Autonomous driving method

Abstract

The application provides an automatic driving and docking method, comprising the following steps: S1, starting a docking vehicle; S2, a data acquisition module acquires information through a sensor; S3, a data processing module calculates a relative azimuth angle; S4, the data processing module performs a data fusion operation; S5, an execution module controls the docking vehicle to move towards an aircraft hatch; S6, the execution module controls an upper loading platform to rise to the height of the aircraft hatch; S7, the execution module controls the docking platform to extend forward until the docking platform contacts the aircraft; and S8, a communication module receives a docking completion signal and sends the docking completion signal to an external system. The application integrates a data acquisition module, a data processing module, an execution module and a communication module into the docking vehicle, and designs corresponding methods, so that the automatic driving and safe docking of an airport special vehicle are realized.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to an autonomous driving method. Background Art

[0002] Airport special vehicles are used exclusively within airports, operating in enclosed environments. They include passenger elevators, food trucks, vehicles for boarding passengers with limited mobility, and bulk cargo loaders. These vehicles utilize specialized superstructures mounted on their chassis to provide them with the necessary functions.

[0003] During airport special vehicle operations, docking is a critical operation: controlling the vehicle's approach to the aircraft and precisely aligning the superstructure with the aircraft's door. Currently, docking lacks automation, relying primarily on manual observation and sensor-assisted control. While the vehicle possesses some perception capabilities, these capabilities serve only as a protective measure to limit manual operation. Operator inexperience or errors often lead to inaccurate docking positioning, repeated docking, and even scratching the aircraft's exterior. This can delay ramp scheduling and flight delays, or even damage the aircraft, resulting in substantial compensation.

[0004] Therefore, existing technologies lack automation in the docking process of airport special vehicles, resulting in high apron scheduling pressure and loss costs. Considering that airport special vehicles operate on a closed airport apron and dock only one object (involving only the vehicle and the aircraft, without any third party involved), a sensor system and its control method can be used to achieve automatic and safe docking. Summary of the Invention

[0005] The purpose of the present invention is to provide an automatic driving docking method, which realizes the automatic driving and safe docking of special vehicles at airports by integrating data acquisition module, data processing module, execution module and communication module, integrating these modules into the docking vehicle, and designing corresponding utilization methods.

[0006] An automatic driving system, comprising:

[0007] A vehicle for docking, comprising a chassis and a superstructure, wherein the superstructure comprises a lifting component and a docking platform, and the superstructure is arranged on the chassis;

[0008] A data acquisition module, the data acquisition module collects information, the data acquisition module is arranged on the platform, wherein the data acquisition module includes a plurality of sensors;

[0009] A data processing module, which processes the collected information and obtains data processing information. The data processing module includes a data receiver, a data processor, and a unit controller. The data processing module is disposed in the chassis;

[0010] An execution module controls the vehicle, and includes a wire-controlled throttle, wire-controlled steering, wire-controlled brakes, a body controller, and a limit switch;

[0011] A communication module is provided in the chassis and is used for information communication.

[0012] In one embodiment of the present invention, the multiple sensors include cameras, ultrasonic radars, and laser radars, wherein the multiple cameras include:

[0013] A first camera, wherein the first camera is arranged on the front panel of the vehicle;

[0014] a second camera, the second camera being arranged on both sides of the lifting component;

[0015] The third camera is arranged on the facade of the machine platform.

[0016] In one embodiment of the present invention, the field of view angles of the first camera and the third camera are configured to be 70°-80° upward in the horizontal direction, the field of view angle of the second camera is configured to be configured in the horizontal direction, and the limit switch is configured to be triggered when the docking platform contacts the aircraft.

[0017] An automatic driving method comprises the following steps:

[0018] S1. Start the vehicle. After the vehicle reaches the starting position, the external system issues a start signal. The data processing module receives the start signal and transmits it to the data acquisition module and the execution module. The execution module starts the vehicle.

[0019] S2. The data acquisition module collects information through sensors. The first and third cameras collect visual information in front of the vehicle, the second camera collects visual information on the side of the vehicle, and the ultrasonic radar and lidar collect environmental information and transmit it to the data processing module.

