Automatic obstacle avoidance method and system based on user-defined coordinate system navigation

Abstract

The invention provides an automatic obstacle avoidance method and system based on user-defined coordinate system navigation, and relates to the technical field of automatic navigation, and the method comprises the following steps: obtaining global position information of a moving unit in a global coordinate system; judging whether the moving unit enters a new scene internal area according to the global position information; if yes, acquiring spatial morphological characteristics of an internal area of the scene into which the moving unit newly enters; according to the acquired information, judging the area type of the internal area of the newly entered scene; establishing a local coordinate system corresponding to the internal area of the newly entered scene according to the area type; and generating a local navigation path in the internal area of the scene based on the established local coordinate system. According to the method, a local coordinate system is defined, a coordinate system better matched with specific task requirements is adopted in a small range, meanwhile, the data processing amount is reduced, path planning and an obstacle avoidance algorithm are easier to implement, and higher flexibility, lower calculation cost and better precision are provided under specific conditions.

Description

Technical Field

[0001] The present invention relates to the field of automatic navigation technology, and in particular to an automatic obstacle avoidance method and system based on custom coordinate system navigation. Background Art

[0002] Custom coordinate system path navigation refers to planning and executing a path from one point to another based on a user-defined coordinate system in a specific application scenario. This type of navigation is widely used in robotics, self-driving cars, drone control, and various applications that require precise position tracking and movement. Automatic obstacle avoidance based on custom coordinate system navigation is one of the key technologies to ensure that mobile robots or self-driving vehicles reach their target locations safely and efficiently in unknown environments. However, in a constantly changing environment, how to efficiently update the map information in the custom coordinate system and maintain the effectiveness and safety of path planning is a difficult problem.

[0003] Therefore, how to provide a more efficient and accurate automatic obstacle avoidance method based on custom coordinate system navigation is a problem that needs to be solved urgently. Summary of the invention

[0004] In order to improve the above problems, the present invention provides an automatic obstacle avoidance method and system based on custom coordinate system navigation.

[0005] According to a first aspect of an embodiment of the present invention, there is provided an automatic obstacle avoidance method based on navigation in a custom coordinate system, the method comprising: Get the global position information of the mobile unit in the global coordinate system; Determining whether the mobile unit has entered a new scene internal area according to the global position information; If yes, then obtain the spatial morphological features of the internal area of ​​the scene newly entered by the mobile unit; Determine the area type of the newly entered internal area of ​​the scene based on the acquired information; Establishing a local coordinate system corresponding to the newly entered internal area of ​​the scene according to the area type; A local navigation path within the internal area of ​​the scene is generated based on the established local coordinate system.

[0006] Optionally, the step of determining the area type of the newly entered internal area of ​​the scene according to the acquired information specifically includes: Get the internal area classification standard of the building where the mobile unit is located; The internal area of ​​the scene is classified according to the spatial morphological features of the newly entered internal area of ​​the scene and the global position information to obtain a corresponding area type.

[0007] Optionally, the step of establishing a local coordinate system corresponding to the newly entered internal area of ​​the scene according to the area type specifically includes: Determining the number of dimensions and obstacle avoidance accuracy of the corresponding local coordinate system according to the area type; A local coordinate system is established based on the number of dimensions and the obstacle avoidance accuracy.

[0008] Optionally, the method of establishing a local coordinate system includes: Select a stable feature point in the newly entered internal area of ​​the scene as the origin; Determine the direction of the coordinate axis according to the behavior target of the mobile unit in the newly entered internal area of ​​the scene; Establishing a local coordinate system based on the origin and the coordinate axis directions; Determine the mapping relationship between the local coordinate system and the global coordinate system.

[0009] Optionally, the method for determining the behavior target includes: Determine whether the destination of the mobile unit is within the newly entered internal area of ​​the scene according to the global navigation path of the mobile unit in the global coordinate system; If not, the behavior goal is to pass through the internal area of ​​the scene; If yes, the behavior goal is to enter the internal area of ​​the scene.

[0010] Optionally, the step of generating a local navigation path within the internal area of ​​the scene based on the established local coordinate system specifically includes: Determine the navigation end point of the mobile unit in the local coordinate system according to the behavior target; Obtain obstacle feature data corresponding to the current area type according to the area type; Perform obstacle detection within the internal area of ​​the scene based on the obstacle feature data; A local navigation path with a target as a navigation end point is generated according to the obstacle detection result and the obstacle avoidance accuracy.

[0011] Optionally, the step of determining the navigation endpoint of the mobile unit in the local coordinate system according to the behavior target specifically includes: When the behavior goal is to pass through the internal area of ​​the scene, the navigation end point is the coordinates of the exit of the internal area of ​​the scene in the local coordinate system; When the behavior goal is to enter the internal area of ​​the scene, the navigation end point is the coordinates of the destination of the mobile unit in the local coordinate system.

