CN116300971B - Traction sliding control method and device for civil aircraft, tractor and storage medium - Google Patents

Abstract

The invention discloses a traction sliding control method and device of a civil aircraft, a tractor and a medium. The method comprises the following steps: controlling a wheel holding mechanism on a tractor to approach and accurately butt-joint a front landing gear of a target aircraft in a first-fast-then-slow mode, and planning a global planning path from a sliding starting point to a sliding end point after the butt-joint is completed; adopting a pre-established traction sliding controller to track and control the global planning path so as to enable the tractor to stably slide the traction target plane; during traction and taxiing, if an obstacle is detected to appear in front of a taxiing path and the obstacle cannot be avoided in a slow-down mode, a new global planning path is planned again according to the current taxiing position point, the obstacle information and the taxiing target point. The technical scheme of the embodiment of the invention ensures the accuracy, the rapidity and the safety of the automatic docking of the tractor and the airplane and improves the safety and the reliability of the tractor and the airplane in the combined sliding process.

Description

Towing taxiing control method, device, tractor and storage medium for civil aircraft

Technical Field

The present invention relates to the technical field of towing navigation of civil aircraft, and in particular to a towing taxiing control method, device, towing vehicle and storage medium of a civil aircraft.

Background Art

The number of flights in my country has increased dramatically with the development of the country. With the increase in the number of flights, the speed of airports in the aircraft dispatching process also needs to be improved. In order to ensure the improvement of work efficiency, a tractor is needed to tow the aircraft to the target location, that is, the preparation point before the aircraft takes off.

In the related technology, an unmanned barless tractor can be used to automatically dock the wheel-holding mechanism with the front landing gear of the aircraft. After the docking is completed and the wheel-holding mechanism is lifted, the tractor tows the aircraft from the apron to the preparation point before the aircraft takes off. One of the difficulties is the accuracy, speed and safety of the barless tractor and the aircraft in the automatic docking process; another difficulty is the safety and reliability of the barless tractor and the aircraft in the combined taxiing process.

Summary of the invention

The embodiment of the present invention provides a towing taxiing control method, device, towing vehicle and storage medium for a civil aircraft, so as to realize that after the wheel holding mechanism on the towing vehicle accurately docks with the front landing gear of the target aircraft, the towing vehicle stably tows the target aircraft to taxi to the taxiing end point.

According to one aspect of an embodiment of the present invention, a towing taxiing control method for a civil aircraft is provided, comprising:

Control the wheel-holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a fast-first-and-slow-later manner, and after docking, plan a global planning path from the taxiing start point to the taxiing end point;

Using a pre-established traction and taxiing controller, the global planning path is tracked and controlled so that the tractor can smoothly tow the target aircraft for taxiing;

During the traction sliding process, if an obstacle is detected in front of the sliding path, it is detected whether the obstacle can be avoided by slowing down;

If not, a new global planning path is replanned according to the current sliding position, obstacle information and sliding target point, and then the operation of using the pre-established traction sliding controller to perform trajectory tracking control on the global planning path is returned to the execution until the entire traction process is completed.

According to another aspect of an embodiment of the present invention, there is also provided a traction taxiing control device for a civil aircraft, comprising:

The global planning path planning module is used to control the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a fast-first-then-slow manner, and after the docking is completed, plan the global planning path from the taxiing start point to the taxiing end point;

A trajectory tracking control module is used to use a pre-established traction and taxiing controller to perform trajectory tracking control on the global planning path so that the tractor can stably tow the target aircraft for taxiing;

The avoidance detection module is used to detect whether the obstacle can be avoided by slowing down if an obstacle is detected in front of the sliding path during the traction sliding process;

The path replanning module is used to replan a new global planning path according to the current sliding position point, obstacle information and sliding target point, and then return to execute the operation of using the pre-established traction sliding controller to perform trajectory tracking control on the global planning path until the entire traction process is completed.

According to another aspect of an embodiment of the present invention, a tractor is further provided, the tractor comprising:

at least one processor; and

a memory communicatively connected to the at least one processor; wherein,

The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the traction taxiing control method for a civil aircraft described in any embodiment of the present invention.

According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the traction taxiing control method for a civil aircraft described in any embodiment of the present invention when executed.

The technical solution of the embodiment of the present invention controls the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a first fast and then slow manner, and after the docking is completed, plans a global planning path from the taxiing start point to the taxiing end point; adopts a pre-established traction taxiing controller to track the global planning path so that the tractor can stably tow the target aircraft for taxiing; during the traction taxiing process, if an obstacle is detected in front of the taxiing path, it is detected whether the obstacle can be avoided by slowing down; if not, according to the current taxiing position point, obstacle information and taxiing target point, a new global planning path is re-planned, and then returns to execute the operation of using the pre-established traction taxiing controller to track the global planning path until the entire traction process is completed. The technical means achieves the technical effect that after the wheel holding mechanism on the tractor accurately docks with the front landing gear of the target aircraft, the tractor stably tows the target aircraft to the taxiing end point, while ensuring the accuracy, speed and safety of the automatic docking of the tractor and the target aircraft, the safety and reliability of the tractor and the aircraft in the combined taxiing process are improved.

It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

FIG1 is a flow chart of a traction taxiing control method for a civil aircraft according to a first embodiment of the present invention;

FIG2 is a flow chart of a traction taxiing control method for a civil aircraft according to a second embodiment of the present invention;

3 is a flow chart of a method for calculating a target deflection angle of a tractor relative to a front landing gear according to a second embodiment of the present invention;

FIG4 is a front view of a tractor to which the technical solution of an embodiment of the present invention is applicable, with the tractor deflected at an angle α relative to the front landing gear;

FIG5 is a top view of a tractor to which the technical solution of an embodiment of the present invention is applicable, with the tractor deflected at an angle α relative to the front landing gear;

FIG6 is a schematic diagram of a dynamic model of a traction sliding system applicable to the technical solution of an embodiment of the present invention;

7 is a structural diagram of a traction taxiing control device for a civil aircraft provided in Embodiment 3 of the present invention;

FIG8 is a schematic diagram of a tractor provided in Embodiment 4 of the present invention.

