Trajectory planning method, apparatus and system, aircraft, movable platform, and computer-readable storage medium

Abstract

The present application provides a trajectory planning method, apparatus and system, an aircraft, a movable platform, and a computer-readable storage medium. In the method, an aircraft determines a recommended path for a movable platform on the basis of collected observation data; the aircraft sends the recommended path to the movable platform; the movable platform determines a movement planning path for the movable platform on the basis of the recommended path, and sends the movement planning path to the aircraft; and the aircraft receives the movement planning path, and on the basis of the movement planning path, determines a planned landing trajectory for landing on the movable platform. The present embodiment can improve the accuracy and safety of dynamic landing of aircrafts.

Description

Trajectory planning method, device, system, aircraft, movable platform and computer readable storage medium TECHNICAL FIELD

[0001] The present application relates to the technical field of aircraft, in particular to a trajectory planning method, device, system, aircraft, movable platform and computer readable storage medium. BACKGROUND

[0002] The aircraft is currently widely used in various fields such as aerial photography, agriculture, surveying and mapping, logistics, rescue, performance and the like. The intelligence level of the aircraft is also getting higher and higher, which can automatically identify the environment, avoid obstacles, plan the flight path, and automatically land, etc. This greatly improves the use efficiency and safety of the aircraft. However, how to make the aircraft safely and accurately land on the moving movable platform is still a big challenge.

[0003] SUMMARY

[0004] Therefore, the present application provides a trajectory planning method, device, system, aircraft, movable platform and computer readable storage medium to solve the problem that the aircraft cannot accurately land on the moving movable platform in the related art.

[0005] In a first aspect, a trajectory planning method is provided, applied to an aircraft, and the method comprises:

[0006] Based on the collected aircraft observation data, a recommended path of the movable platform is determined; the recommended path is sent to the movable platform, a movement planning path planned based on the recommended path is obtained from the movable platform; and based on the movement planning path, a landing planning trajectory landing on the movable platform is determined.

[0007] In a second aspect, a trajectory planning method is provided, applied to an aircraft, and the method comprises:

[0008] The path planning related information is sent to the movable platform; a movement planning path planned based on the path planning related information is obtained from the movable platform; and based on the movement planning path, a landing planning trajectory landing on the movable platform is determined.

[0009] In a third aspect, a trajectory planning method is provided, applied to a movable platform, and the method comprises:

[0010] receiving a recommended path sent by an aerial vehicle, the recommended path being determined by the aerial vehicle based on collected aerial vehicle observation data; planning a movement planning path based on the recommended path; and sending the movement planning path to the aerial vehicle, so that the aerial vehicle determines a landing planning trajectory to the movable platform based on the movement planning path.

[0011] In a fourth aspect, a trajectory planning method is provided, the method comprising:

[0012] an aerial vehicle determining a recommended path of a movable platform based on collected observation data; the aerial vehicle sending the recommended path to the movable platform; the movable platform determining a movement planning path of the movable platform based on the recommended path, and sending the movement planning path to the aerial vehicle; and the aerial vehicle receiving the movement planning path, and determining a landing planning trajectory to the movable platform based on the movement planning path.

[0013] In a fifth aspect, a trajectory planning apparatus is provided, comprising: a processor; and a memory storing computer program code executable on the processor; wherein the processor implements the steps of the method of the first aspect when executing the computer program.

[0014] In a sixth aspect, a trajectory planning apparatus is provided, comprising: a processor; and a memory storing computer program code executable on the processor; wherein the processor implements the steps of the method of the second aspect when executing the computer program.

[0015] In a seventh aspect, an aerial vehicle is provided, comprising: a body; a power device provided on the body, for providing power to the aerial vehicle; and the trajectory planning apparatus of the fifth aspect.

[0016] In an eighth aspect, a movable platform is provided, comprising: a body; a power device provided on the body, for providing power to the movable platform; and the trajectory planning apparatus of the sixth aspect.

[0017] In a ninth aspect, a trajectory planning system is provided, comprising the aerial vehicle of the seventh aspect and the movable platform of the eighth aspect.

[0018] In a tenth aspect, a computer readable storage medium is provided, having stored thereon computer instructions which, when executed by a processor, implement the steps of the method of the first aspect, the second aspect or the third aspect.

[0019] By means of the scheme provided in the application, the aerial vehicle can collect observation data in the air, generate a recommended path for the movable platform to be landed and provide the recommended path to the movable platform. The movable platform can plan a movement planning path based on the recommended path of the aerial vehicle, and then provide the movement planning path to the aerial vehicle, so that the aerial vehicle can determine a landing planning trajectory that can land on the movable platform more accurately and efficiently based on the movement planning path planned by the movable platform, and improve the accuracy and safety of dynamic landing of the aerial vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0021] FIG. 1 is a schematic diagram of a trajectory planning system according to an embodiment of the present application.

[0022] FIG. 2 is a schematic diagram of an aerial vehicle according to an embodiment of the present application.

[0023] FIG. 3A is a flowchart of a trajectory planning method according to an embodiment of the present application.

[0024] FIG. 3B is a schematic diagram of a trajectory planning method according to an embodiment of the present application.

[0025] FIG. 4 is a flowchart of a trajectory planning method according to an embodiment of the present application.

[0026] FIG. 5 is a flowchart of a trajectory planning method according to an embodiment of the present application.

[0027] FIG. 6 is a flowchart of a trajectory planning method according to an embodiment of the present application.

[0028] FIG. 7 is a structural diagram of a trajectory planning device according to an embodiment of the present application.

[0029] FIG. 8 is a structural diagram of a trajectory planning device according to an embodiment of the present application.

[0030] FIG. 9 is a structural diagram of an aerial vehicle according to an embodiment of the present application.

[0031] FIG. 10 is a structural diagram of a movable platform according to an embodiment of the present application. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments.

[0033] To solve the problems in the related art, the embodiments of the present application provide a trajectory planning system. Referring to FIG. 1, the trajectory planning system can include a movable platform 100 and an aircraft 200. In FIG. 1, for the convenience of illustration, the number of the movable platform 100 and the number of the aircraft 200 are both taken as an example of 1. It can be understood that in the actual scene, the number of the movable platform 100 and the number of the aircraft 200 can both be any number, and the embodiments of the present application do not limit this.

[0034] It will be apparent to those skilled in the art that any type of aircraft can be used without limitation, and the embodiments of the present application can be applied to various types of aircraft. For example, the aircraft can be a small or large aircraft. The aircraft can be a manned aircraft or an unmanned aircraft. In some embodiments, the aircraft can be a rotorcraft, for example, a multicopter aircraft propelled by a plurality of propulsion devices through air. The aircraft can also be a fixed-wing aircraft, and can also be a combination of a rotor and a fixed wing. The embodiments of the present application are not limited thereto, and the aircraft can also be other types of aircraft.

[0035] In an exemplary embodiment, referring to FIG. 2, the aircraft 200 can include a power system 210, a flight control system 220, a frame, and a gimbal 230 carried on the frame.

[0036] The frame can include a fuselage and a landing leg (also referred to as a landing gear). The fuselage can include a central frame and one or more arms connected to the central frame, and the one or more arms can extend radially from the central frame. The landing leg is connected to the fuselage and serves to support the aircraft 200 when it lands.

[0037] The power system 210 can include one or more electronic speed controllers (abbreviated as ESCs) 211, one or more propellers 213, and one or more motors 212 corresponding to the one or more propellers 213, where the motor 212 is connected between the ESC 211 and the propeller 213, and the motor 212 and the propeller 213 are disposed on the arm of the aerial vehicle 200; the ESC 211 is configured to receive a driving signal generated by the flight control system 220, and provide a driving current to the motor 212 according to the driving signal, so as to control the rotation speed of the motor 212. The motor 212 is configured to drive the propeller to rotate, thereby providing power for the flight of the aerial vehicle 200, and the power enables the aerial vehicle 200 to realize movement in one or more degrees of freedom. In some embodiments, the aerial vehicle 200 can rotate around one or more rotation axes. For example, the rotation axes can include a roll axis, a yaw axis, and a pitch axis. It should be understood that the motor 212 can be a direct current motor or an alternating current motor. In addition, the motor 212 can be a brushless motor or a brushed motor.

[0038] The flight control system 220 can include a flight controller 221 and a sensing system 222. The sensing system 222 is configured to measure attitude information of the aerial vehicle 200, i.e., position information and state information of the aerial vehicle 200 in space, such as three-dimensional position, three-dimensional angle, three-dimensional velocity, three-dimensional acceleration, and three-dimensional angular velocity, etc. The sensing system can also be configured to collect spatial environment to obtain observation data. The sensing system 222 can include one or more of a gyroscope, an ultrasonic sensor, an electronic compass, an inertial measurement unit (IMU), a vision sensor, an infrared sensor, a TOF (Time of Flight) sensor, a laser radar, a millimeter wave radar, a thermal imager, a global navigation satellite system, a barometer, etc. For example, the global navigation satellite system can be a global positioning system (GPS). The flight controller 221 is configured to control the flight of the aerial vehicle 200, for example, can control the flight of the aerial vehicle 200 according to the attitude information and the observation data measured by the sensing system 222. It should be understood that the flight controller 221 can control the aerial vehicle 200 according to pre-programmed instructions; or can also control the aerial vehicle 200 by responding to one or more remote control signals from other devices, which can be devices capable of communicating with the aerial vehicle 200, including but not limited to the movable platform 100, the remote control device of the aerial vehicle 200, the mobile device of the user, etc.

[0039] The gimbal 230 can include a motor 232. The gimbal is used to carry a photographing device 233. The flight controller 221 can control the motion of the gimbal 230 through the motor 232. It should be understood that the gimbal 230 can be independent of the aerial vehicle 200, or can be part of the aerial vehicle 200. It should be understood that the motor 232 can be a direct current motor, or an alternating current motor. In addition, the motor 232 can be a brushless motor, or a brushed motor. It should also be understood that the gimbal can be located on the top of the aerial vehicle 200, or on the bottom of the aerial vehicle 200.

[0040] The photographing device 233, for example, can be a camera or a video camera, or the like, for capturing images. The photographing device 233 can communicate with the flight controller, and take photographs under the control of the flight controller. The photographing device 233 of the present embodiment at least includes a photosensitive element, for example, a Complementary Metal Oxide Semiconductor (CMOS) sensor or a Charge-coupled Device (CCD) sensor. It can be understood that the photographing device 233 can also be directly fixed to the aerial vehicle 200, so that the gimbal 230 can be omitted.

[0041] The movable platform 100 of the present embodiment can refer to any device capable of movement, including but not limited to land vehicles, water vehicles, air vehicles, and other types of motorized conveyances. Taking a vehicle as an example, the vehicle generally includes a chassis, a body, an engine, and electrical equipment. The engine is a power device of the vehicle, for generating power; the chassis is used to support the transmitter and the body, and the chassis can drive the vehicle to move according to the power generated by the engine; the body is mounted on the frame of the chassis, for the driver, passengers to ride or load goods; the electrical equipment includes a power supply and an electrical device, for example, the power supply includes a storage battery and a generator, and the electrical device includes a starting system of the engine or other electrical devices. Optionally, the vehicle also includes a vehicle-mounted sensor for sensing environmental information of the surrounding environment of the vehicle; the vehicle-mounted sensor includes but is not limited to an image sensor, an infrared sensor, a thermal imager, a laser radar, a millimeter wave radar, or an ultrasonic sensor photographing device, etc., which can be used to collect observation data. Optionally, the vehicle also includes an auxiliary / automatic driving system for assisting the driver to drive or for automatic driving.

[0042] As an example, the movable platform 100 can be provided with a parking apron for parking the aircraft 200, and the aircraft 200 can take off from the parking apron and land in the parking apron of the aircraft 200. In actual scenarios, the parking apron for parking the aircraft 200 can be different based on the type of the movable platform 100. Taking the movable platform 100 as a vehicle example, the parking apron can be the top of the vehicle, such as a parking apron in the trunk of the vehicle, and the like; for vehicles such as pickup trucks, the parking apron can be the cargo area with an open area. For the movable platform 100 such as a ship, the parking apron can be the deck of the ship, and the like.

[0043] The aircraft 200 and the movable platform 100 both have the ability to communicate with each other, and the information interaction between the two can be relayed through a cloud server or through a remote control device of the aircraft 200; or a communication link is established between the aircraft 200 and the movable platform 100, and the two sides interact information based on the established communication link.

[0044] In some embodiments, as shown in FIG. 3A, it is a flowchart of a trajectory planning method in this embodiment, which can include the following steps:

[0045] In step 302, the aircraft 200 determines the recommended path of the movable platform 100 based on the collected observation data.

[0046] In step 304, the aircraft 200 sends the recommended path to the movable platform 100.

[0047] In step 306, the movable platform 100 determines the movement planning path of the movable platform 100 based on the recommended path, and sends the movement planning path to the aircraft 200;

[0048] In step 308, the aircraft 200 receives the movement planning path, and determines the landing planning trajectory for landing on the movable platform 100 based on the movement planning path.

[0049] In this embodiment, the aircraft 200 can collect observation data in the air, generate a recommended path for the movable platform 100 to be landed, and provide it to the movable platform 100. The movable platform 100 can plan its movement planning path based on the recommended path of the aircraft 200, and then provide it to the aircraft 200. The aircraft 200 can determine the landing planning trajectory that can land on the movable platform 100 more accurately and efficiently based on the movement planning path planned by the movable platform 100, thereby improving the accuracy and safety of the dynamic landing of the aircraft 200.

