A high-precision container yard truck positioning system and method

Abstract

The application discloses a high-precision container yard truck positioning system and method, and relates to the technical field of vehicle positioning.The system comprises a camera installation end, a vehicle simulation end and a target adjustment end.The camera installation end is used to generate a transformation matrix from camera pixel coordinates to relative world coordinates.The vehicle simulation end is used to establish a simulation environment by using a high-precision container yard map, and to obtain real-time GPS positioning information of a truck.The target adjustment end is used to match, remove duplicates, correct deviations and update positions of detection results of multiple time slices, and to identify and process camera detection abnormalities in a timely manner.The high-precision container yard truck positioning system and method can detect whether a vehicle in a corresponding detection area range has problems such as drift, jitter or data loss in real time, monitor abnormal back-and-forth movement of a gantry crane in real time, and can process line-of-sight blocking problems between gantry cranes in real time in combination with camera detection abnormalities.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle positioning, in particular to a high-precision container yard truck positioning system and method. BACKGROUND

[0002] With the development of global economy and technological progress, ports as an important node of the supply chain are undergoing rapid transformation. The application of automation technology aims to improve the operational efficiency of ports, reduce manpower dependence, reduce operating costs and improve safety. To meet the needs of automated driving test zone construction and automated wharf production operation, realize intelligent scheduling of truck cranes, gantry cranes and other equipment, and super-visual perception of automated driving trucks, safety and operational efficiency need to be improved. Real-time and accurate acquisition of the position and state of the truck is required.

[0003] Currently, to achieve accurate positioning of the truck in the container yard, there are some deficiencies: 1. The Beidou positioning data has the problems of drift, jitter or data loss in the container yard, which cannot be directly used in the production system, resulting in limitations in positioning the truck in the container yard; 2. Since the container yard cannot be erected, the existing sensing equipment can only be installed on the gantry crane. At the same time, the gantry crane moves back and forth for operation, and there is vibration during the movement, which may weaken the accuracy of the truck positioning in the container yard; 3. There are multiple gantry cranes operating in a container yard, and there is occlusion between them, resulting in weak truck positioning effect in the container yard.

[0004] Therefore, the present application provides a high-precision container yard truck positioning system and method to solve the above problems. SUMMARY

[0005] The main purpose of the present application is to provide a high-precision container yard truck positioning system and method to solve the problems raised in the background.

[0006] To achieve the above purpose, the technical solution adopted by the present application is as follows: a high-precision container yard truck positioning system and method, comprising a camera installation end, a vehicle simulation end and a target adjustment end.

[0007] The camera installation end is used to set the truck positioning detection area in the container yard, install a camera on the gantry crane, and set the shape and parameters of the corresponding vehicle on the gantry crane. The camera installation end is used to set the truck positioning detection area in the container yard, install a camera on the gantry crane, and set the shape and parameters of the corresponding vehicle on the gantry crane. Real-time detection of the truck in the corresponding area is performed, and a transformation matrix from the camera pixel coordinates to the relative world coordinates is generated.

[0008] The vehicle simulation end is used for detecting the positions of the truck vehicles in the corresponding area in real time, judging whether the gantry crane equipment operation is normal according to the pixel coordinates of the target vehicle in the image and the transformation matrix, and performing real-time vibration detection on the running state of the gantry crane, and real-time acquisition of the GPS positioning information of the truck vehicle by establishing a simulation environment by using the high-precision yard map.

[0009] The target adjustment end is used for detecting the target credibility according to the effective sensing range of each camera, and using a multi-target matching algorithm to match, remove duplicates, correct deviation and update the position of the detection results of a plurality of continuous time slices to obtain the optimal matching optimized result, and establishing a physical simulation model for each vehicle to identify and process camera detection abnormalities in time.

[0010] The camera installation end comprises a region setting module, a region detection module and a coordinate conversion module.

[0011] The region setting module comprises a region setting unit and a multi-region combination unit.

[0012] The region setting unit is used for setting the truck vehicle positioning detection region in the container yard, and the region corresponding to a plurality of ranges in the region, and three cameras are installed on each gantry crane in the region corresponding to a plurality of ranges, two cameras are installed in the direction of the arriving vehicle of the gantry crane, one of which is a long-focus camera for detecting truck vehicles within a range of 100-250 meters from the gantry crane, and the other is a short-focus camera for detecting truck vehicles within a range of 120 meters from the gantry crane, and one short-focus camera is installed in the direction of the departing vehicle of the gantry crane for detecting truck vehicles within a range of 110 meters from the gantry crane, and the cameras are connected to the network switch on the gantry crane and the video stream is returned to the system center control region server.

