An airport flight area heterogeneous ground service vehicle cooperative control method and system

Abstract

A method for cooperative control of heterogeneous ground support vehicles in an airport flight area is disclosed. The method includes: acquiring real-time operational status information of multiple ground support vehicles of different types within the flight area, relative status information of neighboring aircraft and vehicles, task status information, environmental status information, and reference trajectory information; and finally enabling the corresponding ground support vehicles to perform cooperative obstacle avoidance and task passage. A cooperative control system for heterogeneous ground support vehicles in an airport flight area is also disclosed. This invention offers advantages in both safety and efficiency.

Description

Technical Field

[0001] This invention relates to the field of intelligent civil aviation and autonomous driving cooperative control technology, and in particular to a method and system for cooperative control of heterogeneous service vehicles in airport flight areas. Background Technology

[0002] With the increasing demand for intelligent and unmanned support in airport flight areas, autonomous or assisted driving ground support vehicles are gradually being applied to various business scenarios, including baggage transfer, passenger shuttle, aircraft refueling, aircraft towing, platform support, aircraft maintenance inspection, and guidance support. Flight areas differ from ordinary roads, characterized by high safety constraints, high aircraft priority, semi-enclosed road networks, dense work areas, traffic organization affected by air traffic control orders, and frequent weather disturbances. Various types of ground support vehicles must maintain a safe distance from aircraft, other ground support vehicles, and restricted areas while adhering to flight area operating rules and completing support tasks with clearly defined time requirements.

[0003] Existing ground support vehicle scheduling and control schemes mostly employ preset rules, static path planning, or homogeneous vehicle unified control strategies. While these schemes can achieve basic control in scenarios with a single vehicle type and minimal traffic disturbance, significant differences can arise in large, busy airport flight zones due to variations in vehicle length, width, wheelbase, minimum turning radius, maximum speed, braking capacity, task priority, and operating area. For example, refueling trucks and lifting platform trucks typically have large overall dimensions and low steering agility, while baggage tractors and guide vehicles may have different speed ranges and following requirements. Using homogeneous vehicle control models in such cases can easily lead to problems such as control commands becoming unexecutable, overly conservative avoidance maneuvers, or insufficient safety margins.

[0004] In recent years, methods such as multi-agent collaborative decision-making, reinforcement learning, and model predictive control have been applied in traffic cooperative control, but there are still shortcomings in directly transferring them to airport flight areas. On the one hand, the candidate behaviors output by the learning-based decision-making model may not meet the vehicle kinematic boundaries, safety protection intervals, or local operating rules of the flight area; on the other hand, limiting the collaborative decision-making model to a single network structure or training algorithm can easily narrow the scope of technical protection and make it difficult to adapt to different airport vehicle types, communication conditions, weather levels, and task organization methods.

[0005] Therefore, a collaborative control method and system for heterogeneous support vehicles in the airport flight area is needed. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for collaborative control of heterogeneous support vehicles in the airport flight area, which has the advantages of safety and efficiency.

[0007] To achieve the above objectives, this invention provides a method for cooperative control of heterogeneous ground support vehicles in an airport flight area. The method includes: Step S1, acquiring real-time operational status information of multiple ground support vehicles of different types within the flight area, relative status information of neighboring aircraft and vehicles, task status information, environmental status information, and reference trajectory information; Step S2, configuring vehicle kinematic constraints, safety protection interval constraints, and flight area operation rule constraints according to vehicle type; Step S3, constructing a multi-vehicle... State representation; Step S4, based on the state representations of multiple ground support vehicles, output the candidate cooperative behaviors corresponding to each vehicle; Step S5, map the candidate cooperative behaviors to control vectors or local trajectories, and perform feasibility verification and constraint correction on the control vectors or local trajectories according to the vehicle kinematic constraints, safety protection interval constraints, and flight area operation rule constraints configured according to vehicle type in Step S2; Step S6, based on the corrected control vectors, generate speed control commands, steering control commands, or trajectory adjustment commands, and send them to the vehicle execution interface of the corresponding ground support vehicle, so that the corresponding ground support vehicle performs cooperative avoidance and mission passage.

[0008] Preferably, in step S2, for vehicle i, its vehicle type parameter can be expressed as:

[0009] θ i =(l i ,w i ,b i ,v i max ,a i max ,d i max ,R i min );

[0010] Among them, l i Indicates vehicle length; w i Indicates vehicle width; b i Indicates wheelbase; v i max Indicates maximum speed; a i max Indicates maximum acceleration; d i max R represents the maximum braking deceleration. i min This indicates the minimum turning radius.

[0011] Preferably, in step S2, the safety protection interval is the relationship between vehicle i and target j, where target j can be an aircraft, an aircraft protection zone, other ground support vehicles, or the boundary of a restricted area.

