Method for conflict determination between aircraft and vehicle in the same location based on multi-keypoint detection

Abstract

The application discloses a method for judging the conflict between an aircraft and a vehicle in a parking position based on multi-key point detection, which comprises the following steps: obtaining the images of the aircraft and the support vehicle in the parking position through a parking position monitoring camera; identifying the aircraft and the support vehicle in the parking position respectively by using a target detection algorithm, and obtaining the position information of the aircraft and the support vehicle in the parking position by using a coordinate conversion algorithm; drawing the safety zone of the aircraft nose, engine, wing tip, tail wing tip and the safety zone of the support vehicle based on an aircraft parking deceleration model and a support vehicle motion model; judging whether the support vehicle invades the safety zone of the detected aircraft according to the safety zone of the detected aircraft, and issuing a warning if the invasion occurs. Through the technical scheme provided by the application, the target pixel coordinates can be converted into the actual coordinates of the parking position in real time, the conflict judgment detection precision is improved, and the fine identification of the key components of the aircraft and the conflict pre-warning are realized.

Description

Technical Field

[0001] This invention relates to the field of aviation, specifically to a method for determining conflicts between aircraft and vehicles within a parking position based on multi-keypoint detection. Background Technology

[0002] Airports, as crucial nodes in aviation operations, handle numerous tasks such as aircraft taxiing, takeoff, and landing. Coupled with the large number of people, vehicles, and obstacles within airports, dangerous close encounters and even collisions are frequent occurrences. The apron, being the most complex operational environment, requires special attention. Data shows that 96% of dangerous situations at airports occur on the apron, and globally, losses due to aircraft damage, injuries, and flight cancellations or delays caused by unsafe apron incidents exceed US$10 billion annually. Therefore, effectively providing early warnings of potential conflicts between aircraft and support vehicles on the apron is of paramount importance for ensuring airport operational safety.

[0003] With the advent of deep learning technology and the emergence of numerous object detection algorithms such as Fast R-CNN, Faster R-CNN, YOLO, SSD, and Retina-Net, research in this field has seen explosive growth. Currently, most research focuses on improving the accuracy of surface target detection, such as optimizing the YOLO algorithm's network structure and loss function, and improving the SSD algorithm's basic features and extraction network. These approaches can effectively enhance the target detection capabilities of airport remote control towers, but such research has not addressed the detection of critical aircraft components.

[0004] In summary, current research focuses primarily on optimizing traditional target detection algorithms. However, these algorithms can only identify the aircraft as a whole and cannot perform refined identification of critical components prone to conflict, such as wingtips and engines. Furthermore, while there is a wealth of literature on taxiway and runway conflicts, little attention is paid to conflicts that may occur on the apron, especially during aircraft taxiing into parking positions. Currently, there is no commercially available technology suitable for refined determination of conflicts between aircraft and vehicles within parking positions. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for determining aircraft and vehicle conflicts within a parking position based on multi-keypoint detection, comprising the following steps:

[0006] Step 1: Obtain images of aircraft and support vehicles on the tarmac using the aircraft stand monitoring camera;

[0007] Step 2: Use target detection algorithms to identify aircraft and support vehicles within the parking position, and use coordinate transformation algorithms to obtain the position information of aircraft and support vehicles on the apron.

[0008] Step 3: Based on the aircraft deceleration model and the support vehicle motion model, the safety zones for the aircraft's nose, engine, wingtips, horizontal stabilizer wingtips, and support vehicles are delineated.

[0009] Step 4: Security Zone Conflict Determination and Alarm. Based on the security zone of the detected aircraft, determine whether the support vehicle has intruded into the security zone of the detected aircraft. If intrusion occurs, an alarm will be issued.

[0010] Furthermore, the method of obtaining the position information of each aircraft and support vehicle within the apron through coordinate transformation algorithm includes:

[0011] The acquired pixel coordinates are transformed into and vice versa, based on the target pixel coordinates (x, y) in the image data. a ,y a The actual coordinates (x, y) of the detected target within the apron are calculated using the following coordinate transformation formula:

[0012]

[0013] s c p1, p2, p3, p4, p5.

[0014] Furthermore, the aircraft deceleration model for entering position includes: the aircraft deceleration model for entering position is divided into five stages:

[0015] During the reaction phase, the distance γ1 (m) traveled by the aircraft satisfies γ1 = v0t1, where v0 is the aircraft's taxiing speed when it is not braking, and t1 (s) is the reaction time required by the pilot.

[0016] During the initial braking phase, the pilot begins to depress the pedals, the aircraft decelerates a = 0, and travels a distance γ2 = v0t2, where t2(s) is the braking system transmission time;

[0017] During the deceleration growth phase, the aircraft begins to brake, and the deceleration 'a' increases from 0 to 'a'. max During this stage, the aircraft travels a certain distance. t3 is the time for the braking force to increase, a max This represents the maximum aircraft deceleration after the brake pedal is depressed.

[0018] During the braking hold phase, the aircraft's deceleration will remain at its maximum until the end of this phase.

[0019]

[0020] During the parking phase, the aircraft deceleration a = 0, and the aircraft displacement is 0.

