Unmanned aerial vehicle autonomous route planning and collision avoidance decision method and system

By constructing an evaluation function based on the angle change of the drone and obstacle density, and combining the A* algorithm and QP solver to carry out the drone route planning, the operational risks and navigation performance degradation caused by the irrationality of the navigation route in the existing technology are solved. , achieving safe and economical route planning and maneuver obstacle avoidance decisions.

CN118915804APending Publication Date: 2024-11-08CIVIL AVIATION UNIV OF CHINA
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411076796.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The existing drone route planning methods fail to effectively consider the operating risks caused by the irrational route. Especially in complex low-altitude airspace environments, it is easy to cause problems such as degradation in navigation performance and insufficient computing resources, and the route planning is relatively to the maneuvering obstacle avoidance function. Independence makes it difficult to achieve safe and economical route planning.

Method used

By reading the terrain data of the airspace of the UAV, a global terrain database is established, a three-dimensional route planning is carried out, and an evaluation function based on angle changes, obstacle density and navigation performance is constructed. The A* algorithm and QP solver are combined to perform global and local route planning. Ensure the safety and economics of the route and make maneuvering obstacle avoidance decisions when dynamic obstacles appear.

Benefits of technology

Effectively avoid the problems of degraded navigation performance and insufficient computing resources in route planning, ensure the safety and flightability of the UAV route, and realize the organic integration of route planning and maneuvering obstacle avoidance, reducing the computing resources and Communication link requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118915804A_ABST
    Figure CN118915804A_ABST
Patent Text Reader

Abstract

The invention provides an unmanned aerial vehicle autonomous route planning and collision avoidance decision-making method, which comprises the steps of reading in a static airspace environment of an unmanned aerial vehicle and maneuvering turning performance data of the unmanned aerial vehicle, and establishing an airspace and aircraft performance constraint information base, constructing an air route planning economy and safety evaluation function based on airspace obstacle density and unmanned aerial vehicle navigation performance, and performing global strategic planning according to unmanned aerial vehicle task requirements and information constraints; according to the real-time position of an unmanned aerial vehicle and the sensing range of an airborne situation sensing sensor, the local route planning range of the unmanned aerial vehicle is delimited, the starting point and the ending point of local route planning and other route points, determined in global planning, in the area are determined, and fine-grained fusion is carried out on airspace static information and local dynamic information. Adjusting and modifying a local airspace obstacle information base, establishing dynamics and kinematics equations of the unmanned aerial vehicle, performing time division, performing local route planning according to the kinematics equations of the unmanned aerial vehicle, and ensuring the maneuverability of the planned route.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of dynamic route planning for unmanned aerial vehicles (UAVs), and in particular relates to a method and system for autonomous route planning and collision avoidance decision-making for UAVs. Background Art

[0002] With the increasing use of drones in aviation logistics, low-altitude tourism, emergency rescue, agriculture and fishery operations, low-altitude airspace traffic is rapidly increasing. At the same time, the high concentration of birds, airborne objects, and other low-altitude aircraft in low-altitude airspace can easily compromise the operational safety of drones. Furthermore, due to obstructions from terrain or buildings, complex low-altitude operating environments can easily lead to data link interruptions between drones and ground stations, increasing operational risks. Highly secure autonomous route planning and obstacle avoidance capabilities for airborne drones are crucial for navigating these complex low-altitude airspace environments.

[0003] Most existing route planning methods use graph search methods or intelligent bionics methods to perform economic-oriented two-dimensional static route planning, and basically do not consider the UAV operation risks caused by the irrationality of the planned route itself.

[0004] If the planned route is unreasonable, for example, if part of the route is in a highly complex airspace environment with dense terrain or buildings around, it will block the radio wave signal of the satellite navigation system and cause multipath interference, resulting in reduced performance of the drone's onboard navigation system or even failure to work. If the planned route is too tortuous, it will make it difficult or even impossible for the drone to fly. In addition, the existing drone route planning function is usually relatively independent of the maneuvering and obstacle avoidance function. The high frequency of dynamic obstacle encounters and collision avoidance maneuvers in low-altitude airspace will cause the drone to frequently restart route planning, consuming a large amount of computing resources of the onboard equipment, resulting in insufficient computing resources.

[0005] In response to the above problems, the present invention provides a method for autonomous route planning and collision avoidance decision-making for UAVs, which effectively solves many problems caused by unreasonable route planning, provides a safe, economical and feasible route planning method for UAVs, and realizes the organic integration of UAV route planning and maneuverable collision avoidance. Summary of the Invention

[0006] In view of this, the present invention aims to overcome the deficiencies of the above-mentioned problems in the prior art and proposes a method and system for autonomous route planning and collision avoidance decision-making of unmanned aerial vehicles.

