Airplane avoidance method, device and equipment of automatic driving vehicle and medium

Abstract

The application discloses an airplane avoidance method and device of an automatic driving vehicle, equipment and a medium. The method comprises the following steps: when it is detected that the automatic driving vehicle enters a preset parking avoidance area, airplane detection is performed on a preset detection area to determine whether there is an airplane in the preset detection area; if it is determined that there is an airplane in the preset detection area, the moving direction and position information of the airplane are obtained; when it is determined that the airplane is in a taxiing state according to the moving direction and position information of the airplane, a parking avoidance signal is generated, and the automatic driving vehicle is controlled to park and avoid according to the parking avoidance signal. The technical scheme of the embodiment can realize effective avoidance of the airplane in the airport unmanned logistics scene, and can improve the safety of the airport unmanned logistics, by controlling the automatic driving vehicle to park and avoid when it is detected that there is an airplane in the set detection area and the airplane is in a taxiing state in the airport unmanned logistics scene.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a method, apparatus, device, and medium for autonomous vehicles to avoid aircraft collisions. Background Technology

[0002] Airports have a large demand for cargo transportation, and smart logistics systems can greatly improve cargo transportation efficiency. Therefore, adopting smart logistics systems in airports is significant for improving the efficiency of cargo handling.

[0003] Currently, smart logistics systems primarily utilize autonomous vehicles that, based on their onboard sensor and positioning modules, deliver goods along pre-planned routes. However, in the unique environment of airports, these autonomous vehicles may interfere with aircraft takeoffs and landings, posing safety hazards. Therefore, in airport unmanned logistics scenarios, how to control autonomous vehicles to safely avoid aircraft has become a pressing issue that needs to be addressed when applying smart logistics systems to airport settings. Summary of the Invention

[0004] This invention provides a method, device, equipment, and medium for autonomous vehicles to avoid aircraft, which can enable autonomous vehicles to safely avoid aircraft in unmanned logistics scenarios at airports, thereby improving the safety of unmanned logistics at airports and increasing logistics delivery efficiency.

[0005] According to one aspect of the present invention, an aircraft avoidance method for an autonomous vehicle is provided, comprising:

[0006] When an autonomous vehicle is detected to have entered a preset parking and avoidance area, an aircraft detection is performed on the preset detection area to determine whether an aircraft exists in the preset detection area.

[0007] If it is determined that an aircraft exists in the preset detection area, then the aircraft's direction of movement and position information are obtained;

[0008] When it is determined that the aircraft is in a taxiing state based on the aircraft's direction of movement and position information, a stop and avoidance signal is generated, and the autonomous vehicle is controlled to stop and avoid the obstacle based on the stop and avoidance signal.

[0009] According to another aspect of the present invention, an aircraft avoidance device for an autonomous vehicle is provided, comprising:

[0010] The aircraft detection module is used to detect aircraft in the preset detection area when an autonomous vehicle is detected to have entered the preset parking and avoidance area, so as to determine whether there is an aircraft in the preset detection area.

[0011] The aircraft information acquisition module is used to acquire the aircraft's movement direction and position information if it is determined that an aircraft exists in the preset detection area.

[0012] The parking avoidance control module is used to generate a parking avoidance signal when it is determined that the aircraft is in a taxiing state based on the aircraft's direction of movement and position information, and to control the autonomous vehicle to perform parking avoidance based on the parking avoidance signal.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the aircraft avoidance method for an autonomous vehicle according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the aircraft avoidance method for an autonomous vehicle according to any embodiment of the present invention.

[0018] The technical solution of this invention, when an autonomous vehicle is detected entering a preset stopping and yielding area, if an aircraft is detected in the preset detection area, obtains the aircraft's movement direction and position information. When it is determined that the aircraft is in a taxiing state based on the aircraft's movement direction and position information, a stopping and yielding signal is generated. Based on the stopping and yielding signal, the autonomous vehicle is controlled to stop and yield. By determining whether the aircraft is in a taxiing state based on the aircraft's movement direction and position information when the aircraft is successfully detected, and then determining whether to stop and yield, the safe yielding of autonomous vehicles to aircraft can be achieved, improving the safety of unmanned logistics at airports, while reducing the probability of ineffective yielding and improving logistics and distribution efficiency.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1A This is a flowchart of an aircraft avoidance method for an autonomous vehicle according to Embodiment 1 of the present invention;

[0022] Figure 1B This is a schematic diagram illustrating the positional relationship between a preset detection area and a preset parking avoidance area according to Embodiment 1 of the present invention;

[0023] Figure 1C This is a schematic diagram of an aircraft taxiing state detection according to Embodiment 1 of the present invention;

[0024] Figure 2 This is a flowchart of an aircraft avoidance method for an autonomous vehicle according to Embodiment 2 of the present invention;

[0025] Figure 3 This is a schematic diagram of the structure of an aircraft avoidance device for an autonomous vehicle according to Embodiment 3 of the present invention;