[0013] The multi-region combination unit is used for real-time acquisition of the running parameters of the gantry crane in multiple regions and multiple angles, and automatic storage of the parameters by the data logger in real time.

[0014] The region detection module comprises a truck vehicle unit and a vehicle detection unit.

[0015] The truck vehicle unit is used for setting the appearance image model of the gantry crane in the vehicle positioning system.

[0016] The vehicle detection unit is used for real-time capture of the running parameters of the gantry crane by the camera, difference calculation of the captured running parameters of the gantry crane with the standard value, real-time detection of the gantry crane abnormality, and installation of a GPS positioning device on the gantry crane if the difference is not equal to 0.

[0017] The coordinate conversion module comprises a pixel coordinate unit, a transformation matrix unit and a data recording unit.

[0018] The pixel coordinate unit is configured to acquire pixel coordinates (u, v) of the target vehicle in the image in real time through image acquisition;

[0019] The transformation matrix unit is configured to convert the pixel coordinates into world coordinates (x, y, z) by setting a transformation matrix T and in combination with the pixel coordinates (u, v) of the target vehicle in the image;

[0020] The data recording unit is configured to record the calculated data parameters in real time through a data recorder.

[0021] The vehicle simulation end comprises a position detection module, a vibration detection module and a simulation environment module.

[0022] The position detection module comprises a position detection unit and a data receiving unit.

[0023] The position detection unit is configured to position the current position of the gantry crane in real time through a GPS device on the gantry crane, and arrange the positioning parameters of a plurality of gantry cranes in sequence by Arabic numerals, the sequence arrangement being set in ascending order.

[0024] The data receiving unit is configured to receive the positioning parameters of a plurality of gantry cranes, and pixel coordinates and world coordinate parameters of the target vehicle in the image in real time through a data receiver.

[0025] The vibration detection module comprises a vibration detection unit and an abnormality judgment unit.

[0026] The vibration detection unit is configured to detect vibration parameters in the running process of the gantry crane in real time through a calculation formula, and set standard parameters of the movement of the gantry crane.

[0027] The abnormality judgment unit is configured to calculate the difference between the effective value of acceleration and the standard acceleration parameter of the movement of the gantry crane, judge whether the vibration intensity in the movement of the gantry crane is abnormal, if the difference between the effective value of acceleration and the standard acceleration parameter of the movement of the gantry crane is greater than 1, it indicates that the vibration intensity is abnormal, if the difference between the effective value of acceleration and the standard acceleration parameter of the movement of the gantry crane is less than or equal to 1, it indicates that the vibration intensity is normal.

[0028] The simulation environment module comprises a simulation simulation unit and a vehicle positioning unit.

[0029] The simulation simulation unit is configured to set a simulation environment simulation model of the positioning of the container truck in the container area, and project the sensing results of a plurality of camera lenses into the simulation environment simulation model.

[0030] The vehicle positioning unit is used for obtaining high-precision positioning parameters of the vehicle target according to the sensing results of multiple camera heads and combining the real-time positioning of the target vehicle captured by the multiple-area and multi-angle camera heads.

[0031] The target adjustment end comprises a trusted detection module, a target matching module, a multi-effect processing module and a simulation identification module.

[0032] The trusted detection module is used for calculating the trustworthiness of the target vehicle positioning result of the gantry crane at the current time in real time through a calculation formula.

[0033] The target matching module comprises a target matching algorithm and a matching sorting unit.

[0034] The target matching algorithm is used for matching, deduplicating, rectifying and position updating the multiple target vehicle positioning results of the continuous multiple time slices through the multi-target matching algorithm, and outputting the optimal matching optimization result.

[0035] The matching sorting unit is used for arranging the optimal matching optimization result of the multi-target matching algorithm in sequence numbers through Arabic numerals.

[0036] The multi-effect processing module is used for calculating the difference in real time through the multi-target matching algorithm for the matching, deduplication, rectification and position updating of the target vehicle positioning result, and arranging the sequence numbers according to Arabic numerals.

