Vehicle anti-collision method and device based on unmanned aerial vehicle, electronic equipment and medium

Abstract

The embodiment of the invention provides a vehicle anti-collision method and device based on an unmanned aerial vehicle, electronic equipment and a medium, and the method comprises the steps: determining the current working information of the unmanned aerial vehicle and a cabin according to the unmanned aerial vehicle type of the unmanned aerial vehicle and the cabin type of the cabin in the running process of a target vehicle; under the current working information, information in the view field is processed based on the unmanned aerial vehicle, so that obstacle information of the environment to which the target vehicle belongs is determined; when the obstacle information meets a preset condition, an identification result corresponding to the obstacle information is sent to a processing module, so that the processing module determines target control information corresponding to the target vehicle based on the identification result; wherein the target control information at least comprises a deceleration instruction and / or a braking instruction. According to the technical scheme, the advantage of the installation position of the unmanned aerial vehicle is utilized, the aerial target recognition range is expanded, the blind area above the rear portion of the vehicle is effectively eliminated, and the precise recognition capacity of small and non-standard targets is improved.

Description

Technical Field

[0001] This invention relates to the fields of automotive safety and intelligent driving technology, and in particular to vehicle collision avoidance methods, devices, electronic devices and media based on unmanned aerial vehicles (UAVs). Background Technology

[0002] With the increasing popularity of drone technology, many vehicle models are equipped with drone systems to meet users' needs for travel recording. The drones can be housed in a drone cabin on the vehicle's roof and, upon receiving instructions, can accompany the vehicle to scout and film. However, these vehicles generally have higher suspensions and larger bodies, resulting in a significantly greater vehicle height than ordinary vehicles, potentially leading to collisions with aerial obstacles such as height restriction barriers, pipes, and signs.

[0003] Currently, most vehicles are equipped with features such as Autonomous Emergency Braking (AEB), side collision warning, parking radar warning, automatic parking, and 360° AEB. However, these features typically focus on obstacles such as pillars, walls, and adjacent vehicles. Due to limitations in sensor installation angles and recognition algorithms, the system has blind spots in detecting obstacles behind the vehicle, especially in the space above and behind it, and its ability to recognize small and non-standard targets such as height restriction poles and tree branches is insufficient. Summary of the Invention

[0004] This invention provides a vehicle collision avoidance method, device, electronic device, and medium based on unmanned aerial vehicles (UAVs), which can utilize the advantages of the UAV's installation location to expand the range of aerial target recognition, effectively eliminate the blind spot above and behind the vehicle, and increase the accuracy of identifying small and non-standard targets.

[0005] In a first aspect, embodiments of the present invention provide a vehicle collision avoidance method based on an unmanned aerial vehicle (UAV). The target vehicle has an UAV deployed on its roof and a cabin for use with the UAV. The cabin can be used to charge the UAV. The UAV communicates with a processing module in the target vehicle to send collected and / or processed data to the processing module, including:

[0006] During the target vehicle's movement, the current operational information of the drone and its cabin is determined based on the drone type and cabin type.

[0007] Based on the current working information, the drone processes the information within its field of vision to determine the obstacle information of the target vehicle's environment;

[0008] When the obstacle information meets the preset conditions, the recognition result corresponding to the obstacle information is sent to the processing module so that the processing module can determine the target control information corresponding to the target vehicle based on the recognition result.

[0009] The target control information includes at least deceleration commands and / or braking commands.

[0010] Secondly, embodiments of the present invention also provide a vehicle collision avoidance device based on an unmanned aerial vehicle (UAV), the device comprising:

[0011] The current working information determination module is used to determine the current working information of the drone and the cabin based on the drone type and cabin type during the target vehicle's movement.

[0012] The obstacle information determination module is used to process information within the field of view of the UAV under the current working conditions in order to determine the obstacle information of the environment to which the target vehicle belongs.

[0013] The target control information determination module is used to send the recognition result corresponding to the obstacle information to the processing module when the obstacle information meets the preset conditions, so that the processing module can determine the target control information corresponding to the target vehicle based on the recognition result;

[0014] The target control information includes at least deceleration commands and / or braking commands.

