An eye movement-based virtual reality unmanned aerial vehicle flight control device and method
By generating 3D point cloud data and solving user eye and head movement data, and mapping it to drone control commands, the problems of drone operation complexity and lack of immersion are solved, achieving natural and precise drone control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING FANMEILI ROBOT TECH CO LTD
- Filing Date
- 2026-02-05
- Publication Date
- 2026-07-10
AI Technical Summary
Existing drone control methods are complex and have a high learning curve. Traditional display devices cannot provide an immersive flight experience, and eye-tracking control solutions cannot achieve precise target designation and environmental depth perception, resulting in low control accuracy and unnatural human-computer interaction.
By acquiring real-time depth video data of the drone environment to generate 3D point cloud data, solving user eye and head movement data, mapping 2D gaze points to 3D coordinates, selecting the nearest point cloud point as the flight target, calculating action and attitude control commands, and using a PID control model that coordinates eye and head movements to drive the drone flight.
It achieves a natural and intuitive immersive interactive experience, improves the accuracy and precision of target selection and control, simplifies the operation logic, conforms to human intuition, and enhances the user's sense of connection with the environment.
Smart Images

Figure CN121657710B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to an eye-tracking-based virtual reality UAV flight control device and method. Background Technology
[0002] With the maturation of drone technology, its applications in aerial photography, surveying, inspection, and emergency rescue are becoming increasingly widespread. Currently, the mainstream drone control methods rely on physical remote controllers or mobile device applications. Users manually operate joysticks, sliders, or touchscreens to control the drone's flight attitude, heading, and gimbal angle. However, these traditional control methods have significant drawbacks: First, they are complex to operate and have a high learning curve, requiring users to undergo extensive training to master them, especially for complex flight missions, resulting in a heavy mental burden; second, traditional display devices (such as mobile phone screens and remote controller screens) cannot provide an immersive first-person perspective flight experience, creating a strong sense of disconnect between the user and the flight environment, making it difficult to achieve a truly "human-machine integrated" control experience.
[0003] To improve user experience, several improvement solutions have emerged in existing technologies: one type of solution introduces virtual reality technology into drone control, allowing users to view the first-person perspective of the drone through a head-mounted display, which enhances the sense of presence, but its operation still relies on traditional controllers and does not fundamentally simplify the interaction logic; another type of solution explores control methods based on gestures, body sensing, or eye tracking, especially eye-tracking-based control solutions that allow users to interact with the interface through their gaze.
[0004] However, existing eye-tracking control solutions cannot enable users to accurately and directly specify flight target points in three-dimensional physical space. Furthermore, the system lacks a deep perception and understanding of the surrounding environment, and gaze commands cannot be associated with the environment, resulting in low control precision. It cannot fully utilize the most natural human visual attention mechanism, making human-computer interaction unnatural and unable to directly and naturally convert the user's true intentions into drone flight commands. Consequently, it cannot provide users with an intuitive, natural, and immersive interactive experience where what you see is what you control. Summary of the Invention
[0005] The purpose of this invention is to provide an eye-tracking-based virtual reality drone control device and method, thereby solving the technical problems of lack of immersion and unnatural interaction in existing technologies. To achieve the above objective, this invention discloses an eye-tracking-based virtual reality drone flight control method, comprising: S1: acquiring depth video data of the drone's environment in real time, and calculating the depth video data to generate three-dimensional point cloud data of the environment in real time; S2: acquiring user eye data and head movement data, and calculating the eye data to obtain the user's two-dimensional gaze point on the virtual reality display interface; S3: mapping the two-dimensional gaze point from the coordinate system of the virtual reality display interface to the coordinate system of the three-dimensional point cloud data to obtain the corresponding three-dimensional gaze coordinates; S4: searching and selecting the point cloud point closest to the three-dimensional gaze coordinates in the three-dimensional point cloud data as the drone's flight target point; S5: calculating the drone's current position based on the flight target point. With the flight target point Euclidean distance between Based on the Euclidean distance, the action speed control quantity is calculated. The speed control quantity The following nonlinear function is used to automatically attenuate the velocity at close range:
[0006]
[0007] in, This is the speed control proportional coefficient. The maximum distance threshold, An adjustment coefficient is used to automatically decay the speed at close range; to generate action control commands for controlling the UAV to move toward the three-dimensional flight target point, determine the head rotation angle based on the head motion data, and calculate attitude control commands for controlling the UAV's yaw motion and gimbal pitch motion; S6: send out the action control commands and the attitude control commands to drive the UAV to fly.