[0012] Optionally, the step of performing obstacle detection within the internal area of ​​the scene based on the obstacle feature data specifically includes: Obtain obstacle information within the scene through the sensors of the mobile units and perform feature extraction; Obtaining an obstacle classification model based on the extracted potential obstacle features and obstacle feature data; Obstacle detection is performed through obstacle classification models.

[0013] Optionally, the step of generating a local navigation path with a target as a navigation end point according to the obstacle detection result and the obstacle avoidance accuracy specifically includes: Determine a corresponding path algorithm based on the obstacle avoidance accuracy; Determine obstacle avoidance rules for various obstacles in the local coordinate system based on the results of obstacle detection; A local navigation path to a navigation end point is generated based on the path algorithm and obstacle avoidance rules.

[0014] A second aspect of an embodiment of the present invention provides an automatic obstacle avoidance system based on custom coordinate system navigation, comprising: A position acquisition unit, used to acquire global position information of the mobile unit in the global coordinate system; An area determination unit, used to determine whether the mobile unit has entered a new scene internal area according to the global position information; an area detection unit, for obtaining, if yes, the spatial morphological features of the internal area of ​​the scene newly entered by the mobile unit; A type determination unit, used to determine the area type of the newly entered internal area of ​​the scene according to the acquired information; A coordinate system establishing unit, used to establish a local coordinate system corresponding to the newly entered internal area of ​​the scene according to the area type; The path generation unit is used to generate a local navigation path within the internal area of ​​the scene based on the established local coordinate system.

[0015] In summary, the present invention provides an automatic obstacle avoidance method and system based on custom coordinate system navigation, which can adopt a coordinate system that better matches the specific task requirements within a small range, while reducing the amount of data processing, making path planning and obstacle avoidance algorithms easier to implement, and providing higher flexibility, lower computing costs and better accuracy under specific conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0017] Figure 1A method flow chart of an automatic obstacle avoidance method based on custom coordinate system navigation according to an embodiment of the present invention; Figure 2 The figure is a functional module block diagram of an automatic obstacle avoidance system based on custom coordinate system navigation according to an embodiment of the present invention.

[0018] Reference numerals: Position acquisition unit 110; area determination unit 120; area detection unit 130; type determination unit 140; coordinate system establishment unit 150; path generation unit 160. DETAILED DESCRIPTION

[0019] Custom coordinate system path navigation refers to planning and executing a path from one point to another based on a user-defined coordinate system in a specific application scenario. This type of navigation is widely used in robotics, self-driving cars, drone control, and various applications that require precise position tracking and movement. Automatic obstacle avoidance based on custom coordinate system navigation is one of the key technologies to ensure that mobile robots or self-driving vehicles reach their target locations safely and efficiently in unknown environments. However, in a constantly changing environment, how to efficiently update the map information in the custom coordinate system and maintain the effectiveness and safety of path planning is a difficult problem.

[0020] Therefore, how to provide a more efficient and accurate automatic obstacle avoidance method based on custom coordinate system navigation is a problem that needs to be solved urgently.

[0021] In view of this, the designers of the present invention have designed an automatic obstacle avoidance method and system based on custom coordinate system navigation.

[0022] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0025] In the description of the present invention, it should be noted that the terms "top", "bottom", "inside", "outside", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the invented product is usually placed when in use, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply 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 to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0026] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0027] 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.

[0028] The following is a detailed description of an automatic obstacle avoidance method based on custom coordinate system navigation provided by this embodiment.

[0029] See also Figure 1 This embodiment provides an automatic obstacle avoidance method based on custom coordinate system navigation, the method comprising: Step S101, obtaining the global position information of the mobile unit in the global coordinate system.

[0030] In this embodiment, the mobile unit may be a drone, a robot, or other equipment with automatic navigation function. Considering that conventional coordinate systems (such as the world geographic coordinate system, the GPS coordinate system, the local Cartesian coordinate system, etc.) are mostly used in outdoor scenes, the embodiment of the present invention is mainly aimed at custom coordinate system navigation in indoor, underground or closed space scenes.

[0031] In some special environments (such as indoors, underground, or enclosed spaces), it may not be possible to use the Global Positioning System (GPS) or other technologies that rely on the Earth's reference frame. In this case, a custom coordinate system can better adapt to these environmental characteristics and provide more accurate location information.

[0032] This solution is a further optimization of the conventional custom coordinate system navigation method. In this embodiment, the global coordinate system can be any of the conventional coordinate systems mentioned above, or it can be a new coordinate system custom-established based on the current scene.

[0033] In the global coordinate system, based on the current position of the mobile unit, the global position information of the mobile unit, that is, the coordinates of the mobile unit in the global coordinate system, can be obtained. In most application scenarios, there is a regional division, and there may be different types of obstacles or terrain differences in different areas, which leads to different obstacle avoidance requirements when the mobile unit moves in the corresponding area. Based on the differences in these obstacle avoidance requirements, in the implementation of this method, a local coordinate system is established separately for different spaces or areas based on their specific obstacle avoidance requirements, so that the mobile unit can use obstacle avoidance and navigation algorithms that better match the requirements in the local coordinate system to improve the accuracy and efficiency of the mobile unit during navigation.