[0050] As an example, the aerial vehicle 200 can collect one or more observation data through one or more sensors in step 302, including but not limited to point cloud data collected by a lidar, image data collected by a vision sensor, data collected by an infrared sensor, data collected by a millimeter wave radar, data collected by an ultrasonic sensor, etc. The aerial vehicle 200 can identify one or more movable platforms 100 from the collected observation data, and determine a recommended path for the movable platform 100 based on the observation data.

[0051] The aerial vehicle 200 of the present embodiment can determine a recommended path for the movable platform 100, which can be a path determined by the aerial vehicle 200 based on the observation data, and suitable for the aerial vehicle 200 to safely or accurately land on the movable platform 100. For example, on the recommended path, there are relatively fewer low-altitude obstacles around the movable platform 100, the traffic density around the movable platform 100 is small, the movable platform 100 can move smoothly, etc. Thus, the movable platform 100 can plan a movement planning path based on the recommended path, which is more suitable for the aerial vehicle 200 to safely land, so that the movement planning path of the movable platform 100 can cooperate with the landing process of the aerial vehicle 200. The movable platform 100 subsequently provides the movement planning path to the aerial vehicle 200, thereby facilitating the aerial vehicle 200 to determine a landing planning trajectory for safely or accurately landing on the movable platform 100, and improving the success rate of landing of the aerial vehicle 200. In addition, the recommended path can also be a path determined by the aerial vehicle 200 based on the observation data, which is suitable for the movable platform 100 to safely and efficiently move. Since the aerial vehicle 200 can observe in the air, it can provide different or more distant observation ranges, such as observing environmental information such as obstacles, traffic density, etc. beyond the line of sight of the movable platform 100 in advance, and generating a recommended path based thereon, thereby helping the movable platform 100 to plan a movement planning path that is more suitable for the movable platform 100 to safely and efficiently move, so that the subsequent aerial vehicle 200 landing on the movable platform 100 can also be safer and more efficient.

[0052] Optionally, the movable platform 100 includes an autonomous driving system or a navigation system, which can plan a movement planning path for the movable platform 100, which can be a movement path calculated by the movable platform 100 from the current position to the target position according to its own movement ability and environmental conditions, etc. In the present embodiment, the movable platform 100 can also plan a movement planning path suitable for the aerial vehicle 200 to land in combination with the recommended path provided by the aerial vehicle 200, so that the aerial vehicle 200 can plan a landing planning trajectory for safely and accurately landing on the movable platform 100.

[0053] As an example, the aircraft 200 takes a certain time length to land on the mobile platform 100 which is in a moving state, and the aircraft 200 of the embodiment can obtain the moving planning path provided by the mobile platform 100 in real time and with high reliability, thereby improving the efficiency of planning the landing planning track and the success rate of landing, and improving the accuracy and safety of dynamic landing of the aircraft 200.

[0054] As an example, the aircraft 200 has a larger field of view than the mobile platform 100, and the recommended path of the mobile platform 100 determined by the aircraft 200 can be longer, so that after being provided to the mobile platform 100, the mobile platform 100 can obtain path information in a larger range. For example, the aircraft 200 can observe environmental information farther in front of the mobile platform 100. For example, 400 meters in front of the mobile platform 100 is a tunnel, and after passing through the tunnel, the number of lanes increases, and the aircraft 200 can plan a recommended path for the mobile platform 100 to drive to the rightmost lane after passing through the tunnel, because the light brightness is lower in the tunnel, and the risk of landing outside the tunnel is relatively lower. And after perceiving the surrounding environmental information of the lane, the aircraft 200 determines that there are more obstacles on the left lane, and the rightmost lane is more suitable for the aircraft 200 to land. The aircraft 200 provides the recommended path suitable for the aircraft 200 to land to the mobile platform 100, and the mobile platform 100 also plans a navigation path. However, because the field of view of the mobile platform 100 is limited, the mobile platform 100 cannot know the information of each lane and its surroundings after passing through the tunnel, so the navigation path plans to drive in the leftmost lane after passing through the tunnel. Based on the recommended path, the navigation path is optimized in the embodiment, and the originally planned driving in the leftmost lane is optimized to the rightmost lane in the recommended path, so that the optimized moving planning path suitable for the aircraft 200 to land is obtained. Therefore, after receiving the moving planning path of the mobile platform 100, the aircraft 200 can efficiently plan the landing planning track to safely land on the mobile platform 100.

[0055] In some examples, after receiving the recommended path, the mobile platform 100 can output a reminder to the user that the recommended path is received. The output method can be voice output, or image or text output in the display system of the mobile platform 100. As an example, the mobile platform 100 can output "the current navigation path is being optimized based on the recommended path of the aircraft 200". Or, it can also output "whether to select the recommended path of the aircraft 200 to make the aircraft 200 land" and other types of reminders.

[0056] Further, the movable platform 100 can also receive a confirmation instruction from the user. For example, the movable platform 100 can comprise at least one input device. The input device can comprise one or more external input devices and / or one or more built-in input devices. The instruction from the user can be generated by the user manipulating the built-in input device or the external input device within the movable platform 100. After receiving the confirmation instruction from the user, the movable platform 100 confirms the planning of the movement planning path based on the recommended path.

[0057] Optionally, in the case where the aerial vehicle 200 identifies a plurality of movable platforms 100 from the collected observation data, the aerial vehicle 200 can:

[0058] In some examples, the aerial vehicle 200 can determine a recommended path for one of the movable platforms 100, and send the recommended path to the movable platform 100, so that the movable platform 100 determines a movement planning path based on the recommended path and sends the movement planning path to the aerial vehicle 200, and then the aerial vehicle 200 determines a landing planning trajectory for landing on the movable platform 100 based on the movement planning path.

[0059] In other examples, the aerial vehicle 200 can determine a recommended path for each of the plurality of movable platforms 100, and send the recommended path to the corresponding movable platform 100, so that each of the movable platforms 100 determines a movement planning path based on the recommended path and sends the movement planning path to the aerial vehicle 200. Then, the aerial vehicle 200 can determine a landing planning trajectory for landing on one of the plurality of movable platforms 100 based on the movement planning paths.

[0060] Optionally, as to the timing of starting to execute the path planning method of the present embodiment, in actual scenarios, the aerial vehicle 200 can start to execute the path planning method of the present embodiment after detecting a landing instruction. The landing instruction can be generated by the aerial vehicle 200. For example, the aerial vehicle 200 can generate the landing instruction after completing a flight task. Or the aerial vehicle 200 can generate the landing instruction after detecting that a set landing condition is met. Optionally, the set landing condition can be various. For example, the aerial vehicle 200 can detect that the power is lower than a set power threshold, or detect that it is in a dangerous environment, or detect that it is about to enter a no-fly zone, etc. Or, the landing instruction can be issued by the user. For example, the user can issue the instruction by using a remote control device of the movable platform 100 or the aerial vehicle 200, or a client in communication connection with the aerial vehicle 200.

[0061] Optionally, the method of steps 302 to 308 in this embodiment can be executed in a loop during the process that the aerial vehicle 200 is in the air to finally land on the movable platform 100. For example, at T1, the aerial vehicle 200 starts to execute the method in this embodiment to obtain a landing planning trajectory of the aerial vehicle 200 in response to a landing instruction, and then the aerial vehicle 200 starts to land based on the landing planning trajectory. Since the landing needs to last for a certain period of time, during the landing process, the movable platform 100 and the aerial vehicle 200 are dynamically changing, and the surrounding environment of the movable platform 100 and the surrounding environment of the aerial vehicle 200 are also dynamically changing. Therefore, at T2 (at this time, the aerial vehicle 200 has not landed on the movable platform 100), the method in this embodiment can be executed again to obtain a landing planning trajectory that is newly planned at T2, and the aerial vehicle 200 continues to land based on the newly planned landing planning trajectory. In this way, until the aerial vehicle 200 finally successfully lands on the movable platform 100.

[0062] When the aerial vehicle 200 determines the landing planning trajectory based on the movement planning path, there can be various implementation manners. For example, the landing planning trajectory can be determined based on only the movement planning path. The movement planning path of the movable platform 100 can indicate trajectory points (the trajectory points can be position information or pose information) of the movable platform 100 at multiple future time points, and the aerial vehicle 200 can plan to land on the movement planning path of the movable platform 100 based on this.

[0063] Alternatively, the landing planning trajectory can also be determined based on the movement planning path in combination with one or more combinations of the following information: the recommended path, the movable platform observation data sent by the movable platform 100, and the aerial vehicle observation data, which are described in the following.

[0064] (1) In a first manner, the landing planning trajectory can be determined based on the recommended path and the movement planning path at the same time, so as to improve the accuracy of the landing trajectory planning.

[0065] In some examples, the landing planning trajectory that lands on the movable platform 100 can be determined based on the recommended path and the movement planning path, which can include: fusing the recommended path and the movement planning path, and determining the landing planning trajectory that lands on the movable platform 100 according to the fusion result.

[0066] The fusion manner of the recommended path and the movement planning path can be set as needed, and this embodiment does not limit this. For example, the recommended path and the movement planning path can be compared to determine the difference between the two paths, and the landing planning trajectory can be planned according to the comparison result.

[0067] Alternatively, the recommended path can indicate trajectory points of the movable platform 100 predicted by the aircraft 200 at multiple future time points, and the movement planning path can be a trajectory path planned by the movable platform 100 at multiple future time points. The trajectory points of the movable platform 100 at multiple future time points included in the two paths can be fused. For example, the same trajectory points of the movable platform 100 at the same future time points in the two paths can be retained, and the trajectory points of the movable platform 100 at the same future time points that are significantly different can be fused.

[0068] Based on this, the fusion result of the recommended path and the movement planning path can determine a landing planning trajectory with higher reliability to improve the success rate of the aircraft 200 landing.

[0069] In some examples, fusing the recommended path and the movement planning path, and determining the landing planning trajectory for landing on the movable platform 100 according to the fusion result can include: performing weighted calculation on the trajectory points in the recommended path and the trajectory points in the movement planning path, and determining the landing planning trajectory for landing on the movable platform 100 according to the calculation result.

[0070] The weights of the trajectory points in the recommended path and the trajectory points in the movement planning path correspond to each other, which can be set according to actual needs, and the embodiment is not limited in this regard. For example, the weight of the trajectory points in the movement planning path can be higher than the weight of the trajectory points in the recommended path, and the like.

[0071] As an example, a weighted average calculation method can be used. The weighted average calculation can be performed on all trajectory points included in the recommended path and all trajectory points included in the movement planning path. Alternatively, multiple time windows can be divided for the recommended path and the movement planning path, the same or different weights can be used in each time window, and the weighted average calculation can be performed on the trajectory points of the two paths in each time window.

[0072] (2) The second method can determine the landing planning trajectory based on the movable platform observation data and the recommended path to improve the success rate of the aircraft 200 landing.

[0073] As an example, the aircraft 200 can obtain the movable platform observation data sent by the movable platform 100. The landing planning trajectory for landing on the movable platform 100 can be determined based on the movement planning path and the movable platform observation data.

[0074] In the embodiment, the movable platform 100 can send the movable platform observation data to the aircraft 200. The movable platform 100 can send the movable platform observation data to the aircraft 200 at the same time as the movement planning path of the movable platform 100, or the two can be sent separately. The sending frequency of the observation data by the movable platform 100 can also be set according to actual needs.

[0075] The observation range of the movable platform 100 is different from the observation range of the aerial vehicle 200. The movable platform 100 can accurately sense the surrounding low-altitude environment information. The aerial vehicle 200 can also have the ability to cover the observation range of the movable platform 100 in the air. However, because the aerial vehicle 200 is at a distance from the movable platform 100, the movable platform observation data can provide the aerial vehicle 200 with more abundant, clear, and reliable surrounding low-altitude environment information of the movable platform 100.

[0076] Based on this, in the embodiment, the aerial vehicle 200 can determine a landing planning trajectory based on the movable platform observation data and the recommended path at the same time, so that the aerial vehicle 200 is safer and more reliable when landing based on the landing planning trajectory, and the success rate of landing of the aerial vehicle 200 is improved.

[0077] In some examples, the movable platform observation data can include at least one of the following: information of low-altitude obstacles in the direction of the movement planning path, information of surrounding movable platforms 100. As an example, the information of low-altitude obstacles can be obstacle information appearing in the observation range of the sensor of the movable platform 100. The information contained in the movable platform observation data can be identified by the movable platform 100 from the collected data of each sensor and provided to the aerial vehicle 200 in the observation data. The movable platform 100 can also obtain the collected data of each sensor and provide it to the aerial vehicle 200 as movable platform observation data, and the aerial vehicle 200 can identify the information from the observation data.

[0078] In some examples, the information of low-altitude obstacles can include at least one of the following: movement information of moving obstacles, position information of stationary obstacles. Based on different application scenarios, the moving obstacles and the stationary obstacles can be various. For example, the moving obstacles can be pedestrians, etc., and the stationary obstacles can be trees, road signs, traffic lights, or power poles, etc. Based on this, the embodiment can provide the aerial vehicle 200 with the information of low-altitude obstacles around the movable platform 100 to improve the success rate of landing of the aerial vehicle 200.