[0020] S3. The data processing module calculates the relative azimuth angle, wherein the data processing module determines whether the docking vehicle is facing the aircraft hatch based on the visual information in front of the vehicle and the calculated relative azimuth angle between the docking vehicle and the aircraft hatch;

[0021] S4. The data processing module performs a data fusion operation, wherein the data fusion operation fuses the visual information in front of the vehicle and the environmental information to obtain the horizontal distance and vertical distance between the docking platform and the aircraft hatch, as well as environmental obstacle information;

[0022] S5. The execution module controls the docking vehicle to move toward the aircraft hatch. When the docking platform reaches a first preset distance from the aircraft hatch, the execution module controls the docking vehicle to move forward at a speed lower than the first preset speed and to stop at a second preset distance.

[0023] S6. The execution module controls the upper loading platform to rise to the height of the aircraft hatch. When the number of hatch feature points detected by the first camera decreases by more than 50%, it is determined that the first camera's field of view is partially blocked by the aircraft hatch. Based on the blockage determination, the priority of the first camera's data is lowered, and the third camera, ultrasonic radar, and lidar data are preferentially used for position calibration.

[0024] S7, the execution module controls the docking platform to extend forward until the docking platform contacts the aircraft, the execution module triggers the limit switch, the docking platform stops extending forward, and the execution module sends a docking completion signal;

[0025] S8. The communication module receives the docking completion signal and sends the docking completion signal to an external system. The recipients of the docking completion signal may be the apron control center, the aircraft, the apron manager, and other work vehicles.

[0026] In one embodiment of the present invention, in step S2,

[0027] The vehicle front visual information includes hatch feature points, hatch center pixel coordinates, aircraft hatch outline feature point pixel coordinates, hatch sign features, and ground guide line information;

[0028] The vehicle side visual information includes the edge features of obstacles on the vehicle side and feature points of adjacent equipment;

[0029] The environmental information includes near-field obstacle distance information, 3D point cloud data, and ultrasonic radar ranging data.

[0030] In one embodiment of the present invention, in the step S3, the relative azimuth angle is the angle between the longitudinal axis of the docking vehicle and the normal direction of the aircraft hatch in the horizontal plane. If the deviation angle of the relative azimuth angle is greater than 30°, the docking is terminated and a repositioning instruction is issued; if the deviation angle of the relative azimuth angle is less than or equal to 30°, the next step is performed.

[0031] In one embodiment of the present invention, the calculation method of the relative azimuth angle in step S3 includes:

[0032] S31. Identify the aircraft hatch through a sensor and obtain the pixel coordinates of the hatch center point;

[0033] S32, using the sensor's intrinsic parameter matrix and extrinsic parameter matrix, converting the hatch center pixel coordinates into three-dimensional coordinates in the vehicle coordinate system;

[0034] S33, calculating the direction vector of the line between the hatch center point and the origin of the vehicle coordinate system;

[0035] S34. Project the direction vector onto a horizontal plane, calculate the angle between the projection vector and the longitudinal axis of the docking vehicle, and obtain the relative azimuth angle between the docking vehicle and the aircraft hatch.

[0036] In one embodiment of the present invention, in step S5, the first preset distance is 500 mm to 700 mm, and the first preset speed is 0.6 km / h to 0.9 km / h; the second preset distance is dynamically set according to the physical dimensions of the approaching vehicle and the aircraft.

[0037] In one embodiment of the present invention, the step S6 includes:

[0038] S61, performing monocular visual positioning based on hatch feature points collected by a third camera;

[0039] S62. Fit the hatch plane equation using 3D point cloud data and calculate the plane normal vector;

[0040] S63. Combine the ultrasonic radar ranging data and optimize the posture parameters of the platform using the least squares method.

[0041] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements any of the methods described above.

