Return flight path planning method and device, unmanned aerial vehicle and storage medium

Abstract

The invention relates to the technical field of unmanned aerial vehicles, in particular to a homeward flight path planning method and device, an unmanned aerial vehicle and a storage medium. The homeward flight path planning method comprises the following steps: acquiring sensor data of an unmanned aerial vehicle, and preprocessing the sensor data; acquiring land parcel information according to the sensor data, and generating a global return path according to the land parcel information; judging whether an unmarked obstacle is encountered or not according to the sensor data, if so, generating a local return path for obstacle avoidance, returning to the original route after obstacle avoidance, and continuing to return; and if not, continuing to execute the global return path. According to the homeward voyage path planning method provided by the embodiment of the invention, global path planning and local path planning are combined to ensure that the unmanned aerial vehicle can avoid fixed obstacles and can cope with temporarily appearing obstacles in the homeward voyage process, so that safe, rapid and efficient homeward voyage is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a return path planning method, device, unmanned aerial vehicle and storage medium. Background Art

[0002] In UAV agricultural operations, the return process after completing the sowing mission often faces obstacle risks, including fixed obstacles (such as trees, telephone poles, ditches, buildings) and temporary obstacles (such as farm tools, personnel, and other agricultural machinery). If the return path is not properly planned, the UAV is likely to collide with these obstacles, causing equipment damage. This risk is particularly prominent in complex plot environments.

[0003] In particular, the information disclosed in this background technology section is only intended to increase the understanding of the overall background of the invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to ordinary technicians in the field. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a return path planning method, device, drone and storage medium to solve the problem of unreasonable return path planning in the related art.

[0005] In order to solve the above technical problems, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a return path planning method, comprising: Collecting sensor data of the drone and preprocessing the sensor data; Acquire land parcel information according to the sensor data, and generate a global return path according to the land parcel information; It is determined according to the sensor data whether an unmarked obstacle is encountered. If so, a local return path is generated to bypass the obstacle. After the obstacle is bypassed, the original route is returned to continue returning. If not, the global return path is continued.

[0006] Furthermore, the collecting of sensor data of the drone and preprocessing of the sensor data include: Collect the drone's real-time dynamic positioning RTK data, inertial measurement unit IMU data and millimeter wave radar data; Performing time synchronization processing and coordinate system conversion on the real-time dynamic positioning RTK data, inertial measurement unit IMU data and millimeter wave radar data; Extracting useful data from the point cloud information obtained from the millimeter wave radar; Use the processed millimeter-wave radar data to build different scene maps.

[0007] Further, acquiring the land parcel information according to the sensor data, and generating a global return path according to the land parcel information includes: Acquire the geometric shape, obstacle area and starting point information of the land plot according to the sensor data; The obstacle area is stored as polygonal boundary points, and the polygonal obstacle area is expanded; Generate a global return path based on the starting point, target point and expanded obstacle area.

[0008] Further, generating a global return path according to the starting point, the target point and the expanded obstacle area includes: Divide the plot into cells according to the starting point and the target point, wherein the cell division should avoid cutting the obstacle area; Identify hub cells from multiple cells; Generate multiple global return paths based on hub cells; An optimal global return path is determined from multiple global return paths.

[0009] Furthermore, it also includes: Use sensors to perceive the drone’s current environment; If the starting point of the drone is within the expanded obstacle area, a new starting point away from the obstacle area is selected; After exiting the obstacle area, re-plan the global return path.

[0010] Further, judging whether an unmarked obstacle is encountered according to the sensor data, and if so, generating a local return path to bypass the obstacle, returning to the original route after the obstacle is bypassed, and continuing to return includes: If an unmarked obstacle is encountered, determine whether horizontal obstacle avoidance can be performed; If yes, a local return path is generated to circumvent the obstacle; if no, the vehicle stops and waits for human intervention.

[0011] Further, judging whether an unmarked obstacle is encountered according to the sensor data, and if so, generating a local return path to bypass the obstacle, returning to the original route after the obstacle is bypassed, and continuing to return further includes: In the global return path, obstacles are continuously detected in real time through sensors; When encountering unmarked obstacles, a local grid map is constructed based on the Bayesian model, and the SVM model is loaded to identify the obstacle type; Generate a local return path to perform horizontal obstacle avoidance based on obstacle location and type; After the obstacle avoidance is completed, reconnect with the original return route.