[0079] In some examples, the information about the surrounding movable platform 100 can include at least one of: motion information of the movable platform 100 in front, density information of the surrounding movable platform 100. Considering that there can be other movable platforms 100 around the movable platform 100 in the actual scene, the movable platform observation data in this embodiment also includes the motion information of the movable platform 100 in front and the density information of the surrounding movable platform 100. The aircraft 200 can determine the future pose of the movable platform 100 to be landed more accurately based on the motion information of the movable platform 100 in front, and can accurately determine the obstacles that can be encountered during landing based on the density information of the surrounding movable platform 100, so as to improve the success rate of landing of the aircraft 200.

[0080] In actual implementation, there can be multiple ways to determine the landing planning trajectory for landing on the movable platform 100 based on the movement planning path and the movable platform observation data, such as directly determining the landing planning trajectory based on the movement planning path and the movable platform observation data.

[0081] Alternatively, in some examples, determining the landing planning trajectory for landing on the movable platform 100 based on the movement planning path and the movable platform observation data can include: determining a preliminary landing planning trajectory based on the movement planning path; and optimizing the preliminary landing planning trajectory based on the movable platform observation data to obtain the landing planning trajectory.

[0082] In this embodiment, the preliminary landing planning trajectory can be determined based on the movement planning path first, and then the preliminary landing planning trajectory can be optimized based on the movable platform observation data to obtain the landing planning trajectory, which can improve the success rate of landing of the aircraft 200.

[0083] As an example, the trajectory points in the landing planning trajectory can be optimized based on the information of low-altitude obstacles such as road signs and power poles or the density information of the surrounding movable platform 100, such as optimizing the pose information in the trajectory points.

[0084] As an example, the preliminary landing planning trajectory of the aircraft 200 includes a first trajectory segment for continuously reducing the height of the aircraft 200 from the current high-altitude position, and a second trajectory segment for approaching the movable platform 100 at a low-altitude position and landing on the movable platform 100 from the low-altitude position. The movable platform observation data mainly includes low-altitude environment information around the movable platform 100. Therefore, the second trajectory segment can be optimized using the movable platform observation data, so that the aircraft 200 can safely avoid obstacles and safely land on the movable platform 100 at the low-altitude position.

[0085] (3) The third mode can determine the landing planning trajectory based on the aircraft observation data and the moving planning path, so as to improve the success rate of the landing of the aircraft 200.

[0086] In this embodiment, after the aircraft 200 obtains the moving planning path of the movable platform 100, the aircraft 200 can continuously acquire the aircraft observation data when planning the landing planning trajectory, perceive the information such as obstacles around the aircraft 200 through the aircraft observation data, and thus plan a safe and reliable landing planning trajectory.

[0087] For example, the planning of the landing planning trajectory includes the position planning of the landing trajectory points and the speed planning corresponding to the landing trajectory points. Based on this, the landing planning trajectory planned in this embodiment can enable the aircraft 200 to land on the movable platform 100 at a safe and reliable position and speed.

[0088] In some examples, the aircraft observation data includes at least one of the following: obstacle information near the movable platform 100, obstacle information in the landing direction of the aircraft 200, moving speed information of the movable platform 100, and density information of the movable platform 100 around the movable platform 100.

[0089] For example, the image sensor of the aircraft 200 can continuously acquire image data at a set acquisition frequency, and the aircraft 200 can identify the obstacles and the movable platform 100 from the continuously acquired image data. For the moving obstacles and the movable platform 100, the speed information of the moving obstacles and the moving speed information of the movable platform 100 can also be determined according to the positions of the identified moving obstacles and the movable platform 100 in the images acquired at different time points. The density information of the movable platform 100 around the movable platform 100 can also be determined based on the number of the movable platform 100 identified in the image data.

[0090] As another example, the laser radar of the aircraft 200 can acquire point cloud data at a set frequency. Similarly, the aircraft 200 can identify the obstacles and the movable platform 100 from the continuously acquired point cloud data, and further obtain the obstacle information, the moving speed information of the movable platform 100, and the density information of the movable platform 100.

[0091] The aircraft 200 can also carry other sensors, and the above-mentioned embodiment can also determine one or more types of information through the continuously acquired data.

[0092] Based on this, through the various types of information included in the above-mentioned aircraft observation data, the aircraft 200 can determine a safe and reliable landing planning trajectory to improve the success rate of the landing of the aircraft 200.

[0093] In actual implementation, the landing planning trajectory for landing on the movable platform 100 can be determined in various manners based on the mobile planning path and the aircraft observation data. For example, the landing planning trajectory can be directly determined based on the mobile planning path and the aircraft observation data.

[0094] Alternatively, in some examples, determining the landing planning trajectory for landing on the movable platform 100 based on the aircraft observation data and the mobile planning path includes: determining a preliminary landing planning trajectory based on the mobile planning path; and optimizing the preliminary landing planning trajectory based on the aircraft observation data to obtain the landing planning trajectory.

[0095] In this embodiment, the preliminary landing planning trajectory can be determined based on the mobile planning path first, and then the preliminary landing planning trajectory can be optimized based on the aircraft observation data to obtain the landing planning trajectory, which can improve the success rate of landing of the aircraft 200.

[0096] As an example, the future pose of the aircraft 200 in the preliminary landing planning trajectory can be optimized, or the speed of the aircraft 200 represented by the preliminary landing planning trajectory can be optimized, based on information such as obstacle information near the movable platform 100, obstacle information in the landing direction of the aircraft 200, speed information of the movable platform 100, or density information of the movable platform 100 around the movable platform 100, to determine a safe and reliable landing planning trajectory.

[0097] As an example, the aircraft 200 identifies from the aircraft observation data that the movable platform 100 is about to drive to a fork (for example, the intersection of two highways) of a bend, and the aircraft 200 judges the driving direction of the movable platform 100 in advance to optimize the landing planning trajectory. For example, the aircraft 200 can directly go straight and fly to the vicinity of the fork to be selected by the movable platform 100.

[0098] In some examples, optimizing the preliminary landing planning trajectory based on the aircraft observation data can include optimizing the position information of the trajectory points of the preliminary landing planning trajectory based on the aircraft observation data. In this embodiment, the position information of the trajectory points of the preliminary landing planning trajectory can be optimized based on the aircraft observation data, so that the aircraft 200 can reach a safer position when landing, thereby ensuring the safety of landing of the aircraft 200.

[0099] In some examples, the preliminary landing planning trajectory can be optimized based on the aircraft observation data, including: optimizing the speed information of the trajectory points of the preliminary landing planning trajectory based on the aircraft observation data. In this embodiment, the speed information of the trajectory points of the preliminary landing planning trajectory can be optimized based on the aircraft observation data, so that the aircraft 200 can land at a safe speed, thereby ensuring the safety of landing of the aircraft 200.

[0100] The above provides three embodiments of the aircraft 200 planning the landing planning trajectory. It can be understood that in actual implementation, the aircraft 200 can also combine the above three ways as needed to determine the landing planning trajectory. For example, the first way and the second way can be combined to plan the landing planning trajectory, the first way and the third way can be combined to plan the landing planning trajectory, the second way and the third way can be combined to plan the landing planning trajectory, or the three ways can be combined to plan the landing planning trajectory.

[0101] In some examples, based on the movement planning path, determining the landing planning trajectory for landing on the movable platform 100 can include: based on the movement planning path, determining a future pose of the movable platform 100; obtaining a current pose of the aircraft 200; based on the current pose of the aircraft 200 and the future pose of the movable platform 100, determining the landing planning trajectory for landing on the movable platform 100.

[0102] As an example, the pose can include position information and attitude information. Specifically, the future pose of the movable platform 100 can be determined from the movement planning path sent by the movable platform 100, such as one or more combinations of geographic position coordinates, height, speed, acceleration, or attitude angle.

[0103] As an example, the aircraft 200 can obtain its current pose information in real time through sensors such as GPS or inertial measurement unit carried by the aircraft 200, such as one or more combinations of geographic position coordinates, height, speed, acceleration, or attitude angle (such as pitch angle, roll angle, or yaw angle).

[0104] As an example, the future pose of the movable platform 100 and the current pose of the aircraft 200 can be input to calculate an optimal landing planning trajectory through an embedded trajectory planning algorithm, that is, a trajectory from the current pose of the aircraft 200 to the future pose of the movable platform 100. Based on this, the embodiment can determine a safe and reliable landing planning trajectory by determining the future pose of the movable platform 100 and the current pose of the aircraft 200.

[0105] In some examples, based on the current pose of the aircraft 200 and the future pose of the movable platform 100, determining the landing planning trajectory for landing on the movable platform 100 can include: determining a starting point of the landing planning trajectory to be determined based on the current pose of the aircraft 200; determining an end point of the landing planning trajectory to be determined based on the future pose of the movable platform 100; determining the landing planning trajectory from the starting point to the end point.

[0106] As an example, the geographic position information contained in the current pose of the aerial vehicle 200 can be determined as the starting point of the landing planning trajectory to be determined. The geographic position information contained in the future pose of the movable platform 100 can be determined as the ending point of the landing planning trajectory to be determined, which represents the expected position and attitude of the movable platform 100 at a specific future time point. The ending point of the landing planning trajectory is determined based on the future pose of the movable platform 100, which can ensure that the ending point of the landing planning trajectory matches the expected position of the movable platform 100. Based on the starting point and the ending point, an optimal landing planning trajectory from the starting point to the ending point can be calculated by using a path planning algorithm or the like, so as to determine the aerial vehicle 200 to safely land at the precise position.

[0107] In some examples, determining the landing planning trajectory from the starting point to the ending point can include determining at least one intermediate point between the starting point and the ending point, and determining the landing planning trajectory from the starting point, through the intermediate point, to the ending point.

[0108] In this embodiment, in addition to the starting point and the ending point, intermediate points can also be determined, and the number of intermediate points can be set according to actual needs. These intermediate points are trajectory points that the aerial vehicle 200 needs to consider when planning the landing planning trajectory, and the landing planning trajectory will pass through these intermediate points. The determination of the intermediate points can be based on various factors, including but not limited to the flight performance of the aerial vehicle 200, the expected flight time, the user's settings, the flight task, the movement speed of the movable platform 100, the environmental obstacles, and the possible wind speed and direction changes, and the like. Based on this, the landing trajectory planning manner considering the intermediate points provided by this embodiment can ensure that the aerial vehicle 200 safely and accurately lands on the movable platform 100 in a complex dynamic environment, or meets the specific needs of the user, and the like.

[0109] In some examples, determining the at least one intermediate point between the starting point and the ending point can include obtaining a combination of one or more of the following trajectory points: a trajectory point for obstacle avoidance of the aerial vehicle 200, a trajectory point set by the user, or a trajectory point of a flight task performed by the aerial vehicle 200 during the landing process; and determining the at least one intermediate point between the starting point and the ending point based on the obtained trajectory points.

[0110] In this embodiment, the trajectory points for obstacle avoidance of the aerial vehicle 200 can be obtained to determine the intermediate points based on the need for obstacle avoidance, so that the planned movement planning trajectory can avoid obstacles. The trajectory points set by the user can also be obtained to determine the intermediate points based on the need to meet the user's needs, so that the planned movement planning trajectory can pass through the trajectory points set by the user, thereby meeting the user's needs.

[0111] Alternatively, the aerial vehicle 200 can need to perform one or more specific flight tasks during the landing process, and embodiments are not limited in this regard. By way of example, the flight tasks can include, but are not limited to, one or more of a combination of aerial photography tasks, environmental information collection tasks, inspection tasks, item delivery tasks, or flight game tasks. For example, the aerial photography tasks can use at least one camera to collect aerial photography data; the environmental information collection tasks can use at least one camera and one or more sensors for sensing the surrounding environment to collect environmental information (e.g., traffic information, information of other movable platforms, etc.). The inspection tasks can be that the aerial vehicle 200 flies to certain specific areas and identifies the state of specific targets in the specific areas through observation data, such as identifying the state of a cable, etc. The item delivery tasks of the aerial vehicle 200 can be that the aerial vehicle 200 carries items and delivers them to a specified location (e.g., a certain fixed geographic location or other movable platform 100, etc.). The flight game tasks can use at least one camera to collect real-time pictures, or use at least one audio collection component to collect real-time audio data, or use at least one camera and at least one audio collection component to collect real-time audio and video data. In actual implementation, the aerial vehicle 200 can be controlled to perform different types of flight tasks according to actual needs, so as to obtain different trajectory points and then obtain different intermediate points.

[0112] In some examples, determining the landing planning trajectory from the starting point, through the intermediate point, to the ending point includes: dividing the landing planning trajectory to be determined into a plurality of sub-landing planning trajectories to be determined based on the starting point, the intermediate point, and the ending point; and planning each of the sub-landing planning trajectories, thereby obtaining the landing planning trajectory from the starting point, through the intermediate point, to the ending point.

[0113] By way of example, the number of sub-landing planning trajectories can be determined according to actual needs, and can be determined based on one or more of the number of the starting point, the intermediate point, and the ending point, or the distance between the starting point and the ending point, or the total landing time, or the surrounding environment, etc. Embodiments are not limited in this regard. The entire landing planning trajectory to be planned can be divided into a plurality of continuous sub-trajectories, i.e., the sub-trajectories can be spliced to obtain the landing planning trajectory.

[0114] Each of the divided sub-trajectories to be planned can have a corresponding starting point and ending point, so that embodiments can plan each of the sub-trajectories to be planned respectively, for example, can be sequentially planned by a path planning algorithm, and can also be parallelly planned, thereby improving the generation efficiency of the landing planning trajectory.

[0115] By dividing the landing planning trajectory into multiple sub-trajectories to be planned, fine path planning can be performed to obtain an accurate landing planning trajectory, thereby improving the success rate of landing of the aircraft 200 and ensuring safe landing of the aircraft 200.

