicle and trailer, as well as the front and rear buffer distance parameters, and calculates the length and width of the collision risk area based on the read data.

[0058] The width dimension (denoted as risky_width) can be calculated using the following formula: risky_width=outward_risky_buffer+inward_risky_buffer.

[0059] Wherein, outward_risky_buffer refers to the distance from the front of the vehicle to the front of the trailer, representing the risk width extending outward in front of the vehicle before turning; inward_risky_buffer refers to the effective length from the rear of the trailer to the front of the trailer, representing the risk width extending inward in front of the trailer before turning.

[0060] The length dimension (denoted as risky_length) can be calculated using the following formula: risky_length=forward_risky_buffer+backward_risky_buffer.

[0061] Among them, forward_risky_buffer refers to the safe distance reserved in front of the vehicle after turning, which can be determined according to the operation scenario and vehicle speed; backward_risky_buffer is composed of half the vehicle width plus the absolute value of the distance along the normal direction between the starting point of the turn and the center point of the turn, which can be obtained through the following operation: backward_risky_buffer=GetAdvWidth()*0.5+|cross_dir_norm.CrossProd(turn_start_pos-turn_center_pos)|.

[0062] CrossProd represents the cross product of two vectors (in two-dimensional space, the absolute value of the cross product is equal to the area of ​​the parallelogram formed by the two vectors).

[0063] (3) Determine the center position of the collision risk area based on the center point position, lateral direction vector, normal direction vector, length dimension, width dimension, and turning direction of the target turning section; The center position refers to the coordinates of the geometric center point of the collision risk area, which can be calculated using a specific offset formula based on the turning center point, direction vector, and size parameters.

[0064] In this step, the control system determines the center point coordinates (risky_area_center) of the risky area based on the turning center point position, lateral direction vector, normal direction vector, length dimension, width dimension, and the turning direction of the target turning segment through the following two-step offset calculation: Step 1: Offset along the path direction (lateral direction). First, calculate the temporary center point (denoted as risky_area_center_temp) using the following formula: risky_area_center_temp=turn_center_pos+cross_dir*(forward_risky_buffer-backward_risky_buffer)*0.5.

[0065] The above formula is used to adjust the center of the risk area along the driving direction, so that the risk area is centered relative to the turning center point in the path direction (i.e. the front buffer zone and the rear buffer zone are symmetrically distributed on both sides of the center).

[0066] Step 2: Offset along the normal direction (horizontal perpendicular direction).

[0067] Specifically, the control system can further adjust the temporary center point based on the turning direction, as follows: a. If the target turning segment is a left turn, adjust the temporary center point based on the following formula: risky_area_center=risky_area_center_temp+cross_dir_norm*(inward_risky_buffer-outward_risky_buffer)*0.5.

[0068] When turning left, vehicles and trailers tend to deviate towards the inside of the curve (i.e., the positive or negative direction of the normal, depending on the coordinate definition). The above formula can move the center of the risk area inward to cover the risk range of the trailer swinging inward.

[0069] b. If the target turning segment is a right turn, adjust the temporary center point based on the following formula: risky_area_center=risky_area_center_temp+cross_dir_norm*(-inward_risky_buffer+outward_risky_buffer)*0.5.

[0070] When turning right, the center of the risk area shifts outward to cover the risk range of the vehicle's front end swinging outward.

[0071] After the above two offset steps are completed, the final geometric center coordinates of the collision risk area, risky_area_center, are obtained.

[0072] (4) The length, width and center position are used as the shape and position parameters of the collision risk area.

[0073] In this step, the control system uses the length (risky_length) and width (risky_width) as shape parameters and the center position (risky_area_center) as position parameters.

[0074] This embodiment defines `outward_risky_buffer` (front risk of the vehicle) and `inward_risky_buffer` (rear risk of the trailer) separately, and performs differentiated offsets based on the turning direction when determining the center position. This accurately simulates the asymmetric risk area caused by the inner and outer wheel differences when a vehicle with a trailer turns left or right. Compared to the conventional method using symmetrical rectangular frames, this scheme improves the consistency between the risk area and the actual collision risk. Furthermore, by using the lateral and normal direction vectors as orthogonal bases and combining them with a simple linear offset formula to calculate the center position of the risk area, the entire process involves only vector addition and subtraction, dot product, and cross product operations, resulting in low computational complexity and completion in milliseconds, meeting the stringent real-time requirements of autonomous driving systems. In addition, the calculation of length and width dimensions depends entirely on the configurable vehicle and trailer size parameters and buffer distance parameters, allowing adaptation to vehicle types with different wheelbases and trailer lengths without modifying the algorithm logic. This parameterized design gives this embodiment good versatility and scalability.

