A method for coordinated scheduling and traffic flow control of low-altitude UAVs and ground-based roads and bridges.

Abstract

This invention discloses a method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles (UAVs) and ground-based roads and bridges, relating to the fields of intelligent transportation, low-altitude UAV traffic management, bridge structural health monitoring, vehicle-road cooperation, and active traffic flow control. The method acquires UAV operational data, road traffic flow data, bridge structural status data, traffic control equipment status data, and environmental data to construct a road-bridge-air spatiotemporal resource map; calculates the air-ground-bridge coupling risk value based on the overlap relationship between the risk shadow domain and ground road segment units and bridge structural units; generates an air-ground interlocking passage token; and synchronously generates UAV scheduling instructions and ground traffic flow control instructions under the safety constraints of a barrier function. This invention can transform the risks of low-altitude UAV operation into controllable ground traffic risks, achieving closed-loop coordinated management and control of low-altitude air routes, ground traffic flow, and bridge structural safety.

Description

Technical Field

[0001] This invention relates to the fields of intelligent transportation, low-altitude unmanned aerial vehicle (UAV) traffic management, bridge structural health monitoring, vehicle-road cooperation, and active traffic flow control, and particularly to a method for coordinated scheduling and traffic flow management of low-altitude UAVs and ground-based roads and bridges. Background Technology

[0002] Low-altitude drones have been used for road inspection, bridge inspection, accident investigation, traffic guidance, emergency material delivery, and rescue guidance. Meanwhile, traffic conditions on urban expressways, bridges spanning rivers and seas, mountain bridge and tunnel complexes, and transportation hubs are complex. Bridge congestion, sudden vehicle stops, concentrated traffic of heavy vehicles, crosswinds, bridge vibrations, and structural anomalies can all affect traffic safety.

[0003] Existing low-altitude drone management technologies typically focus on airspace conflicts between drones, route planning, and flight permits; existing road traffic flow control technologies typically focus on signal timing, ramp control, variable speed limits, lane control, and path guidance; and existing bridge structural health monitoring technologies typically focus on monitoring bridge strain, vibration, cable tension, deflection, and wind field.

[0004] While the aforementioned technologies can achieve airspace management, road traffic control, or bridge safety monitoring respectively, the following problems still exist in road and bridge areas: First, the risks associated with low-altitude drone operations are not being translated into ground traffic control objectives. When drones operate above bridges or roads, their failure landing zones, wind drift zones, and load consequence zones create dynamic risk projections on the ground. If these risk projections overlap with high-density traffic flow, bridge queues, or areas with concentrated heavy vehicles, they can create compound risks.

[0005] Second, ground traffic control does not serve the safety mission of drones. Emergency drones need to quickly reach accident or abnormal bridge areas, but if the traffic density on the bridge is too high or the risk projection area has too many exposed vehicles, relying solely on drones to detour will reduce the efficiency of emergency response; if ground traffic diversion is relied upon alone, there is a lack of rapid confirmation capability from the drone's perspective.

[0006] Third, there is a lack of a unified closed-loop mechanism between bridge structural condition, bridge deck traffic flow, and UAV flight path capacity. When the safety margin of the bridge structure decreases, crosswinds on the bridge deck increase, or vibrations are abnormal, the UAV flight path voxel capacity and ground inlet traffic should be affected simultaneously; otherwise, air-to-ground risks may overlap.

[0007] Therefore, a method is needed that can incorporate the low-altitude failure risk of UAVs, the exposure status of ground traffic, and the safety status of bridge structures into the same closed-loop control framework. Summary of the Invention

[0008] Technical problems to be solved The technical problem to be solved by this invention is: how to transform the dynamic flight risk of low-altitude UAVs into a calculable, constrainable, and executable ground traffic control object in the road and bridge area, and reduce the compound risk caused by the superposition of UAV failure risk and high-density traffic flow and bridge structural anomalies through the synchronous action of UAV scheduling and ground traffic flow control.

[0009] Technical solution To address the aforementioned technical problems, this invention provides a method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles (UAVs) and ground-based roads and bridges, comprising the following steps: 1. Data Collection The collaborative management and control platform acquires data within the target road and bridge area. The target road and bridge area includes roads, bridges, bridge entrances, bridge lanes, bridge exits, low-altitude air routes around the bridge, and UAV take-off and landing service points.

[0010] The drone operation data includes drone number, location, speed, heading, altitude, payload level, battery level, mission type, expected flight path, flight envelope, communication link quality, available alternate landing points, and flight control status.

