Vehicle-mounted unmanned aerial vehicle control method and vehicle

Abstract

The invention discloses a control method of a vehicle-mounted unmanned aerial vehicle and a vehicle. The control method of the vehicle-mounted unmanned aerial vehicle comprises the steps of obtaining a vehicle driving state and driving surrounding environment information; judging whether the vehicle has a collision risk according to the vehicle driving state and the driving surrounding environment information; and when the vehicle has the collision risk, controlling the vehicle-mounted unmanned aerial vehicle to put an air bag at the collision prediction position. Therefore, the vehicle-mounted unmanned aerial vehicle is controlled to put the safety air bag at the collision prediction position when the vehicle has the collision risk, so that the position limitation of fixing the air bag in the prior art can be solved, the safety air bag can be suitable for being used in an omnibearing collision scene, and thus protection on vehicle passengers and the vehicle can be enhanced; damage to the vehicle during collision is reduced, and the maintenance cost of the vehicle is reduced.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and in particular to a control method and vehicle for a vehicle-mounted unmanned aerial vehicle. Background Technology

[0002] With the rapid development of the automotive industry, airbags are no longer only used to protect the safety of people inside the vehicle. Some cars have designed airbags under the front of the vehicle to reduce the damage to the front of the vehicle in the event of a collision. In addition, some cars are equipped with drones to monitor the road conditions and predict possible collisions.

[0003] Existing airbags have fixed placement locations, poor mobility, and high repair costs after vehicle collisions. In addition, existing vehicle-mounted drones only serve a cruising function. Summary of the Invention

[0004] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a control method for a vehicle-mounted drone that can improve the mobility of airbag deployment.

[0005] The present invention further proposes a vehicle.

[0006] According to an embodiment of the present invention, a control method for a vehicle-mounted drone includes: acquiring vehicle driving status and surrounding environment information; determining whether the vehicle has a collision risk based on the vehicle driving status and surrounding environment information; and when the vehicle has a collision risk, controlling the vehicle-mounted drone to deploy airbags at the collision prediction location.

[0007] Therefore, by controlling an onboard drone to deploy airbags at the predicted collision location when a vehicle is at risk of collision, the limitations of fixed airbag placement in existing technologies can be overcome. This allows airbags to be used in all-around collision scenarios, thereby enhancing the protection of vehicle occupants and the vehicle itself, reducing injuries to the vehicle during a collision, and lowering vehicle maintenance costs.

[0008] According to some embodiments of the present invention, the step of controlling the vehicle-mounted drone to deploy airbags at a collision prediction location when the vehicle is at risk of collision includes: controlling the vehicle-mounted drone to move to the deployment prediction location; controlling the airbag to pop out from the bottom of the vehicle-mounted drone, the airbag inflating at the collision prediction location; and controlling the vehicle-mounted drone to rise and survey the area around the vehicle.

[0009] According to some embodiments of the present invention, the step of controlling the vehicle-mounted drone to rise and patrol the area around the vehicle further includes: controlling the vehicle-mounted drone to acquire at least one piece of information, including the collision process, vehicle damage, personnel status, and the environment around the vehicle; and controlling the warning lights to flash and the alarm sound to play.

[0010] According to some embodiments of the present invention, the step of controlling the vehicle-mounted drone to deploy airbags at the collision prediction location when the vehicle is at risk of collision includes: reminding the driver of the deployment direction of the airbags.

[0011] According to some embodiments of the present invention, the step of determining whether the vehicle has a collision risk based on the vehicle's driving state and the surrounding environment information includes: using a spatiotemporal graph neural network model to predict the expected driving trajectory of the vehicle and the multimodal probability trajectory distribution map of dynamic targets around the vehicle within a first preset time period; calculating the minimum distance, relative speed, and collision time between the vehicle and all potential collision targets in the multimodal probability trajectory distribution map; and determining whether the vehicle has a collision risk based on the collision time.