[0007] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0008] A first aspect of the present invention provides a method for autonomous route planning and collision avoidance decision-making for a UAV, comprising the following steps:

[0009] Step 1: Read the terrain raster data information of the airspace where the UAV is located, establish a global terrain database, and determine the angle change limit of the UAV based on the UAV's maneuverability characteristics;

[0010] Step 2: Perform a 3D route planning node search based on the UAV mission information, calculate the horizontal and vertical angle changes required for the UAV to reach the neighboring node, and search for nodes that meet the UAV's angle change performance constraints;

[0011] Step 3: Calculate the diagonal distance between nodes, the density function of obstacles around the route segment, and the navigation performance parameters of the onboard navigation system, and construct an evaluation function for global route optimization through weighted average calculation;

[0012] Step 4: Use the global terrain database and the evaluation function calculated in step 3 to perform global route planning and search for a global route that meets economic and safety requirements;

[0013] Step 5: Read the real-time location information of the UAV, define the UAV situational awareness range, determine the starting and ending points of the UAV local route planning, and select the waypoint information in the route planned in step 4 as the key nodes of the local dynamic route planning;

[0014] Step 6: Establish the UAV motion equation and use the QP solver to calculate the optimization parameters of the UAV's local airspace trajectory based on the optimization goal of the UAV's minimum jerk operation;

[0015] Step 7: Combined with the surrounding dynamic obstacle information provided by the UAV perception system, the constraint inequality of the QP solver is established according to the obstacle avoidance principle, and local dynamic path planning based on the obstacle avoidance strategy is performed to determine the local area motion trajectory;

[0016] Step 8: Calculate the speed and acceleration range and maximum tilt angle of the UAV in the local route planning, and determine the feasibility of the planned route. If the route is feasible, output the UAV local trajectory polynomial. If not, adjust the time allocation and re-plan until a feasible route is determined.

[0017] Step 9: Calculate the lateral deviation of the local route planning endpoint relative to the global planning route. If the deviation is greater than the specified limit, use this point as the starting point of the new UAV global route planning, and start the global route optimization fine-tuning from the current position to the destination;

[0018] Step 10: Read the current position of the UAV, perform rolling time domain planning calculations, update the start and end points of the local route planning, and perform local route planning.

[0019] Furthermore, in step 3, the evaluation function is as follows:

[0020] h(n)=w1×h1(n)+w2×h2(n)+w3×h3(n);

[0021] where h1(n) = dis diagonal (n, target) is the diagonal distance between nodes; h2(n) = cost obstacle_density is the obstacle density cost; h3(n)=cost nav_uncertainty is the navigation uncertainty cost at the node, w1, w2, and w3 are weighting coefficients, and the weighting coefficients can be appropriately adjusted according to the different levels of safety and economy requirements.

[0022] Furthermore, in step 4, global route planning is performed using the terrain database and the evaluation function calculated in step 3, and a global route that meets the economic and safety requirements is searched and stored using the A* algorithm.

[0023] Furthermore, in step 7, according to the obstacle avoidance principle, the obstacle avoidance requirement is modeled as an inequality constraint of the QP problem, which stipulates that the relationship between the planned trajectory and the dynamic obstacle must satisfy the following formula:

[0024] (x own (t)-x obs (t)) 2 +(y own (t)-y obs (t)) 2 +(z own (t)-z obs (t)) 2 ≥radius own +radius obs ;

[0025] Where (x own ,y own ,z own ) represents the position of the drone, (x obs ,y obs ,z obs ) represents the position of the dynamic obstacle, radius own and radius obs Represent the real-time equivalent radius of the drone and obstacle respectively.

[0026] A second aspect of the present invention provides a system for autonomous route planning and collision avoidance decision-making for a UAV, comprising:

[0027] The scenario simulation unit is used to construct the static airspace environment in which the UAV operates and the UAV's maneuverability performance parameters, and to generate dynamic obstacles output by the situational awareness sensor;

[0028] A data reading and processing unit is used to read and process the data output by the scenario simulation unit to form route planning constraint information for calculating the route planning evaluation function;

[0029] The global route optimization unit is used to generate a global planning route that meets economic and safety requirements based on the UAV flight mission requirements and operation scenario constraints;

[0030] The local dynamic route planning unit is used to integrate the static environment and dynamic obstacle information of the local airspace, perform fine-grained normalization calculations, form local airspace constraint information, calculate the proximity relationship between the UAV itself and dynamic obstacles, and perform local route planning based on the UAV's motion characteristics to ensure the route's feasibility. At the same time, it determines the route deviation that may be caused by the local route planning collision avoidance maneuver and timely initiates fine-tuning of the subsequent segments of the global planned route based on the degree of deviation;

[0031] The rolling time domain planning unit is used to rollingly adjust the input information of the local dynamic route planning according to the position information of the UAV, so as to achieve a smooth transition and inheritance between the global route and the local route planning.