[0012] In a second aspect, the present invention provides a return path planning device, comprising: Data acquisition module, used to collect sensor data from drones; A data processing module, used for preprocessing the sensor data; A land parcel information acquisition module, used to acquire land parcel information according to the sensor data; A return path generation module, used to generate a global return path according to the land parcel information; The obstacle avoidance path planning module is used to determine whether an unmarked obstacle is encountered based on the sensor data. If so, a local return path is generated to avoid the obstacle. After the obstacle is avoided, the original route is returned and the return is continued; if not, the global return path is continued.

[0013] In a third aspect, the present invention provides a drone, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the return path planning method described in any embodiment of the present invention.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the return path planning method described in any embodiment of the present invention when executed.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The return path planning method provided in this embodiment aims to design a path that ensures that the UAV can quickly and safely return from the current position (i.e., the starting point) to the designated location (i.e., the target point) after completing the sowing task. The core of this method is to combine global path planning and local path planning to ensure that the UAV can avoid fixed obstacles and deal with temporary obstacles during the return process, thereby achieving safe, fast and efficient return. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of a return path planning method provided in Embodiment 1 of the present invention; Figure 2 is a schematic structural diagram of the obstacle zone after expansion provided by the first embodiment of the present invention; Figure 3 is a schematic diagram of the position of a new starting point of the drone provided in the first embodiment of the present invention; Figure 4 is a schematic diagram of a situation where a drone provided by Embodiment 1 of the present invention encounters an unmarked obstacle area; Figure 5 This is a schematic diagram of a drone horizontally circumventing obstacles provided by the first embodiment of the present invention; Figure 6 This is a schematic diagram of the braking situation of the drone provided by the first embodiment of the present invention; Figure 7 is a structural diagram of a return path planning device provided in Embodiment 2 of the present invention; Figure 8 4 is a block diagram of a drone provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0017] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. In addition, the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict. It is also necessary to explain that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all structures.

[0018] It should be mentioned before discussing the exemplary embodiments in more detail that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the steps as sequential processes, many of the steps therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the steps can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0019] It should be noted that the concepts such as "first" and "second" mentioned in the embodiments of the present invention are only used to distinguish different devices, modules, units or other objects, and are not used to limit the order or interdependence of the functions performed by these devices, modules, units or other objects. Embodiment 1:

[0020] Figure 1 This is a flow chart of the return path planning method provided in Example 1 of the present invention. This embodiment can be applied to return path planning for a UAV. The method can be executed by a return path planning device. The return path planning device can be implemented in the form of hardware and / or software. The return path planning device can be configured in a UAV.

[0021] The return path planning method provided in this embodiment aims to design a path that ensures that the UAV can quickly and safely return from the current position (i.e., the starting point) to the designated location (i.e., the target point) after completing the sowing task. The core of this method is to combine global path planning and local path planning to ensure that the UAV can avoid fixed obstacles and deal with temporary obstacles during the return process, thereby achieving safe, fast and efficient return.

[0022] like Figure 1 As shown, the return path planning method includes: S110: Collect sensor data of the drone and pre-process the sensor data; In this embodiment, the sensor data includes real-time dynamic positioning RTK data, inertial measurement unit IMU data and millimeter wave radar data. The RTK (real-time dynamic positioning), IMU (inertial measurement unit) and millimeter wave radar are three commonly used sensors in drone technology. RTK is used to provide high-precision location information, IMU is used to provide attitude information of the drone, and millimeter wave radar is used to monitor the surrounding environment in real time to ensure the safety of drone obstacle avoidance and stable operation. This embodiment integrates multiple sensors to solve the problem of unstable perception of mapping model and unsafe flight control of drones in complex environments by the return obstacle avoidance algorithm with a single data source. The overall accuracy and robustness of the system can be improved through multi-sensor fusion, and the ability to cope with complex environmental changes can be enhanced.

[0023] For example, RTK (real-time dynamic positioning), IMU (inertial measurement unit) and millimeter-wave radar are installed on a plant protection UAV, and the UAV is placed in an area with flat ground and open areas on all sides for random flight to collect data.

[0024] S111: Preprocessing the sensor data includes: The real-time dynamic positioning RTK data, inertial measurement unit IMU data and millimeter wave radar data are subjected to time synchronization processing and coordinate system conversion; useful data are extracted from the point cloud information obtained from the millimeter wave radar; and different scene maps are established using the processed millimeter wave radar data.