DETAILED DESCRIPTION

In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

Embodiment 1

FIG1 is a flow chart of a towing taxiing control method for a civil aircraft provided in Embodiment 1 of the present invention. This embodiment is applicable to the case where a towing vehicle is used to automatically align the wheel-holding mechanism on the vehicle with the front landing gear of the target aircraft to be towed, and to tow the target aircraft to taxi to the target position after the docking is completed and the wheel-holding lifting is completed. The method can be executed by a towing taxiing control device of a civil aircraft, which can be implemented in the form of hardware and/or software, and can generally be configured in a towing vehicle and executed by a controller on the towing vehicle. As shown in FIG1 , the method includes:

S110, controlling the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a first fast and then slow manner, and after the docking is completed, planning a global planning path from the taxiing start point to the taxiing end point.

In this embodiment, in order to improve the speed and accuracy of the tractor (typically, an unmanned poleless tractor) docking with the target aircraft, a new implementation method is proposed to control the wheel holding mechanism on the tractor to gradually approach and dock with the front landing gear of the target aircraft at at least two speeds. Specifically, when the tractor is far away from the target aircraft, the tractor can be controlled to move quickly toward the direction where the target aircraft is located. When the tractor is close to the target aircraft, the tractor can be controlled to move slowly to accurately dock with the target aircraft.

Specifically, two or more groups of visual sensors can be pre-installed on the tractor, and each group of visual sensors is used to detect the relative distance and relative position relationship of target aircraft at different distances. Accordingly, the movement speed and movement direction of the tractor can be flexibly set according to the detection results of different groups of visual sensors to achieve an effective balance between docking speed and docking accuracy.

It is understandable that those skilled in the art can set the number of visual sensor groups and select the movement speed to be used at different distances according to actual conditions, and this embodiment does not limit this.

After the wheel-holding mechanism on the tractor is docked with the front landing gear of the target aircraft, the position of the whole composed of the tractor and the target aircraft (hereinafter referred to as the traction taxiing system) in the airport can be determined through the positioning module set on the target aircraft or the tractor as the taxiing starting point. At the same time, it is also necessary to determine the taxiing end point through air traffic control instructions, that is, the preparation point before the target aircraft takes off. After obtaining the taxiing start point and taxiing end point, by combining the airport map data and the preset path planning algorithm, the global planning path of the traction taxiing system from the taxiing start point to the taxiing end point can be planned.

Among them, when planning to obtain the global planning path, constraints such as minimum time or minimum turns can be adopted to obtain a global planning path that meets actual needs.

S120: Using a pre-established traction and taxiing controller, the global planning path is tracked and controlled so that the tractor can stably tow the target aircraft for taxiing.

After the global path planning is completed, the tractor can be controlled to steadily tow the target aircraft along the global planned path from the taxiing starting point to the taxiing end point.

In this embodiment, in order to ensure that the tractor can smoothly tow the target aircraft for taxiing, it is proposed to perform effective trajectory tracking control during the entire taxiing process. Specifically, a traction taxiing controller is first constructed, and by setting a cost control function and constraint conditions in the traction taxiing controller, the optimal control amount can be solved in different control cycles to effectively control the real-time speed and real-time turning angle of the tractor.

S130. During the traction sliding process, it is detected whether there is an obstacle in front of the sliding path. If so, execute S140; otherwise, return to execute S120 until the entire traction process is completed.

In this embodiment, considering the dynamic change characteristics of the airport environment, it is necessary to detect obstacles in front of the taxiing path in real time. If there is no obstacle in front, the taxiing can continue according to the original global planning path. If an obstacle appears in front, it is necessary to avoid the obstacle safely, for example, by slowing down or detouring.

Furthermore, considering that the implementation cost of deceleration and avoidance is much lower than that of detour and avoidance, in this embodiment, it is first detected whether the traction sliding system can safely avoid the obstacle by deceleration and avoidance: if so, it can continue to slide by decelerating and driving slowly while maintaining the existing global planned route and safely avoiding the obstacle; if not, it is necessary to re-plan a new global planned route from the current sliding position to the sliding end point based on information such as the position and speed of the obstacle, so as to safely avoid the obstacle by detouring.

It is understandable that the operation of S130 needs to be performed in real time during the entire taxiing process until the tractor successfully tows the target aircraft to the taxiing end point.

S140, detecting whether the obstacle can be avoided by decelerating and driving slowly: if so, executing S150; otherwise, executing S160.

In this embodiment, by combining the current position point, movement speed and direction of the obstacle, the movement speed and direction of the traction sliding system, and the deceleration performance of the traction sliding system and other parameters, it can be evaluated whether the traction sliding system can avoid the obstacle by slowing down while maintaining the existing global planned route.

Specifically, a neural network model can be pre-trained, the input of which is the current position, movement speed and direction of the obstacle, the movement speed, current position, deceleration performance parameters and global planned route of the traction sliding system, and the output is the identification result of whether the traction sliding system can avoid the obstacle by decelerating.

S150. After executing the deceleration and avoidance strategy, return to execute S120 until the entire traction process is completed.

Among them, the deceleration avoidance strategy can define the time period and deceleration method (for example, specific braking time or throttle setting value, etc.) in which the tractor needs to decelerate, or the time point in which the tractor needs to reduce the current speed to a certain speed value, etc.

Similarly, the deceleration and avoidance strategy can be directly output by the above-mentioned neural network model, or it can be calculated in real time according to a preset calculation formula, combined with the current position point, movement speed and movement direction of the obstacle, the movement speed, current position point, deceleration performance parameters of the traction sliding system, and the global planned route and other parameters.

Specifically, after the obstacle is successfully avoided by executing the deceleration avoidance strategy, the speed of the traction sliding system can be gradually increased according to the preset acceleration processing strategy to continue to move toward the sliding end point at a normal sliding speed.

S160, after replanning a new global planning path according to the current sliding position point, obstacle information and sliding target point, return to execute S120 until the entire traction process is completed.