[0032] Compared with the existing technology, the method and system for autonomous route planning and collision avoidance decision-making for unmanned aerial vehicles described in the present invention have the following advantages:

[0033] In the present invention, by searching and calculating the diagonal distance between nodes, a route planning safety evaluation function based on the airspace obstacle density and the navigation performance of the UAV is constructed, which effectively avoids the occurrence of narrow tunnel sections with dense obstacles in the planned route, solves the problems of navigation performance degradation and communication interruption caused by radio wave signal obstruction and multipath that the UAV may face in a highly complex environment, and ensures the safe operation of the UAV.

[0034] The present invention establishes the dynamic and kinematic equations of the UAV, performs local route planning of the UAV through minimum jerk optimization constraint calculation, and solves the problem of UAV flight difficulties that may exist in route planning.

[0035] The present invention organically integrates the local route planning of the UAV with dynamic obstacle avoidance, effectively avoiding the mutual influence and restraint between the two functions.

[0036] The present invention effectively connects the global planning and local route planning of the UAV through rolling time domain planning, performs fine-tuning of the global route planning and allocation and management of computing resources, and ensures the smooth implementation of the UAV's autonomous route planning and obstacle avoidance decision-making method.

[0037] The present invention will play an important role in the low-altitude autonomous flight of UAVs and has important market value. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which constitute part of the present invention, are provided to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are provided to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0039] Figure 1 This is the overall framework diagram of the autonomous route planning and obstacle avoidance decision-making of the UAV provided by the present invention. Figure 2 This is a schematic diagram of the composition of the autonomous route planning and obstacle avoidance decision-making system for unmanned aerial vehicles provided by the present invention. DETAILED DESCRIPTION

[0040] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0041] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0042] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0043] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0044] The embodiment of the present invention discloses a method for autonomous route planning and collision avoidance decision-making of a UAV, referring to Figure 1 As shown, the method mainly includes the following steps:

[0045] Step 1: Read the terrain grid data information of the airspace where the drone is located, establish the OPENList and CLOSE List of the global terrain data; and determine the angle change limit of the drone.

[0046] Step 2: Search for three-dimensional neighbor nodes based on the drone's planned location information. The horizontal and vertical angle changes required for the drone to reach the neighbor nodes are calculated. Neighbor nodes that meet the drone's angle change performance constraints are added to the open list. This method calculates the horizontal and vertical angle changes required for the drone to reach the neighbor nodes in three-dimensional space and uses the drone's angle change constraints to improve search speed and reduce redundant nodes.

[0047] Step 3: Construct an evaluation function for global route optimization. Calculate the diagonal distance between nodes, the density function of obstacles around the route segment, and the navigation performance parameters of the onboard navigation system. Calculate the route optimization evaluation function by weighted average. The evaluation function is shown in the following formula:

[0048] h(n)=w1×h1(n)+w2×h2(n)+w3×h3(n);

[0049] where h1(n) = dis diagonal (n, target) is the diagonal distance between nodes; h2(n) = cost obstacle_density is the obstacle density cost; h3(n)=cost nav_uncertainty is the navigation uncertainty cost at the node, w1, w2, and w3 are weighting coefficients, and the weighting coefficients can be appropriately adjusted according to the different levels of safety and economy requirements.

[0050] In order to improve the safety and feasibility of the planned route, the present invention introduces the density function of obstacles around the route segment and the navigation performance parameters of the airborne navigation system into the evaluation function of the route optimization through weighted averaging.

[0051] Step 4: Use the terrain database and the evaluation function calculated in step 3 to perform global route planning, and use the A* algorithm to search for and store a global route that meets the economic and safety requirements.

[0052] In step 5, the real-time location information of the UAV is read in, the situational awareness range of the UAV is defined, the starting and ending points of the local route planning of the UAV are determined, and the waypoint information in the route planned in step 4 is selected as the key nodes of the local dynamic route planning.