[0034] Step S102: judging whether the mobile unit has entered a new scene internal area according to the global position information. If yes, executing step S103.

[0035] The choice of the global coordinate system corresponds to the current scene. Therefore, by obtaining the global position information of the mobile unit in the global coordinate system, the specific area where the mobile unit is located can be effectively determined, thereby facilitating subsequent custom coordinate navigation.

[0036] In actual applications, there are two situations. One is that all the internal areas of the current scene are known areas, and each area has a clear corresponding coordinate range in the global coordinate system. By obtaining the global position information, it can be judged whether the mobile unit has entered a new internal area of ​​the scene. The other situation is that there is an unknown area in the current scene. When the mobile unit is in a known area, the judgment method is the same as the first situation. When the mobile unit enters an unknown area, its global position information does not belong to any known area. At this time, it can be judged that the mobile unit has entered a new internal area of ​​the scene.

[0037] When it is determined that a mobile unit enters from an internal area of ​​a scene to a new internal area of ​​the scene, it means that the obstacle avoidance requirements at this time may change. Therefore, it is necessary to establish a local coordinate system that matches the requirements based on the obstacle avoidance requirements of entering the new internal area of ​​the scene, and then use the obstacle avoidance strategy and navigation method that match the requirements based on the newly established local coordinate system.

[0038] Step S103, obtaining the spatial morphological features of the internal area of ​​the scene newly entered by the mobile unit.

[0039] Spatial morphological features are used to describe the morphological structure of the internal space of the scene, mainly including parameters such as geometric size and shape.

[0040] In different application scenarios, the spatial morphological characteristics of the internal area of ​​the scene determine the main function or positioning of the area (for example, in a warehouse, rooms of different sizes correspond to different storage types, and in a hospital, rooms of different sizes correspond to different functional rooms), which in turn makes the internal obstacles and obstacle avoidance requirements different. For example, in a hospital, the main obstacles in the internal area of ​​the scene corresponding to the corridor are people coming and going and facilities in fixed positions, and the main obstacles in the internal area of ​​the scene corresponding to the ward are the beds and cabinets in the ward. In the corridor, you can move at a relatively higher speed than in the ward, but the obstacle avoidance distance also needs to be greater than the obstacle avoidance distance in the ward.

[0041] By obtaining the spatial morphological features of the area inside the scene that the mobile unit has newly entered, it is possible to quickly and easily determine the function or positioning of the area inside the scene that has newly entered, and then determine the corresponding obstacle avoidance requirements.

[0042] Specifically, the acquisition of spatial morphological features mainly relies on various sensors installed on mobile units, including but not limited to LiDAR, RGB-D cameras, ultrasonic sensors, etc. Among them, LiDAR calculates the distance by emitting laser beams and measuring the time difference of reflected light, generating high-resolution point cloud maps, which can quickly cover large areas and obtain three-dimensional models of the internal areas of the scene in a short time. RGB-D cameras capture color images and depth information at the same time, and use time of flight (ToF), structured light or stereo vision technology to obtain depth data. Ultrasonic sensors determine the distance of objects by emitting high-frequency sound waves and based on the echo time.

[0043] It should be noted that, considering the reasons such as the occlusion by obstacles, the acquisition of the spatial morphological features of the internal area of ​​the scene can be carried out continuously, and the spatial morphological feature data can be continuously corrected based on the acquired information. For example, when the mobile unit just enters a new internal area of ​​the scene, the currently available spatial morphological features can be immediately acquired, and then as the mobile unit continues to move in the internal area of ​​the scene, more spatial morphological feature data can be acquired.

[0044] Therefore, in order to obtain spatial morphological features, the inertial measurement unit (IMU) or odometer can be combined with the above-mentioned lidar, RGB-D camera, ultrasonic sensor and other equipment to obtain more accurate spatial morphological features.

[0045] For example, during the initial quick scan, a one-time quick full-range scan is performed using LiDAR to obtain the general outline and main structural features of the entire interior area of ​​the scene. This method can be completed in seconds and is very suitable for obtaining basic information when entering a new space for the first time. Detailed measurement and verification, combined with RGB-D cameras, carefully scan key areas (such as walls, doors, windows, facilities, etc.) to obtain richer color and texture information, and further confirm the specific size and layout of the interior area of ​​the scene. This step helps to correct possible errors in the initial scan and provide a more accurate map for subsequent tasks. Real-time monitoring and adjustment, using ultrasonic sensors and IMU / odometers to continuously monitor changes in the surrounding environment to ensure that mobile units or equipment always accurately obtain spatial morphological features.

[0046] Step S104: determine the area type of the newly entered scene internal area according to the acquired information.

[0047] In the application scenario, the spatial morphological features of the internal areas of the scene with different functions, that is, different area types, can be set in advance, and the range of the spatial morphological features corresponding to each area type can be set based on the current application scenario. Taking the smart warehousing and logistics warehouse as an example, the sizes and shapes of warehouses with different functions are different. The size and shape of the newly entered warehouse can be obtained to determine what function the warehouse belongs to.