[0012] According to some embodiments of the present invention, the step of determining whether the vehicle has a collision risk based on the collision time includes: when the collision time is greater than a first preset time, determining that the degree of collision of the vehicle is low risk, and the vehicle-mounted drone does not start; when the collision time is less than the first preset time but greater than a second preset time, determining that the degree of collision of the vehicle is medium risk, controlling the vehicle-mounted drone to rise and patrol the area around the vehicle; when the collision time is less than the second preset time, determining that the degree of collision of the vehicle is high risk, controlling the vehicle-mounted drone to deploy airbags, wherein the second preset time is less than the first preset time, and the first preset time is less than the second preset time.

[0013] According to some embodiments of the present invention, before the step of determining whether the vehicle has a collision risk based on the collision time, the method further includes: obtaining the collision prediction location based on the intersection of the expected driving trajectory of the vehicle and the multimodal probability trajectory distribution map of dynamic targets around the vehicle.

[0014] According to some embodiments of the present invention, when the collision time is less than a first preset time and greater than a second preset time, the step of determining that the degree of collision of the vehicle is of medium risk and controlling the vehicle-mounted drone to rise and patrol the area around the vehicle includes: reminding the driver of the direction of the predicted collision location of the vehicle.

[0015] According to some embodiments of the present invention, the step of obtaining vehicle driving status and surrounding environment information includes: obtaining at least one piece of information from the vehicle's gear position, speed, power supply, and historical collision accident records of other vehicles, pedestrians, static obstacles, road information, and surrounding environment around the vehicle.

[0016] According to an embodiment of the present invention, a vehicle is suitable for the above-described vehicle-mounted drone control method, comprising: a vehicle sensing module; a vehicle infotainment system control module, the vehicle infotainment system control module being electrically connected to the vehicle sensing module; a vehicle-mounted drone, the vehicle-mounted drone being communicatively connected to the vehicle infotainment system control module; and an airbag, the airbag being disposed at the bottom of the vehicle-mounted drone.

[0017] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a partial schematic diagram of a vehicle and its surrounding environment according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a light on a vehicle instrument panel that reminds the driver of the direction of the predicted collision location of the vehicle, according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the lights in the vehicle instrument panel displaying the airbag deployment direction according to an embodiment of the present invention; Figure 4 This is a partial schematic diagram of a control method for a vehicle-mounted unmanned aerial vehicle according to an embodiment of the present invention; Figure 5 This is a partial schematic diagram of a control method for a vehicle-mounted unmanned aerial vehicle according to an embodiment of the present invention.

[0019] Figure label: 1000, vehicles; 100. Vehicle-mounted drones; 10. Warning lights; 20. Cameras; 200. Airbags. Detailed Implementation

[0020] The embodiments of the present invention are described in detail below. The embodiments described with reference to the accompanying drawings are exemplary. The embodiments of the present invention are described in detail below.

[0021] The following is for reference. Figures 1-5 A control method for a vehicle-mounted unmanned aerial vehicle 100 according to an embodiment of the present invention is described, and the control method for the vehicle-mounted unmanned aerial vehicle 100 is applicable to a vehicle 1000.

[0022] According to an embodiment of the present invention, in combination Figure 1 , Figure 4 and Figure 5 As shown, the control method of the vehicle-mounted unmanned aerial vehicle 100 according to an embodiment of the present invention may mainly include: Acquire vehicle 1000's driving status and surrounding environment information; Based on the driving status of vehicle 1000 and the surrounding environment information, determine whether vehicle 1000 is at risk of collision. When there is a risk of collision with vehicle 1000, control the vehicle-mounted drone 100 to deploy airbag 200 at the collision prediction location.

[0023] Specifically, in the control method of the vehicle-mounted drone 100 of the present invention, the current driving state of the vehicle 1000 and the environmental information around the vehicle are first acquired to determine whether there is a collision risk for the vehicle 1000. If there is no collision risk for the vehicle 1000, the control method of the vehicle-mounted drone 100 acquires the current driving state of the vehicle 1000 and the environmental information around the vehicle again. That is, during the driving process of the vehicle 1000, the control method of the vehicle-mounted drone 100 will repeatedly monitor the driving state of the vehicle 1000 and the surrounding environment to detect the risk of collision of the vehicle 1000 in a timely manner.

