Low-altitude airspace obstacle management system and method based on multi-dimensional information fusion

Abstract

This application provides a low-altitude airspace obstacle management system and method based on multi-dimensional information fusion. The system includes a geometric description module, a semantic level module, a risk assessment module, and a behavior constraint module that are interconnected. The geometric description module is used to collect and store basic spatial information of each obstacle in the low-altitude airspace and transmit the basic spatial information to the semantic level module, the risk assessment module, and the behavior constraint module. The semantic level module is used to generate semantic weight coefficients for each obstacle based on the basic spatial information and transmit the semantic weight coefficients to the risk assessment module. The risk assessment module is used to calculate the comprehensive risk score of each obstacle based on the basic spatial information and the semantic weight coefficients, and determine the risk level of each obstacle based on the comprehensive risk score. The behavior constraint module is used to generate and output the flight behavior constraints of the aircraft when approaching or bypassing each obstacle based on the basic spatial information and the risk level.

Description

Technical Field

[0001] This invention relates to the field of low-altitude airspace management technology, and in particular to a low-altitude airspace obstacle management system and method based on multi-dimensional information fusion. Background Technology

[0002] With the widespread application of drones in urban low-altitude logistics, emergency rescue, and inspection mapping, the usage density and flight frequency of low-altitude airspace have significantly increased. In these scenarios, drones typically need to navigate complex environments such as densely built-up areas and infrastructure corridors. This results in numerous fixed obstacles such as buildings, communication towers, power lines, and bridges within the low-altitude airspace, posing a continuous challenge to drone flight safety. Currently, low-altitude obstacle information is typically described in the following ways: 1. Spatial description methods based on three-dimensional coordinates or grids are only used to determine whether space is occupied by obstacles; 2. Based on the geometric representation of GIS vector elements or 3D models (such as polygons, triangular meshes), the focus is on describing the shape and height of obstacles; 3. Collision detection models based on simple bounding volumes (such as cuboids and cylinders) are used for real-time obstacle avoidance calculations.

[0003] The aforementioned solutions primarily focus on the geometric properties of obstacles, typically treating them as non-traversable areas or areas requiring a fixed safe distance during UAV path planning and flight decision-making. However, in actual low-altitude flight, different obstacles have varying impacts on UAV flight safety and behavior. For example, obstacles such as communication towers and power lines pose high risks and require greater safety margins; ordinary building rooftops allow for close-range flight below a certain height; and constraints on some obstacles can be temporarily relaxed or strengthened under specific times or flight missions. However, existing technologies fail to uniformly express these semantic information and behavioral constraints at the system level, leading to the need for multiple sets of rule logic in the flight decision-making system, resulting in high system complexity and poor scalability. Furthermore, existing technologies also have significant shortcomings in the following aspects: 1. Information silo problem: Geometric data, risk assessment results, and flight constraint rules are stored in different systems or modules, making it difficult to achieve unified access and collaborative decision-making; 2. Lack of semantic understanding: Unable to distinguish between ordinary buildings that can be bypassed at close range and high-risk facilities that must be strictly avoided, resulting in overly conservative flight strategies or potential safety hazards; 3. Static constraint rules: Behavioral constraints are usually fixed in the code as fixed parameters and cannot be dynamically adjusted according to task type, drone performance, and environmental changes; 4. Poor decision interpretability: The path planning results lack a traceable relationship with obstacle attributes, which is not conducive to flight safety audits and accident analysis. Summary of the Invention

[0004] The purpose of this invention is to provide a low-altitude airspace obstacle management system and method based on multi-dimensional information fusion, aiming to solve the above-mentioned problems in the prior art.

[0005] This invention provides a low-altitude airspace obstacle management system based on multi-dimensional information fusion, comprising: a geometric description module, a semantic level module, a risk assessment module, and a behavior constraint module that are interconnected. The geometric description module is used to collect and store basic spatial information of various obstacles in the low-altitude airspace, and transmit the basic spatial information to the semantic level module, risk assessment module and behavior constraint module. The semantic level module is used to generate semantic weight coefficients for each obstacle based on the basic spatial information, and transmit the semantic weight coefficients to the risk assessment module. The risk assessment module is used to calculate the comprehensive risk score of each obstacle based on the basic spatial information and the semantic weight coefficient, determine the risk level of each obstacle based on the comprehensive risk score, and send the risk level to the behavior constraint module. The behavior constraint module is used to generate and output flight behavior constraints for the aircraft when approaching or circling various obstacles based on the basic spatial information and the risk level.

[0006] This invention provides a method for low-altitude airspace obstacle management based on multi-dimensional information fusion, comprising: The basic spatial information of each obstacle in the low-altitude airspace is collected and stored by the geometric description module, and the basic spatial information is transmitted to the semantic level module, risk assessment module and behavior constraint module. The semantic weight coefficients of each obstacle are generated by the semantic level module based on the basic spatial information, and the semantic weight coefficients are transmitted to the risk assessment module. The risk assessment module calculates a comprehensive risk score for each obstacle based on the basic spatial information and the semantic weight coefficient, determines the risk level of each obstacle based on the comprehensive risk score, and sends the risk level to the behavior constraint module. The behavior constraint module generates and outputs flight behavior constraints for the aircraft when approaching or circling various obstacles based on the basic spatial information and the risk level. The path planning module maps each obstacle to its corresponding cost region based on its basic spatial information, risk level, and flight behavior constraints, constructs a spatial cost model, and performs path search based on the spatial cost model and flight behavior constraints to output the flight path that meets safety requirements and has the lowest total cost value.

[0007] This invention also provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the above-described low-altitude airspace obstacle management method based on multi-dimensional information fusion.