[0012] The safety protection interval is expressed as:

[0013] D ij safe =D0+(v i 2 ) / (2d i max )+(v j 2 ) / (2d j max )+μ1Q+μ2B+μ3A ij ;

[0014] Where D0 represents the basic safety distance; Q represents the regional flow saturation; B represents the weather level; and A represents the regional flow saturation. ij The identifier indicates whether target j is an aircraft or an aircraft protected area; μ1, μ2, and μ3 are weighting coefficients; v i d represents the speed of vehicle i; i v represents the braking deceleration of vehicle i; j d represents the velocity of target j; j Indicates the braking deceleration of target j; when target j is an aircraft or an aircraft protection zone, A ij Take a larger value or 1 to increase the safety protection interval; when in low visibility, precipitation or local congestion scenarios, the weather level B or regional flow saturation Q increases, and the corresponding safety protection interval also increases accordingly.

[0015] Preferably, in step S3, the state representation construction module constructs a vehicle state representation based on the vehicle's own state, the states of neighboring objects, the task state, the environment state, and the reference trajectory deviation; for vehicle i, the following state vector can be used:

[0016] s i =[x i ,y i ,v i ,a i ,ψ i ,e i path ,τ i ,ρ i ,z i ,θ i ];

[0017] τ i =T i due -t now -T̂ i ;

[0018] Among them, (x i ,yi ) indicates the vehicle's position; v i Indicates speed; a i Indicates acceleration; ψ i Indicates the heading angle, e i path τ represents the deviation from the reference trajectory. i Indicates the remaining time margin of the task; ρ i Indicates neighborhood traffic density; z i Represents the environment state code; θ i Indicates the vehicle type parameter; τ i =T i due -t now -T̂ i Indicates the remaining time margin of the task; T i due Indicates the expected completion time; t now Indicates the current moment; T̂ i This indicates the estimated remaining travel time for the vehicle under the current reference trajectory; additionally, ρ i It can be determined based on the number of vehicles and aircraft within the vehicle's neighborhood, the congestion level of key intersection areas, or the road occupancy rate; z i It can indicate weather level, low visibility status, temporary control status, or restricted area status.

[0019] Preferably, in step S4, the state representations of multiple ground support vehicles are input into the collaborative decision-making model, and the candidate collaborative behaviors corresponding to each ground support vehicle are output. The evaluation function of the collaborative decision-making model may include safety, task efficiency, energy consumption, comfort, and rule compliance, and its reward or evaluation function may be expressed as:

[0020] R i =ω s R i safe +ω t R i task +ω e R i energy +ω c R i comfort +ω r R i rule ;

[0021] Among them, R i safe R i task R i energy R i comfortand R i rule These represent evaluation items for safety, task efficiency, energy consumption, comfort, and rule compliance, respectively; ω s ω t ω e ω c and ω r The weighting coefficient can be adaptively adjusted according to weather level, regional traffic saturation, and task level.

[0022] Preferably, in step S5, the control vector for vehicle i is:

[0023] u i =[v i cmd ,a i cmd ,δ i cmd kappa i cmd ];

[0024] Among them, v i cmd a represents the target speed of vehicle i; i cmd δ represents the target acceleration of vehicle i; i cmd This represents the steering control amount of vehicle i; kappa i cmd This represents the curvature correction amount for vehicle i.

[0025] Preferably, in step S5, the feasibility verification includes determining whether the control vector of vehicle i satisfies:

[0026] 0≤ v i cmd ≤ v i max , -d i max ≤ a i cmd ≤ a i max , R i cmd ≥ R i min ;

[0027] And determine whether the distance between vehicle i and the aircraft, vehicle, or restricted area meets the following requirements:

[0028] ||p i -p j ||≥ D ij safe ;

[0029] Where, p i and p j Let i and j represent the positions of vehicle i and target j, respectively.

[0030] Preferably, in step S5, if the vehicle kinematic constraints are not met, speed correction, acceleration correction, braking boundary trimming, turning radius correction, or curvature correction are performed; if the safety protection interval constraints are not met, deceleration avoidance, stopping and waiting, widening the interval, path replacement, or constraint projection are performed; if the flight area operation rules constraints are not met, candidate trajectories that violate restricted areas, aircraft protection zones, or temporary control rules are eliminated, and a control vector or local trajectory that meets the flight area operation rules is reselected; if the constraints are still not met, a stopping and waiting instruction or a command to maintain the current safe state is output; if the control vector mapped by the candidate cooperative behavior exceeds the speed, acceleration, braking, or turning boundary corresponding to the vehicle type, boundary trimming or speed correction is performed; if the local trajectory violates restricted areas, aircraft priority passage rules, or minimum safety protection interval, path replacement, constraint projection, or stopping and waiting correction are performed.

[0031] Preferably, in step S6, if the candidate cooperative behavior still does not meet the safety protection interval constraint or the flight area operation rule constraint after constraint correction, a stop-and-wait instruction or a maintain current safety state instruction is output; a speed control instruction, steering control instruction, or trajectory adjustment instruction is generated based on the corrected control vector and sent to the vehicle execution interface of the corresponding ground support vehicle, so that the corresponding ground support vehicle performs cooperative avoidance and mission passage; if the candidate cooperative behavior still cannot meet the safety protection interval constraint or the flight area operation rule constraint after constraint correction, a safety hold logic is triggered; the safety hold logic can output a stop-and-wait instruction, a maintain current safety state instruction, a speed reduction instruction, or a request for manual confirmation instruction, until a control instruction that meets the constraints is obtained again.

