WO2021212462A1 - 移动控制方法、移动装置及移动平台 - Google Patents

移动控制方法、移动装置及移动平台 Download PDF

Info

Publication number
WO2021212462A1
WO2021212462A1 PCT/CN2020/086658 CN2020086658W WO2021212462A1 WO 2021212462 A1 WO2021212462 A1 WO 2021212462A1 CN 2020086658 W CN2020086658 W CN 2020086658W WO 2021212462 A1 WO2021212462 A1 WO 2021212462A1
Authority
WO
WIPO (PCT)
Prior art keywords
movement
trajectory
mobile platform
obstacle avoidance
instruction
Prior art date
Application number
PCT/CN2020/086658
Other languages
English (en)
French (fr)
Inventor
田原原
颜江
Original Assignee
深圳市大疆创新科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to PCT/CN2020/086658 priority Critical patent/WO2021212462A1/zh
Priority to CN202080004614.0A priority patent/CN112639656A/zh
Publication of WO2021212462A1 publication Critical patent/WO2021212462A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/10—Simultaneous control of position or course in three dimensions
    • G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft

Definitions

  • This application relates to the field of control technology, and in particular to a mobile control method, mobile device and mobile platform.
  • the existing assisted driving technology generally performs braking operations automatically when the UAV encounters an obstacle, even in some scenes where braking is not required, and the user's driving experience is poor.
  • the embodiments of the present application provide a mobile control method, a mobile device, and a mobile platform to improve user experience and improve the obstacle avoidance ability and accuracy of the mobile platform.
  • the first aspect of the embodiments of the present application is to provide a mobility control method, including:
  • a single-step trajectory with a predetermined step length is generated based on the user's control command and the obstacle avoidance assistance command, and the predetermined step length is related to the duration of action of the obstacle avoidance assistance command;
  • the single-step trajectory meets the preset obstacle avoidance condition, continue to generate a single-step trajectory with a predetermined step length based on the obstacle avoidance assistance instruction and the single-step trajectory;
  • a movement trajectory of the mobile platform that can bypass obstacles is generated, and the movement of the mobile platform is controlled according to the movement trajectory.
  • the second aspect of the embodiments of the present application is to provide a mobile control device, including:
  • the memory is used to store program instructions
  • the processor calls the program instructions, and when the program instructions are executed, is used to perform the following operations:
  • a single-step trajectory with a predetermined step length is generated based on the user's control command and the obstacle avoidance assistance command, and the predetermined step length is related to the duration of action of the obstacle avoidance assistance command;
  • the single-step trajectory meets the preset obstacle avoidance condition, continue to generate a single-step trajectory with a predetermined step length based on the obstacle avoidance assistance instruction and the single-step trajectory;
  • a movement trajectory of the mobile platform that can bypass obstacles is generated, and the movement of the mobile platform is controlled according to the movement trajectory.
  • the third aspect of the embodiments of the present application is to provide a mobile platform, including:
  • a power system installed on the fuselage and used to provide power for the mobile platform
  • the fourth aspect of the embodiments of the present application is to provide a computer-readable storage medium on which a computer program is stored, and the computer program is executed by a processor to implement the method described in the first aspect.
  • a movement track of the mobile platform that can bypass obstacles is generated, thereby controlling the movement of the mobile platform, so that the user can avoid It is necessary to consider obstacle avoidance to control the movement of the mobile platform, while ensuring the safe movement of the mobile platform, avoiding the situation in the prior art that predicts that the mobile platform is about to hit an obstacle and immediately takes the decision to stop, which can extend the mobile platform’s Moving distance.
  • Fig. 1 is a schematic diagram of a mobile platform provided by an embodiment of the present application.
  • FIGS. 2a and 2b are schematic diagrams of a mobile platform and obstacles provided by an embodiment of the present application.
  • FIG. 3 is a schematic top view of a mobile platform provided by an embodiment of the application.
  • Figure 4 is a flowchart of a mobility control method provided by an embodiment of the present application.
  • FIG. 5 is a schematic diagram of an iterative generation flow of a movement trajectory provided by an embodiment of the present application.
  • Fig. 6 is a schematic diagram of a movement track provided by an embodiment of the present application.
  • FIG. 7a is a schematic diagram of comparison between a trajectory predicted based on user instructions and a trajectory predicted after superimposing obstacle avoidance assistance instructions provided by an embodiment of the present application;
  • Fig. 7b is a schematic diagram of the obstacle avoidance assistance command superimposed in Fig. 7a;
  • Fig. 8 is a schematic structural diagram of a mobile platform provided by an embodiment of the present application.
  • This embodiment provides a method for controlling a mobile platform.
  • the control method of the embodiment of the present application is suitable for controlling various mobile platforms.
  • a drone is taken as an example for description, but the mobile platform is not limited to drones, but is suitable for various other movable vehicles, such as unmanned vehicles, unmanned ships, robots, etc. Carrier.
  • the drone 10 is equipped with a photographing device 101.
  • the photographing device 101 is mounted on the fuselage of the drone 10 through a supporting device 102.
  • the supporting device 102 may specifically be a pan/tilt, and the photographing device 101 is used in the drone. Capture images or record videos during exercise.
  • the user can control the flight of the drone through a remote control device 11 on the ground.
  • the remote control device 11 can be a remote control device corresponding to the drone (for example, a remote control with a screen), a mobile phone, a tablet computer, a notebook computer, and other devices.
  • the remote control device may be provided with an operating element for the user to operate, such as a joystick.
  • the remote control device generates a control stick amount according to the user's operation of the joystick.
  • the remote control device sends the control stick amount to the drone 10 through wireless communication.
  • the drone 10 is provided with a communication module 103 and a flight controller 104. After receiving the control stick amount from the remote control device through the communication module 103, the flight controller 104 generates a control instruction according to the control stick amount to control the flight of the drone.
  • the operating elements set on the remote control device for user operation are not limited to joysticks, for example, they can also be icons or virtual keys displayed on the screen of the remote control device, or the operating elements can also be set on the body of the remote control device. Physical keys or buttons.
  • the flight controller may include one or more processors, and the processors may individually or jointly generate control instructions to control the flight of the drone.
  • the drone is also provided with a driver, and the driver is, for example, a motor.
  • the motor can be coupled with one or more power units of the drone.
  • the power unit may include a rotor.
  • the flight controller can control the action of the driver, so that the driver drives the rotor to rotate, thereby generating power for the drone.
  • the drone is also equipped with sensors.
  • the sensors may include, but are not limited to, GPS receivers, Inertial Measurement Units (IMU, Inertial Measurement Unit) and other types of sensors.
  • the UAV can obtain the position and attitude parameters of the UAV through the sensor.
  • the position parameters may include: the position, linear velocity, and linear acceleration of the drone.
  • the attitude parameters may include the attitude angle, attitude angular velocity, and attitude angular acceleration of the UAV, and the attitude angle may include the yaw angle, pitch angle, and roll angle of the UAV.
  • the user can control the flight of the drone through a remote control device, and the remote control device communicates wirelessly with the flight controller of the drone.
  • the remote control device is provided with an operating element for the user to operate, such as a joystick.
  • the user generates the control stick amount by operating the joystick, and the control stick amount is sent to the flight controller through wireless communication.
  • the flight controller generates control instructions according to the control stick amount to control the action of the driver, thereby controlling the flight of the drone.
  • the embodiment of the present application provides a movement control method, which is used to generate a movement trajectory that can bypass obstacles based on the user's control instructions and obstacle avoidance assistance instructions when the mobile platform is in a user control mode, and to control the movement of the mobile platform , It is convenient for users to control the movement of the mobile platform without considering obstacle avoidance, while ensuring the safe flight of the mobile platform, avoiding the situation that the prior art predicts that the mobile platform is about to hit an obstacle and immediately takes the decision to stop. Can extend the flying distance of the mobile platform.
  • the state space of the mobile platform is discretized through a series of small single-step trajectories, which can generate a larger number and finer-grained trajectories, which can facilitate the screening of smoother and circumstantial trajectories, and the degree of freedom of the circumvention plan is greater. High, higher accuracy, and the probability of successfully flying around obstacles is also greater.
  • the mobile platform stores map information of the current environment.
  • the mobile platform detects that the current position or the future movement track is less than a predetermined distance from the obstacle, or predicts that it will be at the current speed within a predetermined time
  • the mobile platform starts to perform the operation of generating obstacle avoidance assistance instructions.
  • the map information of the current environment stored on the mobile platform may be downloaded from a server or acquired based on detection data of sensors on the mobile platform.
  • the sensor may include a vision sensor (for example, a binocular camera, a monocular camera) and/or a distance sensor (for example, a TOF camera, a lidar).
  • the map information may be acquired by the unmanned aerial vehicle based on the detection data of the sensor in the same flight or in different flights, where the unmanned aerial vehicle is located next to each other. The flight between the take-off and landing as a flight.
  • the activation of the auxiliary obstacle avoidance mode on the mobile platform may be triggered based on an instruction input by the user.
  • the user for example, the user’s operation interface used to control the mobile platform is provided with a physical button or a virtual button, or the operation interface is set with an auxiliary obstacle avoidance mode option.
  • the user detects the physical button or virtual button or auxiliary obstacle avoidance mode When operating the option, make sure to enter the auxiliary obstacle avoidance mode of the mobile platform.
  • the mobile platform when the mobile platform turns on the assisted obstacle avoidance mode, it can also be detected when the distance between the current position or the future movement track and the obstacle is less than a predetermined distance, or it is predicted that it will hit the obstacle within a predetermined time at the current speed. It is automatically turned on by default when the distance between the current position or the future movement track and the obstacle is detected to be less than the predetermined distance, and the current speed direction of the mobile platform is facing the obstacle. In some embodiments, the user may choose to turn off the function of automatically turning on the auxiliary obstacle avoidance mode by default.
  • the obstacle avoidance assistance instruction is always generated during the movement of the mobile platform, but only under certain conditions is the obstacle avoidance assistance instruction generated to generate the orbiting trajectory of the mobile platform (the trajectory that can bypass obstacles) .
  • the obstacle avoidance assistance commands include linear acceleration commands, and the magnitude, direction of action, or duration of action of each linear acceleration command are not completely the same.
  • the obstacle avoidance assistance instruction set may also include a linear acceleration instruction with a magnitude of 0, that is, no intervention is made to the user's control instruction.
  • the obstacle avoidance assistance command set includes linear acceleration commands of various sizes, various directions of action, and various durations of action.
  • a plurality of different obstacle avoidance auxiliary commands can be selected from the obstacle avoidance auxiliary command set to intervene the control commands input by the user according to the current relative position relationship with the obstacle.
  • the selected obstacle avoidance assistance instruction can increase the linear acceleration component of the mobile platform along the first direction, that is, the linear acceleration direction of the mobile platform corresponding to the selected obstacle avoidance assistance instruction is the first direction, or, When the mobile platform moves under the control of the manipulation instruction input by the user, when the obstacle avoidance assistance instruction is added, the acceleration component of the linear velocity of the mobile platform in the first direction increases.
  • the mobile platform After the mobile platform increases the control of the obstacle avoidance assistance instruction, it will change the original movement trajectory (that is, the movement trajectory of the mobile platform only under the control of the manipulation instruction input by the user).
  • the first direction is different from the direction in which the mobile platform faces the obstacle, for example, the first direction is perpendicular to the direction in which the mobile platform faces the obstacle.
  • the direction of the mobile platform towards the obstacle can be the direction of the shortest connection between the mobile platform and the obstacle, or the direction of the connection between a certain point on the mobile platform and a certain point of the obstacle.
  • the direction of the mobile platform facing the obstacle can be defined as the direction of the mobile platform facing the obstacle in the horizontal direction.
  • the direction of movement, or the linear direction of the movement of the mobile platform to the obstacle is not limited to the definition shown in Figs. 2a and 2b.
  • FIG. 3 is a schematic top view of a mobile platform provided by an embodiment of the application.
  • the direction x is the direction of the mobile platform toward the obstacle
  • the direction y is the direction of linear acceleration applied by the obstacle avoidance assistance instruction.
  • the direction y is perpendicular to the direction x, and the mobile platform changes the current movement trajectory under the action of the linear acceleration component in the direction y.
  • the angle between the linear acceleration in the direction y and the speed in the direction x is 90 degrees as an example. In this way, the speed of the mobile platform toward the obstacle caused by the speed command applied by the user in the x direction does not need to be reduced or offset.
  • the component (that is, the velocity component in the direction x), only changes the moving trajectory of the mobile platform, so as to achieve the purpose of avoiding obstacles or allowing the mobile platform to move for a period of time before the collision.
  • the angle between the linear acceleration in the direction y and the velocity in the direction x can also be greater than 90 degrees, and the acceleration in the direction y can be decomposed into the acceleration component perpendicular to the direction x and the direction.
  • the acceleration component opposite to the x direction, where the acceleration component perpendicular to the direction x can change the moving trajectory of the mobile platform, and the acceleration component opposite to the direction x can reduce or offset the speed component of the mobile platform towards the obstacle caused by the user's control instructions.
  • FIG. 3 is only an example for illustration, rather than a sole limitation to the present invention.