[0025] Specifically, the time synchronization processing operation includes: combining the high-precision position information provided by RTK and the attitude information provided by IMU to obtain the real-time attitude information data of the drone; adding accurate timestamps to the attitude information provided by RTK and IMU and the millimeter-wave radar data, and finding the two most recent frames of RTK / IMU attitude data for the timestamp of the i-th frame of millimeter-wave radar data, and using the linear interpolation method to calculate the accurate attitude information corresponding to the i-th frame of millimeter-wave radar data, by ensuring that the timestamps of the millimeter-wave radar data are synchronized with those of other sensors to ensure that all data reflect the status at the same moment.

[0026] The coordinate system conversion operation includes: according to the current posture of the drone, converting the point cloud provided by the millimeter wave radar from the radar coordinate system to the world coordinate system or the global coordinate system.

[0027] The extracting useful data from the point cloud information obtained from the millimeter wave radar includes: performing a practicality judgment on the point cloud information provided by each frame of the millimeter wave radar according to the current posture of the unmanned aerial vehicle to filter out useless information.

[0028] More specifically, through a series of processing steps, including distance screening, noise removal, static / dynamic object differentiation, and redundant data elimination, useless millimeter-wave radar data is filtered out to reduce the amount of data that needs to be processed subsequently.

[0029] The method of establishing different scene maps using the processed millimeter-wave radar data includes: establishing different scene maps based on the screened radar data, including constructing a local grid map through scene classification, that is, pre-defining a series of typical application scenarios according to the mission type of the drone (such as urban patrol, forest monitoring, agricultural spraying, etc.), and setting specific map construction rules for each scenario.

[0030] Exemplarily, the current position information of the UAV is obtained, and the radar point cloud data is converted from the radar coordinate system to the world coordinate system. The radar point cloud data is filtered to ensure that only useful information is retained. The rules suitable for the scenario are located according to the task type, and a grid map is generated. The filtered point cloud data is mapped to the grid, and potential obstacle areas are marked through probability estimation. Features are extracted and incorporated into the map construction process. As the UAV flies, new radar feedback is continuously received, and the map is updated in real time to ensure that it reflects the latest environmental conditions.

[0031] S120: Acquiring land parcel information according to the sensor data, and generating a global return path according to the land parcel information, including: The geometric shape, obstacle area and starting point information of the land parcel are obtained according to the sensor data; the obstacle area is stored as polygonal boundary points, and the polygonal obstacle area is expanded; and a global return path is generated according to the starting point, the target point and the expanded obstacle area.

[0032] In this embodiment, map data of the land parcel may be collected by collecting sensor data, including the geometric shape of the terrain and the positions of known fixed obstacles in the land parcel, and the land parcel and the obstacle information inside the land parcel may be represented by polygons.

[0033] Specifically, for each obstacle, a series of ordered two-dimensional coordinate points (i.e., boundary points) are used to define its outline to form a closed polygon, and then each polygonal obstacle area is expanded according to the safety requirements of the drone.

[0034] For example, see Figure 2 , according to the actual size of the drone, sensor error, control accuracy and other factors, select the appropriate expansion distance, expand the polygon obstacle area, and form a new polygon boundary by expanding a certain distance outward on the basis of the original polygon. The expansion process must ensure that the expanded boundary does not conflict with other facilities. The expansion process can effectively increase the minimum distance between the drone and the obstacle, reducing the possibility of collision.

[0035] In this embodiment, by determining the current location of the drone (i.e., the starting point) and the designated safe landing point (target point), the global return path can be generated by combining the land plot and obstacle area information.

[0036] S121: Generating a global return path according to the starting point, the target point and the expanded obstacle area includes: The plot is divided into cells according to the starting point and the target point, wherein the cell division should avoid cutting the obstacle area; a hub cell is identified from multiple cells; multiple global return paths are generated according to the hub cell; and the optimal global return path is determined from the multiple global return paths.

[0037] Specifically, the path of the drone flying from the starting point to the target point should avoid crossing the obstacle area. If it crosses the obstacle area, the path is considered to be an invalid path. Therefore, in order to find the most efficient return path, this embodiment divides the land into several connected small cells without internal obstacle areas according to the expanded obstacle area. Each cell should be an obstacle-free space for the drone to pass freely. When dividing, the cell shape should be kept regular as much as possible to facilitate calculation and path planning.

[0038] In the divided cells, hub cells are further identified. Hub cells are nodes located at the intersection of multiple cells and used to connect different path segments. In path planning, hub cells are important nodes around complex terrain or obstacles, which provide safe path options to bypass obstacle areas. Hub cells are key points connecting different path segments and are crucial to ensure that the drone can successfully reach the target point from the starting point.