[0116] In some examples, planning the sub-landing planning trajectory of each segment includes: determining a total landing time from the starting point to the ending point, and assigning a landing sub-time to each sub-landing planning trajectory to be determined according to the total landing time; and planning the sub-landing planning trajectory of each segment according to the landing sub-time of each sub-landing planning trajectory to be determined.

[0117] As an example, the total landing time of the entire landing process from the starting point to the ending point can be determined according to the flight performance of the aircraft 200, the landing requirements, the moving speed of the movable platform 100, the environmental information, the needs of the user, and other related factors, which are not limited in this embodiment. After obtaining the total landing time, a landing sub-time is assigned to each sub-landing planning trajectory, and the landing sub-time of each landing planning trajectory can be the same or different, which can be configured according to actual needs. For example, when assigning, factors such as the complexity of each sub-trajectory, environmental information, or the expected flight speed of the aircraft 200 in the segment can be considered for assignment. For example, if a certain sub-landing planning trajectory needs to avoid obstacles or perform specific flight maneuvers, a longer landing sub-time can be assigned to the segment to ensure sufficient time to complete these operations. For example, in the multiple sub-landing planning trajectories, the landing sub-time assigned to some sub-landing planning trajectories starting from the starting point can be shorter because the aircraft 200 is farther away from the movable platform 100 in the air; and in the last few sub-landing planning trajectories, the aircraft 200 approaches the movable platform 100, and a larger landing sub-time can be assigned to make the aircraft 200 land more safely.

[0118] For each sub-landing planning trajectory, the aircraft 200 can be planned in detail according to the assigned landing sub-time, based on which the final landing planning trajectory can meet the time length requirement of the total landing time and can be finely planned based on the landing sub-time of each segment.

[0119] In some examples, the landing sub-time of each sub-landing planning trajectory to be determined is determined according to a preset condition. The preset condition can be one or more, which can be pre-configured in the aircraft 200, or can be generated in real time by the aircraft 200 during planning, or can be received in real time. Based on this, this embodiment can assign a reasonable landing sub-time to each sub-landing planning trajectory to be determined based on the preset condition, to ensure that an accurate landing planning trajectory is planned.

[0120] In some examples, the preset condition includes a condition determined based on the obstacle avoidance information and / or a condition determined based on the motion speed information of the movable platform 100.

[0121] For example, if some to-be-determined sub-descent planning trajectory contains a trajectory point for the aircraft 200 to avoid obstacles, the descent sub-duration of the segment of the sub-descent planning trajectory can be relatively large, for example, higher than those sub-trajectories that do not contain trajectory points for the aircraft 200 to avoid obstacles.

[0122] For example, the movement planning path of the movable platform 100 can represent the movement speed of the movable platform 100 at different times. In order to cooperate with the movable platform 100 and safely land on the movable platform 100, the possible position change of the movable platform 100 during the landing of the aircraft 200 can be determined based on the motion speed information of the movable platform 100, so as to assign appropriate descent sub-durations to each segment of the sub-descent planning trajectory.

[0123] Based on this, the embodiment can plan each sub-descent planning trajectory based on the obstacle avoidance requirement to ensure the safe landing of the aircraft 200. In addition, the embodiment can plan each sub-descent planning trajectory based on the motion speed information of the movable platform 100 to ensure that the aircraft 200 has enough time to adapt to the movement of the movable platform 100, thereby ensuring that the aircraft 200 can land accurately on the movable platform 100.

[0124] In some examples, the aircraft 200 can also send the aircraft observation data and the recommended path to the movable platform 100 to improve the accuracy of the movement planning path of the movable platform 100. For example, the method can further include: sending the aircraft observation data to the movable platform 100; and the movement planning path planned based on the recommended path can include: a movement planning path planned based on the aircraft observation data and the recommended path.

[0125] The movable platform 100 can also carry sensors to collect surrounding environment information, but the field of view of the sensors of the movable platform 100 is limited. The aircraft observation data of the embodiment can provide the movable platform 100 with environment information in a larger or different field of view, so that the movable platform 100 can plan a safer and more suitable movement planning path for the aircraft 200 to land based on the multiple information contained in the aircraft observation data and the recommended path, thereby helping the aircraft 200 to plan a safer and more reliable descent planning trajectory.

[0126] As an example, the aerial vehicle observation data can be data collected by sensors onboard the aerial vehicle 200, or data processed from the data collected by sensors onboard the aerial vehicle 200. The aerial vehicle observation data can include various information. In some examples, the aerial vehicle observation data can include at least one of the following: obstacle information around the movable platform 100, obstacle information in the direction of movement of the movable platform 100, movement information of the movable platform 100, movement information of the movable platform 100 around the movable platform 100.

[0127] As an example, the image sensor of the aerial vehicle 200 can continuously collect image data at a set frequency, and the aerial vehicle 200 can identify one or more movable platforms 100 from the continuously collected image data. The identified movable platforms 100 include the movable platform 100 to be landed, and other movable platforms 100. The obstacle information around the movable platform 100 to be landed, or the obstacle information in the direction of movement of the movable platform 100 to be landed can be identified. According to the positions of the identified movable platforms 100 in the images collected at different times, the movement information of each movable platform 100, such as the movement information of the movable platform 100 to be landed and the movement information of the movable platforms 100 around the movable platform 100 to be landed, can be determined.

[0128] As another example, the laser radar of the aerial vehicle 200 can collect point cloud data at a set frequency. Similarly, the aerial vehicle 200 can identify the movable platforms 100 from the continuously collected point cloud data, and can further obtain the above-mentioned information. Other sensors can also be carried on the aerial vehicle 200, and the above-mentioned one or more information can also be determined from the continuously collected data according to the above-mentioned embodiments.

[0129] In some examples, the aerial vehicle observation data includes sensing data from the bird's eye view of the aerial vehicle 200. The sensing data includes at least one of the following: laser radar data, visual image data, microwave radar data, ultrasonic data. In this embodiment, since the sensing data from the bird's eye view (BEV) of the aerial vehicle 200 can be provided to the movable platform 100, the perception range of the movable platform 100 to the environment can be expanded, the blind area of the movable platform 100 can be reduced, and a safer movement planning path can be planned by obtaining more obstacle information around the movable platform 100.

[0130] In some examples, the movement planning path planned based on the aerial vehicle observation data and the recommended path includes a movement planning path obtained by optimizing a preliminary movement planning path of the movable platform 100 based on the aerial vehicle observation data and the recommended path.

[0131] As an example, the preliminary movement planning path of the movable platform 100 can be generated by an autonomous driving system or a navigation system, and the embodiment can optimize the preliminary movement planning path of the movable platform 100 based on the aerial vehicle observation data and the recommended path, so that the movement planning path obtained after optimization is more suitable for the aerial vehicle 200 to land.

[0132] In some examples, the preliminary movement planning path is determined based on input information of a user or / and sensing information of the movable platform 100. As an example, the autonomous driving system or the navigation system of the movable platform 100 can generate the preliminary movement planning path from the current position to the destination based on the input information of the user, such as the destination information, etc., so that the preliminary movement planning path meets the needs of the user. As an example, the autonomous driving system or the navigation system of the movable platform 100 can determine the preliminary movement planning path based on the sensing information of the movable platform 100, so that the preliminary movement planning path can avoid obstacles. As an example, the movable platform 100 can also determine the preliminary movement planning path based on the input information of the user and the sensing information of the movable platform 100.

[0133] In some examples, the aerial vehicle 200 can communicate with multiple movable platforms 100, and the method can further include determining the movable platform 100 to be landed from the multiple movable platforms 100. In actual scenarios, the aerial vehicle 200 can communicate with multiple movable platforms 100, and the aerial vehicle 200 can determine the movable platform 100 to be landed from the multiple movable platforms 100.

[0134] Among them, the determination method can be various, and the embodiment does not limit it. As an example, the aerial vehicle 200 can determine the movable platform 100 to be landed specified by the user. Or, the aerial vehicle 200 can determine it by itself based on some factors, for example, it can determine the movable platform 100 most suitable for landing from the multiple movable platforms 100 to ensure the success rate of landing. Or, it can determine the closest movable platform 100 from the multiple movable platforms 100 to achieve the fastest landing. Or, it can determine the movable platform 100 with the strongest communication signal from the multiple movable platforms 100 to ensure as much as possible that the movable platform 100 and the aerial vehicle 200 can keep communication during the landing process, thereby ensuring the success rate of the aerial vehicle 200 landing.

[0135] In some examples, based on the aerial vehicle observation data collected, the recommended path of the movable platform 100 is determined, including: determining the movable platform 100 to be landed based on the aerial vehicle observation data, and determining the recommended path of the movable platform 100 to be landed.

[0136] As an example, the aerial vehicle 200 can determine the movable platform 100 to be landed based on aerial vehicle observation data. For example, the aerial vehicle 200 can identify one or more movable platforms 100 based on sensor collected data such as image data of an image sensor and / or point cloud data of a lidar. For example, the aerial vehicle 200 can communicate with multiple movable platforms 100, but only observe one movable platform 100 in the current field of view, and thus take the observed movable platform 100 as the movable platform 100 to be landed.

[0137] In some examples, determining the movable platform 100 to be landed based on aerial vehicle observation data includes: identifying multiple movable platforms 100 based on the aerial vehicle observation data, and determining the movable platform 100 to be landed from the identified multiple movable platforms 100.

[0138] In the embodiment, the aerial vehicle 200 can identify multiple movable platforms 100 from the observation data, and thus can determine the movable platform 100 to be landed from the identified multiple movable platforms 100.

[0139] Further, the determined movable platform 100 to be landed can be determined with a recommended path. In the embodiment, the aerial vehicle 200 can determine the recommended path of the movable platform 100 based on the collected aerial vehicle observation data, and thus can reduce user operation and autonomously complete landing.

[0140] In actual applications, there can be a situation that the aerial vehicle 200 loses communication with the movable platform 100 during landing. To improve the safety of the aerial vehicle 200, the aerial vehicle 200 can switch to a non-interactive mode. As an example, the method can further include: in response to disconnection of communication with the movable platform 100, switching to a preset non-interactive mode; and in the non-interactive mode, re-determining a landing planning trajectory for landing on the movable platform 100 based on aerial vehicle observation data.

[0141] As an example, the aerial vehicle 200 continuously monitors the communication state between the aerial vehicle 200 and the movable platform 100. In actual applications, there can be multiple reasons for the aerial vehicle 200 to lose communication with the movable platform 100 during landing. For example, the movable platform 100 enters a tunnel or other signal-poor range, or there is a sudden failure of network infrastructure, or network congestion, or a sudden failure of the communication module of the movable platform 100 or the aerial vehicle 200. Therefore, the embodiment can set a non-interactive mode for the aerial vehicle 200, so that the aerial vehicle 200 can switch to the non-interactive mode after disconnection of communication with the movable platform 100. The foregoing embodiments, such as the embodiments shown in steps 302 to 308, can be referred to as an interactive mode.

[0142] In the non-interaction mode, the aircraft 200 fails to obtain the movement planning path of the movable platform 100, but the aircraft 200 can observe the movable platform 100 in the air, and in the process of observing the movable platform 100 in the air, the aircraft 200 can obtain aircraft observation data, can determine the movement state of the movable platform 100 and the environmental information around the movable platform 100 based on the aircraft observation data, and thus can redetermine the landing planning trajectory for landing on the movable platform 100, so that the aircraft 200 can still plan the landing planning trajectory by itself in the case of disconnection with the movable platform 100, and realize the landing of the aircraft 200 on the movable platform 100.

[0143] In some examples, the aircraft observation data includes at least one of the following: current position information of the movable platform 100, current movement direction information of the movable platform 100, and movement trend information of the movable platform 100.

[0144] As an example, the current position information of the movable platform 100 can be absolute position information, such as three-dimensional space coordinates containing latitude, longitude and height, or relative position information relative to the aircraft 200. The current movement direction information of the movable platform 100 can represent the direction in which the movable platform 100 is currently moving. The movement trend information of the movable platform 100 can include speed, acceleration, possible turning or movement mode (such as straight line movement, curve movement, acceleration or deceleration, etc.), and can represent the future movement state of the movable platform 100.

[0145] Through the above information, the aircraft 200 can track the movable platform 100 in real time, and can also plan a landing planning trajectory that can accurately land on the movable platform 100 based on the information.

[0146] In some examples, the method can further include: in response to the communication connection with the movable platform 100, exiting the preset non-interaction mode, and determining the landing planning trajectory for landing on the movable platform 100 based on the movement planning path.

[0147] In actual application, after the aircraft 200 loses the communication with the movable platform 100 during the landing process of the aircraft 200, the aircraft 200 can continue to attempt to restore the communication connection with the movable platform 100 during the landing process of the aircraft 200 in the non-interaction mode. If the aircraft 200 successfully re-establishes the communication connection with the movable platform 100, the aircraft 200 can land more safely and accurately with the cooperation of the movable platform 100. Based on this, the aircraft 200 can exit the non-interaction mode and switch to the aforementioned interaction mode again, i.e., re-determine the landing planning trajectory for landing on the movable platform 100 based on the planned landing planning trajectory. Thus, after the communication connection with the movable platform 100 is restored, the aircraft 200 lands based on the newly planned landing planning trajectory with the cooperation of the movable platform 100, thereby ensuring the success rate of the landing of the aircraft 200.

