Unmanned aerial vehicle illumination control method and system based on three-dimensional model data

Abstract

The invention relates to an unmanned aerial vehicle illumination control method and system based on three-dimensional model data. The method comprises the steps of obtaining three-dimensional model data, environment perception data and flight positioning data, and analyzing a target scene and an unmanned aerial vehicle illumination demand. And based on the data, planning the flight path of the unmanned aerial vehicle, and generating flight path planning data. And in combination with environmental perception information, performing illumination analysis on the main illumination end, calculating main illumination power density distribution and a main illumination angle, and identifying a main illumination blind area. And for the blind area, light supplement analysis is performed by using environmental perception data, and blind area illumination power density distribution and a blind area illumination angle are calculated. And finally, synthesizing the main illumination data and the blind area light supplementing data to generate illumination control information of the unmanned aerial vehicle. By adopting the method, the illumination strategy can be intelligently adjusted under different environmental conditions, the illumination effect, the illumination precision and the illumination uniformity of the unmanned aerial vehicle can be effectively improved, the energy utilization efficiency is also improved, and unnecessary energy consumption is reduced.

Description

Technical Field

[0001] The present application relates to the field of automatic control technology, and in particular to a method and system for controlling lighting of an unmanned aerial vehicle based on three-dimensional model data. Background Art

[0002] In traditional technology, drone lighting control technology mainly relies on traditional remote control methods or preset programs for operation. Usually, drones are equipped with LED lights or other light sources, and the opening, closing and brightness changes of the light sources are adjusted by manual remote control to meet basic lighting needs. In some application scenarios, timing control or simple automatic adjustment mechanisms are also used to enable drones to perform lighting tasks according to preset paths and times. However, traditional technologies for drone lighting control still have certain limitations in terms of precise control, remote operation and adaptive adjustment. For example, the direction of light is difficult to flexibly adjust, the linkage with environmental changes is poor, and there is a lack of intelligent lighting management solutions, which makes it difficult to meet human needs in terms of drone lighting effects, lighting accuracy and lighting uniformity. Summary of the invention

[0003] Based on this, it is necessary to provide a UAV lighting control method and system based on three-dimensional model data that can effectively improve the lighting effect, lighting accuracy and lighting uniformity of the UAV in response to the above technical problems.

[0004] In a first aspect, the present application provides a method for controlling lighting of a drone based on three-dimensional model data, comprising:

[0005] Acquire three-dimensional model data, environmental perception data, and flight positioning data of a target scene; wherein a lighting drone is flying in the target scene;

[0006] Planning a flight path of the UAV in the target scene according to the three-dimensional model data, the environmental perception data, and the flight positioning data to obtain flight path planning data;

[0007] According to the flight path planning data and the environmental perception data, a lighting analysis is performed on the main lighting end of the UAV to obtain a main lighting power density distribution and a main lighting angle;

[0008] Determining each main lighting blind area in the target scene according to the main lighting power density distribution and the main lighting angle;

[0009] According to each of the main lighting blind spots and the environmental perception data, a fill light analysis is performed on each of the main lighting blind spots to obtain a lighting power density distribution of each blind spot and a lighting angle of each blind spot;

[0010] UAV lighting control information is generated according to the main lighting power density distribution, the main lighting angle, the blind spot lighting power density distribution and the blind spot lighting angle.