[0075] In some embodiments, generating a collision risk area corresponding to the target turning segment based on shape parameters and position parameters includes: constructing a two-dimensional rectangle with the center position as the center, the lateral direction vector as the direction of the rectangle, the length dimension as the length along the path direction, and the width dimension as the width along the normal direction; and using the two-dimensional rectangle as the collision risk area corresponding to the target turning segment.

[0076] The center position refers to the geometric center coordinates of the collision risk area (risky_area_center), which is the center point of the two-dimensional rectangle. The normal direction is the unit vector obtained by rotating the lateral direction vector 90 degrees clockwise (cross_dir_norm), representing the direction of the short side of the rectangle (i.e., the lateral direction perpendicular to the driving direction).

[0077] In this embodiment, since the lateral direction vector (cross_dir) is a unit vector, it can be directly used to define the x-axis direction in the local coordinate system of the two-dimensional rectangle. Based on this, the control system first uses the lateral direction vector (cross_dir) as the "direction vector" of the two-dimensional rectangle (this vector determines the direction of the long side of the two-dimensional rectangle, that is, the orientation of the two-dimensional rectangle is consistent with the driving direction of the vehicle on the turning section), and uses the center position (risky_area_center) as the geometric center of the two-dimensional rectangle (this center point is located at the intersection of the two diagonals of the two-dimensional rectangle).

[0078] Then, using the rectangle construction function provided by a 2D geometry library (such as the common math::Box2d class), the following parameters are passed to construct a 2D rectangle: Center point coordinates: risky_area_center; Direction vector: cross_dir (defines the local x-axis direction of the rectangle); Half-length: half_length; Half width: half_width.

[0079] The half-length (half_length) is equal to half the length dimension (risky_length), i.e., half_length = risky_length / 2. This value represents the distance of half_length extended from the center point in both the horizontal and reverse directions to the midpoints of the two longer sides of the rectangle. The half-width (half_width) is equal to half the width dimension (risky_width), i.e., half_width = risky_width / 2. This value represents the distance of half_width extended from the center point in both the normal and reverse directions to the midpoints of the two shorter sides of the rectangle.

[0080] The rectangle construction function internally calculates the coordinates of the four corner points of the two-dimensional rectangle based on the received parameters. The specific operation process includes: using the center point coordinates as a reference, moving horizontally by +half_length and -half_length respectively to obtain the midpoints of the two longer sides; then moving horizontally by +half_width and -half_width respectively to obtain the offsets of the four corner points; finally, combining these coordinates to form a complete polygon description of the two-dimensional rectangle. Finally, the control system uses the constructed two-dimensional rectangle as the collision risk area corresponding to the target turning segment. Specifically, for left-turn segments, this two-dimensional rectangle is the first risk area; for right-turn segments, it is the second risk area.

[0081] It should be noted that the construction of the two-dimensional rectangle can be achieved using any equivalent geometric method, such as directly calculating the coordinates of the four corner points and storing them as polygons, or using a parameterized representation of the center point plus half the length, half the width, and the direction. Regardless of the specific implementation, as long as it can accurately represent the spatial range of the rectangular area, it falls within the scope of this embodiment.

[0082] This embodiment simplifies the complex vehicle turning collision risk area into a two-dimensional rectangle, requiring only four parameters: center point, direction vector, length, and width. This concise geometric representation significantly reduces the computational complexity of subsequent overlap detection (intersection judgment between rectangles and polygons). Compared to using irregular polygons or sector regions, the computational speed is significantly improved, meeting the high real-time requirements of autonomous driving systems. Furthermore, the orientation of the rectangle is dynamically determined by the lateral direction vector, allowing it to adaptively rotate according to the actual direction of the turning segment. Regardless of the turning radius or arc length, the rectangle is always aligned with the path direction, ensuring that the risk area always covers the buffer zone in front of and behind the vehicle's direction of travel, avoiding the coverage deviation problem that occurs with fixed-orientation rectangles on curved paths.