[0021] The aircraft deceleration is at its maximum when it is on the main road surface and anti-skid braking is in use; at this time, a d =7.37m / s 2 ;

[0022] The deceleration is minimal on a wet track surface without anti-skid braking; at this point, a w =2.38m / s 2 ;

[0023] The aircraft braking distance is then γ1+γ2+γ3+γ4. Since t3 is extremely small, the formula can be further simplified to:

[0024]

[0025]

[0026] Where Γ min For the minimum braking distance, Γ max This is the maximum braking distance.

[0027] Furthermore, the aforementioned vehicle motion model includes:

[0028] Based on the vehicle motion model, a vehicle kinematic model is constructed, from which we can obtain:

[0029]

[0030]

[0031] The formula for calculating the sliding angle ξ is as follows:

[0032]

[0033] When a vehicle brakes, its total stopping distance can be divided into reaction distance and braking distance. Within the driver's reaction time, the vehicle travels at a V-shaped distance. cr When traveling at a constant speed, the reaction distance h1 satisfies:

[0034] h1 = t r ×V cr

[0035] Where t r This refers to the driver's reaction time;

[0036] Braking distance h2 satisfies:

[0037] h2=f(M,V cr )

[0038] Where M represents the distance between the vehicles, and V cr Let F represent the vehicle's center of gravity speed; during braking, the work done by the braking force F should be the same as the change in kinetic energy, i.e., satisfying:

[0039]

[0040] Where F = M × a c When the maximum braking force is applied to the vehicle, the deceleration remains constant, and the stopping distance H of the vehicle satisfies:

[0041] H = h1 + h2.

[0042] Furthermore, the aforementioned safety zones for the aircraft's nose, engine, wingtips, horizontal stabilizer wingtips, and support vehicle are defined based on the aircraft's deceleration model and support vehicle motion model, including:

[0043] The nose safety zone of an aircraft includes the nose warning zone and the nose dynamic warning zone. The nose warning zone is as follows:

[0044] The minimum clearance at the apron is λ, and the maximum ground turning angle of the aircraft is... A fan-shaped safety zone with radius ρ is drawn at the nose of the aircraft, and the maximum braking distance Γ required by the aircraft is... max The sum of the minimum clearances from the apron is the radius R of the nose warning zone. N ;

[0045] The radius of the dynamic warning zone for the nose is: R ′ N =R N +Γ min ;

[0046] The wingtip safety area of ​​an aircraft includes the wingtip safety area when the aircraft is stationary and the wingtip safety area when the aircraft is in motion. When the aircraft is stationary, the area is maintained at L. W ×L W The alarm range, of which L W It is twice the minimum clearance of the apron, i.e., L W =λ×2;

[0047] When the aircraft is in motion, the length of the longer side of the warning zone The size of the alarm zone is Length of the long side of the warning zone The size of the alarm zone is

[0048] The aircraft engine safety zone includes the engine intake safety zone and the engine exhaust safety zone. The size of the aircraft engine intake warning zone is as follows:

[0049]

[0050] For the engine dynamic warning zone, the length of the intake warning zone is extended to the minimum braking distance of the aircraft. minThe rectangular area forms an area of ​​size The safe zone is designated as the engine dynamic early warning zone;

[0051] The safety zone for the vehicle is based on twice the aircraft's horizontal stabilizer sweep angle, supplemented by 2×(180°). ° -ω), with the intersection of the extended lines of the two hypotenuses of the aircraft's horizontal stabilizer as the center, a fan-shaped warning zone for the horizontal stabilizer is drawn; the length of the hypotenuse of a single horizontal stabilizer can be obtained from the distance l between the wingtips of the aircraft's horizontal stabilizer and the width q of the wingtips of the horizontal stabilizer. Based on the vehicle motion model, the horizontal tail warning zone is defined according to the vehicle stopping distance, with a sector radius R. t =m+H;

[0052] Based on the vehicle motion model, a vehicle safety zone is defined at the front of the vehicle. Let the length of the vehicle be *c*, and the angle between the vehicle body and the x-axis be *Λ*. The coordinates of the center of the vehicle safety zone are then calculated as follows: Where (x) c ,y c The coordinates of the vehicle's center point are obtained through target detection; the maximum slip angle ξ of the vehicle in the vehicle motion model is selected. max The sector angle of the vehicle safety zone; the stopping distance of the vehicle is reserved when defining the vehicle safety zone, and the radius R of the vehicle safety zone sector. c =H.

[0053] The beneficial effects of this invention are: (1) HRNet is selected as the key point detection network, which ensures high-precision recognition while meeting the requirements of real-time detection. The camera coordinate transformation algorithm is applied to transform the target pixel coordinates into the actual coordinates of the apron scene in real time, thereby improving the accuracy of collision detection.

[0054] (2) Based on the geometric characteristics and conflict types of different locations such as the aircraft nose, wingtips, engine, and horizontal stabilizer wingtips, corresponding safety zones are delineated using aircraft braking models and support vehicle motion models. Combined with the real-time operating speeds of the aircraft and vehicles, the safety zones are further subdivided into dynamic early warning zones, alarm zones, and vehicle safety zones. Attached Figure Description

[0055] Figure 1 Flowchart of a method for determining aircraft and vehicle conflicts within a parking space based on multi-keypoint detection.