  • the mobile platform determines the target direction of the mobile platform based on the manipulation instruction currently input by the user, and generates at least one predicted trajectory that bypasses the obstacle and can move toward the target direction according to the manipulation instruction and the obstacle avoidance assistance instruction, from A target predicted trajectory is determined in the at least one predicted trajectory, a control instruction capable of enabling the mobile platform to move along the target predicted trajectory is generated based on the target predicted trajectory, and the movement of the mobile platform is controlled based on the control instruction.
  • the target direction is the same as the speed direction of the mobile platform corresponding to the manipulation instruction currently input by the user.
  • the mobile platform predicts the user-input control instruction in a certain time window in the future based on the current control instruction input by the user, and determines the target direction of the mobile platform based on the predicted control instruction.
  • the target direction is the same as the speed direction of the mobile platform corresponding to the predicted control instruction. It is understandable that when the manipulation instruction input by the user changes, the obstacle avoidance assistance instruction may change along with it.
  • the speed direction of the mobile platform corresponding to the instruction mentioned in this article refers to the direction of movement of the mobile platform when the mobile platform controls the movement based on the instruction when the mobile platform is stationary.
  • the mobile platform predicts the manipulation instructions input by the user in a certain time window in the future based on the manipulation instructions input by the user, and selects a variety of obstacle avoidance assistance instructions based on the manipulation instructions input by the user based on specific rules;
  • the movement state of the user and at least one of the following instructions: the current control instruction input by the user, the current obstacle avoidance assist instruction used to control the movement of the mobile platform, the predicted control instruction input by the user in a certain time window in the future, for a certain time window in the future.
  • the condition determines a target movement trajectory from the multiple candidate movement trajectories, and controls the movement of the mobile platform in a certain time window in the future according to the target movement trajectory.
  • the mobile platform regards the currently inputted manipulation instruction as the predicted manipulation instruction input in a certain time window in the future.
  • the user inputs control instructions through the joystick on the remote control, and the amount of stick input by the user includes the amount of roll, pitch, yaw, and thr. ).
  • the physical model of the remote control rocker is established through the Kalman filter, and the physical model can add factors such as rocker spring, resistance, and so on. The user's force on each stick on the remote control rocker is input as input into the physical model, and the forecast of each stick on the remote remote control is output.
  • Fig. 4 is a flowchart of a mobility control method provided by an embodiment of the present application. As shown in Fig. 4, the method includes the following steps:
  • Step S401 In the user manipulation mode, a single-step trajectory of a predetermined step length is generated based on the user's manipulation instruction and the obstacle avoidance assistance instruction, and the predetermined step length is related to the duration of action of the obstacle avoidance assistance instruction.
  • Step S402 If the single-step trajectory meets the preset obstacle avoidance condition, continue to generate a single-step trajectory with a predetermined step length based on the obstacle avoidance assistance instruction and the single-step trajectory.
  • Step S403 Generate a movement trajectory of the mobile platform that can bypass obstacles according to the single-step trajectory, and control the movement of the mobile platform according to the movement trajectory.
  • the user control mode involved in this embodiment refers to a control mode in which the user controls the movement track and/or movement state of the mobile platform through a handheld remote control or other control devices.
  • the mobile platform involved in this embodiment may be a device such as a drone, a car, etc., which has a certain processing capability and can be controlled by a control device.
  • the following embodiments mainly take a drone as an example for description, and the implementation process of other mobile platforms is similar, and will not be repeated.
  • the user's manipulation instruction may be triggered by controlling the amount of the control lever of the mobile platform, and the amount of the control lever may be generated at the remote control device.
  • the user can manually control the flight of the drone by pushing and pulling the joystick of the remote control device to generate the amount of control stick, which corresponds to the desired trajectory of the user.
  • the remote rod for example, a pitch rod
  • the remote control device for example, the pitch stick and the yaw stick
  • the drone After the drone receives the control stick amount sent by the remote control device, it will generate a control command based on the control stick amount, but it will not directly control the drone according to the generated control command, but will make trajectory prediction based on the control stick amount.
  • the generated control instruction is used as the initial control instruction sequence, and the obstacle avoidance assistance instruction is superimposed based on the initial control instruction sequence to predict and generate the movement trajectory of the drone that can fly around the obstacle, and then generate the control instruction according to the movement trajectory.
  • the control command controls the drone. Specifically:
  • the initial control command sequence is generated according to the amount of control lever.
  • this embodiment can obtain an instruction sequence according to the control lever amount, and use the instruction sequence as the initial control instruction sequence.
  • the initial manipulation instruction sequence of this embodiment may include: an initial linear velocity instruction sequence and an initial linear acceleration instruction sequence.
  • This embodiment does not limit the number of commands in the initial linear velocity command sequence and the initial linear acceleration command sequence (ie the value of N) and the time interval between adjacent commands, which can be set according to actual conditions and control effects. It is understandable that the initial control command sequence may be a control command predicted by the drone in a period of time T in the future based on the control command currently input by the user.
  • the lever amount can also be mapped to the linear acceleration command, and the linear velocity corresponding to the linear acceleration in the initial linear acceleration command sequence can be searched to obtain the initial linear velocity command sequence; alternatively, the control lever amount can be Simultaneously map to linear velocity command and linear acceleration command.
  • the obstacle avoidance assistance instruction is superimposed on the initial control instruction sequence to perform trajectory prediction to obtain the predicted trajectory of the UAV.
  • the process of determining the predicted trajectory first obtain the kinematics model of the UAV, and then use the kinematics model to predict the trajectory of the initial control instruction sequence superimposed with the obstacle avoidance assistance instruction, and obtain the predicted trajectory.
  • the predicted trajectory points are characterized by predicted location parameters.
  • the predicted position parameters include: predicted position; or, predicted position and predicted linear velocity; or, predicted position, predicted linear velocity, and predicted linear acceleration.
  • the obstacle avoidance assistance command includes a large number of linear acceleration commands, and the linear acceleration commands selected at different moments may be the same or different, and the obstacle avoidance assistance commands that can be used at each moment may also be one or more.
  • the obstacle avoidance auxiliary linear acceleration command can be superimposed to optimize or modify the initial linear acceleration command, so as to avoid the collision of the mobile platform with the obstacle. That is, after the auxiliary obstacle avoidance line acceleration command a_det k is superimposed on the initial linear acceleration command sequence a_cmd k , the intervened or corrected linear acceleration command sequence a_cmd' k can be obtained.
  • the initial linear acceleration command is also corrected based on at least one of the following factors: the linear acceleration control amount of the UAV flight controller, the centripetal acceleration of the UAV, and the air resistance experienced by the UAV.
  • the initial linear acceleration command sequence is corrected by using at least one of the following corrections: the linear acceleration control variable acc_ctrl of the flight controller of the drone, the centripetal acceleration acc_cent of the drone, and the drone received The air resistance acc_air.
  • the corrected initial linear acceleration command sequence a_cmd' k a_cmd k + a_det k + acc_ctrl-acc_cent-acc_air .
  • the following takes a uniform acceleration model as an example for description, but this embodiment is not limited to this, and any other type of kinematic model may also be used, such as, but not limited to: a uniform velocity model, a nonlinear model, etc.
  • the uniform acceleration model is as follows:
  • ⁇ t represents the time interval between adjacent moments
  • p k represents the position at time k
  • v k represents the linear velocity at time k
  • a k represents the linear acceleration at time k
  • v k+1 represents the linear velocity at time k+1
  • a k+1 represents the linear acceleration at time k+1.
  • the predicted position parameter of the predicted track point Specifically, the position p 0 , linear velocity v 0 and linear acceleration a 0 of the UAV at the current moment can be obtained through the sensors of the UAV.
  • the predicted position parameters include two parameters: predicted position and predicted linear velocity.
  • the initial control command sequence of this embodiment includes not only the initial linear velocity command sequence and the initial linear acceleration command sequence, but also the initial yaw rate command sequence and/or The initial yaw angular acceleration command sequence.
  • the predicted attitude parameters include: predicted yaw angle and predicted yaw rate.
  • the predicted trajectory of the UAV is determined according to the initial control instruction sequence. Similar to the previous embodiment, in the process of determining the predicted trajectory, first obtain the kinematics model of the UAV, and then use the kinematics model to predict the trajectory of the initial control instruction sequence superimposed with the obstacle avoidance assistance instruction to obtain the mobile platform
  • the predicted trajectory can bypass obstacles, and the predicted trajectory points in the predicted trajectory are characterized by predicted position parameters and predicted attitude parameters.
  • the predicted attitude parameters include: predicted yaw angle and predicted yaw rate.
  • the uniform acceleration model is as follows:
  • formulas (3) and (4) are the same as formulas (1) and (2) of the previous embodiment, and are used to predict the position parameters of the predicted trajectory.
  • Formulas (5) and (6) are used to predict the attitude parameters of the predicted trajectory; Represents the yaw rate at time k, Represents the angular velocity of the route at time k+1, ⁇ k represents the yaw angle at time k, and ⁇ k+1 represents the yaw angle at time k+1, Represents the yaw angular acceleration at time k.
  • the position parameter and the posture parameter of the predicted trajectory can be decoupled, and the position parameter and the posture parameter can be predicted respectively.
  • the prediction process of the position parameter is similar to the previous embodiment. For details, please refer to the description of the previous embodiment.
  • the position parameter at the current moment, the initial linear velocity command sequence, and the initial line with the acceleration command of the obstacle avoidance auxiliary line are superimposed.
  • the prediction process of the attitude parameters is as follows: first obtain the attitude parameters of the UAV at the current moment; input the current attitude parameters, the initial yaw angular velocity command sequence, and the initial yaw angular acceleration command sequence into the formula (5) and (6) Obtain the predicted attitude parameters of each predicted trajectory point in the predicted trajectory, and the predicted attitude parameters include: the predicted yaw angle and the predicted yaw angular velocity of the predicted trajectory point.
  • the UAV’s yaw angle ⁇ 0 and the yaw angle velocity at the current moment can be obtained through the UAV’s IMU And yaw angular acceleration Change the yaw angle ⁇ 0 and the yaw angle speed And yaw angular acceleration
  • the initial value and the initial yaw rate command sequence The initial yaw angular acceleration command sequence is substituted into formulas (5) and (6) for iterative operations to obtain the predicted yaw angle and predicted yaw rate at each predicted trajectory point of the predicted trajectory.
  • the predicted trajectories P 1 , P 2 ,..., P N predicted according to the initial position command and the initial posture command are obtained, and the predicted position, predicted linear velocity, and predicted linear acceleration of the predicted trajectory point , Predicted yaw angle, predicted yaw angular velocity, predicted yaw angular acceleration and other parameters can be obtained.
  • a smooth and continuous initial position command and initial attitude command can be generated, so that the predicted trajectory determined according to the initial position command and the initial attitude command is smooth and continuous, and the control effect on the flight trajectory of the UAV can be ensured.
  • a short segment of motion primitives is used to iteratively generate a complete movement trajectory in a single-step trajectory manner. For example, based on the UAV's current position, linear velocity, linear acceleration, and yaw angular velocity, traverse the obstacle avoidance assistance instruction set (including linear acceleration instruction set), combined with the predicted instructions, including linear velocity, linear acceleration, and yaw angular velocity , Yaw angular acceleration, substituted into the motion model to execute a fixed step length, and generate a single-step trajectory (that is, the motion primitive, the trajectory points of the single-step trajectory include position, linear velocity, linear acceleration, and yaw angle parameters) After the collision detection, only the trajectory that has not collided with the obstacle is retained, and the trajectory that collides with the obstacle is discarded.
  • the obstacle avoidance assistance instruction set including linear acceleration instruction set
  • a single-step trajectory is used to iteratively generate the above-mentioned movement trajectory. For example, traversing the obstacle avoidance assistance instruction set (including linear acceleration instruction set), combined with the predicted instructions, including linear velocity and linear acceleration, is substituted into the motion model to execute a fixed step length, and generate a single-step trajectory (that is, the motion primitive, on the trajectory
  • the trajectory points include position, linear velocity, and linear acceleration.
  • generating a complete movement trajectory requires many iterations.
  • Each small step and single-step trajectory generated requires collision detection.
  • multiple movement trajectories are obtained, and all feasible trajectories are searched for that are in line with the user's intention. If the target moving trajectory meets the obstacle avoidance condition, the drone is controlled to move according to the target moving trajectory. If the moving trajectory meeting the obstacle avoidance condition is not generated after the timeout, the drone is controlled to slow down and hover or return home.
  • the preset end condition includes that the total time length of generating the multiple movement trajectories reaches a first preset time length or the total number of the multiple movement trajectories generated reaches a preset number. That is to say, in this application, a time threshold or quantity threshold is set in advance. Within the time threshold or quantity threshold, the drone can continuously generate different obstacle avoidance conditions in the aforementioned single-step trajectory iteration method. Movement trajectory, when the time threshold or quantity threshold is reached, the generation of the movement trajectory will stop. In this way, a larger number of movement trajectories that meet the obstacle avoidance conditions can be iterated, and the range of options for the detour trajectory is larger. In a scene with complex obstacles, the possibility of choosing a detour trajectory is greater.
  • satisfying the obstacle avoidance condition may include that the distance between the mobile platform and the obstacle when moving on the moving track is greater than or equal to a set distance threshold.