[0039] Based on the identified hub cells, a search algorithm is used to explore all possible path combinations. Each path should ensure passing through the selected hub cells, thereby providing multiple safe path options to bypass obstacles. For example, methods such as binary depth-first search (DFS) and Monte Carlo tree search (MCTS) are used to explore all possible global return path combinations.

[0040] Each generated global return path is evaluated, and the path sequence is optimized through the A* algorithm or the Euclidean distance algorithm in large-scale plots to find the most efficient path as the optimal global return path, thereby reducing the length of the drone's return path, saving time and power consumption.

[0041] Specifically, in the generated path, the system calculates the total length of the global return path through an algorithm based on the starting point and target point information, and selects the path with the shortest distance as the final global return path.

[0042] S122: Generating a global return path according to the starting point, the target point and the expanded obstacle area further includes: The current environment of the drone is perceived in real time through sensors. If the starting point of the drone is in the expanded obstacle area, a new starting point away from the obstacle area is selected; after exiting the obstacle area, the global return path planning is re-performed.

[0043] Specifically, when the drone is located inside the expanded obstacle zone, it needs to exit the obstacle zone first. At this time, a circle with a specified radius is drawn with the drone as the center. The radius should take into account the maximum turning radius and safety distance of the drone. Then, a suitable point is found at the intersection of this circle and the boundary of the expanded obstacle zone as the new starting point. The drone flies from the current position to the newly selected point, and then the global return path is planned with this point as the starting point.

[0044] It should be noted that the conditions for finding a new starting point include: the point is reachable, that is, there are no other obstacles blocking the straight-line path from the current position of the drone to the point; the point needs to ensure that the drone has enough space for turning and other operations after leaving the obstacle area; if there are multiple points that meet the conditions, the point closest to the original starting point or most conducive to subsequent route planning can be selected as the new starting point.

[0045] For example, see Figure 3 , the current position of the drone is inside the expanded obstacle area. At this time, the new starting point 1 and the new starting point 2 away from the obstacle area are selected, and then the drone is determined to fly to the new starting point 1 or the new starting point 2 according to the conditions for finding the new starting point. After exiting the obstacle area, the new starting point is used as the current position of the drone for global return path planning.

[0046] S130: Determine whether an unmarked obstacle is encountered based on the sensor data. If so, generate a local return path to bypass the obstacle. After bypassing the obstacle, return to the original route and continue to return. If not, continue to execute the global return path.

[0047] In this embodiment, the millimeter-wave radar is capable of continuously scanning an area within a certain range ahead even in adverse weather conditions such as rain and fog. By sensing the current environment of the drone in real time, it can be determined whether the drone encounters unmarked obstacles in front of it. If not, the drone can continue to execute the global return path until it returns to the target location.

[0048] If the drone encounters an unmarked obstacle during its return journey, it can generate a local return path to circumvent the obstacle. After circumventing the obstacle, it will return to the original route and continue its return journey.

[0049] Specifically, the drone is equipped with a millimeter-wave radar, which can continuously scan a certain range of areas ahead without being affected by bad weather such as rain and fog. Through real-time radar feedback, the drone can detect obstacles on the path, and by analyzing the distance, size and direction of the obstacle in the radar echo, determine the position coordinates of the obstacle relative to the drone, and accurately locate the position of the obstacle; further, by using machine learning or deep learning algorithms, the drone can classify obstacles according to their characteristics, such as using the SVM model to identify the type of obstacle, so as to select the optimal obstacle avoidance strategy.

[0050] In this embodiment, when sensors such as millimeter-wave radar detect an unmarked obstacle ahead, the system triggers local return path planning to bypass the obstacle.

[0051] Specifically, based on the Bayesian probability grid or similar environmental modeling methods, the drone can generate an obstacle avoidance path in real time within a local range. More specifically, with the current position of the drone as the center, a two-dimensional or three-dimensional grid map covering a certain range is generated, and the radar echo data is mapped into these grid cells, each cell represents a small area, and the state of the area is updated according to the received information.

[0052] Each grid cell is initialized using the Bayesian formula to calculate the probability that it contains an obstacle. As the drone continues to receive new sensor data, the probability values ​​in the grid cells are constantly adjusted. For example, if multiple scans show that there is an object in a grid cell, the probability that the cell is an obstacle area is increased; otherwise, the probability is reduced. Once the probability of an obstacle in a grid cell exceeds a preset threshold, the cell is officially marked as an obstacle area. Through probability estimation, the drone marks high-probability areas in the grid as obstacle areas, thereby avoiding flying to these areas.