[0008] This invention also provides a computer-readable storage medium storing an information transmission implementation program, which, when executed by a processor, implements the steps of the aforementioned low-altitude airspace obstacle management method based on multi-dimensional information fusion.

[0009] The following beneficial effects can be achieved by adopting the embodiments of the present invention: The embodiments of the present invention propose a system and method for digital representation and intelligent management of fixed or semi-fixed obstacles in low-altitude flight scenarios of UAVs. The system integrates the geometric information, semantic level, risk assessment results and behavioral constraint rules of the obstacles, and can be applied to UAV path planning, obstacle avoidance decision-making, digital management of low-altitude airspace and collaborative flight of unmanned systems. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in one or more embodiments of this specification or in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram of a low-altitude airspace obstacle management system based on multi-dimensional information fusion according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the system structure composition according to an embodiment of the present invention; Figure 3 This is a system principle block diagram of an embodiment of the present invention; Figure 4 This is a schematic diagram of drone path planning according to an embodiment of the present invention; Figure 5 This is a flowchart of a low-altitude airspace obstacle management method based on multi-dimensional information fusion, according to an embodiment of the present invention. Detailed Implementation

[0012] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0013] System Implementation Examples According to embodiments of the present invention, a low-altitude airspace obstacle management system based on multi-dimensional information fusion is provided. Figure 1 This is a schematic diagram of a low-altitude airspace obstacle management system based on multi-dimensional information fusion, according to an embodiment of the present invention. Figure 1 As shown, the low-altitude airspace obstacle management system based on multi-dimensional information fusion according to an embodiment of the present invention specifically includes: a geometric description module 10, a semantic level module 12, a risk assessment module 14, and a behavior constraint module 16 that are interconnected with each other; The geometric description module 10 is used to collect and store basic spatial information of various obstacles in the low-altitude airspace, and transmit the basic spatial information to the semantic level module, risk assessment module and behavior constraint module. The basic spatial information includes the obstacle's unique identifier ID, geographical coordinates, spatial morphology description type, geometric data set, and actual height or height range.

[0014] The semantic level module 12 is used to generate semantic weight coefficients for each obstacle based on the basic spatial information, and transmit the semantic weight coefficients to the risk assessment module; specifically, it is used for: Based on the basic spatial information, obstacle type labels, criticality levels, and population or asset density levels are generated for each obstacle, and semantic weight coefficients for each obstacle are generated according to at least one of the obstacle type labels, criticality levels, and population or asset density levels. The obstacle type labels include buildings, communication towers, power lines, bridges, temporary facilities, or trees; The criticality levels include ordinary, important, critical, and extremely high risk; The population or asset density levels include low density, medium density, high density, and very high density.

[0015] The risk assessment module 14 is used to calculate a comprehensive risk score for each obstacle based on the basic spatial information and the semantic weight coefficient, determine the risk level of each obstacle based on the comprehensive risk score, and send the risk level to the behavior constraint module; specifically, it is used for: The altitude risk factor is calculated using Formula 1 based on the ratio of obstacle height to aircraft flight altitude and the altitude difference. The morphological risk factor is calculated using Formula 2 based on the morphological complexity of the obstacle's geometry. The semantic risk factor is calculated using Formula 3 based on the semantic weight coefficients. Based on the uncertainty and validity period of obstacle location information, the positioning uncertainty risk factor is calculated using Formula 4. The environmental risk factor is calculated using Formula 5 based on the environmental parameters of the environment in which the obstacle is located. Formula 1; Formula 2; Formula 3; Formula 4; Formula 5; in, Indicates the height of the obstacle. Indicates the maximum safe flight altitude. Indicates the current flight altitude of the aircraft. Indicates a high sensitivity coefficient. Represents the morphological complexity coefficient. This represents the projected area of ​​the obstacle on the horizontal plane. Indicates the reference area. , Indicates the weighting coefficient. Indicates the obstacle type weight. Indicates the key level weight. Indicates the weight of population or asset intensity. , , These represent the weight coefficients of each factor in semantic risk. Indicates the uncertainty of positioning. Indicates the uncertainty threshold. This indicates the time interval since the last position update. Indicates the validity period of location information. Indicates the current wind speed. Indicates the safe wind speed threshold. Indicates the current rainfall intensity. Indicates the safe rainfall threshold. Indicates the current electromagnetic interference intensity. Indicates the safe electromagnetic interference threshold. , , These represent the weighting coefficients of each factor in environmental risk; The weighted fusion model shown in Formula 6 is used to fuse and calculate the high risk factor, morphological risk factor, semantic risk factor, positioning uncertainty risk factor and environmental risk factor to obtain a comprehensive risk score; Formula 6; in, This represents the overall risk score. , , , , These represent the weighting coefficients of each risk factor; The comprehensive risk score is mapped to a risk level based on a preset risk scoring threshold; wherein the risk level includes low risk, medium risk, high risk and extremely high risk.

[0016] The behavior constraint module 16 is used to generate and output flight behavior constraint conditions for the aircraft when approaching or circling various obstacles based on the basic spatial information and the risk level, specifically for: Define the minimum safe distance that must be maintained between the aircraft and the obstacle; wherein, the minimum safe distance includes the minimum horizontal safe distance, the minimum vertical safe distance, and the radius of the safety buffer zone; Define flight restriction rules for aircraft around obstacles; wherein, the flight restriction rules include no-fly signs, altitude limits, speed limits and permitted flight directions; Define a set of conditional rules that are triggered under specific conditions; each conditional rule includes the triggering condition, the action to be performed, and the rule priority. Define mission adaptation constraints related to specific flight missions of the aircraft; wherein, the mission adaptation constraints include mission type, aircraft capability requirements, and dynamic adjustment flags; Based on the basic spatial information and the risk level, and in combination with at least one of the minimum safe distance, flight restriction rules, condition rule set and mission adaptation constraints, flight behavior constraints are generated and output.