[0011] Road traffic flow data includes cross-sectional flow, average speed, lane occupancy, vehicle density, queue length, proportion of heavy vehicles, vehicle trajectory, accident information, construction information, and event level.

[0012] Bridge structural condition data includes bridge strain, deflection, cable force, vibration, bridge deck wind speed, wind direction, structural health index, structural safety margin, bridge deck slipperiness, and traffic restriction level.

[0013] Traffic control equipment status data includes traffic light status, ramp control equipment status, variable speed limit sign status, variable lane sign status, variable information board status, roadside unit status, and drone take-off and landing service unit status.

[0014] Environmental data includes meteorological data, no-fly zones, restricted-fly zones, temporary control zones, map data, and emergency event data.

[0015] 2. Construct a road-bridge-space-time resource map The collaborative management and control platform divides the target road and bridge area into low-altitude airway voxel units, ground road section units, bridge structure units, and UAV take-off and landing service units.

[0016] Low-altitude airway voxel units consist of horizontal grids, altitude layers, and time windows. Ground road segment units can be lanes, ramps, bridge approach sections, intersection entrance lanes, or tollbooth lanes. Bridge structure units can be bridge spans, bridge towers, main beams, cable-stayed areas, bridge deck lane areas, expansion joint areas, or structural monitoring sections. UAV take-off and landing service units include UAV nests, charging and swapping points, alternate landing points, and waiting points.

[0017] The nodes in the road-bridge-air spacetime resource map include low-altitude airway voxel nodes, ground road segment nodes, bridge structure nodes, traffic control nodes, take-off and landing service nodes, and roadside communication nodes; the edges include low-altitude airway edges, road passage edges, bridge passage edges, risk mapping edges, structurally sensitive edges, information gain edges, and take-off and landing service edges.

[0018] Risk mapping edges represent the correspondence between the UAV risk shadow domain in the low-altitude flight path voxel unit and the ground road segment unit or bridge structure unit. Structure-sensitive edges represent the constraints imposed by the bridge structure status on the available capacity of the low-altitude voxel, flight altitude, or UAV speed. Information gain edges represent the UAV's ability to observe traffic accidents, bridge anomalies, or rescue channel conditions after entering a voxel.

[0019] 3. Generate a risk shadow domain For the low-altitude flight path v to be occupied by UAV u at time step k, generate the risk shadow domain Ω(u,v,k).

[0020] Where u represents the UAV number, v represents the low-altitude airway voxel number, k represents the time step in the rolling time window, and Ω(u,v,k) represents the potential risk coverage area formed on the ground road segment unit or bridge structure unit when UAV u occupies the low-altitude airway voxel unit at time step k.

[0021] The risk shadow domain Ω(u,v,k) includes: (1) Normal flight projection area Ω1. Ω1 represents the horizontal projection area of ​​the UAV's normal flight envelope on the ground or bridge surface.

[0022] (2) Wind drift extension domain Ω2. Ω2 represents the extended area that the UAV may cover after deviating from the normal projection under the influence of crosswinds, gusts, or bridge surface wind fields.

[0023] (3) Failure landing reachable area Ω3. Ω3 represents the ground or bridge area that the UAV may reach under the preset emergency landing model when it experiences power failure, communication abnormality, low battery return failure, or control abnormality.

[0024] Ω(u,v,k)=Ω1∪Ω2∪Ω3 The symbol “∪” represents the union of regions, that is, the risk shadow domain Ω(u,v,k) is composed of the normal flight projection domain Ω1, the wind drift extension domain Ω2, and the failure landing reachability domain Ω3.

[0025] The higher the drone's altitude, the larger its payload, the stronger the crosswinds, the worse the communication link, or the lower the battery level, the greater the risk shadow domain. When a drone has a parachute, redundant power, controlled alternate landing capability, or high-reliability positioning capability, the failure landing reachable domain can be correspondingly reduced.

[0026] When the risk shadow domain overlaps with ground road segment units and bridge structural units, the system establishes a risk mapping relationship. This risk mapping relationship not only represents spatial overlap, but also the coupling strength between the consequences of UAV failure and the exposure of ground vehicles and the sensitivity of bridge structures.