  • the obstacle avoidance assistance command set contains multiple different linear acceleration commands, and the acceleration value, acceleration acting direction, or acceleration acting duration of each linear acceleration command are not completely the same.
  • the multiple linear acceleration instructions may further include an instruction that the acceleration is zero.
  • the acting direction of the linear acceleration can be determined according to the current relative position relationship between the drone and the obstacle, and the acting direction of the linear acceleration can be different from the direction the drone faces the obstacle, for example, it can be perpendicular to the direction of the drone The direction of the obstacle.
  • the mobile platform can preset the time length of each movement track (for example, 10 seconds), or can set the number of single-step trajectories for each movement track iteration (for example, 10 times), when a single movement track meets the time length or When the number is selected, the single movement track is generated.
  • the obstacle avoidance assistance command has a duration of action, for example, 1 second.
  • a short single-step trajectory ie motion primitive
  • the duration of the single-step trajectory is related to the role of the obstacle avoidance assistance command
  • the time is related, for example, it can be the same.
  • a short single-step trajectory is generated by superimposing the obstacle avoidance assistance command based on the predicted control instruction, and then performing collision detection and retaining The trajectory that does not collide with the obstacle is discarded, and the trajectory that will collide with the obstacle is discarded.
  • This process is executed in a loop until multiple trajectories that can bypass the obstacle are generated based on multiple single-step trajectories.
  • FIG. 5 is a schematic flow diagram of the iterative generation of a single-step trajectory during a curve movement provided by an embodiment of the present application.
  • Angular velocity, yaw angular acceleration and other information in the current state of the UAV, traverse the obstacle avoidance assistance instruction set (linear acceleration a), perform motion model prediction, and predict a future trajectory (including position p, linear velocity v, Linear acceleration a, yaw angular velocity, yaw angular acceleration).
  • the above-mentioned mapped linear velocity, linear acceleration superimposed with auxiliary commands, yaw angular velocity, yaw angular acceleration, and the current state of the UAV can be substituted into the motion model to execute a fixed step length, generate a single-step trajectory, and pass the collision.
  • the cost can be determined according to the size of the obstacle avoidance assistance instruction superimposed during the generation of the trajectory, the size of the change of the trajectory, the movable distance of the trajectory, or the energy consumed by the drone.
  • the larger the obstacle avoidance assistance commands superimposed during the trajectory generation process, or the greater the number of overlaps, or the greater the force of the superimposed obstacle avoidance assistance commands the greater the intervention of the user's intention commands.
  • the movable distance can be understood as the movable distance before the collision of the mobile platform.
  • the energy consumed can represent the energy consumed when the mobile platform moves on the moving track. For example, when the superimposed obstacle avoidance assistance commands are larger, the number of stacking times is greater, the force of the superimposed obstacle avoidance assistance commands is greater, the trajectory changes more drastically, the movable distance is smaller, or the energy consumed by the mobile platform is greater. The corresponding cost is higher.
  • the trajectory with the lowest cost can be selected from multiple trajectories as the target trajectory.
  • the final movement trajectories are all trajectories that the mobile platform will not hit obstacles, and they will not hit obstacles. Choose the one with the lowest obstacle avoidance intervention, the smoother the change, the larger the movable distance, or the lower the energy consumption among the trajectories, to achieve the effect of matching the user's intention and smooth movement.
  • obstacle collision detection may not be performed, that is, after the single-step trajectory is generated, iteratively generate other single-step trajectories based on the obstacle avoidance assistance instruction until the complete movement is generated. Trajectory. After obtaining multiple movement trajectories, when the target movement trajectory is finally selected, the distance to the obstacle is also used as a dimension to measure the cost. When the distance between the movement trajectory and the obstacle is greater, the corresponding cost The higher the distance, the smaller the distance between the movement track and the obstacle, the lower the corresponding cost.
  • Figure 6 is a schematic diagram of a single-step trajectory provided by an embodiment of the present application.
  • the generation process of a single-step trajectory is similar to a tree diagram.
  • the root of the tree can be understood as the initial predicted trajectory, which is superimposed on the basis of the initial control instructions.
  • Different obstacle avoidance auxiliary instructions generate multiple branches, and each branch corresponds to a single-step trajectory.
  • the length of each branch is equal to the duration of the obstacle avoidance auxiliary command.
  • the generated branches will not collide with obstacles. Branches, discard the branches that will collide with obstacles, and continue to superimpose different obstacle avoidance auxiliary commands at the ends of the reserved branches, thereby generating more branches that meet the obstacle avoidance conditions.
  • the searchable space is larger, and the freedom and room for choice are greater.
  • the branch stops growing, and when the number of all branches reaches the predetermined number or reaches the specified time period, the branch growth stops. Finally, the best one is selected from the multiple branches to control the drone flight.
  • the single-step trajectory method is used to iteratively generate the moving trajectory of the mobile platform. There is no need to plan in advance the change shape and quantity of the speed used to intervene in the auxiliary instruction to predict the trajectory. It can freely generate more pieces that meet the obstacle avoidance conditions based on the actual environment. Detour trajectory, the search density is larger, the granularity is smaller, and it is more free, and the probability of searching the detour trajectory is higher. And by setting the linear acceleration set used for obstacle avoidance assistance, the speed change of the mobile platform between single-step trajectories can be made smoother, and the situation where the speed sudden change causes the mobile platform to change drastically can be avoided.
  • the embodiment of the present application has greater freedom, more range of options, and a higher probability of searching for the detour trajectory. high.
  • the acceleration scheme of the intervention trajectory can achieve smoother and more stable movement of the mobile platform.
  • FIG. 7a is a schematic diagram of a trajectory predicted based on original instructions and an auxiliary detour trajectory under the action of an obstacle avoidance assist instruction provided by an embodiment of the present application.
  • a trajectory predicted based on original instructions and an auxiliary detour trajectory under the action of an obstacle avoidance assist instruction provided by an embodiment of the present application.
  • the mobile platform will collide with an obstacle.
  • the mobile platform can Move around obstacles, which not only conforms to the user's intention, that is, move forward, but also avoid obstacles and avoid collisions.
  • the detour trajectory is the entire detour trajectory generated iteratively based on a single-step trajectory.
  • Figure 7b is a schematic diagram of the obstacle avoidance assistance instruction involved in Figure 7a.
  • the magnitude of acceleration is not exactly the same.
  • it can randomly select or traverse the linear acceleration command from the linear acceleration command set to generate a single-step trajectory. Only the single-step trajectory that will not collide with obstacles is retained, and the iteration continues at the end of the single-step trajectory Generate a single-step trajectory until multiple complete movement trajectories are generated.
  • the controlling the movement of the mobile platform according to the movement trajectory includes: selecting a target movement trajectory from the plurality of movement trajectories; and controlling the movement according to the target movement trajectory The platform moves.
  • the target movement trajectory is selected according to one or more of the following selection strategies: the one or more movement trajectories have the smallest obstacle avoidance assistance instruction; the one or more movement trajectories The most gradual change among the ones; the one selected by the user from the one or more movement tracks; the one with the longest movable distance among the one or more movement tracks; the one with the least energy consumption among the one or more movement tracks; Among the one or more moving tracks, the movable distance is greater than the first preset threshold and the energy consumption is less than the second preset threshold.
  • the following uses an example to illustrate how to select the target movement track from the obtained movement tracks to control the movement of the mobile platform:
  • the one or more movement trajectories obtained by the above prediction are selected to obtain the movement trajectory with the longest movable distance as the target movement trajectory, or the one or more movement trajectories obtained by the above prediction are selected to select the mobile platform to consume the least energy
  • the movement trajectory of is regarded as the target movement trajectory.
  • the energy consumption refers to the energy consumed by the execution of various control commands when the mobile platform moves along the trajectory.
  • a control instruction is generated according to the target movement trajectory, thereby controlling the mobile platform to move along the target movement trajectory.
  • the rotation speed of the rotor of the mobile platform can be adjusted according to the position, linear velocity, linear acceleration, yaw angular velocity, and yaw angular acceleration of the track point in the target movement trajectory to achieve the linear velocity and angular velocity on the target movement trajectory.
  • the movement of the target's trajectory For example, when the user operates the pitch rod and yaw rod of the remote control device, the drone can be controlled to turn, and the drone can adjust the rotation speed of the rotor, so that the drone can change the roll angle to turn.
  • the attitude angle of the pan/tilt on the mobile platform can be adjusted according to the movement trajectory of the mobile platform, so as to achieve relatively stable changes between the pan/tilt and the mobile platform.
  • the roll angle of the gimbal can be adjusted according to the yaw angular velocity of the mobile platform.
  • the roll angle setting of the drone's gimbal is related to the drone's yaw angular velocity.
  • the gimbal roll will tilt to achieve
  • the shooting picture of the camera on the gimbal adjusts as the drone turns.
  • the user can experience the effect of driving the drone inside the drone to make a turn.
  • the pan/tilt tilts to the left along the roll axis, that is, the pan/tilt is low to the left and high to the right.
  • the gimbal tilts to the right along the roll axis, that is, the gimbal is higher left and lower right.
  • the yaw angular velocity of the drone is proportional to the angle at which the gimbal rotates. For example, when the drone's yaw angular velocity is greater, the angle of the gimbal's rotation along the roll axis is larger, and when the drone's yaw angular velocity is smaller, the angle of the gimbal's rotation along the roll axis is smaller.
  • the movement trajectory of the mobile platform is controlled based on the control instructions and obstacle avoidance assistance instructions input by the user, so that in the user control mode, the active obstacle avoidance of the mobile platform can also be realized, so that the mobile platform It can avoid obstacles under the combined action of the control commands input by the user and the obstacle avoidance assistance commands. It improves the security and user experience of the mobile platform.
  • FIG. 8 is a schematic structural diagram of a mobile device provided by an embodiment of the present application.
  • the mobile device 80 includes a processor 81 and a memory 82.
  • the processor 81 is configured to generate a predetermined step size based on a user's manipulation instruction and obstacle avoidance assistance instruction.
  • the single-step trajectory of the mobile platform is generated iteratively, and the moving trajectory of the mobile platform can bypass obstacles, and the mobile platform is controlled to move according to the moving trajectory.
  • the mobile device 80 may also include a detection device. When the detection device detects that the distance between the mobile platform and the obstacle is less than a predetermined distance, the processor 81 is triggered to generate an obstacle avoidance assistance instruction, which is then based on the user's manipulation instruction and avoidance.
  • the obstacle assistance instruction generates a single-step trajectory with a predetermined step length.
  • one or more obstacle avoidance assistance instructions are selected. Among them, one or more of the linear acceleration value, the linear acceleration acting direction or the acting time length corresponding to different obstacle avoidance auxiliary commands is different.
  • the processing methods of the processor 81 include the following:
  • the processor 81 determines whether to generate an obstacle avoidance assistance instruction according to the manipulation instruction input by the user. For example, when the processor 81 determines that the manipulation instruction input by the user causes a collision risk between the mobile platform and the obstacle , Select the obstacle avoidance assistance command to change the movement track of the mobile platform through the obstacle avoidance assistance command. If it is determined that the control command input by the user will not cause a collision, the obstacle avoidance assistance command is not selected.
  • the processor 81 when the detection device detects that the distance between the mobile platform and the obstacle is less than a predetermined distance, the processor 81 directly selects the obstacle avoidance assistance instruction without detecting whether the user input control instruction will be Cause a collision.
  • the processor 81 automatically generates obstacle avoidance assistance instructions by default, that is, regardless of whether there is a danger of collision between the mobile platform and the obstacle due to the manipulation instructions input by the user, the processor 81 automatically based on the obstacle avoidance assistance instructions And the control instructions input by the user control the movement of the mobile platform.
  • the mobile device provided in this embodiment can execute the movement control method provided in the foregoing embodiment, and its execution mode and beneficial effects are similar, and details are not described herein again.
  • the embodiment of the present application also provides a mobile platform, which includes:
  • a power system installed on the fuselage and used to provide power for the mobile platform
  • the mobile platform may further include a sensor installed on the body for detecting and obtaining map information of the environment in which the mobile platform is located.
  • the senor includes a vision sensor and/or a distance sensor.
  • the mobile platform further includes:
  • the communication equipment is installed on the fuselage and is used for information interaction with the ground station.
  • the mobile platform includes at least one of the following: unmanned aerial vehicle and automobile.
  • this embodiment also provides a computer-readable storage medium on which a computer program is stored, and the computer program is executed by a processor to implement the control method of the mobile platform described in the foregoing embodiment.
  • the disclosed device and method may be implemented in other ways.
  • the device embodiments described above are merely illustrative, for example, the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or It can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit may be implemented in the form of hardware, or may be implemented in the form of hardware plus software functional units.
  • the above-mentioned integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium.
  • the above-mentioned software functional unit is stored in a storage medium, and includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute the method described in each embodiment of the present application. Part of the steps.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program instructions .