[0037] The simulation identification module comprises a simulation identification unit and a camera detection recording unit.

[0038] The simulation identification unit is used for establishing a physical simulation model for each vehicle in a simulation environment, and identifying and processing the camera detection abnormality in real time through a simulation simulation judgment formula.

[0039] The camera detection recording unit is used for recording the vehicle positioning abnormality in real time through a data recorder, and making a traffic control decision for the target vehicle in real time when the vehicle positioning is abnormal.

[0040] A high-precision container yard truck positioning method comprises the following steps:

[0041] Step 1: configure the IP address information of the vehicle positioning remote control area server.

[0042] Step 2: enter the camera installation end, install the camera on the gantry crane, set the detection area, detect the truck in the corresponding area in real time, and generate a transformation matrix from the camera pixel coordinates to the relative world coordinates;

[0043] Step 3: enter the vehicle simulation end, detect the truck position in the corresponding area in real time, obtain the high-precision positioning data of the gantry crane in real time according to the pixel coordinates of the target vehicle in the image and the transformation matrix, establish a simulation environment using the high-precision map of the container area, and obtain the GPS positioning information of the truck in real time;

[0044] Step 4: enter the target adjustment end, detect the target reliability according to the effective sensing range of each camera in the simulation environment, use a multi-target matching algorithm to match, remove duplicates, correct deviation, and update the position of the detection results of multiple time slices in succession, obtain the optimal matching optimized result, and establish a physical simulation model for each vehicle in the simulation environment, and timely identify and handle camera detection abnormalities.

[0045] The present application has the following beneficial effects:

[0046] 1. In the present application, by setting the camera installation end, when positioning the truck in the high-precision container area, the truck in the corresponding area is detected in real time, and a transformation matrix from the camera pixel coordinates to the relative world coordinates is generated, so that the existence of drift, jitter or data loss in the corresponding detection area can be detected in real time, and the limitations of the truck positioning in the container are reduced.

[0047] 2. In the present application, by setting the vehicle simulation end, when positioning the truck in the high-precision container area, the running state of the gantry crane is detected in real time, the running state of the gantry crane is judged, and the GPS positioning information of the truck is obtained in real time, so that the system can monitor the abnormal operation of the gantry crane moving back and forth, and can adjust or issue a warning reminder in real time when the vibration is abnormal, thereby increasing the accuracy and intelligence of the truck positioning in the container area.

[0048] 3. In the present application, by setting the target adjustment end, when positioning the truck in the high-precision container area, the target reliability is detected according to the effective sensing range of each camera, a multi-target matching algorithm is used to match, remove duplicates, correct deviation, and update the position of the detection results of multiple time slices in succession, obtain the optimal matching optimized result, and timely identify and handle camera detection abnormalities, so that when multiple gantry cranes in the container area are working simultaneously, the line-of-sight blocking problem between the gantry cranes can be handled in real time by combining the camera detection abnormalities, the positioning effect of the truck in the container is guaranteed, and the traffic operation when multiple gantry cranes are working is better. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 A flow architecture diagram of a high-precision container yard truck positioning method of the present application is shown in the figure.

[0050] Figure 2 A camera installation end architecture diagram of a high-precision container yard truck positioning system of the present application is shown in the figure.

[0051] Figure 3 A vehicle simulation end architecture diagram of a high-precision container yard truck positioning system of the present application is shown in the figure.

[0052] Figure 4 A target adjustment end architecture diagram of a high-precision container yard truck positioning system of the present application is shown in the figure. DETAILED DESCRIPTION

[0053] In order to make the technical means, creative features, purposes and effects achieved by the present application easy to understand, the present application will be further described below in conjunction with specific embodiments.

[0054] Embodiment One

[0055] Please refer to Figures 1-2 A high-precision container yard truck positioning system and method, including a camera installation end, a vehicle simulation end, and a target adjustment end.

[0056] The camera installation end is used to set the container yard truck positioning detection area, install cameras on the gantry crane, and set the shape and parameters of the corresponding vehicle, real-time detect the truck in the corresponding area, and generate a transformation matrix from the camera pixel coordinates to the relative world coordinates.