[0027] 4. Calculate the coupling risk value between the open ground and the bridge. For the low-altitude airway voxel element and the ground road segment element or bridge structure element g, the air-ground-bridge coupling risk value R(v,g,k) is calculated at time step k: R(v,g,k)=a1A(v,g,k)+a2Q(g,k)+a3H(g,k)+a4B(g,k)+a5F(v,k)+a6W(v,k)-a7O(v,g,k) Wherein, R(v,g,k) represents the air-ground-bridge coupling risk value of the low-altitude airway voxel unit v and the ground road segment unit or bridge structure unit g at time step k; g represents the number of the ground road segment unit or bridge structure unit; A(v,g,k) represents the degree of overlap between the risk shadow domain and g, which can be determined by the overlap area ratio, overlap length ratio, or exposed vehicle number ratio; Q(g,k) represents the normalized value of vehicle density or queue length; H(g,k) represents the normalized value of the proportion of heavy-load vehicles; B(g,k) represents the normalized value of bridge structure risk, which can be determined based on structural safety margin, bridge vibration, crosswind on the bridge deck, and traffic restriction level; F(v,k) represents the UAV failure consequence level, which can be determined based on UAV mass, payload, flight altitude, and speed; W(v,k) represents wind field or communication environment risk; O(v,g,k) represents the observation gain of the UAV on traffic events, bridge anomalies, or rescue channel status; a1 to a7 are non-negative weights, which can be obtained from historical operating data, expert experience, or offline calibration.

[0028] By introducing an observation gain O(v,g,k), this invention does not simply prohibit drones from entering when the risk is high. Instead, it considers the benefits that drones can bring in emergency missions, such as accident identification, bridge anomaly confirmation, or rescue guidance. Emergency drones are only permitted to pass when the observation gain is sufficient to offset part of the risk and ground traffic control can reduce vehicle exposure to a safe level.

[0029] 5. Generate an airspace-ground interlocking access token The system generates an air-ground interlock access token based on the air-ground bridge coupling risk value.

[0030] The air-to-ground interlock access token refers to an electronic permission status that simultaneously restricts the UAV's entry into the low-altitude airway voxel unit and ground traffic control actions. The token includes at least three types: (1) Passage token: UAVs can enter the corresponding low-altitude airway voxel unit according to the planned time window, and ground traffic maintains normal control.

[0031] (2) Conditional Passage Token: UAVs can only enter the corresponding low-altitude airway voxel unit after the ground traffic control equipment has completed the preset control actions. The preset control actions include reducing the speed limit, restricting ramp inflow, closing risk lanes, diverting heavy vehicles, issuing vehicle-road cooperative avoidance prompts, or adjusting signal timing.

[0032] (3) No-pass token: UAVs for ordinary missions are not allowed to enter the corresponding low-altitude airway voxel units, and UAVs for emergency missions must also wait for the risk to decrease or change to conditional passage through ground pre-clearing measures.

[0033] The token mechanism interlocks drone flight path permits with ground traffic flow control, preventing drones from entering risky voxels while ground vehicles remain in a state of high-density exposure.

[0034] 6. Barrier Function Security Constraints To prevent risks from suddenly exceeding limits during the rolling optimization process, the system sets up a barrier function for safety constraints.

[0035] The safety constraints for the barrier function are: G(g,k)=Rmax(g)-ΣR(v,g,k)x(v,k)-c1D(g,k)-c2B(g,k) Wherein, G(g,k) represents the residual safety margin function of the ground road segment unit or bridge structure unit g at time step k; Rmax(g) represents the maximum allowable risk budget of the ground road segment unit or bridge structure unit g; Σ represents the summation over all relevant low-altitude airway voxel units v; R(v,g,k) represents the air-ground-bridge coupling risk value of the low-altitude airway voxel unit v and the ground road segment unit or bridge structure unit g at time step k; x(v,k) represents whether the low-altitude airway voxel unit v is occupied by a UAV at time step k, x(v,k)=1 indicates that it is occupied, and x(v,k)=0 indicates that it is not occupied; D(g,k) represents the degree of traffic congestion; B(g,k) represents the normalized value of bridge structure risk; c1 and c2 are non-negative coefficients.

[0036] The barrier function also satisfies: G(g,k+1)≥ηG(g,k) Where η represents the safety maintenance coefficient, with a value ranging from 0 to 1; G(g,k+1) represents the remaining safety margin function for the next time step. This constraint indicates that during the rolling control process, the remaining safety margin of the ground road segment unit or bridge structure unit must not decrease at a rate exceeding the preset limit.