[0042] The present invention has the following beneficial effects. By integrating a data acquisition module, a data processing module, an execution module, and a communication module, integrating these modules into the docking vehicle, and designing a corresponding utilization method, the present invention realizes the automatic and safe docking of airport special vehicles. This achievement fills the automation gap in the last link of the workflow of airport special vehicles, can effectively help the apron relieve scheduling pressure, reduce various losses caused by human factors, and significantly improve the automation level of the apron. The present invention is also equipped with a variety of sensors that can effectively transmit back a variety of data and perform data calculations. The present invention integrates multi-source data of visual sensors, laser radars, and ultrasonic radars to achieve high-precision positioning of the aircraft hatch and posture optimization of the docking platform, providing reliable technical support for automatic and safe docking. In addition, the sensor layout is highly flexible and can be applied to most airport special vehicles with docking functions. At the same time, it focuses on the use of visual solutions to achieve stable information acquisition at a relatively low cost in a relatively single airport environment, and has good practicality and economy. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A side view of a docking vehicle is disclosed, showing an automatic driving docking method according to an embodiment of the present invention;

[0044] Figure 2 A front view of a docking vehicle is disclosed for an automatic driving docking method according to an embodiment of the present invention;

[0045] Figure 3 A module diagram of an automatic driving method according to an embodiment of the present invention is disclosed;

[0046] Figure 4 A detailed module diagram of an automatic driving method according to an embodiment of the present invention is disclosed;

[0047] Figure 5 A schematic diagram of relative azimuth angles of an automatic driving approach method according to an embodiment of the present invention is disclosed;

[0048] Figure 6 A step diagram of an automatic driving method according to an embodiment of the present invention is disclosed.

[0049] Attached photos

[0050] 1. Vehicle for support; 11. Chassis; 12. Superstructure; 13. Front of vehicle; 14. Lifting components; 15. Platform for support; 2. Data acquisition module; 21. First camera; 22. Second camera; 23. Third camera; 24. Ultrasonic radar; 25. LiDAR; 3. Data processing module; 31. Data receiver; 32. Data processor; 33. Unit controller; 4. Execution module; 41. Wire-controlled throttle; 42. Wire-controlled steering; 43. Wire-controlled brake; 44. Superstructure controller; 45. Limit switch; 5. Communication module. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the invention and are not intended to limit the invention.

[0052] like Figure 1 、 Figure 2 As shown, an automatic driving docking system includes a docking vehicle 1, a data acquisition module 2, a data processing module 3, an execution module 4, and a communication module 5. The docking vehicle 1 includes a chassis 11, a superstructure 12, and a vehicle head 13. The superstructure 12 includes a lifting component 14 and a docking platform 15. The superstructure 12 is arranged on the chassis 11. Figure 3As shown, data acquisition module 2, located on the docking platform 15, collects information and includes multiple sensors. Data processing module 3 processes the collected information and generates data processing information. Data processing module 3 includes a data receiver 31, a data processor 32, and a unit controller 33. Data processing module 3 is located within chassis 11. Actuation module 4 controls the docking vehicle 1 and includes a drive-by-wire throttle 41, drive-by-wire steering 42, drive-by-wire brake 43, a bodywork controller 44, and a limit switch 45. A communication module 5 is located within chassis 11 and is used for information communication.

[0053] In one embodiment of the present invention, the multiple sensors include cameras, ultrasonic radar 24, and lidar 25. The multiple cameras include a first camera 21, a second camera 22, and a third camera 23. The first camera 21 is located on the front panel of the vehicle head 13. The second camera 22 is located on both sides of the lifting component 14. The third camera 23 is located on the vertical surface of the platform 15. In this embodiment, the first and third cameras 21 and 23 are preferably Sony IMX490 models, with specifications of 12 megapixels, a 1 / 1.3" sensor, and a 100° horizontal field of view. The second camera 22 is preferably Sony IMX390, with specifications of -8 megapixels, a 1 / 1.7" sensor, and a 180° horizontal field of view. The lidar 25 is a Bosch SRR500 model, operating at 77 GHz and with a detection range of 0.3-20 meters. The data processor 32 is an NVIDIA Orin AGX model, supporting 275 TOPS of computing power and 16 GMSL2 camera inputs. In this embodiment, the first camera 21 is preferably mounted on the exterior front panel of the vehicle head 13 at a 15° elevation angle. The second camera 22 is mounted on the right rearview mirror bracket of the upper body 12 lifting component 14, facing horizontally toward the vehicle body. The third camera 23 is mounted in the middle of the bumper at a 10° elevation angle. The laser radar 25 is mounted on the front end of the platform 15. The ultrasonic radar 24 is embedded in the four corners or the front end of the platform 15.