[0083] In some embodiments, detecting whether a quay bridge exists in a first risk area and a second risk area includes: (1) Detect the quay cranes in the current environment of the vehicle and obtain the geometry of each quay crane; The quay cranes are identified and reported by the perception module using sensors such as radar and cameras. The geometric shape refers to the polygonal outline output by the perception module for each quay crane obstacle, composed of a set of ordered two-dimensional coordinate points, accurately describing the area occupied by the quay crane in space. The first detection result is a comprehensive judgment obtained after overlapping detection of the first risk area with each quay crane, indicating whether at least one quay crane exists within the first risk area. The second detection result is a comprehensive judgment obtained after overlapping detection of the second risk area with each quay crane, indicating whether at least one quay crane exists within the second risk area.

[0084] In this step, the control system acquires obstacle information around the vehicle in real time through the perception module. The perception module outputs an obstacle list, and each obstacle in the list includes the following key fields: obstacle type (such as "shore bridge", "other vehicles", "pedestrian", "cone", etc.), geometry (i.e., polygon outline), position coordinates, speed, etc.

[0085] The control system iterates through the obstacle list, filtering by the obstacle type field and retaining only obstacles of type "quay crane" (or "crane"). For each selected quay crane, its geometric shape information is extracted from the obstacle data structure. This geometry is stored in the form of polygons, where the vertex coordinates are a set of two-dimensional points in the world coordinate system, arranged in clockwise or counterclockwise order, completely covering the quay crane's projection area on the ground.

[0086] (2) Detect whether the geometry of the first risk area and each quay bridge overlaps, obtain the first detection result, and determine whether there is a quay bridge in the first risk area based on the first detection result; In this step, the control system acquires the first risk area, namely the two-dimensional rectangle corresponding to the left-turn section. For each acquired quay bridge geometry, the control system performs an overlap detection between the two-dimensional rectangle and the geometry. For example, the detection operation may include: first determining whether any vertex of the two-dimensional rectangle is inside the geometry, or whether any vertex of the geometry is inside the two-dimensional rectangle, or whether the edges of the two intersect. In other examples, since the two-dimensional rectangle is a convex polygon, the separating axis theorem can also be used for rapid detection.

[0087] The overlap detection returns a Boolean value indicating whether the two-dimensional rectangle and the geometric shape overlap or not. Specifically, if the two-dimensional rectangle and the geometric shape have at least one common point (including the case where the boundary is tangent), it is determined to be "overlapping"; otherwise, it is determined to be "non-overlapping".

[0088] The control system sequentially performs the aforementioned overlap detection on all quay cranes. If at least one quay crane overlaps with the first risk area, the first detection result is "quay crane exists"; if none of the quay cranes overlap with the first risk area, the first detection result is "quay crane does not exist".

[0089] (3) Detect whether the geometry of the second risk area and each quay bridge overlaps, obtain the second detection result, and determine whether there is a quay bridge in the second risk area based on the second detection result.

[0090] In this step, the control system iterates through the geometry of all quay cranes and sequentially determines whether they overlap with the second risk area. If at least one quay crane overlaps with the second risk area, the second detection result is "quay crane exists"; if none of the quay cranes overlap with the second risk area, the second detection result is "quay crane does not exist".

[0091] This embodiment directly uses the actual polygonal geometry of the quay crane to determine the intersection with the rectangular bounding box of the risk area, rather than simply simplifying the quay crane to a point or circle. This precise geometric detection method avoids misjudgments or omissions caused by approximate obstacle outlines, accurately identifying whether the quay crane has truly intruded into the risk area, thus improving the reliability of the blockage determination. Furthermore, this embodiment independently detects the first and second risk areas, outputting two independent detection results. This design allows the control system to clearly distinguish between three different scenarios: "blockage on the left only," "blockage on the right only," and "blockage on both sides," providing a refined basis for subsequent decision-making regarding whether to initiate a route change request.