[0032] A heterogeneous ground support vehicle cooperative control system for an airport flight area is provided to implement the aforementioned heterogeneous ground support vehicle cooperative control method for an airport flight area. The system can execute a feedback update module to obtain control execution results and environmental feedback, update the vehicle state representation for the next control cycle, and repeat steps S1 to S6 within a preset control cycle to form a rolling cooperative control closed loop. It includes: a state acquisition module, a heterogeneous constraint configuration module, a state representation construction module, a cooperative decision-making module, an action mapping and constraint correction module, a control execution module, and a feedback update module. The modules communicate with each other. The state acquisition module acquires real-time operating status information of multiple different types of ground support vehicles within the flight area, relative status information of neighboring aircraft and vehicles, task status information, environmental status information, and reference trajectory information. The heterogeneous constraint configuration module configures vehicle kinematic constraints, safety protection interval constraints, and flight area operation rule constraints for each ground support vehicle according to vehicle type. The state representation construction module... Based on the real-time operating status information, relative status information, task status information, environmental status information, reference trajectory information, and constraints configured according to vehicle type, the module constructs a vehicle status representation for the corresponding ground support vehicle. The collaborative decision-making module inputs the vehicle status representation into the collaborative decision-making model and outputs candidate collaborative behaviors for each ground support vehicle. The action mapping and constraint correction module performs action mapping, feasibility verification, and constraint correction on the candidate collaborative behaviors based on the vehicle kinematic constraints, safety protection interval constraints, and flight zone operation rule constraints for each ground support vehicle, generating speed control commands, steering control commands, or trajectory adjustment commands for the corresponding ground support vehicle. The control execution module sends the speed control commands, steering control commands, or trajectory adjustment commands to the vehicle execution interface of the corresponding ground support vehicle. The feedback update module obtains the control execution results and environmental feedback, and updates the vehicle status representation for the next control cycle based on the control execution results and environmental feedback, forming a rolling collaborative control closed loop.

[0033] In summary, compared with the prior art, the cooperative control method and system for heterogeneous support vehicles in airport flight areas provided by the present invention has the following beneficial effects:

[0034] First, the present invention proposes a collaborative control method and system for heterogeneous tactical support vehicles in airport flight areas. By configuring kinematic constraints, safety protection intervals, and flight area operation rule constraints according to vehicle type, it can avoid the problem that isomorphic control models are difficult to adapt to heterogeneous tactical support vehicles. By integrating vehicle type parameters, relative states of neighboring aircraft and neighboring vehicles, remaining mission time, environmental state, and reference trajectory deviation in the state representation, it can improve the ability of collaborative decision-making to express the complex dynamic environment of the flight area.

[0035] Second, the present invention proposes a method and system for cooperative control of heterogeneous service vehicles in airport flight areas. It performs action mapping and constraint correction on candidate cooperative behaviors, so that the learning or rule-based decision results are converted into executable control commands that satisfy vehicle boundaries and flight area rules. It can reduce the computational load of multi-vehicle cooperative control and improve real-time performance through local shared communication, neighboring vehicle screening and key intersection area priority decision-making mechanisms.

[0036] Third, the collaborative control method and system for heterogeneous ground support vehicles in the airport flight area proposed in this invention allows for the use of multiple collaborative decision-making models and action mapping methods, avoiding limiting the scope of protection to a specific neural network structure or a single training algorithm; it can incorporate aircraft targets and ground support vehicle targets into a unified collaborative control framework, thereby improving the safety and support efficiency in scenarios where aircraft and heterogeneous ground support vehicles travel together. Attached Figure Description

[0037] Figure 1 This is a flowchart of a collaborative control method for heterogeneous support vehicles in an airport flight area proposed in this invention. Detailed Implementation

[0038] The following will be combined with the appendix in the embodiments of the present invention. Figure 1 The technical solutions, structural features, objectives and effects achieved in the embodiments of the present invention will be described in detail.

[0039] It should be noted that the accompanying drawings are in a very simplified form and use non-precise proportions. They are only used to facilitate and clarify the purpose of illustrating the embodiments of the present invention, and are not intended to limit the implementation conditions of the present invention. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportional relationship, or adjustments to the size should still fall within the scope of the technical content disclosed in the present invention, provided that they do not affect the effects and objectives that the present invention can produce.

[0040] It should be noted that, in this invention, relational terms such as "first" and "second" are used merely 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. 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 the expressly listed elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0041] like Figure 1 As shown, this invention proposes a cooperative control method for heterogeneous support vehicles in an airport flight area, the method comprising:

[0042] Step S1: Obtain real-time operational status information of multiple different types of ground support vehicles in the flight area, relative status information of nearby aircraft and vehicles, mission status information, environmental status information, and reference trajectory information.

[0043] The real-time operating status information includes vehicle position, speed, acceleration, heading angle, and vehicle type;

[0044] The task status information includes the task start point, task end point, task urgency level, and remaining task time.

[0045] The environmental status information includes weather level, regional traffic saturation, local control status, and restricted area status.