[0056] Figure 2 This is a schematic diagram illustrating the implementation process of the conflict determination method between aircraft and vehicles within the parking space in an embodiment of the present invention.

[0057] Figure 3 This is a schematic diagram of the coordinate transformation algorithm in an embodiment of the present invention, wherein (a) the y-coordinate of the aircraft at its actual position on the apron is calculated, and (b) the x-coordinate of the aircraft at its actual position on the apron is calculated.

[0058] Figure 4 This is a schematic diagram of a vehicle motion protection model in an embodiment of the present invention;

[0059] Figure 5 This is a schematic diagram illustrating the division of each safety zone in an embodiment of the present invention;

[0060] Figure 6 This is a schematic diagram illustrating a multi-keypoint detection application scenario in one embodiment of the present invention;

[0061] Figure 7 This is a screenshot of a scenario where the multi-keypoint detection part is used in one embodiment of the present invention. Detailed Implementation

[0062] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0063] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention; that is, the described embodiments are only a part of the embodiments of the invention, and not all of them. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0064] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention. It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0065] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0066] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0067] like Figure 1 As shown, the method for determining aircraft and vehicle conflicts within a parking position based on multi-keypoint detection includes the following steps:

[0068] Step 1: Obtain images of aircraft and support vehicles on the tarmac using the aircraft stand monitoring camera;

[0069] Step 2: Use target detection algorithms to identify aircraft and support vehicles within the parking position, and use coordinate transformation algorithms to obtain the position information of aircraft and support vehicles on the apron.

[0070] Step 3: Based on the aircraft deceleration model and the support vehicle motion model, the safety zones for the aircraft's nose, engine, wingtips, horizontal stabilizer wingtips, and support vehicles are delineated.

[0071] Step 4: Security Zone Conflict Determination and Alarm. Based on the security zone of the detected aircraft, determine whether the support vehicle has intruded into the security zone of the detected aircraft. If intrusion occurs, an alarm will be issued.

[0072] The method of obtaining the position information of each aircraft and support vehicle within the apron through coordinate transformation algorithm includes: converting the obtained pixel coordinates to and from the actual coordinates, and based on the target pixel coordinates (x, y) in the image data. a ,y a The actual coordinates (x, y) of the detected target within the apron are calculated using the following coordinate transformation formula:

[0073]

[0074] s c p1, p2, p3, p4, p5.

[0075] The aircraft deceleration model includes five stages:

[0076] During the reaction phase, the distance γ1 (m) traveled by the aircraft satisfies γ1 = v0t1, where v0 is the aircraft's taxiing speed when it is not braking, and t1 (s) is the reaction time required by the pilot.

[0077] During the initial braking phase, the pilot begins to depress the pedals, the aircraft decelerates a = 0, and travels a distance γ2 = v0t2, where t2(s) is the braking system transmission time;

[0078] During the deceleration growth phase, the aircraft begins to brake, and the deceleration 'a' increases from 0 to 'a'. max During this stage, the aircraft travels a certain distance. t3 is the time for the braking force to increase, a maxThis represents the maximum aircraft deceleration after the brake pedal is depressed.

[0079] During the braking hold phase, the aircraft's deceleration will remain at its maximum until the end of this phase.

[0080]

[0081] During the parking phase, the aircraft deceleration a = 0, and the aircraft displacement is 0.

[0082] The aircraft deceleration is at its maximum when it is on the main road surface and anti-skid braking is in use; at this time, a d =7.37m / s 2 ;

[0083] The deceleration is minimal on a wet track surface without anti-skid braking; at this point, a w =2.38m / s 2 ;

[0084] The aircraft braking distance is then γ1+γ2+γ3+γ4. Since t3 is extremely small, the formula can be further simplified to:

[0085]

[0086]

[0087] Where Γ min For the minimum braking distance, Γ max This is the maximum braking distance.

[0088] The aforementioned vehicle motion model includes:

[0089] Based on the vehicle motion model, a vehicle kinematic model is constructed, from which we can obtain:

[0090]

[0091]

[0092] The formula for calculating the sliding angle ξ is as follows:

[0093]

[0094] When a vehicle brakes, its total stopping distance can be divided into reaction distance and braking distance. Within the driver's reaction time, the vehicle travels at a V-shaped distance. cr When traveling at a constant speed, the reaction distance h1 satisfies:

[0095] h1 = t r ×V cr

[0096] Where t r This refers to the driver's reaction time;

[0097] Braking distance h2 satisfies:

[0098] h2=f(M,V cr )

[0099] Where M represents the distance between the vehicles, and V cr Let F represent the vehicle's center of gravity speed; during braking, the work done by the braking force F should be the same as the change in kinetic energy, i.e., satisfying:

[0100]

[0101] Where F = M × a c When the maximum braking force is applied to the vehicle, the deceleration remains constant, and the stopping distance H of the vehicle satisfies:

[0102] H = h1 + h2.