[0026] Figure 4 This is a schematic diagram of the structure of an electronic device that implements the aircraft avoidance method for autonomous vehicles according to embodiments of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," "target," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] Example 1

[0030] Figure 1A This is a flowchart illustrating an aircraft avoidance method for an autonomous vehicle, as provided in Embodiment 1 of the present invention. This embodiment is applicable to unmanned logistics delivery at airports. The method can be executed by an aircraft avoidance device of the autonomous vehicle, which can be implemented in hardware and / or software and can be configured in an electronic device. For example... Figure 1A As shown, the method includes:

[0031] S110. When an autonomous vehicle is detected to have entered a preset parking and avoidance area, an aircraft detection is performed on the preset detection area to determine whether an aircraft exists in the preset detection area.

[0032] The autonomous vehicle can be a driverless vehicle used for logistics transportation. In this embodiment, the autonomous vehicle can drive automatically according to a pre-planned path. Optionally, the autonomous vehicle can be equipped with a positioning module, such as a BeiDou satellite navigation system module or a Global Positioning System module, to obtain the real-time location information of the autonomous vehicle. In addition, the autonomous vehicle can also be equipped with a perception module, such as LiDAR or a depth camera, to perceive the vehicle's environmental information in real time.

[0033] The preset parking avoidance area can be a pre-defined area within the planned path for parking and waiting. In a specific example, the preset parking avoidance area can be defined by setting a virtual stop line and a distance. Specifically, a position at a set distance from the virtual stop line can be used as a virtual starting line, and the preset parking avoidance area can be determined based on the virtual starting line, the virtual stop line, and the left and right boundaries of the planned path.

[0034] In this embodiment, when the autonomous vehicle is detected to be entering the preset parking avoidance area based on its location information and the location information of the preset parking avoidance area, aircraft detection can be performed on the preset detection area. The preset detection area can be a pre-defined aircraft detection area within an airport, such as a runway or apron where aircraft may be operating. Specifically, an image of the preset detection area can be acquired through a pre-configured perception module, and aircraft detection can be performed on this image.

[0035] In a specific example, the positional relationship between the preset detection area and the preset parking avoidance area can be as follows: Figure 1B As shown in the diagram, the preset parking avoidance area is determined by a virtual stop line and a set distance, and is part of the planned path of the autonomous vehicle, while the preset detection area is located on one side of the planned path. It is understandable that the preset detection area and the preset parking avoidance area can be adaptively set according to the actual airport scenario.

[0036] Specifically, when detecting aircraft in a preset detection area, a 3D point cloud image corresponding to the preset detection area can first be acquired through a pre-configured perception module. Then, a pre-set aircraft detection method can be used to determine whether an aircraft exists in the preset detection area based on the 3D point cloud image. For example, the 3D point cloud image corresponding to the preset detection area can be input into a pre-trained aircraft detection model, and the presence of an aircraft in the preset detection area can be determined based on the confidence level of the aircraft detection result output by the aircraft detection model. Alternatively, the 3D point cloud image corresponding to the preset detection area can be converted into a 2D grid image, and the 2D grid image can be detected using relative height and absolute height methods, thereby determining whether an aircraft exists in the preset detection area based on the detection result.

[0037] It is worth noting that after the autonomous vehicle begins to enter the pre-designated stopping and avoidance zone, if the autonomous vehicle's perception module can only detect a portion of the pre-designated detection area, then aircraft detection can be performed on the portion of the pre-designated detection area covered by the perception module to determine the presence of an aircraft. The portion of the pre-designated detection area that the perception module cannot cover is considered not to affect the autonomous vehicle's operation.

[0038] S120. If it is determined that an aircraft exists in the preset detection area, then the aircraft's movement direction and position information are obtained.

[0039] The aircraft's direction of movement and position information can refer to the aircraft's direction of movement and position coordinates within the current autonomous vehicle coordinate system. The autonomous vehicle coordinate system can be a Cartesian coordinate system established with a point on the autonomous vehicle as the origin and the vehicle's heading as the y-axis. It is understandable that as the autonomous vehicle moves, the transformation relationship between the autonomous vehicle coordinate system and the world coordinate system is continuously updated.

[0040] Specifically, after confirming the presence of an aircraft within the preset detection area, the aircraft can be continuously tracked to obtain its position coordinates at different times in the world coordinate system. Then, based on the aircraft's position coordinates at different times, a direction vector from the initial position coordinates to the final position coordinates can be obtained. Further, after obtaining the aircraft's position coordinates at different times, the current vehicle coordinate system can be obtained, and the counterclockwise angle between the aforementioned direction vector and the positive x-axis of the current vehicle coordinate system can be determined as the aircraft's direction of movement in the current vehicle coordinate system. Additionally, coordinate system transformation can be performed on the aircraft's position coordinates at different times in the world coordinate system to obtain the transformed position coordinates of each position coordinate in the current vehicle coordinate system.