[0046] Step S2: Configure vehicle kinematic constraints, safety protection interval constraints, and flight zone operation rule constraints according to vehicle type;

[0047] Step S3: Construct state representations for multiple vehicles based on vehicle type parameters, real-time operating status, relative status information, task status information, environmental status information, and reference trajectory information;

[0048] Step S4: Based on the state representations of multiple ground support vehicles, output the candidate cooperative behaviors corresponding to each vehicle.

[0049] Step S5: Map the candidate cooperative behavior to a control vector or local trajectory, and perform feasibility verification and constraint correction on the control vector or local trajectory based on the vehicle kinematic constraints, safety protection interval constraints and flight zone operation rule constraints configured according to vehicle type in step S2.

[0050] Step S6: Based on the corrected control vector, generate speed control commands, steering control commands, or trajectory adjustment commands, and send them to the vehicle execution interface of the corresponding ground support vehicle, so that the corresponding ground support vehicle can perform cooperative obstacle avoidance and mission passage.

[0051] Specifically, in step S2, for vehicle i, its vehicle type parameter can be expressed as:

[0052] θ i =(l i ,w i ,b i ,v i max ,a i max ,d i max ,R i min );

[0053] Among them, l i Indicates vehicle length; wi Indicates vehicle width; b i Indicates wheelbase; v i max Indicates maximum speed; a i max Indicates maximum acceleration; d i max R represents the maximum braking deceleration. i min Indicates the minimum turning radius;

[0054] For example, a refueling truck can be configured with a larger vehicle length, a lower maximum speed, and a larger minimum turning radius; a baggage tractor can be configured with smaller overall dimensions and lower braking deceleration under traction conditions; and a guide vehicle can be configured with a higher speed limit and stricter heading stability requirements. These type parameters are used for subsequent state coding, candidate action boundary determination, and constraint correction.

[0055] In addition, in step S2, the safety protection interval is the relationship between vehicle i and target j, where target j can be an aircraft, an aircraft protection zone, other ground support vehicles, or the boundary of a restricted area.

[0056] The safety protection interval is expressed as:

[0057] D ij safe =D0+(v i 2 ) / (2d i max )+(v j 2 ) / (2d j max )+μ1Q+μ2B+μ3A ij ;

[0058] Where D0 represents the basic safety distance; Q represents the regional flow saturation; B represents the weather level; and A represents the regional flow saturation. ij The identifier indicates whether target j is an aircraft or an aircraft protected area; μ1, μ2, and μ3 are weighting coefficients; v i d represents the speed of vehicle i; i v represents the braking deceleration of vehicle i; j d represents the velocity of target j; j This represents the braking deceleration of target j;

[0059] When target j is an aircraft or an aircraft protection zone, A ij Take a larger value or 1 to increase the safety protection interval; when in low visibility, precipitation or local congestion scenarios, the weather level B or regional flow saturation Q increases, and the corresponding safety protection interval also increases accordingly.

[0060] In addition, the vehicle kinematic constraints include upper speed limits, upper acceleration limits, upper braking deceleration limits, minimum turning radius, vehicle dimensions, and steering stability requirements for the corresponding vehicle type; the airfield operation rule constraints include permitted passage areas, restricted areas, aircraft protection zone avoidance rules, aircraft priority passage rules, speed limit rules, temporary control rules, and priority passage rules for critical intersection areas. Different types of ground support vehicles can be configured with the above kinematic constraints and operation rule constraints according to their vehicle size, mission type, load status, and operating area.

[0061] Specifically, in step S3, the state representation construction module constructs a vehicle state representation based on the vehicle's own state, the states of neighboring objects, the task state, the environment state, and the reference trajectory deviation. For vehicle i, the following state vector can be used:

[0062] s i =[x i ,y i ,v i ,a i ,ψ i ,e i path ,τ i ,ρ i ,z i ,θ i ];

[0063] τ i =T i due -t now -T̂ i ;

[0064] Among them, (x i ,y i ) indicates the vehicle's position; v i Indicates speed; a i Indicates acceleration; ψ i Indicates the heading angle, e i path τ represents the deviation from the reference trajectory. i Indicates the remaining time margin of the task; ρ i Indicates neighborhood traffic density; z i Represents the environment state code; θ i Indicates the vehicle type parameter; τ i =T i due -t now -T̂ i Indicates the remaining time margin of the task; T i due Indicates the expected completion time; t nowIndicates the current moment; T̂ i This indicates the estimated remaining travel time for the vehicle under the current reference trajectory.

[0065] In addition, ρ i It can be determined based on the number of vehicles and aircraft within the vehicle's neighborhood, the congestion level of key intersection areas, or the road occupancy rate; z i It can indicate weather level, low visibility status, temporary control status, or restricted area status.

[0066] For different types of vehicles, the state representation can employ either a dedicated state encoding method or a unified state vector superimposed with vehicle type encoding. For scenarios with many heterogeneous vehicle categories, a dedicated state encoding method can be configured for each vehicle category; for scenarios requiring a unified model input dimension, a unified state vector superimposed with vehicle type embedding encoding can be used.