[0048] Different types of areas have different obstacles in their areas, which means that when moving in different types of scenes, the obstacle avoidance requirements will also be different. Based on different obstacle avoidance requirements, mobile units can adopt more appropriate obstacle avoidance strategies and navigation algorithms based on customized coordinate systems.

[0049] On this basis, for different area types, in addition to the range of pre-set spatial morphological features, the position of each area in the global coordinate system can be determined based on previously acquired data, which is also helpful for judging the type of newly entered area through global position information. At the same time, with the help of accumulated historical data, obstacle feature data of obstacles appearing in different area types can also be obtained, which helps mobile units to identify obstacles more accurately during navigation.

[0050] Specifically, as a preferred method of the embodiment of the present invention, step S104 specifically includes: Get the internal area classification standard of the building where the mobile unit is located; The internal area of ​​the scene is classified according to the spatial morphological features of the newly entered internal area of ​​the scene and the global position information to obtain a corresponding area type.

[0051] By confirming the area type of the newly entered internal area of ​​the scene, the obstacle avoidance requirements of the internal area of ​​the scene where the mobile target is currently located can be obtained, which facilitates the subsequent customized coordinate system and path navigation.

[0052] Step S105 , establishing a local coordinate system corresponding to the newly entered internal area of ​​the scene according to the area type.

[0053] A new local coordinate system is established for each newly entered scene internal area, so that the mobile unit can navigate and avoid obstacles based on the local coordinate system in the current scene internal area. It is of great significance to dynamically establish different custom coordinate systems based on the current space. This method allows the system to flexibly adjust its reference frame according to environmental changes, so as to better adapt to the needs of different scenes. Custom coordinate systems can better adapt to the characteristics of each scene internal area and provide more accurate location information. For example, AGV (automatic guided vehicle) scheduling in a warehouse or the operation of drones in a small space can establish the most suitable coordinate system according to specific task requirements, thereby improving work efficiency. In addition, by defining a local coordinate system, more sophisticated operations can be performed in a small range without processing the entire large-scale map data, reducing unnecessary computational burden. When using a global coordinate system in a large-scale environment, due to the limitations of the measurement equipment itself and the influence of external interference factors, large cumulative errors may be introduced. The custom coordinate system can calibrate these errors through relative position relationships to improve positioning accuracy. At the same time, when the environment changes (such as new obstacles), the map under the local coordinate system can be quickly updated, and the path planning in the scene internal area can be adjusted accordingly to ensure the response speed and flexibility of the system.

[0054] Specifically, as a preferred method of an embodiment of the present invention, step S105 specifically includes:

[0055] Determining the number of dimensions and obstacle avoidance accuracy of the corresponding local coordinate system according to the area type; A local coordinate system is established based on the number of dimensions and the obstacle avoidance accuracy.

[0056] Based on the characteristics of the region type, considering the specific geometric shape and symmetry of the region type, select the coordinate system that can best simplify the problem statement.

[0057] In some areas, only a two-dimensional plane navigation path may be required. For example, most ground robots or indoor navigation systems usually use a two-dimensional rectangular coordinate system (XY). For environments with significant height differences or irregular surfaces, even a two-dimensional custom coordinate system may not be sufficient to describe all situations. In this case, additional height information may need to be introduced to enhance the system's adaptability, that is, a three-dimensional coordinate system is used. For the internal spaces of some building structures with special shapes, such as circular buildings, piping systems, hemispherical dome buildings, etc., polar coordinate systems, cylindrical coordinate systems, or spherical coordinate systems can be used.

[0058] As other embodiments of the present invention, in addition to determining the coordinate system type of the local coordinate system according to the area type, it can also be determined according to the task requirements currently performed by the mobile unit, and according to the characteristics of the task (such as high obstacle avoidance accuracy requirements), the coordinate system most suitable for completing the task can be selected. For example, when high-precision obstacle avoidance and local adjustment are required, a relative coordinate system can be used. The relative coordinate system allows the mobile unit to use itself as a reference point and monitor changes in the surrounding environment in real time. It is convenient to quickly respond to new obstacles and make local path adjustments to ensure safe movement.

[0059] In addition, you can also ensure that the selected coordinate system is convenient for processing sensor data based on the type of sensor used by the mobile unit and its data output format. For example, when the sensor used is an ultrasonic sensor to avoid obstacles in a narrow channel, a polar coordinate system is used. The ultrasonic sensor mainly provides distance information, while the polar coordinate system is centered at its origin and uses radius and angle to describe the position. It is very suitable for this application scenario and can simplify the design of close-range obstacle detection and obstacle avoidance strategies.

[0060] After determining the type of local coordinate system based on the above coordinate system selection principles and obstacle avoidance accuracy, the coordinate system can be established.