[0041] S130. When it is determined that the aircraft is in a taxiing state based on the aircraft's direction of movement and position information, a stop and avoidance signal is generated, and the autonomous vehicle is controlled to stop and avoid the obstacle based on the stop and avoidance signal.

[0042] In this embodiment, the aircraft's taxiing state can include a taxiing-in state and a taxiing-out state. The taxiing-in state can be an aircraft taxiing state in which the taxiing direction is towards the planned path of the autonomous vehicle; correspondingly, the taxiing-out state can be an aircraft taxiing state in which the taxiing direction is away from the planned path of the autonomous vehicle.

[0043] Specifically, when the parallel distance between the aircraft and the y-axis of the current vehicle coordinate system gradually decreases based on the aircraft's position coordinates at different times in the current vehicle coordinate system, it can be determined that the aircraft is in a taxiing state; or, when the aircraft's position information is determined to be within a set position range and the aircraft's direction of movement is within a set angle range, it can be determined that the aircraft is in a taxiing state.

[0044] The stop and avoidance signal can be a control signal for the driving device of an autonomous vehicle, used to control the autonomous vehicle to stop. In this embodiment, when it is determined that an aircraft is taxiing in a preset detection area, a stop and avoidance signal can be generated to control the autonomous vehicle to stop and avoid the aircraft.

[0045] In this embodiment, after determining that an aircraft exists in the preset detection area, it is further determined that the aircraft is in a taxiing state before controlling the autonomous vehicle to stop and avoid it. This can reduce the probability of ineffective avoidance and improve logistics and delivery efficiency.

[0046] The technical solution of this invention, when an autonomous vehicle is detected entering a preset stopping and yielding area, if an aircraft is detected in the preset detection area, obtains the aircraft's movement direction and position information. When it is determined that the aircraft is in a taxiing state based on the aircraft's movement direction and position information, a stopping and yielding signal is generated. Based on the stopping and yielding signal, the autonomous vehicle is controlled to stop and yield. By determining whether the aircraft is in a taxiing state based on the aircraft's movement direction and position information when the aircraft is successfully detected, and then determining whether to stop and yield, the safe yielding of autonomous vehicles to aircraft can be achieved, improving the safety of unmanned logistics at airports, while reducing the probability of ineffective yielding and improving logistics and distribution efficiency.

[0047] In an optional implementation of this embodiment, determining that the aircraft is in a taxiing state based on the aircraft's movement direction and position information may include: when it is detected that the aircraft's position information belongs to a preset quadrant of the vehicle coordinate system and the aircraft's movement direction belongs to a preset angle range of the vehicle coordinate system, the aircraft is determined to be in a taxiing state.

[0048] Among them, the preset quadrant can be a pre-set quadrant of the vehicle coordinate system; the preset angle range can be a pre-set angle range used to determine whether the aircraft is in a taxiing state.

[0049] In a specific example, the detection of the aircraft's taxiing state can be as follows: Figure 1C As shown in the diagram. The preset quadrants can be the first and fourth quadrants of the vehicle coordinate system, and the preset angle range can be 90 to 270 degrees. Therefore, when the aircraft's position coordinates at different times are detected to belong to the first or fourth quadrant of the vehicle coordinate system, and the aircraft's direction of movement is within the range of 90 to 270 degrees, it can be determined that the aircraft is in a taxiing state.

[0050] Furthermore, the preset quadrants can also be the second and third quadrants of the vehicle coordinate system, and the corresponding preset angle ranges can be 0 to 90 degrees and 270 to 360 degrees. In this case, when the aircraft's position coordinates at different times are detected to belong to the second or third quadrant of the vehicle coordinate system, and the aircraft's direction of movement is within the range of 0 to 90 degrees or 270 to 360 degrees, it can be determined that the aircraft is in a taxiing state.

[0051] Optionally, when the aircraft's position information is detected to be within a preset quadrant of the vehicle coordinate system, and the aircraft's direction of movement is not within a preset angle range of the vehicle coordinate system, it can be determined that the aircraft is in a taxiing state.

[0052] The advantage of the above settings is that they enable accurate judgment of the aircraft's taxiing status, thereby improving the accuracy of controlling autonomous vehicles to stop and avoid obstacles, and enhancing the safety of unmanned logistics at airports.

[0053] In another optional embodiment of this example, after controlling the autonomous vehicle to stop and avoid a collision based on the stop avoidance signal, the method may further include:

[0054] If, within a preset time threshold, it is determined that there is no aircraft in the preset detection area, or if it is determined that the aircraft is in a taxiing state based on the aircraft's movement direction and position information, a passage signal is generated; based on the passage signal, the autonomous vehicle is controlled to pass through the preset parking and avoidance area.

[0055] The preset time threshold can be a pre-defined time length, such as 1 minute.

[0056] In this embodiment, after controlling the autonomous vehicle to stop and avoid an obstacle, continuous aircraft detection can be performed on the preset detection area. If no aircraft is detected in the preset detection area within a preset time threshold, or if all detected aircraft are in a taxiing state, a passage signal can be generated to control the autonomous vehicle to pass through the preset stopping and avoiding area.