[0017] The system also includes: The path planning module, connected to the geometry description module, risk assessment module, and behavior constraint module, is used to obtain the basic spatial information, risk level, no-fly sign, safety buffer radius, and flight behavior constraint conditions of each obstacle from the geometry description module, risk assessment module, and behavior constraint module, respectively. Based on the risk level and the no-fly sign, it maps each obstacle to the corresponding cost region and assigns a corresponding cost multiplier to each cost region. A spatial cost model is constructed based on the cost multiplier, the safety buffer radius, and the distance from the current position of the aircraft to the geometric surfaces of each obstacle; wherein, the spatial cost model is used to calculate the basic cost of any point in three-dimensional space affected by obstacles; An improved A* algorithm is used to perform path search. During the search process, the basic cost is used as the cost of node expansion, and the flight behavior constraints are used as the basis for verifying the feasibility of nodes to generate candidate flight paths. The candidate flight paths are subjected to constraint verification and cost evaluation, and the flight path that meets the safety requirements and has the lowest total value is output.

[0018] The following describes the specific details of the low-altitude airspace obstacle management system based on multi-dimensional information fusion, as described in the embodiments of the present invention. Figure 2 As shown, the above technical solutions of the embodiments of the present invention will be described in detail.

[0019] This invention addresses the problems of existing low-altitude obstacle information representation methods, missing semantic information, separation of risk assessment and behavioral constraints, and lack of dynamic adaptation capabilities. It proposes a low-altitude airspace obstacle management system that combines obstacle semantic levels, risk assessment, and behavioral constraints. The technical problems to be solved include: 1. How to uniformly express the geometric features, semantic attributes, risk levels, and behavioral constraints of obstacles at the system level to eliminate information silos; 2. How to enable the flight decision-making system to directly perform risk assessment and path planning based on a unified information model, reducing the fragmented implementation of rules and logic; 3. How to achieve dynamic adaptation of behavioral constraints to different task types and different drone capabilities; 4. How to improve the interpretability and traceability of path planning results and obstacle attributes.

[0020] To address the aforementioned issues, this invention proposes a low-altitude airspace obstacle management system. This system employs a four-layer interconnected architecture, consisting of four core modules. These modules are logically interconnected and interact uniformly at the data level, forming a complete closed-loop system from physical space perception to flight decision execution. 1. Geometric Description Module: This module stores basic spatial information about obstacles and forms the basis for the system's spatial judgment and collision detection. Specifically, this module includes: A. Obstacle Unique Identifier ID: Used to uniquely identify each obstacle object in the system, supporting indexing and querying in multi-obstacle scenarios; B. Geographic location coordinates: Supports both latitude and longitude coordinates and three-dimensional Cartesian coordinates, and can adapt to the data formats of different positioning systems; C. Spatial morphology description types: including polygons, bounding boxes, cylinders, voxel sets and other morphology types. The system can select the most appropriate expression method according to the actual physical shape of the obstacle. D. Geometric Data Set: Stores corresponding geometric data according to the spatial shape type, such as the vertex coordinate set of a polygon, the length, width and height parameters of a bounding box, the coordinates of the center of the base of a cylinder and its radius, etc. E. Actual height or height range: Records the absolute height value or height variation range of obstacles for vertical collision detection and path planning.

[0021] This module provides the system with basic spatial judgment capabilities and serves as a spatial positioning reference for other modules. Through this module, the system can quickly determine whether the drone has entered a space occupied by an obstacle or calculate the spatial distance between the drone and the obstacle.

[0022] 2. Semantic Level Module: This module describes the classification and importance level of obstacles in flight safety semantics, serving as the core basis for the system to make differentiated flight decisions. Specifically, this module includes: A. Obstacle Type Labels: Obstacles are semantically classified, including but not limited to buildings, communication towers, power lines, bridges, temporary facilities, trees, etc. Different types correspond to different flight safety considerations.

[0023] B. Criticality Level: Based on the degree of impact of obstacles on flight safety, levels are classified into four categories: Ordinary, Important, Critical, and Extremely High Risk. For example, ordinary residential buildings can be classified as Ordinary, hospitals or government buildings as Important, communication towers as Critical, and high-voltage power lines as Extremely High Risk.

[0024] C. Population or Asset Density Level: This level is assigned based on the population density or asset value surrounding the obstacle, and is categorized as low density, medium density, high density, and very high density. This level is used to assess the risk of secondary disasters that may result from an obstacle falling.

[0025] D. Semantic weight coefficient: The discrete semantic labels mentioned above are transformed into continuous values ​​that can participate in quantitative calculations by combining the analytic hierarchy process and expert scoring method. The value ranges from 0 to 1. This coefficient is used for the calculation of the risk assessment module, thereby organically integrating semantic information and quantitative assessment.

[0026] The semantic level module enables the system to understand the safety implications of different obstacles, thereby enabling differentiated management in subsequent risk assessments and behavioral constraints.

[0027] 3. Risk Assessment Module: This module describes the obstacle risk value calculated based on multiple factors and is a core component for the system to achieve safety quantification. Specifically, this module includes: A. Risk Score: A quantitative risk value calculated by combining multiple risk factors, ranging from 0 to 1, with a higher value indicating greater risk.

[0028] B. Risk Level: Qualitative risk level is obtained by mapping the risk score, including four levels: low risk, medium risk, high risk, and extremely high risk, which facilitates rapid decision-making by the system.