[0037] When G(g,k) approaches 0, it indicates that the remaining risk budget for the ground road and bridge unit is insufficient. At this point, the system must reduce drone occupancy, lower ground vehicle density, reduce speed limits, divert heavy-duty vehicles, or close some lanes to ensure that the safety boundary is not breached.

[0038] 7. Rolling Cooperative Scheduling Model The system establishes a collaborative scheduling model with a rolling time window period. Decision variables include UAV route selection, flight altitude, entry time window, speed, holding point, alternate landing point, ground signal timing, ramp control rate, variable speed limit, lane opening status, and variable information prompts.

[0039] The objective function is: J=b1Tuav+b2Troad+b3Euav+b4Ccontrol+b5Rtotal-b6Ototal Where J represents the comprehensive optimization objective value of the rolling collaborative scheduling model; Tuav represents UAV mission delay; Troad represents ground vehicle delay; Euav represents UAV energy consumption; Ccontrol represents traffic control disturbance; Rtotal represents the total coupling risk of the air-ground-bridge system; Ototal represents the total observation gain of the UAV; and b1 to b6 are non-negative weights.

[0040] The rolling collaborative scheduling model aims to minimize J. Since Rtotal is a positive term and Ototal is a negative term in the objective function, the model tends to select drone entry strategies with higher value for traffic event identification or bridge anomaly identification while reducing risk and delay.

[0041] The constraints include UAV safety interval constraints, low-altitude airway voxel capacity constraints, UAV endurance constraints, take-off and landing service unit capacity constraints, road traffic flow conservation constraints, traffic control equipment operation constraints, bridge structure safety constraints, air-ground interlocking access token constraints, and barrier function safety constraints.

[0042] 8. Command execution and feedback updates After the model is solved, the system sends the UAV route, altitude layer, speed, entry time window, waiting point, alternate landing point and return strategy; and sends signal timing, ramp control, variable speed limit, lane opening status, variable message prompts and vehicle-to-infrastructure (V2I) broadcasts to traffic control equipment.

[0043] After execution, the system receives feedback from drones, roadside sensing devices, bridge structure monitoring devices, and traffic control devices, updates the risk shadow domain, air-ground-bridge coupling risk value, air-ground interlocking access token, and rolling scheduling model, and then enters the next rolling time window.

[0044] Beneficial effects Compared with the prior art, the present invention provides a method for coordinated scheduling and traffic flow control of low-altitude UAVs and ground roads and bridges, which has the following beneficial effects: 1. The flight risks of drones are concretized into a ground-calcifiable risk shadow domain, enabling the low-altitude failure risk of drones to directly participate in road traffic flow control.

[0045] 2. By using air-to-ground interlock access tokens, the drone's entry into low-altitude voxels can be interlocked with ground-based speed reduction, flow restriction, lane adjustment, or diversion operations to prevent air-to-ground control from becoming disconnected.

[0046] 3. By using barrier function safety constraints, the risks of bridge structure, ground congestion, and drone occupancy are constrained within a unified risk budget.

[0047] 4. For emergency drone missions, the approach is not simply to prohibit flights or to prioritize their passage. Instead, a safe observation corridor is formed by pre-clearing the ground, reducing density, and limiting flow, thereby improving the efficiency of accident identification and bridge anomaly confirmation.

[0048] 5. In the event of abnormal bridge structure, crosswinds on the bridge deck, concentrated heavy vehicles, or abnormal UAV communication, it can simultaneously adjust the low-altitude airway capacity and ground traffic control strategies to reduce compound risks. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating the overall process of the method of the present invention.

[0050] Figure 2 This is a schematic diagram of the road-bridge-space spatiotemporal resource map of the present invention.

[0051] Figure 3 This is a schematic diagram illustrating the generation of the risk shadow domain in this invention.

[0052] Figure 4 This is a schematic diagram illustrating the generation of the air-ground interlocking access token according to the present invention.

[0053] Figure 5 This is a flowchart of the rolling collaborative scheduling process under the security constraints of the barrier function in this invention.

[0054] Figure 6 This is a flowchart of the abnormal interlock mode of the present invention. Detailed Implementation

[0055] The present invention will be further described below with reference to the accompanying drawings and embodiments. The following embodiments are used to illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.

[0056] Example 1: Routine drone inspection of cross-river bridges See Figure 1 and Figure 2 A certain cross-river bridge is equipped with bridge deck traffic detectors, a bridge structure monitoring system, variable speed limit signs, variable message signs, roadside communication units, and drone nests. The collaborative management and control platform divides the space above the bridge into multiple low-altitude airway voxel units, divides the bridge deck into bridge structure units and ground road section units according to the bridge spans and lanes, and establishes a road-bridge-air spatiotemporal resource map.