In this embodiment, if the obstacle cannot be successfully avoided by deceleration alone, it is necessary to re-plan a new global planning route and use a pre-established traction and taxiing controller to continue trajectory tracking control of the newly planned global planning path so that the tractor can smoothly tow the target aircraft for taxiing.

It should be noted that after determining that the obstacle cannot be avoided by slowing down, the first avoidance cost of the traction taxiing system to avoid the obstacle and the second avoidance cost of the traction taxiing system to avoid the obstacle can be evaluated first. If the second avoidance cost is much greater than the first avoidance cost, the air traffic control center can be requested to prompt the obstacle to be avoided and at the same time execute the corresponding deceleration avoidance strategy.

The technical solution of the embodiment of the present invention controls the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a first fast and then slow manner, and after the docking is completed, plans a global planning path from the taxiing start point to the taxiing end point; adopts a pre-established traction taxiing controller to track the global planning path so that the tractor can stably tow the target aircraft for taxiing; during the traction taxiing process, if an obstacle is detected in front of the taxiing path, it is detected whether the obstacle can be avoided by slowing down; if not, according to the current taxiing position point, obstacle information and taxiing target point, a new global planning path is re-planned, and then returns to execute the operation of using the pre-established traction taxiing controller to track the global planning path until the entire traction process is completed. The technical means achieves the technical effect that after the wheel holding mechanism on the tractor accurately docks with the front landing gear of the target aircraft, the tractor stably tows the target aircraft to the taxiing end point, while ensuring the accuracy, speed and safety of the automatic docking of the tractor and the target aircraft, the safety and reliability of the tractor and the aircraft in the combined taxiing process are improved.

Embodiment 2

FIG2 is a flow chart of a towing taxiing control method for a civil aircraft provided in a second embodiment of the present invention. This embodiment is refined on the basis of the above embodiment. In this embodiment, the operation of controlling the wheel holding mechanism on the towing vehicle to approach and accurately dock with the front landing gear of the target aircraft in a fast-first-then-slow manner, and after the docking is completed, planning a global planning path from the taxiing start point to the taxiing end point is concretized.

Accordingly, as shown in FIG2 , the method includes:

S210, collecting a first front landing gear image of the target aircraft in real time through a long-distance visual camera arranged on the tractor, and obtaining a first real-time distance and relative position relationship between the tractor and the front landing gear according to the first front landing gear image.

In this embodiment, two groups of visual sensors are arranged on the tractor, and each group of visual sensors includes only one visual camera, so as to control the movement of the tractor at two speeds and dock with the target aircraft.

Optionally, a long-distance visual camera may be arranged near the center line of the tractor roof, and the front wheels of the aircraft (ie, the front landing gear of the aircraft) at a relatively long distance may be visually identified by the long-distance visual camera to obtain a first front landing gear image.

After acquiring the first front landing gear image, the image position of the front landing gear of the target aircraft and the pixel size occupied can be acquired through image recognition technology. After acquiring the above information, the mapping relationship between the real space and the pixel space of the long-distance vision camera can be determined based on the camera calibration result of the long-distance vision camera and the installation position of the long-distance vision camera in the tractor. Furthermore, the first real-time distance and relative position relationship between the tractor and the front landing gear can be identified based on the above mapping relationship.

S220, determine whether the first real-time distance is greater than a distance threshold: if so, execute S230; otherwise, execute S240.

In this embodiment, the distance threshold is used as the switching boundary of the tractor's moving speed. When the distance between the tractor and the front landing gear is greater than the distance threshold, the tractor can quickly approach the front landing gear of the target aircraft at a higher speed. When the distance between the tractor and the front landing gear is less than or equal to the distance threshold, the tractor can slowly approach the front landing gear of the target aircraft at a lower speed to take into account both the docking speed and the docking accuracy.

The distance threshold may be preset according to actual scenarios or requirements for docking speed or accuracy, and this embodiment does not limit this.

S230. According to the relative position relationship, control the wheel holding mechanism on the tractor to move toward the front landing gear at a first speed, and return to execute S210.

After the relative position relationship between the tractor and the front landing gear is acquired, the wheel holding mechanism on the tractor can be controlled to move toward the front landing gear at a first speed based on the relative position relationship to achieve rapid coarse alignment.

S240, collecting a second front landing gear image of the target aircraft in real time through a close-range visual camera arranged on the tractor, and calculating a target deflection angle of the tractor relative to the front landing gear according to the second front landing gear image.

Optionally, a close-range vision camera may be fixedly mounted on the frame in front of the wheel-holding mechanism on the central axis of the rodless tractor. The close-range vision camera performs a visual scan of the environment when working and quickly identifies the position of the front landing gear of the aircraft in front.

Specifically, the models and technical parameters of long-distance vision cameras and short-distance vision cameras can be selected according to actual application scenarios or application requirements. Generally speaking, the acquisition accuracy of short-distance vision cameras is generally higher than that of long-distance vision cameras.

When the wheel-holding mechanism on the tractor is very close to the front landing gear of the target aircraft, the relative position relationship between the wheel-holding mechanism and the front landing gear needs to be accurately determined so that the two can be accurately docked. In other words, the target deflection angle of the tractor relative to the front landing gear needs to be calculated.

In an optional implementation of this embodiment, as shown in FIG3 , the method of calculating the target deflection angle of the tractor relative to the front landing gear according to the second front landing gear image may include:

S2401. Identify a front landing gear contour in a second front landing gear image, and identify, based on features of the front landing gear contour, whether the deflection type of the tractor relative to the front landing gear is left deflection or right deflection.

In this embodiment, the nose landing gear contour can be identified in the second nose landing gear image by image recognition technology. Considering that the right side of the wheel hub can be seen in the second nose landing gear image with a left deflection angle, and the left side of the wheel hub can be seen in the second nose landing gear image with a right deflection angle, based on the features of the identified nose landing gear contour, it can be determined whether the deflection type of the tractor relative to the nose landing gear is left deflection or right deflection.

S2402. According to the front landing gear profile, obtain the maximum lateral dimension w', maximum longitudinal dimension d and front wheel width b of the aircraft's front wheel.