[0053] In this embodiment, when encountering an unmarked obstacle, the drone generates a local return path planning to avoid obstacles in real time, ensuring that the drone can pass safely when encountering an unmarked obstacle. That is, the drone searches for the best path from the current drone position to the nearest safe point (rather than directly to the final target point) within a certain area.

[0054] Specifically, local return path planning aims to find a temporary safe detour solution to allow the drone to avoid obstacles in front of it and return to the original return route as soon as possible. On the local grid map, the drone uses the A* or Dijkstra algorithm to plan the local return path. The algorithm can calculate the optimal path from the current position of the drone to the safe point, thereby achieving the purpose of bypassing temporary obstacles, and based on the latest sensor feedback, the detour path can be updated in real time to cope with new obstacles or other changes that may appear. In the case of complex and changeable obstacles, randomized path planning algorithms such as the RRT (rapid random tree extension) algorithm can be used to generate a more flexible path.

[0055] The local return path planning includes: when the drone detects an obstacle ahead, the system will evaluate the location, size and shape of the obstacle based on sensor data. By calculating the possibility of bypassing the left and right sides, a better path is selected. Usually, the side with the smaller obstacle is selected for bypassing, which can shorten the obstacle bypass distance and reduce unnecessary flight time.

[0056] Specifically, during the obstacle avoidance process, the drone continuously uses sensor data to monitor changes in the environment around it, and adjusts the flight path through real-time radar feedback to avoid collisions with new obstacles. This shows that even after the initial local return path is determined, the drone can still respond to new situations, such as new obstacles that suddenly appear or previously unrecognized objects. For example, if the drone initially chooses to bypass the right side but then finds a new obstacle on the right side, the system will dynamically switch to bypass the left side quickly to ensure that a safe flight path is always maintained.

[0057] After completing the obstacle avoidance, the drone automatically adjusts its flight attitude and direction and returns to the global return path. Specifically, at the intersection of the local return path and the global return path, the drone adjusts its attitude and realigns itself with the return direction so as to seamlessly connect to the original route and continue the return mission.

[0058] For example, see Figure 4-6 ,When the drone encounters an unmarked obstacle in front, the system first determines whether ,horizontal obstacle avoidance can be performed. If it can, for example Figure 5 , a local return path is generated to avoid obstacles horizontally. When avoiding obstacles, the flight height of the drone above the ground should be kept constant. When the horizontal obstacle avoidance path fails to be searched, refer to Figure 6 , the drone should brake to a stop in front of an obstacle and wait for human intervention. When braking, the distance between the drone and the obstacle must be kept at a safe distance.

[0059] This embodiment aims to improve the path planning and obstacle avoidance capabilities of drones in complex and changing environments, such as rainy days, foggy days, and poor light conditions. In order to improve the path planning and obstacle avoidance capabilities of drones, an environmental perception and obstacle avoidance system model based on multi-sensor fusion is proposed. By combining the data of millimeter-wave radar, RTK (real-time dynamic positioning), and IMU (inertial measurement unit) sensors, and taking advantage of the fact that millimeter-wave radar is not subject to interference in bad weather, it provides high-precision positioning and attitude information to build a more robust drone perception system. The construction of this system further ensures the flight safety of drones in complex environments and effectively improves the rationality and efficiency of return path planning. Embodiment 2:

[0060] Figure 7 A schematic diagram of the structure of a return path planning device provided in Embodiment 2 of the present invention.

[0061] like Figure 7 As shown, the device comprises: Data acquisition module, used to collect sensor data from drones; A data processing module, used for preprocessing the sensor data; A land parcel information acquisition module, used to acquire land parcel information according to the sensor data; A return path generation module, used to generate a global return path according to the land parcel information; The obstacle avoidance path planning module is used to determine whether an unmarked obstacle is encountered based on the sensor data. If so, a local return path is generated to avoid the obstacle. After the obstacle is avoided, the original route is returned and the return is continued; if not, the global return path is continued.

[0062] The return path planning device provided in the embodiment of the present invention can execute the return path planning method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. Embodiment three:

[0063] Figure 8 A schematic diagram of an electronic device that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0064] like Figure 8 As shown, the electronic device includes at least one processor, and a memory connected to the at least one processor, such as a read-only memory (ROM), a random access memory (RAM), etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) or the computer program loaded from the storage unit to the random access memory (RAM). In the RAM, various programs and data required for the operation of the electronic device can also be stored. The processor, ROM and RAM are connected to each other through a bus. The input / output (I / O) interface is also connected to the bus.