[0092] In some embodiments, before detecting collision risk areas in left-turn and right-turn sections, the method further includes: (1) Obtain the first label information related to the first lane segment group and the second label information related to the second lane segment group from the map lane data.

[0093] The first annotation information is the attribute information pre-stored in the map data for each lane segment within the first lane segment group. It includes at least the turning attribute (e.g., left turn), arc continuity indicator (whether it belongs to a complete turning arc), and geometric parameters (e.g., length, curvature). The second annotation information is similar attribute information pre-stored in the map data for each lane segment within the second lane segment group, used to describe the characteristics of the right turn segment.

[0094] In this step, the control system requests map annotation information for each lane segment from the map module based on the first lane segment group (a set of consecutive lane segments corresponding to left-turn segments) and the second lane segment group (a set of consecutive lane segments corresponding to right-turn segments). The map module returns detailed attributes for each lane segment, including but not limited to: the lane segment's turning attribute (left turn, right turn, straight), whether it belongs to a complete turning arc (e.g., a pre-marked "turning arc ID" on the map), the lane segment's geometric length (in meters), and the mileage values ​​of the starting and ending points on the path's centerline. This ultimately yields the first and second annotation information.

[0095] (2) Detect whether the left-turn section is a complete turning arc based on the first annotation information and whether the difference between the starting and ending mileage of the left-turn section is greater than the preset mileage threshold.

[0096] A complete turning arc is an arc segment formed by lane segments marked on the map as continuous turns in the same direction (such as continuous left turns) that geometrically constitute a complete turning action. For example, if multiple left-turn lane segments are connected end to end and the curvature changes continuously, they are considered a complete turning arc segment; if they are interrupted by a straight-ahead lane segment or the turning direction changes, they are not considered a complete turning arc segment.

[0097] The difference between the start and end path mileage refers to the difference between the end path mileage (turn_end_s) and the start path mileage (turn_start_s) of a turning segment, which is the actual length of the turning segment along the centerline of the path. This difference reflects the spatial span of the turning segment.

[0098] The preset mileage threshold is a pre-defined length value (e.g., 5 meters, 10 meters, etc.) used to determine whether a turning section is long enough to warrant collision risk area detection. If the turning section is too short, its actual collision risk is negligible, and subsequent detection is unnecessary to save computational resources.

[0099] In this step, the control system analyzes the first annotation information and performs two checks: a. Integrity check: This involves checking whether all lane segments in the first lane segment group are marked on the map as belonging to the same complete turning arc. Specifically, this is done by checking if each lane segment has the same "turning arc ID," or by checking if the turning attribute of consecutive lane segments is always left turn and if there are no abrupt changes in turning direction between adjacent lane segments. If the lane segments in the first lane segment group can form a geometrically continuous, uninterrupted left-turn arc, it is considered a "complete turning arc"; otherwise, it is considered "incomplete."

[0100] b. Length detection, i.e., calculating the difference in mileage between the start and end points of the left-turn segment. The control system obtains the starting mileage (turn_start_s) of the first lane segment and the ending mileage (turn_end_s) of the last lane segment in the first lane segment group, and calculates the difference (denoted as delta_s_left) using the following formula: delta_s_left = turn_end_s - turn_start_s.

[0101] Then, the difference is compared with the system's preset mileage threshold (e.g., THRESHOLD_LENGTH = 5.0 meters). If delta_s_left > THRESHOLD_LENGTH, the length is determined to meet the requirement; otherwise, it is determined not to meet the requirement.

[0102] (3) Detect whether the right turn segment is a complete turning arc segment based on the second annotation information and whether the difference between the starting and ending path mileage of the right turn segment is greater than the preset mileage threshold. The operation process of this step is as follows: check whether the lane segment in the second lane segment group constitutes a complete right turn arc; calculate the difference in mileage between the start and end paths of the right turn segment, delta_s_right = turn_end_s - turn_start_s, and determine whether it is greater than the preset mileage threshold.

[0103] (4) When both the left-turn and right-turn segments can form a complete turning arc, and the difference in the starting and ending mileage of the left-turn and right-turn segments is greater than the preset mileage threshold, it is determined to perform the step of detecting the collision risk area in the left-turn and right-turn segments.