[0057] In this embodiment, a bridge inspection drone is planned to enter the low-altitude flight path voxel above the main span. See also Figure 3 The collaborative management platform generates a risk shadow domain Ω(u,v,k) based on the drone's altitude, speed, payload, crosswind, and battery level. Here, Ω(u,v,k) represents the risk coverage area on the bridge or ground corresponding to the drone u occupying the low-altitude flight path v at time step k.

[0058] The system further calculates the overlap between the risk shadow domain and the monitoring sections of the second and third lanes of the bridge deck, as well as the main girder. If low traffic density, normal bridge structural safety margin, and wind speed below a threshold are detected in this area, the air-ground-bridge coupling risk value R(v,g,k) is lower than the first threshold. See also Figure 4 The system generates a pass token. The drone enters the corresponding voxel as planned to perform inspections, while ground traffic maintains normal control.

[0059] After execution, see Figure 5 The system receives images from UAV inspections, bridge structure monitoring data, and traffic detection data, updates the road-bridge-air spatiotemporal resource map, and then enters the next scrolling time window.

[0060] Example 2: Conditional passage under bridge congestion conditions See Figure 1 , Figure 3 and Figure 4 When congestion occurs on the bridge, the system detects an increase in vehicle density and queue length in the lane corresponding to the risk shadow domain. Although the UAV can still pass through from the perspective of low-altitude airway voxel capacity, its failed landing reachability domain overlaps with the high-density traffic flow, causing the air-ground-bridge coupling risk value R(v,g,k) to rise to between the first and second thresholds.

[0061] At this point, the system generates a conditional access token. The conditional access token includes the conditions for drone entry and the ground traffic control conditions. The ground traffic control conditions include: the ramp control equipment before the bridge reduces the entrance flow, the variable speed limit sign reduces the speed limit on the bridge to a preset value, the variable message sign prompts vehicles to maintain a safe distance, and the roadside unit broadcasts a low-altitude operation prompt to the vehicle terminal.

[0062] See Figure 5 Once the system confirms, based on feedback from traffic detection equipment, that the vehicle density within the risk shadow domain has decreased below the threshold and the barrier function G(g,k) satisfies the safety maintenance constraint, it issues an entry time window and flight altitude layer to the UAV. The UAV then enters the voxel above the bridge to perform inspection or passage tasks.

[0063] In this embodiment, ground traffic flow control is not executed independently, but is a prerequisite for UAVs to obtain low-altitude passage permission, thereby forming an air-ground interlock.

[0064] Example 3: Emergency Detection of Traffic Accidents See Figure 1 , Figure 3 , Figure 4 and Figure 5 After a traffic accident occurs on the bridge, the roadside sensing equipment detects abnormal parking and a rapid increase in queues. The collaborative management platform marks the lane on the bridge where the accident occurred as an event unit and includes the corresponding bridge structure unit and ground road section unit in the key calculation scope.

[0065] When a regular UAV enters the voxel above the accident area and its air-to-ground bridge coupling risk value R(v,g,k) exceeds the second threshold, the system initially generates a no-pass token. However, an emergency reconnaissance UAV, whose mission is accident detection, has an observation gain O(v,g,k) higher than the preset observation threshold. The observation gain O(v,g,k) represents the contribution of the UAV, after entering the low-altitude airway voxel, to the identification of the accident-occupied area, personnel stranded status, hazardous material leakage, or rescue channel status.

[0066] After determining that the drone meets the emergency priority conditions, the system converts the no-pass token into a conditional pass token. The conditional pass token requires ground traffic equipment to perform pre-clearance control, including reducing the speed limit upstream of the accident, restricting traffic flow at the bridge entrance, closing adjacent high-risk lanes, and issuing detour and avoidance prompts to vehicles.

[0067] Once the risk of ground vehicle exposure decreases and the barrier function G(g,k) satisfies the safety maintenance constraint, the emergency drone enters the voxel above and to the side of the accident area for detection. The drone transmits back the location of the accident vehicles, occupied lanes, stranded personnel, and passage conditions for rescue vehicles. Based on the transmitted results, the system further adjusts the timing of the bridge-front signals, the access lanes for rescue vehicles, and the lane control on the bridge.