[0067] Specifically, in step S4, the state representations of multiple ground support vehicles are input into the collaborative decision-making model, which outputs candidate collaborative behaviors corresponding to each ground support vehicle. The collaborative decision-making model generates candidate collaborative behaviors based on vehicle states, neighboring vehicle states, relative aircraft states, and task constraints, and is not limited to a specific neural network structure, a specific training algorithm, or a unique multi-agent reinforcement learning implementation. The candidate collaborative behaviors include one or more of the following: accelerating through, decelerating to avoid, stopping and waiting, maintaining following distance, path switching, partial detour, yielding, and priority passage requests in key intersection areas. As an optional implementation, a centralized training and distributed execution multi-agent collaborative decision-making model can be used; during the training phase, the joint states, joint actions, and environmental feedback of multiple ground support vehicles are aggregated; during the execution phase, candidate collaborative behaviors are output based on individual vehicle states, neighboring object states, and locally shared information.

[0068] In a specific embodiment, the evaluation function of the collaborative decision-making model may include safety, task efficiency, energy consumption, comfort, and rule compliance, and its reward or evaluation function may be expressed as:

[0069] R i =ω s R i safe +ω t R i task +ω e R i energy +ω c R i comfort +ω r R i rule ;

[0070] Among them, Ri safe R i task R i energy R i comfort and R i rule These represent evaluation items for safety, task efficiency, energy consumption, comfort, and rule compliance, respectively; ω s ω t ω e ω c and ω r The weighting coefficient can be adaptively adjusted according to weather level, regional traffic saturation, and task level.

[0071] Safety evaluation items can be determined based on the minimum distance between vehicles and aircraft, and between vehicles; mission efficiency evaluation items can be determined based on remaining mission time and on-time arrival; energy consumption evaluation items can be determined based on speed and acceleration changes; comfort evaluation items can be determined based on rapid acceleration, deceleration, and steering changes; and rule compliance evaluation items can be determined based on violations of restricted areas, aircraft priority passage, and speed limits. Each weighting coefficient can be adaptively adjusted based on weather level, regional traffic saturation, and mission level.

[0072] Specifically, in step S5, the control vector for vehicle i is:

[0073] u i =[v i cmd ,a i cmd ,δ i cmd kappa i cmd ];

[0074] Among them, v i cmd a represents the target speed of vehicle i; i cmd δ represents the target acceleration of vehicle i; i cmd This represents the steering control amount of vehicle i; kappa i cmd This represents the curvature correction amount for vehicle i.

[0075] In step S5, the feasibility verification includes determining whether the control vector of vehicle i satisfies the following:

[0076] 0≤ v i cmd ≤ v imax , -d i max ≤ a i cmd ≤ a i max , R i cmd ≥ R i min ;

[0077] And determine whether the distance between vehicle i and the aircraft, vehicle, or restricted area meets the following requirements:

[0078] ||p i -p j ||≥ D ij safe ;

[0079] Where, p i and p j These represent the positions of vehicle i and target j, respectively.

[0080] The above-mentioned inequalities of speed, acceleration, braking, and turning radius correspond to vehicle kinematic constraints; the distance inequalities correspond to safety protection interval constraints; candidate paths must not enter restricted areas, aircraft protection zones, or temporary control areas, and must comply with aircraft priority passage and speed limit rules, corresponding to flight zone operation rule constraints.

[0081] Furthermore, step S2 is responsible for establishing and configuring various constraints, including vehicle kinematic constraints, safety protection interval constraints, and flight area operation rule constraints; step S5 is responsible for calling the constraints configured in step S2 to perform feasibility verification and constraint correction after the candidate cooperative behavior is mapped to a control vector or local trajectory.

[0082] Constraint correction can be achieved through at least one of the following methods: boundary trimming, velocity correction, path replacement, model predictive control, sampling optimization, or constraint projection. To improve the executability of collaborative decision-making results, this embodiment performs feasibility verification and constraint correction on the candidate collaborative behaviors output by the collaborative decision-making model.

[0083] Specifically, if the vehicle kinematic constraints are not met, speed correction, acceleration correction, braking boundary trimming, turning radius correction, or curvature correction are performed; if the safety protection interval constraints are not met, deceleration avoidance, stopping and waiting, widening the interval, path replacement, or constraint projection are performed; if the flight area operation rules constraints are not met, candidate trajectories that violate restricted areas, aircraft protection zones, or temporary control rules are eliminated, and a control vector or local trajectory that meets the flight area operation rules is reselected; if the constraints are still not met, a stopping and waiting instruction or a instruction to maintain the current safe state is output.

[0084] If the control vector mapped by the candidate cooperative behavior exceeds the speed, acceleration, braking, or turning boundaries corresponding to the vehicle type, boundary clipping or speed correction is performed. If the local trajectory violates restricted areas, aircraft priority rules, or minimum safety protection intervals, path replacement, constraint projection, or stop-and-wait correction is performed. For example, when the candidate cooperative behavior is "deceleration avoidance," the system calculates the required deceleration based on the current speed, the relative distance to the conflicting target ahead, the permissible braking deceleration, and the safety protection interval. Similarly, when the candidate cooperative behavior is "path switching," the system determines whether the local trajectory is executable based on reachable road segments, minimum turning radius, vehicle dimensions, and flight zone operating rules; if the candidate path crosses a restricted area or aircraft protection zone, the path is discarded, and a new local trajectory that meets the constraints is selected.