[0103] The aforementioned safety zones for the aircraft's nose, engines, wingtips, horizontal stabilizer wingtips, and support vehicle are defined based on the aircraft's deceleration model and support vehicle motion model, including:

[0104] The nose safety zone of an aircraft includes the nose warning zone and the nose dynamic warning zone. The nose warning zone is as follows:

[0105] The minimum clearance at the apron is λ, and the maximum ground turning angle of the aircraft is... A fan-shaped safety zone with radius ρ is drawn at the nose of the aircraft, and the maximum braking distance Γ required by the aircraft is... max The sum of the minimum clearances from the apron is the radius R of the nose warning zone. N ;

[0106] The radius of the dynamic warning zone for the nose is: R ′ N =R N +Γ min ;

[0107] The wingtip safety area of ​​an aircraft includes the wingtip safety area when the aircraft is stationary and the wingtip safety area when the aircraft is in motion. When the aircraft is stationary, the area is maintained at L. W ×L W The alarm range, of which L W It is twice the minimum clearance of the apron, i.e., L W =λ×2;

[0108] When the aircraft is in motion, the length of the longer side of the warning zone The size of the alarm zone is Length of the long side of the warning zone The size of the alarm zone is

[0109] The aircraft engine safety zone includes the engine intake safety zone and the engine exhaust safety zone. The size of the aircraft engine intake warning zone is as follows:

[0110]

[0111] For the engine dynamic warning zone, the length of the intake warning zone is extended to the minimum braking distance of the aircraft. min The rectangular area forms an area of ​​size The safe zone is designated as the engine dynamic early warning zone;

[0112] The safety zone for the vehicle is based on twice the aircraft's horizontal stabilizer sweep angle, supplemented by 2×(180°). ° -ω), with the intersection of the extended lines of the two hypotenuses of the aircraft's horizontal stabilizer as the center, a fan-shaped warning zone for the horizontal stabilizer is drawn; the length of the hypotenuse of a single horizontal stabilizer can be obtained from the distance l between the wingtips of the aircraft's horizontal stabilizer and the width q of the wingtips of the horizontal stabilizer. Based on the vehicle motion model, the horizontal tail warning zone is defined according to the vehicle stopping distance, with a sector radius R. t =m+H;

[0113] Based on the vehicle motion model, a vehicle safety zone is defined at the front of the vehicle. Let the length of the vehicle be *c*, and the angle between the vehicle body and the x-axis be *Λ*. The coordinates of the center of the vehicle safety zone are then calculated as follows: Where (x) c ,y c The coordinates of the vehicle's center point are obtained through target detection; the maximum slip angle ξ of the vehicle in the vehicle motion model is selected. max The sector angle of the vehicle safety zone; the stopping distance of the vehicle is reserved when defining the vehicle safety zone, and the radius R of the vehicle safety zone sector. c =H.

[0114] Specifically, this invention provides a method for determining conflicts between aircraft and vehicles within a parking space based on key point detection, which mainly includes the following steps:

[0115] Step S1: Image data acquisition; Based on the aircraft monitoring camera, acquire real-time images of aircraft and support vehicles on the tarmac.

[0116] Step S2: Target detection; Target detection algorithm is used to accurately identify aircraft and support vehicles within the parking position, and coordinate transformation algorithm is used to obtain high-precision position information of each target within the apron.

[0117] Step S3: Key point detection and safe zone generation; using a key point detection network to identify key points such as the nose, engine, wingtip, and horizontal tail wingtip of the aircraft under inspection, generating corresponding safe zones based on the aircraft's deceleration model, and building a vehicle kinematic model based on the support vehicle motion model to generate the support vehicle's safe zone.

[0118] Step S4: Safe Zone Conflict Determination and Alarm. Based on the aircraft, determine whether the vehicle has intruded into the safe zone and issue corresponding early warning or alarm information. A flowchart of the implementation process is shown below. Figure 2 As shown.

[0119] The coordinate transformation algorithm described in step S2 converts the acquired pixel coordinates to and from the actual coordinates. Collision detection in video images relies on the accurate determination of the actual distance between targets. Unlike the three-dimensional world we live in, cameras acquire two-dimensional image data; therefore, it is necessary to establish a conversion relationship between pixel coordinates and actual coordinates to obtain the actual position of the target. By designing a coordinate transformation algorithm to convert the acquired pixel coordinates to and from the actual coordinates, the accuracy of collision detection is improved. A schematic diagram of the coordinate transformation algorithm is shown below. Figure 3 To simplify subsequent calculations, the lengths of each side in the model are quantified. By convention: AB = s c , BC=p1, CD=p2, O1C1=p3, C1C2=p4, D1D2=p5. Based on the target pixel coordinates (x, y) in the image data. a ,y a The actual coordinates (x, y) of the detected target within the apron are obtained through calculation.