[0015] Thirdly, embodiments of the present invention also provide a vehicle, characterized in that it includes:

[0016] The drone and at least one processor; and

[0017] A memory that is communicatively connected to at least one processor; wherein,

[0018] The memory is used to receive processing information sent by the processor. The drone communicates with the processor and is used to receive information sent by the drone so that at least one processor can execute the drone-based vehicle collision avoidance method provided in any embodiment of the present invention.

[0019] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer instructions for causing a processor to execute the UAV-based vehicle collision avoidance method provided in any embodiment of the present invention.

[0020] This invention, in its embodiments, determines the current operational information of the drone and its cabin based on the drone type and cabin type during the target vehicle's movement. Under this current operational information, the drone processes information within its field of view to determine obstacle information in the target vehicle's environment. When the obstacle information meets preset conditions, a corresponding identification result is sent to the processing module, enabling the module to determine target control information corresponding to the target vehicle. This target control information includes at least deceleration and / or braking commands. By leveraging the drone's installation location, the aerial target recognition range is expanded, effectively eliminating blind spots above and behind vehicles and increasing the accuracy of identifying small, non-standard targets.

[0021] 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

[0022] 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.

[0023] Figure 1 A flowchart illustrating a vehicle collision avoidance method based on an unmanned aerial vehicle (UAV) provided in an embodiment of the present invention;

[0024] Figure 2 An overall framework diagram of a vehicle collision avoidance method based on an unmanned aerial vehicle (UAV) provided in an embodiment of the present invention;

[0025] Figure 3 A schematic diagram of a vehicle collision avoidance device based on an unmanned aerial vehicle (UAV) provided in an embodiment of the present invention;

[0026] Figure 4 This is a structural schematic diagram of a vehicle provided in an embodiment 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," etc., 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 the 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] To meet the increasingly diverse travel needs of users, some vehicles, such as SUVs or off-road vehicles, can be equipped with drone systems. These vehicles have high chassis and exceed a certain height threshold, requiring a correspondingly expanded detection range for aerial obstacles. Therefore, drones can be mounted on the vehicle roof, and to ensure the drone's safety and concealment, a drone bay is provided within the target vehicle. This bay has at least one observation port, allowing the drone to continuously collect aerial environmental information through its onboard sensors to ensure driving safety. Considering that the drone needs to remain operational throughout the vehicle's journey, environmental factors can cause the observation port to be obstructed by mud, water mist, ice, and leaves. Therefore, the drone needs to monitor the observation port's visibility in real time and automatically trigger a cleaning mechanism during operation. The specific monitoring and cleaning process for the drone bay's observation port is as follows:

[0030] During the operation of the drone, if an abnormality is detected in the drone's observation port, a cleaning signal is sent to the processing module, which then sends a cleaning instruction to the cleaning module of the target vehicle. The cleaning instruction is used to spray windshield washer fluid into the observation port, and the windshield washer fluid is installed on the target vehicle.

[0031] The target vehicle is a vehicle equipped with a drone and its associated cabin, characterized by a high chassis and body. The drone is an unmanned aircraft controlled by radio remote control equipment and program control devices, used for tasks such as aerial photography, equipment inspection, and environmental monitoring. The cabin is a closed or semi-closed space for accommodating the drone; in this embodiment, the cabin may be located on the roof of the target vehicle. The cleaning module is a built-in glass cleaning device in the target vehicle system, containing a reserve of dedicated windshield washer fluid.

[0032] Specifically, during operation, if the onboard sensors detect obstructions at the observation port of the drone's cabin, or if a large area of ​​abnormal shadows appears in the acquired images of the environment surrounding the target vehicle, the system determines that the observation port is in an abnormal state. At this time, the drone sends a cleaning request signal to the target vehicle's processing module. Upon receiving the signal, the processing module issues an instruction to the cleaning module to use its stored windshield washer fluid to spray and clean the observation port to restore the field of view.

[0033] Figure 1 This is a flowchart illustrating a vehicle collision avoidance method based on an unmanned aerial vehicle (UAV) according to an embodiment of the present invention. This embodiment is applicable to using UAVs to detect aerial targets in the area where a vehicle is located, thereby improving the target vehicle's field of vision. This method can be executed by a UAV-based vehicle collision avoidance device, which can be implemented in hardware and / or software and can be configured in the vehicle's communication module. Figure 1 As shown, a drone and a cabin for use with the drone are deployed on the roof of the target vehicle. The cabin can be used to charge the drone. The drone communicates with a processing module in the target vehicle to send the collected and / or processed data to the processing module. The method includes:

[0034] S110. During the movement of the target vehicle, determine the current working information of the drone and the cabin based on the drone type and cabin type.