[0085] Specifically, in step S6, if the candidate cooperative behavior, after constraint correction, still does not meet the safety protection interval constraint or the flight area operation rule constraint, a stop-and-wait instruction or a maintain current safety state instruction is output. Speed ​​control instructions, steering control instructions, or trajectory adjustment instructions are generated based on the corrected control vector and sent to the vehicle execution interface of the corresponding ground support vehicle, enabling the corresponding ground support vehicle to perform cooperative obstacle avoidance and mission passage. If the candidate cooperative behavior, after constraint correction, still cannot meet the safety protection interval constraint or the flight area operation rule constraint, the safety-maintaining logic is triggered. The safety-maintaining logic can output a stop-and-wait instruction, a maintain current safety state instruction, a speed reduction instruction, or a request for manual confirmation instruction, until a control instruction that meets the constraints is obtained again.

[0086] In addition, the present invention also proposes a collaborative control system for heterogeneous support vehicles in airport flight areas, which is used to implement the aforementioned collaborative control method for heterogeneous support vehicles in airport flight areas; the system can execute a feedback update module to obtain control execution results and environmental feedback, update the vehicle status representation for the next control cycle, and repeat steps S1 to S6 within a preset control cycle to form a rolling collaborative control closed loop.

[0087] In a specific embodiment, the heterogeneous support vehicle collaborative control system for airport flight area includes: a state acquisition module, a heterogeneous constraint configuration module, a state representation construction module, a collaborative decision-making module, an action mapping and constraint correction module, a control execution module, and a feedback update module; the modules communicate with each other.

[0088] The status acquisition module acquires real-time operational status information of multiple different types of ground support vehicles in the flight area, relative status information of nearby aircraft and vehicles, mission status information, environmental status information, and reference trajectory information.

[0089] The heterogeneous constraint configuration module configures vehicle kinematic constraints, safety protection interval constraints, and flight zone operation rule constraints for each local support vehicle according to vehicle type.

[0090] The state representation construction module constructs the vehicle state representation of the corresponding ground support vehicle based on the real-time operating state information, relative state information, task state information, environmental state information, reference trajectory information, and constraints configured according to vehicle type.

[0091] The collaborative decision-making module inputs the vehicle state representation into the collaborative decision-making model and outputs the candidate collaborative behaviors corresponding to each local support vehicle.

[0092] The motion mapping and constraint correction module performs motion mapping, feasibility verification and constraint correction on the candidate cooperative behaviors based on the vehicle kinematic constraints, safety protection interval constraints and flight zone operation rule constraints corresponding to each ground support vehicle, and generates the corresponding speed control command, steering control command or trajectory adjustment command for the ground support vehicle.

[0093] The control execution module sends the speed control command, steering control command, or trajectory adjustment command to the vehicle execution interface of the corresponding ground support vehicle;

[0094] The feedback update module obtains the control execution results and environmental feedback, and updates the vehicle state representation for the next control cycle based on the control execution results and environmental feedback, forming a rolling collaborative control closed loop.

[0095] The control cycle can be from 50 milliseconds to 500 milliseconds, or it can be adjusted adaptively based on vehicle type, communication latency, or flight zone operation level. To reduce the computational load of multi-vehicle cooperative control and improve real-time performance, the system can employ a local shared communication mechanism and a neighborhood object filtering mechanism. For vehicle i, objects with a distance less than a preset neighborhood radius, expected arrival at the same critical intersection area, existing trajectory, or priority influence relationship with aircraft or vehicles, are identified as neighborhood objects. Within each control cycle, the system completes state acquisition, heterogeneous constraint configuration, state representation construction, collaborative decision-making, action mapping, constraint correction, and control execution. It then acquires the vehicle's actual speed, actual heading, current position deviation, changes in the states of neighboring aircraft and vehicles, communication status changes, and local environmental changes, and updates the input for the next control cycle. This forms a rolling cooperative control closed loop, enabling heterogeneous servicing vehicles to achieve online adaptive cooperative avoidance and mission passage in complex dynamic flight zone scenarios. The control cycle can be from 50 milliseconds to 500 milliseconds, or it can be adaptively adjusted based on vehicle type, communication latency, or flight zone operation level. To reduce the computational load of multi-vehicle cooperative control and improve real-time performance, the system can employ a local shared communication mechanism and a neighborhood object filtering mechanism. For vehicle i, objects whose distance is less than a preset neighborhood radius, are expected to arrive at the same critical intersection area, have trajectory intersection relationships, or have priority influence relationships with aircraft are identified as neighborhood objects. Neighborhood objects can include ground support vehicles, taxiing aircraft, pushback aircraft, aircraft protection zones, and temporary control areas. The system inputs the relative position, relative speed, estimated arrival time at the intersection area, and priority information of neighborhood objects into the collaborative decision-making model. For non-neighborhood objects, simplified representations can be achieved using aggregated traffic density, regional flow saturation, or key area occupancy status, thereby reducing the state dimension and improving computational efficiency within the control cycle.