[0029] C. Risk Factor Decomposition: The system decomposes the overall risk into multiple quantifiable risk factors, including but not limited to: High risk: Calculated based on the relationship between obstacle height and drone flight altitude, the higher the obstacle or the closer it is to the flight altitude, the greater the risk; Morphological risk: Calculated based on the complexity of the obstacle's geometry. The more irregular the shape of the obstacle and the easier it is to cause a collision, the greater the risk. Location uncertainty risk: Based on the accuracy and confidence level of obstacle location information, the more inaccurate the location information, the greater the risk of the obstacle. Environmental risk: Calculated based on factors such as meteorological conditions and electromagnetic interference in the environment where the obstacle is located. The more severe the environment, the greater the risk of the obstacle. Semantic risk: Calculated based on the semantic weight coefficients passed by the semantic level module. The higher the semantic importance of an obstacle, the greater its risk.

[0030] D. Risk confidence level: This indicates the reliability of the risk assessment results. The value ranges from 0 to 1, with a higher value indicating a more reliable assessment result.

[0031] E. Validity period information: Records the valid time range of the risk assessment results. The system can automatically trigger an assessment update based on the validity period.

[0032] F. Last Update Time: Records the timestamp of the last risk assessment, used to determine the timeliness of the assessment results.

[0033] This module adopts a multi-factor fusion mechanism, which comprehensively considers multiple dimensions such as geometric features, semantic information, and dynamic environment, and supports dynamic updates based on flight environment parameters, UAV performance parameters, or historical data.

[0034] 4. Behavior Constraint Module: This module defines the flight behavior rules that the UAV must follow when approaching or circumventing obstacles. It is the execution layer for the system's flight strategy control. This module specifically includes: A. Safety Distance Constraints: Define the minimum safe distance that the drone must maintain between itself and obstacles, specifically including: Minimum horizontal safe distance: The minimum distance that a drone must maintain in the horizontal direction from an obstacle; Minimum vertical safe distance: The minimum distance that a drone must maintain in the vertical direction from an obstacle; Safety buffer radius: The radius of the safe zone centered on the obstacle. Drones entering this zone must follow special strategies.

[0035] B. Flight Restriction Rules: Define the flight restrictions for drones around obstacles, specifically including: No-fly sign: Indicates whether the obstacle constitutes a no-fly zone. If it is true, no drone may enter the zone. Height limit: The maximum permissible flight altitude of a drone around the obstacle; Speed ​​limit: The maximum permissible flight speed of the drone around the obstacle; Permitted flight direction: Defines the permitted direction for a drone when approaching or circling an obstacle, used to guide the drone to circumvent it from the safe side.

[0036] C. Conditional Rule Set: Defines dynamic behavior rules that are triggered under specific conditions. Each rule includes: Triggering conditions: Use logical expressions to describe the conditions that trigger the rule, such as drone type, weather conditions, mission type, time window, etc. Execution Action: Define the action the system should take when the conditions are met, including allowing passage, requiring detour, and prohibiting passage; Rule priority: When multiple rules are satisfied at the same time, the system selects the rule to execute according to its priority.

[0037] D. Mission-Specific Constraints: Define constraint rules related to specific flight missions, including: Mission type: Identifies the type of the current flight mission, such as logistics delivery, inspection, emergency rescue, etc. Drone capability requirements: Define the performance parameters of the drone required to perform the task, such as minimum turning radius, maximum climb rate, etc. Dynamic adjustment flag: Indicates whether the system is allowed to dynamically adjust constraint parameters based on actual conditions.

[0038] The behavior constraint module of this invention introduces a two-layer mechanism of condition rule set and task-specific constraint, enabling the system to have condition triggering capability and task adaptation capability, breaking through the limitation of static and fixed constraint rules in the prior art.

[0039] The modules described above are logically independent, uniformly encapsulated in a data structure, and collectively provide structured input to the UAV flight control system or path planning system to support path search, risk weight calculation, and flight behavior decision-making. Based on this system architecture, as follows... Figure 3 As shown, the embodiments of the present invention propose the following three core application methods: I. Multi-dimensional integrated obstacle risk assessment This method achieves a quantitative assessment of obstacle risk through the following steps: Step 1: Extract the height features and morphological complexity of obstacles from the geometric description module, and calculate the height risk factor and morphological risk factor: ; ; In the formula, The height of the obstacle. The maximum safe flight altitude is a system preset value, such as 120 meters; This is the current flight altitude of the drone. This is a high sensitivity coefficient, ranging from 0.5 to 2.0, adjusted according to the type of drone. The morphological complexity coefficient is preset according to the shape type. In this embodiment of the invention, it is 0.2 for cylinder / bounding box, 0.5 for polygonal prism, 0.8 for irregular polyhedron, and 1.0 for voxel set. Let be the projected area of ​​the obstacle on the horizontal plane. This is the baseline area, which is also a system preset value, such as 100 square meters. , Let be the weighting coefficient, satisfying .

[0040] Step 2: Extract obstacle type weights, criticality weights, and population density weights from the semantic level module, and calculate the semantic risk factor: ; In the formula, The weights for obstacle types are preset according to the type, such as 0.3 for ordinary buildings, 0.9 for high-voltage transmission lines, and 0.4 for temporary facilities; As a critical level weight, Weighted by population or asset intensity. , , Let be the weight coefficients of each factor, satisfying... .

[0041] Step 3: Obtain positioning uncertainty parameters and environmental parameters from the positioning system and environmental sensors, and calculate the uncertainty risk factor and environmental risk factor: ; ; In the formula, To determine the location uncertainty, For uncertainty threshold, This is the time interval since the last location update. The validity period of location information; The current wind speed, For safe wind speed thresholds, Given the current rainfall intensity, To ensure a safe rainfall threshold, Given the current electromagnetic interference intensity, To ensure a safe electromagnetic interference threshold, , , Let be the weight coefficients of each factor, satisfying a sum of 1.