For the sake of convenience, FIG. 4 shows a tractor deflected relative to the front landing gear to which the technical solution of the embodiment of the present invention is applicable. FIG5 shows a front view of a tractor to which the technical solution of an embodiment of the present invention is applicable, and FIG6 shows a front view of a tractor to which the front landing gear is deflected relative to the front landing gear. Top view of the corner.

Specifically, the cross section of an aircraft tire is composed of complex arcs and line segments, and thus a cylindrical model with a fillet radius of R can be used to approximate a single tire, that is, an arc with a radius of R is used to fit the tire fillet. Generally speaking, the contour of the front landing gear identified in the second front landing gear image is shown in FIG5 , and therefore, the maximum lateral dimension w', the maximum longitudinal dimension d (that is, the diameter of the front wheel tire), and the width b of the front wheel tire of the aircraft can be directly measured in FIG5 . In the example shown in FIG5 , the deflection angle of the tractor relative to the front landing gear can be calculated by the following formulas: .

;

;

;in, .

Obviously, in order to calculate , there is an unknown quantity R that cannot be directly obtained from the front landing gear profile.

S2403. Determine a fixed proportional relationship among the tire diameter d1, tire width b1, and tire arc radius R1 of the front wheel of the aircraft according to the wheel and tire model of the target aircraft.

It is understandable that the civil aircraft of the set model has a specific model of aircraft wheels. Therefore, after the model of the target aircraft is determined, the wheel and tire model of the target aircraft is also uniquely determined. After the wheel and tire model of the target aircraft is determined, it can be known that there is a fixed proportional relationship between its tire diameter d1, tire width b1 and arc radius R1, and this proportional relationship still exists in the front landing gear profile identified in the second front landing gear image. Therefore, the proportional factor can be introduced. and ,Right now: ;

S2404. Calculate the deflection angle of the tractor relative to the front landing gear according to the maximum lateral dimension w', the maximum longitudinal dimension d, the front wheel width b and the fixed proportional relationship of the aircraft's front wheels.

By adding the scaling factor and Substitute the above The calculation formula can be used to obtain the deflection angle of the tractor relative to the front landing gear. .

Specifically, according to the maximum lateral dimension w', the maximum longitudinal dimension d, the width b of the front wheel of the aircraft, and the fixed proportional relationship, calculating the deflection angle of the tractor relative to the front landing gear may include:

According to the fixed proportional relationship, a first proportional factor is obtained. , and the second scaling factor ;

According to the first scale factor and the second scale factor , calculate the deflection adjustment factor ;

in, ;

According to the maximum lateral dimension w', maximum longitudinal dimension d, width b, and first proportional factor of the aircraft's front wheel , Second scale factor and the deflection adjustment factor , calculate the deflection angle of the tractor relative to the front landing gear ;

in, .

S2405: Use the combination of the deflection type and the deflection angle as a target deflection angle of the tractor relative to the front landing gear.

The target deflection angle describes the deflection angle of the tractor relative to the front landing gear in a left-biased or right-biased manner.

S250, controlling the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft at a second speed according to the target deflection angle, wherein the first speed is greater than the second speed.

After the target deviation angle is accurately determined by the close-range visual camera, the wheel-holding mechanism on the tractor can be controlled to approach and accurately dock with the front landing gear of the target aircraft at the second speed.

Similarly, those skilled in the art may customize the first speed and the second speed according to actual application scenarios and actual needs, and this embodiment does not limit this.

It should be noted that when the wheel-holding mechanism on the tractor is automatically docked with the front landing gear of the target aircraft, the wheel-holding mechanism can lift the front wheels of the aircraft and drive the target aircraft to taxi after receiving the next taxiing instruction.

S260: Determine the taxiing starting point through the positioning module on the tractor, and identify the taxiing end point in the received air traffic control instruction.

S270, use As a heuristic function, the global optimal path from the taxiing start point to the taxiing end point is planned in combination with the taxiing start point, the taxiing end point and the airport map as the global planning path.

In this embodiment, the existing A* algorithm can be improved to complete the planning of the global optimal path.

in, As the inspiration factor, The value of is determined by the size of the airport map; The angle between the vector formed by the current node and the starting point and the vector formed by the end point and the current node; It is the angle between the vector formed by the previous node and the current node and the vector formed by the next node and the current node; is the safety factor, The value of The size is determined by , are adjustment parameters, The value of The size is determined by The value of The size is determined.

Specifically, first obtain the coordinate information of the taxi start point and the taxi end point in the airport map, and then calculate the global optimal path based on the airport map and the cost evaluation function f(n). Among them, the search can be performed according to the cost evaluation function f(n), and the searched point that satisfies the search is used as the starting point of the next search, and this process is repeated until the taxi end point is found, and the points found in the search process form the optimal path.

Optionally, the cost evaluation function can be defined as: ;

Among them, f(n) is the cost evaluation function of the current node n, g(n) is the actual cost from the sliding starting point to the current node n, is the cost from the current node to the target point, which is also the heuristic function. In order to improve the calculation efficiency and the safety of the sliding process, it is necessary to consider the search direction and the angle information during the sliding process. The heuristic function constructed is:

.

in, is the heuristic factor, which determines the weight of the direction information in the traction sliding process. Its value is determined according to the size of the actual map and is generally 5; It is the angle between the vector formed by the current node and the starting point and the vector formed by the end point and the current node, which reflects that the search direction is pointing to the end point of the sliding. The smaller the value, the closer the search is to the end of the glide path; It is the angle between the vector formed by the previous node and the current node and the vector formed by the next node and the current node. The larger it is, the greater the angle is, and the higher the risk factor is; is the safety factor, which determines the influence weight of the longitudinal angle between the tractor and the target aircraft on taxiing safety.

when Angle less than hour, The value is 0. hour, The value is 5. hour, The value is 10. hour The value is 20; , are all preset parameters. Small hour, is 0; when Greater than hour, is 1; when Less than hour, is 0; when Greater than hour, is 1.

S280: Using a pre-established traction and taxiing controller, the global planning path is tracked and controlled so that the tractor can stably tow the target aircraft for taxiing.