[0096] Although the present invention has been described in detail through the preferred embodiments above, it should be understood that the above description should not be considered as a limitation of the present invention. Various modifications and substitutions to the present invention will be apparent to those skilled in the art after reading the above description. Therefore, the scope of protection of the present invention should be defined by the appended claims.

Claims

1. A method for cooperative control of heterogeneous support vehicles in an airport flight area, characterized in that, The method includes: Step S1: Obtain real-time operational status information of multiple different types of ground support vehicles in the flight area, relative status information of nearby aircraft and vehicles, mission status information, environmental status information, and reference trajectory information. Step S2: Configure vehicle kinematic constraints, safety protection interval constraints, and flight zone operation rule constraints according to vehicle type; Step S3: Construct state representations for multiple vehicles based on vehicle type parameters, real-time operating status, relative status information, task status information, environmental status information, and reference trajectory information; Step S4: Based on the state representations of multiple ground support vehicles, output the candidate cooperative behaviors corresponding to each vehicle. Step S5: Map the candidate cooperative behavior to a control vector or local trajectory, and perform feasibility verification and constraint correction on the control vector or local trajectory based on the vehicle kinematic constraints, safety protection interval constraints and flight zone operation rule constraints configured according to vehicle type in step S2. Step S6: Based on the corrected control vector, generate speed control commands, steering control commands, or trajectory adjustment commands, and send them to the vehicle execution interface of the corresponding ground support vehicle, so that the corresponding ground support vehicle can perform cooperative obstacle avoidance and mission passage.

2. The method for coordinated control of heterogeneous support vehicles in an airport flight area according to claim 1, characterized in that, In step S2, for vehicle i, its vehicle type parameter can be expressed as: θ i =(l i ,w i ,b i ,v i max ,a i max ,d i max ,R i min ); Among them, l i Indicates vehicle length; w i Indicates vehicle width; b i Indicates wheelbase; v i max Indicates maximum speed; a i max Indicates maximum acceleration; d i max R represents the maximum braking deceleration. i min This indicates the minimum turning radius.

3. The method for coordinated control of heterogeneous support vehicles in an airport flight area according to claim 2, characterized in that, In step S2, the safety protection interval is the relationship between vehicle i and target j. Target j can be an aircraft, an aircraft protection zone, other ground support vehicles, or the boundary of a restricted area. The safety protection interval is expressed as: D ij safe =D0+(v i 2 ) / (2d i max )+(v j 2 ) / (2d j max )+μ1Q+μ2B+μ3A ij ; Where D0 represents the basic safety distance; Q represents the regional flow saturation; B represents the weather level; and A represents the regional flow saturation. ij The identifier indicates whether target j is an aircraft or an aircraft protected area; μ1, μ2, and μ3 are weighting coefficients; v i d represents the speed of vehicle i; i v represents the braking deceleration of vehicle i; j d represents the velocity of target j; j This represents the braking deceleration of target j; When target j is an aircraft or an aircraft protection zone, A ij Take a larger value or 1 to increase the safety protection interval; when in low visibility, precipitation or local congestion scenarios, the weather level B or regional flow saturation Q increases, and the corresponding safety protection interval also increases accordingly.

4. The method for coordinated control of heterogeneous support vehicles in an airport flight area according to claim 3, characterized in that, In step S3, the state representation construction module constructs a vehicle state representation based on the vehicle's own state, the states of neighboring objects, the task state, the environment state, and the reference trajectory deviation; for vehicle i, the following state vector can be used: s i =[x i ,y i ,v i ,a i ,ψ i ,e i path ,t i ,r i ,z i ,i i ]; τ i =T i due -t now -T̂ i ; Among them, (x i ,y i ) indicates the vehicle's position; v i Indicates speed; a i Indicates acceleration; ψ i Indicates the heading angle, e i path τ represents the deviation from the reference trajectory. i Indicates the remaining time margin of the task; ρ i Indicates neighborhood traffic density; z i Represents the environment state code; θ i Indicates the vehicle type parameter; τ i =T i due -t now -T̂ i Indicates the remaining time margin of the task; T i due Indicates the expected completion time; t now Indicates the current moment; T̂ i This indicates the estimated remaining travel time for the vehicle under the current reference trajectory; In addition, ρ i It can be determined based on the number of vehicles and aircraft within the vehicle's neighborhood, the congestion level of key intersection areas, or the road occupancy rate; z i It can indicate weather level, low visibility status, temporary control status, or restricted area status.

5. The method for coordinated control of heterogeneous support vehicles in an airport flight area according to claim 4, characterized in that, In step S4, the state representations of multiple ground support vehicles are input into the collaborative decision-making model, and the candidate collaborative behaviors corresponding to each ground support vehicle are output. The evaluation function of a collaborative decision-making model can include safety, task efficiency, energy consumption, comfort, and rule compliance. Its reward or evaluation function can be expressed as: R i =ω s R i safe +oh t R i task +oh e R i energy +oh c R i comfort +oh r R i rule ; Among them, R i safe R i task R i energy R i comfort and R i rule These represent evaluation items for safety, task efficiency, energy consumption, comfort, and rule compliance, respectively; ω s ω t ω e ω c and ω r The weighting coefficient can be adaptively adjusted according to weather level, regional traffic saturation, and task level.