[0120] The conversion formula is as follows:

[0121]

[0122] Step S3 describes the safety zone designation based on the actual coordinates of the target within the parking position, delineating the aircraft's nose, engines, wingtips, horizontal stabilizer wingtips, and support vehicles. Traditional studies of aircraft safety zones often only consider the scenario where the aircraft is completely stationary, defining the safety zone solely based on the minimum clearance between the aircraft entering the parking position and adjacent aircraft, as well as nearby buildings and other objects. However, in actual apron operations, aircraft and support vehicles are often in a complex and dynamic process. Therefore, this invention incorporates the dynamic movement of the aircraft and support vehicles into the designation through aircraft deceleration models and support vehicle motion models, considering different conflict scenarios to define the safety zone accordingly.

[0123] The aforementioned aircraft deceleration model:

[0124] To address the braking scenario during aircraft taxiing into the parking position, an aircraft deceleration model is introduced as the basis for safety zone delineation. This model mainly consists of five stages:

[0125] During the reaction phase, the pilot has not yet processed the hazard information, and the aircraft maintains its original speed. The distance γ1 (m) traveled by the aircraft during this phase satisfies γ1 = v0t1, where v0 (m / s) is the aircraft's taxiing speed before braking, and t1 (s) is the pilot's required reaction time.

[0126] During the initial braking phase, the pilot begins to depress the pedals, but the aircraft has not yet begun braking; the aircraft deceleration a = 0, and the distance traveled is γ2 = v0t2. Here, t2(s) is the braking system's transmission time.

[0127] During the deceleration growth phase, the aircraft begins to brake, and the deceleration 'a' increases from 0 to 'a'. max During this stage, the aircraft travels a certain distance. t3(s) is the time for the braking force to increase, a max (m / s 2 () represents the maximum aircraft deceleration after the brake pedal is pressed;

[0128] During the braking hold phase, the aircraft's deceleration will remain at its maximum until the end of this phase.

[0129] During the parking phase, the aircraft deceleration a = 0, and the aircraft displacement is 0.

[0130] The braking deceleration of an aircraft is determined by pavement conditions and the type of braking. According to FAA surface braking tests, the aircraft deceleration is greatest on a dry pavement when anti-skid braking is employed; at this point, a d =7.37m / s 2 The deceleration is minimal on a wet track surface without anti-skid braking; at this time, a w =2.38m / s 2 .

[0131] Based on the above model, the aircraft braking distance γ1+γ2+γ3+γ4 is calculated. Since t3 is extremely small, the formula can be further simplified to:

[0132]

[0133]

[0134] Where Γ min (m) represents the minimum braking distance, Γ max (m) represents the maximum braking distance.

[0135] The aforementioned vehicle motion model:

[0136] To accurately reflect vehicle characteristics and ensure the model's ease of use, a vehicle motion model is used as the basis for defining the vehicle's safety zones. This model does not consider the vehicle's movement along the Z-axis, treating it as moving in a two-dimensional plane. The left and right front wheels are combined into a single point, denoted as point M, and the left and right rear wheels are combined into a single point N. The transfer of loads between the front and rear axles is ignored, and point P is the vehicle's center of mass. cr Let ε be the vehicle's center of gravity speed, and R be the turning radius. O is the intersection of ON and OM, which is the vehicle's instantaneous rolling center. Line segments OM and ON are perpendicular to the directions of the two rolling wheels, respectively. The angle between the vehicle's velocity direction and its body direction is the slip angle ξ, and the angle between the body and the x-axis is defined as the heading angle Λ. f and ε r These are the front and rear wheel deflection angles of the vehicle, e f and e r These refer to the front and rear overhang lengths of the vehicle, such as... Figure 4 As shown.

[0137] Based on the vehicle motion model, a vehicle kinematic model is constructed, from which we can obtain:

[0138]

[0139]

[0140] To determine the maximum steering angle of the vehicle, the slip angle is solved. From equations (4) and (5), the formula for calculating the slip angle β is:

[0141]

[0142] When a vehicle brakes, its total stopping distance can be divided into reaction distance and braking distance. This article assumes that within the driver's reaction time, the vehicle moves at a speed of V... cr Under the assumption of constant speed, the reaction distance h1 satisfies h1 = t r ×V cr , where t r The driver's reaction time is taken as 2 seconds.

[0143] Regarding braking distance, considering the vehicle's own weight during actual operation, the braking distance h2 should satisfy h2=f(M,V) cr ), where M represents the distance between the vehicles, and V cr This represents the vehicle's center of gravity speed. During braking, the work done by the braking force F should be equal to the change in kinetic energy, i.e., satisfying:

[0144]

[0145] Where F = M × ac When the maximum braking force is applied to the vehicle, the deceleration remains constant. Based on the standard braking distance of 40m per 100km / h, calculate the deceleration 'a' of the vehicle during emergency braking. c Approximately 10 m / s 2 .

[0146] Therefore, the stopping distance H of the vehicle should satisfy H = h1 + h2.

[0147] Step S3 involves defining the safety zones for the aircraft's nose, engines, wingtips, horizontal stabilizer wingtips, and support vehicles based on the aircraft's deceleration model and support vehicle motion model. A schematic diagram of the safety zone delineation for each key point is shown below. Figure 5 As shown.

[0148] According to the International Civil Aviation Organization (ICAO) regulations, the minimum clearance between aircraft and other obstacles on the apron is affected by airport flight area indicators. Based on airport flight area indicator II, let the minimum clearance on the apron be λ.