[0061] Specifically, the method of establishing a local coordinate system includes: Select a stable feature point in the newly entered internal area of ​​the scene as the origin; Determine the direction of the coordinate axis according to the behavior target of the mobile unit in the newly entered internal area of ​​the scene; Establishing a local coordinate system based on the origin and the coordinate axis directions; Determine the mapping relationship between the local coordinate system and the global coordinate system.

[0062] When establishing a local coordinate system, except for some special coordinate systems, the origin and the direction of the coordinate axis need to be determined first. The origin is usually a feature point in the selection area that is very easy to identify and has a stable position. The direction of the coordinate axis is related to the task performed by the mobile unit in the internal area of ​​the current scene, that is, the behavioral goal. After selecting the appropriate coordinate axis direction to establish the coordinate system, it can simplify the design of obstacle detection and obstacle avoidance strategies to a certain extent, and facilitate faster generation and change of navigation paths.

[0063] Specifically, the method for determining the behavior target includes: Determine whether the destination of the mobile unit is within the newly entered internal area of ​​the scene according to the global navigation path of the mobile unit in the global coordinate system; If not, the behavior goal is to pass through the internal area of ​​the scene; If yes, the behavior goal is to enter the internal area of ​​the scene.

[0064] When a mobile unit is performing a task, a global navigation path will be generated based on the global coordinate system. The global navigation path indicates the starting position and the destination position in the global coordinate system. According to the internal area of ​​the scene where the destination position is located, it can be determined which internal areas of the scene the mobile unit needs to pass through and which internal area of ​​the scene it will finally stay in to reach the destination. Based on this, the behavior goals of the mobile unit in a certain internal area of ​​the scene can be divided into two categories: one is to pass through the internal area of ​​the scene, and the other is to enter the internal area of ​​the scene and stay.

[0065] In the first case, the coordinate system is established based on the need to pass through the entire scene area. In the second case, the consideration is how to navigate to the destination in the scene area. It is necessary to determine the appropriate coordinate axis direction based on different behavior goals and then establish a local coordinate system.

[0066] It should be noted that after establishing the local coordinate system, the coordinate conversion mechanism between the local coordinate system and the global coordinate system needs to be considered, that is, to determine the mapping relationship between the position data in different coordinate systems to ensure that all sensor data can be accurately mapped to the corresponding local coordinate system. The transformation matrix can be used to implement translation and rotation operations, and the Kalman filter can be used to compensate for possible errors.

[0067] Through the above process, after the local coordinate system of the newly entered internal area of ​​the scene is established, navigation based on the custom local coordinate system can be performed.

[0068] Step S106: generating a local navigation path within the internal area of ​​the scene based on the established local coordinate system.

[0069] When performing obstacle avoidance navigation in a local coordinate system, it is necessary to determine the obstacle avoidance rules or strategies based on the current obstacle avoidance requirements, including speed adjustment, steering angle control, etc., based on the characteristics of the current scene's internal area (including obstacle type, number, etc.) and the behavioral goals in the current scene's internal area, to ensure that you can respond quickly when you encounter obstacles. For example, slow down when approaching pedestrians, or bypass medical equipment that suddenly appears in front. Then, based on the current position of the mobile unit and the navigation end point in the local coordinate system, select a suitable navigation algorithm to generate a local navigation path.

[0070] Specifically, as a preferred embodiment of the present invention, step S106 specifically includes: Determine the navigation end point of the mobile unit in the local coordinate system according to the behavior target; Obtain obstacle feature data corresponding to the current area type according to the area type; Perform obstacle detection within the internal area of ​​the scene based on the obstacle feature data; A local navigation path with a target as a navigation end point is generated according to the obstacle detection result and the obstacle avoidance accuracy.

[0071] Specifically, the step of performing obstacle detection within the internal area of ​​the scene based on the obstacle feature data specifically includes: Obtain obstacle information within the scene through the sensors of the mobile units and perform feature extraction; Obtaining an obstacle classification model based on the extracted potential obstacle features and obstacle feature data; Obstacle detection is performed through obstacle classification models.

[0072] When generating a local navigation path, obstacle detection in the internal area of ​​the scene is also a very important link. The current common classification method is to train a deep learning model based on a large number of labeled data sets to obtain an obstacle classification model for the internal area of ​​the scene. Therefore, the data set used for training has a great influence on the accuracy of the classification. The obstacle feature data corresponding to the current area type is obtained through the area type. These obstacle feature data correspond to the characteristics of the obstacles that appear in the area type. Based on the obstacle classification model trained with these obstacle feature data, obstacle detection is performed on the potential obstacle features in the internal area of ​​the current scene, and obstacles can be quickly classified. Based on the set obstacle avoidance accuracy and obstacle avoidance rules, the corresponding navigation path can be obtained. As a preferred implementation, a lightweight target detection algorithm, such as YOLOv5, can also be deployed to achieve real-time obstacle monitoring, ensure rapid response to new obstacles, and dynamically adjust the navigation path.