[0054] In one embodiment of the present invention, Figure 1 As shown, the first camera 21 and the third camera 23 have a field of view angle of 70°-80° upward from the horizontal, while the second camera 22 has a field of view angle of 180° horizontally. The limit switch 45 is configured to be triggered when the docking platform 15 contacts the aircraft. In this embodiment, the first camera 21 and the third camera 23 preferably have a field of view angle of 75° upward from the horizontal, while the second camera 22 has a field of view angle of 180° horizontally.

[0055] In one embodiment of the present invention, the autonomous driving system utilizes a four-layer distributed architecture, with each module achieving coordinated control via a hybrid communication network. The modules of this system exchange data via the CAN bus and Ethernet. The system comprises a data acquisition layer, a core processing layer, an execution control layer, and a communication interface layer. The data acquisition layer deploys various sensors, including cameras and radars. The cameras are high-definition industrial cameras, installed on the vehicle's front and sides; the radars are millimeter-wave radars, located on the vehicle's front and sides, to detect the distance and relative position to the aircraft. The data acquisition layer transmits data in real time via Ethernet to the data processing module 3 in the core processing layer, providing basic information for subsequent data fusion and decision-making. In the core processing layer, data processing module 3 completes data fusion and decision-making, and sends control instructions to the execution module 4 via the CAN bus. In the execution control layer, the execution module 4 receives the instructions to control vehicle movement, and feedback signals, such as limit switches 45, are transmitted back via hardwire or CAN. Upon receiving the control instructions from the core processing layer, the execution module 4 controls the vehicle to perform the corresponding actions. Limit switches 45 are installed at key locations on the vehicle's upper structure 12 and body. When the vehicle approaches an aircraft to a preset safe distance or when the upper structure 12 is properly docked with the aircraft's door, the limit switches 45 trigger feedback signals. These feedback signals are transmitted back to the data processing module 3 in the core processing layer via hardwire or CAN bus. At the communication interface layer, the communication module 5 exchanges data with external systems via 4G / 5G / Wi-Fi. Through these communication methods, the system can exchange data with external systems such as the airport's apron dispatch system and aircraft information management system.

[0056] like Figure 6 As shown, an automatic driving method includes the following steps:

[0057] S1. Start the docking vehicle 1. After the docking vehicle 1 reaches the docking starting position, the external system issues a docking start signal. The data processing module 3 receives the docking start signal and transmits it to the data acquisition module 2 and the execution module 4. The execution module 4 starts the docking vehicle 1.

[0058] S2, the data acquisition module 2 collects information through sensors, wherein the first camera 21 and the third camera 23 collect visual information in front of the vehicle, the second camera 22 collects visual information on the side of the vehicle, the ultrasonic radar 24 and the laser radar 25 collect environmental information, and transmits it to the data processing module 3;

[0059] S3. The data processing module 3 calculates the relative azimuth angle. The data processing module 3 determines whether the docking vehicle 1 is facing the aircraft hatch based on the visual information in front of the vehicle and the calculated relative azimuth angle between the docking vehicle 1 and the aircraft hatch.

[0060] S4, the data processing module 3 performs a data fusion operation, wherein the data fusion operation fuses the visual information in front of the vehicle and the environmental information to obtain the horizontal distance and vertical distance between the docking platform 15 and the aircraft hatch and the environmental obstacle information;

[0061] S5. The execution module 4 controls the docking vehicle 1 to move toward the aircraft hatch. When the docking platform 15 reaches a first preset distance from the aircraft hatch, the docking vehicle 1 is controlled to move forward at a speed lower than the first preset speed and stop at a second preset distance.

[0062] S6. The execution module 4 controls the upper platform 12 to rise to the height of the aircraft hatch. When the number of hatch feature points detected by the first camera 21 decreases by more than 50%, it is determined that the field of view of the first camera 21 is partially blocked by the docking platform 15. Based on the blockage determination, the data priority of the first camera 21 is lowered, and the data from the third camera 23, ultrasonic radar 24, and lidar 25 are preferentially used for position calibration.