[0035] The current operational information may include, but is not limited to, the orientation of the drone's sensors, the orientation of the cabin observation ports, and the drone's operational status. In this embodiment, the drone type can be divided into two types: omnidirectional recognition and unidirectional recognition. Omnidirectional recognition can be understood as installing multiple sensors at different locations on the drone's body to achieve 360° omnidirectional recognition, meaning it can identify targets from all directions. Unidirectional recognition only supports recognition in one direction, meaning that sensors are installed only in one direction on the drone's body. This type of drone can only identify targets in a single direction.

[0036] In this embodiment, the cabin type can be divided into two types: rotating and non-rotating. The rotating type is a cabin that can rotate relative to the drone's fuselage; when the drone is parked in the cabin, it can rotate along with the cabin or internal platform, with the direction of rotation being fixed. The non-rotating type, on the other hand, has a fixed position after the drone is parked and cannot rotate. When the target vehicle is in motion, the orientation and operating status of the drone's sensors within the cabin can be matched in real time based on the drone type and cabin type to determine the current operating information of the drone and the cabin.

[0037] Specifically, when the drone needs to return after completing its mission while the target vehicle is in motion, the system makes a decision based on the drone's orientation and the rotatability of the cabin, and then plans and determines accordingly.

[0038] S120. Under the current working information, the information within the field of vision of the UAV is processed to determine the obstacle information of the environment to which the target vehicle belongs.

[0039] The field of view (FVR) refers to the maximum angular range of the external space effectively covered by the environmental monitoring sensors onboard the UAV, such as LiDAR, TOF structured light, and binocular stereo vision cameras, through the cabin observation ports. Furthermore, the range of environmental information that can be acquired in a single observation can be expanded by increasing the number of cabin observation ports. Obstacle information includes data on all targets identified by the UAV that conflict with the vehicle, including the target's size, shape, attribute category, and position parameters relative to the target vehicle. If the target is a dynamic object, motion parameters such as its velocity and acceleration can be obtained. Obstacles detected by the UAV system are in the aerial environment. For example, in a vehicle driving scenario, obstacles identified by the UAV may include aerial targets such as trees, pipes, height restriction bars, and signs, as well as conventional targets such as walls and vehicles.

[0040] Specifically, based on the established working information of the drone and its cabin, the drone is used to perform real-time obstacle identification in the external space of the vehicle within the maximum angular range, and to obtain the corresponding obstacle information.

[0041] Optionally, based on the information within the drone's field of view, obstacle information in the target vehicle's environment can be determined, including:

[0042] The drone collects point cloud data and / or image data within its field of view using sensors mounted on it; obstacle information is determined based on the point cloud data and / or image data.

[0043] The obstacle information includes at least one or more of the following: walls, height restrictions, signs, and pipes.

[0044] The sensors carried by the drone can be monocular, binocular, multi-view cameras, event cameras, TOF structured light cameras, infrared cameras, LiDAR, ultrasonic radar, and millimeter-wave radar, etc. Point cloud data is a three-dimensional spatial structure composed of a large number of discrete three-dimensional spatial data points. These data points collectively construct the three-dimensional contour of the surrounding environment, and each data point contains at least its three-dimensional coordinates. Image data is a two-dimensional visual feature recorded in the form of a pixel array, and each pixel contains at least its brightness and color values. Both point cloud data and image data originate from the sensors carried by the drone, and both are used to acquire obstacle objects in the vehicle's external environment.

[0045] Specifically, the UAV collects point cloud data and / or image data of the vehicle's external space within its field of view through the observation port of the cabin. Based on this, it identifies obstacles and obtains their positions in the sensor coordinate system. Through a series of coordinate transformations, it fuses the real-time position and attitude information of the target vehicle, transforming this position to a unified world coordinate system. By calculating the geometric difference between its own position and the obstacle's position in the world coordinate system, the UAV ultimately determines the distance and angle of the obstacle relative to the target vehicle, thereby generating obstacle information.

[0046] For example, if walls and signs exist in the environment surrounding the vehicle, a drone inside the roof-mounted nacelle collects point cloud data and image data of the external space of the vehicle through an observation port. Information about the walls and signs is then determined based on the point cloud data and image data.