6. The method for coordinated control of heterogeneous support vehicles in an airport flight area according to claim 5, characterized in that, In step S5, the control vector for vehicle i is: you i =[v i cmd ,a i cmd ,δ i cmd ,kappa i cmd ]; Among them, v i cmd a represents the target speed of vehicle i; i cmd δ represents the target acceleration of vehicle i; i cmd This represents the steering control amount of vehicle i; kappa i cmd This represents the curvature correction amount for vehicle i.

7. The method for cooperative control of heterogeneous support vehicles in an airport flight area according to claim 6, characterized in that, In step S5, the feasibility verification includes determining whether the control vector of vehicle i satisfies the following: 0≤ v i cmd ≤ v i max , -d i max ≤ a i cmd ≤ a i max , R i cmd ≥ R i min ; And determine whether the distance between vehicle i and the aircraft, vehicle, or restricted area meets the following requirements: ‖p i -p j ‖≥ D ij safe ; Where, p i and p j Let i and j represent the positions of vehicle i and target j, respectively.

8. The method for cooperative control of heterogeneous support vehicles in an airport flight area according to claim 7, characterized in that, In step S5, if the vehicle kinematic constraints are not met, speed correction, acceleration correction, braking boundary trimming, turning radius correction, or curvature correction are performed; if the safety protection interval constraints are not met, deceleration avoidance, stopping and waiting, widening the interval, path replacement, or constraint projection are performed; if the flight area operation rules constraints are not met, candidate trajectories that violate restricted areas, aircraft protection zones, or temporary control rules are eliminated, and a control vector or local trajectory that meets the flight area operation rules is reselected; if the constraints are still not met, a stopping and waiting instruction or a command to maintain the current safe state is output; if the control vector mapped by the candidate cooperative behavior exceeds the speed, acceleration, braking, or turning boundary corresponding to the vehicle type, boundary trimming or speed correction is performed; if the local trajectory violates restricted areas, aircraft priority passage rules, or minimum safety protection interval, path replacement, constraint projection, or stopping and waiting correction are performed.

9. A method for coordinated control of heterogeneous support vehicles in an airport flight area according to claim 8, characterized in that, In step S6, if the candidate cooperative behavior still does not meet the safety protection interval constraint or the flight area operation rule constraint after constraint correction, a stop waiting instruction or a maintain current safety state instruction is output; a speed control instruction, steering control instruction, or trajectory adjustment instruction is generated based on the corrected control vector and sent to the vehicle execution interface of the corresponding ground support vehicle, so that the corresponding ground support vehicle performs cooperative avoidance and mission passage; if the candidate cooperative behavior still cannot meet the safety protection interval constraint or the flight area operation rule constraint after constraint correction, a safety hold logic is triggered; the safety hold logic can output a stop waiting instruction, a maintain current safety state instruction, a speed reduction instruction, or a request for manual confirmation instruction, until a control instruction that meets the constraints is obtained again.

10. A collaborative control system for heterogeneous support vehicles in an airport flight area, characterized in that, This system is used to implement a collaborative control method for heterogeneous support vehicles in an airport flight area as described in any one of claims 1 to 9. The system can execute a feedback update module to obtain control execution results and environmental feedback, update the vehicle state representation for the next control cycle, and repeat steps S1 to S6 within a preset control cycle to form a rolling collaborative control closed loop. It includes: a state acquisition module, a heterogeneous constraint configuration module, a state representation construction module, a collaborative decision-making module, an action mapping and constraint correction module, a control execution module, and a feedback update module; the modules communicate with each other. The status acquisition module acquires real-time operational status information of multiple different types of ground support vehicles in the flight area, relative status information of nearby aircraft and vehicles, mission status information, environmental status information, and reference trajectory information. The heterogeneous constraint configuration module configures vehicle kinematic constraints, safety protection interval constraints, and flight zone operation rule constraints for each local support vehicle according to vehicle type. The state representation construction module constructs the vehicle state representation of the corresponding ground support vehicle based on the real-time operating state information, relative state information, task state information, environmental state information, reference trajectory information, and constraints configured according to vehicle type. The collaborative decision-making module inputs the vehicle state representation into the collaborative decision-making model and outputs the candidate collaborative behaviors corresponding to each local support vehicle. The motion mapping and constraint correction module performs motion mapping, feasibility verification and constraint correction on the candidate cooperative behaviors based on the vehicle kinematic constraints, safety protection interval constraints and flight zone operation rule constraints corresponding to each ground support vehicle, and generates the corresponding speed control command, steering control command or trajectory adjustment command for the ground support vehicle. The control execution module sends the speed control command, steering control command, or trajectory adjustment command to the vehicle execution interface of the corresponding ground support vehicle; The feedback update module obtains the control execution results and environmental feedback, and updates the vehicle state representation for the next control cycle based on the control execution results and environmental feedback, forming a rolling collaborative control closed loop.

Images