[0011] In a second aspect, the present application also provides a UAV lighting control system based on three-dimensional model data, including a computer, a UAV, and a data interaction terminal;

[0012] The drone is used to obtain three-dimensional model data, environmental perception data and flight positioning data of the target scene; there is a lighting drone flying in the target scene;

[0013] The computer is used to plan a flight path of the UAV in the target scene according to the three-dimensional model data, the environmental perception data and the flight positioning data to obtain flight path planning data;

[0014] The computer is used to perform lighting analysis on the main lighting end of the UAV according to the flight path planning data and the environmental perception data to obtain the main lighting power density distribution and the main lighting angle;

[0015] The computer is used to determine each main lighting blind area in the target scene according to the main lighting power density distribution and the main lighting angle;

[0016] The computer is used to perform fill light analysis on each of the main lighting blind areas according to each of the main lighting blind areas and the environmental perception data, so as to obtain the lighting power density distribution of each blind area and the lighting angle of each blind area;

[0017] The computer is used to generate the UAV lighting control information according to the main lighting power density distribution, the main lighting angle, the blind spot lighting power density distribution and the blind spot lighting angle;

[0018] The data interaction terminal is used for data transmission between the computer and the drone.

[0019] The above-mentioned UAV lighting control method and system based on three-dimensional model data accurately constructs the spatial information and environmental characteristics of the target area by acquiring the three-dimensional model data of the target scene, the environmental perception data and the flight positioning data of the UAV, and realizes intelligent path optimization in combination with the flight trajectory planning of the lighting UAV, so that the UAV can perform lighting tasks efficiently and safely. Based on the flight path planning data and environmental perception data, the main lighting end of the UAV is accurately analyzed for lighting, and the main lighting power density distribution and lighting angle are optimized to ensure the lighting uniformity and effectiveness of the main lighting area. In addition, the main lighting blind area in the target scene is automatically detected, and the fill light analysis is performed based on the environmental perception data to calculate the optimal blind area lighting power density distribution and lighting angle, so as to effectively reduce the lighting dead angle and improve the overall lighting coverage and quality. Finally, by integrating the data of the main lighting and the blind area fill light, the UAV lighting control information is generated, so that the UAV can intelligently adjust the lighting strategy under different environmental conditions, which can not only effectively improve the lighting effect, lighting accuracy and lighting uniformity of the UAV, but also improve the energy utilization efficiency, reduce unnecessary energy consumption, and significantly enhance the visibility and safety at night or in low light environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 A diagram of an application environment of a method for controlling lighting of a drone based on three-dimensional model data in one embodiment;

[0022] Figure 2 1 is a flow chart of a method for controlling lighting of a UAV based on three-dimensional model data in one embodiment;

[0023] Figure 3 It is a schematic flow chart of a method for obtaining main lighting power density distribution and main lighting angle in one embodiment;

[0024] Figure 4 A schematic flow chart of a method for obtaining main lighting power density distribution in another embodiment;

[0025] Figure 5 is a schematic flow chart of a method for obtaining a main lighting angle in another embodiment;

[0026] Figure 6It is a flowchart of a method for obtaining the lighting power density distribution of each blind spot and the lighting angle of each blind spot in one embodiment;

[0027] Figure 7 A schematic diagram of a flow chart of a method for obtaining blind spot lighting power density distribution and lighting angles of each blind spot in another embodiment;

[0028] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0030] The present invention provides a method for controlling the lighting of a UAV based on three-dimensional model data, which can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the server 104 can be implemented with an independent server or a server cluster composed of multiple servers.

[0031] In an exemplary embodiment, Figure 2 As shown, a UAV lighting control method based on three-dimensional model data is provided, and the method is applied to Figure 1 The server in the example is used to illustrate, including the following steps 202 to 212. Among them:

[0032] Step 202, obtaining three-dimensional model data, environmental perception data and flight positioning data of the target scene.

[0033] Among them, the target scene can be the specific area where the drone performs lighting tasks, including construction sites, road patrol sections, rescue sites, stadiums and other different application environments.

[0034] Among them, the three-dimensional model data can be a digital representation of the target scene, which is usually acquired by devices such as laser radar (LiDAR), depth camera, structured light sensor, etc., and constructed through point cloud data processing, mesh reconstruction, texture mapping and other methods.

[0035] Among them, environmental perception data can be external environmental information collected in real time by sensors carried by drones, including the current light intensity measured by light sensors, temperature, humidity and wind speed information provided by weather sensors, and haze concentration provided by air quality monitoring sensors.