[0063] S7, the execution module 4 controls the docking platform 15 to extend forward until the docking platform 15 contacts the aircraft, the execution module 4 triggers the limit switch 45, the docking platform 15 stops extending forward, and the execution module 4 sends a docking completion signal;

[0064] S8. The communication module 5 receives the docking completion signal and sends the docking completion signal to the external system. The recipients of the docking completion signal may be the apron control center, the aircraft, the apron manager and other work vehicles.

[0065] In one embodiment of the present invention, in step S1, after the vehicle is moved to the docking starting position by manual driving or automatic driving, the automatic docking process is started manually or automatically.

[0066] In one embodiment of the present invention, in step S2, the visual information in front of the vehicle includes hatch feature points, hatch center pixel coordinates, aircraft hatch outline feature point pixel coordinates, hatch sign features, and ground guideline information. The visual information to the side of the vehicle includes edge features of obstacles to the side of the vehicle and feature points of adjacent equipment. Environmental information includes near-field obstacle distance information, 3D point cloud data, and ultrasonic radar 24 ranging data. In this embodiment, the 3D point cloud data preferably includes obstacle spatial coordinates, plane fitting parameters, and lidar 25 point cloud data. Furthermore, the 3D point cloud data is obtained by scanning the lidar 25, and the near-field obstacle distance information and ultrasonic radar 24 ranging data are obtained by scanning the ultrasonic radar 24.

[0067] In one embodiment of the present invention, Figure 5As shown, in step S3, the relative azimuth angle is the projection angle between the longitudinal axis of the docking vehicle 1 and the normal direction of the aircraft hatch on the horizontal plane. If the deviation angle of the relative azimuth angle is greater than 30°, the docking is terminated and a repositioning instruction is issued; if the deviation angle of the relative azimuth angle is less than or equal to 30°, the next step is performed.

[0068] In one embodiment of the present invention, in step S3, the method for calculating the relative azimuth angle includes: S31, identifying the aircraft hatch through a sensor and obtaining the pixel coordinates of the hatch center point. S32, using the sensor's intrinsic parameter matrix and extrinsic parameter matrix, converting the hatch center point pixel coordinates into three-dimensional coordinates in the coordinate system of the docking vehicle 1. In this embodiment, the sensor is preferably a camera. In step S32, the hatch center point pixel coordinate system is converted to the camera coordinate system through the camera's intrinsic parameter matrix, and the camera coordinate system is converted to the vehicle coordinate system through the camera's extrinsic parameter matrix. S33, calculating the direction vector of the line between the hatch center point and the origin of the coordinate system of the docking vehicle 1. S34, projecting the direction vector onto the horizontal plane, calculating the angle between the projection vector and the longitudinal axis of the docking vehicle 1, and obtaining the relative azimuth angle between the docking vehicle 1 and the aircraft hatch.

[0069] In one embodiment of the present invention, in step S5, the first preset distance is 500 mm to 700 mm, and the first preset speed is 0.6 km / h to 0.9 km / h; the second preset distance is dynamically set according to the physical dimensions of the approaching vehicle 1 and the aircraft.

[0070] In one embodiment of the present invention, step S6 includes:

[0071] S61. Perform monocular visual positioning based on hatch feature points captured by the third camera 23. Furthermore, the third camera 23 is mounted on the facade of the airport's special vehicle docking platform 15, with its lens facing the aircraft hatch, enabling clear capture of the aircraft hatch area. During the docking process, the third camera 23 captures hatch images in real time and uses a preset image recognition algorithm to extract key hatch feature points, such as the corner points of the hatch edge and the contour points where the hatch door connects to the fuselage. The pixel coordinates of these feature points in the image are recorded in real time. Combined with the third camera 23's internal parameters (such as focal length and principal point coordinates) and external parameters (such as the camera's mounting position and angle relative to the docking platform 15), the three-dimensional coordinates of the feature points in the vehicle coordinate system are calculated using the principles of monocular visual positioning, thereby preliminarily determining the position of the aircraft hatch relative to the docking platform 15.