[0008] Furthermore, the specific steps in step S2 to calculate the two-dimensional gaze coordinates from the eye data are as follows: identifying the coordinates of the pupil center point in the user's eye image. and the coordinates of the corneal reflector ;
[0009]
[0010] , These are the abscissa and ordinate components of the pupil center point in the coordinate system to which the eye image belongs, respectively; and These represent the x-coordinate and y-coordinate components of the corneal reflection point in the coordinate system of the eye image, respectively. Indicates transpose processing; calculates the coordinates of the pupil center point. Coordinates of the corneal reflection point The pupil-corneal reflex vector
[0011]
[0012] The pupil-corneal reflection vector is mapped using a pre-defined eye feature mapping model. Convert to 3D line-of-sight vector The expression is:
[0013]
[0014] in, Indicates the tilt angle of the user's line of sight; Indicates the horizontal angle of the user's line of sight; Represents the linear mapping coefficients in the eye biometric calibration matrix; Represents the bias compensation term in eye biometric calibration; solves for the three-dimensional gaze direction vector. The intersection point with the virtual reality display interface is used to determine the two-dimensional gaze point.
[0015] Furthermore, the specific steps for generating action control commands in step S5 are as follows: based on the action speed control amount... and the direction vector components of the flight target point in the body coordinate system and The roll rate control quantity is calculated using a PID control model. and pitch rate control quantity :
[0016]
[0017] This is the proportionality coefficient. These are the differential coefficients. is the integral coefficient.
[0018] Furthermore, the attitude control commands in S5 include yaw rate control quantities. The specific generation steps are as follows: based on the horizontal rotation angle in the head rotation angle... The yaw rate control quantity is generated through the PID control model. ,
[0019]
[0020] in, Forward coefficients, This is the proportionality coefficient. These are the differential coefficients. is the integral coefficient.
[0021] Furthermore, the attitude control commands in S5 include gimbal pitch rate control parameters. The specific generation steps are as follows: based on the pitch angle in the head rotation angle... The gimbal pitch rate control quantity is generated through a PID control model. ,
[0022]
[0023] in, This is the proportionality coefficient. These are the differential coefficients. is the integral coefficient.
[0024] Furthermore, the method also includes: calculating the eye data to obtain the user's blink frequency; when the blink frequency meets a preset trigger condition, activating the flight mode selection interface; selecting the flight mode based on the user's eye data; and outputting control commands in conjunction with the current drone flight status.
[0025] Furthermore, when it is detected that the duration of a user's gaze at the flight target point exceeds a preset locking threshold... When the target point is locked, the drone is allowed to fly around the target point.
[0026] Furthermore, the drone's attitude angle, angular velocity, and position deviation are monitored in real time, and protection mechanisms are triggered when anomalies are detected, including: hovering in place, slow landing, and returning to home.
[0027] This invention also discloses an eye-tracking-based virtual reality drone flight control device, comprising:
[0028] The eye-tracking and head-movement hybrid perception processing module includes:
[0029] The eye data acquisition unit is used to acquire eye data.
[0030] The head data acquisition unit is used to acquire head movement data.
[0031] The processing terminal is used to process eye-tracking data;
[0032] Virtual reality perception and mixed display control module, including:
[0033] The hybrid head-mounted display unit is used to acquire and display the first-person view of the drone, drone flight status information, user gaze point, and target lock;
[0034] The central processing unit, equipped with a large-scale situational awareness model and a PID control algorithm, receives and processes the output data from the eye-tracking and head-motion hybrid perception processing module and the hybrid head-mounted display unit to calculate the drone's current position. With the flight target point Euclidean distance between Based on the Euclidean distance, the action speed control quantity is calculated. The speed control quantity The following nonlinear function is used to automatically attenuate the velocity at close range:
[0035]
[0036] in, This is the speed control proportional coefficient. The maximum distance threshold, An adjustment coefficient is used to automatically reduce speed at close range; and action control commands and attitude control commands are generated.
[0037] The UAV control execution module is used to receive and execute action control commands and attitude control commands, acquire UAV flight status information, and send it to the virtual reality perception and hybrid display and control module.