[0073] Specifically, the step of determining the navigation end point of the mobile unit in the local coordinate system according to the behavior target specifically includes: When the behavior goal is to pass through the internal area of ​​the scene, the navigation end point is the coordinates of the exit of the internal area of ​​the scene in the local coordinate system; When the behavior goal is to enter the internal area of ​​the scene, the navigation end point is the coordinates of the destination of the mobile unit in the local coordinate system.

[0074] According to the above behavior-goal division method, when it is the first case, it means that the mobile unit has entered the current scene internal area and needs to leave again, so the navigation end point is the coordinates of the exit of the scene internal area in the local coordinate system. When it is the second case, it means that the destination of the mobile unit is in the current scene internal area. According to the mapping relationship between the local coordinate system and the global coordinate system, the coordinates of the destination in the global coordinate system need to be converted to the coordinates in the local coordinate system, and then the coordinates of the destination in the local coordinate system are used as the navigation end point.

[0075] As a preferred implementation, the step of generating a local navigation path with the target being the navigation end point according to the obstacle detection result and the obstacle avoidance accuracy specifically includes: Determine a corresponding path algorithm based on the obstacle avoidance accuracy; Determine obstacle avoidance rules for various obstacles in the local coordinate system based on the results of obstacle detection; A local navigation path to a navigation end point is generated based on the path algorithm and obstacle avoidance rules.

[0076] After determining the navigation end point and obstacle avoidance accuracy, we can select a suitable path algorithm to generate a local navigation path based on the classification results of different types of obstacles according to the obstacle avoidance rules. Common path algorithms include A* algorithm, Dijkstra algorithm, RRT (fast random tree) algorithm, DWA (dynamic window method), VFH (Vector Field Histogram) algorithm, etc. Among them, the A* algorithm is suitable for situations where the map is known and the shortest path needs to be found, and can quickly find the optimal path from the starting point to the end point. The Dijkstra algorithm is suitable for undirected graphs or environments with consistent weights, and is very effective for simple and small maps. The RRT (fast random tree) algorithm is suitable for path planning in high-dimensional space or unknown environments, and is suitable for real-time path planning, especially when the environment changes, the path can be adjusted quickly. DWA (dynamic window method) is suitable for scenarios with fast moving speeds and frequent path adjustments. The VFH (Vector Field Histogram) algorithm is suitable for dense environments with a large number of obstacles, and is particularly effective for close-range obstacle detection, facilitating rapid response.

[0077] Obstacle avoidance rules mainly consider dimensions such as speed adjustment, direction adjustment, and safe distance, and require actual application requirements to match obstacle avoidance accuracy. For example, speed adjustment can include the range and speed of deceleration approach and acceleration principles, direction adjustment ensures detour strategies and actions such as following the wall, and safe distance includes minimum spacing and dynamic adjustment with changes in the environment or the state of the mobile unit.

[0078] The specific implementation process of the above steps is as follows: 1. Initialization parameters According to the mission requirements and sensor data, set key parameters such as initial position, target position, obstacle position, etc.

[0079] 2. Execution path planning Apply the selected path algorithm to calculate the best path from the current position to the navigation end point in the local coordinate system.

[0080] 3. Integrate obstacle avoidance rules Obstacle avoidance rules are incorporated into the path planning process to ensure that the path is not only optimal but also safe.

[0081] 4. Real-time updates As the environment changes or mobile units move, path planning is continuously updated to ensure that it always moves along the safest and most efficient route.

[0082] 5. Smoothing The initially generated path is optimized and smoothed to reduce sharp turns or sudden accelerations, making the movement more fluid and natural.

[0083] For example, in a hospital environment, an intelligent medicine delivery robot needs to move from the current ward (starting point) to another ward (end point), avoiding pedestrians and other equipment in the corridor. In this scenario: Generate local coordinate system: Construct local coordinate systems in the current ward, corridor, and another ward respectively.

[0084] Select path algorithm: Considering that the hospital environment is relatively fixed and known, the A* algorithm can be selected to find the shortest path.

[0085] Create obstacle avoidance rules: Use speed regulation and direction adjustment strategies to ensure that the robot can slow down and steer around pedestrians or other obstacles when approaching them.

[0086] Generate local navigation paths: Combine the A* algorithm and obstacle avoidance rules to calculate safe and efficient paths in three local coordinate systems and smooth them to ensure that the robot can complete the task smoothly.

[0087] The above is all the implementation schemes of the automatic obstacle avoidance method based on custom coordinate system navigation provided by the embodiments of the present invention, which will be explained below through a case study of a specific scenario.

[0088] Application scenario: Automatic obstacle avoidance method for hospital intelligent medicine delivery robot.

[0089] A large general hospital introduced an intelligent medicine delivery robot to optimize the in-hospital medicine delivery process and reduce the workload of nurses. The robot needs to autonomously navigate in complex indoor environments (such as corridors, wards, elevators, etc.), and be able to safely avoid obstacles (such as pedestrians, beds, carts, etc.) to deliver medicines accurately to designated locations.

[0090] The specific steps are as follows: 1. Environmental Perception and Modeling Sensor selection: Use LIDAR to scan the environment from a distance.