[0042] Step 4: Using a weighted fusion model, the above risk factors are fused and calculated to obtain a comprehensive risk score: ; In the formula, , , , , Let be the weight coefficients of each risk factor, satisfying... ; Furthermore, the system in this embodiment of the invention also supports dynamically adjusting the weighting coefficients based on the task type and the capabilities of the UAV: ; In the formula, The base weight of the i-th risk factor. The task adaptation factor for the i-th risk factor is dynamically preset based on the task type and the drone's capabilities, and the same applies to j; n is the total number of risk factors.

[0043] Step 5: Map the risk score to a risk level based on the preset risk scoring threshold, as shown in Table 1: Table 1 Risk Level Mapping Relationship Table

[0044] Step 6: Update the risk score, risk level, confidence level, validity period, and last update time in the risk assessment module.

[0045] II. Evaluation and Dynamic Adjustment of Condition-Triggered Behavioral Constraints This method achieves the verification and adaptive adjustment of flight behavior constraints through the following steps: Step 1: Read static rules such as safety distance constraints and flight restriction rules from the behavior constraint module; that is, the system first reads predefined static rules from the behavior constraint module, including parameters such as minimum horizontal safety distance, minimum vertical safety distance, safety buffer radius, no-fly sign, altitude limit value, and speed limit value, which constitute the basic threshold of behavior constraints.

[0046] Step 2: Perform geometric constraint verification on the candidate flight path to determine whether it meets the minimum safe distance and altitude requirements; The system calculates the shortest distance from each waypoint in the candidate path to the geometric surface of the obstacle, and takes the minimum value as the minimum distance between the path and the obstacle. This minimum distance is compared with the minimum safe horizontal distance read in step 1: if the minimum distance is greater than or equal to the safe distance threshold, the distance constraint is satisfied, and subsequent verification continues; if the minimum distance is less than the safe distance threshold, a constraint violation state is triggered, and the system performs detour adjustment. Detour adjustment refers to inserting new waypoints before and after the violation point to ensure that the adjusted path always maintains a safe distance from the obstacle. The detour direction is determined based on the relative orientation of the UAV's current position and the target point, prioritizing the side with the shorter detour distance.

[0047] Meanwhile, the system checks whether the maximum flight altitude of the candidate path meets the altitude constraint requirements. The minimum safe flight altitude required is equal to the height of the obstacle plus the minimum vertical safe distance. If the maximum flight altitude of the path is lower than this requirement, an altitude violation state is triggered, and the system raises all waypoints in the path that are below the safe altitude to the safe altitude while maintaining the horizontal track unchanged.

[0048] Step 3: Traverse the set of conditional rules and determine whether the rule triggering conditions are met based on the current state of the drone, flight mission parameters, and environmental conditions; The system iterates through the set of conditional rules in the behavior constraint module. Each rule contains three elements: triggering condition, execution action (allow passage, require detour, prohibit passage), and rule priority.

[0049] The system determines whether the triggering conditions of each rule are met based on the current drone status (including drone type, current location, speed, remaining battery power, etc.), flight mission parameters (mission type, priority, etc.), and environmental conditions (wind speed, rainfall intensity, electromagnetic interference, time of day, etc.). The triggering conditions are described using logical expressions, such as the drone being a light logistics model with a wind speed of less than 10 m / s, or performing an emergency rescue mission during daytime.

[0050] Step 4: If the rule triggering conditions are met, execute the corresponding actions, including allowing passage, requiring detours, or prohibiting passage. From all rules whose trigger conditions are true, sort them in descending order of priority, and execute the action corresponding to the rule with the highest priority: the priority can be dynamically set according to actual needs; If the action is to allow passage, the system keeps the current candidate path unchanged and sets the constraint state to satisfied; If the action requires a detour, the system triggers a constraint violation state and performs a detour adjustment. The detour adjustment can carry specific detour parameters, such as the detour radius or the detour lateral direction, and the system generates an adjusted path based on these parameters; If the action is to prohibit crossing, the system determines that the current candidate path is not feasible, sets the constraint status to prohibited, and terminates the subsequent verification process.

[0051] Step 5: If the task-specific constraints allow for dynamic adjustment, then adaptively optimize the constraint parameters based on the actual capabilities of the UAV; The system checks the dynamic adjustment flag in the task-specific constraints. If the flag is true, the system adaptively optimizes the constraint parameters based on the actual capabilities of the UAV, specifically including: Adjustment of safe distance: The basic safe distance is scaled according to the drone's maneuverability coefficient (such as maximum acceleration, minimum turning radius, etc.). The stronger the drone's capabilities, the safe distance can be appropriately reduced; the weaker the capabilities, the safe distance should be increased accordingly.

[0052] Flight speed adjustment: Optimize flight speed according to mission type. For example, the speed coefficient can be appropriately increased for emergency rescue missions, while the speed coefficient can be decreased for inspection missions.

[0053] Detour radius adjustment: The detour radius is adapted to the minimum turning radius of the drone to ensure that the adjusted path is within the drone's maneuverability range.

[0054] Step 6: Output the constraint-satisfied state and the adjusted flight path; The system outputs the final constraint status (satisfied, violated, or prohibited) and the adjusted flight path. If the constraint status is satisfied or violated, the adjusted path can be used as a valid path for subsequent use; if the constraint status is prohibited, the current candidate path is not feasible and needs to be replanned.

[0055] III. Joint path planning based on semantics, risk, and constraints, such as... Figure 4 As shown This method achieves deep fusion of path planning and obstacle data through the following steps: Step 1: Load the obstacle set containing the unified information model; The system loads information model objects of all obstacles from the low-altitude airspace database. Each obstacle information model contains a geometric description module, a semantic level module, a risk assessment module, and a behavioral constraint module.