[0038] Furthermore, the eye data acquisition unit includes an infrared light source and a miniature infrared camera; the head data acquisition unit includes an IMU; the hybrid head-mounted display unit includes VR glasses, a drone camera, a drone gimbal, and a 5G image transmission module mounted on the drone frame; the drone control and execution module includes a communication module, a flight controller, an integrated electronic speed controller, a battery, and motors, and the flight controller includes an IMU, a barometer, a magnetometer, and a GPS locator.
[0039] Compared with the prior art, the present invention has at least the following beneficial effects:
[0040] 1. The operator's eye data is processed and mapped into action control commands for the UAV in three-dimensional space, i.e., the flight target point; at the same time, the head movement data is processed and mapped into attitude control commands for the UAV, i.e., yaw motion and gimbal pitch. This avoids the error caused by the operator's intention to turn their head to observe, which leads to the wrong flight turning action in the existing technology. Moreover, the head-eye coordination but decoupling design is more in line with human intuition, realizing the control method of flying wherever you look, making human-computer interaction more natural.
[0041] 2. In existing technologies, the two-dimensional gaze point corresponds to an infinitely extending ray in three-dimensional space, making it impossible to determine the target depth. This invention searches for the nearest point between the two-dimensional gaze and the real-time constructed three-dimensional point cloud model, anchoring the user's visual commands to the three-dimensional physical environment model. This ensures that each gaze of the user can be calculated in the real environment model with unique and accurate three-dimensional physical coordinates, solving the problem of target positioning ambiguity and improving the accuracy of target selection and the precision of subsequent control. Attached Figure Description
[0042] Figure 1 This is a schematic diagram of the process of the present invention.
[0043] Figure 2 This is a schematic diagram illustrating the device module architecture and working principle of the present invention.
[0044] Figure 3 This is a schematic diagram illustrating the data interaction and display control between the hybrid head-mounted display and the cloud / processor according to the present invention.
[0045] Figure 4 This is a schematic diagram illustrating the acquisition of eye data and calculation of gaze direction according to the present invention.
[0046] Figure 5 This is a schematic diagram of the eye-tracking-head-tracking control command mapping and target-locking fly-around method of the present invention. Detailed Implementation
[0047] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments:
[0048] S1: Acquire depth video data of the environment in which the UAV is located in real time, and calculate the depth video data to generate three-dimensional point cloud data of the environment in real time;
[0049] The drone acquires depth video data of its environment through its onboard depth camera. It then uses a context-aware big data model to perform real-time point cloud reconstruction and object recognition, generating 3D point cloud data of the environment. This provides a foundation for subsequent eye-tracking-based target selection. The big data model can identify common environmental elements such as buildings, trees, vehicles, and people, and mark them to facilitate user selection through gaze.
[0050] S2: Acquire user eye data and head movement data, and calculate the user's two-dimensional gaze point on the virtual reality display interface from the eye data;
[0051] Acquire user's eye and head movement data.
[0052] Using a pre-defined pupil tracking algorithm, the image data is processed to identify the coordinates of the pupil center point. and the coordinates of the corneal reflection point ,in:
[0053]
[0054] Calculate the pupil-corneal reflection vector ,in:
[0055]
[0056] Using an eye feature mapping model, the pupil-corneal reflectance vector is... Mapped to line-of-sight vector Its expression is:
[0057]
[0058] in, Indicates the tilt angle of the user's line of sight; Indicates the horizontal angle of the user's line of sight; These represent the linear mapping coefficients in the eye biometric calibration matrix, obtained through calibration. This represents the bias compensation term in the calibration of ocular biometrics, obtained through calibration.