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

本申请实施例提供一种移动控制方法、移动装置及移动平台，其中该方法包括在用户操控模式中，基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹，所述预定步长与所述避障辅助指令的作用时间长度相关；若所述单步轨迹满足预设避障条件，则继续基于所述避障辅助指令以及所述单步轨迹生成预定步长的单步轨迹，根据所述单步轨迹生成移动平台可绕开障碍物的移动轨迹，并根据所述移动轨迹控制所述移动平台移动。本申请实施例能够提高避障的精度。

Description

移动控制方法、移动装置及移动平台 技术领域
本申请涉及控制技术领域，尤其涉及一种移动控制方法、移动装置及移动平台。
背景技术
随着无人机越来越普及，更多的人加入了无人机航拍的行列。但是对于之前从未使用过无人机的用户来说，操作是个问题，稍有不慎容易造成坠毁撞击。因此，对于这些用户来说需要一些辅助驾驶技术来帮助用户进行避障。
现有的辅助驾驶技术一般是在无人机一遇到障碍物时就自动执行刹车操作，即使是在一些不需要执行刹车的场景中，也会执行刹车，用户的驾驶体验较差。
发明内容
本申请实施例提供一种移动控制方法、移动装置及移动平台，用以提高用户体验，提高移动平台避障的能力和精准度。
本申请实施例的第一方面是提供一种移动控制方法，包括：
在用户操控模式中，基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹，所述预定步长与所述避障辅助指令的作用时间长度相关；
若所述单步轨迹满足预设避障条件，则继续基于所述避障辅助指令以及所述单步轨迹生成预定步长的单步轨迹；
根据所述单步轨迹生成移动平台可绕开障碍物的移动轨迹，并根据所述移动轨迹控制所述移动平台移动。
本申请实施例的第二方面是提供一种移动控制装置，包括：
存储器和处理器；
所述存储器用于存储程序指令；
所述处理器，调用所述程序指令，当程序指令被执行时，用于执行以下操作：
在用户操控模式中，基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹，所述预定步长与所述避障辅助指令的作用时间长度相关；
若所述单步轨迹满足预设避障条件，则继续基于所述避障辅助指令以及所述单步轨迹生成预定步长的单步轨迹；
根据所述单步轨迹生成移动平台可绕开障碍物的移动轨迹，并根据所述移动轨迹控制 所述移动平台移动。
本申请实施例的第三方面是提供一种移动平台，包括：
机身；
动力系统，安装在所述机身，用于为所述移动平台提供动力；
以及上述第二方面提供的移动装置。
本申请实施例的第四方面是提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行以实现第一方面所述的方法。
本申请实施例，在移动平台处于用户操控模式中时，基于用户输入的操控指令以及避障辅助指令来生成移动平台可绕开障碍物的移动轨迹，从而控制移动平台的移动，使得用户可以不需要考虑避障来控制移动平台的移动的同时，保障移动平台的安全移动，避免了现有技术中预测到移动平台即将撞上障碍物时立即采取刹住的决策的情况，可以延长移动平台的移动距离。
附图说明
图1是本申请实施例提供的移动平台的示意图；
图2a和图2b是本申请实施例提供的移动平台与障碍物的示意图；
图3为本申请实施例提供的一种移动平台的俯视示意图；
图4是本申请实施例提供的移动控制方法的流程图；
图5是本申请实施例提供的一种移动轨迹的迭代生成流程示意图；
图6是本申请实施例提供的一种移动轨迹的示意图；
图7a是本申请实施例提供的一种基于用户指令预测的轨迹与叠加避障辅助指令后预测的轨迹的对比示意图；
图7b是图7a中叠加的避障辅助指令的示意图；
图8是本申请实施例提供的移动平台的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚地描述。
本实施例提供了一种移动平台的控制方法。本申请实施例的控制方法适应于对各种移动平台进行控制。为描述方便，在本申请实施例中以无人机为例进行说明，但该移动平台并不限于无人机，而是适用于例如无人车、无人船、机器人等其他各种可移动载体。
首先对本实施例控制方法适用的无人机进行说明。如图1所示，给出无人机的示例。 无人机10上搭载有拍摄装置101，具体的，拍摄装置101通过支撑设备102搭载在无人机10的机身上，支撑设备102具体可以是云台，拍摄装置101用于在无人机运动过程中捕捉图像或录制视频。另外，用户可以通过地面的遥控设备11控制无人机的飞行，该遥控设备11可以是无人机对应的遥控设备(例如，带屏遥控器)、手机、平板电脑、笔记本电脑等设备。具体的，遥控设备可设置有供用户操作的操作元件，例如摇杆。遥控设备根据用户对摇杆的操作生成控制杆量，进一步，遥控设备将该控制杆量通过无线通信发送给无人机10，例如，无人机10内设置有通信模块103和飞行控制器104，飞行控制器104通过通信模块103从遥控设备接收到控制杆量后，根据该控制杆量生成控制指令，以控制无人机的飞行。可以理解的是，遥控设备上设置的供用户操作的操作元件不限于摇杆，例如，还可以是遥控设备屏幕上显示的图标或者虚拟按键，或者该操作元件还可以是遥控设备机身上设置的物理按键或按钮。
其中，所述飞行控制器可以包括一个或多个处理器，所述处理器可以单独或共同产生控制指令，以控制所述无人机的飞行。
无人机还设置有驱动器，驱动器例如是电机。电机可以与无人机的一个或多个动力单元耦合。所述动力单元可以包括旋翼。所述飞行控制器可以控制驱动器的动作，以使驱动器带动旋翼旋转，从而为无人机产生动力。
无人机还设置有传感器。所述传感器可以包括但不限于GPS接收机、惯性测量单元(IMU,Inertial Measurement Unit)等各种类型的传感器。无人机可通过传感器得到无人机的位置和姿态参数。位置参数可以包括：无人机的位置、线速度、线加速度。姿态参数可以包括：无人机的姿态角、姿态角速度、姿态角加速度，姿态角可以包括无人机的偏航角、俯仰角、横滚角。
示例性的，用户可通过遥控设备控制无人机的飞行，遥控设备与无人机的飞行控制器无线通信。遥控设备设置有供用户操作的操作元件，例如摇杆。用户通过操作摇杆生成控制杆量，控制杆量通过无线通信发送给飞行控制器，飞行控制器根据控制杆量生成操控指令，以控制驱动器的动作，从而控制无人机的飞行。
下面结合附图，对本发明的一些实施方式作详细说明。在不冲突的情况下，下述的实施例及实施例中的特征可以相互组合。
本申请实施例提供一种移动控制方法，用于在移动平台处于用户操控模式中时，基于用户的操控指令和避障辅助指令来生成可绕开障碍物的移动轨迹，并控制移动平台的移动，便于用户可以不需要考虑避障来控制移动平台的移动的同时，保障移动平台的安全飞行， 避免了现有技术中预测到移动平台即将撞上障碍物时立即采取刹住的决策的情况，可以延长移动平台的飞行距离。并且，通过一系列小段的单步轨迹，来离散化移动平台的状态空间，可以生成数量更加庞大且粒度更加精细的轨迹，可便于筛选更加平滑可绕行的轨迹，绕行方案的自由度更高，精准度更高，可成功绕开障碍物飞行的概率也更大。
本申请实施例中，生成避障辅助指令的触发方式有多种。
在一些实施例中，移动平台存储有当前环境的地图信息，当移动平台检测到当前位置或未来的移动轨迹与障碍物的距离小于预定距离时，或者预测到以当前的速度在预定时间内会撞上障碍物时，或者检测到当前位置或未来的移动轨迹与障碍物的距离小于预定距离，且移动平台的当前速度方向朝向障碍物时，或者在用户设置开启或移动平台默认设置开启了移动平台的辅助避障模式之后，移动平台开始执行生成避障辅助指令的操作，。
其中，移动平台所存储的当前环境的地图信息可以是从服务器上下载的，或者是基于移动平台上的传感器的探测数据所获取的。其中，该传感器可以包括视觉传感器(例如双目相机、单目相机)和/或距离传感器(例如TOF相机、激光雷达)。例如，在移动平台为无人飞行器的实施例中，该地图信息可以是无人飞行器基于传感器在同一次飞行或者在不同次飞行中的探测数据所获取的，其中，以无人飞行器在相邻的起飞和降落之间的飞行作为一次飞行。
其中，移动平台开启辅助避障模式可以是基于用户输入的指令触发的。例如，用户用于控制移动平台的操作界面上设置有物理按键或者虚拟按键，或者该操作界面上设置有辅助避障模式的选项，当检测到用户对该物理按键或虚拟按键或辅助避障模式的选项的操作时，确定进入移动平台的辅助避障模式。
可选的，移动平台开启辅助避障模式也可以是在检测到当前位置或未来的移动轨迹与障碍物的距离小于预定距离时，或者预测到以当前的速度在预定时间内会撞上障碍物时，或者检测到当前位置或未来的移动轨迹与障碍物的距离小于预定距离，且移动平台的当前速度方向朝向障碍物时默认自动开启的。在一些实施例中，用户可以选择关闭该默认自动开启辅助避障模式的功能。
在一些实施例中，避障辅助指令在移动平台的移动中一直产生，但仅在某些条件下才基于该避障辅助指令来生成移动平台的绕行轨迹(可绕开障碍物的轨迹)。
本申请实施例中，避障辅助指令有多种，组成避障辅助指令集，其中，每种避障辅助指令的值、作用方向或作用时长不完全相同。示例性的，避障辅助指令包括线加速度指令，每个线加速度指令的大小、作用方向或作用时长不完全相同。可选的，该避障辅助指令集 中还可以包括大小为0的线加速度指令，即不对用户的操控指令进行干预。避障辅助指令集中包含各个大小、各个作用方向、各个作用时长的线加速度指令。在移动平台实际使用过程中，可以根据当前与障碍物的相对位置关系来从避障辅助指令集中选取多个不同的避障辅助指令来干预用户输入的操控指令。其中，选取的避障辅助指令能够增加移动平台沿第一方向上的线加速度分量，也即是说，选取的避障辅助指令所对应的移动平台的线加速度方向为该第一方向，或者，移动平台在基于用户输入的操控指令的控制下移动时，当增加该避障辅助指令后，移动平台的线速度在第一方向上的加速度分量增加。移动平台在增加避障辅助指令的控制后，将改变原有的移动轨迹(也即移动平台仅仅在用户输入的操控指令控制下的移动轨迹)。例如，第一方向与移动平台朝向障碍物的方向不同，例如，第一方向与移动平台朝向障碍物的方向垂直。
其中，移动平台朝向障碍物的方向可以是移动平台与障碍物的最短连线所在的方向，或者是移动平台上的某一个点与障碍物的某一个点的连线所在的方向，在此不做限制。以图2a和图2b是本申请实施例提供的移动平台移动的侧视图为例，如图2a和图2b所示，移动平台朝向障碍物的方向可以定义为移动平台在水平方向上朝向障碍物移动的方向，或者移动平台向障碍物移动的直线方向，但不局限于图2a和图2b所示的定义方式。
示例的，图3为本申请实施例提供的一种移动平台的俯视示意图，如图3所示，方向x为移动平台朝向障碍物的方向，方向y为避障辅助指令施加的线加速度方向，方向y与方向x之间垂直，移动平台在方向y上的线加速度分量的作用下改变当前的移动轨迹。本实施例以方向y上的线加速度与方向x上的速度之间的角度为90度为例，如此可以不用减少或抵消用户在x方向上施加的速度指令造成的移动平台朝向障碍物的速度分量(即方向x上的速度分量)，只改变移动平台的移动轨迹，从而达到避障或者让移动平台在碰撞前多移动一段时间的目的。在其他可能的实现方式中，方向y上的线加速度与方向x上的速度之间的角度还可以是大于90度，且方向y上的加速度可以分解为垂直于方向x的加速度分量和与方向x方向相反的加速度分量，其中，垂直于方向x的加速度分量可以改变移动平台的移动轨迹，与方向x方向相反的加速度分量可以减小或抵消用户操控指令造成的移动平台朝向障碍物的速度分量(即方向x上的速度分量)，从而达到避障或者让移动平台在碰撞前多移动一段时间的目的。当然图3仅为示例说明，而不是对本发明的唯一限定。
在一些实施例中，移动平台基于用户当前输入的操控指令确定移动平台的目标方向，根据操控指令和避障辅助指令生成至少一个绕过该障碍物且能够朝向该目标方向移动的 预测轨迹，从该至少一个预测轨迹中确定目标预测轨迹，基于该目标预测轨迹生成能够使得移动平台能够沿着该目标预测轨迹移动的控制指令，并基于该控制指令控制移动平台的移动。
其中，可选的，目标方向与用户当前输入的操控指令对应的移动平台的速度方向相同。或者，移动平台基于用户当前输入的操控指令预测未来一定时间窗口内的用户输入的操控指令，并基于该预测的操控指令确定移动平台的目标方向。可选的，目标方向与预测的操控指令对应的移动平台的速度方向相同。可以理解的是，当用户输入的操控指令改变时，避障辅助指令可能会随之一起改变。
需要说明的是，本文中提到的指令对应的移动平台的速度方向，指的是移动平台在静止的情况下基于该指令控制移动时移动平台的移动方向。
在一些实施例中，移动平台根据用户输入的操控指令来预测未来一定时间窗口内用户输入的操控指令，以及基于特定规则根据用户输入的操控指令来选择多种避障辅助指令；移动平台基于当前的运动状态，以及以下至少一项指令：用户当前输入的操控指令、当前用于控制移动平台移动的避障辅助指令、预测用户在未来一定时间窗口内输入的操控指令、用于未来一定时间窗口内的多种备选避障辅助指令，来分别预测在未来一定时间窗口内移动平台在不同的指令或者指令组合作用下的移动轨迹，该预测的多种移动轨迹作为备选移动轨迹，根据预定条件从该多种备选移动轨迹中确定出一条目标移动轨迹，并根据该目标移动轨迹在未来一定时间窗口内控制移动平台的运动。