[0053] In step 6, equality constraints are established between key nodes of the local path through time allocation. Based on the UAV's equations of motion, the minimum jerk of the UAV is optimized. A QP solver is used to calculate the optimal parameters for the UAV's local path trajectory, and local dynamic path planning is performed. If there are fewer than two local waypoints, interpolation is used to resolve the problem.

[0054] Step 7: Based on the obstacle avoidance principle, the obstacle avoidance requirement is modeled as an inequality constraint of the QP problem, which stipulates that the relationship between the planned trajectory and the dynamic obstacle must satisfy the following equation:

[0055] (x own (t)-x obs (t)) 2 +(y own (t)-y obs (t)) 2 +(z own (t)-z obs (t)) 2 ≥radius own +radius obs ;

[0056] Where (x own ,y own ,z own ) represents the position of the drone, and (x obs ,y obs ,z obs ) represents the position of the dynamic obstacle, radius own and radius obs Represent the real-time equivalent radius of the drone and obstacle respectively.

[0057] According to the surrounding dynamic obstacle information provided by the drone's perception system, local waypoints are added or adjusted to ensure that the drone avoids obstacles and satisfies the kinematic equations. The smooth obstacle avoidance trajectory connecting the waypoints is calculated to determine the local route planning that satisfies the drone's obstacle avoidance maneuvers.

[0058] Step 8: Calculate the speed and acceleration range and maximum tilt angle of the UAV in the local route planning, and determine the feasibility of the planned route. If it is not feasible, adjust the time allocation and re-plan until a local trajectory polynomial that is feasible is output.

[0059] Step 9: Calculate the lateral deviation of the local route planning endpoint relative to the global route planning. If the deviation is greater than the specified requirement, start the global route planning from the current position to the destination with the UAV as the starting point, and fine-tune the global route.

[0060] Step 10: Read the UAV's current position, perform a rolling horizon planning calculation, update the starting and ending points of the local route planning, and then perform a new local route planning. The rolling horizon planning method achieves a smooth transition and inheritance between the UAV's global route optimization and local dynamic route planning.

[0061] This invention effectively integrates the global route optimization of drones with local dynamic route planning through rolling time domain planning, ensuring the economy, safety, and feasibility of the planned route, and enabling effective autonomous collision avoidance decisions when encountering dynamic obstacles. This distributed rolling planning calculation method can effectively reduce the requirements for onboard computing resources. At the same time, the airborne autonomous route planning and collision avoidance decision-making method greatly reduces the drone's requirements for the real-time and continuity of air-to-ground communication links, and can be used for autonomous operation of drones in complex low-altitude airspace and mixed airspace, and is particularly suitable for operation requirements in complex airspace environments such as mountainous areas and over cities.

[0062] Another embodiment of the present invention provides a system for autonomous route planning and collision avoidance decision-making of a UAV, such as Figure 2 As shown, the system includes:

[0063] The scenario simulation unit is used to construct the static airspace environment in which the UAV operates and the UAV's maneuverability performance parameters, and to generate dynamic obstacles output by the situational awareness sensor;

[0064] A data reading and processing unit is used to read and process the data output by the scenario simulation unit to form route planning constraint information for calculating the route planning evaluation function;

[0065] The global route optimization unit is used to generate a global planning route that meets economic and safety requirements based on the UAV flight mission requirements and operation scenario constraints;

[0066] The local dynamic route planning unit is used to integrate the static environment and dynamic obstacle information of the local airspace, perform fine-grained normalization calculations, form local airspace constraint information, calculate the proximity relationship between the UAV itself and dynamic obstacles, and perform local route planning based on the UAV's motion characteristics to ensure the route's feasibility. It calculates and determines the route deviation that may be caused by the collision avoidance maneuver of the local route planning, and timely initiates fine-tuning of the subsequent segments of the global route based on the degree of deviation;

[0067] The rolling time domain planning unit is used to rollingly adjust the input information of the local dynamic route planning according to the position information of the UAV, connect the global route planning of the UAV with the local route planning, and allocate computing resources.