[0104] In this step, if the left-turn segment is determined to be a "complete turning arc" and delta_s_left > preset mileage threshold, and the right-turn segment is determined to be a "complete turning arc" and delta_s_right > preset mileage threshold, then the execution conditions are met, and step S103 is continued, that is, collision risk areas in the left-turn and right-turn segments are detected.

[0105] If any of the above conditions are not met (e.g., the left-turn segment is incomplete, or the right-turn segment is not long enough), the control system will skip the subsequent collision risk area detection and path change request construction, and directly end the process without performing any path change operation.

[0106] This embodiment effectively eliminates situations where there is no actual collision risk due to map labeling errors, abnormal lane segment splicing, or insufficient turning radius by detecting whether the turning segment constitutes a complete turning arc and whether its length is sufficient. This pre-filtering mechanism avoids invalid calculations and improves the robustness and reliability of the method. Furthermore, for turning segments that do not meet the completeness or length conditions, subsequent collision risk area calculations and quay bridge overlap detection are skipped directly, thereby reducing unnecessary computational resource consumption and ensuring the real-time performance of the autonomous driving system.

[0107] It should be noted that, unless otherwise explicitly stated herein, the execution order of the various steps included in the port terminal quay crane congestion sensing method provided in any of the above embodiments is not strictly limited, and these steps can be executed in other orders. Moreover, at least some of these steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0108] Based on the same inventive concept, this application also provides a port terminal quay crane congestion sensing device. In some embodiments, such as Figure 2 As shown, the port terminal quay crane congestion sensing device includes the following modules: The perception condition detection module 110 is used to detect whether the perception conditions for the quay crane blockage are met; the perception conditions are that the quay crane blockage path changing function is enabled and the vehicle has trailer information. The turning segment determination module 120 is used to determine the nearest left-turn and right-turn segments to the vehicle when the perception conditions are met. The risk area detection module 130 is used to detect collision risk areas in left-turn and right-turn road sections, which are denoted as the first risk area and the second risk area. The quay crane presence detection module 140 is used to detect whether a quay crane exists in the first risk area and the second risk area. The route change triggering module 150 is used to construct and send a route change request message when there are quay bridges in both the first risk area and the second risk area.

[0109] In some embodiments, the turning segment determination module 120 is specifically used for: obtaining the planned path of the vehicle, the planned path including lane segments arranged in driving order; traversing each lane segment in the planned path, and filtering out lane segments with left-turn attributes as a first lane segment set and lane segments with right-turn attributes as a second lane segment set according to the map label attributes of the lane segments; selecting the group of consecutive lane segments closest to the current position of the vehicle along the driving direction from the first lane segment set as a first lane segment group, and determining the left-turn segment closest to the vehicle based on the first lane segment group; selecting the group of consecutive lane segments closest to the current position of the vehicle along the driving direction from the second lane segment set as a second lane segment group, and determining the right-turn segment closest to the vehicle based on the second lane segment group.

[0110] In some embodiments, the risk area detection module 130 is specifically used to: acquire the size parameters of the vehicle and its towed trailer, the pre-set front and rear buffer distances, and the turning directions of the left-turn and right-turn sections; and, taking the left-turn and right-turn sections as target turning sections respectively, perform the following operations: extract the spatial geometric information of the target turning sections based on the start and end position information of the target turning sections; determine the shape parameters and position parameters of the collision risk area based on the spatial geometric information, the size parameters of the vehicle and its towed trailer, the front and rear buffer distances, and the turning directions of the target turning sections; and generate the collision risk area corresponding to the target turning section based on the shape parameters and position parameters.

[0111] In some embodiments, the spatial geometric information of the target turning segment includes the center point position, start position, and end position of the target turning segment determined based on the start and end position information of the target turning segment. The risk area detection module 130 is further configured to: determine the lateral direction vector and normal direction vector of the target turning segment based on the center point position, start position, and end position; the lateral direction vector is a unit vector pointing from the center point to the end point, and the normal direction vector is a unit vector obtained by rotating the lateral direction vector 90 degrees clockwise; determine the length and width dimensions of the collision risk area based on the size parameters of the vehicle and its towed trailer, as well as the front and rear buffer distances; determine the center position of the collision risk area based on the center point position, lateral direction vector, normal direction vector, length dimension, width dimension, and the turning direction of the target turning segment; and use the length dimension, width dimension, and center position as the shape and position parameters of the collision risk area.