[0056] Step 2: Perform a multi-dimensional risk assessment on each obstacle to obtain a risk score and risk level; Step 3: Based on the risk assessment results and the no-fly signs in the behavior constraint module, map the obstacles into different area types, including no-fly zones, high-cost zones, and restricted detour zones; The system maps each obstacle to a specific area type based on its risk level and the no-fly sign in the behavior constraint module, and assigns a cost multiplier to each type. : If the risk level is extremely high, or the no-fly sign in the behavior constraint module is true, then the obstacle will be mapped as a no-fly zone. Setting it to infinity indicates that it is impossible to traverse; If the risk level is high, it is mapped to a high-cost area. Setting it to 10 indicates a strong recommendation to avoid it; If the risk level is medium risk, it is mapped to a medium cost region. Setting it to 5 indicates a suggestion to avoid it; If the risk level is low, it is mapped to a low-cost area. Setting it to 2 indicates that it is possible to traverse, but caution is advised. Unobstructed spaces are free spaces. Set to 1 to indicate normal passage.

[0057] Step 4: Construct a constrained spatial cost model, where the cost function comprehensively considers factors such as obstacle risk score, safety distance requirements, and UAV flight efficiency; Define three-dimensional planning space ,starting point ,end There is a set of obstacles. Each obstacle It includes information such as geometric description, risk score, and behavioral constraints. (Path definition follows.) ,in Starting from, As the endpoint, Let i be the i-th node on the path.

[0058] From node Move to node The cost of moving is: ; In the formula, For Euclidean distance, For nodes The cost field value at that location.

[0059] Single obstacle cost field: For any point Its obstacles The cost of the impact is: ; In the formula, For obstacles No-fly zone For point arrive The shortest distance on the geometric surface, for The radius of the safety buffer zone, The attenuation coefficient; Total Cost Field: .

[0060] Step 5: Perform path search and generate candidate flight paths while satisfying the behavioral constraints; The system uses an improved A* algorithm to perform path search in three-dimensional space. Each node in the search space represents a spatial location, and the heuristic function uses the three-dimensional Euclidean distance from the current node to the destination.

[0061] During node expansion, the system performs a feasibility check on each neighboring node: it checks whether the node meets all obstacle behavior constraints, including whether it has entered a no-fly zone, whether it meets the minimum safe distance requirement, and whether it meets the height constraint. If it does not meet these constraints, the node is pruned and not added to the search queue.

[0062] The system generates neighbor nodes in three-dimensional space, and the neighbor step size is dynamically adjusted according to the UAV's maneuverability and safety requirements: UAVs with high maneuverability can use a larger step size to improve search efficiency, while UAVs with low maneuverability use a smaller step size to ensure path accuracy.

[0063] During the search process, the system prioritizes expanding the node with the lowest total cost until the destination is reached or the search queue is empty. After the search is completed, the system backtracks from the destination to obtain the original path.

[0064] Step 6: Perform constraint verification and cost evaluation on the candidate paths, and output the optimal or suboptimal flight path that meets the safety requirements; The system performs post-processing optimization on the original path generated in step 5: First, it performs path smoothing by using cubic spline interpolation or B-spline curves to smooth the original path, eliminating polylines and redundant waypoints, making the path more consistent with the actual flight characteristics of the UAV; then, it removes redundant points by traversing the path nodes and deleting the intermediate nodes if three consecutive nodes are collinear or have an angle close to 180 degrees, thereby reducing the path length.

[0065] The system performs a complete constraint check on the smoothed path again, calling the behavioral constraint evaluation method in Method 2 to confirm that the path meets all safety requirements; at the same time, the system calculates the total cost of the path, including path length cost, obstacle crossing cost, height change cost, etc.

[0066] Ultimately, the system outputs the optimal flight path that satisfies all constraints and minimizes the total cost. If multiple feasible paths are found during the search process, the system can sort them by cost from smallest to largest and output the optimal path and several suboptimal paths for the operator to choose from. If the search queue is empty and no feasible path is found, the system returns a failure status and triggers a replanning or manual intervention process.

[0067] Example 1: Urban Low-Altitude Fixed Building Obstacle Management System In this embodiment, the system of the present invention is applied to urban low-altitude drone inspection and logistics flight scenarios to manage fixed obstacles such as urban buildings.

[0068] 1. System Configuration For a single building obstacle, the system constructs the following modular information: Geometric Description Module: The obstacle ID is OB-001, the geometric type is a polygon bounding volume, the spatial position is based on a set of three-dimensional coordinates of latitude, longitude and height, and the obstacle height is 60 meters.

[0069] Semantic rating module: The obstacle category is urban buildings, the criticality level is important, the population density level is medium, and the semantic weight coefficient is 0.7.

[0070] Risk assessment module: After multi-dimensional integrated assessment, the risk factor is 0.5, the morphological risk factor is 0.3, the semantic risk factor is 0.7, the comprehensive risk score is 0.7, the risk level is medium to high risk, the risk confidence level is 0.85, and the validity period is long-term.

[0071] Behavioral constraint module: minimum horizontal safe distance 20 meters, minimum vertical safe distance 10 meters, flight height constraint prohibits flying below 10 meters from the top of buildings, and the condition rules define a dynamic adjustment rule that if the drone is a light logistics drone, the safe distance can be reduced to 15 meters.

[0072] 2. System operation process Before the drone executes its mission, the system loads a set of obstacle objects containing information from the aforementioned modules from the low-altitude airspace database. During path planning, the system first performs spatial accessibility assessment based on the geometric description module, then determines the risk weight of the obstacle by combining the semantic level module and the risk assessment module, and finally filters and corrects candidate flight paths based on the behavioral constraint module. When the drone type is a light logistics drone, the system's conditional rules are automatically triggered, dynamically adjusting the safety distance from 20 meters to 15 meters to achieve mission adaptation.