[0036] Among them, the flight positioning data can be the real-time position and attitude information of the drone in the target scene, including GPS (Global Positioning System) coordinates, heading angle, pitch angle, roll angle provided by the inertial navigation system (INS), and relative position data calculated by visual SLAM (Simultaneous Localization and Mapping).

[0037] Specifically, high-precision laser radar (LiDAR), stereo camera (Stereo Camera) or structured light sensor and other equipment are used to scan the target scene to build a complete three-dimensional model data, so as to accurately obtain information such as buildings, obstacles, terrain undulations in the scene. These data are processed through point cloud processing, deep image conversion and grid reconstruction to generate a refined three-dimensional model of the target scene. At the same time, environmental perception equipment such as light sensors, temperature and humidity sensors, wind speed sensors, etc. deployed in drones or scenes collect environmental perception data such as light intensity, meteorological conditions, and air quality in the target area in real time to adapt to the lighting adjustment needs under different weather and scene conditions. In addition, through the global positioning system (GPS), inertial navigation system (INS), visual SLAM (simultaneous localization and mapping) and other technologies, the spatial position, flight trajectory and attitude information of the lighting drone are tracked in real time to obtain flight positioning data.

[0038] Step 204, planning the flight path of the UAV in the target scene according to the three-dimensional model data, the environmental perception data and the flight positioning data, and obtaining flight path planning data.

[0039] Among them, the flight path planning data can be the optimal flight trajectory generated by the UAV according to the preset lighting task, including its take-off point, hovering point, path point, flight altitude, speed and attitude adjustment plan.

[0040] Specifically, based on the three-dimensional scene data, the optimal initial flight path of the drone to avoid obstacles is calculated using path search methods such as the A algorithm, Dijkstra algorithm, or rapid search random tree (RRT), ensuring that it can smoothly and safely perform lighting tasks. However, during the flight, the drone may encounter dynamically changing environmental factors (such as pedestrians, vehicles, wind speed changes, etc.). Therefore, it is necessary to combine real-time environmental perception data and use dynamic obstacle avoidance algorithms such as reinforcement learning and adaptive artificial potential field (APF) to enable the drone to autonomously adjust the optimal initial flight path, avoid sudden obstacles, and improve flight flexibility and safety. In addition, in order to maximize the endurance of the drone, the system optimizes the adjusted flight path based on the drone's battery power, energy consumption model, and mission requirements to ensure that while providing sufficient lighting, energy consumption is minimized to improve the continuity of mission execution.

[0041] When there are multiple drones working together, a distributed task allocation algorithm (such as auction algorithm, ant colony algorithm) needs to be used to ensure that each drone has a reasonable division of labor, avoid path overlap or coverage blind spots, and improve overall lighting efficiency and resource utilization. The flight path planning data finally generated will be used for the actual flight control of the drone, guiding it to move along the optimized trajectory, and combined with the task scheduling system to achieve intelligent and adaptive lighting task execution.

[0042] Step 206: Perform lighting analysis on the main lighting end of the drone according to the flight path planning data and the environment perception data to obtain the main lighting power density distribution and the main lighting angle.

[0043] Among them, the main lighting end can be the main lighting equipment installed on the drone, such as high-power LED lights, laser lighting systems, multi-beam lighting equipment, etc.

[0044] The main lighting power density distribution can be the distribution of the light intensity of the main lighting end in the target scene, usually expressed in terms of luminous flux per unit area (lm / m 2 ). This data is determined by the flight path of the drone, light source power, beam angle, light attenuation model and other parameters, and is calculated through ray tracing or radiosity analysis.

[0045] The main lighting angle can be the beam direction of the drone lighting device, including the pitch angle (vertical direction of the beam), yaw angle (horizontal rotation angle of the beam) and beam spread angle (coverage of the light). This angle is adjusted by the servo or gimbal control to ensure that the light is accurately projected to the target area.