[0112] In some embodiments, the risk area detection module 130 is further configured to: construct a two-dimensional rectangle with the center position as the center, the lateral direction vector as the direction of the rectangle, the length dimension as the length along the path direction, and the width dimension as the width along the normal direction; and use the two-dimensional rectangle as the collision risk area corresponding to the target turning section.

[0113] In some embodiments, the quay crane detection module 140 is specifically used to: detect quay cranes in the environment where the vehicle is currently located, and obtain the geometry of each quay crane; detect whether the geometry of the first risk area and each quay crane overlaps, obtain a first detection result, and determine whether there is a quay crane in the first risk area based on the first detection result; detect whether the geometry of the second risk area and each quay crane overlaps, obtain a second detection result, and determine whether there is a quay crane in the second risk area based on the second detection result.

[0114] In some embodiments, the device further includes a turning segment validity detection module. The turning segment validity detection module is configured to: obtain first annotation information related to a first lane segment group and second annotation information related to a second lane segment group from map lane data; detect whether a left-turn segment is a complete turning arc and whether the difference in mileage between the start and end paths of the left-turn segment is greater than a preset mileage threshold based on the first annotation information; detect whether a right-turn segment is a complete turning arc and whether the difference in mileage between the start and end paths of the right-turn segment is greater than a preset mileage threshold based on the second annotation information; and determine to perform the step of detecting collision risk areas in the left-turn and right-turn segments when both the left-turn and right-turn segments can form complete turning arcs and the difference in mileage between the start and end paths of both the left-turn and right-turn segments is greater than the preset mileage threshold.

[0115] Specific limitations regarding the port terminal quay crane congestion sensing device can be found in the limitations of the port terminal quay crane congestion sensing method described above, and will not be repeated here. Each module in the aforementioned port terminal quay crane congestion sensing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0116] This application also provides a computer device. In some embodiments, the computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it can implement the port terminal quay crane congestion perception method provided in any of the above embodiments.

[0117] In some embodiments, the internal structure diagram of a computer device may be as follows: Figure 3As shown, the computer device includes a processor, memory, and network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores configuration files, planning world information, and other data; specific data stored may also be defined in the above method embodiments. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for sensing the congestion status of port terminal quay cranes.

[0118] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0119] This application also provides a computer-readable storage medium, in some embodiments of which a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements the port terminal quay crane congestion perception method provided in any of the above embodiments.

[0120] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0121] Those skilled in the art will understand that implementing all or part of the processes in the above method embodiments can be accomplished by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchlink, DRAM (SLDRAM), memory bus, direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0122] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0123] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for sensing the congestion status of quay cranes at port terminals, characterized in that, The method includes: The detection checks whether the sensing conditions for the quay crane blockage are met; the sensing conditions are that the quay crane blockage path changing function is enabled and the vehicle has trailer information. When the perception conditions are met, determine the nearest left-turn and right-turn road segments to the vehicle; The collision risk areas in the left-turn and right-turn sections are detected and denoted as the first risk area and the second risk area. Detect whether there are quay bridges in the first risk area and the second risk area; When there are quay bridges in both the first risk area and the second risk area, construct a route change request message.

2. The method according to claim 1, characterized in that, Determining the nearest left-turn and right-turn sections to the vehicle includes: Obtain the planned route of the vehicle, the planned route including lane segments arranged in driving order; Traverse each lane segment in the planned path, and based on the map label attributes of the lane segments, filter out the lane segments with left turn attributes as the first set of lane segments and the lane segments with right turn attributes as the second set of lane segments. Select the group of consecutive lane segments closest to the current position of the vehicle along the direction of travel from the first set of lane segments as the first lane segment group, and determine the left turn segment closest to the vehicle based on the first lane segment group; Select the group of consecutive lane segments closest to the vehicle's current position along the driving direction from the second set of lane segments as the second lane segment group, and determine the right turn segment closest to the vehicle based on the second lane segment group.