[0091] Equipped with an RGB-D camera to capture color images and depth information.

[0092] Install ultrasonic sensors for close-range obstacle detection.

[0093] Data collection and processing: Data from various sensors are collected in real time, and after pre-processing such as filtering and denoising, an environmental point cloud map is generated.

[0094] Map building: Using SLAM (Simultaneous Localization and Mapping) algorithms, we create and continuously update maps of the hospital interior, which include not only fixed structures such as walls and doors, but also dynamic objects such as moving people or equipment.

[0095] 2. Dynamic custom coordinate system establishment Origin selection: A fixed reference point is set at the entrance of each floor as the origin of the global coordinate system.

[0096] When the robot enters a new room or area, it temporarily selects a stable feature point (such as the door of the ward, elevator button, etc.) as the origin of the local coordinate system.

[0097] Coordinate axis direction: The global coordinate system sets the X-axis along the main direction of travel of the corridor, and the direction perpendicular to the X-axis is the Y-axis; for three-dimensional applications, a Z-axis pointing to the ceiling is added.

[0098] The local coordinate system can be flexibly adjusted according to specific task requirements. In the ward, the bedside position can be set as the origin to facilitate operations close to the patient's bedside.

[0099] Coordinate transformation mechanism: Determine the coordinate transformation algorithm to ensure that all sensor data can be accurately mapped to the corresponding coordinate system. Use the transformation matrix to implement translation and rotation operations, and use the Kalman filter to compensate for possible errors.

[0100] 3. Obstacle recognition and classification Feature extraction: A convolutional neural network (CNN) is used to extract features of potential obstacles from images acquired by an RGB-D camera.

[0101] Classifier training: Based on the hospital scenario, we obtain labeled obstacle feature datasets for different areas and train the deep learning model to distinguish between static obstacles (such as furniture and equipment) and dynamic obstacles (such as pedestrians and moving vehicles).

[0102] Real-time detection: Deploy lightweight object detection algorithms to achieve real-time obstacle monitoring and ensure rapid response to newly emerging obstacles.

[0103] 4. Path planning and re-planning Local path generation: The A* search algorithm is used to calculate the optimal path on the map of each local coordinate system, taking into account factors such as corridor width and turning radius to ensure that the robot can pass smoothly in the limited space.

[0104] Local obstacle avoidance strategy: Based on the characteristics of each area, local obstacle avoidance rules are determined, including speed adjustment, steering angle control, etc., to ensure quick response when encountering obstacles. Smoothing: The initially generated path is optimized by spline interpolation to reduce sharp turns or sudden accelerations, making the movement smoother and more natural.

[0105] 5. Control execution and feedback adjustment Motion Control: Adjust the direction and speed according to the deviation between the current position and the expected trajectory to ensure that the predetermined path is followed. Use PID controller to achieve precise speed and direction control.

[0106] State Estimation: The Extended Kalman Filter (EKF) is used to fuse multi-source information (such as IMU inertial measurement unit and wheel odometer) to provide a more accurate and reliable current position estimate.

[0107] Feedback Control: A closed-loop control system is established to continuously monitor the difference between the actual driving route and the planned route and make corrections in a timely manner. For example, when a deviation from the expected path is detected, the direction is immediately adjusted to get back on track.

[0108] In summary, the automatic obstacle avoidance method based on custom coordinate system navigation provided by the implementation of the present invention can adopt a coordinate system that better matches the specific task requirements within a small range, while reducing the amount of data processing, making path planning and obstacle avoidance algorithms easier to implement, and providing higher flexibility, lower computing costs and better accuracy under specific conditions.

[0109] like Figure 2 As shown, the automatic obstacle avoidance system based on custom coordinate system navigation provided by the present invention is provided, and the system comprises: A position acquisition unit 110, used to acquire global position information of the mobile unit in the global coordinate system; An area determination unit 120, configured to determine whether the mobile unit has entered a new scene internal area according to the global position information; The area detection unit 130 is used to obtain the spatial morphological characteristics of the internal area of ​​the scene newly entered by the mobile unit if yes; A type determination unit 140 is used to determine the area type of the newly entered scene internal area according to the acquired information; A coordinate system establishing unit 150, configured to establish a local coordinate system corresponding to a newly entered internal area of ​​the scene according to the area type; The path generation unit 160 is used to generate a local navigation path within the internal area of ​​the scene based on the established local coordinate system.

[0110] The automatic obstacle avoidance system based on custom coordinate system navigation provided by an embodiment of the present invention is used to implement the above-mentioned automatic obstacle avoidance method based on custom coordinate system navigation. Therefore, the specific implementation method is the same as the above-mentioned method and will not be repeated here.

[0111] In summary, the present invention provides an automatic obstacle avoidance method and system based on custom coordinate system navigation, which can adopt a coordinate system that better matches the specific task requirements within a small range, while reducing the amount of data processing, making path planning and obstacle avoidance algorithms easier to implement, and providing higher flexibility, lower computing costs and better accuracy under specific conditions.