[0024] When the control method of the vehicle-mounted drone 100 determines that the vehicle 1000 is at risk of collision based on the driving status of the vehicle 1000 and the surrounding environment, the control method of the vehicle-mounted drone 100 immediately activates the vehicle-mounted drone 100 and controls it to deploy the airbag 200 towards the predicted collision location. In the event of an actual collision, the airbag will deploy between the vehicle 1000 and the object being collided with, and the deformation and deflation of the airbag will reduce the impact on the vehicle 1000, thereby reducing the collision injuries to the vehicle 1000 and its occupants, and thus significantly reducing the repair costs after a collision.

[0025] Furthermore, in an embodiment of the present invention, the drone is mobile, and using the drone for the deployment of the airbag 200 can overcome the limitations of the airbag 200's location setting. Thus, the airbag 200 is also mobile, so that the airbag 200 can be used to reduce the damage to the vehicle 1000 in the event of a collision at any location of the vehicle 1000.

[0026] Therefore, this invention controls the vehicle-mounted drone 100 to deploy airbags 200 at the collision prediction location when there is a collision risk to the vehicle 1000. This solves the positional limitations of fixed airbags in the prior art, and makes the airbags 200 suitable for use in all-around collision scenarios. This can enhance the protection of the occupants of the vehicle 1000 and the vehicle 1000, reduce the damage to the vehicle 1000 in the event of a collision, and reduce the maintenance cost of the vehicle 1000.

[0027] Combination Figure 1 and Figure 4 As shown, when there is a collision risk to vehicle 1000, the steps for controlling the vehicle-mounted drone 100 to deploy airbag 200 at the predicted collision location include: Control the vehicle-mounted drone 100 to move to the predicted deployment location; The airbag 200 is deployed from the bottom of the vehicle-mounted drone 100, and the airbag 200 inflates at the collision prediction location; Control the vehicle-mounted drone 100 to rise and patrol around vehicle 1000.

[0028] Specifically, when the control method of the vehicle-mounted drone 100 controls the vehicle-mounted drone 100 to deploy the airbag 200, it first controls the vehicle-mounted drone 100 to move to the predicted deployment position. In an embodiment of the present invention, the predicted deployment position is 0.3m above the predicted collision position, and the control method of the vehicle-mounted drone 100 controls the vehicle-mounted drone 100 to arrive at the predicted deployment position 1 second before the predicted collision occurs.

[0029] Furthermore, when the vehicle-mounted drone 100 reaches the predicted deployment location, the control method of the vehicle-mounted drone 100 controls the airbag 200 inside the vehicle-mounted drone 100 to deploy from the bottom of the vehicle-mounted drone 100 to the predicted collision location. When the airbag 200 reaches the predicted collision location, it rapidly inflates to form a spherical protective buffer barrier.

[0030] When a collision occurs, the airbag 200 absorbs the impact energy by deforming and deflating itself. This improves the safety factor of the vehicle 1000, effectively reduces the injury to the driver and passengers in a collision, protects pedestrians and cyclists outside the vehicle, reduces the injury caused by direct impact to the vehicle body, and significantly reduces direct damage to the vehicle 1000 itself, thereby reducing maintenance costs.

[0031] Furthermore, after the vehicle-mounted drone 100 deploys the airbag 200, the control method of the vehicle-mounted drone 100 controls the vehicle-mounted drone 100 to rise from the predicted deployment position to inspect the situation around the vehicle 1000, so as to monitor the surrounding environment after the collision of the vehicle 1000.

[0032] Combination Figure 1 and Figure 4 As shown, the steps for controlling the vehicle-mounted drone 100 to rise and patrol around the vehicle 1000 also include: Control the vehicle-mounted drone 100 to acquire at least one piece of information, including the collision process, vehicle 1000 damage, personnel status, and the environment surrounding the vehicle 1000. Control the alarm light 10 to flash and play an alarm sound.

[0033] Specifically, when the vehicle-mounted drone 100 is controlled by the control method to deploy the airbag 200 and raise it, the vehicle-mounted drone 100 is equipped with a camera 20, a warning light 10, and a speaker. The control method of the vehicle-mounted drone 100 controls the camera 20 to record the collision process, vehicle 1000 damage, personnel status, and information about the surrounding environment of the vehicle 1000, and controls the warning light 10 to flash and the speaker to emit an alarm sound. This can warn other vehicles 1000 or pedestrians that a vehicle 1000 collision has occurred, which can reduce the risk of secondary collisions.