Optionally, using a pre-established traction and sliding controller to perform trajectory tracking control on the global planning path may include:

As shown in Figure 6, in the geodetic coordinate system XOY, the center of the front axle of the tractor is defined as point A, the center of the rear axle is defined as point B, the front and rear wheel tracks of the tractor are defined as L1, the hinge point between the tractor and the target aircraft is defined as point H, and the front wheel turning angle of the tractor is defined as The midpoint of the line connecting the two main landing gears of the target aircraft is point C, the distance from point C to the hinge point is L2, the distance from the rear axle center point B to the hinge point H is La, and the yaw angle of the tractor is , the yaw angle of the target aircraft is , the articulation angle between the tractor and the target aircraft is , and obtain the schematic diagram of the dynamic model of the traction sliding system.

by is the state quantity, Model the aircraft traction for the control quantity:

;in, represents the derivative with respect to Z;

According to the aircraft traction model, the state space expression of the four-degree-of-freedom linear error model is constructed: ;in, ;

; ;

; ;

;

;

The state space expression of the four-degree-of-freedom linear error model is discretized, and the discretized model is obtained under the sampling period T:

;

in, , , ; is the state quantity at the current time t0, It is the state quantity at time t0+T.

Constructing control increments within each control cycle , and the discrete state quantity and control increment , combined into a new state quantity , the new state space equation is:

;

in, , , ;

Let Np be the prediction time domain step size, Nc be the control time domain step size, and Nc Np, the target cost function is constructed as:

;

in, It is the reference value at time k+i predicted at time k; is the actual value of the output at time k+i at time k; is the control increment at time k+i calculated at time k; is the weight coefficient of the relaxation factor, which takes a value of 1;

in, ;

;

; is the preset relaxation factor; where Q and R are weight matrices, , , , , as well as They are all coefficients in the weight matrix, and their values are all 1;

Based on the target cost function and the following two constraints:

,as well as , a traction and sliding controller is constructed;

Through the traction sliding controller, based on the two constraints, the target cost function is solved in each control cycle to obtain the following optimal control increment sequence:

;

Through the traction sliding controller, the first control increment in the optimal control increment sequence is added to the control amount at the previous moment as the control amount input at the current moment. The above process is repeated in the next control cycle, and online rolling optimization is performed to obtain the actual control amount at each moment, and finally the trajectory tracking control of the system is realized.

S290: During the traction sliding process, if an obstacle is detected in front of the sliding path, it is detected whether the obstacle can be avoided by slowing down.

In this embodiment, when the tractor and the target aircraft are controlled to travel according to the global planned path based on the optimal global path planning, obstacles are sensed in real time to adjust the local path in real time to generate a smooth path.

Specifically, the positions of the aircraft and the tractor in the airport can be first determined using the positioning module, and the profile of the aircraft traction system can be constructed by storing the aircraft parameters and the tractor parameters in the module and combining the articulation angle information measured in the sensor module.

Afterwards, the sensor module can be used to detect in real time whether there are obstacles on the taxiing path of the aircraft traction system. When there are obstacles, the processing and judgment module determines the safety risk based on the speed and size of the obstacle and the speed and size of the aircraft traction system. Due to the high turning risk factor of the aircraft traction system, slow deceleration is preferred to avoid it. When deceleration cannot avoid it, the taxiing path needs to be replanned. That is, when the processing and judgment module concludes that the aircraft cannot avoid it by decelerating, it is necessary to use the current node as the initial node, add the information of the obstacle to the airport map, and replan a new taxiing path.

In an optional implementation of this embodiment, detecting whether the obstacle can be avoided by slowing down may include:

The moving speed of the obstacle and the current traction and sliding speed are obtained, and the sum of the moving speed and the traction and sliding speed is calculated; if the sum of the speeds does not exceed a preset speed threshold, it is determined that the obstacle can be avoided by slowing down; otherwise, it is determined that the obstacle cannot be avoided by slowing down.

In a specific example, when the sum of the obstacle's moving speed and the current traction taxiing speed is less than 50km/h, slow deceleration can be used as a preferred method for avoidance; when the sum of the obstacle's moving speed and the current traction taxiing speed is greater than 50km/h, deceleration may not be able to avoid it. At this time, the obstacle's position coordinates and speed need to be updated to the airport map.

S2100, if not, then re-plan a new global planning path according to the current sliding position point, obstacle information and sliding target point, and then return to execute the operation of using the pre-established traction sliding controller to perform trajectory tracking control on the global planning path until the entire traction process is completed.

The technical solution of the embodiment of the present invention controls the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a first fast and then slow manner, and after the docking is completed, plans a global planning path from the taxiing start point to the taxiing end point; adopts a pre-established traction taxiing controller to track the global planning path so that the tractor can smoothly tow the target aircraft for taxiing; during the traction taxiing process, if an obstacle is detected in front of the taxiing path, it is detected whether the obstacle can be avoided by slowing down; if not, according to the current taxiing position point, obstacle information and taxiing target point, a new global planning path is re-planned, and then returns to execute the operation of using the pre-established traction taxiing controller to track the global planning path until the entire traction process is completed. The technical means achieves the technical effect that after the wheel holding mechanism on the tractor accurately docks with the front landing gear of the target aircraft, the tractor smoothly tows the target aircraft to the taxiing end point, while ensuring the accuracy, rapidity and safety of the automatic docking of the tractor and the target aircraft, the safety and reliability of the tractor and the aircraft in the combined taxiing process are improved.

Embodiment 3

FIG7 is a schematic diagram of the structure of a traction environment control device for a civil aircraft provided in Embodiment 3 of the present invention. As shown in FIG7 , the device includes: a global planning path planning module 710, a trajectory tracking control module 720, an avoidance detection module 730, and a path re-planning module 740, wherein:

The global planning path planning module 710 is used to control the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a first fast and then slow manner, and after the docking is completed, plan a global planning path from the taxiing start point to the taxiing end point;

The trajectory tracking control module 720 is used to use a pre-established traction and taxiing controller to perform trajectory tracking control on the global planning path so that the tractor can stably tow the target aircraft for taxiing;

The avoidance detection module 730 is used to detect whether the obstacle can be avoided by slowing down if an obstacle is detected in front of the sliding path during the traction sliding process;

The path replanning module 740 is used to, if not, replan a new global planning path based on the current sliding position point, obstacle information and sliding target point, and then return to execute the operation of using the pre-established traction sliding controller to perform trajectory tracking control on the global planning path until the entire traction process is completed.