[0059] Establish a coordinate system V for the VR glasses lens display, with the origin at the lens center, the x-axis pointing to the right, the y-axis upwards, and the z-axis perpendicular to the lens outwards. Vector the gaze direction. Intersecting with the display plane, the coordinates of the gaze point are obtained. Its calculation expression is:
[0060]
[0061] in, The nominal distance from the eyeball to the display plane of the VR glasses lenses;
[0062] By analyzing the user's eye opening and closing state through a time window, the blink frequency is obtained by calculating the number of blinks per unit time. Its expression is:
[0063]
[0064] in, For time window The number of blinks within the eye, Set to 1 second;
[0065] S3: Map the two-dimensional gaze point from the coordinate system of the virtual reality display interface to the coordinate system of the three-dimensional point cloud data to obtain the corresponding three-dimensional gaze coordinates;
[0066] Establish a coordinate system E for the 3D point cloud data. The origin is the current position of the UAV, the x-axis is along the direction of the UAV's nose, the z-axis is vertically upward, and the y-axis is determined by the right-hand rule. Map the user's gaze point from the VR glasses' display coordinate system V to coordinate system E to obtain the 3D gaze coordinates. The expression is:
[0067] in, This is the rotation matrix from the VR glasses lens display coordinate system V to coordinate system E. The nominal distance from the eyeball to the display plane of the VR glasses lenses;
[0068] S4: Search for and select the point cloud point closest to the three-dimensional gaze coordinates in the three-dimensional point cloud data as the flight target point of the UAV;
[0069] In the three-dimensional point cloud data, calculate each point cloud point. To 3D gaze coordinates The Euclidean distance is expressed as:
[0070] Select the point cloud point with the shortest distance as the drone's flight target point. The expression is:
[0071]
[0072] S5: Based on the flight target point, generate action control commands to control the UAV to move toward the three-dimensional flight target point, determine the head rotation angle based on the head motion data, and calculate attitude control commands to control the UAV's yaw motion and gimbal pitch motion.
[0073] Define the current location of the drone as The control vector for the direction of movement of the UAV for:
[0074]
[0075] Define the speed control quantity of the drone Distance from drone to target point The relationship is:
[0076] in, This is the speed control proportional coefficient. The maximum distance threshold, This is an adjustment factor used to ensure lower speeds at close range;
[0077] Define a U-shaped coordinate system for the UAV, with the origin at the UAV's geometric center. The x-axis points forward along the nose, the z-axis points vertically upward, and the y-axis is determined by the right-hand rule. The control vector for the UAV's movement direction is... and the angle of horizontal rotation of the user's head and pitch angle Mapped to the UAV body coordinate system U, the expression is:
[0078]
[0079] in, The rotation matrix from coordinate system E to the UAV body coordinate system U is expressed as:
[0080]
[0081] Define the motion state vector of the UAV and control vector :
[0082]
[0083] in, For position coordinates, These are roll angle, pitch angle, and yaw angle, respectively. These are the three-axis angular velocity control quantities around the body coordinate system U. Total thrust;
[0084] Construct a calculation model for angular velocity control quantity based on PID control algorithm:
[0085]
[0086] in, This is the proportionality coefficient. These are the differential coefficients. Here, T is the integral coefficient, and T is the UAV speed control variable calculated in S5. Forward coefficients;
[0087] Introducing a gimbal control algorithm to control the tilt and rotation angle of the user's head. Mapped to UAV gimbal pitch rate control quantity :
[0088]
[0089] in, This is the proportionality coefficient. These are the differential coefficients. The integral coefficient;
[0090] To ensure the flight stability of the drone, dynamic attitude constraints are introduced:
[0091]
[0092] in, This is a dynamically adjusted coefficient.
[0093] In this step, if the user continues to stare at the same target point for more than a preset threshold time... In this embodiment, it is preferably 2 seconds. The system locks onto the target and displays it on the VR glasses lens. The user can control the drone to fly around the target point by turning their head.
[0094] S6: Issue the action control command and the attitude control command to drive the UAV to fly, i.e., the command calculated in step S5. .
[0095] This invention includes four flight modes: takeoff, landing, hovering, and movement. It calculates the eye data to obtain the user's blink frequency. When the blink frequency meets the preset trigger conditions, it brings up the flight mode selection interface, selects the flight mode based on the user's eye data, and outputs control commands in combination with the current UAV flight status.
[0096] The system detects whether the user blinks twice or more within one second. If so, a flight mode selection interface is displayed on the VR glasses lenses. The interface includes four selection areas: takeoff, landing, hovering, and movement. Each area has a range of [range missing] in the display coordinate system V. , These correspond to four flight modes; based on the user's gaze coordinates. Determine its region and define a selection function. The expression is:
[0097]
[0098] When a user stares at a certain area for a duration exceeding a preset threshold (Set to 1 second) The visual feedback in this area is enhanced; if the user blinks once at this time, the selection of the corresponding flight mode for this area is confirmed, and drone control commands are received. ;
[0099] like If so, determine whether the drone is in flight. If so, the final output control command is to hover; otherwise, the final output control command is to take off. If the drone is stationary on the ground, then determine whether it is. If so, execute S2; otherwise, output the control command to land. If so, the output control command is fixed-point hovering; if The final output control command is to execute S3 and hide the flight mode selection interface.