[0047] S130. When the obstacle information meets the preset conditions, the recognition result corresponding to the obstacle information is sent to the processing module so that the processing module can determine the target control information corresponding to the target vehicle based on the recognition result.

[0048] The target control information includes at least deceleration commands and / or braking commands.

[0049] The preset conditions are pre-defined safety criteria used to determine whether the target vehicle's trajectory overlaps with an obstacle, or whether the distance between them is below the minimum safe distance. These preset conditions can be customized based on the maneuverability and braking requirements of different vehicle models. The processing module receives the recognition results from the perception system, makes decisions accordingly, and generates vehicle control commands. The recognition results corresponding to obstacles include at least whether the obstacle is on the vehicle's current trajectory and whether there is a risk of collision with the target vehicle. Target control information directly affects the vehicle's operating state and can instruct the vehicle to perform actions such as deceleration or braking.

[0050] Specifically, the drone compares the obstacle information it collects with preset conditions. Once the preset conditions are met, a corresponding recognition result is generated and sent along with the obstacle information to the target vehicle processing module. The processing module then performs further collision risk assessment based on the recognition result and determines target control information accordingly, such as issuing an alert to the user and, if necessary, actively braking the vehicle to a stop to avoid a collision.

[0051] Optionally, when the obstacle information meets preset conditions, a recognition result corresponding to the obstacle information is sent to the processing module, including:

[0052] When the obstacle information is pipe information and / or wall information, determine whether the target vehicle has collided with the obstacle corresponding to the obstacle information based on the target vehicle's speed information and steering angle information; or, when the obstacle information is pipe information and / or wall information, send the obstacle information to the processing module so that the processing module can determine whether the target vehicle has collided with the obstacle corresponding to the obstacle information; if so, receive the identification result and / or obtain the identification result.

[0053] The pipeline information may include its geometry and spatial orientation, while the wall information may include its planar dimensions and spatial location. Vehicle speed information refers to the real-time speed and acceleration data obtained by the target vehicle processing module at the current moment. Steering angle information refers to the real-time steering wheel angle information obtained by the target vehicle processing module at the current moment.

[0054] Specifically, when the drone identifies an obstacle as a pipe, wall, or any combination thereof, it can directly determine whether there is a collision risk between the target vehicle and the obstacle based on the real-time vehicle speed and steering angle information received by the drone. Alternatively, the obstacle information can be sent to the target vehicle's processing module, which will then perform the collision risk assessment. If either method determines that a collision risk exists, the processing module will ultimately obtain or confirm the identification result.

[0055] This can be understood as two methods for collision risk assessment. First, the processing module within the drone independently assesses the risk based on acquired obstacle information, vehicle speed, and steering angle information, and sends the assessment result to the processing module. Second, the drone sends obstacle information to the target vehicle's processing module, which then combines this information with the target vehicle's real-time speed and steering angle information to perform a comprehensive calculation and arrive at the assessment result.

[0056] The technical solution provided by this invention determines the current operating information of the drone and its cabin based on the drone type and cabin type during the target vehicle's operation. Under this current operating information, the drone uses its onboard sensors to collect and process information within its field of view, obtaining point cloud data and / or image data. Obstacle information is determined based on the point cloud data and / or image data. Preset conditions are set, and the obstacle information is compared with these conditions. When the obstacle information meets the preset conditions, a corresponding recognition result is sent to the processing module. The target vehicle's processing module determines the target control information corresponding to the target vehicle based on the recognition result and the obstacle information. This target control information includes at least deceleration and / or braking commands. By leveraging the drone's installation location, the aerial target recognition range is expanded, effectively eliminating blind spots above and behind the vehicle, and increasing the accuracy of identifying small, non-standard targets.

[0057] Based on the above embodiments, it is known that the current working information is determined based on the drone type and cabin type, and then the obstacle information of the target vehicle's environment is determined based on the working information. In this embodiment, the drone can include two types: all-view recognition type and one-way recognition type. The cabin can include two types: rotating type and non-rotating type. Due to different drone types, different sensor schemes are used, which can match different types of cabins. The working information of the drone and cabin differs under different combinations. The specific content of the working information under different working types is described below. The current working information of the drone and cabin is determined based on the drone type and cabin type, including:

[0058] The first implementation method is as follows: When the drone type is full-view recognition type and the cabin type is rotating type or non-rotating type, the current working information of the drone and the cabin is to enable the drone in the current placement position and control the drone's acquisition view to be related to the driving direction of the target vehicle.