[0072] S62. Fit the hatch plane equation using the 3D point cloud data and calculate the plane normal vector. In this preferred embodiment, in step S62, the hatch plane equation is fitted using the LiDAR 25 point cloud data to calculate the plane normal vector. Furthermore, the LiDAR 25 mounted on the vehicle's roof simultaneously scans the aircraft hatch area, acquiring LiDAR 25 point cloud data for the hatch and surrounding areas. Data processing module 3 filters the raw LiDAR 25 point cloud data to remove outliers caused by noise and environmental interference, retaining valid LiDAR 25 point cloud data belonging to the hatch plane.

[0073] S63. Optimize the pose parameters of the docking platform 15 using the least squares method, combining the ranging data from the ultrasonic radar 24. Four ultrasonic radars 24 are evenly mounted around the docking platform 15, with their detection direction perpendicular to the platform surface. They can measure the distance between the platform and the aircraft hatch in real time. After obtaining the locations of the feature points from monocular vision positioning and the hatch plane normal vector, the data processing module 3 combines these two to preliminarily determine the pose parameters of the docking platform 15. These pose parameters include position parameters (x, y, and z coordinates) and attitude parameters (roll, pitch, and yaw angles). Subsequently, the ranging data from the ultrasonic radar 24 is used as a constraint to construct an error function using the pose parameters as variables. The input to the error function is the deviation between the theoretical ranging value calculated from the preliminary pose parameter calculation and the actual value measured by the ultrasonic radar 24. The error function is iteratively optimized using the least squares method, continuously adjusting the pose parameters until the error function value is minimized, ultimately obtaining the optimal pose parameters for the docking platform 15. Based on the parameters, the execution module 4 controls the docking platform 15 to fine-tune its position and posture to achieve docking with the aircraft hatch.

[0074] In one embodiment of the present invention, a computer-readable storage medium is also included, on which a computer program is stored. When the program is executed by a processor, the automatic driving method as described above is implemented.

[0075] The present invention has the following beneficial effects. By integrating data acquisition modules, data processing modules, execution modules, and communication modules, integrating these modules into docking vehicles, and designing corresponding utilization methods, the present invention successfully realizes the automatic and safe docking of airport special vehicles. This achievement fills the automation gap in the last link of the workflow of airport special vehicles, can effectively help the apron relieve scheduling pressure, reduce various losses caused by human factors, and significantly improve the automation level of the apron. The present invention is also equipped with a variety of sensors that can effectively transmit back a variety of data and perform data calculations. The present invention integrates multi-source data of visual sensors, laser radars, and ultrasonic radars to achieve high-precision positioning of the aircraft hatch and posture optimization of the docking platform, providing reliable technical support for automatic and safe docking. In addition, the sensor layout is highly flexible and can be applied to most airport special vehicles with docking functions. At the same time, it focuses on the use of visual solutions to achieve stable information acquisition at a relatively low cost in a relatively single airport environment, and has good practicality and economy.

[0076] It should be noted that, unless otherwise clearly specified and limited, the words "install", "connect", "connect" and similar terms used in the description of this application should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, an indirect connection through an intermediate medium, or a connection between two components. Those skilled in the art can understand their specific meanings in this application according to the specific circumstances.

[0077] The above embodiments are merely further explanations of the present invention and are not intended to limit the present invention in any other manner. The present invention may also have various other embodiments. Those skilled in the art may make various corresponding modifications and variations based on the present invention without departing from the spirit and substance of the present invention, and such corresponding modifications and variations shall fall within the scope of protection of the present invention.