[0112] In several embodiments disclosed in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and a module, a program segment or a part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0113] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0114] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

Claims

1. An automatic obstacle avoidance method based on custom coordinate system navigation, characterized in that: The method comprises: Get the global position information of the mobile unit in the global coordinate system; Determining whether the mobile unit has entered a new scene internal area according to the global position information; If yes, then obtain the spatial morphological features of the internal area of ​​the scene newly entered by the mobile unit; Determine the area type of the newly entered internal area of ​​the scene based on the acquired information; Establishing a local coordinate system corresponding to the newly entered internal area of ​​the scene according to the area type; A local navigation path within the internal area of ​​the scene is generated based on the established local coordinate system.

2. The automatic obstacle avoidance method based on custom coordinate system navigation according to claim 1, characterized in that: The step of determining the area type of the newly entered internal area of ​​the scene according to the acquired information specifically includes: Get the internal area classification standard of the building where the mobile unit is located; The internal area of ​​the scene is classified according to the spatial morphological features of the newly entered internal area of ​​the scene and the global position information to obtain a corresponding area type.

3. The automatic obstacle avoidance method based on custom coordinate system navigation according to claim 2, characterized in that: The step of establishing a local coordinate system corresponding to the newly entered internal area of ​​the scene according to the area type specifically includes: Determining the number of dimensions and obstacle avoidance accuracy of the corresponding local coordinate system according to the area type; A local coordinate system is established based on the number of dimensions and the obstacle avoidance accuracy.

4. The automatic obstacle avoidance method based on custom coordinate system navigation according to claim 3, characterized in that: Methods for establishing a local coordinate system include: Select a stable feature point in the newly entered internal area of ​​the scene as the origin; Determine the direction of the coordinate axis according to the behavior target of the mobile unit in the newly entered internal area of ​​the scene; Establishing a local coordinate system based on the origin and the coordinate axis directions; Determine the mapping relationship between the local coordinate system and the global coordinate system.

5. The automatic obstacle avoidance method based on custom coordinate system navigation according to claim 4, characterized in that: The method for determining the behavioral target includes: Determine whether the destination of the mobile unit is within the newly entered internal area of ​​the scene according to the global navigation path of the mobile unit in the global coordinate system; If not, the behavior goal is to pass through the internal area of ​​the scene; If yes, the behavior goal is to enter the internal area of ​​the scene.

6. The automatic obstacle avoidance method based on custom coordinate system navigation according to claim 5, characterized in that: The step of generating a local navigation path within the internal area of ​​the scene based on the established local coordinate system specifically includes: Determine the navigation end point of the mobile unit in the local coordinate system according to the behavior target; Obtain obstacle feature data corresponding to the current area type according to the area type; Perform obstacle detection within the internal area of ​​the scene based on the obstacle feature data; A local navigation path with a target as a navigation end point is generated according to the obstacle detection result and the obstacle avoidance accuracy.

7. The automatic obstacle avoidance method based on custom coordinate system navigation according to claim 6, characterized in that: The step of determining the navigation endpoint of the mobile unit in the local coordinate system according to the behavior target specifically includes: When the behavior goal is to pass through the internal area of ​​the scene, the navigation end point is the coordinates of the exit of the internal area of ​​the scene in the local coordinate system; When the behavior goal is to enter the internal area of ​​the scene, the navigation end point is the coordinates of the destination of the mobile unit in the local coordinate system.

8. The automatic obstacle avoidance method based on custom coordinate system navigation according to claim 6, characterized in that: The step of performing obstacle detection within the internal area of ​​the scene based on the obstacle feature data specifically includes: Obtain obstacle information within the scene through the sensors of the mobile units and perform feature extraction; Obtaining an obstacle classification model based on the extracted potential obstacle features and obstacle feature data; Obstacle detection is performed through obstacle classification models.

9. The automatic obstacle avoidance method based on custom coordinate system navigation according to claim 8, characterized in that: The step of generating a local navigation path with the target being the navigation end point according to the obstacle detection result and the obstacle avoidance accuracy specifically includes: Determine a corresponding path algorithm based on the obstacle avoidance accuracy; Determine obstacle avoidance rules for various obstacles in the local coordinate system based on the results of obstacle detection; A local navigation path to a navigation end point is generated based on the path algorithm and obstacle avoidance rules.

10. An automatic obstacle avoidance system based on custom coordinate system navigation, characterized in that: include: A position acquisition unit, used to acquire global position information of the mobile unit in the global coordinate system; An area determination unit, used to determine whether the mobile unit has entered a new scene internal area according to the global position information; an area detection unit, for obtaining, if yes, the spatial morphological features of the internal area of ​​the scene newly entered by the mobile unit; A type determination unit, used to determine the area type of the newly entered internal area of ​​the scene according to the acquired information; A coordinate system establishing unit, used to establish a local coordinate system corresponding to the newly entered internal area of ​​the scene according to the area type; The path generation unit is used to generate a local navigation path within the internal area of ​​the scene based on the established local coordinate system.

Images