[0059] The installation location can include the top, bottom, and sides of the drone's fuselage. The acquisition perspective is the effective observation range formed by the drone's onboard sensors through the observation ports on the fuselage.

[0060] Specifically, when the drone type is an all-view recognition type, and the cabin type is either rotating or non-rotating, the drone can perform target recognition from any direction. Therefore, regardless of whether the cabin type is rotating or non-rotating, the drone can be parked anywhere after flying back to the cabin, without a fixed placement direction. The drone can then be launched at that location, and its data collection perspective adjusted to align with its direction of travel.

[0061] The second implementation method: When the drone type is one-way recognition type and the cabin type is non-rotatable type, the current working information of the drone and the cabin is to control the drone's acquisition perspective to be opposite to the front direction of the target vehicle in order to collect obstacle information behind the target vehicle.

[0062] Specifically, when the drone is a one-way recognition type and the cabin type is fixed, the drone needs to adjust its sensor position to face backward after flying back to the target vehicle. This can also be understood as positioning the drone's field of view opposite to the direction of the vehicle's front, making it easier to use the drone's sensors to identify obstacles behind the vehicle and collect information about those obstacles.

[0063] The third implementation method: When the drone type is one-way recognition type and the cabin is rotating type, based on the received reversing signal, the drone is controlled to adjust from the first direction to the second direction, where the first direction is the same as the direction of travel of the target vehicle, and the second direction is the opposite direction of travel of the target vehicle.

[0064] The reversing signal is a status signal sent from the target vehicle's processing module to the drone when the driver shifts into reverse, indicating that the vehicle is in reverse. The first direction is the direction the drone was in before receiving this signal, aligned with the target vehicle's direction of travel. The second direction is the direction the drone needs to adjust to after receiving the reversing signal, opposite to the vehicle's forward direction. The target vehicle's direction of travel is the direction recorded by the drone just moments before receiving the gear shift signal.

[0065] Specifically, when the drone type is unidirectional and the nacelle is rotating, the observation direction of the drone's sensors is determined based on the target vehicle's direction of travel before shifting gears and the real-time received gear position signal. When the drone receives a reversing signal, it needs to be controlled to adjust from the first direction, which was originally in the same direction as the driving direction, to a second direction opposite to the direction of the vehicle's front. When the drone receives a forward gear or other gear position signal, there is no need to change the drone's orientation; it only needs to be consistent with the target vehicle's direction of travel before shifting gears.

[0066] For example, when the driver shifts the gear to reverse, the target vehicle changes from forward to reverse. At this time, the drone receives a reversing signal from the processing module, and the system controls the drone to adjust its observation direction from the original direction of travel to the opposite direction of the vehicle's movement. Conversely, when the driver shifts the gear to drive (D) or another gear, the drone's observation angle remains unchanged, maintaining the same forward direction as the vehicle's original direction of travel, without requiring any directional adjustment.

[0067] The technical solutions provided in this invention determine different working information based on different combinations of drone types and cabin types. The first implementation involves, when the drone type is a full-view recognition type and the cabin type is either rotating or non-rotating, the current working information for both the drone and the cabin is to activate the drone at its current placement position and control the drone's acquisition angle to be related to the target vehicle's driving direction. The second implementation involves, when the drone type is a one-way recognition type and the cabin type is rotating, the current working information for both the drone and the cabin is to control the drone's acquisition angle to be opposite to the target vehicle's front direction to collect information about obstacles behind the target vehicle. The third implementation involves, when the drone type is a one-way recognition type and the cabin type is non-rotating, based on the received reversing signal, controlling the drone to adjust from a first direction to a second direction, where the first direction is consistent with the target vehicle's driving direction and the second direction is opposite to the target vehicle's driving direction. By combining different types of drones with different cabin types, the sensor orientation after the drone returns to the cabin is set differently. This fully utilizes limited sensors and effectively eliminates blind spots above and behind vehicles.

[0068] Figure 2 This is an overall framework diagram of a vehicle collision avoidance method based on unmanned aerial vehicles (UAVs) provided in an embodiment of the present invention, combined with... Figure 2 Understand the technical solutions of the embodiments of the present invention.