[0057] Example 2

[0058] Figure 2 This is a flowchart illustrating an aircraft avoidance method for an autonomous vehicle according to Embodiment 2 of the present invention. This embodiment is a further refinement of the above technical solution, and the technical solution in this embodiment can be combined with one or more of the above implementation methods. For example... Figure 2 As shown, the method includes:

[0059] S210, Start, and execute S220.

[0060] S220. When an autonomous vehicle is detected to have entered a preset parking avoidance area, a three-dimensional point cloud corresponding to the preset detection area is obtained, and the three-dimensional point cloud corresponding to the preset detection area is processed into a two-dimensional grid to obtain a two-dimensional grid image corresponding to the preset detection area, and then S230 is executed.

[0061] The two-dimensional grid image may include at least one grid, and each grid may include at least one projection point. A projection point is a two-dimensional point obtained by dimensionality reduction of a three-dimensional point. In this embodiment, the size of the grid can be adaptively set according to accuracy requirements.

[0062] In this embodiment, when an autonomous vehicle is detected to have entered a preset parking avoidance area, a 3D point cloud corresponding to the preset detection area can be obtained through a pre-configured perception module. Then, each 3D point can be dimensionality-reduced to project the 3D point cloud onto a rectangular 2D grid image in a vertical plane, thereby obtaining the 2D grid image corresponding to the preset detection area.

[0063] S230: Obtain the height corresponding to each projection point, and based on the height corresponding to each projection point, obtain the relative height corresponding to each grid, and execute S240.

[0064] Specifically, after obtaining the two-dimensional grid image corresponding to the preset detection area, the height of each projection point in the two-dimensional grid image can be obtained. This height can be relative to the height of the autonomous vehicle. Then, based on the height of each projection point, the maximum and minimum heights of each projection point in each grid cell can be obtained. Furthermore, the height difference between the maximum and minimum heights can be calculated to obtain the relative height of each grid cell.

[0065] S240. If the relative height of a certain grid cell is detected to be greater than or equal to a preset relative height threshold, the detected grid cell is determined as a candidate grid cell, and S250 is executed.

[0066] Specifically, the relative height of each grid cell is compared with a pre-set relative height threshold. If the relative height of the current grid cell is greater than or equal to the relative height threshold, the current grid cell can be identified as a candidate grid cell. If the relative height of the current grid cell is less than the relative height threshold, the next grid cell is detected until all grid cells are detected.

[0067] S250: Based on the height of each projection point in each candidate grid, obtain the number of target projection points in each candidate grid whose corresponding height is greater than a preset height threshold, and execute S260.

[0068] Specifically, after the candidate grids are selected, the height of each projection point in each candidate grid is compared with a pre-set height threshold. If the height of the projection point is greater than or equal to the height threshold, the projection point is determined as the target projection point, thereby obtaining the number of target projection points in each candidate grid.

[0069] S260. If the number of target projection points in a candidate grid is greater than or equal to a preset number threshold, the detected candidate grid is determined as the target grid, and S270 is executed.

[0070] Furthermore, it is determined whether the number of target projection points in the current candidate grid is greater than or equal to a preset threshold. If so, the current candidate grid can be identified as the target grid; otherwise, it can continue to determine whether the number of target projection points in the next candidate grid is greater than or equal to the threshold, until the detection of all candidate grids is completed.

[0071] S270. Perform clustering processing on each of the target grids to obtain at least one grid cluster, and determine whether there is an aircraft in the preset detection area based on each grid cluster.

[0072] Specifically, after obtaining all target rasters, a preset clustering method can be used to cluster them to obtain multiple raster clusters. These preset clustering methods can include k-means clustering, fuzzy C-means (FCM) clustering, etc. Then, the obtained raster clusters can be further filtered based on information such as cluster area and maximum cluster height to determine if there is a raster cluster corresponding to an aircraft, thereby determining whether an aircraft exists within the preset detection area.

[0073] If it is determined that an aircraft exists in the preset detection area, then S280 is executed; if it is determined that no aircraft exists in the preset detection area, then S2100 is executed.

[0074] Optionally, after obtaining the target projection points in each candidate grid, clustering can be performed on each target projection point to obtain multiple projection point clusters. Then, each projection point cluster can be filtered based on the pre-set cluster area and maximum cluster height to determine whether there is a projection point cluster corresponding to the aircraft, thereby determining whether an aircraft exists in the preset detection area.

[0075] S280. Obtain the aircraft's direction of movement and position information, and determine whether the aircraft is in a taxiing state based on the aircraft's direction of movement and position information.

[0076] If the aircraft is determined to be in a taxiing state based on its direction of movement and position information, then S290 is executed; if the aircraft is determined not to be in a taxiing state based on its direction of movement and position information, then S2100 is executed.