[0057] The vehicle simulation end is used to real-time detect the position of the truck in the corresponding area, according to the pixel coordinates of the target vehicle in the image and the transformation matrix, and real-time detect the running state of the gantry crane, judge whether the gantry crane equipment is running normally, and real-time obtain the GPS positioning information of the truck by establishing a simulation environment using the high-precision map of the yard.

[0058] The target adjustment end is used to detect the target credibility according to the effective perception range of each camera, and use a multi-target matching algorithm to match, remove duplicates, correct deviation, and update the position of the detection results of multiple time slices, obtain the optimal matching optimization result, and establish a physical simulation model for each vehicle to timely identify and process camera detection abnormalities.

[0059] The camera installation end includes a region setting module, a region detection module, and a coordinate conversion module.

[0060] The area setting module comprises an area setting unit and a multi-area combination unit.

[0061] The area setting unit is used for setting a truck positioning detection area in a container area, and a plurality of ranges corresponding to the area, and three cameras are installed on each gantry crane in the plurality of ranges corresponding to the area, two of which are installed in the direction of the approaching vehicle, one of which is a long-focus camera for detecting trucks within a range of 100-250 meters from the gantry crane, and the other is a short-focus camera for detecting trucks within a range of 120 meters from the gantry crane, and one is installed in the direction of the departing vehicle, which is a short-focus camera for detecting trucks within a range of 110 meters from the gantry crane, and the cameras are connected to a network switch on the gantry crane, and the video stream is returned to the system center control area server.

[0062] The multi-area combination unit is used for real-time acquisition of multi-angle gantry crane operation parameters, and automatic storage of the parameters by a data logger.

[0063] The area detection module comprises a truck unit and a vehicle detection unit.

[0064] The truck unit is used for setting an appearance image model of the gantry crane in the vehicle positioning system.

[0065] The vehicle detection unit is used for real-time capture of the operation parameters of the gantry crane by the camera, and difference calculation of the captured operation parameters of the gantry crane with standard values, real-time detection of the abnormality of the gantry crane, if the difference is equal to 0, it indicates that the gantry crane is running normally, if the difference is not equal to 0, it indicates that the gantry crane is running abnormally, and a GPS positioning device is installed on the gantry crane.

[0066] The coordinate conversion module comprises a pixel coordinate unit, a transformation matrix unit and a data recording unit.

[0067] The pixel coordinate unit is used for real-time acquisition of the pixel coordinates (u, v) of the target vehicle in the image by the image.

[0068] The transformation matrix unit is used for setting a transformation matrix T, and combining the pixel coordinates (u, v) of the target vehicle in the image, and converting the pixel coordinates into world coordinates (x, y, z), the method is as follows:

[0069] Step one, extend the pixel coordinates to homogeneous coordinates:

[0070] P_{image}=\begin{bmatrix}u\\ v\\1\\

[0071] Step two, apply the transformation matrix:

[0072] Convert the homogeneous coordinates to world coordinates by the transformation matrix T:

[0073] ;

[0074] The specific calculation is as follows:

[0075] P_{world}=\begin{bmatrix}a&b&c\\

[0076] d&e&f\\

[0077] g&h&i\\

[0078] \end{bmatrix}\begin{bmatrix}

[0079] u\\

[0080] v\\

[0081] 1\\

[0082] \end{bmatrix}=\begin{bmatrix}

[0083] a\cdot u+b\cdot v+c\\

[0084] d\cdot u+e\cdot v+f\\

[0085] g\cdot u+h\cdot v+i\\

[0086] Step three, extracting world coordinates:

[0087] Extract the actual world coordinates from the homogeneous coordinates:

[0088] x=a⋅u+b⋅v+c;

[0089] y=d⋅u+e⋅v+f;

[0090] z=g⋅u+h⋅v+i;

[0091] The data recording unit is used to record the calculated data parameters in real time through the data recorder, detect the corresponding area of the truck vehicle in real time, and generate a transformation matrix from the camera pixel coordinates to the relative world coordinates, so that when the truck vehicle in the box area is positioned, the converted relative world coordinate transformation matrix can detect whether there is a problem of drift, jitter or data loss in the corresponding detection area range. Through the serial number arrangement of the target vehicle, the pixel coordinates are converted into the world coordinate transformation matrix in real time, and the standard transformation matrix parameters are set, the converted transformation matrix and the standard transformation matrix parameters are difference calculated, if the difference is equal to 0, it means that the vehicle is normal, if the difference is not equal to 0, it means that the vehicle is abnormal.