[0073] Example 2: Path planning with fusion of multiple obstacle types In another embodiment, the system of the present invention can be applied to a unified management scenario of multiple fixed obstacles such as buildings, communication towers and power transmission lines.

[0074] 1. System Configuration For communication tower obstacles, the criticality level in the semantic level module of the system is set to extremely high risk, with a semantic weight coefficient of 0.95; the comprehensive risk score in the risk assessment module is 0.88, and the risk level is extremely high risk; in the behavior constraint module, the minimum safe horizontal distance is set to 50 meters, and the no-fly sign is set to true.

[0075] For power transmission line obstacles, the system's geometric description module adopts a linear voxel set form, the semantic level module sets the criticality level to critical, and the semantic weight coefficient is 0.85; the risk assessment module has a comprehensive risk score of 0.82, and the risk level is high risk; the minimum safe vertical distance in the behavioral constraint module is set to 15 meters, and the condition rule is set: if the wind speed is greater than 10m / s, crossing is prohibited.

[0076] 2. System operation process Although different types of obstacles are represented using different forms in the geometric description module, they all use a unified field structure in the semantic level module, risk assessment module, and behavior constraint module. This allows the system to perform risk assessments and make behavioral decisions for different obstacles in a consistent manner. When the UAV plans its path, the system dynamically adjusts its flight altitude, detour radius, and flight speed based on the semantic level and risk score of different obstacles, achieving a balance between safety and efficiency.

[0077] Preferably, the semantic level module and risk assessment module of this embodiment can be merged into a unified risk semantic module; the constraint parameters in the behavior constraint module can be expressed in the form of rule tables, interval functions or probability models; the system architecture can be tailored according to specific application scenarios, such as simplifying the semantic level module in scenarios that do not involve semantic distinction.

[0078] In summary, this invention proposes an obstacle management system that combines obstacle semantic levels, risk assessment, and behavioral constraints. By describing the geometric attributes, semantic risk attributes, and flight behavior constraint rules of obstacles in a unified data structure, the UAV flight control system can directly perform path planning and obstacle avoidance decisions based on this data structure, thereby improving flight safety, decision-making efficiency, and system scalability.

[0079] Method Implementation Examples According to embodiments of the present invention, a method for low-altitude airspace obstacle management based on multi-dimensional information fusion is provided. Figure 5 This is a flowchart of a low-altitude airspace obstacle management method based on multi-dimensional information fusion, according to an embodiment of the present invention. Figure 5 As shown, the low-altitude airspace obstacle management method based on multi-dimensional information fusion according to an embodiment of the present invention specifically includes: Step S501: Collect and store basic spatial information of each obstacle in the low-altitude airspace through the geometric description module, and transmit the basic spatial information to the semantic level module, risk assessment module and behavior constraint module; Step S502: The semantic level module generates semantic weight coefficients for each obstacle based on the basic spatial information, and transmits the semantic weight coefficients to the risk assessment module. Step S503: The risk assessment module calculates the comprehensive risk score of each obstacle based on the basic spatial information and the semantic weight coefficient, determines the risk level of each obstacle based on the comprehensive risk score, and sends the risk level to the behavior constraint module. Step S504: The behavior constraint module generates and outputs the flight behavior constraint conditions of the aircraft when approaching or bypassing various obstacles based on the basic spatial information and the risk level. Step S505: The path planning module maps each obstacle to its corresponding cost region based on the basic spatial information, risk level, and flight behavior constraints of each obstacle, constructs a spatial cost model, and performs path search based on the spatial cost model and the flight behavior constraints to output the flight path that meets the safety requirements and has the minimum total cost value.

[0080] The embodiments of the present invention are method embodiments corresponding to the system embodiments described above. The specific operations of each step can be understood by referring to the description of the system embodiments, and will not be repeated here.

[0081] In summary, compared with the prior art, the beneficial effects of the embodiments of the present invention include: 1. Unify the representation of the geometry, semantics, risk, and constraint information of obstacles at the data level to reduce system complexity; 2. Support differentiated flight decisions to improve the safety of UAVs flying at low altitudes; 3. It facilitates the expansion and collaboration of multi-task, multi-UAV systems; 4. It is easy to reuse in different UAV platforms and airspace management systems.

[0082] Device Example 1 This invention provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it performs the steps described in the method embodiment.

[0083] Device Example 2 This invention provides a computer-readable storage medium storing an information transmission implementation program, which, when executed by a processor, performs the steps described in the method embodiment.

[0084] The computer-readable storage media described in this embodiment include, but are not limited to, ROM, RAM, disk, or optical disk.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A low-altitude airspace obstacle management system based on multi-dimensional information fusion, characterized in that, It includes a geometric description module, a semantic level module, a risk assessment module, and a behavioral constraint module that are interconnected with each other; The geometric description module is used to collect and store basic spatial information of various obstacles in the low-altitude airspace, and transmit the basic spatial information to the semantic level module, risk assessment module and behavior constraint module. The semantic level module is used to generate semantic weight coefficients for each obstacle based on the basic spatial information, and transmit the semantic weight coefficients to the risk assessment module. The risk assessment module is used to calculate the comprehensive risk score of each obstacle based on the basic spatial information and the semantic weight coefficient, determine the risk level of each obstacle based on the comprehensive risk score, and send the risk level to the behavior constraint module. The behavior constraint module is used to generate and output flight behavior constraints for the aircraft when approaching or circling various obstacles based on the basic spatial information and the risk level.

2. The system according to claim 1, characterized in that, The basic spatial information includes the obstacle's unique identifier ID, geographical coordinates, spatial morphology description type, geometric data set, and actual height or height range.