[0148] In some examples, to improve the safety of the aircraft 200, the method can further include: in response to the disconnection of the communication with the movable platform 100 or the failure to identify the movable platform 100 from the aircraft observation data, increasing the flight height or flying to a specific position.

[0149] In this embodiment, if the aircraft 200 loses the communication with the movable platform 100 or fails to identify the movable platform 100 from the aircraft observation data, the aircraft 200 can also increase the flight height, so that the aircraft 200 can have a larger field of view to attempt to observe whether there is a movable platform 100 in the larger field of view. Specifically, the increased flight height can be pre-set or determined by the aircraft 200 based on the actual scene, for example, based on the limitation of the flight height by the obstacles around the aircraft 200 and the actual flight position.

[0150] In some examples, increasing the flight height includes flying the aircraft 200 to a preset height; the method further includes: after the aircraft 200 flies to the preset height, continuing to identify the movable platform 100.

[0151] In this embodiment, the preset height can be pre-set, and the preset height can be absolute height information or relative height information. Taking the relative height information as an example, the aircraft 200 can increase h meters at the current height, and h is the preset height. Since the field of view of the aircraft 200 is expanded after the flight height of the aircraft 200 is increased, the aircraft 200 can continue to attempt to identify the movable platform 100 based on the aircraft observation data.

[0152] As an example, in addition to the lifting of the flight height, the aircraft 200 can also fly to a specific location. For example, the aircraft 200 can fly directly to a specific location after the communication is disconnected or the movable platform 100 is not identified from the aircraft observation data. It can also be that the flight height is lifted first, and the movable platform 100 is not identified after the aircraft observation data is collected, and then the aircraft 200 flies to a specific location.

[0153] Based on this, the safety measures of lifting the flight height or flying to a specific location described above can be used by the aircraft 200 in the above-mentioned sudden situation of disconnection of communication or failure to observe the movable platform 100, so as to attempt safe landing or fly to a specific location to ensure the safety of the aircraft 200.

[0154] The specific location can be flexibly set according to actual needs, and the determination strategy of the specific location can be pre-configured in the aircraft 200, so that the aircraft 200 can determine the specific location by itself. The specific location is not limited in the embodiment. As an example, the specific location can include any one of the following: a user-specified location, a preset return location, a location of another movable platform 100 that can communicate with the aircraft 200, and a location of another movable platform 100 identified by the aircraft 200 from the aircraft observation data.

[0155] As an example, the user-specified location can be provided by the user to the aircraft 200 in advance, for example, before the aircraft 200 takes off or before the aircraft 200 lands without performing the trajectory planning method of the embodiment. The specific location can be provided in advance through a remote control device of the aircraft 200 or through a client that can interact with the aircraft 200, and the like. Alternatively, the specific location can also be determined by the aircraft 200 by querying the user, for example, flying to a specific location can include: outputting an inquiry request to the user whether to fly to a specific location; and flying to a specific location in response to receiving an instruction from the user to fly to a specific location.

[0156] In the embodiment, the aircraft 200 can interact with the user to determine whether to fly to a specific location in the above-mentioned sudden situation. The inquiry request can be output to the user in various ways, for example, the inquiry request can be sent to a remote control device, or sent to a client that can interact with the aircraft 200, or sent to a wearable device (such as a smart helmet, smart glasses, or smart earphones, etc.) that can communicate with the aircraft 200. Alternatively, it can be directly sent or sent through a cloud server. The user of the embodiment can be a user located in the movable platform 100 to be landed, or can be another user, as long as the user has a use right for the aircraft 200. The embodiment is not limited in this regard.

[0157] Optionally, the remote control device can have a display screen, and display content can be generated in the display screen based on the inquiry request, so that the user can check the instruction; as an example, the display content can be "aircraft 200 loses communication with movable platform 100, whether to fly to a specific location" or "aircraft 200 fails to observe movable platform 100, whether to fly to a specific location". Similarly, the mobile device where the client is located also has a display screen, and the above display content can also be displayed in the user interface of the client; the smart helmet and the smart glasses are also the same.

[0158] Optionally, the output of the inquiry request can also be output in the form of voice, for example, the remote control device, the client and the above wearable device have language output capability, and voice can also be output in the form of language, so that the user knows the inquiry request.

[0159] The user can issue an instruction of whether the aircraft 200 flies to a specific location in various ways. For example, the above display content of the remote control device or the client can contain controls for the user to trigger "yes" and "no", and the remote control device or the client can obtain the instruction of whether the aircraft 200 flies to a specific location sent by the user in response to detecting that the above controls are triggered by the user. Alternatively, for wearable devices, the instruction of whether the aircraft 200 flies to a specific location issued by the user can be determined by detecting whether a set physical button is triggered or recognizing the user's action. Alternatively, the instruction of the user can also be recognized according to the voice signal collected by the voice collection component.

[0160] If the instruction received by the aircraft 200 indicates not to fly to a specific location, the aircraft 200 does not fly to a specific location. If the instruction indicates to fly to a specific location, the aircraft 200 can fly to a specific location. Based on this, the embodiment can determine whether to fly to a specific location under the instruction of the user through interaction with the user, so that it can fly under the instruction of the user, which not only meets the needs of the user, but also ensures the safety of the aircraft 200.

[0161] In some examples, based on the movement planning path, a landing planning trajectory for landing on the movable platform 100 is determined, including: based on the movement planning path, determining a preferred landing section; following the movable platform 100 in the preferred landing section; in the process of following, based on the aircraft observation data, adjusting the landing planning trajectory in real time until landing on the movable platform 100.

[0162] As an example, the aircraft 200 can analyze the movement planning path from the movable platform 100, and the aircraft 200 can determine one or more preferred landing sections based on one or more factors.

[0163] As an example, the mobile planning path can include various trajectory points at future time instances, and the aerial vehicle 200 can determine the positions of the various trajectory points to select the landing segment. Alternatively, the considerations can include current environmental information of the various trajectory points in the mobile planning path, future environmental information, speed of the mobile platform 100, etc.

[0164] As an example, determining the preferred landing segment based on the mobile planning path can include determining the preferred landing segment based on the mobile planning path and the aerial vehicle observation data. For example, the aerial vehicle observation data can be used to determine the environmental information of the various segments in the mobile planning path to determine the preferred landing segment to ensure that the aerial vehicle 200 can select a landing segment that can be safely landed.

[0165] In some examples, the preferred landing segment can include any one of a segment with a relatively small change in the moving direction of the mobile platform 100, a segment with a relatively small speed of the mobile platform 100, a segment with a relatively small number of low-altitude obstacles around the mobile platform 100, and a segment with a relatively small density of the mobile platform 100 around the mobile platform 100.

[0166] After the aerial vehicle 200 determines the preferred landing segment from the mobile planning path, the aerial vehicle 200 can start to follow the mobile platform 100 when entering the segment. The purpose of the following is to enable the aerial vehicle 200 to determine the best landing conditions during the following process. For example, during the following process of the aerial vehicle 200 following the mobile platform 100, the aerial vehicle 200 can continuously track the mobile platform 100 through the aerial vehicle observation data, can identify the obstacle information around the mobile platform 100 in real time, and can maintain the relative position with the mobile platform 100 to prepare for landing.

[0167] During the following process, the aerial vehicle 200 can continuously adjust the landing planning trajectory to adapt to the motion state and environmental changes of the mobile platform 100, and to determine the moving direction and speed of the mobile platform 100. For example, the aerial vehicle 200 can adjust the landing planning trajectory in real time to adjust the flight position, flight height, speed, or heading, etc. in real time to ensure safe and accurate landing on the mobile platform 100.

[0168] After ensuring that the landing planning trajectory is safe and adapted to the current conditions, the aerial vehicle 200 can perform the landing operation to safely land on the mobile platform 100. For example, the aerial vehicle 200 can maintain the relative consistency in position, moving speed, and moving direction with the mobile platform 100 in the last stage of landing, and then land on the mobile platform 100.

[0169] As can be seen from the above embodiments, the aircraft 200 can select the optimal landing section based on the movement planning path provided by the movable platform 100, continuously follow the movable platform 100 in the landing section, and adjust the landing planning trajectory in real time through the aircraft observation data to find the optimal landing opportunity and improve the success rate of the aircraft 200 landing.

[0170] In some examples, the landing planning trajectory can include a trailing trajectory for trailing the movable platform 100, a climbing trajectory for lifting the height of the aircraft 200, and a landing trajectory for descending to the movable platform 100; the aircraft 200 executes the trailing trajectory, the climbing trajectory, and the landing trajectory in sequence. This embodiment specifically designs the landing planning trajectory, which can include different types of trajectories in three stages:

[0171] ①First, the trailing trajectory for trailing the movable platform 100; as an example, the trailing trajectory can be updated in real time based on the aircraft observation data or / and the movable platform observation data.

[0172] For example, when the aircraft 200 executes the trailing trajectory, it can collect aircraft observation data through its own sensors such as cameras, radars, GPS, etc. These data can include the position, speed, surrounding environment information, etc. of the aircraft 200 itself, as well as the position and motion state of the movable platform 100, etc.

[0173] If the movable platform 100 can communicate with the aircraft 200, the aircraft 200 can also obtain the movable platform observation data to update the trailing trajectory in combination with the movable platform observation data.

[0174] The aircraft 200 can analyze these data in real time and adjust the trailing trajectory according to the analysis results. For example, if the movable platform 100 suddenly accelerates or changes direction, the aircraft 200 updates the trailing trajectory in real time according to these changes to ensure that the aircraft 200 can safely follow the platform.

[0175] As can be seen from the above embodiments, the aircraft 200 updates the trailing trajectory in real time, which can maintain a safe distance between the aircraft 200 and the movable platform 100, while ensuring that the aircraft 200 can adapt to the motion changes of the movable platform 100 and the changes of the surrounding environment, and prepare for the next landing.

[0176] ②The climbing trajectory for lifting the height of the aircraft 200. As an example, the highest height of the climbing trajectory can be determined based on the visual sensor on board the aircraft 200.

[0177] For example, the aerial vehicle 200 can be equipped with various sensors, such as vision sensors, which can collect image data of the surrounding environment. The vision sensors can help the aerial vehicle 200 determine its distance relative to the ground or other reference objects. By analyzing the images obtained from the vision sensors, the aerial vehicle 200 can estimate a safe height to avoid collision with obstacles. In addition, the vision sensors can also observe the state of the movable platform 100 to be landed to determine the highest height of the climb trajectory based on the observed movable platform 100, for example, to determine the highest height such that the landing position on the movable platform 100 can be observed.

[0178] ③Descending to the landing trajectory on the movable platform 100. As an example, the landing trajectory can be updated in real time based on aerial vehicle observation data or / and movable platform observation data.

[0179] For example, the aerial vehicle 200 can use its own sensors to collect aerial vehicle observation data to identify surrounding environment information during the landing process. The movable platform 100 and the aerial vehicle 200 can communicate, and the movable platform observation data can also be collected in real time and provided to the aerial vehicle 200 in real time. The aerial vehicle 200 can be based only on aerial vehicle observation data or movable platform observation data, or can be fused to obtain rich surrounding environment information.

[0180] Based on this, the aerial vehicle 200 can update the landing trajectory in real time according to the observation data to adapt to the position, speed or environmental changes of the landing position of the movable platform 100, to ensure the safety of landing and avoid collision with the movable platform 100 or other obstacles.

[0181] As an example, the aerial vehicle 200 can first perform the trailing trajectory, and in this stage, the aerial vehicle 200 can be kept behind or laterally behind the movable platform 100, etc. to maintain a certain relative position and distance. At this stage, the aerial vehicle 200 follows the movable platform 100 while maintaining a safe distance.

[0182] After trailing for a period of time, the aerial vehicle 200 will perform the climb trajectory. In this stage, the aerial vehicle 200 can gradually increase its flight height to prepare for landing. The purpose of the climb trajectory is to let the aerial vehicle 200 reach a safe height to find an appropriate landing position for landing while avoiding possible collision risks.

[0183] When the aerial vehicle 200 reaches an appropriate height, it can start to perform the landing trajectory. In this stage, the aerial vehicle 200 will gradually reduce its height and adjust its position to accurately land on the movable platform 100. The landing trajectory needs to be accurately controlled to ensure that the aerial vehicle 200 can land safely and smoothly.

[0184] Based on this, the aircraft 200 passes through different trajectories in the above three stages, and in the landing process, it first follows the movable platform 100, then raises the flight height, and then performs the landing action, which can ensure the success rate of landing.

[0185] Next, an embodiment will be described. In this embodiment, the aircraft 200 is a UAV, and the movable platform 100 is a vehicle. As shown in FIG. 3B, it is a schematic diagram of the trajectory planning method in this embodiment.

[0186] In this embodiment, the vehicle can be:

[0187] a. The vehicle interacting with the UAV, including the vehicle with automatic driving or assisted driving. Such a vehicle can perform autonomous positioning and path planning, can select the UAV observation data (such as images, point clouds, etc.) from the UAV perspective, can send the UAV observation data to the vehicle to assist the vehicle in positioning and road condition perception and behavior prediction of other vehicles around, thereby improving the path planning efficiency. The vehicle can also directly upload the vehicle's own trajectory to the UAV for generating the UAV's own trajectory.

[0188] b. For the vehicle without interaction with the UAV, vehicle detection can be performed from the UAV perspective, and the relative pose can be converted to the global coordinate system according to the UAV pose, and the vehicle trajectory can be predicted for generating the UAV's own trajectory.