[0069] like Figure 2 As shown, this embodiment explains the overall implementation of the solution based on the above optional implementation methods. Specifically, it includes:

[0070] The target vehicle is equipped with a drone system. The drone is positioned on the roof of the vehicle, which has a designated bay. This bay not only powers the drone but also provides an observation port for the drone to continuously scan and identify external targets using its sensors. Simultaneously, the drone and the target vehicle maintain communication. The drone communicates with the target vehicle wirelessly, such as via Bluetooth or Wi-Fi. The drone sends its identification results to the target vehicle's processing module, which in turn sends information such as vehicle speed, direction, and gear changes to the drone. Based on this information, the drone automatically determines its final orientation upon returning to the vehicle. If the drone only supports one-way recognition (one-way recognition type) and the bay supports rotation (non-rotatable type), the drone's direction can be changed according to the vehicle's gear position. If the drone only supports one-way recognition and the bay is fixed, the drone is fixed in the opposite direction to the vehicle's front. If the drone supports omnidirectional recognition, it can be parked anywhere after returning to the vehicle. During the target vehicle's operation, the system continuously monitors for obstructions in front of the sensors. If obstructions are detected, such as mud, sand, or water mist, an automatic cleaning function is activated to maintain good visibility. Simultaneously, the drone's sensors identify targets in the vehicle's direction of travel, such as overhead height restrictions, and determine if there is a collision risk. Once a risk is identified, the information is reported to the vehicle, and the vehicle controller issues an alarm or automatically brakes to avoid a collision.

[0071] The technical solution provided by this invention determines the current operating information of the drone and its cabin based on the drone type and cabin type during the target vehicle's operation. Under this current operating information, the drone uses its onboard sensors to collect and process information within its field of view, obtaining point cloud data and / or image data. Obstacle information is determined based on the point cloud data and / or image data. Preset conditions are set, and the obstacle information is compared with these conditions. When the obstacle information meets the preset conditions, a corresponding recognition result is sent to the processing module. The target vehicle's processing module determines the target control information corresponding to the target vehicle based on the recognition result and the obstacle information. This target control information includes at least deceleration and / or braking commands. By leveraging the drone's installation location, the aerial target recognition range is expanded, effectively eliminating blind spots above and behind the vehicle, and increasing the accuracy of identifying small, non-standard targets.

[0072] Figure 3 This is a schematic diagram of a vehicle collision avoidance device based on an unmanned aerial vehicle (UAV) according to an embodiment of the present invention. Figure 3 As shown, the device includes: a current working information determination module 210, an obstacle information determination module 220, and a target control information determination module 230.

[0073] The current working information determination module 210 is used to determine the current working information of the UAV and the cabin based on the UAV type and cabin type during the target vehicle's movement. The obstacle information determination module 220 is used to process information within the UAV's field of vision based on the current working information to determine the obstacle information of the target vehicle's environment. The target control information determination module 230 is used to send the recognition result corresponding to the obstacle information to the processing module when the obstacle information meets preset conditions, so that the processing module can determine the target control information corresponding to the target vehicle based on the recognition result.

[0074] The technical solution provided by this invention determines the current operating information of the drone and its cabin based on the drone type and cabin type during the target vehicle's movement. Under this current operating information, the drone processes information within its field of view to determine obstacle information in the target vehicle's environment. When the obstacle information meets preset conditions, a corresponding recognition result is sent to the processing module, enabling the processing module to determine target control information corresponding to the target vehicle based on the recognition result. The target control information includes at least deceleration and / or braking commands. This leverages the drone's installation location to expand the aerial target recognition range, effectively eliminating blind spots above and behind vehicles and increasing the accuracy of identifying small, non-standard targets.

[0075] Based on the above technical solutions, optionally, the current working information determination module is further configured to: when the UAV type is a full-view recognition type and the cabin type is a rotating or non-rotating type, the current working information of the UAV and cabin is to activate the UAV at the current placement position and control the UAV's acquisition view to be related to the target vehicle's driving direction; or, when the UAV type is a one-way recognition type and the cabin type is a non-rotating type, the current working information of the UAV and cabin is to control the UAV's acquisition view to be opposite to the front direction of the target vehicle in order to acquire obstacle information behind the target vehicle; or, when the UAV type is a one-way recognition type and the cabin type is a rotating type, based on the received reversing signal, control the UAV to adjust from a first direction to a second direction, wherein the first direction is the same as the target vehicle's driving direction and the second direction is the opposite to the target vehicle's driving direction; and use the second direction as the current working information of the UAV and cabin.