[0034] Combination Figure 1 , Figure 3 , Figure 4 and Figure 5 As shown, when there is a collision risk to vehicle 1000, the steps for controlling the vehicle-mounted drone 100 to deploy airbag 200 at the predicted collision location include: Remind the driver of the direction in which the airbag 200 is deployed.

[0035] Specifically, when there is a collision risk to vehicle 1000, the control method of vehicle-mounted drone 100 not only controls vehicle-mounted drone 100 to deploy airbag 200 to the collision prediction location, but also controls the instrument panel of vehicle 1000 to display the deployment direction of airbag 200 to the driver, so as to inform the driver of information such as airbag deployment and impending collision of vehicle 1000, so that the driver can grasp the real-time status of vehicle 1000.

[0036] Furthermore, the control method of the vehicle-mounted drone 100 transmits collision signals, such as the collision location, the object of collision, and the collision type, to the vehicle-mounted drone 100. Simultaneously, the in-vehicle instrument panel displays the deployment direction of the airbag 200 in red and provides a voice prompt. The voice prompt can be designed as needed, displaying the deployment direction of the airbag 200 in eight directions, including but not limited to the front, rear, left, right, left-front, right-front, left-rear, and right-rear of the vehicle 1000.

[0037] Combination Figure 1 and Figure 5 As shown, the steps for determining whether vehicle 1000 is at risk of collision based on its driving status and surrounding environment information include: The spatiotemporal graph neural network model is used to predict the expected driving trajectory of vehicle 1000 and the multimodal probability trajectory distribution map of dynamic targets around vehicle 1000 within a first preset time period. Calculate the minimum distance, relative speed, and collision time between vehicle 1000 and all potential collision targets in the multimodal probability trajectory distribution map; Determine whether vehicle 1000 is at risk of collision based on the time of collision.

[0038] Specifically, after the vehicle-mounted UAV 100 obtains information such as the driving status of the vehicle 1000 and the surrounding environment, the spatiotemporal graph neural network model can be used to pre-determine the driving trajectory of the vehicle 1000 and the movement trajectory of the dynamic targets around the vehicle 1000 within a first preset time period, and construct a multimodal probability trajectory distribution map including the outline shape and movement trajectory of the dynamic targets around the vehicle 1000.

[0039] Furthermore, in the multimodal probability trajectory distribution map, it can be assumed that all non-stationary targets may collide with the vehicle. The control method of the vehicle-mounted UAV 100 can calculate the minimum distance, relative speed and collision time between all potential collision targets in the multimodal probability trajectory distribution map, and determine whether there are potential targets around the vehicle 1000 that may collide with the vehicle based on the calculated collision time, thereby determining whether the vehicle 1000 has a collision risk.

[0040] Furthermore, the control method of the vehicle-mounted drone 100 also collects massive amounts of real driving data and simulation data of high-risk scenarios such as congestion, intersections, and pedestrian crossings, and inputs them into the spatiotemporal graph neural network model for data training and optimization.

[0041] Combination Figure 1 and Figure 5 As shown, the steps to determine whether vehicle 1000 is at risk of collision based on the time of collision include: If the collision time is longer than the first preset time, the degree of collision of vehicle 1000 is judged to be low risk, and vehicle-mounted drone 100 will not start. When the collision time is less than the first preset time but greater than the second preset time, the degree of collision of vehicle 1000 is judged to be medium risk, and the vehicle-mounted drone 100 is controlled to rise and patrol the area around vehicle 1000. When the collision time is less than the second preset time, the degree of collision of vehicle 1000 is judged to be high risk, and the vehicle-mounted drone 100 is controlled to deploy airbag 200. The second preset time is less than the first preset time, and the first preset time is less than the second preset time.

[0042] Specifically, after the vehicle-mounted drone 100 obtains the collision time, it determines the magnitude of the collision time and the first preset time. If the collision time is greater than the first preset time, it means that the potential collision target in the environment around the vehicle 1000 is far away from the vehicle 1000. At this time, the risk of collision of the vehicle 1000 is low, and there is no need for the drone to patrol or deploy the airbag 200. Therefore, the control method of the vehicle-mounted drone 100 controls the drone not to start.