[0189] In this embodiment, the parameterization method of the trajectory can be an N-order polynomial, or a B-Spline, etc., which is not limited in this embodiment.

[0190] As an example, the trajectory planning method embodiment can include the following steps:

[0191] (1) State estimation of the UAV itself

[0192] The state parameters estimated by the UAV include the position, attitude, and speed of the UAV itself, and can also include drift parameters selected according to the sensor model. The specific fusion method can use an extended Kalman filter based on a filter, an unscented Kalman filter, a particle filter, etc., which has fast speed and low operation resource requirement; or a nonlinear optimization based method can be selected, key frames are selected to form a sliding window, and the state quantity is optimized in the sliding window.

[0193] (2) Vehicle detection and state conversion

[0194] ① Vehicle detection

[0195] Vehicle detection refers to that a UAV identifies a vehicle of interest based on UAV observation data, which can include image data of a camera, point cloud data of a laser radar, etc. Three-dimensional position and attitude information (referred to as attitude information for short) in a sensor coordinate system of the UAV can be output.

[0196] In actual implementation, a traditional computer vision method can be used, such as a certain feature descriptor or image segmentation technology, plus a classifier, combined with depth information, to restore the three-dimensional position and attitude of the target vehicle. A deep learning method can also be used, a convolutional neural network is used for sample classification based on a candidate region, or a direct end-to-end structure is used to directly define the problem as a bounding box regression.

[0197] ②State transition

[0198] The result of vehicle detection is a relative position and attitude in a UAV coordinate system, which can be converted to a global coordinate system according to a real-time position and attitude of the UAV to obtain a position and attitude in the global coordinate system:

[0199] G R C = G R U U R C

[0200] Among them, G R C refers to a vehicle attitude in a global coordinate system; G R U refers to a UAV attitude in the global coordinate system; U R C refers to a vehicle attitude relative to the UAV in a UAV coordinate system.

[0201] Among them, G P C = G R U U R C U P C + G P U ;

[0202] Among them, G represents a global coordinate system, U represents a UAV coordinate system, and C represents a vehicle coordinate system.

[0203] G P C refers to a vehicle position in a global coordinate system; G R U refers to a UAV attitude in the global coordinate system; U R Cis the attitude of the vehicle relative to the UAV in the UAV coordinate system; U P C is the position of the vehicle relative to the UAV; G P U is the position of the UAV in the global coordinate system.

[0204] ③ Vehicle trajectory prediction

[0205] The vehicle trajectory prediction can be obtained according to the historical vehicle state X t-N:t-1 and the system input U t-N:t-1 , based on the maximum likelihood estimation method, to obtain the trajectory of the vehicle prediction, to obtain P(X t |X t-N:t- 1,U t-N:t-1 ,O t ). Wherein, the formula represents the probability principle of maximum likelihood estimation. Where X is the state quantity of the system, P represents the probability, U represents the control information, and O represents the observation. t is the time, and t-N represents the time of N time units before t.

[0206] The traditional EKF can be used in combination with the constant speed model to predict the vehicle state at t time in combination with the estimated prior speed at the current time.

[0207] Or use the deep learning method to define the problem as a 3D multi-object tracking (MOT) problem, for example, first obtain 3D detection from image and radar point cloud, direct combination of Kalman filter and Hungarian algorithm for state estimation and data association, to obtain the prediction of the vehicle state.

[0208] (3) Planning trajectory

[0209] In the field of trajectory planning, most of the controllers used are linearized from the model under hovering conditions, and are stable only at reasonable small roll and pitch angles. Some work in this regard has addressed acrobatic maneuvers, however without stability and convergence guarantees, the attitude of the UAV deviates greatly from the horizontal hovering condition. Although some progress has been made in the field of machine learning techniques using reinforcement to improve performance, these methods do not seem suitable for motion planning or trajectory generation in an environment with obstacles. Model predictive control (MPC) methods can only guarantee convergence when the linearized model is completely controllable, or a control Lyapunov function can be obtained, so it is difficult to directly apply to UAV trajectory generation technology.

[0210] Trajectory planning problem consists in finding a relationship between two elements belonging to different domains: time and space. Therefore, a trajectory is usually expressed as a parametric function of time, which provides the corresponding desired position at each instant. The desired motion can be defined by assuming only the initial and final points (point-to-point trajectory), or considering a set of intermediate points that must be correctly interpolated / approximated (multipoint trajectory).

[0211] In the case of point-to-point trajectories, a complex motion is obtained by connecting several point-to-point trajectories, which are individually optimized by considering the initial and final boundary conditions (velocity, acceleration, etc.) of each trajectory and constraints on their maximum values. In the case of multipoint trajectories, an arbitrary complex motion can be defined by specifying intermediate points, and finding the trajectory as a solution of a global optimization problem, which depends on the profile of conditions imposed at each intermediate point and globally. Moreover, different criteria can be adopted to define the motion curve according to the given intermediate points, which are not necessarily crossed by the trajectory. In particular, two types of fitting can be distinguished: the interpolating way, in which the curve must pass through the intermediate points at certain time values; and the approximating way, in which the curve does not exactly pass through the intermediate points, which can be particularly useful in multidimensional trajectories, where it is necessary to reduce the velocity / acceleration values along the curve, but errors can be generated by specifying a given tolerance.

[0212] The present embodiment takes the current position of the UAV as the starting position, and the future position of the vehicle as the reference position of the end point, so as to generate the landing planning trajectory of the UAV. Every time interval, the UAV can refresh the target position according to the future position of the vehicle, and obtain a new landing trajectory planning. The trajectory generation problem can be defined as solving the polynomial parameter pj describing the trajectory, while a series of conditions that the desired trajectory equation satisfies are turned into constraint conditions to constrain the solving process.

[0213] The planning task from the current position to the current predicted vehicle trajectory can be divided into M segments, and B-Spline or polynomial can be selected for parameterization fitting of each segment of the trajectory. Here, N-order polynomial is selected for illustration, and B-Spline is similar.

[0214] Here, the road marker points are divided at equal intervals, and the time allocation method is uniform allocation or trapezoidal allocation, assuming that the velocity in each polynomial segment satisfies uniform or trapezoidal velocity change, and the total time T is allocated to each segment according to the distance of each segment. The trajectory segmentation and time allocation here are sample allocations. In an iterative algorithm, if various conditions such as obstacle avoidance or vehicle speed demand can be met, the time of a certain segment can be reduced or increased.

[0215] For example, setting the position, velocity, acceleration or higher order derivative of a certain point to a specific value can constitute an equality constraint; or the continuity of position, velocity and acceleration between adjacent segments can constitute an equality constraint.

[0216] The problem of trajectory generation is defined as a traditional mathematical quadratic programming (QP) solving problem.

[0217] The UAV without interaction with the vehicle can only determine the target trajectory to be tracked through detection and trajectory prediction of the vehicle. However, the UAV with interaction with the vehicle can obtain a more reliable target trajectory through the positioning and trajectory planning information uploaded by the vehicle. The difference in reliability will be reflected in:

[0218] 1. When the target vehicle has a sharp state change, such as sharp turning or drifting, the UAV prediction of the target vehicle will have a certain lag, and the tracking trajectory generated only according to the UAV observation of the target vehicle will have a certain overshoot.

[0219] 2. In the scene where the surrounding environment has a large amount of occlusion, the UAV has intermittent detection of the vehicle, and at this time there is a possibility of losing the tracking target.

[0220] 3. The scene where the same vehicle model is stacked will also increase the difficulty of detection, which may cause the false detection rate to increase.

[0221] If the UAV can establish interaction with the target vehicle to obtain the vehicle's own planning, it can greatly reduce the lag caused by the sudden change of the target state, or the tracking failure caused by detection failure, and can greatly improve the efficiency of the following trajectory planning and the success rate of landing.

[0222] If no interaction is established with the target vehicle, after losing the tracking of the target vehicle, the UAV can:

[0223] 1. Request to increase the flight height to try to obtain a better field of view to improve the detection success rate.

[0224] 2. Send a request to the user to return, allow the UAV to fly back to the return point to wait for the next instruction; send a request to the user to the target location, fly to the target location to wait for the next instruction after the response.

[0225] 3. When it is unable to establish contact with the user, the UAV directly flies to the destination set by the user in advance, or returns to the return point to wait for the user to connect.

[0226] In actual application, the UAV with interaction with the vehicle:

[0227] 1. When the signal is blocked, the UAV that can interact with the target vehicle can switch to the non-interaction state. Wait for the signal to recover and switch back to the interaction state, and integrate the trajectory obtained from the vehicle.

[0228] 2. After establishing interaction, the UAV can provide closer following, such as in environments with a large number of obstructions and complex obstacles, or in situations where the target vehicle will undergo a sharp state change. The UAV can adjust its state in time to closely follow the vehicle and land at any time. A non-interaction UAV that relies solely on its own observations for tracking may not be able to keep up with a sudden state change in time, resulting in blocked following and task failure.

[0229] 3. It can be accurately landed on the designated position on the target vehicle, such as the landing gear, charging pile, or side-hanging structure device due to appearance or other design requirements, etc. Such applications must require the UAV to establish interaction with the vehicle to ensure the success rate of safe landing.

[0230] 4. When the UAV establishes interaction with the target vehicle, it can also provide the vehicle with sensor data such as image point cloud of bird of view (BEV) of the UAV, which can improve the vehicle's perception range and reduce blind areas, and is beneficial to the behavior prediction of surrounding vehicles, pedestrians, etc. Plan a safer and more efficient driving scheme.

[0231] The various technical features in the above embodiments can be combined in any way, as long as there is no conflict or contradiction between the features. However, due to space limitations, they are not described one by one, so any combination of the various technical features in the above embodiments also belongs to the scope disclosed in this specification.

[0232] As shown in FIG. 4, it is a flow chart of another trajectory planning method provided by the embodiment, which can be applied to a flying vehicle. The method can include:

[0233] In step 402, based on the collected observation data of the flying vehicle, a recommended path of the movable platform is determined;

[0234] In step 404, the recommended path is sent to the movable platform, and a movement planning path planned based on the recommended path is obtained from the movable platform; and

[0235] In step 406, based on the movement planning path, a landing planning trajectory for landing on the movable platform is determined.

[0236] The specific implementation process of the method of the embodiment can refer to the description of the foregoing embodiments, which will not be described here.

[0237] As shown in FIG. 5, it is a flow chart of another trajectory planning method provided by the embodiment, which can be applied to a flying vehicle. The method can include:

[0238] In step 502, the path planning related information is sent to the movable platform;

[0239] In step 505, the movement planning path planned by the movable platform based on the path planning related information is acquired; and

[0240] In step 506, the landing planning trajectory for landing on the movable platform is determined based on the movement planning path.

[0241] In this embodiment, the aircraft can send the path planning related information to the movable platform to provide the movable platform with relevant environmental information or to provide the movable platform with information suitable for the landing of the aircraft. Thus, the movable platform can plan a movement planning path suitable for the safe and efficient movement of the movable platform, and can also plan a movement planning path more suitable for the landing of the aircraft. The movement planning path of the movable platform 100 can cooperate with the landing process of the aircraft, and the movable platform subsequently provides the movement planning path to the aircraft, thereby helping the aircraft to determine a landing planning trajectory for landing on the movable platform more safely or more accurately, and improving the success rate of the landing of the aircraft. After the movable platform plans a movement planning path more suitable for the safe and efficient movement of the movable platform, the subsequent landing of the aircraft on the movable platform can also be more safe and efficient.

[0242] The aircraft needs a certain time length to land on the movable platform in motion, and the movable platform is in motion. In this embodiment, the aircraft can obtain the movement planning path provided by the movable platform in real time and with high reliability, thereby improving the efficiency of planning the landing planning trajectory and the success rate of landing, and improving the accuracy and safety of dynamic landing of the aircraft.

[0243] In some examples, the path planning related information includes at least one of the following: environmental information sensed by the aircraft and related to the path planning of the movable platform; movement information of the movable platform sensed by the aircraft; and a recommended path of the movable platform determined by the aircraft.

[0244] In some examples, the path planning related information includes at least one of the following: the path planning related information is determined based on the observation data collected by the aircraft.

[0245] In some examples, the landing planning trajectory for landing on the movable platform is determined based on the movement planning path, including: the landing planning trajectory for landing on the movable platform is determined based on the recommended path and the movement planning path.

[0246] In some examples, the landing planning trajectory to the movable platform is determined based on the recommended path and the movement planning path by fusing the recommended path and the movement planning path, and determining the landing planning trajectory to the movable platform according to a fusion result.

[0247] In some examples, the landing planning trajectory to the movable platform is determined based on the recommended path and the movement planning path by fusing the recommended path and the movement planning path, and determining the landing planning trajectory to the movable platform according to a fusion result.

[0248] In some examples, the landing planning trajectory to the movable platform is determined based on the movement planning path by obtaining movable platform observation data sent by the movable platform, and determining the landing planning trajectory to the movable platform based on the movement planning path and the movable platform observation data.

[0249] In some examples, the movable platform observation data includes at least one of information of low-altitude obstacles in a direction of the movement planning path, and information of surrounding movable platforms.

[0250] In some examples, the information of low-altitude obstacles includes at least one of movement information of moving obstacles, and position information of stationary obstacles.

[0251] In some examples, the information of surrounding movable platforms includes at least one of movement information of front movable platforms, and density information of surrounding movable platforms.