[0092] Embodiment two

[0093] Please refer to Figure 3 As shown in the figure: based on the basis of embodiment one, the vehicle simulation end includes a position detection module, a vibration detection module and a simulation environment module;

[0094] The position detection module includes a position detection unit and a data receiving unit;

[0095] The position detection unit is used to position the current position of the gantry crane in real time through the GPS device on the gantry crane, and arrange the positioning parameters of multiple gantry cranes in sequence through Arabic numerals, and the sequence is set in ascending order;

[0096] The data receiving unit is used to receive the positioning parameters of multiple gantry cranes and the pixel coordinates and world coordinate parameters of the target vehicle in the image in real time through the data receiver.

[0097] The vibration detection module includes a vibration detection unit and an abnormality judgment unit;

[0098] The vibration detection unit is used to detect the vibration parameters in the running process of the gantry crane in real time through a calculation formula, and set the standard parameters of the movement of the gantry crane, and the calculation formula is as follows:

[0099] ;

[0100] Among them, is the acceleration effective value, T is the period, and a(t) is the function of acceleration changing with time;

[0101] The abnormality judgment unit is used to calculate the difference between the acceleration effective value and the standard acceleration parameter of the movement of the gantry crane, judge whether the vibration intensity in the movement process of the gantry crane is abnormal, if The difference between the acceleration and the standard acceleration parameter of the movement of the gantry crane is greater than 1, which indicates that the vibration intensity is abnormal, if The difference between the acceleration and the standard acceleration parameter of the movement of the gantry crane is less than or equal to 1, which indicates that the vibration intensity is normal.

[0102] The simulation environment module includes a simulation simulation unit and a vehicle positioning unit;

[0103] The simulation simulation unit is used to set the simulation environment simulation model of the positioning of the container truck in the container area, and project the perception results of multiple camera lenses into the simulation environment simulation model;

[0104] The vehicle positioning unit is used for obtaining high-precision positioning parameters of the vehicle target according to the sensing results of multiple camera heads and combining the real-time positioning of the target vehicle captured by the multiple-area and multiple-angle camera heads, establishing a simulation environment by using the high-precision box area map, and obtaining the GPS positioning information of the truck in real time, so that the system can monitor the abnormal movement of the gantry crane in real time and adjust or issue a warning reminder in real time when the vibration is abnormal, thereby increasing the accuracy and intelligence of the truck positioning in the container box area.

[0105] Embodiment three

[0106] Please refer to Figure 4 Based on the basis of embodiment one, the target adjustment end includes a trusted detection module, a target matching module, a multi-effect processing module and a simulation identification module.

[0107] The trusted detection module is used for real-time calculation of the reliability of the gantry crane target vehicle positioning result at the current time through a calculation formula, and the calculation formula is as follows:

[0108] K = P (x) / P (y) ;

[0109] Wherein, K represents the reliability coefficient, x and y respectively represent the conditions of meeting and not meeting the positioning result of the gantry crane target vehicle at the current time, and the condition here represents the standard value difference and the standard parameter difference.

[0110] The target matching module includes a target matching algorithm and a matching sorting unit.

[0111] The target matching algorithm is used for matching, deduplication, deviation correction and position updating processing of multiple gantry crane target vehicle positioning results in continuous multiple time slices through a multi-target matching algorithm, and outputs an optimal matching optimization result. The multi-target matching algorithm is to calculate the difference value of two parameters of the multiple gantry crane target vehicle positioning results. If the difference value is equal to 0, it means that the processing is abnormal, and if the difference value is not equal to 0, it means that the processing is normal.

[0112] The matching sorting unit is used for arranging the optimal matching optimization result of the multi-target matching algorithm in sequence number by Arabic numerals.

[0113] The multi-effect processing module is used for real-time calculation of the difference value of the matching, deduplication, deviation correction and position updating of the target vehicle positioning result by the multi-target matching algorithm, and arranging the sequence number according to Arabic numerals.

[0114] The simulation identification module includes a simulation identification unit and a camera detection recording unit.