[0043] Furthermore, when the control method of the vehicle-mounted drone 100 determines that the collision time is less than a first preset time, the control method of the vehicle-mounted drone 100 continues to determine the magnitude of the collision time relative to a second preset time. If the collision time is greater than the second preset time, it indicates that the distance between the potential collision target in the environment surrounding the vehicle 1000 and the vehicle 1000 has decreased, but the predicted collision can be avoided by taking measures. In this case, the collision situation of the vehicle 1000 is classified as medium risk. Under medium risk conditions, the control method of the vehicle-mounted drone 100 controls the drone to start and take off, and patrols around the vehicle 1000. At the same time, the vehicle-mounted drone 100 makes a voice broadcast to prepare for the deployment of the airbag 200 and to remind the user of the collision risk. The prompt voice can be set as needed.

[0044] Furthermore, when the control method of the vehicle-mounted drone 100 determines that the collision time is less than a second preset time, it indicates that the potential collision object is about to hit the vehicle and the collision is unavoidable. At this time, the risk of collision for the vehicle 1000 is high. The control method of the vehicle-mounted drone 100 controls the drone to move to the predicted deployment position and deploys the airbag 200 at the predicted collision position to place a protective barrier between the vehicle and the collision object before the collision occurs.

[0045] In an embodiment of the present invention, if the collision risk of the vehicle 1000 is reduced from high risk or medium risk to low risk, the control method of the vehicle-mounted drone 100 controls the vehicle-mounted drone 100 to reset.

[0046] In embodiments of the present invention, the first preset time includes, but is not limited to, 5 seconds, and the second preset time includes, but is not limited to, 3 seconds.

[0047] Combination Figure 5 As shown, the steps to determine whether vehicle 1000 is at risk of collision based on the time of collision also include: The collision prediction location is derived from the intersection of the expected driving trajectory of vehicle 1000 and the multimodal probability trajectory distribution map of dynamic targets around vehicle 1000.

[0048] Specifically, while the control method of the vehicle-mounted drone 100 calculates the collision time, the control method of the vehicle-mounted drone 100 also obtains the collision prediction position based on the intersection of the preset driving trajectory of the vehicle 1000 and the preset moving trajectory of the potential collision target. In this way, the control method of the vehicle-mounted drone 100 can control the vehicle-mounted drone 100 to deploy the airbag 200 at the preset deployment position based on the collision prediction position.

[0049] Combination Figure 2 and Figure 5 As shown, when the collision time is less than a first preset time but greater than a second preset time, the degree of collision of vehicle 1000 is determined to be medium risk. The steps to control the vehicle-mounted drone 100 to rise and patrol the area around vehicle 1000 include: The system alerts the driver to the direction of the predicted collision location for vehicle 1000.

[0050] Specifically, when the collision risk of vehicle 1000 is moderate, the control method of vehicle-mounted drone 100 controls the vehicle-mounted drone 100 to only patrol around vehicle 1000 without deploying airbags 200. At this time, the instrument panel of vehicle 1000 is controlled to display the direction of the collision prediction location to the driver, so as to remind the driver that there is a collision risk in that direction and that there is a potential collision object in that direction of vehicle 1000, so that the driver can take measures to reduce the collision risk before the collision occurs, thereby reducing the loss of vehicle 1000 and improving the driving safety of vehicle 1000.

[0051] In an embodiment of the present invention, the light on the instrument panel of the vehicle 1000 that reminds the driver of the direction of the collision prediction location of the vehicle 1000 may be including but not limited to yellow light, and may display the direction of the collision prediction location in eight directions including but not limited to the front, rear, left, right, left front, right front, left rear and right rear of the vehicle 1000.

[0052] According to an embodiment of the present invention, the steps for obtaining the driving status and surrounding environment information of vehicle 1000 include: Obtain at least one piece of information from historical collision records of vehicle 1000, including vehicle gear, speed, power, and other vehicles 1000, pedestrians, static obstacles, road information, and the surrounding environment.