The technical solution of the embodiment of the present invention controls the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a first fast and then slow manner, and after the docking is completed, plans a global planning path from the taxiing start point to the taxiing end point; adopts a pre-established traction taxiing controller to track the global planning path so that the tractor can stably tow the target aircraft for taxiing; during the traction taxiing process, if an obstacle is detected in front of the taxiing path, it is detected whether the obstacle can be avoided by slowing down; if not, according to the current taxiing position point, obstacle information and taxiing target point, a new global planning path is re-planned, and then returns to execute the operation of using the pre-established traction taxiing controller to track the global planning path until the entire traction process is completed. The technical means achieves the technical effect that after the wheel holding mechanism on the tractor accurately docks with the front landing gear of the target aircraft, the tractor stably tows the target aircraft to the taxiing end point, while ensuring the accuracy, speed and safety of the automatic docking of the tractor and the target aircraft, the safety and reliability of the tractor and the aircraft in the combined taxiing process are improved.

Based on the above embodiments, the global planning path planning module 710 may specifically include:

A first image recognition unit is used to collect a first front landing gear image of the target aircraft in real time through a long-distance visual camera arranged on the tractor, and obtain a first real-time distance and relative position relationship between the tractor and the front landing gear according to the first front landing gear image;

A first speed moving unit, configured to control the wheel holding mechanism on the tractor to move toward the front landing gear at a first speed according to the relative position relationship if the first real-time distance is greater than a distance threshold;

a second image recognition unit, configured to collect a second front landing gear image of the target aircraft in real time through a close-range visual camera provided on the tractor if the first real-time distance is less than or equal to the distance threshold, and calculate a target deflection angle of the tractor relative to the front landing gear according to the second front landing gear image;

The second speed moving unit is used to control the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft at a second speed according to the target deflection angle, wherein the first speed is greater than the second speed.

On the basis of the above embodiments, the second image recognition unit is specifically used for:

recognizing a front landing gear profile in the second front landing gear image, and recognizing, based on features of the front landing gear profile, whether the deflection type of the tractor relative to the front landing gear is left deflection or right deflection;

According to the front landing gear profile, obtain the maximum lateral dimension w', the maximum longitudinal dimension d and the width b of the front wheel of the aircraft;

According to the wheel and tire model of the target aircraft, a fixed proportional relationship between the tire diameter d1, the tire width b1 and the tire arc radius R1 of the front wheel of the aircraft is determined;

Calculate the deflection angle of the tractor relative to the front landing gear according to the maximum lateral dimension w', the maximum longitudinal dimension d, the width b of the front wheel of the aircraft and the fixed proportional relationship;

A combination of the deflection type and the deflection angle is used as a target deflection angle of the tractor relative to the front landing gear.

On the basis of the above embodiments, the second image recognition unit is further specifically used for:

According to the fixed proportional relationship, a first proportional factor is obtained. , and the second scaling factor ;

According to the first scale factor and the second scale factor , calculate the deflection adjustment factor ;

in, ;

According to the maximum lateral dimension w', maximum longitudinal dimension d, width b, and first proportional factor of the aircraft's front wheel , Second scale factor and the deflection adjustment factor , calculate the deflection angle of the tractor relative to the front landing gear ;

in, .

Based on the above embodiments, the global planning path planning module 710 can be further specifically used for:

Determine the taxiing start point through the positioning module on the tractor, and identify the taxiing end point in the received air traffic control instructions;

use As a heuristic function, the global optimal path from the taxi start point to the taxi end point is planned in combination with the taxi start point, the taxi end point and the airport map as the global planning path;

in, As the inspiration factor, The value of is determined by the size of the airport map; The angle between the vector formed by the current node and the starting point and the vector formed by the end point and the current node; It is the angle between the vector formed by the previous node and the current node and the vector formed by the next node and the current node; is the safety factor, The value of The size is determined by , are adjustment parameters, The value of The size is determined by The value of The size is determined.

Based on the above embodiments, the trajectory tracking control module 720 can be specifically used for:

In the geodetic coordinate system XOY, the center of the front axle of the tractor is defined as point A, the center of the rear axle is defined as point B, the front and rear wheelbase of the tractor is defined as L1, the hinge point between the tractor and the target aircraft is defined as point H, and the front wheel turning angle of the tractor is defined as The midpoint of the line connecting the two main landing gears of the target aircraft is point C, the distance from point C to the hinge point is L2, the distance from the rear axle center point B to the hinge point H is La, and the yaw angle of the tractor is , the yaw angle of the target aircraft is , the articulation angle between the tractor and the target aircraft is ;

by is the state quantity, Model the aircraft traction for the control quantity:

;in, represents the derivative with respect to Z;

According to the aircraft traction model, the state space expression of the four-degree-of-freedom linear error model is constructed: ;in, ;

; ;

; ;

;

;

The state space expression of the four-degree-of-freedom linear error model is discretized, and the discretized model is obtained under the sampling period T:

;

in, , , ;

Constructing control increments within each control cycle , and the discrete state quantity and control increment , combined into a new state quantity , the new state space equation is:

;

in, , , ;

Let Np be the prediction time domain step size, Nc be the control time domain step size, and Nc Np, the target cost function is constructed as:

;

in, It is the reference value at time k+i predicted at time k; is the actual value of the output at time k+i at time k; is the control increment at time k+i calculated at time k; is the weight coefficient of the relaxation factor, which takes a value of 1;

in, ;

;

; is the preset relaxation factor; where Q and R are weight matrices, , , , , as well as They are all coefficients in the weight matrix, and their values are all 1;

Based on the target cost function and the following two constraints:

,as well as , a traction and sliding controller is constructed;

Through the traction sliding controller, based on the two constraints, the target cost function is solved in each control cycle to obtain the following optimal control increment sequence:

;

Through the traction sliding controller, the first control increment in the optimal control increment sequence is added to the control amount at the previous moment as the control amount input at the current moment. The above process is repeated in the next control cycle, and online rolling optimization is performed to obtain the actual control amount at each moment, and finally the trajectory tracking control of the system is realized.