[0100] The present invention also provides an eye-tracking-based virtual reality drone flight control device, comprising:
[0101] The eye-tracking and head-movement hybrid perception processing module includes:
[0102] The eye data acquisition unit is used to acquire eye data, including an infrared light source and a miniature infrared camera. The infrared light source illuminates the user's eyes to generate corneal reflection points and enhance pupil contrast. The miniature infrared camera collects infrared image information of the user's eyes and transmits the image data to the processing terminal at a sufficiently high frame rate, such as no less than 60 frames per second, to ensure the real-time performance and accuracy of eye tracking.
[0103] The head data acquisition unit is used to acquire head motion data, including an IMU (Induction Unit), which is used to acquire the user's head horizontal rotation angle. and pitch angle ,in As a yaw control variable for drones. As a pitch control variable for the drone gimbal;
[0104] The processing terminal is equipped with a pupil tracking algorithm for processing eye movement data;
[0105] Virtual reality perception and mixed display control module, including:
[0106] The hybrid head-mounted display unit includes VR glasses, a drone camera, a drone gimbal, and a 5G image transmission module mounted on the drone frame. It is used to acquire and display the drone's first-person perspective view, drone flight status information, user gaze point, and target lock; and to push real-time video data to the cloud server via RTXP.
[0107] The central processing unit, equipped with a large-scale situational awareness model and a PID control algorithm, receives and processes the output data from the eye-tracking and head-motion hybrid perception processing module and the hybrid head-mounted display unit to calculate the drone's current position. With the flight target point Euclidean distance between Based on the Euclidean distance, the action speed control quantity is calculated. The speed control quantity The following nonlinear function is used to automatically attenuate the velocity at close range:
[0108]
[0109] in, This is the speed control proportional coefficient. The maximum distance threshold, An adjustment coefficient is used to automatically reduce speed at close range; and action control commands and attitude control commands are generated.
[0110] The UAV control and execution module includes a communication module, a flight controller, an integrated electronic speed controller, a battery, and motors. The flight controller includes an IMU, a barometer, a magnetometer, and a GPS locator, which are used to receive and execute action control commands and attitude control commands, acquire UAV flight status information, and send it to the virtual reality perception and hybrid display and control module.
[0111] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.
Claims
1. A virtual reality drone flight control method based on eye tracking, characterized in that, include: S1: Acquire depth video data of the environment in which the UAV is located in real time, and calculate the depth video data to generate three-dimensional point cloud data of the environment in real time; S2: Acquire user eye data and head movement data, and calculate the user's two-dimensional gaze point on the virtual reality display interface from the eye data; S3: Map the two-dimensional gaze point from the coordinate system of the virtual reality display interface to the coordinate system of the three-dimensional point cloud data to obtain the corresponding three-dimensional gaze coordinates; S4: Search for and select the point cloud point closest to the three-dimensional gaze coordinates in the three-dimensional point cloud data as the flight target point of the UAV; S5: Calculate the current position of the UAV based on the flight target point. With the flight target point Euclidean distance between Based on the Euclidean distance, the action speed control quantity is calculated. The speed control quantity The following nonlinear function is used to automatically attenuate the velocity at close range: in, This is the speed control proportional coefficient. The maximum distance threshold, This is an adjustment factor used to automatically reduce speed at close range; To generate action control commands for controlling the UAV to move toward the three-dimensional flight target point, determine the head rotation angle based on the head motion data, and calculate attitude control commands for controlling the UAV's yaw motion and gimbal pitch motion; S6: Issue the action control command and the attitude control command to drive the UAV to fly.