3. The system according to claim 1, characterized in that, The semantic level module is specifically used for: Based on the basic spatial information, obstacle type labels, criticality levels, and population or asset density levels are generated for each obstacle, and semantic weight coefficients for each obstacle are generated according to at least one of the obstacle type labels, criticality levels, and population or asset density levels. The obstacle type labels include buildings, communication towers, power lines, bridges, temporary facilities, or trees; The criticality levels include ordinary, important, critical, and extremely high risk; The population or asset density levels include low density, medium density, high density, and very high density.

4. The system according to claim 1, characterized in that, The risk assessment module is specifically used for: The altitude risk factor is calculated using Formula 1 based on the ratio of obstacle height to aircraft flight altitude and the altitude difference. The morphological risk factor is calculated using Formula 2 based on the morphological complexity of the obstacle's geometry. The semantic risk factor is calculated using Formula 3 based on the semantic weight coefficients. Based on the uncertainty and validity period of obstacle location information, the positioning uncertainty risk factor is calculated using Formula 4. The environmental risk factor is calculated using Formula 5 based on the environmental parameters of the environment in which the obstacle is located. Official 1; Official 2; Official 3; Official 4; Official 5; in, Indicates the height of the obstacle. Indicates the maximum safe flight altitude. Indicates the current flight altitude of the aircraft. Indicates a high sensitivity coefficient. Represents the morphological complexity coefficient. This represents the projected area of ​​the obstacle on the horizontal plane. Indicates the reference area. , Indicates the weighting coefficient. Indicates the obstacle type weight. Indicates the key level weight. Indicates the weight of population or asset intensity. , , These represent the weight coefficients of each factor in semantic risk. Indicates the uncertainty of positioning. Indicates the uncertainty threshold. This indicates the time interval since the last position update. Indicates the validity period of location information. Indicates the current wind speed. Indicates the safe wind speed threshold. Indicates the current rainfall intensity. Indicates the safe rainfall threshold. Indicates the current electromagnetic interference intensity. Indicates the safe electromagnetic interference threshold. , , These represent the weighting coefficients of each factor in environmental risk.

5. The system according to claim 4, characterized in that, The risk assessment module is specifically used for: The weighted fusion model shown in Formula 6 is used to fuse and calculate the high risk factor, morphological risk factor, semantic risk factor, positioning uncertainty risk factor and environmental risk factor to obtain a comprehensive risk score; Official 6; in, This represents the overall risk score. , , , , These represent the weighting coefficients of each risk factor; The comprehensive risk score is mapped to a risk level based on a preset risk scoring threshold; wherein the risk level includes low risk, medium risk, high risk and extremely high risk.

6. The system according to claim 1, characterized in that, The behavior constraint module is specifically used for: Define the minimum safe distance that must be maintained between the aircraft and the obstacle; wherein, the minimum safe distance includes the minimum horizontal safe distance, the minimum vertical safe distance, and the radius of the safety buffer zone; Define flight restriction rules for aircraft around obstacles; wherein, the flight restriction rules include no-fly signs, altitude limits, speed limits and permitted flight directions; Define a set of conditional rules that are triggered under specific conditions; each conditional rule includes the triggering condition, the action to be performed, and the rule priority. Define mission adaptation constraints related to specific flight missions of the aircraft; wherein, the mission adaptation constraints include mission type, aircraft capability requirements, and dynamic adjustment flags; Based on the basic spatial information and the risk level, and in combination with at least one of the minimum safe distance, flight restriction rules, condition rule set and mission adaptation constraints, flight behavior constraints are generated and output.

7. The system according to claim 1, characterized in that, The system also includes: The path planning module, connected to the geometry description module, risk assessment module, and behavior constraint module, is used to obtain the basic spatial information, risk level, no-fly sign, safety buffer radius, and flight behavior constraint conditions of each obstacle from the geometry description module, risk assessment module, and behavior constraint module, respectively. Based on the risk level and the no-fly sign, it maps each obstacle to the corresponding cost region and assigns a corresponding cost multiplier to each cost region. A spatial cost model is constructed based on the cost multiplier, the safety buffer radius, and the distance from the current position of the aircraft to the geometric surfaces of each obstacle; wherein, the spatial cost model is used to calculate the basic cost of any point in three-dimensional space affected by obstacles; An improved A* algorithm is used to perform path search. During the search process, the basic cost is used as the cost of node expansion, and the flight behavior constraints are used as the basis for verifying the feasibility of nodes to generate candidate flight paths. The candidate flight paths are subjected to constraint verification and cost evaluation, and the flight path that meets the safety requirements and has the lowest total value is output.

8. A method for managing obstacles in low-altitude airspace based on multi-dimensional information fusion, characterized in that, include: The basic spatial information of each obstacle in the low-altitude airspace is collected and stored by the geometric description module, and the basic spatial information is transmitted to the semantic level module, risk assessment module and behavior constraint module. The semantic weight coefficients of each obstacle are generated by the semantic level module based on the basic spatial information, and the semantic weight coefficients are transmitted to the risk assessment module. The risk assessment module calculates a comprehensive risk score for each obstacle based on the basic spatial information and the semantic weight coefficient, determines the risk level of each obstacle based on the comprehensive risk score, and sends the risk level to the behavior constraint module. The behavior constraint module generates and outputs flight behavior constraints for the aircraft when approaching or circling various obstacles based on the basic spatial information and the risk level. The path planning module maps each obstacle to its corresponding cost region based on its basic spatial information, risk level, and flight behavior constraints, constructs a spatial cost model, and performs path search based on the spatial cost model and flight behavior constraints to output the flight path that meets safety requirements and has the lowest total cost value.

9. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the low-altitude airspace obstacle management method based on multi-dimensional information fusion as described in claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an information transmission implementation program, which, when executed by a processor, implements the steps of the low-altitude airspace obstacle management method based on multi-dimensional information fusion as described in claim 8.

Images