Based on the above embodiments, the avoidance detection module 730 can be specifically used for:

Obtaining the moving speed of the obstacle and the current traction sliding speed, and calculating the speed sum of the moving speed and the traction sliding speed;

If the speed sum value does not exceed the preset speed threshold, it is determined that the obstacle can be avoided by slowing down; otherwise, it is determined that the obstacle cannot be avoided by slowing down.

The traction taxiing control device for a civil aircraft provided in an embodiment of the present invention can execute the traction taxiing control method for a civil aircraft provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

Embodiment 4

FIG8 shows a schematic diagram of the appearance of a tractor applicable to the fourth embodiment of the present invention. As shown in FIG8, 1-wheel holding mechanism, 2-close-range visual camera, 3-accelerometer gyroscope, 4-controller, 5, 6-laser radar ranging sensor, 7-angle sensor, 8-long-range visual camera.

In this embodiment, the tractor mainly includes a wheel-holding mechanism, a controller, two visual cameras and other sensors; the wheel-holding mechanism is located near the center of the rodless tractor and is used to clamp and lift the front wheel of the aircraft.

The controller includes a positioning module, a storage module, at least one processor and a motion controller; the positioning module is used to determine the positions of the aircraft and the tractor at the airport, the storage module is used to store aircraft parameters and tractor parameters, and the at least one processor is used to execute the method according to any embodiment of the present invention, that is:

The wheel holding mechanism on the tractor is controlled to approach and accurately dock with the front landing gear of the target aircraft in a fast-first-then-slow manner, and after the docking is completed, a global planning path from the taxiing start point to the taxiing end point is planned; a pre-established traction taxiing controller is used to perform trajectory tracking control on the global planning path, so that the tractor can stably tow the target aircraft for taxiing; during the traction taxiing process, if an obstacle is detected in front of the taxiing path, it is detected whether the obstacle can be avoided by slowing down; if not, a new global planning path is re-planned according to the current taxiing position point, obstacle information and taxiing target point, and then the operation of using the pre-established traction taxiing controller to perform trajectory tracking control on the global planning path is returned to execute until the entire traction process is completed.