[0068] This embodiment illustrates that the present invention does not simply allow emergency drones to enter, but rather determines whether to allow them to enter under certain conditions through a combination of risk shadow domain, observation gain, and ground pre-clearing actions.

[0069] Example 4: Protective Management under Bridge Structural Anomalies See Figure 2 , Figure 3 , Figure 4 and Figure 6 When the bridge structure monitoring system detects an increase in the vibration level of a bridge span and the proportion of heavily loaded vehicles on the bridge deck exceeds a threshold, the system increases the bridge structure risk normalization value B(g,k) of the corresponding bridge structural unit. B(g,k) represents the bridge structure risk normalization value; the higher the value, the higher the bridge structure risk.

[0070] As B(g,k) increases, the air-to-ground bridge coupling risk value R(v,g,k) also increases. The system expands the risk shadow domain associated with the bridge span, reduces the available capacity of relevant low-altitude route voxel units, and guides ordinary mission UAVs to the bridge-side bypass route.

[0071] For mandatory bridge anomaly inspection drones, the system generates conditional access tokens. These tokens require drones to use higher altitudes, lower speeds, one-way routes, and waiting points outside the bridge; simultaneously, the ground side implements heavy vehicle diversion, entrance flow control, and bridge speed reduction.

[0072] See Figure 5 If the barrier function G(g,k) is close to 0, it indicates that the remaining risk budget of the bridge structural unit is insufficient. The system further closes the relevant low-altitude voxels and restricts the flow at the bridge deck inlet until the safety margin of the bridge structure recovers to above the threshold.

[0073] This embodiment illustrates that abnormal bridge structures not only affect vehicle traffic on the bridge deck, but also directly impact the voxel capacity of low-altitude airways and the entry conditions for unmanned aerial vehicles.

[0074] Example 5: Abnormal Interlock Mode under UAV Communication Anomalies See Figure 3 , Figure 4 and Figure 6 When the communication quality of the drone above the bridge degrades and falls below a threshold, the system activates an abnormal interlock mode.

[0075] The system regenerates the risk shadow domain based on the drone's current location, speed, altitude, battery level, and expected return path. Ω(u,v,k). If the risk shadow domain overlaps with the high-density traffic flow on the bridge, the system will adjust the token of the corresponding low-altitude airway voxel from a pass-allow token to a conditional pass token or a prohibit token.

[0076] When the conditional pass token is active, the system issues speed reduction, risk warning, lane adjustment, or entrance flow restriction commands to traffic control equipment; simultaneously, it issues waiting points or alternate landing points outside the bridge to the drone. If the drone enters the loss-of-connection return procedure, the system dynamically updates the risk shadow domain based on the predicted return path and continuously adjusts ground traffic control commands.

[0077] This embodiment illustrates that, in the event of an abnormal drone condition, the present invention does not only require the drone to return to its home location, but also simultaneously incorporates the ground areas that it may affect into traffic flow protection and control.

[0078] Example 6: A scenario where bridge construction and drone logistics coexist. See Figure 1 , Figure 2 and Figure 4 When temporary construction occupies the road in front of the bridge, the traffic capacity of the ground road segment decreases and the queue length of vehicles increases. At this time, if a regular logistics drone plans to pass through the low-altitude voxel above the road in front of the bridge, the system calculates the air-ground-bridge coupling risk value R(v,g,k) based on the construction area, queued vehicles, and the drone's failure landing reachability domain.

[0079] When R(v,g,k) exceeds the second threshold, the system generates a no-pass token for ordinary logistics drones and guides them to a detour route or waiting point. If the drone mission can be postponed, the system recalculates the risk value in a subsequent rolling time window; if the construction queue eases and the barrier function G(g,k) returns to a safe range, the system adjusts the no-pass token to a pass token or a conditional pass token.

[0080] This embodiment illustrates that the present invention is also applicable to scenarios where there is increased exposure to ground traffic but not bridge-level accidents, and can avoid the compound risk of ordinary drone missions being combined with ground construction congestion.

[0081] This invention can be applied to cross-river bridges, cross-sea bridges, urban expressways, highway bridge and tunnel complexes, port roads, roads surrounding airports, and traffic support areas for large-scale events. The system can be connected to traffic operation management platforms, bridge health monitoring platforms, UAV nesting platforms, vehicle-road cooperative platforms, and low-altitude flight service platforms, and can be implemented through existing UAV flight control interfaces, roadside communication units, traffic signal controllers, variable speed limit signs, and bridge monitoring systems.