[0115] The simulation identification unit is used for establishing a physical simulation model for each vehicle in the simulation environment, and real-time identification and processing of camera detection abnormalities through a simulation simulation judgment formula, as follows:

[0116] ;

[0117] wherein, represents the matching value of the corresponding vehicle positioning parameter and the optimal matching optimization result, represents the corresponding target vehicle detected at the corresponding position, represents the standard parameter of the target vehicle at the corresponding position, represents the repetition rate of the actual value and the standard value of the corresponding target vehicle retrieved at the corresponding position, represents the repetition rate of the standard value and the actual value detected in real time;

[0118] If the calculated =0, it is determined that the vehicle positioning at the corresponding position matches the optimal matching optimization result, that is, the camera detection result is normal, and if ≠0, it is determined that the vehicle positioning at the corresponding position does not match the optimal matching optimization result, that is, the camera detection result is abnormal, and the system is reported, and the corresponding vehicle positioning parameters in the determination formula are replaced until the determination result =0, and the corresponding vehicle positioning parameters include the position parameter, the serial number parameter, the vehicle movement trajectory parameter and the vehicle positioning parameter;

[0119] The camera detection recording unit is used to record vehicle positioning abnormalities in real time through a data recorder, and to make traffic control decisions for target vehicles in real time when vehicle positioning abnormalities occur. The traffic control decisions are made according to the optimal matching optimization result. By establishing a physical simulation model for each vehicle in a simulation environment, camera detection abnormalities can be identified and processed in a timely manner. When multiple gantry cranes are operating simultaneously in a box area, the line-of-sight blocking problem between the gantry cranes can be handled in real time in combination with camera detection abnormalities, ensuring the positioning effect of the truck vehicles in the container and the traffic running condition when multiple gantry cranes are operating.

[0120] The application discloses a high-precision container yard truck positioning system and method, which comprises the following steps: configuring vehicle positioning remote control area server IP address information; entering a camera installation end, installing a camera on a gantry crane, setting a detection area, and detecting a truck in the corresponding area in real time, and generating a camera pixel coordinate to a relative world coordinate transformation matrix; entering a vehicle simulation end, detecting the position of the truck in the corresponding area in real time, obtaining high-precision positioning data of the gantry crane according to the pixel coordinates of the target vehicle in the image and the transformation matrix, establishing a simulation environment by using a high-precision container yard map, and obtaining GPS positioning information of the truck in real time; when the truck is positioned in the container yard, the transformed relative world coordinate transformation matrix can detect whether the vehicle in the corresponding detection area range exists problems such as drift, jitter or data loss, and the limitations of the truck positioning in the container are reduced; entering a target adjustment end, detecting target credibility according to the effective perception range of each camera in the simulation environment, matching, deduplicating, correcting and updating the position of the detection results of continuous multiple time slices by using a multi-target matching algorithm, obtaining an optimal matching and optimization result, establishing a physical simulation model for each vehicle in the simulation environment, and identifying and processing camera detection abnormalities in a timely manner; by establishing the simulation environment by using the high-precision container yard map and obtaining the GPS positioning information of the truck in real time, the system can monitor the abnormal operation of the back-and-forth movement of the gantry crane in real time, and can adjust or issue a warning reminder in real time when the vibration is abnormal, thereby increasing the accuracy and intelligence of the truck positioning in the container yard; by establishing the physical simulation model for each vehicle in the simulation environment and identifying and processing the camera detection abnormalities in a timely manner, when multiple gantry cranes in the container yard operate simultaneously, the line-of-sight blocking problem between the gantry cranes can be processed in real time by combining the camera detection abnormalities, the positioning effect of the truck in the container is ensured, and the traffic operation condition is better when multiple gantry cranes operate.

[0121] The basic principles and main features of the application and the advantages of the application are shown and described above. It should be understood by those skilled in the art that the application is not limited by the above examples, and the above examples and descriptions in the specification are only illustrative of the principles of the application. Without departing from the spirit and scope of the application, various changes and improvements can be made to the application, and these changes and improvements all fall within the scope of the claimed application. The scope of protection of the application is defined by the appended claims and their equivalents.