需要说明的是，在一些场景下，例如基于用户输入的操控指令不会导致移动平台在一定时间内撞上障碍物的场景中，该目标移动轨迹对应的指令中可能没有避障辅助指令，那么在未来一定时间窗口内，仅仅基于用户输入的操控指令控制移动平台的移动，避障辅助指令没有被用于控制移动平台。
其中，移动平台预测用户在未来一定时间窗口内输入的操控指令的方法有多种。例如，移动平台将当前输入的操控指令作为在未来一定时间窗口内输入的预测操控指令。又例如，用户通过遥控器上的摇杆来输入操控指令，用户输入的杆量包括横滚杆量(roll)、俯仰杆量(pitch)、偏航杆量(yaw)、油门杆量(thr)。通过卡尔曼滤波器建立遥控器摇杆的物理模型，该物理模型可以加入摇杆弹簧、阻力等等因素。将用户对遥控器摇杆上在各杆量上的力作为输入输入到该物理模型中，输出对未来遥控器的各杆量预测。
下面对本申请实施例中的移动控制方法进行举例描述。
本申请实施例提供一种移动控制方法。图4是本申请实施例提供的移动控制方法的流 程图，如图4所示，该方法包括如下步骤：
步骤S401、在用户操控模式中，基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹，所述预定步长与所述避障辅助指令的作用时间长度相关。
步骤S402、若所述单步轨迹满足预设避障条件，则继续基于所述避障辅助指令以及所述单步轨迹生成预定步长的单步轨迹。
步骤S403、根据所述单步轨迹生成移动平台可绕开障碍物的移动轨迹，并根据所述移动轨迹控制所述移动平台移动。
本实施例涉及的用户操控模式是指用户通过手持遥控器或其他操控设备控制移动平台的移动轨迹和/或移动状态的操控模式。其中，本实施例涉及的移动平台可以是诸如无人机、汽车等具有一定处理能力，且可通过操控设备进行操控的设备。下述实施例主要以无人机为例进行说明，其他移动平台的实现过程类似，不再赘述。
在步骤S401中，用户的操控指令可以通过控制移动平台的控制杆量来触发，控制杆量可以在遥控设备处生成。以无人机为例，用户可手动控制无人机的飞行，通过推拉遥控设备的摇杆生成控制杆量，控制杆量对应用户的期望轨迹。例如，当用户想要控制无人机飞直线时，可推拉遥控设备的遥杆(例如pitch杆)，从而生成期望的直线轨迹对应的控制杆量。用户想要控制无人机转弯时，也可以推拉遥控设备的摇杆(例如pitch杆和yaw杆)，从而生成期望的曲线轨迹对应的控制杆量。
无人机接收到遥控设备发送的控制杆量后，会根据该控制杆量生成操控指令，但不会直接根据生成的该操控指令控制无人机，而是根据控制杆量进行轨迹预测。将生成的该操控指令作为初始操控指令序列，并基于初始操控指令序列叠加避障辅助指令，预测生成无人机可绕开障碍物飞行的移动轨迹，进而根据该移动轨迹生成控制指令，根据该控制指令对无人机进行控制。具体来说：
首先，根据控制杆量生成初始操控指令序列。
假设在当前时刻接收到遥控设备生成的控制杆量，本实施例可根据所述控制杆量得到一个指令序列，并将该指令序列作为初始操控指令序列。
当该控制杆量为直线轨迹对应的控制杆量时，本实施例的初始操控指令序列可以包括：初始线速度指令序列和初始线加速度指令序列。在一个示例中，可将控制杆量映射为线速度指令。并根据上一个时刻滤波后的线速度指令对当前时刻的线速度指令进行滤波，并根据匀加速模型得到初始线速度指令序列v_cmd k,k＝1…N，并根据空气阻力模型，查找与初始线速度指令序列中的线速度相对应的线加速度，从而得到初始线加速度指令序列 a_cmd k,k＝1…N，其中，k代表当前时刻之后的各个指令时刻。本实施例不对初始线速度指令序列和初始线加速度指令序列中的指令数量(即N的数值)、以及相邻指令之间的时间间隔作限定，其可根据实际情况和控制效果进行设置。可以理解的是，初始操控指令序列可以是无人机根据用户当前输入的操控指令预测出的未来一段时间T内的操控指令。
在其它示例中，也可以将控制杆量映射为线加速度指令，查找与初始线加速度指令序列中的线加速度相对应的线速度，从而得到初始线速度指令序列；或者，也可以将控制杆量同时映射为线速度指令和线加速度指令。通过上述方式可以生成平滑连的初始操控指令，使得根据初始操控指令确定出的预测轨迹平滑连续，可保证对无人机飞行轨迹的控制效果。
生成初始操控指令序列后，在初始操控指令序列的基础上叠加避障辅助指令，进行轨迹预测，得到无人机的预测轨迹。其中，在确定预测轨迹的过程中，首先获取无人机的运动学模型，再利用运动学模型对叠加了避障辅助指令后的初始操控指令序列进行轨迹预测，得到预测轨迹，预测轨迹中的预测轨迹点通过预测位置参数进行表征。预测位置参数包括：预测位置；或者，预测位置和预测线速度；或者，预测位置、预测线速度和预测线加速度。其中，避障辅助指令包括很多个线加速度指令，在不同时刻选取的线加速度指令可以相同或不同，每个时刻可使用的避障辅助指令也可以是一个或多个。
示例性的，可以是初始线加速度指令序列的基础上，叠加避障辅助线加速度指令来优化或修正初始线加速度指令，从而避免移动平台与障碍物发生碰撞。即在初始线加速度指令序列a_cmd k的基础上叠加了辅助避障线加速度指令a_det k后可以得到干预或者修正后的线加速度指令序列a_cmd' k。其中，还基于以下至少一个因素对初始线加速度指令修正：无人机飞行控制器的线加速度控制量、无人机的向心加速度、无人机受到的空气阻力。在本实施例中，利用以下的至少一个修正量对初始线加速度指令序列进行修正：无人机的飞行控制器的线加速度控制量acc_ctrl、无人机的向心加速度acc_cent、以及无人机受到的空气阻力acc_air。例如，当考虑以上三个因素对初始线加速度指令序列进行修正时，修正后的初始线加速度指令序列a_cmd' k＝a_cmd k+a_det k+acc_ctrl-acc_cent-acc_air 。
以下以匀加速模型为例进行说明，但本实施例并不限于此，还可以采用其他任何类型的运动学模型，例如但不限于：匀速模型、非线性模型等。
匀加速模型如下所示：
p k+1＝p k+v k·Δt+0.5·a k·Δt 2        (1)
v k+1＝v k+a k·Δt            (2)
其中，Δt表示相邻时刻间的时间间隔；p k表示k时刻的位置，v k表示k时刻的线速度，a k表示k时刻的线加速度，v k+1表示k+1时刻的线速度，a k+1表示k+1时刻的线加速度。
本实施例中，首先获取无人机在当前时刻的位置参数，再将当前时刻的位置参数、初始线速度指令和初始线加速度指令输入公式(1)和(2)，即可得到预测轨迹的预测轨迹点的预测位置参数。具体来说，可通过无人机的传感器得到无人机在当前时刻的位置p 0、线速度v 0和线加速度a 0。然后将位置p 0、线速度v 0和线加速度a 0作为公式(1)和(2)的初始值，并将初始值以及初始线速度指令v_cmd k中的各个线速度指令、修正后的线加速度指令序列a_cmd' k中的各个线加速度指令代入公式(1)和(2)迭代运算，即可得到预测轨迹的各个预测轨迹点P 1,P 2,...,P n，预测轨迹点的预测位置参数包括预测位置和预测线速度两个参数。
当该控制杆量为曲线轨迹对应的控制杆量时，本实施例的初始操控指令序列除了包括：初始线速度指令序列和初始线加速度指令序列以外，还包括初始偏航角速度指令序列和/或初始偏航角加速度指令序列。在确定预测轨迹的过程中，首先获取无人机的运动学模型，再利用运动学模型对叠加了避障辅助指令的初始操控指令序列进行轨迹预测，得到预测轨迹，预测轨迹中的预测轨迹点通过预测位置参数和预测姿态参数进行表征。其中，预测姿态参数包括：预测偏航角和预测偏航角速度。
生成初始操控指令序列后，根据初始操控指令序列确定无人机的预测轨迹。与上一实施例类似，在确定预测轨迹的过程中，首先获取无人机的运动学模型，再利用运动学模型对叠加了避障辅助指令后的初始操控指令序列进行轨迹预测，得到移动平台可绕开障碍物的预测轨迹，预测轨迹中的预测轨迹点通过预测位置参数和预测姿态参数进行表征。其中，预测姿态参数包括：预测偏航角和预测偏航角速度。
对于匀加速模型，匀加速模型如下所示：
p k+1＝p k+v k·Δt+0.5·a k·Δt 2         (3)
v k+1＝v k+a k·Δt             (4)
Figure PCTCN2020086658-appb-000001
Figure PCTCN2020086658-appb-000002
其中，公式(3)和(4)与上一实施例的公式(1)和(2)相同，用于对预测轨迹的位置参数进行预测。公式(5)和(6)用于对预测轨迹的姿态参数进行预测；
Figure PCTCN2020086658-appb-000003
表示k 时刻的偏航角速度，
Figure PCTCN2020086658-appb-000004
表示k+1时刻的航线角速度，ψ k表示k时刻的偏航角，ψ k+1表示k+1时刻的偏航角，
Figure PCTCN2020086658-appb-000005
表示k时刻的偏航角加速度。
本实施例中，可将预测轨迹的位置参数与姿态参数解耦，分别对位置参数和姿态参数进行预测。
位置参数的预测过程与上一实施例类似，具体可参见上一实施例的描述，简单来说，将当前时刻的位置参数、初始线速度指令序列、叠加了避障辅助线加速度指令的初始线加速度指令序列代入公式(3)和(4)迭代运算，即可得到预测轨迹点的预测位置和预测线速度。
姿态参数的预测过程如下：首先获取无人机在当前时刻的姿态参数；将当前时刻的姿态参数、初始偏航角速度指令序列、初始偏航角加速度指令序列输入匀加速模型的公式(5)和(6)，得到预测轨迹中的各个预测轨迹点的预测姿态参数，预测姿态参数包括：预测轨迹点的预测偏航角和预测偏航角速度。
具体来说，可通过无人机的IMU得到无人机在当前时刻的偏航角ψ 0、偏航角速度
Figure PCTCN2020086658-appb-000006
和偏航角加速度
Figure PCTCN2020086658-appb-000007
将偏航角ψ 0、偏航角速度
Figure PCTCN2020086658-appb-000008
和偏航角加速度
Figure PCTCN2020086658-appb-000009
作为公式(5)和(6)的初始值，并将初始值以及初始偏航角速度指令序列
Figure PCTCN2020086658-appb-000010
初始偏航角加速度指令序列代入公式(5)和(6)迭代运算，即可得到预测轨迹的各个预测轨迹点的预测偏航角和预测偏航角速度。
通过上述位置参数和姿态参数的预测，得到根据初始位置指令和初始姿态指令预测的预测轨迹P 1,P 2,...,P N，预测轨迹点的预测位置、预测线速度、预测线加速度、预测偏航角、预测偏航角速度、预测偏航角加速度等几个参数即可获得。
通过上述方式可以生成平滑连续的初始位置指令和初始姿态指令，使得根据初始位置指令和初始姿态指令确定出的预测轨迹平滑连续，可保证对无人机飞行轨迹的控制效果。
具体的，在本申请实施例中，利用一小段的运动基元，采用单步轨迹的方式来迭代生成完整的移动轨迹。例如，基于无人机当前的位置、线速度、线加速度、偏航角速度，遍历避障辅助指令集(包括线加速度指令集)，结合预测的指令，即包括线速度、线加速度、偏航角速度、偏航角加速度，代入运动模型执行一个固定步长，生成单步轨迹(即运动基元，单步轨迹的轨迹点包括位置、线速度、线加速度、偏航角度这几个维度的参数)，经过碰撞检测后只保留与障碍物未碰撞的轨迹，舍弃与障碍物碰撞的轨迹，在保留下来的单步轨迹的基础上继续遍历避障辅助指令集，代入运动模型执行一个固定步长，生成单步轨迹，循环执行上述步骤直至满足预设结束条件或超时。
针对直线运动为例，利用一小段的运动基元，采用单步轨迹的方式来迭代生成上述移动轨迹。例如，遍历避障辅助指令集(包括线加速度指令集)，结合预测的指令，即包括线速度、线加速度，代入运动模型执行一个固定步长，生成单步轨迹(即运动基元，轨迹上的轨迹点包括位置、线速度、线加速度这几个维度的参数)，经过碰撞检测后只保留与障碍物未碰撞的轨迹，舍弃与障碍物碰撞的轨迹，在保留下来的单步轨迹的基础上继续遍历避障辅助指令集，代入运动模型执行一个固定步长，生成单步轨迹，循环执行上述步骤直至满足预设结束条件或超时。
其中，生成一条完整的移动轨迹需要很多次的迭代，生成的每一小步单步轨迹，都需要进行碰撞检测，通过上述方式得到多条移动轨迹，在所有可行轨迹中搜索出既符合用户意图又符合避障条件的目标移动轨迹，控制无人机按照该目标移动轨迹进行移动，若超时仍未生成符合避障条件的移动轨迹，则控制无人机减速悬停或者返航。
其中，预设结束条件包括生成所述多条移动轨迹的总时长达到第一预设时长或生成的所述多条移动轨迹的总数量达到预设数量。也即是说，本申请中，提前设定一个时间阈值或者数量阈值，在该时间阈值或数量阈值内，无人机可以按照前述单步轨迹迭代方式持续不断的生成不同的满足避障条件的移动轨迹，当该时间阈值或数量阈值到达时，才会停止生成移动轨迹。如此可以迭代出数量较多的满足避障条件的移动轨迹，绕行轨迹的可选择范围则更大，在障碍物错综复杂的场景中，可选择出绕行轨迹的可能性则更大。
其中，满足避障条件可以包括移动平台在移动轨迹上移动时与障碍物的距离大于等于设定距离阈值。