[0068] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for autonomous route planning and collision avoidance decision-making for unmanned aerial vehicles. Features: The steps include: Step 1: Read the terrain grid data information of the airspace where the UAV is located, establish a global terrain database, and determine the angle change limit of the UAV according to the UAV's maneuverability characteristics; Step 2: Search for nodes in the three-dimensional route planning based on the UAV mission information, calculate the horizontal and vertical angle changes required for the UAV to reach the neighboring nodes, and search for nodes that meet the UAV angle change performance constraints; Step 3: Calculate the diagonal distance between nodes, the density function of obstacles around the flight segment, and the navigation performance parameters of the onboard navigation system, and construct an evaluation function for global route optimization through weighted average calculation; Step 4: Use the global terrain database and the evaluation function calculated in step 3 to perform global route planning and search for a global route that meets the requirements of economy and safety; Step 5: Read the real-time location information of the drone, define the drone situation awareness range, determine the starting point and end point of the drone's local route planning, and select the waypoint information in the route planned in step 4 as the key node of the local dynamic route planning; Step 6: Establish the UAV motion equation, and according to the optimization goal of the UAV's minimum jerk operation, use the QP solver to calculate the optimization parameters of the UAV's local airspace trajectory to determine the local area motion trajectory; Step 7: Combined with the dynamic obstacle information perceived by the drone, a constraint inequality is established according to the obstacle avoidance principle, and local dynamic route planning based on the obstacle avoidance strategy is performed; Step 8: Calculate the speed and acceleration range and maximum tilt angle of the UAV in the local route planning, and determine the feasibility of the planned route. If the route is feasible, output the UAV local trajectory polynomial. If not, adjust the time allocation and re-plan until the feasible route is determined; Step 9: Calculate the lateral deviation of the local route planning endpoint relative to the global planning route. If the deviation is greater than the limit requirement, use this point as the starting point of the new UAV global route planning, and start the global route optimization fine-tuning from the current position to the destination; Step 10: Read the current position of the drone, perform rolling time domain planning calculations, update the starting and ending points of the local route planning, and perform new local route planning.

2. A method for autonomous route planning and collision avoidance decision-making for a UAV according to claim 1, Features: In step 3, the evaluation function is as follows: h(n)=w 1 ×h 1 (n)+w 2 ×h 2 (n)+w 3 ×h 3 (n); where h 1 (n) = dis diagonal (n, target) is the diagonal distance between nodes; h 2 (n) = cost obstacle_density is the obstacle density cost; h 3 (n) = cost nav_uncertainty is the navigation uncertainty cost at the node, w 1 、w 2 、w 3 It is a weighting factor, which can be adjusted appropriately according to the different levels of safety and economic requirements.

3. The method for autonomous route planning and collision avoidance decision-making of a UAV according to claim 1, Features: In step 4, the terrain database and the evaluation function calculated in step 3 are used to perform global route planning, and the global route that meets the economic and safety requirements is searched and stored through the A* algorithm.

4. The method for autonomous route planning and collision avoidance decision-making of a UAV according to claim 1, Features: In step 7, according to the obstacle avoidance principle, the obstacle avoidance requirement is modeled as an inequality constraint of the QP problem, and the relationship between the planned trajectory and the dynamic obstacle is limited to satisfy the following formula: (x own (t)-x obs (t)) 2 +(y own (t)-y obs (t)) 2 +(z own (t)-z obs (t)) 2 ≥radius own +radius obs ; Where (x own ,y own ,z own ) represents the position of the drone, (x obs ,y obs ,z obs ) represents the position of the dynamic obstacle, radius own and radius obs Represent the real-time equivalent radius of the drone and obstacle respectively; Combined with the surrounding dynamic obstacle information provided by the drone perception system, the drone obstacle avoidance maneuver is achieved through local waypoint adjustment and smoothing of obstacle avoidance trajectories between waypoints.

5. A system for autonomous route planning and collision avoidance decision-making of unmanned aerial vehicles. Features: include: The scenario simulation unit is used to construct the static airspace environment in which the UAV operates and the maneuverability parameters of the UAV, and to generate dynamic obstacles output by the situation awareness sensor; A data reading and processing unit is used to read and process the data output by the scene simulation unit to form route planning constraint information for calculating the route planning evaluation function; The global route optimization unit is used to generate a global planning route that meets economic and safety requirements based on the UAV flight mission requirements and operation scenario constraints; The local dynamic route planning unit is used to integrate the local airspace static environment and dynamic obstacle information, perform fine-grained normalization calculations, form local airspace constraint information, calculate the proximity relationship between the UAV and dynamic obstacles, and perform local route planning in combination with the UAV's motion characteristics to ensure the feasibility of the route. At the same time, it determines the route deviation caused by the UAV's collision avoidance maneuver, and timely initiates fine-tuning of the subsequent segments of the global planned route according to the degree of deviation; The rolling time domain planning unit is used to rollingly adjust the input information of the local dynamic route planning according to the position information of the UAV, so as to achieve a smooth transition and inheritance between the global route and the local route planning.

Citation Information

Cited By

  • Local target point selection method and device for rolling route planning of unmanned aerial vehicle

    CN120010502A

  • Security-enhanced trajectory planning method based on Astar algorithm

    CN120779970A