Claims

1. A method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles (UAVs) and ground-based road and bridge infrastructure, characterized in that: The process is executed by the collaborative management platform and includes the following steps: S1. Acquire drone operation data, road traffic flow data, bridge structure status data, traffic control equipment status data, and environmental data within the target road and bridge area; S2. Divide the target road and bridge area into low-altitude airway voxel units, ground road segment units, bridge structure units, and UAV take-off and landing service units, and establish a road-bridge-air spatiotemporal resource map including the low-altitude airway voxel units, ground road segment units, bridge structure units, and UAV take-off and landing service units. S3. For each low-altitude airway voxel unit that the UAV is to occupy in the future rolling time window, generate risk shadow domains corresponding to ground road segment units and bridge structure units based on the UAV's flight envelope, payload level, speed, altitude, battery status, wind field drift amount and failure landing reachability domain. S4. Based on the overlap between the risk shadow domain and the ground road segment unit and bridge structure unit, and combined with vehicle density, vehicle speed, proportion of heavy vehicles, bridge structure safety margin, bridge vibration level, wind speed, communication quality and UAV mission priority, calculate the air-ground-bridge coupling risk value. S5. Based on the air-ground bridge coupling risk value, generate an air-ground interlocking access token for the low-altitude airway voxel unit, ground road segment unit, and bridge structure unit; wherein, the air-ground interlocking access token is used to limit the conditions for UAVs to enter the corresponding low-altitude airway voxel unit within the same rolling time window, as well as the traffic flow control conditions for the corresponding ground road segment unit or bridge structure unit. S6. Taking UAV mission delay, ground vehicle delay, UAV energy consumption, traffic control disturbance and air-ground-bridge coupling risk as optimization objectives, and taking the air-ground interlocking access token, UAV safety interval, low-altitude airway capacity, UAV endurance, take-off and landing service capacity, road traffic capacity and bridge structural safety margin as constraints, solve the rolling collaborative scheduling model to generate UAV scheduling instructions and ground traffic flow control instructions. S7. Issue at least one UAV dispatching instruction to the UAV, including route, flight altitude layer, speed, entry time window, waiting point, alternate landing point or return strategy, and issue at least one ground traffic flow control instruction to traffic control equipment, including signal timing, ramp control, variable speed limit, lane open status, variable information prompt or vehicle-road cooperative broadcast. S8. Based on the feedback results from the drone, traffic detection equipment, bridge structure monitoring equipment, and traffic control equipment, update the road-bridge-air spatiotemporal resource map, risk shadow domain, and air-ground interlocking access token, and enter the next rolling time window to repeat S3 to S8.

2. The method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles and ground roads and bridges according to claim 1, characterized in that, The risk shadow domain includes the normal flight projection domain, the wind drift extension domain, and the failed landing reach domain; wherein, the normal flight projection domain is determined by the horizontal projection of the UAV in the low-altitude route voxel unit, the wind drift extension domain is determined by wind speed, wind direction, UAV speed, and UAV altitude, and the failed landing reach domain is determined by UAV altitude, speed, remaining battery power, payload level, and preset emergency landing model.

3. The method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles and ground roads and bridges according to claim 1, characterized in that, The air-ground bridge coupling risk value R(v,g,k) is calculated as follows: R(v,g,k)=a1A(v,g,k)+a2Q(g,k)+a3H(g,k)+a4B(g,k)+a5F(v,k)+a6W(v,k)-a7O(v,g,k) Wherein, R(v,g,k) represents the air-ground-bridge coupling risk value of the low-altitude airway voxel unit v and the ground road segment unit or bridge structure unit g at time step k; A(v,g,k) represents the degree of overlap between the risk shadow domain and the ground road segment unit or bridge structure unit; Q(g,k) represents the normalized value of vehicle density or queue length; H(g,k) represents the normalized value of the proportion of heavy-load vehicles; B(g,k) represents the normalized value of bridge structure risk; F(v,k) represents the UAV failure consequence level; W(v,k) represents the wind field or communication environment risk; O(v,g,k) represents the observation gain of UAV on ground traffic events or bridge anomalies; a1 to a7 are non-negative weights.