The motion controller is used to control the wheel holding mechanism to approach the front landing gear quickly and then slowly, as well as to control the direction and throttle of the tractor during the towing taxiing process.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A method for controlling the traction taxiing of a civil aircraft, characterized by comprising: Control the wheel-holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a fast-first-and-slow-later manner, and after docking, plan a global planning path from the taxiing start point to the taxiing end point; Using a pre-established traction and taxiing controller, the global planning path is tracked and controlled so that the tractor can smoothly tow the target aircraft for taxiing; During the traction sliding process, if an obstacle is detected in front of the sliding path, it is detected whether the obstacle can be avoided by slowing down; If not, after replanning a new global planning path according to the current sliding position point, obstacle information and sliding target point, return to execute the operation of using the pre-established traction sliding controller to perform trajectory tracking control on the global planning path until the entire traction process is completed; If yes, after executing the deceleration avoidance strategy, the process returns to executing the operation of using the pre-established traction sliding controller to perform trajectory tracking control on the global planning path until the entire traction process is completed, wherein the deceleration avoidance strategy includes when and in what deceleration mode the tractor needs to decelerate, and the deceleration mode includes the braking time or the throttle setting value; Among them, detecting whether the obstacle can be avoided by slowing down includes: The current position, speed and direction of the obstacle, the speed, current position, deceleration performance parameters and global planning route of the traction sliding system are input into a pre-trained neural network model, and the neural network model outputs an identification result of whether the obstacle can be avoided by slowing down; The method further comprises: If the recognition result is that the obstacle cannot be avoided by slowing down, the first avoidance cost of the traction taxiing system to avoid the obstacle and the second avoidance cost of the traction taxiing system to avoid the obstacle are calculated. If the second avoidance cost is much greater than the first avoidance cost, the air traffic control center is requested to prompt the obstacle to be avoided and at the same time execute the deceleration avoidance strategy. 2. The method according to claim 1 is characterized in that controlling the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a manner of first fast and then slow, comprises: The first front landing gear image of the target aircraft is collected in real time by a long-distance visual camera arranged on the tractor, and the first real-time distance and relative position relationship between the tractor and the front landing gear are obtained according to the first front landing gear image; If the first real-time distance is greater than the distance threshold, controlling the wheel holding mechanism on the tractor to move toward the front landing gear at a first speed according to the relative position relationship; If the first real-time distance is less than or equal to the distance threshold, a close-range visual camera provided on the tractor is used to collect a second front landing gear image of the target aircraft in real time, and a target deflection angle of the tractor relative to the front landing gear is calculated based on the second front landing gear image; The wheel-holding mechanism on the tractor is controlled to approach and accurately dock with the front landing gear of the target aircraft at a second speed according to the target deflection angle, wherein the first speed is greater than the second speed. 3. The method according to claim 2, characterized in that calculating the target deflection angle of the tractor relative to the front landing gear according to the second front landing gear image comprises: recognizing a front landing gear profile in the second front landing gear image, and recognizing, based on features of the front landing gear profile, whether the deflection type of the tractor relative to the front landing gear is left deflection or right deflection; According to the front landing gear profile, obtain the maximum lateral dimension w', the maximum longitudinal dimension d and the width b of the front wheel of the aircraft; According to the wheel and tire model of the target aircraft, a fixed proportional relationship between the tire diameter d1, the tire width b1 and the tire arc radius R1 of the front wheel of the aircraft is determined; Calculate the deflection angle of the tractor relative to the front landing gear according to the maximum lateral dimension w', the maximum longitudinal dimension d, the width b of the front wheel of the aircraft and the fixed proportional relationship; A combination of the deflection type and the deflection angle is used as a target deflection angle of the tractor relative to the front landing gear. 4. The method according to claim 3, characterized in that the deflection angle of the tractor relative to the front landing gear is calculated according to the maximum lateral dimension w', the maximum longitudinal dimension d, the front wheel width b and the fixed proportional relationship of the front wheel of the aircraft, comprising: According to the fixed proportional relationship, a first proportional factor is obtained. , and the second scaling factor ; According to the first scale factor and the second scale factor , calculate the deflection adjustment factor ; in, ; According to the maximum lateral dimension w', maximum longitudinal dimension d, width b, and first proportional factor of the aircraft's front wheel , Second scale factor and the deflection adjustment factor , calculate the deflection angle of the tractor relative to the front landing gear ; in, . 5. The method according to claim 1, wherein planning a global planning path from a taxiing start point to a taxiing end point comprises: Determine the taxiing start point through the positioning module on the tractor, and identify the taxiing end point in the received air traffic control instructions; use As a heuristic function, the global optimal path from the taxi start point to the taxi end point is planned in combination with the taxi start point, the taxi end point and the airport map as the global planning path; in, As the inspiration factor, The value of is determined by the size of the airport map; The angle between the vector formed by the current node and the starting point and the vector formed by the end point and the current node; It is the angle between the vector formed by the previous node and the current node and the vector formed by the next node and the current node; is the safety factor, The value of The size is determined by , are adjustment parameters, The value of The size is determined by The value of The size is determined. 6. The method according to claim 1, characterized in that the pre-established traction sliding controller is used to perform trajectory tracking control on the global planning path, comprising: In the geodetic coordinate system XOY, the center of the front axle of the tractor is defined as point A, the center of the rear axle is defined as point B, the front and rear wheelbase of the tractor is defined as L1, the hinge point between the tractor and the target aircraft is defined as point H, and the front wheel turning angle of the tractor is defined as The midpoint of the line connecting the two main landing gears of the target aircraft is point C, the distance from point C to the hinge point is L2, the distance from the rear axle center point B to the hinge point H is La, and the yaw angle of the tractor is , the yaw angle of the target aircraft is , the articulation angle between the tractor and the target aircraft is ; by is the state quantity, Model the aircraft traction for the control quantity: ;in, The table takes the derivative with respect to Z; According to the aircraft traction model, the state space expression of the four-degree-of-freedom linear error model is constructed: ;in, ; ; ; ; ; ; ; The state space expression of the four-degree-of-freedom linear error model is discretized, and the discretized model is obtained under the sampling period T: ; in, , , ; Constructing control increments within each control cycle , and the discrete state quantity and control increment , combined into a new state quantity , the new state space equation is: ; in, , , ; Let Np be the prediction time domain step size, Nc be the control time domain step size, and Nc Np, the target cost function is constructed as: ; in, It is the reference value at time k+i predicted at time k; is the actual value of the output at time k+i at time k; is the control increment at time k+i calculated at time k; is the weight coefficient of the relaxation factor, which takes a value of 1; in, ; ; ; is the preset relaxation factor; where Q and R are weight matrices, , , , , as well as They are all coefficients in the weight matrix, and their values are all 1; Based on the target cost function and the following two constraints: ,as well as , a traction and sliding controller is constructed; Through the traction sliding controller, based on the two constraints, the target cost function is solved in each control cycle to obtain the following optimal control increment sequence: ; Through the traction sliding controller, the first control increment in the optimal control increment sequence is added to the control amount at the previous moment as the control amount input at the current moment. The above process is repeated in the next control cycle, and online rolling optimization is performed to obtain the actual control amount at each moment, and finally the trajectory tracking control of the system is realized. 7. The method according to claim 1, characterized in that detecting whether the obstacle can be avoided by slowing down comprises: Obtaining the moving speed of the obstacle and the current traction sliding speed, and calculating the speed sum of the moving speed and the traction sliding speed; If the speed sum value does not exceed the preset speed threshold, it is determined that the obstacle can be avoided by slowing down; otherwise, it is determined that the obstacle cannot be avoided by slowing down. 8. A traction taxiing control device for a civil aircraft, characterized by comprising: The global planning path planning module is used to control the wheel holding mechanism on the tractor to approach and accurately dock with the front landing gear of the target aircraft in a fast-first-then-slow manner, and after the docking is completed, plan the global planning path from the taxiing start point to the taxiing end point; A trajectory tracking control module is used to use a pre-established traction and taxiing controller to perform trajectory tracking control on the global planning path so that the tractor can stably tow the target aircraft for taxiing; The avoidance detection module is used to detect whether the obstacle can be avoided by slowing down if an obstacle is detected in front of the sliding path during the traction sliding process; A path re-planning module, which is used to, if not, re-plan a new global planning path according to the current sliding position point, obstacle information and sliding target point, and then return to execute the operation of using the pre-established traction sliding controller to perform trajectory tracking control on the global planning path until the entire traction process is completed; a deceleration avoidance module, for returning to the operation of using a pre-established traction and sliding controller to perform trajectory tracking control on the global planning path after executing the deceleration avoidance strategy, until the entire traction process is completed, wherein the deceleration avoidance strategy includes when and in what deceleration mode the tractor needs to decelerate, and the deceleration mode includes a braking time or a throttle setting value; The avoidance detection module is specifically used to: input the current position point, movement speed and movement direction of the obstacle, the movement speed, current position point, deceleration performance parameters and global planning route of the traction sliding system into a pre-trained neural network model, and the neural network model outputs the recognition result of whether the obstacle can be avoided by slowing down; The device also includes: an obstacle avoidance module, which is used to: if the recognition result is that the obstacle cannot be avoided by slowing down, calculate the first avoidance cost of the traction taxiing system for avoiding the obstacle, and the second avoidance cost of the traction taxiing system for avoiding the obstacle; if the second avoidance cost is much greater than the first avoidance cost, request the air traffic control center to prompt the obstacle to be avoided, and simultaneously execute the deceleration avoidance strategy. 9. A tractor, characterized in that the tractor comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the traction taxiing control method for a civil aircraft according to any one of claims 1 to 7. 10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the traction taxiing control method for a civil aircraft according to any one of claims 1 to 7 when executed.