[0149] When an aircraft taxis on the tarmac, nose-to-nose collisions mostly occur directly in front or to the side. To address this, the aircraft's maximum ground turning angle should be considered. A sector-shaped safety zone with radius ρ is demarcated at the nose of the aircraft. This safety zone is further divided into a nose warning zone and a nose dynamic warning zone. The radius of the sector-shaped safety zone conforms to the aircraft's deceleration model and minimum clearance requirements. For the nose warning zone, in the event of a conflict, the sector radius should provide sufficient tolerance for the pilot while ensuring that the normal operation of other vehicles on the apron is not affected. Therefore, the maximum braking distance Γ required by the aircraft is... max The sum of the minimum clearance between the aircraft and the apron is used as the radius R of the nose warning zone. N For the dynamic warning zone at the nose, to ensure sufficient elastic response time, its radius R is specified. ′ N R is the sum of the warning zone length and the minimum braking distance. ′ N =R N +Γ min .

[0150] To address wingtip conflicts, and considering the geometric characteristics of the aircraft wing, a rectangular safety zone is designated at the wingtip. The sides of this safety zone are parallel or perpendicular to the aircraft's velocity direction. The design principles for the wingtip dynamic warning zone and wingtip alert zone are largely the same as those for the nose. When the aircraft is stationary, to ensure clearance requirements, an area of ​​L should be maintained. W ×L W The alarm range, of which L W It is twice the minimum clearance of the apron, i.e., L W =λ×2(m). The length of the longer side of the warning zone when the aircraft is in motion. The size of the alarm zone at this time is Length of the long side of the warning zone The size of the alarm zone at this time is

[0151] During aircraft engine operation, the high airflow velocity at the engine intake creates a significant pressure difference with the surrounding airflow, generating a "suction" force near the engine intake. As the aircraft taxis into its parking position, the engine remains running continuously. If support vehicles and personnel within the parking position accidentally enter the engine's hazardous area, they are highly susceptible to property damage or even injury. Furthermore, the large area of ​​heat radiation generated behind the engine during operation can easily cause harm to vehicles and personnel. Therefore, assuming the aircraft is only using idle thrust when parking, safety zones for both the engine intake and exhaust are defined.

[0152] For the air intake safety zone, engine manufacturers typically define it based solely on engine design parameters, without providing sufficient margin for early warning. Therefore, this invention combines the aircraft's motion state during taxiing, engine thrust, and the range of the engine air intake safety zone to divide the air intake safety zone into an air intake warning zone and an engine dynamic early warning zone. The air intake warning zone is an improvement upon the original engine safety zone. Specifically, it extends beyond the originally defined radius of r... e Based on the semi-circular region and extending a distance d behind the fairing, the maximum braking distance Γ of the aircraft is extended in the direction of aircraft movement. max Width is 2r e A rectangular area is designated to ensure pilots have sufficient time to avoid a collision. The size of the aircraft engine intake warning zone is:

[0153]

[0154] For the engine dynamic warning zone, under the premise of ensuring the orderly operation of ground activities, the length of the intake warning zone is extended to the minimum braking distance of the aircraft. min The rectangular area forms an area of ​​size The safe zone is designated as the engine dynamic warning zone, as shown in the diagram. Figure 5 As shown.

[0155] Based on the influence of airflow at the engine exhaust, an exhaust protection zone is delineated, divided into a prohibited exhaust zone and a hazardous exhaust zone. The prohibited exhaust zone is delineated starting from the exhaust ends of both engines, extending at a 15° angle to the engine centerline to both sides and connecting to each other, where z1 = max{Z t1 Z v1The exhaust hazard zone starts at the tail ends of both engines, extends outward and backward at a 30° angle along the engine block to z2, with the engine block as the axis. The end of the extended straight line is approximated by a straight line connection, where z2 = max{Z t2 Z v2}

[0156] Z t1 Z represents the distance at which the exhaust temperature is reduced to 300K under idle thrust. v1 Z represents the distance at which the exhaust gas velocity corresponds to 10 m / s under idle thrust; t2 Z represents the distance at which the exhaust temperature is reduced to 400K under idle thrust. v2 The distance corresponding to the exhaust gas velocity of 45 m / s under slow-speed thrust.

[0157] During the aircraft's taxiing into its parking position, the support vehicle may also be involved in a rear-end collision with the aircraft. This is based on a supplementary angle of twice the aircraft's horizontal stabilizer sweep angle: 2 × (180°). ° -ω), with the intersection of the extended lines of the two hypotenuses of the aircraft's horizontal stabilizer as the center, a fan-shaped warning zone for the horizontal stabilizer is delineated. The length of the hypotenuse on one side of the horizontal stabilizer can be obtained from the distance l between the wingtips of the aircraft's horizontal stabilizer and the width q of the wingtips. Based on the vehicle motion model, the horizontal tail warning zone is defined according to the vehicle stopping distance, with a sector radius R. t =m+H, such as Figure 5 As shown, the size of the horizontal stabilizer warning zone can be automatically adjusted according to the aircraft's speed and the speed of vehicles approaching from behind.