2. The eye-tracking-based virtual reality drone flight control method according to claim 1, characterized in that, The specific steps in step S2 to calculate the two-dimensional gaze coordinates from the eye data are as follows: Identify the coordinates of the pupil center point in the user's eye image. and the coordinates of the corneal reflection point ; , These are the abscissa and ordinate components of the pupil center point in the coordinate system to which the eye image belongs, respectively; and These represent the x-coordinate and y-coordinate components of the corneal reflection point in the coordinate system of the eye image, respectively. Indicates transpose processing; Calculate the coordinates of the pupil center point Coordinates of the corneal reflection point The pupil-corneal reflex vector The pupil-corneal reflection vector is mapped using a pre-defined eye feature mapping model. Convert to 3D line-of-sight vector The expression is: in, Indicates the tilt angle of the user's line of sight; Indicates the horizontal angle of the user's line of sight; Represents the linear mapping coefficients in the eye biometric calibration matrix; This represents the bias compensation term in the calibration of ocular biometrics; Solve for the three-dimensional line-of-sight direction vector The intersection point with the virtual reality display interface is used to determine the two-dimensional gaze point.
3. The eye-tracking-based virtual reality drone flight control method according to claim 1, characterized in that, The specific steps for generating action control instructions in step S5 are as follows: Based on the aforementioned action speed control amount and the direction vector components of the flight target point in the body coordinate system and The roll rate control quantity is calculated using a PID control model. and pitch rate control quantity : This is the proportionality coefficient. The differential coefficients are... is the integral coefficient.
4. The eye-tracking-based virtual reality drone flight control method according to claim 1, characterized in that, The attitude control commands in S5 include yaw rate control. The specific generation steps are as follows: Based on the horizontal rotation angle in the head rotation angle The yaw rate control quantity is generated through the PID control model. , in, Forward coefficients, This is the proportionality coefficient. The differential coefficients are... is the integral coefficient.
5. The eye-tracking-based virtual reality drone flight control method according to claim 1, characterized in that, The attitude control commands in S5 include gimbal pitch rate control. The specific generation steps are as follows: Based on the pitch angle in the head rotation angle The gimbal pitch rate control quantity is generated through a PID control model. , in, This is the proportionality coefficient. The differential coefficients are... is the integral coefficient.
6. The eye-tracking-based virtual reality drone flight control method according to claim 1, characterized in that, The method further includes: The eye data is calculated to obtain the user's blink frequency. When the blink frequency meets the preset trigger conditions, the flight mode selection interface is brought up, the flight mode is selected based on the user's eye data, and control commands are output in combination with the current flight status of the drone.
7. The eye-tracking-based virtual reality drone flight control method according to claim 1, characterized in that, When it is detected that the duration of the user's gaze at the flight target point exceeds a preset locking threshold. When the target point is locked, the drone is allowed to fly around the target point.
8. The eye-tracking-based virtual reality drone flight control method according to claim 1, characterized in that, The drone's attitude angle, angular velocity, and position deviation are monitored in real time. When an anomaly is detected, a protection mechanism is triggered, including hovering in place, slow landing, and returning to home.
9. A virtual reality drone flight control device based on eye tracking, characterized in that, include The eye-tracking and head-movement hybrid perception processing module includes: The eye data acquisition unit is used to acquire eye data. The head data acquisition unit is used to acquire head movement data. The processing terminal is used to process eye-tracking data; Virtual reality perception and mixed display control module, including: The hybrid head-mounted display unit is used to acquire and display the first-person view of the drone, drone flight status information, user gaze point, and target lock; The central processing unit, equipped with a large-scale situational awareness model and a PID control algorithm, receives and processes the output data from the eye-tracking and head-motion hybrid perception processing module and the hybrid head-mounted display unit to calculate the drone's current position. With the flight target point Euclidean distance between Based on the Euclidean distance, the action speed control quantity is calculated. The speed control quantity The following nonlinear function is used to automatically attenuate the velocity at close range: in, This is the speed control proportional coefficient. The maximum distance threshold, An adjustment coefficient is used to automatically reduce speed at close range; and action control commands and attitude control commands are generated. The UAV control execution module is used to receive and execute action control commands and attitude control commands, acquire UAV flight status information, and send it to the virtual reality perception and hybrid display and control module.
10. The eye-tracking-based virtual reality drone flight control device according to claim 9, characterized in that, The eye data acquisition unit includes an infrared light source and a miniature infrared camera; the head data acquisition unit includes an IMU; the hybrid head-mounted display unit includes VR glasses, a drone camera, a drone gimbal, and a 5G image transmission module mounted on the drone frame; the drone control and execution module includes a communication module, a flight controller, an integrated electronic speed controller, a battery, and motors, and the flight controller includes an IMU, a barometer, a magnetometer, and a GPS locator.
Citation Information
Patent Citations
Unmanned aerial vehicle eye movement control method, device and system
CN117930981A