[0077] S290, Generate a parking avoidance signal, and control the autonomous vehicle to stop and avoid the obstacle according to the parking avoidance signal, and execute S2110.

[0078] Optionally, after controlling the autonomous vehicle to stop and avoid an obstacle, the system can continuously detect aircraft in a preset detection area within a preset time threshold. When an aircraft is detected in the preset detection area but is not present, or if an aircraft is present but is taxiing out, a passage signal can be generated to control the autonomous vehicle to pass through the preset stopping and avoiding area.

[0079] S2100: Generate a passage signal, and control the autonomous vehicle to pass through the preset parking avoidance area according to the passage signal, and execute S2110.

[0080] S2110, End.

[0081] The technical solution of this invention involves obtaining a 3D point cloud corresponding to a preset detection area when an autonomous vehicle is detected entering a preset parking avoidance area. This 3D point cloud is then processed into a 2D grid to obtain a 2D grid image of the preset detection area. Next, the height of each projection point is obtained, and the relative height of each grid is calculated based on the height of each projection point. If the relative height of a grid is detected to be greater than or equal to a preset relative height threshold, the detected grid is identified as a candidate grid, and the number of target projection points in each candidate grid whose height is greater than the preset height threshold is obtained. If a certain... If the number of target projection points in a candidate grid is greater than or equal to a preset threshold, the detected candidate grid is determined as the target grid. Further, each target grid is clustered to obtain multiple grid clusters, and the presence of an aircraft in a preset detection area is determined based on each grid cluster. By using preset relative height and height thresholds to filter each grid in the two-dimensional grid image, and then clustering the filtered grids, aircraft detection can be performed based on each grid cluster. This improves the accuracy of aircraft detection, thereby improving the accuracy of controlling autonomous vehicles to stop and avoid obstacles, and further enhancing the safety of unmanned logistics at airports.

[0082] In an optional implementation of this embodiment, determining whether an aircraft exists in the preset detection area based on each of the grid clusters may include: obtaining the cluster area and maximum height corresponding to each of the grid clusters; if it is detected that the cluster area corresponding to a target grid cluster is greater than a preset cluster area threshold and the corresponding maximum height is greater than a preset maximum height threshold, then determining whether an aircraft exists in the preset detection area based on the three-dimensional point cloud corresponding to the preset detection area and the target grid cluster.

[0083] In this embodiment, after obtaining the grid clusters, the cluster area and maximum height corresponding to each grid cluster can be obtained. Then, the cluster area corresponding to each grid cluster can be compared with a preset cluster area threshold, and the corresponding maximum height can be compared with a preset maximum height threshold. If it is detected that there is a target grid cluster whose cluster area is greater than the preset cluster area threshold and whose corresponding maximum height is greater than the preset maximum height threshold, then based on the 3D point cloud corresponding to the preset detection area, a partial point cloud that may correspond to the aircraft can be obtained. Based on the partial point cloud that may correspond to the aircraft and the target grid cluster, it can be jointly determined whether an aircraft exists in the preset detection area.

[0084] In another optional implementation of this embodiment, determining whether an aircraft exists in the preset detection area based on the clustering of the 3D point cloud corresponding to the preset detection area and the target grid may include:

[0085] The three-dimensional point cloud corresponding to the preset detection area is input into the pre-trained aircraft detection model to obtain at least one candidate three-dimensional point cloud output by the aircraft detection model; the target grid cluster and each candidate three-dimensional point cloud are fused; if it is detected that the target three-dimensional point cloud and the target grid cluster are successfully fused in each candidate three-dimensional point cloud, then it is determined that there is an aircraft in the preset detection area.

[0086] The aircraft detection model can be built based on a convolutional neural network (CNN) method. The input is a 3D point cloud of the region to be detected, and the output is a partial 3D point cloud of the aircraft. In this embodiment, an initial aircraft detection model can first be built based on a CNN method. Then, labeled sample 3D point clouds can be used to perform supervised training on this initial aircraft detection model to obtain a trained aircraft detection model.

[0087] In this embodiment, the 3D point cloud corresponding to the preset detection area is first input into the trained aircraft detection model, and one or more candidate 3D point clouds corresponding to the aircraft are obtained from the output of the aircraft detection model. Then, the target raster cluster is fused with each candidate 3D point cloud to determine whether there is a target 3D point cloud that can be successfully fused with the target raster cluster. Finally, if the target 3D point cloud can be successfully detected, it can be determined that an aircraft exists in the preset detection area.

[0088] Specifically, when determining whether there exists a target 3D point cloud that can be successfully fused with the target grid cluster, the 2D projection corresponding to the current candidate 3D point cloud can be obtained, and it can be determined whether the 2D projection coincides with the target grid cluster. If it is determined that the 2D projection successfully coincides with the target grid cluster, the current candidate 3D point cloud can be determined as the target 3D point cloud. If it is determined that the 2D projection does not successfully coincide with the target grid cluster, the current candidate 3D point cloud can be determined as not being the target 3D point cloud.