Claims

1. An automatic driving method, characterized in that: The following steps are involved: S1. Start the docking vehicle. The docking vehicle includes a chassis and a superstructure. The superstructure includes a lifting component and a docking platform. The superstructure is arranged on the chassis. After the docking vehicle reaches the docking starting position, the external system issues a docking start signal. The data processing module receives the docking start signal and transmits it to the data acquisition module and the execution module. The execution module starts the docking vehicle. S2. A data acquisition module collects information through sensors. The data acquisition module is arranged on the platform, wherein the data acquisition module includes multiple cameras, ultrasonic radars, and laser radars. The first camera is arranged on the front panel of the vehicle, and the third camera is arranged on the facade of the platform. The first and third cameras collect visual information in front of the vehicle. The second camera is arranged on both sides of the lifting component. The second camera collects visual information on the side of the vehicle. The ultrasonic radar and laser radar collect environmental information and transmit it to the data processing module. S3. A data processing module calculates a relative azimuth angle. The data processing module includes a data receiver, a data processor, and a unit controller. The data processing module is disposed within the chassis. The data processing module determines whether the approaching vehicle is facing the aircraft hatch based on visual information in front of the vehicle and the calculated relative azimuth angle between the approaching vehicle and the aircraft hatch. S4. The data processing module performs a data fusion operation, wherein the data fusion operation fuses the visual information in front of the vehicle and the environmental information to obtain the horizontal distance and vertical distance between the docking platform and the aircraft hatch, as well as environmental obstacle information; S5. An execution module controls the docking vehicle to move toward the aircraft hatch. The execution module includes a drive-by-wire throttle, a drive-by-wire steering, a drive-by-wire brake, a bodywork controller, and a limit switch. When the docking platform reaches a first preset distance from the aircraft hatch, the docking vehicle is controlled to move forward at a speed lower than the first preset speed and stop at a second preset distance. S6. The execution module controls the upper loading platform to rise to the height of the aircraft hatch. When the number of hatch feature points detected by the first camera decreases by more than 50%, it is determined that the first camera's field of view is partially blocked by the aircraft hatch. Based on the blockage determination, the priority of the first camera's data is lowered, and the third camera, ultrasonic radar, and lidar data are preferentially used for position calibration. S7, the execution module controls the docking platform to extend forward until the docking platform contacts the aircraft, the execution module triggers the limit switch, the docking platform stops extending forward, and the execution module sends a docking completion signal; S8. The communication module receives the docking completion signal and sends the docking completion signal to the external system. The communication module is set in the chassis. The recipients of the docking completion signal can be the apron control center, aircraft, apron manager and other work vehicles.

2. The automatic driving method according to claim 1, wherein: The field of view angles of the first camera and the third camera are configured to be 70°-80° upward in the horizontal direction, the field of view angle of the second camera is configured to be configured in the horizontal direction, and the limit switch is configured to be triggered when the platform contacts the aircraft.

3. The automatic driving method according to claim 1, characterized in that: In the step S2, The vehicle front visual information includes hatch feature points, hatch center pixel coordinates, aircraft hatch outline feature point pixel coordinates, hatch sign features, and ground guide line information; The vehicle side visual information includes the edge features of obstacles on the vehicle side and feature points of adjacent equipment; The environmental information includes near-field obstacle distance information, 3D point cloud data, and ultrasonic radar ranging data.

4. The automatic driving method according to claim 1, wherein: In step S3, the relative azimuth angle is the angle between the longitudinal axis of the docking vehicle and the normal direction of the aircraft hatch in the horizontal plane. If the deviation angle of the relative azimuth angle is greater than 30°, the docking is terminated and a repositioning instruction is issued; if the deviation angle of the relative azimuth angle is less than or equal to 30°, the next step is performed.

5. The automatic driving method according to claim 1, characterized in that: The calculation method of the relative azimuth angle in step S3 includes: S31. Identify the aircraft hatch through a sensor and obtain the pixel coordinates of the hatch center point; S32, using the sensor's intrinsic parameter matrix and extrinsic parameter matrix, converting the hatch center pixel coordinates into three-dimensional coordinates in the vehicle coordinate system; S33, calculating the direction vector of the line between the hatch center point and the origin of the vehicle coordinate system; S34. Project the direction vector onto a horizontal plane, calculate the angle between the projection vector and the longitudinal axis of the docking vehicle, and obtain the relative azimuth angle between the docking vehicle and the aircraft hatch.

6. The automatic driving method according to claim 1, characterized in that: In step S5, the first preset distance is 500 mm to 700 mm, and the first preset speed is 0.6 km / h to 0.9 km / h; the second preset distance is dynamically set according to the physical dimensions of the approaching vehicle and the aircraft.

7. The automatic driving method according to claim 1, characterized in that: The S6 step includes: S61, performing monocular visual positioning based on hatch feature points collected by a third camera; S62. Fit the hatch plane equation using 3D point cloud data and calculate the plane normal vector; S63. Combine the ultrasonic radar ranging data and optimize the posture parameters of the platform using the least squares method.

Images