[0076] Based on the above technical solutions, the device also includes:

[0077] The cleaning instruction sending module is used to send a cleaning signal to the processing module when an abnormality is detected in the observation port of the UAV during its operation. This signal causes the processing module to send a cleaning instruction to the cleaning module of the target vehicle. The cleaning instruction is used to spray windshield washer fluid into the observation port, and the windshield washer fluid is installed on the target vehicle.

[0078] Based on the above technical solutions, the obstacle information determination module includes:

[0079] The data acquisition module is used to acquire point cloud data and / or image data within the field of view based on the sensors carried on the UAV;

[0080] The obstacle information determination module is used to determine obstacle information based on point cloud data and / or image data;

[0081] The obstacle information includes at least one or more of the following: walls, height restrictions, signs, and pipes.

[0082] Based on the above technical solutions, optionally, the target control information determination module is further used to determine whether the target vehicle has collided with the obstacle corresponding to the obstacle information when the obstacle information is pipeline information and / or wall information, based on the vehicle speed information and steering angle information of the target vehicle; or, when the obstacle information is pipeline information and / or wall information, send the obstacle information to the processing module so that the processing module can determine whether the target vehicle has collided with the obstacle corresponding to the obstacle information; if so, receive the identification result and / or obtain the identification result.

[0083] The vehicle collision avoidance device based on unmanned aerial vehicles (UAVs) provided in the embodiments of the present invention can execute the vehicle collision avoidance method based on UAVs provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0084] Figure 4 This is a schematic diagram of the structure of a vehicle provided in an embodiment of the present invention. Figure 4 As shown, the vehicle includes a drone-based vehicle collision avoidance system 310, a controller 320, a storage device 330, an input device 340, and an output device 350. The number of controllers 320 in the vehicle can be one or more. Figure 4 Taking a controller 320 as an example; the UAV-based vehicle collision avoidance system 310, controller 320, storage device 330, input device 340, and output device 350 can be connected via bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.

[0085] The drone-based vehicle collision avoidance system 310 can be used to detect aerial targets in the area where the vehicle is located using drones, thereby improving the target vehicle's field of vision.

[0086] Storage device 330, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the UAV-based vehicle collision avoidance in this embodiment of the invention (e.g., current working information determination module 210, obstacle information determination module 220, and target control information determination module 230). Controller 320 executes various vehicle functions and data processing by running the software programs, instructions, and modules stored in storage device 330, thereby realizing the aforementioned UAV-based vehicle collision avoidance method.

[0087] Storage device 330 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on terminal usage. Furthermore, storage device 330 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, storage device 330 may further include memory remotely configured relative to controller 320, which can be connected to the vehicle via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0088] Input device 340 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the vehicle. Output device 350 may include display devices such as a display screen.

[0089] This invention also provides a storage medium containing computer-executable instructions. When executed by a computer processor, these instructions are used to perform a drone-based vehicle collision avoidance method. The target vehicle has a drone deployed on its roof and a cabin for use with the drone. The cabin can be used to charge the drone. The drone communicates with a processing module in the target vehicle to send collected and / or processed data to the processing module. The method includes:

[0090] During the target vehicle's movement, the current operational information of the drone and its cabin is determined based on the drone type and cabin type.

[0091] Based on the current working information, the drone processes the information within its field of vision to determine the obstacle information of the target vehicle's environment;

[0092] When the obstacle information meets the preset conditions, the recognition result corresponding to the obstacle information is sent to the processing module so that the processing module can determine the target control information corresponding to the target vehicle based on the recognition result.

[0093] The target control information includes at least deceleration commands and / or braking commands.

[0094] Of course, the computer-executable instructions provided in the embodiments of the present invention are not limited to the above-described method operations, but can also perform related operations in the vehicle collision avoidance method based on UAV provided in any embodiment of the present invention.

[0095] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0096] 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.