[0089] The advantage of the above settings is that they can further improve the accuracy of aircraft detection, thereby further improving the accuracy of controlling autonomous vehicles to stop and avoid obstacles.

[0090] Optionally, determining whether the 2D projection overlaps with the target raster cluster can include: determining whether there is an intersection between the 2D projection corresponding to the current candidate 3D point cloud and the target raster cluster; if so, obtaining the ratio between the area of ​​the intersection region and the area of ​​the 2D projection; if the ratio is detected to be greater than or equal to a preset ratio threshold, it can be determined that the 2D projection overlaps with the target raster cluster; and if the ratio is detected to be less than the preset ratio threshold, it can be determined that the 2D projection does not overlap with the target raster cluster.

[0091] In another optional implementation of this embodiment, obtaining the aircraft's movement direction and position information may include: obtaining a 3D detection box corresponding to the target 3D point cloud, and obtaining the movement direction and position information of the 3D detection box in the vehicle coordinate system; obtaining the aircraft's movement direction and position information based on the movement direction and position information of the 3D detection box in the vehicle coordinate system.

[0092] In this embodiment, after successfully detecting the target 3D point cloud, a 3D detection box corresponding to the target 3D point cloud can be generated, and the position information of the 3D detection box in the world coordinate system (e.g., the position coordinates of each vertex) can be obtained. Then, in each frame of the 3D point cloud image, the 3D detection box corresponding to the target 3D point cloud, and the position information of the 3D detection box, can be obtained. Then, based on the position coordinates of the same point (e.g., a vertex, centroid, etc.) of the 3D detection box in each frame of the 3D point cloud image, the movement direction of the 3D detection box in the vehicle coordinate system can be obtained.

[0093] Furthermore, the position information of each 3D detection box in the world coordinate system can be converted to its position information in the vehicle coordinate system using the coordinate system transformation matrix between the world coordinate system and the vehicle coordinate system. Finally, the movement direction and position information of the 3D detection boxes in the vehicle coordinate system can be used as the movement direction and position information of the aircraft.

[0094] Example 3

[0095] Figure 3 This is a schematic diagram of an aircraft avoidance device for an autonomous vehicle according to Embodiment 3 of the present invention. Figure 3 As shown, the device includes: an aircraft detection module 310, an aircraft information acquisition module 320, and a parking avoidance control module 330; wherein,

[0096] The aircraft detection module 310 is used to detect aircraft in the preset detection area when an autonomous vehicle is detected to have entered the preset parking and avoidance area, so as to determine whether there is an aircraft in the preset detection area.

[0097] The aircraft information acquisition module 320 is used to acquire the aircraft's movement direction and position information if it is determined that an aircraft exists in the preset detection area.

[0098] The parking avoidance control module 330 is used to generate a parking avoidance signal when it is determined that the aircraft is in a taxiing state based on the aircraft's direction of movement and position information, and to control the autonomous vehicle to perform parking avoidance based on the parking avoidance signal.

[0099] The technical solution of this invention, when an autonomous vehicle is detected entering a preset stopping and yielding area, if an aircraft is detected in the preset detection area, obtains the aircraft's movement direction and position information. When it is determined that the aircraft is in a taxiing state based on the aircraft's movement direction and position information, a stopping and yielding signal is generated. Based on the stopping and yielding signal, the autonomous vehicle is controlled to stop and yield. By determining whether the aircraft is in a taxiing state based on the aircraft's movement direction and position information when the aircraft is successfully detected, and then determining whether to stop and yield, the safe yielding of autonomous vehicles to aircraft can be achieved, improving the safety of unmanned logistics at airports, while reducing the probability of ineffective yielding and improving logistics and distribution efficiency.

[0100] Optionally, the aircraft avoidance device for the autonomous vehicle may further include:

[0101] The passage signal generation module is used to generate a passage signal if, within a preset time threshold, it is determined that there is no aircraft in the preset detection area, or if it is determined that the aircraft is in a taxiing state based on the aircraft's movement direction and position information.

[0102] The vehicle control module is used to control the autonomous vehicle to pass through the preset parking and avoidance area according to the traffic signal.

[0103] Optional, the aircraft detection module 310 includes:

[0104] A two-dimensional grid image acquisition unit is used to acquire a three-dimensional point cloud corresponding to the preset detection area, and to perform two-dimensional grid processing on the three-dimensional point cloud corresponding to the preset detection area to obtain a two-dimensional grid image corresponding to the preset detection area.

[0105] The two-dimensional grid image includes at least one grid, and each grid includes at least one projection point;

[0106] A relative height acquisition unit is used to acquire the height corresponding to each of the projection points, and to acquire the relative height corresponding to each of the grids based on the height corresponding to each of the projection points.

[0107] The candidate grid determination unit is used to determine the detected grid as a candidate grid if the relative height of a certain grid is greater than or equal to a preset relative height threshold.