Claims

1. A high-precision container yard truck positioning system, characterized in that, The system comprises a camera mounting end, a vehicle simulation end and a target adjustment end; The camera mounting end is used for setting a truck positioning detection area in a container yard, mounting a camera on a gantry crane, and setting the shape and parameters of the corresponding vehicle, to detect the truck in the corresponding area in real time and generate a transformation matrix from the camera pixel coordinates to the relative world coordinates; The vehicle simulation end is used for detecting the position of the truck in the corresponding area in real time, obtaining the pixel coordinates of the target vehicle in the image and the transformation matrix, and detecting the running state of the gantry crane in real time to determine whether the gantry crane is running normally, and obtaining the GPS positioning information of the truck in real time by establishing a simulation environment based on the high-precision map of the container yard; The target adjustment end is used for detecting the target credibility according to the effective sensing range of each camera, and using a multi-target matching algorithm to match, remove duplicates, correct deviation and update the position of the detection results of multiple time slices to obtain the optimal matching result, and establishing a physical simulation model for each vehicle to identify and process camera detection abnormalities in a timely manner; A physical simulation model is established to identify and process camera detection abnormalities in real time by using a simulation simulation judgment formula, and the specific process is as follows: ； wherein, represents a matching value of determining the corresponding vehicle positioning parameter and the optimal matching optimization result, represents a corresponding target vehicle detected at the corresponding position, represents a standard parameter of the target vehicle at the corresponding position, represents a repetition rate of the standard value and the actual value of the corresponding target vehicle searched at the corresponding position, represents a repetition rate of the standard value and the actual value detected in real time; If the calculated =0, it is determined that the vehicle positioning at the corresponding position matches the optimal matching optimization result, i.e., the camera detection result is normal. If ≠0, it is determined that the vehicle positioning at the corresponding position does not match the optimal matching optimization result, i.e., the camera detection result is abnormal. The system is reported, and the corresponding vehicle positioning parameters in the determination formula are replaced until the determination result =0. The corresponding vehicle positioning parameters include position parameters, serial number parameters, vehicle movement trajectory parameters, and vehicle positioning parameters.

2. The system of claim 1, wherein: The camera mounting end comprises a region setting module, a region detection module and a coordinate conversion module; The region setting module comprises a region setting unit and a multi-region combination unit; The region setting unit is used for setting a truck positioning detection area in a container yard, and a plurality of ranges corresponding to the detection area, and three cameras are installed on each gantry crane in the plurality of ranges corresponding to the detection area, and two cameras are installed in the approach direction of the gantry crane, one of which is a long-focus camera for detecting trucks within 100-250 meters from the gantry crane, and the other is a short-focus camera for detecting trucks within 120 meters from the gantry crane, and a short-focus camera is installed in the departure direction of the vehicle for detecting trucks within 110 meters from the gantry crane, and the cameras are connected to a network switch on the gantry crane and return video streams to the system center control area server; The multi-region combination unit is used for real-time acquisition of multi-region and multi-angle gantry crane running parameters, and automatic storage of the parameters by a data logger.

3. The system of claim 2, wherein: The region detection module comprises a truck unit and a vehicle detection unit; The truck unit is used for setting an appearance image model of the gantry crane in the vehicle positioning system; The vehicle detection unit is used for real-time capture of the running parameters of the gantry crane by the camera, difference calculation of the captured running parameters of the gantry crane with standard values, real-time detection of the gantry crane, and installation of a GPS positioning device on the gantry crane if the difference is not equal to 0.

4. The system of claim 3, wherein: The coordinate conversion module comprises a pixel coordinate unit, a transformation matrix unit and a data recording unit; The pixel coordinate unit is used for real-time acquisition of the pixel coordinates (u, v) of the target vehicle in the image by the image; The transformation matrix unit is configured to convert the pixel coordinates into world coordinates (x, y, z) by setting a transformation matrix T and combining the pixel coordinates (u, v) of the target vehicle in the image; The data recording unit is configured to record the calculated data parameters in real time by a data recorder.

5. The system of claim 1, wherein: The vehicle simulation end comprises a position detection module, a vibration detection module, and a simulation environment module; The position detection module comprises a position detection unit and a data receiving unit; The position detection unit is configured to locate the position of the gantry crane at the current time in real time through a GPS device on the gantry crane, and arrange the positioning parameters of multiple gantry cranes in sequence by Arabic numerals, with the sequence arranged in ascending order; The data receiving unit is configured to receive the positioning parameters of multiple gantry cranes, and the pixel coordinates and world coordinate parameters of the target vehicle in the image in real time through a data receiver.