[0158] During aircraft taxiing and support operations, multiple vehicles need to work together. In actual airport operations, conflicts may arise due to coordination errors among these vehicles. To further ensure apron safety, based on the delineation of safety zones at key aircraft points, a vehicle safety zone is defined at the front of the support vehicle using a support vehicle motion model. Let the length of the support vehicle be *c*, and the angle between the vehicle body and the x-axis be *Λ*. The coordinates of the center of the vehicle safety zone are then obtained as follows: Where (x) c ,y c The coordinates of the vehicle's center point are obtained through target detection. The maximum slip angle ξ of the vehicle in the vehicle motion model is selected. max The sector angle of the vehicle safety zone is defined. To ensure that the detection range of the vehicle safety zone can flexibly change with vehicle speed, a stopping distance for the vehicle is reserved when defining the vehicle safety zone, and the sector radius R of the vehicle safety zone is specified. c =H.

[0159] In one embodiment of the present invention, eight possible scenarios for aircraft taxiing and parking are described using five safety zones: the aircraft nose, engine, wingtip, horizontal stabilizer wingtip, and ground support vehicle. Basic schematic diagrams for each scenario are shown below. Figure 6 As shown.

[0160] Figure 7 This is a screenshot of a scenario where the multi-keypoint detection part is used in one embodiment of the present invention.

[0161] The nose safety zone conflict determination has two application scenarios. Scenario 1 assumes the guide vehicle malfunctions and stops during the aircraft's parking maneuver. Scenario 2 assumes the support vehicle intrudes into the aircraft's nose safety zone from the side during the aircraft's taxiing process. The basic process is as follows: Figure 6 (a), Figure 6 As shown in (b), during the aircraft's taxiing into the parking position in Scenario 1, the lead vehicle malfunctions. As the aircraft gradually approaches a dangerous distance, the system issues a warning message, such as... Figure 7 As shown in (a). In use case two, when a vehicle intrudes into the aircraft's nose safety zone from the side, the system issues an alarm, as shown above. Figure 7 As shown in (b).

[0162] The basic process of determining the conflict in the wingtip safety zone is as follows: Figure 6 As shown in (c) and (d), it is assumed that the support vehicle intrudes into the wingtip dynamic warning zone during the aircraft's taxiing into the parking position, at which point the system issues a warning; when the vehicle continues to enter the wingtip warning zone, the system issues a timely warning.

[0163] The basic process of determining the conflict scenario in the engine safety zone is as follows: Figure 6 As shown in (e) and (f), the design verifies scenarios of a vehicle laterally intruding into the aircraft engine intake safety zone and a vehicle entering the engine exhaust hazard zone from the side and rear. The vehicle enters the engine dynamic warning zone and intake warning zone sequentially from the side, and the system issues warnings and alerts for the intrusion in sequence. When the engine is not shut down, the vehicle enters the exhaust hazard zone and exhaust prohibited zone sequentially from the rear, and the system issues warnings and alerts. Furthermore, when the engine stops running, the protection zone disappears, and the vehicle entering the engine safety zone no longer triggers an alarm.

[0164] The basic process of conflict determination in the Hirao safety zone is as follows: Figure 6 As shown in (g), the system ensures that the vehicle's approach from behind the aircraft does not result in a collision with the aircraft's horizontal stabilizer. When the horizontal stabilizer's warning zone comes into contact with the vehicle's safe zone, the system issues an alarm signal, such as... Figure 7 As shown in (c).

[0165] The basic process of vehicle safety zone conflict determination in various scenarios is as follows: Figure 6As shown in (h), during apron operations, some support vehicles may exceed their designated operating range and come dangerously close to aircraft. To address this conflict, the model classifies support vehicles with different operating ranges into Vehicle1 and Vehicle2. When a Vehicle2 vehicle (i.e., a support vehicle not docking with an aircraft) touches the fuselage within its own protection zone, the system issues an alarm, such as... Figure 7 As shown in (d).

[0166] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A method for determining aircraft and vehicle conflicts within a parking position based on multi-keypoint detection, characterized in that, Includes the following steps: Step 1: Obtain images of aircraft and support vehicles on the tarmac using the aircraft stand monitoring camera; Step 2: Use target detection algorithms to identify aircraft and support vehicles within the parking position, and use coordinate transformation algorithms to obtain the position information of aircraft and support vehicles on the apron. Step 3: Based on the aircraft deceleration model and the support vehicle motion model, the safety zones for the aircraft's nose, engine, wingtips, horizontal stabilizer wingtips, and support vehicles are delineated. Step 4: Security Zone Conflict Determination and Alarm. Based on the security zone of the detected aircraft, determine whether the support vehicle has intruded into the security zone of the detected aircraft. If it has intruded, an alarm will be issued. The aircraft deceleration model includes five stages: The reaction phase, the distance the aircraft travels during this phase. satisfy ,in The taxiing speed of the aircraft when it is not braking. This is the reaction time required by the pilot; During the initial braking phase, the pilot begins to depress the pedals, and the aircraft decelerates. Driving distance ,in Transmitting time to the braking system; During the deceleration growth phase, the aircraft begins to brake, and the deceleration... a From 0 to During this stage, the aircraft travels a certain distance. , For the time of increasing braking force, This represents the maximum aircraft deceleration after the brake pedal is depressed. During the braking hold phase, the aircraft's deceleration will remain at its maximum until the end of this phase. ； During the parking phase, the aircraft decelerates a =0, the aircraft displacement is 0; When the aircraft is on a dry road surface and anti-skid braking is applied, the maximum deceleration is: ; The minimum deceleration on a wet track surface without anti-skid braking is: ; The aircraft braking distance is ,because The formula can be further simplified to: in Minimum braking distance, This is the maximum braking distance.