[0108] The target projection point number acquisition unit is used to acquire the number of target projection points in each candidate grid whose corresponding height is greater than a preset height threshold, based on the height of each projection point in each candidate grid.

[0109] The target grid determination unit is used to determine the detected candidate grid as the target grid if the number of target projection points in a candidate grid is greater than or equal to a preset number threshold.

[0110] An aircraft detection unit is used to perform clustering processing on each of the target grids to obtain at least one grid cluster, and to determine whether an aircraft exists in the preset detection area based on each grid cluster.

[0111] Optional, the aircraft detection unit includes:

[0112] The grid cluster area acquisition subunit is used to acquire the cluster area and maximum height corresponding to each grid cluster;

[0113] The aircraft determination subunit is used to determine whether an aircraft exists in the preset detection area based on the 3D point cloud corresponding to the preset detection area and the target grid cluster if the cluster area corresponding to the target grid cluster is greater than a preset cluster area threshold and the corresponding maximum height is greater than a preset maximum height threshold.

[0114] Optionally, the aircraft judgment subunit is specifically used to input the three-dimensional point cloud corresponding to the preset detection area into the pre-trained aircraft detection model to obtain at least one candidate three-dimensional point cloud output by the aircraft detection model.

[0115] The target grid cluster and each candidate 3D point cloud are fused. If it is detected that the target 3D point cloud and the target grid cluster are successfully fused in each candidate 3D point cloud, then it is determined that there is an aircraft in the preset detection area.

[0116] Optionally, the aircraft information acquisition module 320 includes:

[0117] The three-dimensional detection box acquisition unit is used to acquire the three-dimensional detection box corresponding to the target three-dimensional point cloud, and to acquire the movement direction and position information of the three-dimensional detection box in the vehicle coordinate system;

[0118] The aircraft information acquisition unit is used to acquire the aircraft's movement direction and position information based on the movement direction and position information of the three-dimensional detection box in the vehicle coordinate system.

[0119] Optionally, the parking avoidance control module 330 includes:

[0120] The aircraft taxiing state determination unit is used to determine that the aircraft is in a taxiing state when it detects that the aircraft's position information belongs to a preset quadrant of the vehicle coordinate system and the aircraft's direction of movement belongs to a preset angle range of the vehicle coordinate system.

[0121] The aircraft avoidance device for autonomous vehicles provided in the embodiments of the present invention can execute the aircraft avoidance method for autonomous vehicles provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0122] Example 4

[0123] Figure 4 A schematic diagram of an electronic device 40 that can be used to implement embodiments 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 components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0124] like Figure 4As shown, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory (ROM) 42 or a random access memory (RAM) 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the ROM 42 or loaded into the RAM 43 from storage unit 48. The RAM 43 may also store various programs and data required for the operation of the electronic device 40. The processor 41, ROM 42, and RAM 43 are interconnected via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0125] Multiple components in electronic device 40 are connected to I / O interface 45, including: input unit 46; output unit 47, such as various types of displays, speakers, etc.; storage unit 48, such as disks, optical disks, etc.; and communication unit 49, such as network cards, modems, wireless transceivers, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0126] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as aircraft avoidance methods for autonomous vehicles.

[0127] In some embodiments, the aircraft avoidance method for an autonomous vehicle may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 40 via ROM 42 and / or communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the aircraft avoidance method for an autonomous vehicle described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the aircraft avoidance method for an autonomous vehicle by any other suitable means (e.g., by means of firmware).

[0128] Various embodiments 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-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0129] Computer programs used to implement 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, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0130] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0131] To provide interaction with a user, the systems and techniques described herein can 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 pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0132] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0133] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0134] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0135] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for autonomous vehicles to avoid collisions with aircraft, characterized in that, include: When an autonomous vehicle is detected to have entered a preset parking avoidance area, a three-dimensional point cloud corresponding to the preset detection area is obtained, and the three-dimensional point cloud corresponding to the preset detection area is processed into a two-dimensional grid to obtain a two-dimensional grid image corresponding to the preset detection area. The two-dimensional grid image includes at least one grid, and each grid includes at least one projection point; Obtain the height corresponding to each projection point, and based on the height corresponding to each projection point, obtain the maximum height and minimum height corresponding to each projection point in each grid, and obtain the relative height corresponding to each grid by calculating the height difference between the maximum height and the minimum height; If the relative height of a certain grid cell is detected to be greater than or equal to a preset relative height threshold, the detected grid cell is determined as a candidate grid cell. Based on the height of each projection point in each candidate grid, obtain the number of target projection points in each candidate grid whose corresponding height is greater than a preset height threshold. If the number of target projection points in a candidate grid is greater than or equal to a preset threshold, the detected candidate grid is determined as the target grid. Clustering is performed on each of the target grids to obtain at least one grid cluster, and the cluster area and maximum height corresponding to each grid cluster are obtained; If it is detected that the cluster area corresponding to a target grid cluster is greater than a preset cluster area threshold and the corresponding maximum height is greater than a preset maximum height threshold, then the three-dimensional point cloud corresponding to the preset detection area is input into the pre-trained aircraft detection model to obtain at least one candidate three-dimensional point cloud output by the aircraft detection model. Obtain the two-dimensional projection corresponding to the current candidate 3D point cloud, and determine whether there is an intersection between the two-dimensional projection and the target grid cluster. If so, obtain the ratio between the area of ​​the intersection region and the area of ​​the two-dimensional projection. If the detected ratio is greater than or equal to a preset ratio threshold, it is determined that the two-dimensional projection and the target grid clustering have successfully overlapped, and the current candidate three-dimensional point cloud is determined as the target three-dimensional point cloud, and it is determined that an aircraft exists in the preset detection area; If it is determined that an aircraft exists in the preset detection area, then the aircraft's direction of movement and position information are obtained; When it is determined that the aircraft is in a taxiing state based on the aircraft's direction of movement and position information, a stop and avoidance signal is generated, and the autonomous vehicle is controlled to stop and avoid the obstacle based on the stop and avoidance signal.