其中，避障辅助指令集中包含多个不同的线加速度指令，每个线加速度指令的加速度值、加速度作用方向或者加速度作用的时长不完全相同。可选的，该多个线加速度指令中还可以包括加速度为0的指令。可选的，线加速度的作用方向可以根据当前无人机与障碍物的相对位置关系确定，线加速度的作用方向可以不同于无人机朝向障碍物的方向，例如，可以垂直于无人机朝向障碍物的方向。
其中，移动平台可以预先设置每一条移动轨迹的时间长度(例如10秒钟)，或者可以设置每一条移动轨迹迭代的单步轨迹的数量(例如10次)，当单条移动轨迹满足该时间长度或数量时，该单条移动轨迹即生成完毕。
本申请实施例中，避障辅助指令是有作用时间长度的，例如1秒钟。基于用户的操控指令叠加避障辅助指令，在避障辅助指令的作用时间内可以预测生成一小段的单步轨迹(即运动基元)，该单步轨迹的时长则与避障辅助指令的作用时间相关，例如可以相同。 经过碰撞检测后只保留与障碍物未碰撞的轨迹，舍弃与障碍物会发生碰撞的轨迹。进而，在保留下来的单步轨迹的基础上(比如以单步轨迹的端点为起点)，进一步基于预测的操控指令叠加避障辅助指令继续生成一小段的单步轨迹，然后进行碰撞检测，保留与障碍物未碰撞的轨迹，舍弃与障碍物会发生碰撞的轨迹，如此过程循环执行，直至基于多个单步轨迹生成多条可绕开障碍物的移动轨迹。
可参见图5，是本申请实施例提供的曲线运动时迭代生成单步轨迹的流程示意图，如图5所示，根据遥控设备的遥控指令映射出未来一段时间内的线速度、线加速度、偏航角速度、偏航角加速度等信息，在无人机的当前状态下，遍历避障辅助指令集(线加速度a)，进行运动模型预测，预测一段未来的轨迹(包括位置p、线速度v、线加速度a、偏航角速度、偏航角加速度)。具体的，可以将上述映射的线速度、叠加了辅助指令的线加速度、偏航角速度、偏航角加速度以及无人机的当前状态代入运动模型执行一个固定步长，生成单步轨迹，经过碰撞检测后只保留与障碍物未碰撞的轨迹(p、v、a、偏航角速度)，并记录该步轨迹的代价以及上一段轨迹的索引，循环此步骤直至满足预设结束条件或超时，若是前者则会得到可行轨迹集，利用图搜索算法在所有可行轨迹构成的图中搜索出代价最小的最优轨迹，给到轨迹跟踪控制器控制执行，若是后者则规划减速悬停轨迹进入悬停状态。
其中，代价可以根据轨迹在生成过程中叠加的避障辅助指令的大小、轨迹的变化大小、轨迹可移动距离或者无人机消耗的能量来确定。其中，轨迹在生成过程中叠加的避障辅助指令越大，或者叠加次数越多，或者叠加的避障辅助指令的作用力越大，则代表对用户的意图指令干预的越多，相反的，叠加的避障辅助指令越小、次数越少或者作用力越小，则代表对用户的意图指令干预的越少。轨迹的变化越剧烈，则代表相邻两段单步轨迹之间差别越大，轨迹的变化越平缓，则代表相邻两段单步轨迹之间的差别越小。可移动距离可以理解为在移动平台发生碰撞前的可移动的距离。消耗的能量可以表示移动平台在移动轨迹上移动时所需消耗的能量。示例的，当叠加的避障辅助指令越大、叠加次数越多、叠加的避障辅助指令的作用力越大、轨迹的变化越剧烈、可移动距离越小或者移动平台消耗的能量越大，则对应的代价越高，相反，当叠加的避障辅助指令越小、叠加次数越少、叠加的避障辅助指令的作用力越小、轨迹的变化越平缓、可移动距离越大或者移动平台消耗的能量越小，则对应的代价越低。可从多条移动轨迹中选择代价最低的轨迹作为目标移动轨迹。
需要说明的是，由于单步轨迹在迭代生成的过程中已经进行了碰撞检测，因此最终得到的若干条移动轨迹均是移动平台不会撞到障碍物的轨迹，在这些不会撞到障碍物的轨迹中选择避障干预最低的或者变化越平缓的或者可移动距离越大的或者消耗能量越低的，可 以实现既匹配用户意图且移动平滑的效果。
在其他实现方式中，在迭代生成单步轨迹的过程中，也可以不进行障碍物碰撞检测，即生成单步轨迹后，继续基于避障辅助指令迭代生成其他单步轨迹，直至生成完整的移动轨迹。而在得到多条移动轨迹后，在最终选择目标移动轨迹时，将与障碍物之间的距离也作为衡量代价的一个维度，当移动轨迹与障碍物之间的距离越大时，对应的代价越高，当移动轨迹与障碍物之间的距离越小时，对应的代价越低。
图6是本申请实施例提供的单步轨迹的示意图，如图6所示，单步轨迹的生成过程与树形图类似，树根可以理解为初始预测轨迹，在初始操控指令的基础上叠加不同的避障辅助指令，生成多条树枝，每小段树枝即对应一段单步轨迹，每小段树枝的长度等于避障辅助指令的作用时长，生成的多条树枝中保留不会与障碍物碰撞的树枝，舍弃与障碍物会碰撞的树枝，在保留的树枝的端部，继续叠加不同的避障辅助指令，从而生成更多条满足避障条件的树枝。如此可以生成密密麻麻的网来实现绕行，可查找的空间更大，选择的自由度和余地更大。当树枝的长度达到预定长度时，则该条树枝停止生长，当全部的树枝数量达到预定数量或者到达指定时长时，停止生长树枝。最终从该多条树枝在选择出最优的一条来控制无人机飞行。
采用单步轨迹的方式来迭代生成移动平台的移动轨迹，可以无需事先规划辅助指令中用于干预预测轨迹的速度的变化形状及数量，可以基于实际环境来自由生成更多条满足避障条件的绕行轨迹，搜索的密度更大粒度更小，更加自由，可搜索到绕行轨迹的几率更高。并且通过设置用于避障辅助的线加速度集合，可以使得单步轨迹之间移动平台的速度变化更平滑，避免速度突变造成移动平台移动变化剧烈的情况。相较于事先设定好干预轨迹的速度的变化形状以及设定移动轨迹的数量的方案，本申请实施例的自由度更大，可选择的范围更多，可搜索到绕行轨迹的概率更高。并且，相较于干预轨迹的速度，干预轨迹的加速度方案可实现移动平台的移动更加平滑稳定。
参见图7a，是本申请实施例提供的一种基于原始指令预测的轨迹与在避障辅助指令作用下的辅助绕行轨迹的示意图。图7a中以直线运动为例，可看出如果在用户打杆的指令作用下，移动平台会与障碍物发生碰撞，基于用户打杆的指令，在避障辅助指令的干预下，移动平台可以绕开障碍物移动，从而既符合用户的意图，即向前移动，同时也可以躲开障碍物，避免发生碰撞。图7a中，绕行轨迹是基于单步轨迹迭代生成的整条绕行轨迹，图7b是图7a中涉及的避障辅助指令的示意图，图7b中，每段单步轨迹上所叠加的线加速度的大小不完全相同。可在生成单步轨迹的过程中从线加速度指令集中随机挑选或者遍历线 加速度指令，生成单步轨迹，只保留不会与障碍物发生碰撞的单步轨迹，在单步轨迹的端部继续迭代生成单步轨迹，直至生成多条完整移动轨迹。
基于前述方法生成了多条移动轨迹后，所述根据所述移动轨迹控制所述移动平台移动，包括：从所述多条移动轨迹中选取目标移动轨迹；根据所述目标移动轨迹控制所述移动平台移动。示例性的，所述目标移动轨迹是根据如下选取策略的一种或多种方式选取的：所述一条或多条移动轨迹中增加的避障辅助指令最小的；所述一条或多条移动轨迹中变化最平缓的；用户从所述一条或多条移动轨迹中选择的；所述一条或多条移动轨迹中可移动距离最大的；所述一条或多条移动轨迹中消耗能量最小的；所述一条或多条移动轨迹中可移动距离大于第一预设阈值且消耗能量小于第二预设阈值的。
下面通过一个示例来说明如何从获得的移动轨迹中选择目标移动轨迹来控制移动平台移动：
示例的，从上述预测获得的一个或多个移动轨迹中筛选获得可移动距离最长的移动轨迹作为目标移动轨迹，或者从上述预测获得的一个或多个移动轨迹中筛选出移动平台消耗能量最小的移动轨迹作为目标移动轨迹。其中，消耗能量是指移动平台沿轨迹移动时执行各个控制指令所消耗的能量。在一种可能的方法中，在获得可绕开障碍物的一个或多个移动轨迹后，对该些移动轨迹进行显示，并在显示界面上提供移动轨迹的可选择操作。在用户选择出目标移动轨迹后，将用户选择的目标移动轨迹作为目标移动轨迹。
采用上述方式确定了目标移动轨迹后，根据该目标移动轨迹生成控制指令，从而控制移动平台沿该目标移动轨迹移动。比如，可以根据目标移动轨迹中轨迹点的位置、线速度、线加速度、偏航角速度、偏航角加速度调整移动平台的旋翼的转速，来达到该目标移动轨迹上的线速度和角速度，实现沿该目标移动轨迹的移动。例如，当用户操作遥控设备的pitch杆和yaw杆，可控制无人机转弯，无人机调整旋翼的转速，可实现无人机改变roll角度进行转弯。
可选的，针对曲线运动的情况，在获得了移动平台的移动轨迹后，可以根据移动平台的移动轨迹调整该移动平台上的云台的姿态角，以实现云台与移动平台的变化相对平稳，达到固定翼移动的效果。示例性的，可以根据移动平台的偏航yaw角速度调整云台的翻滚roll角度。
以无人机为例，在无人机转弯时，无人机云台的roll角度设定值和无人机的yaw角速度相关，无人机转弯的时候，云台roll会倾斜一下，以达到云台上相机的拍摄画面随着无人机转弯而调整，用户通过观看无人机返回的拍摄画面即可体验到置身无人机内驾驶无人 机进行转弯的效果。
示例性的，当无人机向左转弯时，云台沿roll轴向左倾斜，即云台左低右高。当无人机向右转弯时，云台沿roll轴向右倾斜，即云台左高右低。
示例性的，无人机的yaw角速度与云台转roll的角度成正比例关系。例如，当无人机的yaw角速度越大，云台沿roll轴转动的角度越大，当无人机的yaw角速度越小，云台沿roll轴转动的角度越小。
当然本领域技术人员应该了解的是上述举例仅是为清楚所做的示例说明而不是对本发明的唯一限定。
本实施例，通过在用户操控模式中，基于用户输入的操控指令和避障辅助指令控制移动平台的移动轨迹，从而使得在用户操控模式中，也可以实现移动平台的主动避障，使得移动平台能够在用户输入的操控指令和避障辅助指令的共同作用下，绕开障碍物，或者在无法绕开障碍物时，能够多移动一段时间，而不是一遇到障碍物就执行刹车操作，提高了移动平台移动的安全性和用户体验。
示例的，图8是本申请实施例提供的一种移动装置的结构示意图，移动装置80包括处理器81和存储器82，处理器81用于基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹，进而迭代生成移动平台可绕开障碍物的移动轨迹，以及根据移动轨迹控制移动平台进行移动。可选的，移动装置80还可以包括检测设备，当检测设备检测到移动平台与障碍物之间的距离小于预定距离时，触发处理器81生成避障辅助指令，进而基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹。当移动平台与障碍物之间的距离小于预定距离时，选取一个或多个避障辅助指令。其中，不同的避障辅助指令对应的线加速度值、线加速度作用方向或作用时间长度中的一项或多项不同。
进一步的，在选择避障辅助指令时，处理器81的处理方法包括如下几种：
在一种可能的处理方法中，处理器81根据用户输入的操控指令确定是否生成避障辅助指令，比如，当处理器81判断用户输入的操控指令使得移动平台与障碍物之间存在碰撞危险时，则选择避障辅助指令，以通过避障辅助指令改变移动平台的移动轨迹。若判断用户输入的操控指令不会导致碰撞时，则不选择避障辅助指令。
在另一种可能的处理方法中，当检测设备检测到移动平台与障碍物之间的距离小于预定距离时，处理器81直接选择避障辅助指令，而不检测用户输入的操控指令会不会导致碰撞。
在另一种可能的处理中，处理器81默认自动生成避障辅助指令，即无论用户输入的 操控指令使得移动平台与障碍物之间是否存在碰撞危险，处理器81均自动基于避障辅助指令和用户输入的操控指令控制移动平台的移动。
本实施例提供的移动装置能够执行前述实施例提供的移动控制方法，其执行方式和有益效果类似，在这里不再赘述。
本申请实施例还提供一种移动平台，该移动平台包括：
机身；
动力系统，安装在所述机身，用于为所述移动平台提供动力；
以及上述实施例提供的移动装置。
可选地，该移动平台还可以包括传感器，安装在所述机身，用于探测获得所述移动平台所处环境的地图信息。
可选地，所述传感器包括视觉传感器和/或距离传感器。
可选地，所述移动平台还包括：
通信设备，安装在所述机身，用于与地面站进行信息交互。
可选地，所述移动平台至少包括如下的一种：无人机、汽车。
本实施例提供的移动平台其执行方式和有益效果与前述实施例提供的移动装置类似，在这里不再赘述。
此外，本实施例还提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行以实现上述实施例所述的移动平台的控制方法。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元，可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器(processor)执行本申请各个实施例所述方法的部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序指令的介质。
本领域技术人员可以清楚地了解到，为描述的方便和简洁，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将装置的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。上述描述的装置的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
最后应说明的是：以上各实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述各实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (40)