[0097] 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 preventing collision of a vehicle based on a UAV, characterized by, A drone and a cabin for the drone are arranged on a roof of a target vehicle, the cabin is used for charging the drone, the drone communicates with a processing module in the target vehicle to send collected and / or processed data to the processing module, and the method comprises: During driving of the target vehicle, current working information of the drone and the cabin is determined according to a drone type of the drone and a cabin type of the cabin; Under the current working information, information within a field of view of the drone is processed to determine obstacle information of an environment to which the target vehicle belongs; When the obstacle information meets a preset condition, an identification result corresponding to the obstacle information is sent to the processing module, so that the processing module determines target control information corresponding to the target vehicle based on the identification result; The target control information at least comprises a deceleration instruction and / or a stop instruction.

2. The method of claim 1, wherein, The current working information of the drone and the cabin is determined according to the drone type of the drone and the cabin type of the cabin, comprising: When the drone type is a full-view recognition type and the cabin type is a rotating type or a non-rotatable type, the current working information of the drone and the cabin is to enable the drone at a current placement position and control a collection view angle of the drone to be related to a driving direction of the target vehicle.

3. The method of claim 1, wherein, The current working information of the drone and the cabin is determined according to the drone type of the drone and the cabin type of the cabin, comprising: When the drone type is a one-way recognition type and the cabin type is a non-rotatable type, the current working information of the drone and the cabin is to control the collection view angle of the drone to be opposite to a head direction of the target vehicle to collect obstacle information behind the target vehicle.

4. The method of claim 1, wherein, The current working information of the drone and the cabin is determined according to the drone type of the drone and the cabin type of the cabin, comprising: When the drone type is a one-way recognition type and the cabin type is a rotating type, based on a received reverse signal, the drone is controlled to be adjusted from a first direction to a second direction, wherein the first direction is consistent with a driving direction of the target vehicle, and the second direction is opposite to the driving direction of the target vehicle; The second direction is taken as the current working information of the drone and the cabin.

5. The method of claim 1, wherein, The method further comprises: During working of the drone, if an observation port of the drone is detected to have an abnormality, a cleaning signal is sent to the processing module to make the processing module send a cleaning instruction to a cleaning module of the target vehicle; wherein the cleaning instruction is used for spraying glass water to the observation port, and the glass water is loaded by the target vehicle.

6. The method of claim 1, wherein, The information within the field of view of the drone is processed to determine the obstacle information of the environment to which the target vehicle belongs, comprising: Point cloud data and / or image data within the field of view are collected based on sensors carried in the drone; According to the point cloud data and / or image data, determine obstacle information; The obstacle information at least includes one or more of a wall, height limit information, sign information, and pipeline information.

7. The method of claim 1, wherein, When the obstacle information meets a preset condition, the recognition result corresponding to the obstacle information is sent to the processing module, including: When the obstacle information is pipeline information and / or wall information, according to the vehicle speed information and the steering angle information of the target vehicle, it is determined whether the target vehicle collides with the obstacle corresponding to the obstacle information; or, When the obstacle information is pipeline information and / or wall information, the obstacle information is sent to the processing module to determine whether the target vehicle collides with the obstacle corresponding to the obstacle information based on the processing module; If so, the recognition result is received and / or obtained.

8. An unmanned aerial vehicle based vehicle anti-collision device, characterized in that, A drone and a cabin matched with the drone are arranged on an upper part of a roof of a target vehicle, the cabin can be used to charge the drone, the drone communicates with a processing module in the target vehicle to send collected and / or processed data to the processing module, and the device includes: A current working information determination module is configured to determine current working information of the drone and the cabin according to a drone type of the drone and a cabin type of the cabin during driving of the target vehicle; An obstacle information determination module is configured to determine obstacle information of an environment to which the target vehicle belongs based on processing of information in a field of view of the drone under the current working information; A target control information determination module is configured to send an identification result corresponding to the obstacle information to the processing module when the obstacle information meets a preset condition, so that the processing module determines target control information corresponding to the target vehicle based on the identification result. The target control information at least includes a deceleration instruction and / or a stop instruction.

9. A vehicle characterized by comprising: The device includes: The drone and at least one processor; And The memory is in communication connection with the at least one processor; wherein The memory is used to receive processing information sent by the processor, the drone and the processor are in communication, and are used to receive information sent by the drone, so that the at least one processor can execute the vehicle anti-collision method based on the drone in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions, and the computer instructions are used to make the processor execute the vehicle anti-collision method based on the drone in any one of claims 1-7.

Images