[0053] It should be noted that the vehicle speed information obtained by the control method of the vehicle-mounted drone 100 includes the vehicle speed and acceleration of the vehicle 1000. In addition, the control method of the vehicle-mounted drone 100 also obtains the yaw angle and vehicle type of the vehicle 1000 to more accurately determine whether the vehicle 1000 has a collision risk.

[0054] According to an embodiment of the present invention, a vehicle 1000 includes a vehicle 1000 sensing module, a vehicle infotainment system central control module, an in-vehicle drone 100, and an airbag 200. The vehicle infotainment system central control module is electrically connected to the vehicle 1000 sensing module, the in-vehicle drone 100 is communicatively connected to the vehicle infotainment system central control module, and the airbag 200 is disposed at the bottom of the in-vehicle drone 100.

[0055] Specifically, the vehicle 1000 perception module is responsible for collecting real-time, all-around data on the vehicle 1000's own status and the surrounding environment, providing data input for risk assessment. The vehicle 1000 perception module includes, but is not limited to, mid-to-long-range millimeter-wave radar, lidar, ultrasonic radar, a three-eye camera 20, and vehicle 1000's own status sensors. Among them, the vehicle 1000's own status sensors include, but are not limited to, the vehicle speed sensor, acceleration sensor, steering angle sensor, gear sensor, throttle position sensor, and pedal position sensor carried by the vehicle 1000 itself.

[0056] The vehicle infotainment system's central control module is responsible for data processing, trajectory prediction, risk assessment, and coordinating the work of various modules and executing instructions. The central control module of the vehicle infotainment system includes, but is not limited to, a central processing unit, memory, and communication bus.

[0057] Furthermore, the vehicle-mounted drone 100 is responsible for video evidence collection, sound and light alarms, drone flight control, and airbag deployment, including but not limited to camera 20, sound and light alarm system, drone control system, airbag 200, electromagnetic lock, energy storage spring, and honeycomb structure plastic plate.

[0058] When the vehicle-mounted drone 100 deploys the airbag 200, the electromagnetic latch receives the arrival signal and instantly releases the compressed energy storage spring. The energy storage spring is compressed by 5mm and has an ejection force of 80N, capable of deploying an airbag weighing up to 2KG. Further, the honeycomb-structured plastic plate pushes the folded airbag pack out from the bottom of the vehicle-mounted drone 100, causing the drone 100 to rise rapidly. The micro gas generator built into the airbag 200 is quickly activated, rapidly inflating between the vehicle 1000 and the collision object to form a spherical protective buffer barrier.

[0059] According to an embodiment of the present invention, vehicle 1000 has the function of protecting itself from collision damage by combining an onboard drone 100 and an airbag 200. In practical applications, when vehicle 1000 is traveling at low speed in congested environments and there is a risk of a minor collision, the vehicle's infotainment system control module can control the onboard drone 100 to quickly fly out and issue an audible warning, alerting the driver of the vehicle, drivers of surrounding vehicles 1000, and pedestrians that there is an imminent collision hazard around the vehicle. If the collision risk persists after the warning, the onboard drone 100 can quickly deploy the airbag 200 at the predicted collision location before the collision time predicted by the control method of the onboard drone 100, forming a protective barrier to reduce the impact on vehicle 1000 and the injuries to occupants. This invention overcomes the limitations of the airbag 200's placement location, achieving all-round protection for vehicle 1000, greatly improving the safety factor of vehicle 1000 in collision accidents, effectively protecting the safety of drivers, passengers, and pedestrians, while also reducing repair costs after a collision.

[0060] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "circumferential," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0061] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example.

[0062] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A control method for a vehicle-mounted unmanned aerial vehicle, comprising: Obtain vehicle (1000) driving status and surrounding environment information; Based on the driving status of the vehicle (1000) and the surrounding environmental information, determine whether the vehicle (1000) is at risk of collision; When the vehicle (1000) is at risk of collision, the vehicle-mounted drone (100) is controlled to deploy airbags (200) at the collision prediction location.