2. The method according to claim 1, characterized in that, After controlling the autonomous vehicle to stop and avoid a collision based on the stop and avoidance signal, the method further includes: If, within a preset time threshold, it is determined that there is no aircraft in the preset detection area, or if it is determined that the aircraft is in a taxiing state based on the aircraft's movement direction and position information, a passage signal is generated. Based on the traffic signal, the autonomous vehicle is controlled to pass through the preset parking and avoidance area.

3. The method according to claim 1, characterized in that, Obtaining the aircraft's direction of movement and position information includes: Obtain the 3D detection box corresponding to the target 3D point cloud, and obtain the movement direction and position information of the 3D detection box in the vehicle coordinate system; Based on the movement direction and position information of the three-dimensional detection box in the vehicle coordinate system, the movement direction and position information of the aircraft are obtained.

4. The method according to claim 1, characterized in that, Determining that the aircraft is in a taxiing state based on its direction of movement and position information includes: When the aircraft's position information is detected to be within a preset quadrant of the vehicle coordinate system, and the aircraft's direction of movement is within a preset angle range of the vehicle coordinate system, the aircraft is determined to be in a taxiing state.

5. An aircraft avoidance device for an autonomous vehicle, characterized in that, include: The aircraft detection module is used to acquire the three-dimensional point cloud corresponding to the preset detection area when an autonomous vehicle is detected to have entered the preset parking and avoidance area, and to perform two-dimensional gridding processing on the three-dimensional point cloud corresponding to the preset detection area to obtain the two-dimensional grid image corresponding to the preset detection area. The two-dimensional grid image includes at least one grid, and each grid includes at least one projection point; Obtain the height corresponding to each projection point, and based on the height corresponding to each projection point, obtain the maximum height and minimum height corresponding to each projection point in each grid, and obtain the relative height corresponding to each grid by calculating the height difference between the maximum height and the minimum height; If the relative height of a certain grid cell is detected to be greater than or equal to a preset relative height threshold, the detected grid cell is determined as a candidate grid cell. Based on the height of each projection point in each candidate grid, obtain the number of target projection points in each candidate grid whose corresponding height is greater than a preset height threshold. If the number of target projection points in a candidate grid is greater than or equal to a preset threshold, the detected candidate grid is determined as the target grid. Clustering is performed on each of the target grids to obtain at least one grid cluster, and the cluster area and maximum height corresponding to each grid cluster are obtained; If it is detected that the cluster area corresponding to a target grid cluster is greater than a preset cluster area threshold and the corresponding maximum height is greater than a preset maximum height threshold, then the three-dimensional point cloud corresponding to the preset detection area is input into the pre-trained aircraft detection model to obtain at least one candidate three-dimensional point cloud output by the aircraft detection model. Obtain the two-dimensional projection corresponding to the current candidate 3D point cloud, and determine whether there is an intersection between the two-dimensional projection and the target grid cluster. If so, obtain the ratio between the area of ​​the intersection region and the area of ​​the two-dimensional projection. If the detected ratio is greater than or equal to a preset ratio threshold, it is determined that the two-dimensional projection and the target grid clustering have successfully overlapped, and the current candidate three-dimensional point cloud is determined as the target three-dimensional point cloud, and it is determined that an aircraft exists in the preset detection area; The aircraft information acquisition module is used to acquire the aircraft's movement direction and position information if it is determined that an aircraft exists in the preset detection area. The parking avoidance control module is used to generate a parking avoidance signal when it is determined that the aircraft is in a taxiing state based on the aircraft's direction of movement and position information, and to control the autonomous vehicle to perform parking avoidance based on the parking avoidance signal.

6. An electronic device, characterized in that, The electronic device includes: 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 to enable the at least one processor to perform the aircraft avoidance method for an autonomous vehicle according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the aircraft avoidance method for an autonomous vehicle according to any one of claims 1-4.

Images