6. The system of claim 5, wherein: The vibration detection module comprises a vibration detection unit and an abnormality judgment unit; The vibration detection unit is configured to detect the vibration parameters in the running process of the gantry crane in real time through a calculation formula, and set the standard parameters of the movement of the gantry crane, the calculation formula being as follows: ； wherein is the effective value of acceleration, T is the period, a(t) is the function of acceleration variation over time; The abnormality judging unit is configured to calculate the difference between the acceleration effective value and the standard acceleration parameter of the gantry crane movement, judge whether the vibration intensity during the gantry crane movement is abnormal, if the difference between the acceleration effective value and the standard acceleration parameter of the gantry crane movement is greater than 1, it indicates that the vibration intensity is abnormal, if the difference between the acceleration effective value and the standard acceleration parameter of the gantry crane movement is less than or equal to 1, it indicates that the vibration intensity is normal.

7. The system of claim 6, wherein: The simulation environment module comprises a simulation simulation unit and a vehicle positioning unit; The simulation simulation unit is configured to set a simulation environment simulation model for the positioning of the container truck in the container area, and project the sensing results of multiple camera lenses into the simulation environment simulation model; The vehicle positioning unit is configured to obtain high-precision positioning parameters of the vehicle target according to the sensing results of multiple camera lenses, and in combination with the camera lenses of multiple regions and multiple angles to capture the real-time positioning of the target vehicle in real time.

8. The system of claim 1, wherein: The target adjustment end comprises a trusted detection module, a target matching module, a multi-effect processing module, and a simulation recognition module; The trusted detection module is configured to calculate the trustworthiness of the positioning result of the gantry crane target vehicle at the current time in real time through a calculation formula, the calculation formula being as follows: K = P(x) / P(y); Wherein, K represents the confidence coefficient, x and y respectively represent the conditions of meeting and not meeting the positioning result of the gantry crane target vehicle at the current time, and the condition here represents the standard value difference and the standard parameter difference; The target matching module comprises a target matching algorithm and a matching sorting unit; The target matching algorithm is configured to match, deduplicate, correct deviation, and update position of the positioning results of multiple gantry crane target vehicles in multiple time slices in succession through a multi-target matching algorithm, output an optimal matching optimization result, and the multi-target matching algorithm is to calculate the difference of two parameters of the positioning results of multiple gantry crane target vehicles, if the difference is equal to 0, it indicates that the processing is abnormal, if the difference is not equal to 0, it indicates that the processing is normal; The matching sorting unit is configured to arrange the optimal matching optimization result of the multi-target matching algorithm in sequence by Arabic numerals.

9. The system of claim 8, wherein: The multi-effect processing module is configured to calculate the difference in real time through the multi-target matching algorithm for the matching, deduplication, deviation correction, and position update of the positioning result of the target vehicle, and arrange the sequence in Arabic numerals; The simulation recognition module comprises a simulation recognition unit and a camera detection recording unit; The simulation recognition unit is configured to establish a physical simulation model for each vehicle in the simulation environment; The camera detection recording unit is used to record the vehicle positioning anomaly in real time through the data recorder, and make traffic control decision of the target vehicle in real time when the vehicle positioning anomaly occurs, and the traffic control decision is made according to the optimal matching optimization result.

10. The positioning method of the high-precision container yard truck positioning system according to any one of claims 1-9, characterized in that, It comprises the following steps: Step 1: configure the vehicle positioning remote control area server IP address information; Step 2: enter the camera installation end, install the camera on the gantry crane, set the detection area, detect the truck in the corresponding area in real time, and generate the transformation matrix of the camera pixel coordinates to the relative world coordinates; Step 3: enter the vehicle simulation end, detect the truck position in the corresponding area in real time, obtain the high-precision positioning data of the gantry crane in real time according to the pixel coordinates of the target vehicle in the image and the transformation matrix, establish a simulation environment by using the high-precision map of the box area, and obtain the GPS positioning information of the truck in real time; Step 4: enter the target adjustment end, detect the target credibility in the simulation environment according to the effective sensing range of each camera, use the multi-target matching algorithm to match, remove duplicates, correct deviation, and update the position of the detection results of continuous multiple time slices, obtain the optimal matching optimization result, and establish a physical simulation model for each vehicle in the simulation environment, and identify and process the camera detection anomaly in time.

Images