2. The control method for a vehicle-mounted unmanned aerial vehicle according to claim 1, characterized in that, The step of controlling the vehicle-mounted drone (100) to deploy airbags (200) at the collision prediction location when there is a collision risk to the vehicle (1000) includes: Control the vehicle-mounted drone (100) to move to the predicted deployment location; The airbag (200) is controlled to deploy from the bottom of the vehicle-mounted drone (100), and the airbag (200) inflates at the collision prediction location; Control the vehicle-mounted drone (100) to rise and patrol around the vehicle (1000).

3. The control method for a vehicle-mounted unmanned aerial vehicle according to claim 2, characterized in that, The steps of controlling the vehicle-mounted drone (100) to rise and patrol around the vehicle (1000) also include: Control the vehicle-mounted drone (100) to acquire at least one piece of information, including the collision process, vehicle (1000) damage, personnel status, and the environment surrounding the vehicle (1000); Control the alarm light (10) to flash and play an alarm sound.

4. The control method for a vehicle-mounted unmanned aerial vehicle according to claim 1, characterized in that, The step of controlling the vehicle-mounted drone (100) to deploy airbags (200) at the collision prediction location when there is a collision risk to the vehicle (1000) includes: The driver is reminded of the deployment direction of the airbag (200).

5. The control method for a vehicle-mounted unmanned aerial vehicle according to claim 1, characterized in that, The step of determining whether the vehicle (1000) has a collision risk based on the vehicle's (1000) driving status and the surrounding environment information includes: The spatiotemporal graph neural network model is used to predict the expected driving trajectory of the vehicle (1000) and the multimodal probability trajectory distribution map of dynamic targets around the vehicle (1000) within a first preset time period. Calculate the minimum distance, relative speed, and collision time between the vehicle (1000) and all potential collision targets in the multimodal probability trajectory distribution map; The collision time is used to determine whether the vehicle (1000) is at risk of collision.

6. The control method for a vehicle-mounted unmanned aerial vehicle according to claim 5, characterized in that, The step of determining whether the vehicle (1000) has a collision risk based on the collision time includes: When the collision time is greater than the first preset time, the degree of collision of the vehicle (1000) is determined to be low risk, and the vehicle-mounted drone (100) is not started; When the collision time is less than the first preset time and greater than the second preset time, the degree of collision of the vehicle (1000) is determined to be medium risk, and the vehicle-mounted drone (100) is controlled to rise and patrol the area around the vehicle (1000). When the collision time is less than the second preset time, it is determined that the degree of collision of the vehicle (1000) is high risk, and the vehicle-mounted drone (100) is controlled to deploy the airbag (200), wherein the second preset time is less than the first preset time, and the first preset time is less than the second preset time.

7. The control method for a vehicle-mounted unmanned aerial vehicle according to claim 6, characterized in that, Before the step of determining whether the vehicle (1000) has a collision risk based on the collision time, the following steps are also included: The collision prediction location is derived from the intersection of the expected driving trajectory of the vehicle (1000) and the multimodal probability trajectory distribution map of the dynamic targets around the vehicle (1000).

8. The control method for a vehicle-mounted unmanned aerial vehicle according to claim 7, characterized in that, The step of determining that the degree of collision of the vehicle (1000) is medium risk when the collision time is less than a first preset time and greater than a second preset time, and controlling the vehicle-mounted drone (100) to rise and patrol the area around the vehicle (1000) includes: The driver is reminded of the direction of the predicted collision location of the vehicle (1000).

9. The control method for a vehicle-mounted unmanned aerial vehicle according to claim 1, characterized in that, The steps for obtaining the vehicle (1000) driving status and surrounding environment information include: Acquire at least one piece of information from the vehicle's (1000) gear position, speed, and power supply, as well as historical collision accident records of other vehicles (1000), pedestrians, static obstacles, road information, and the surrounding environment.

10. A vehicle, suitable for the control method of the vehicle-mounted unmanned aerial vehicle (100) according to any one of claims 1-9, characterized in that, include: Vehicle (1000) perception module; The vehicle infotainment system central control module is electrically connected to the vehicle (1000) sensing module; A vehicle-mounted drone (100) is communicatively connected to the central control module of the vehicle system; An airbag (200) is disposed at the bottom of the vehicle-mounted drone (100).

Images