  1. 一种移动控制方法，其特征在于，包括：
    在用户操控模式中，基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹，所述预定步长与所述避障辅助指令的作用时间长度相关；
    若所述单步轨迹满足预设避障条件，则继续基于所述避障辅助指令以及所述单步轨迹生成预定步长的单步轨迹；
    根据所述单步轨迹生成移动平台可绕开障碍物的移动轨迹，并根据所述移动轨迹控制所述移动平台移动。
  2. 根据权利要求1所述的方法，其特征在于，所述避障辅助指令用于增加所述移动平台沿第一方向上的线加速度分量，其中，所述第一方向不同于所述移动平台朝向所述障碍物的方向。
  3. 根据权利要求2所述的方法，其特征在于，所述第一方向与所述移动平台朝向所述障碍物的方向垂直。
  4. 根据权利要求1至3任一项所述的方法，其特征在于，所述基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹，包括：
    利用运动学模型对叠加了所述避障辅助指令的所述操控指令进行轨迹预测，生成所述预定步长的单步轨迹。
  5. 根据权利要求1至4任一项所述的方法，其特征在于，所述移动轨迹包括多条，所述多条移动轨迹是在满足预设结束条件之前生成的。
  6. 根据权利要求5所述的方法，其特征在于，所述预设结束条件包括生成所述移动轨迹的总时长达到第一预设时长或生成的所述移动轨迹的总数量达到预设数量。
  7. 根据权利要求1至6任一项所述的方法，其特征在于，每条所述移动轨迹的时长满足第二预设时长或迭代次数满足预设次数。
  8. 根据权利要求1至7任一项所述的方法，其特征在于，所述预设避障条件包括所述移动平台与所述障碍物的距离大于等于预设距离。
  9. 根据权利要求1至8任一项所述的方法，其特征在于，所述避障辅助指令包括一个或多个，不同的避障辅助指令对应的加速度值、加速度作用方向或作用时间长度中的一项或多项不同。
  10. 根据权利要求1至9任一项所述的方法，其特征在于，所述移动轨迹包括多条， 所述根据所述移动轨迹控制所述移动平台移动，包括：
    从所述多条移动轨迹中选取目标移动轨迹；
    根据所述目标移动轨迹控制所述移动平台移动。
  11. 根据权利要求10所述的方法，其特征在于，所述目标移动轨迹是根据如下选取策略的一种或多种方式选取的：
    所述一条或多条移动轨迹中叠加的避障辅助指令最小的；
    所述一条或多条移动轨迹中移动状态变化最小的；
    所述一条或多条移动轨迹中消耗能量最小的；
    所述一条或多条移动轨迹中可移动距离最大的；
    所述一条或多条移动轨迹中可移动距离大于第一预设阈值且消耗能量小于第二预设阈值的；
    用户从所述一条或多条移动轨迹中选择的。
  12. 根据权利要求1至11任一项所述的方法，其特征在于，所述操控指令包括线速度指令和线加速度指令。
  13. 根据权利要求1至11任一项所述的方法，其特征在于，所述操控指令包括线速度指令、线加速度指令和偏航角速度指令。
  14. 根据权利要求12所述的方法，其特征在于，所述移动轨迹包括轨迹点的位置、线速度或线加速度中的一种或多种。
  15. 根据权利要求13所述的方法，其特征在于，所述移动轨迹包括轨迹点的位置、线速度、线加速度或偏航角速度中的一种或多种。
  16. 根据权利要求1至15任一项所述的方法，其特征在于，还包括：
    根据所述移动轨迹调整所述移动平台上的云台的姿态角。
  17. 根据权利要求16所述的方法，其特征在于，所述根据所述移动轨迹调整所述移动平台上的云台的姿态角，包括：
    根据所述移动轨迹中的偏航角速度调整所述云台的翻滚角度。
  18. 一种移动装置，其特征在于，包括存储器和处理器；
    所述存储器用于存储程序指令；
    所述处理器，调用所述程序指令，当程序指令被执行时，用于执行以下操作：
    在用户操控模式中，基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹，所述预定步长与所述避障辅助指令的作用时间长度相关；
    若所述单步轨迹满足预设避障条件，则继续基于所述避障辅助指令以及所述单步轨迹生成预定步长的单步轨迹；
    根据所述单步轨迹生成移动平台可绕开障碍物的移动轨迹，并根据所述移动轨迹控制所述移动平台移动。
  19. 根据权利要求18所述的装置，其特征在于，所述避障辅助指令用于增加所述移动平台沿第一方向上的线加速度分量，其中，所述第一方向不同于所述移动平台朝向所述障碍物的方向。
  20. 根据权利要求19所述的装置，其特征在于，所述第一方向与所述移动平台朝向所述障碍物的方向垂直。
  21. 根据权利要求18至20任一项所述的装置，其特征在于，所述基于用户的操控指令和避障辅助指令生成预定步长的单步轨迹，包括：
    利用运动学模型对叠加了所述避障辅助指令的所述操控指令进行轨迹预测，生成所述预定步长的单步轨迹。
  22. 根据权利要求18至21任一项所述的装置，其特征在于，所述移动轨迹包括多条，所述多条移动轨迹是在满足预设结束条件之前生成的。
  23. 根据权利要求22所述的装置，其特征在于，所述预设结束条件包括生成所述移动轨迹的总时长达到第一预设时长或生成的所述移动轨迹的总数量达到预设数量。
  24. 根据权利要求18至23任一项所述的装置，其特征在于，每条所述移动轨迹的时长满足第二预设时长或迭代次数满足预设次数。
  25. 根据权利要求18至24任一项所述的装置，其特征在于，所述预设避障条件包括所述移动平台与所述障碍物的距离大于等于预设距离。
  26. 根据权利要求18至25任一项所述的装置，其特征在于，所述避障辅助指令包括一个或多个，不同的避障辅助指令对应的加速度值、加速度作用方向或作用时间长度中的一项或多项不同。
  27. 根据权利要求18至26任一项所述的装置，其特征在于，所述移动轨迹包括多条，所述根据所述移动轨迹控制所述移动平台移动，包括：
    从所述多条移动轨迹中选取目标移动轨迹；
    根据所述目标移动轨迹控制所述移动平台移动。
  28. 根据权利要求27所述的装置，其特征在于，所述目标移动轨迹是根据如下选取策略的一种或多种方式选取的：
    所述一条或多条移动轨迹中叠加的避障辅助指令最小的；
    所述一条或多条移动轨迹中移动状态变化最小的；
    所述一条或多条移动轨迹中消耗能量最小的；
    所述一条或多条移动轨迹中可移动距离最大的；
    所述一条或多条移动轨迹中可移动距离大于第一预设阈值且消耗能量小于第二预设阈值的；
    用户从所述一条或多条移动轨迹中选择的。
  29. 根据权利要求18至28任一项所述的装置，其特征在于，所述操控指令包括线速度指令和线加速度指令。
  30. 根据权利要求18至28任一项所述的装置，其特征在于，所述操控指令包括线速度指令、线加速度指令和偏航角速度指令。
  31. 根据权利要求29所述的装置，其特征在于，所述移动轨迹包括轨迹点的位置、线速度或线加速度中的一种或多种。
  32. 根据权利要求30所述的装置，其特征在于，所述移动轨迹包括轨迹点的位置、线速度、线加速度或偏航角速度中的一种或多种。
  33. 根据权利要求18至32任一项所述的装置，其特征在于，所述处理器还用于：
    根据所述移动轨迹调整所述移动平台上的云台的姿态角。
  34. 根据权利要求33所述的装置，其特征在于，所述根据所述移动轨迹调整所述移动平台上的云台的姿态角，包括：
    根据所述移动轨迹中的偏航角速度调整所述云台的翻滚角度。
  35. 一种移动平台，其特征在于，包括：
    机身；
    动力系统，安装在所述机身，用于为所述移动平台提供动力；
    以及如权利要求18-34中任一项所述的移动装置。
  36. 根据权利要求35所述的移动平台，其特征在于，所述移动平台还包括：
    传感器，安装在所述机身，用于探测获得所述移动平台所处环境的地图信息。
  37. 根据权利要求36所述的移动平台，其特征在于，所述传感器包括视觉传感器和/或距离传感器。
  38. 根据权利要求35至37任一项所述的移动平台，其特征在于，所述移动平台还包括：
    通信设备，安装在所述机身，用于与地面站进行信息交互。
  39. 根据权利要求35至38任一项所述的移动平台，其特征在于，所述移动平台至少包括如下的一种：无人机、汽车。
  40. 一种计算机可读存储介质，其特征在于，其上存储有计算机程序，所述计算机程序被处理器执行以实现权利要求1至17任一项所述的方法。
PCT/CN2020/086658 2020-04-24 2020-04-24 移动控制方法、移动装置及移动平台 WO2021212462A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2020/086658 WO2021212462A1 (zh) 2020-04-24 2020-04-24 移动控制方法、移动装置及移动平台
CN202080004614.0A CN112639656A (zh) 2020-04-24 2020-04-24 移动控制方法、移动装置及移动平台