3. The method according to claim 2, characterized in that, The detection of collision risk areas in the left-turn and right-turn road sections includes: The size parameters of the vehicle and its towed trailer, the pre-set front and rear buffer distances, and the turning directions of the left-turn and right-turn sections are obtained. Taking the left-turn section and the right-turn section as target turning sections respectively, perform the following operations: The spatial geometric information of the target turning segment is extracted based on the start and end position information of the target turning segment; The shape and location parameters of the collision risk area are determined based on the spatial geometry information, the size parameters of the vehicle and its towed trailer, the front and rear buffer distances, and the turning direction of the target turning section. The collision risk area corresponding to the target turning segment is generated based on the shape parameters and the position parameters.

4. The method according to claim 3, characterized in that, The spatial geometric information of the target turning segment includes the center point position, starting point position, and ending point position of the target turning segment determined based on the starting and ending point position information of the target turning segment; The determination of the shape and location parameters of the collision risk area based on the spatial geometry information, the size parameters of the vehicle and its towed trailer, the front and rear buffer distances, and the turning direction of the target turning section includes: The lateral direction vector and normal direction vector of the target turning segment are determined based on the center point position, the starting point position, and the ending point position; the lateral direction vector is a unit vector pointing from the center point to the ending point, and the normal direction vector is a unit vector obtained by rotating the lateral direction vector 90 degrees clockwise. The length and width of the collision risk zone are determined based on the size parameters of the vehicle and its towed trailer, as well as the front and rear buffer distances. The center position of the collision risk area is determined based on the center point position, the lateral direction vector, the normal direction vector, the length dimension, the width dimension, and the turning direction of the target turning segment. The length, width, and center position are used as the shape and position parameters of the collision risk area.

5. The method according to claim 4, characterized in that, The step of generating the collision risk area corresponding to the target turning segment based on the shape parameters and the position parameters includes: A two-dimensional rectangle is constructed with the center position as the center, the horizontal direction vector as the direction of the rectangle, the length dimension as the length along the path direction, and the width dimension as the width along the normal direction. The two-dimensional rectangle is used as the collision risk area corresponding to the target turning section.

6. The method according to claim 1, characterized in that, The detection of whether a quay bridge exists in the first risk area and the second risk area includes: Detect the quay cranes in the current environment of the vehicle and obtain the geometry of each quay crane; Detect whether the geometry of the first risk area and each of the quay cranes overlaps to obtain a first detection result, and determine whether there is a quay crane in the first risk area based on the first detection result; The second risk area and the geometry of each of the quay cranes are checked for overlap to obtain a second detection result. Based on the second detection result, it is determined whether there is a quay crane in the second risk area.

7. The method according to claim 2, characterized in that, Before detecting collision risk areas in the left-turn and right-turn road segments, the method further includes: Obtain the first annotation information related to the first lane segment group and the second annotation information related to the second lane segment group from the map lane data; Based on the first annotation information, it is detected whether the left-turn segment is a complete turning arc and whether the difference between the start and end path mileage of the left-turn segment is greater than a preset mileage threshold. Based on the second annotation information, it is detected whether the right-turn segment is a complete turning arc and whether the difference between the start and end path mileage of the right-turn segment is greater than a preset mileage threshold. When both the left-turn and right-turn segments can form complete turning arcs, and the difference in mileage between the start and end paths of the left-turn and right-turn segments is greater than a preset mileage threshold, it is determined to execute the step of detecting collision risk areas in the left-turn and right-turn segments.

8. A port terminal quay crane congestion sensing device, characterized in that, The device includes: The sensing condition detection module is used to detect whether the sensing conditions for the quay crane blockage are met; the sensing conditions are that the quay crane blockage path changing function is enabled and the vehicle has trailer information. The turning segment determination module is used to determine the nearest left-turn segment and right-turn segment to the vehicle when the perception conditions are met; The risk area detection module is used to detect collision risk areas in the left-turn section and the right-turn section, which are denoted as the first risk area and the second risk area. The quay crane has a detection module for detecting whether a quay crane exists in the first risk area and the second risk area; The route change triggering module is used to construct and send a route change request message when there are quay bridges in both the first risk area and the second risk area.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 7.

Images