[0065] Multiple components in an electronic device are connected to the I / O interface, including: input units, such as keyboards, mice, etc.; output units, such as various types of displays, speakers, etc.; storage units, such as disks, optical disks, etc.; and communication units, such as network cards, modems, wireless communication transceivers, etc. The communication unit allows the electronic device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunication networks.

[0066] The processor may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of processors include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor executes the various methods and processes described above, such as the return path planning method.

[0067] In some embodiments, the return path planning method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device via a ROM and / or a communication unit. When the computer program is loaded into the RAM 43 and executed by the processor, one or more steps of the return path planning method described above can be performed.

[0068] Alternatively, in other embodiments, the processor may be configured to execute the return path planning method in any other appropriate manner (eg, by means of firmware).

[0069] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0070] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0071] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0072] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0073] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0074] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0075] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0076] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A return route planning method, characterized in that: include: Collecting sensor data of the drone and preprocessing the sensor data; Acquire land parcel information according to the sensor data, and generate a global return path according to the land parcel information; It is determined according to the sensor data whether an unmarked obstacle is encountered. If so, a local return path is generated to bypass the obstacle. After the obstacle is bypassed, the original route is returned to continue returning. If not, the global return path is continued.

2. The method according to claim 1, characterized in that The collecting of the sensor data of the drone and preprocessing of the sensor data include: Collect the drone's real-time dynamic positioning RTK data, inertial measurement unit IMU data and millimeter wave radar data; Performing time synchronization processing and coordinate system conversion on the real-time dynamic positioning RTK data, inertial measurement unit IMU data and millimeter wave radar data; Extracting useful data from the point cloud information obtained from the millimeter wave radar; Use the processed millimeter-wave radar data to build different scene maps.

3. The method according to claim 2, characterized in that The acquiring of land parcel information according to the sensor data and generating of a global return path according to the land parcel information comprises: Acquire the geometric shape, obstacle area and starting point information of the land plot according to the sensor data; The obstacle area is stored as polygonal boundary points, and the polygonal obstacle area is expanded; Generate a global return path based on the starting point, target point and expanded obstacle area.

4. The method according to claim 3, characterized in that The generating of the global return path according to the starting point, the target point and the expanded obstacle area comprises: Divide the plot into cells according to the starting point and the target point, wherein the cell division should avoid cutting the obstacle area; Identify hub cells from multiple cells; Generate multiple global return paths based on hub cells; An optimal global return path is determined from multiple global return paths.

5. The method according to claim 4, characterized in that Also includes: Use sensors to perceive the drone’s current environment; If the starting point of the drone is within the expanded obstacle area, a new starting point away from the obstacle area is selected; After exiting the obstacle area, re-plan the global return path.

6. The method according to claim 5, characterized in that The determining whether an unmarked obstacle is encountered according to the sensor data, and if so, generating a local return path to bypass the obstacle, returning to the original route after the obstacle is bypassed, and continuing to return includes: If an unmarked obstacle is encountered, determine whether horizontal obstacle avoidance can be performed; If yes, a local return path is generated to circumvent the obstacle; if no, the vehicle stops and waits for human intervention.

7. The method according to claim 6, characterized in that The determining whether an unmarked obstacle is encountered according to the sensor data, and if so, generating a local return path to bypass the obstacle, returning to the original route after the obstacle is bypassed, and continuing to return further includes: In the global return path, obstacles are continuously detected in real time through sensors; When encountering unmarked obstacles, a local grid map is constructed based on the Bayesian model, and the SVM model is loaded to identify the obstacle type; Generate a local return path to perform horizontal obstacle avoidance based on obstacle location and type; After the obstacle avoidance is completed, reconnect with the original return route.

8. A return route planning device, characterized in that: include: Data acquisition module, used to collect sensor data from drones; A data processing module, used for preprocessing the sensor data; A land parcel information acquisition module, used to acquire land parcel information according to the sensor data; A return path generation module, used to generate a global return path according to the land parcel information; The obstacle avoidance path planning module is used to determine whether an unmarked obstacle is encountered based on the sensor data. If so, a local return path is generated to avoid the obstacle. After the obstacle is avoided, the original route is returned and the return is continued; if not, the global return path is continued.

9. A drone, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the return path planning method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the return path planning method as described in any one of claims 1-7 is implemented.

Images