4. The method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles and ground roads and bridges according to claim 1, characterized in that, The air-ground interlock access tokens include permission tokens, conditional access tokens, and prohibition tokens. When the air-ground bridge coupling risk value is lower than the first threshold, a passage permit token is generated; When the air-ground-bridge coupling risk value is greater than or equal to the first threshold and lower than the second threshold, a conditional passage token is generated. The conditional passage token requires that before the UAV enters the corresponding low-altitude route voxel unit, at least one of the following traffic flow control measures be implemented on the ground road segment unit or bridge structure unit corresponding to the risk shadow domain: speed reduction, flow restriction, lane adjustment, vehicle diversion, or vehicle-road cooperative prompt. When the air-to-ground bridge coupling risk value is greater than or equal to the second threshold, a no-pass token is generated to restrict ordinary mission UAVs from entering the corresponding low-altitude airway voxel unit.

5. The method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles and ground roads and bridges according to claim 1, characterized in that, The rolling collaborative scheduling model includes a barrier function security constraint; The security constraint of the barrier function is: G(g,k)=Rmax(g)-ΣR(v,g,k)x(v,k)-c1D(g,k)-c2B(g,k); And it satisfies: G(g,k+1)≥ηG(g,k); Wherein, G(g,k) represents the remaining safety margin function of the ground road segment unit or bridge structure unit g at time step k; Rmax(g) represents the maximum allowable risk budget of the ground road segment unit or bridge structure unit g; R(v,g,k) represents the air-ground-bridge coupling risk value of the low-altitude airway v and the ground road segment unit or bridge structure unit g at time step k; x(v,k) represents whether the low-altitude airway v is occupied by a drone at time step k; D(g,k) represents the degree of traffic congestion; B(g,k) represents the normalized value of bridge structure risk; c1 and c2 are non-negative coefficients; η is the safety maintenance coefficient between 0 and 1.

6. The method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles and ground roads and bridges according to claim 1, characterized in that, When the drone mission is to detect traffic accidents, inspect bridge anomalies, deliver emergency supplies, or guide rescue, if the observation gain of the drone is greater than the preset observation threshold, the prohibition token can be converted into a conditional passage token. The conditional access token requires that, before the UAV enters the corresponding low-altitude route voxel unit, at least one of the following operations be performed on the bridge entrance, bridge lane, or adjacent road section corresponding to the risk shadow domain: reducing speed limit, restricting entrance flow, closing risk lanes, diverting heavy vehicles, or issuing vehicle-road cooperative avoidance prompts.

7. The method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles and ground roads and bridges according to claim 1, characterized in that, When the safety margin of the bridge structure is lower than the preset safety margin threshold, the bridge vibration level is higher than the preset vibration threshold, the crosswind on the bridge deck is higher than the preset wind speed threshold, or the proportion of heavy-load vehicles on the bridge deck is higher than the preset proportion threshold, the risk shadow domain associated with the bridge structural unit is expanded, and the available capacity of the relevant low-altitude airway voxel unit is reduced.

8. The method for coordinated scheduling and traffic flow control of low-altitude unmanned aerial vehicles and ground roads and bridges according to claim 1, characterized in that, When an abnormal interlock mode is activated, the drone communication interruption, yaw, battery level below the return threshold, bridge accident level increase, or sudden weather conditions are detected. In the abnormal interlock mode, the air-to-ground interlock access token of the corresponding low-altitude route voxel unit is adjusted to a no-pass token or a conditional pass token, and speed reduction, diversion, lane closure, emergency lane protection or risk warning instructions are issued to the ground road segment unit or bridge structure unit corresponding to the risk shadow domain.

9. A low-altitude unmanned aerial vehicle (UAV) and ground road / bridge collaborative scheduling and traffic flow control system, characterized in that, include: The data access module is used to acquire drone operation data, road traffic flow data, bridge structure status data, traffic control equipment status data, and environmental data; The spatiotemporal resource map construction module is used to construct a road-bridge-space spatiotemporal resource map; The risk shadow domain generation module is used to generate risk shadow domains based on the UAV's flight envelope, payload level, speed, altitude, battery status, wind field drift, and failure landing reachability. The interlock token generation module is used to generate air-ground interlock access tokens based on the risk shadow domain, road traffic flow status, and bridge structure status. The rolling collaborative optimization module is used to generate UAV scheduling instructions and ground traffic flow control instructions under the safety constraints of the barrier function; The execution feedback module is used to issue instructions and update the road-bridge-space spatiotemporal resource map, risk shadow domain, and air-ground interlock access token based on the feedback results.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for coordinated scheduling and traffic flow control of low-altitude UAVs and ground roads and bridges as described in any one of claims 1 to 8.

Images