[0252] In some examples, the landing planning trajectory to the movable platform is determined based on the movement planning path by determining a preliminary landing planning trajectory based on the movement planning path, and optimizing the preliminary landing planning trajectory based on the movable platform observation data to obtain the landing planning trajectory.

[0253] In some examples, the landing planning trajectory to the movable platform is determined based on the movement planning path by obtaining aircraft observation data, and determining the landing planning trajectory to the movable platform based on the aircraft observation data and the movement planning path.

[0254] In some examples, the planning of the landing planning trajectory includes position planning of a landing trajectory point, and velocity planning corresponding to the landing trajectory point.

[0255] In some examples, the aircraft observation data includes at least one of obstacle information near the movable platform, obstacle information in a landing direction of the aircraft, movement speed information of the movable platform, and density information of surrounding movable platforms of the movable platform.

[0256] In some examples, determining, based on the aircraft observation data and the movement planning path, a landing planning trajectory for landing on the movable platform comprises: determining, based on the movement planning path, a preliminary landing planning trajectory; and optimizing, based on the aircraft observation data, the preliminary landing planning trajectory to obtain the landing planning trajectory.

[0257] In some examples, optimizing, based on the aircraft observation data, the preliminary landing planning trajectory comprises: optimizing, based on the aircraft observation data, position information of a trajectory point of the preliminary landing planning trajectory.

[0258] In some examples, optimizing, based on the aircraft observation data, the preliminary landing planning trajectory comprises: optimizing, based on the aircraft observation data, velocity information of a trajectory point of the preliminary landing planning trajectory.

[0259] In some examples, determining, based on the movement planning path, a landing planning trajectory for landing on the movable platform comprises: determining, based on the movement planning path, a future pose of the movable platform; obtaining a current pose of the aircraft; and determining, based on the current pose of the aircraft and the future pose of the movable platform, the landing planning trajectory for landing on the movable platform.

[0260] In some examples, determining, based on the current pose of the aircraft and the future pose of the movable platform, the landing planning trajectory for landing on the movable platform comprises: determining a starting point of the landing planning trajectory to be determined based on the current pose of the aircraft; determining an ending point of the landing planning trajectory to be determined based on the future pose of the movable platform; and determining the landing planning trajectory from the starting point to the ending point.

[0261] In some examples, determining the landing planning trajectory from the starting point to the ending point comprises: determining at least one intermediate point between the starting point and the ending point, and determining the landing planning trajectory from the starting point, through the intermediate point, to the ending point.

[0262] In some examples, determining the at least one intermediate point between the starting point and the ending point comprises: obtaining a combination of one or more of the following trajectory points: a trajectory point for aircraft obstacle avoidance, a trajectory point set by a user, or a trajectory point of a flight task performed by the aircraft during landing; and determining the at least one intermediate point between the starting point and the ending point based on the obtained trajectory points.

[0263] In some examples, determining the landing planning trajectory from the starting point, through the intermediate point, to the ending point comprises: dividing the landing planning trajectory to be determined into a plurality of sub-landing planning trajectories to be determined based on the starting point, the intermediate point, and the ending point; and planning each of the sub-landing planning trajectories to obtain the landing planning trajectory from the starting point, through the intermediate point, to the ending point.

[0264] In some examples, the planning the sub-descent planning trajectory for each segment comprises: determining a total descent time length from the start point to the end point, assigning a sub-descent time length to each segment to be determined according to the total descent time length; and planning the sub-descent planning trajectory for each segment according to the sub-descent time length of each segment to be determined.

[0265] In some examples, the sub-descent time length of each segment to be determined is determined according to a preset condition.

[0266] In some examples, the preset condition comprises: a condition determined based on obstacle avoidance information, and / or a condition determined based on motion speed information of the movable platform.

[0267] In some examples, the method further comprises: sending the aerial vehicle observation data to the movable platform; and planning the movement planning path based on the recommended path comprises: planning the movement planning path based on the aerial vehicle observation data and the recommended path.

[0268] In some examples, the aerial vehicle observation data comprises at least one of: obstacle information near the movable platform, obstacle information in the motion direction of the movable platform, motion information of the movable platform, and motion information of the movable platform around the movable platform.

[0269] In some examples, the aerial vehicle observation data comprises sensing data from an aerial vehicle bird's eye view; and the sensing data comprises at least one of: laser radar data, visual image data, microwave radar data, and ultrasonic data.

[0270] In some examples, the movement planning path planned based on the aerial vehicle observation data and the recommended path comprises: a movement planning path obtained by optimizing a preliminary movement planning path of the movable platform based on the aerial vehicle observation data and the recommended path.

[0271] In some examples, the preliminary movement planning path is determined based on input information of a user or / and sensing information of the movable platform.

[0272] In some examples, the aerial vehicle is capable of communicating with a plurality of movable platforms, and the method further comprises: determining the movable platform to be landed from the plurality of movable platforms.

[0273] In some examples, the determining the recommended path of the movable platform based on the collected aerial vehicle observation data comprises: determining the movable platform to be landed based on the aerial vehicle observation data, and determining the recommended path of the movable platform to be landed.

[0274] In some examples, determining the movable platform to be landed based on the aerial vehicle observation data comprises: identifying a plurality of movable platforms based on the aerial vehicle observation data; and determining the movable platform to be landed from the identified plurality of movable platforms.

[0275] In some examples, the method further comprises: switching to a preset non-interaction mode in response to disconnection of the communication with the movable platform; and re-determining the landing planning trajectory for landing on the movable platform based on the aerial vehicle observation data in the non-interaction mode.

[0276] In some examples, the aerial vehicle observation data comprises at least one of: current position information of the movable platform; current motion direction information of the movable platform; motion trend information of the movable platform.

[0277] In some examples, the method further comprises: exiting the preset non-interaction mode in response to connection of the communication with the movable platform; and determining the landing planning trajectory for landing on the movable platform based on the movement planning path.

[0278] In some examples, the method further comprises: increasing the flight height or flying to a specific location in response to disconnection of the communication with the movable platform or failure to identify the movable platform from the aerial vehicle observation data.

[0279] In some examples, the specific location comprises any one of: a user-specified location; a preset return location; a location of another movable platform capable of communicating with the aerial vehicle; and a location of another movable platform identified by the aerial vehicle from the aerial vehicle observation data.

[0280] In some examples, flying to the specific location comprises: outputting a query request to the user as to whether to fly to the specific location; and flying to the specific location in response to receiving an instruction from the user to fly to the specific location.

[0281] In some examples, increasing the flight height comprises flying the aerial vehicle to a preset height; and the method further comprises: continuing to identify the movable platform after the aerial vehicle flies to the preset height.

[0282] In some examples, determining the landing planning trajectory for landing on the movable platform based on the movement planning path comprises: determining a preferred landing section based on the movement planning path; following the movable platform in the preferred landing section; and adjusting the landing planning trajectory in real time based on the aerial vehicle observation data during the following until landing on the movable platform.

[0283] In some examples, the preferred landing section includes any one of the following: a section with a relatively small change in the moving direction of the movable platform, a section with a relatively small moving speed of the movable platform, a section with a relatively small low-altitude obstacle around the movable platform, and a section with a relatively small density of movable platforms around the movable platform.

[0284] In some examples, the preferred landing section is determined based on the movement planning path, including: determining the preferred landing section based on the movement planning path and the aircraft observation data.

[0285] In some examples, the landing planning trajectory includes: a following trajectory for following the movable platform, a climbing trajectory for lifting the aircraft to a height, and a landing trajectory for descending to the movable platform; and the aircraft sequentially performs the following trajectory, the climbing trajectory, and the landing trajectory.

[0286] In some examples, the following trajectory is updated in real time based on the aircraft observation data or / and the movable platform observation data.

[0287] In some examples, the highest height of the climbing trajectory is determined based on a visual sensor on board the aircraft.

[0288] In some examples, the landing trajectory is updated in real time based on the aircraft observation data or / and the movable platform observation data.

[0289] As shown in FIG. 6, it is a flow chart of another trajectory planning method provided by the embodiment, which can be applied to the movable platform. The method includes:

[0290] In step 602, a recommended path sent by an aircraft is received, the recommended path being determined by the aircraft based on collected aircraft observation data;

[0291] In step 604, a movement planning path is planned based on the recommended path;

[0292] In step 606, the movement planning path is sent to the aircraft, so that the aircraft determines a landing planning trajectory for landing on the movable platform based on the movement planning path.

[0293] In some examples, the method further includes: obtaining movable platform observation data and sending it to the aircraft; and the landing planning trajectory includes a landing planning trajectory of the movable platform determined based on the movement planning path and the movable platform observation data.

[0294] In some examples, the movable platform observation data includes at least one of the following: information of low-altitude obstacles in the direction of the movement planning path, and information of movable platforms around the movable platform.

[0295] In some examples, the information of the low-altitude obstacle includes at least one of: movement information of a moving obstacle, position information of a static obstacle.

[0296] In some examples, the information of the surrounding movable platform includes at least one of: movement information of a front movable platform, density information of the surrounding movable platform.

[0297] In some examples, the planning of the movement planning path based on the recommended path includes: receiving the aerial vehicle observation data sent by the aerial vehicle; and planning the movement planning path based on the aerial vehicle observation data and the recommended path.

[0298] In some examples, the aerial vehicle observation data includes at least one of: obstacle information near the movable platform, obstacle information in the movement direction of the movable platform, movement information of the movable platform, movement information of the surrounding movable platform.

[0299] In some examples, the aerial vehicle observation data includes sensing data of the aerial vehicle bird's-eye view; and the sensing data includes at least one of: laser radar data, visual image data, microwave radar data, ultrasonic data.

[0300] In some examples, the planning of the movement planning path based on the aerial vehicle observation data and the recommended path includes: optimizing the preliminary movement planning path based on the aerial vehicle observation data and the recommended path to obtain the movement planning path.

[0301] In some examples, the preliminary movement planning path is determined based on input information of a user or / and sensing information of the movable platform.

[0302] The above-mentioned trajectory planning method embodiments can be implemented by software, or by hardware or a combination of software and hardware. For example, as a logical device, it is formed by a processor reading corresponding computer program instructions in a non-volatile memory to the memory for running.

[0303] In some embodiments, a trajectory planning device is also provided, as shown in FIG. 7, which is a structure diagram of the trajectory planning device. The trajectory planning device of the present embodiment can be applied to an aerial vehicle. The device can include: a processor 71; and a memory 72 storing computer program code executable on the processor 71; wherein the processor 71 implements the steps of the above-mentioned trajectory planning method embodiments when executing the computer program.

[0304] In some embodiments, a trajectory planning device is also provided, as shown in FIG. 8, which is a structural diagram of the trajectory planning device, the trajectory planning device of the embodiment can be applied to a movable platform, and the device can include: a processor 81; and a memory 82 storing computer program code executable on the processor 81; wherein the processor 81 implements the steps of the foregoing trajectory planning method embodiments when executing the computer program.

[0305] In some embodiments, a flying vehicle is also provided, as shown in FIG. 9, which is a structural diagram of the flying vehicle 90, the flying vehicle 90 can include: a body 91; a power device 92 provided on the body, for providing power for the flying vehicle; and the trajectory planning device embodiment shown in the foregoing FIG. 7.

[0306] In some embodiments, a movable platform is also provided, as shown in FIG. 10, which is a structural diagram of the movable platform 100, and can include: a body 101; a power device 102 provided on the body, for providing power for the movable platform; and the trajectory planning device embodiment shown in the foregoing FIG. 8.

[0307] In some embodiments, a trajectory planning system is also provided, as shown in FIG. 1, which can include the foregoing flying vehicle embodiment and the foregoing movable platform embodiment.

[0308] In some embodiments, a computer readable storage medium having computer instructions stored thereon is also provided, the instructions being executed by a processor to implement the steps of the foregoing trajectory planning method embodiments.

[0309] The embodiments of the present specification can adopt the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. The computer usable storage media includes permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. The information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0310] The step division of the above various methods is only for the purpose of clear description, and can be combined into one step or split into multiple steps in implementation, as long as the same logical relationship is included, which is within the protection scope of the present patent; adding insignificant modifications or introducing insignificant designs in the algorithm or flow, but not changing the core design of the algorithm and flow, are within the protection scope of the present application.

[0311] For the device embodiment, since it basically corresponds to the method embodiment, the relevant part can be referred to the part of the method embodiment. The device embodiment described above is only illustrative, wherein the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present embodiment according to actual needs. The description of "specific examples" or "some examples" means that the specific features, structures, materials or characteristics described in combination with the embodiments or examples are contained in at least one embodiment or example of the present specification. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0312] It should be noted that, in this document, the relationship terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. The terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or device including the element.

[0313] The above describes the method and device provided by the embodiments of the present application in detail, and the principle and implementation manner of the present application are described by applying specific examples; the above embodiment description is only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manner and application range can be changed, and the above description should not be understood as limiting the present application.

Claims

1. A trajectory planning method characterized by, The method is applied to an aircraft, and comprises: determining a recommended path of a movable platform based on collected aircraft observation data; sending the recommended path to the movable platform, and obtaining a movement planning path of the movable platform based on the recommended path; determining a landing planning trajectory of landing on the movable platform based on the movement planning path.

2. The method of claim 1, wherein, The determining of the landing planning trajectory of landing on the movable platform based on the movement planning path comprises: determining the landing planning trajectory of landing on the movable platform based on the recommended path and the movement planning path.

3. The method of claim 1, wherein, The determining of the landing planning trajectory of landing on the movable platform based on the recommended path and the movement planning path comprises: fusing the recommended path and the movement planning path, and determining the landing planning trajectory of landing on the movable platform according to a fusion result.