2. The method for determining aircraft and vehicle conflicts within a parking position based on multi-keypoint detection according to claim 1, characterized in that, The method of obtaining the position information of each aircraft and support vehicle within the apron through coordinate transformation algorithm includes: The acquired pixel coordinates are transformed into and vice versa, based on the target pixel coordinates in the image data. The actual coordinates of the detected target within the apron are obtained through calculation. The coordinate transformation formula is as follows: in, and These are the horizontal and vertical coordinates of the target being detected in the actual coordinate system of the apron; and These are the horizontal and vertical pixel coordinates of the detected target in the image coordinate system, respectively. The installation height of the monitoring camera relative to the apron ground; , , , , The coordinate transformation parameters are obtained by calibration based on the installation location of the monitoring camera and preset reference points and reference distances in the apron scene; The imaging geometry correction angle is determined by the coordinate transformation parameters.

3. The method for determining aircraft and vehicle conflicts within a parking position based on multi-keypoint detection according to claim 1, characterized in that, The aforementioned vehicle motion model includes: Based on the vehicle motion model, a vehicle kinematic model is constructed, from which we can obtain: The sliding angle can be obtained The calculation formula is: in, and These refer to the front and rear wheel deflection angles of the vehicle. and These refer to the front and rear overhang lengths of the vehicle, respectively. When a vehicle brakes, its total stopping distance can be divided into reaction distance and braking distance. Within the driver's reaction time, the vehicle... Constant speed travel, reaction distance satisfy: in This refers to the driver's reaction time; Braking distance satisfy: in M Indicates the distance between vehicles. This indicates the vehicle's center of gravity speed; during braking, the braking force generated by the vehicle's braking... F The work done should be the same as the change in kinetic energy, that is, the following condition must be met: in , The deceleration during emergency braking of a vehicle is the distance the vehicle stops when the maximum braking force is applied and the deceleration remains constant. H satisfy: 。 4. The method for determining aircraft and vehicle conflicts within a parking position based on multi-keypoint detection according to claim 3, characterized in that, The aforementioned safety zones for the aircraft's nose, engines, wingtips, horizontal stabilizer wingtips, and support vehicle are defined based on the aircraft's deceleration model and support vehicle motion model, including: The nose safety zone of an aircraft includes the nose warning zone and the nose dynamic warning zone. The nose warning zone is as follows: The minimum clearance between the apron and the airport is Maximum ground turning angle of aircraft At the nose with radius Delineating a sector-shaped safety zone, the maximum braking distance required for the aircraft The sum of the minimum clearances to the apron is the radius of the nose warning zone. ; The radius of the aforementioned dynamic warning zone for the nose is: ; The wingtip safety area of ​​an aircraft includes the wingtip safety area when the aircraft is stationary and the wingtip safety area when the aircraft is in motion. When the aircraft is stationary, the area maintained is... The alarm range, of which It is twice the minimum clearance of the apron, that is ; When the aircraft is in motion, the length of the longer side of the warning zone The size of the alarm zone is Length of the longer side of the warning zone The size of the alarm zone is ; The aircraft engine safety zone includes the engine intake safety zone and the engine exhaust safety zone. The size of the aircraft engine intake warning zone is as follows: For the engine dynamic warning zone, the length of the intake warning zone is extended to the minimum braking distance of the aircraft. The rectangular area forms an area of ​​size The safe zone is designated as the engine dynamic early warning zone; The safety zone for the vehicle is based on twice the sweep angle of the aircraft's horizontal stabilizer. A fan-shaped warning zone for the horizontal stabilizer is drawn, centered on the intersection of the extended lines of the two diagonal sides of the aircraft's horizontal stabilizer; the warning zone is determined by the distance from the wingtip of the aircraft's horizontal stabilizer. Horizontal tail wingtip width The length of the hypotenuse on one side of the flat tail can be obtained. Based on the vehicle motion model, the horizontal tail warning zone is defined according to the vehicle's stopping distance, with a fan-shaped radius. ; Based on the vehicle motion model, a safe zone is defined at the front of the vehicle. The length of the vehicle body is assumed to be... The angle between the car body and the x-axis is The coordinates of the center of the vehicle's safe zone are obtained as follows: ,in The actual coordinates of the vehicle's center point obtained through target detection are used; the maximum slip angle of the vehicle in the vehicle motion model is selected. The sector angle of the vehicle safety zone; the radius of the vehicle safety zone sector, which allows for a stopping distance when designating the vehicle safety zone. .

Images