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/086658 WO2021212462A1 (zh) 2020-04-24 2020-04-24 移动控制方法、移动装置及移动平台

Publications (1)

Publication Number Publication Date
WO2021212462A1 true WO2021212462A1 (zh) 2021-10-28

Family

ID=75291164

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/086658 WO2021212462A1 (zh) 2020-04-24 2020-04-24 移动控制方法、移动装置及移动平台

Country Status (2)

Country Link
CN (1) CN112639656A (zh)
WO (1) WO2021212462A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112162546A (zh) * 2020-06-24 2021-01-01 上汽通用五菱汽车股份有限公司 遥控车辆的方法、系统和存储介质
CN114035608A (zh) * 2021-11-10 2022-02-11 广东电网能源发展有限公司 基于无人机的巡检培训方法、系统、培训终端和存储介质
CN114333429A (zh) * 2021-12-21 2022-04-12 中国电子科技集团公司第五十四研究所 一种面向多无人机目标覆盖任务的规则提取方法
CN116839591A (zh) * 2023-07-12 2023-10-03 哈尔滨天枢问道技术有限公司 一种轨迹跟踪定位滤波系统及救援无人机的融合导航方法

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114595517B (zh) * 2022-03-07 2024-09-20 南京航空航天大学 一种双发直升机回避区曲线计算方法
CN115291607B (zh) * 2022-08-02 2023-03-14 柳州城市职业学院 复杂未知水域内无人驾驶船路径自主规划生成方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105955298A (zh) * 2016-06-03 2016-09-21 腾讯科技（深圳）有限公司 一种飞行器的自动避障方法及装置
WO2019144298A1 (zh) * 2018-01-23 2019-08-01 深圳市大疆创新科技有限公司 辅助移动方法、移动装置及可移动平台
CN110597271A (zh) * 2019-10-12 2019-12-20 河北工业大学 一种基于分级速度障碍算法的移动机器人避障方法
CN110709792A (zh) * 2018-07-23 2020-01-17 深圳市大疆创新科技有限公司 移动平台的辅助移动方法、移动装置及移动平台

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106289290A (zh) * 2016-07-21 2017-01-04 触景无限科技（北京）有限公司 一种路径导航系统及方法
CN106131790B (zh) * 2016-08-18 2021-01-15 联想(北京)有限公司 移动装置的控制方法和移动装置
CN110703763A (zh) * 2019-11-05 2020-01-17 武汉理工大学 无人车路径跟踪及避障方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105955298A (zh) * 2016-06-03 2016-09-21 腾讯科技（深圳）有限公司 一种飞行器的自动避障方法及装置
WO2019144298A1 (zh) * 2018-01-23 2019-08-01 深圳市大疆创新科技有限公司 辅助移动方法、移动装置及可移动平台
CN110709792A (zh) * 2018-07-23 2020-01-17 深圳市大疆创新科技有限公司 移动平台的辅助移动方法、移动装置及移动平台
CN110597271A (zh) * 2019-10-12 2019-12-20 河北工业大学 一种基于分级速度障碍算法的移动机器人避障方法

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112162546A (zh) * 2020-06-24 2021-01-01 上汽通用五菱汽车股份有限公司 遥控车辆的方法、系统和存储介质
CN114035608A (zh) * 2021-11-10 2022-02-11 广东电网能源发展有限公司 基于无人机的巡检培训方法、系统、培训终端和存储介质
CN114333429A (zh) * 2021-12-21 2022-04-12 中国电子科技集团公司第五十四研究所 一种面向多无人机目标覆盖任务的规则提取方法
CN116839591A (zh) * 2023-07-12 2023-10-03 哈尔滨天枢问道技术有限公司 一种轨迹跟踪定位滤波系统及救援无人机的融合导航方法
CN116839591B (zh) * 2023-07-12 2024-05-28 哈尔滨天枢问道技术有限公司 一种轨迹跟踪定位滤波系统及救援无人机的融合导航方法

Also Published As

Publication number Publication date
CN112639656A (zh) 2021-04-09

Similar Documents

Publication Publication Date Title
WO2021212462A1 (zh) 移动控制方法、移动装置及移动平台
US11914370B2 (en) System and method for providing easy-to-use release and auto-positioning for drone applications
US11726498B2 (en) Aerial vehicle touchdown detection
CN110687902B (zh) 用于免控制器式用户无人机交互的系统和方法
US10241520B2 (en) System and method for vision-based flight self-stabilization by deep gated recurrent Q-networks
CN110825113B (zh) 一种适用于四旋翼无人机集群飞行的队形保持方法
WO2018098678A1 (zh) 飞行器的控制方法、装置和设备以及飞行器
CN107003678B (zh) 控制方法、装置、设备及可移动平台
KR101813697B1 (ko) 딥 러닝을 이용한 무인기 비행 제어 시스템 및 방법
EP3771956B1 (en) Systems and methods for generating flight paths for navigating an aircraft
CN107643758A (zh) 拍摄移动图像的包括无人机和地面站的自主系统及方法
WO2020019110A1 (zh) 移动平台的辅助移动方法、移动装置及移动平台
CN110471442A (zh) 一种目标观察方法、相关设备及系统
WO2021189214A1 (zh) 可移动平台的控制方法、装置、可移动平台及存储介质
CN113795805A (zh) 无人机的飞行控制方法和无人机
US20240076066A1 (en) Fixed-wing unmanned aerial vehicle capable of high angle-of-attack maneuvering
CN115525063A (zh) 用于无人直升机的轨迹规划方法
JP6269735B2 (ja) 飛行装置、方法、及びプログラム
WO2021238743A1 (zh) 一种无人机飞行控制方法、装置及无人机
WO2019144298A1 (zh) 辅助移动方法、移动装置及可移动平台
KR101887314B1 (ko) 무인 항공기의 원격 제어 장치 및 방법과, 무인 항공기에 부착되는 움직임 제어 장치
WO2022047709A1 (zh) 更新限制区域数据的方法、装置、可移动平台和计算机存储介质
WO2021128184A1 (zh) 可移动载体的控制方法、控制装置、计算机可读存储介质
WO2021189215A1 (zh) 可移动平台的控制方法、装置、可移动平台及存储介质
Choi et al. Wearable gesture control of agile micro quadrotors

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20931725

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20931725

Country of ref document: EP

Kind code of ref document: A1