4. The method of claim 3, wherein, The fusing of the recommended path and the movement planning path, and the determining of the landing planning trajectory of landing on the movable platform according to a fusion result comprises: performing weighted calculation on trajectory points in the recommended path and trajectory points in the movement planning path, and determining the landing planning trajectory of landing on the movable platform according to a calculation result.

5. The method of claim 1, wherein, The determining of the landing planning trajectory of landing on the movable platform based on the movement planning path comprises: obtaining movable platform observation data sent by the movable platform; determining the landing planning trajectory of landing on the movable platform based on the movement planning path and the movable platform observation data.

6. The method of claim 5, wherein, The movable platform observation data comprises at least one of the following: information of low-altitude obstacles in a direction of the movement planning path, information of surrounding movable platforms.

7. The method of claim 6, wherein, The information of low-altitude obstacles comprises at least one of the following: movement information of moving obstacles, position information of stationary obstacles.

8. The method of claim 6, wherein, The information of surrounding movable platforms comprises at least one of the following: movement information of front movable platforms, density information of surrounding movable platforms.

9. The method of claim 5, wherein, The determining of the landing planning trajectory of landing on the movable platform based on the movement planning path and the movable platform observation data comprises: determining a preliminary landing planning trajectory based on the movement planning path; optimizing the preliminary landing planning trajectory based on the movable platform observation data to obtain the landing planning trajectory.

10. The method of claim 1, wherein, The determining of the landing planning trajectory of landing on the movable platform based on the movement planning path comprises: determining the landing planning trajectory of landing on the movable platform based on the aircraft observation data and the movement planning path.

11. The method of claim 1, wherein, The planning of the landing planning trajectory comprises: position planning of a landing trajectory point, and velocity planning corresponding to the landing trajectory point.

12. The method of claim 10, wherein, The aircraft observation data comprises at least one of the following: obstacle information near the movable platform, obstacle information in a landing direction of the aircraft, movement speed information of the movable platform, and density information of movable platforms surrounding the movable platform. ​ 13. The method of claim 10, wherein, The determining, based on the aircraft observation data and the movement planning path, of a landing planning trajectory for landing on the movable platform comprises: determining a preliminary landing planning trajectory based on the movement planning path; optimizing the preliminary landing planning trajectory based on the aircraft observation data to obtain the landing planning trajectory.

14. The method of claim 13, wherein, The optimizing the preliminary landing planning trajectory based on the aircraft observation data comprises: optimizing position information of a trajectory point of the preliminary landing planning trajectory based on the aircraft observation data.

15. The method of claim 13, wherein, The optimizing the preliminary landing planning trajectory based on the aircraft observation data comprises: optimizing velocity information of a trajectory point of the preliminary landing planning trajectory based on the aircraft observation data.

16. The method of claim 1, wherein, The determining, based on the movement planning path, of a landing planning trajectory for landing on the movable platform comprises: determining a future pose of the movable platform based on the movement planning path; obtaining a current pose of the aircraft; determining a landing planning trajectory for landing on the movable platform based on the current pose of the aircraft and the future pose of the movable platform.

17. The method of claim 16, wherein, The determining, based on the current pose of the aircraft and the future pose of the movable platform, of a landing planning trajectory for landing on the movable platform comprises: determining a starting point of the landing planning trajectory to be determined based on the current pose of the aircraft; determining an ending point of the landing planning trajectory to be determined based on the future pose of the movable platform; determining the landing planning trajectory from the starting point to the ending point.

18. The method of claim 17, wherein, The determining the landing planning trajectory from the starting point to the ending point comprises: determining at least one intermediate point between the starting point and the ending point, and determining the landing planning trajectory from the starting point, through the intermediate point, to the ending point.

19. The method of claim 18, wherein, The determining the at least one intermediate point between the starting point and the ending point comprises: obtaining a combination of one or more trajectory points, including a trajectory point for aircraft obstacle avoidance, a trajectory point set by a user, or a trajectory point of a flight task performed by the aircraft during landing; determining the at least one intermediate point between the starting point and the ending point based on the obtained trajectory points.

20. The method of claim 18, wherein, The determining the landing planning trajectory from the starting point, through the intermediate point, to the ending point comprises: dividing the landing planning trajectory to be determined into a plurality of sub landing planning trajectories to be determined based on the starting point, the intermediate point, and the ending point; planning each of the sub landing planning trajectories, thereby obtaining the landing planning trajectory from the starting point, through the intermediate point, to the ending point.

21. The method of claim 20, wherein, The planning each of the sub landing planning trajectories comprises: determining a total landing duration from the starting point to the ending point, and assigning a landing sub duration to each of the sub landing planning trajectories to be determined according to the total landing duration; planning each of the sub landing planning trajectories according to the landing sub duration of each of the sub landing planning trajectories to be determined. The landing sub duration of each of the sub landing planning trajectories to be determined is determined according to a preset condition.

22. The method of claim 21, wherein, ​ 23. The method of claim 22, wherein, The preset condition comprises a condition determined based on obstacle avoidance information and / or a condition determined based on motion speed information of the movable platform.

24. The method of claim 1, wherein, The method further comprises: sending the aerial vehicle observation data to the movable platform; The movement planning path planned based on the recommended path comprises the movement planning path planned based on the aerial vehicle observation data and the recommended path.

25. The method of claim 1, wherein, The aerial vehicle observation data comprises at least one of the following: obstacle information near the movable platform, obstacle information in the motion direction of the movable platform, motion information of the movable platform, motion information of the movable platform around the movable platform.

26. The method of claim 1, wherein, The aerial vehicle observation data comprises sensing data in the bird's-eye view of the aerial vehicle. The sensing data comprises at least one of the following: laser radar data, visual image data, microwave radar data, ultrasonic data.

27. The method of claim 24, wherein, The movement planning path planned based on the aerial vehicle observation data and the recommended path comprises the movement planning path obtained by optimizing a preliminary movement planning path of the movable platform based on the aerial vehicle observation data and the recommended path.

28. The method of claim 27, wherein, The preliminary movement planning path is determined based on input information of a user and / or sensing information of the movable platform.

29. The method of claim 1, wherein, The aerial vehicle can communicate with multiple movable platforms, and the method further comprises: determining the movable platform to be landed from the multiple movable platforms.

30. The method of claim 1 or 29, wherein, The method further comprises: determining the recommended path of the movable platform based on the aerial vehicle observation data, comprising:

31. The method of claim 30, wherein, determining the movable platform to be landed based on the aerial vehicle observation data, and determining the recommended path of the movable platform to be landed. The method further comprises:

32. The method of claim 1, wherein, identifying multiple movable platforms based on the aerial vehicle observation data, and determining the movable platform to be landed from the identified multiple movable platforms. The method further comprises: switching to a preset non-interaction mode in response to disconnection with the movable platform; 33. The method of claim 32, wherein, In the non-interaction mode, re-determining the landing planning trajectory for landing on the movable platform based on the aerial vehicle observation data.

34. The method of claim 32, wherein, The aerial vehicle observation data comprises at least one of the following: current position information of the movable platform, current motion direction information of the movable platform, motion trend information of the movable platform. The method further comprises:

35. The method of claim 1, wherein, in response to the communication connection with the movable platform, exiting the preset non-interaction mode, and determining a landing planning trajectory for landing on the movable platform based on the movement planning path. The method further comprises:

36. The method of claim 35, wherein, in response to disconnection with the movable platform or failure to identify the movable platform from the aerial vehicle observation data, increasing the flight height or flying to a specific position. The specific position comprises any one of the following: a position specified by a user, a preset return position, a position of another movable platform capable of communicating with the aerial vehicle, and a position of another movable platform identified by the aerial vehicle from the aerial vehicle observation data.

37. The method of claim 35, wherein, The flying to the specific location comprises: outputting an inquiry request to the user whether to fly to the specific location; in response to receiving an instruction sent by the user to fly to the specific location, flying to the specific location.

38. The method of claim 35, wherein, The lifting of the flight height comprises the aircraft flying to a preset height; the method further comprises: after the aircraft flies to the preset height, continuing to identify the movable platform.

39. The method of claim 1, wherein, The determining of the landing planning trajectory based on the movement planning path comprises: determining a preferred landing section based on the movement planning path; following the movable platform in the preferred landing section; in the process of following, adjusting the landing planning trajectory in real time based on the aircraft observation data until landing on the movable platform.

40. The method of claim 39, wherein, The preferred landing section comprises any one of the following: a section in which the change in the movement direction of the movable platform is relatively small, a section in which the movement speed of the movable platform is relatively small, a section in which the low-altitude obstacles around the movable platform are relatively few, and a section in which the density of the movable platforms around the movable platform is relatively small.

41. The method of claim 39, wherein, The determining of the preferred landing section based on the movement planning path comprises: determining the preferred landing section based on the movement planning path and the aircraft observation data.

42. The method of claim 1, wherein, The landing planning trajectory comprises a following trajectory for following the movable platform, a climbing trajectory for lifting the aircraft height, and a landing trajectory for landing on the movable platform; The aircraft sequentially executes the following trajectory, the climbing trajectory, and the landing trajectory.

43. The method of claim 42, wherein, The following trajectory is updated in real time based on the aircraft observation data or / and movable platform observation data.

44. The method of claim 42, wherein, The highest height of the climbing trajectory is determined based on the visual sensor on the aircraft.

45. The method of claim 42, wherein, The landing trajectory is updated in real time based on the aircraft observation data or / and movable platform observation data.

46. A trajectory planning method, characterized by, The method applied to the aircraft comprises: sending path planning related information to the movable platform; obtaining a movement planning path planned by the movable platform based on the path planning related information; and determining a landing planning trajectory for landing on the movable platform based on the movement planning path.

47. The method of claim 46, wherein, The path planning related information comprises at least one of the following: environmental information sensed by the aircraft and related to the path planning of the movable platform; movement information of the movable platform sensed by the aircraft; a recommended path of the movable platform determined by the aircraft.

48. The method of claim 46, wherein, The path planning related information comprises at least one of the following: the path planning related information is determined based on collected aircraft observation data.

49. A trajectory planning method, characterized by, The method applied to the movable platform comprises: receiving a recommended path sent by the aircraft, the recommended path being determined by the aircraft based on collected aircraft observation data; planning a movement planning path based on the recommended path; sending the movement planning path to the aircraft, so that the aircraft determines a landing planning trajectory for landing on the movable platform based on the movement planning path.

50. The method of claim 49, wherein, The method further comprises: acquiring movable platform observation data and sending the movable platform observation data to the aerial vehicle; the landing planning trajectory is determined based on the moving planning path and the movable platform observation data.

51. The method of claim 50, wherein, The movable platform observation data includes at least one of the following: information of low-altitude obstacles in the direction of the moving planning path, information of surrounding movable platforms.

52. The method of claim 51, wherein, The information of low-altitude obstacles includes at least one of the following: movement information of moving obstacles, position information of stationary obstacles.

53. The method of claim 51, wherein, The information of surrounding movable platforms includes at least one of the following: movement information of front movable platforms, density information of surrounding movable platforms.

54. The method of claim 49, wherein, The moving planning path is planned based on the recommended path, including: receiving aerial vehicle observation data sent by the aerial vehicle; planning a moving planning path based on the aerial vehicle observation data and the recommended path.

55. The method of claim 54, wherein, The aerial vehicle observation data includes at least one of the following: obstacle information near the movable platform, obstacle information in the direction of the movement of the movable platform, movement information of the movable platform, movement information of movable platforms around the movable platform.

56. The method of claim 54, wherein, The aerial vehicle observation data includes sensing data from the aerial vehicle bird's-eye view; The sensing data includes at least one of the following: laser radar data, visual image data, microwave radar data, ultrasonic data.

57. The method of claim 54, wherein, The moving planning path is planned based on the aerial vehicle observation data and the recommended path, including: optimizing a preliminary moving planning path based on the aerial vehicle observation data and the recommended path to obtain the moving planning path.

58. The method of claim 57, wherein, The preliminary moving planning path is determined based on input information of a user or / and sensing information of the movable platform.

59. A trajectory planning method, characterized by, The method includes: an aerial vehicle determines a recommended path of a movable platform based on acquired aerial vehicle observation data; the aerial vehicle sends the recommended path to the movable platform; the movable platform determines a moving planning path of the movable platform based on the recommended path and sends the moving planning path to the aerial vehicle; and the aerial vehicle receives the moving planning path and determines a landing planning trajectory for landing on the movable platform based on the moving planning path.

60. A trajectory planning apparatus characterized by comprising: including: a processor; and a memory storing computer program code executable on the processor; wherein the processor implements the steps of the method of any one of claims 1 to 48 when executing the computer program.

61. A trajectory planning apparatus characterized by comprising: including: a processor; and a memory storing computer program code executable on the processor; wherein the processor implements the steps of the method of any one of claims 49 to 58 when executing the computer program.

62. An aircraft, characterized in that including: a body; a power device provided in the body for providing power for the aerial vehicle; and the trajectory planning device of claim 60.

63. A movable platform, characterized by including: a body; a power device provided in the body for providing power for the movable platform; and the trajectory planning device of claim 61.

64. A trajectory planning system, comprising: including the aerial vehicle of claim 62 and the movable platform of claim 63.

65. A computer readable storage medium having stored thereon computer instructions, wherein, The instructions, when executed by the processor, implement steps of the method of any of claims 1-58.

Images