[0046] Specifically, based on the flight trajectory, hovering position and location of the target area of ​​the flight path planning data, the system extracts key lighting areas. For example, in road patrol missions, the system will extract path points along the road that require uniform lighting, while in construction lighting missions, the system needs to identify work areas and high-altitude work points. At the same time, the system combines environmental perception data to analyze the lighting status of the target area, including natural light intensity, weather conditions (such as haze and rainy days), terrain features and possible light obstruction. By comprehensively analyzing the ambient light level, the system can determine the specific requirements of the lighting task, such as whether high-power illumination is required, or whether the lighting angle needs to be adjusted to adapt to external light changes.

[0047] After clarifying the lighting requirements, the system obtains the optical parameters of the main lighting end of the drone, such as luminous flux (lumens), beam angle, light attenuation model, light source type (spotlight or floodlight), etc., establishes a lighting simulation model, and calculates the light distribution at different heights and angles. For example, if the drone is hovering at high altitudes, the light coverage is larger, but the power density will decrease with increasing distance, so the system may need to increase the lighting power or adjust the beam angle to ensure that the ground illumination meets the mission standards. In addition, the system will dynamically adjust the power in combination with the ambient light compensation algorithm. For example, in areas with sufficient light, the system can appropriately reduce the power to save energy, and in shadow areas or low-light environments, it automatically increases the power to ensure the minimum illumination requirements. Finally, the dynamic main lighting power density distribution is calculated to ensure that the light source can meet the lighting standards of the target area.

[0048] After calculating the power density distribution, the system analyzes the real-time posture information of the drone, including position, altitude, pitch angle, yaw angle and roll angle, to determine the spatial relationship between the light source and the target area. For example, if the drone is directly above the target area, the light beam should be projected vertically downward, and if the drone is in an oblique flight state, the beam angle needs to be adjusted appropriately to ensure that the light can completely cover the target area. Subsequently, the system calculates the optimal beam projection angle based on geometric optics so that the light can cover the target area to the greatest extent while avoiding excessive concentration or diffusion of light. In addition, in a dynamic environment, the system combines a closed-loop control algorithm to adjust the direction of the beam in real time. For example, if the drone deviates due to wind during flight, the system will automatically correct the beam angle to maintain the stability of the lighting. Finally, the dynamic main lighting angle is calculated to ensure that the beam can be accurately projected to the target area.

[0049] Step 208: determining each main lighting blind area in the target scene according to the main lighting power density distribution and the main lighting angle.

[0050] Among them, the main lighting blind area can be the area in the target scene that is not covered by the main lighting end due to factors such as obstacle blocking, beam angle limitation, light attenuation, etc. These blind areas may be located in the shadow of buildings, underground passages, corners, etc., affecting the overall lighting quality.

[0051] Specifically, a spatial illumination analysis method is used to detect areas in the target scene where the illumination is lower than the threshold according to the set minimum illumination requirement threshold (such as the Lux standard) to identify areas with insufficient illumination. Then, the three-dimensional scene illumination projection analysis is further used to combine the spatial information of the obstructions such as buildings, trees, and mechanical equipment in the target scene to identify the shadow areas caused by the obstruction, and avoid the lighting blind spots caused by the influence of fixed obstacles. In addition, in order to improve the accuracy of blind spot detection, light propagation simulation (such as Monte Carlo path tracing and photon mapping) is used to analyze the reflection, refraction and attenuation of the illumination, calculate the possible blind spot boundaries of the above two situations, and ensure that the detected blind spot range is accurate. In order to adapt to the dynamic changes of the environment, the system also needs to perform real-time blind spot detection, that is, when the illumination sensor or environmental perception data detects that the illumination level in a certain area fluctuates abnormally or is unstable, the system will dynamically adjust the blind spot judgment rules, adjust the above blind spots and blind spot boundaries, ensure the real-time and reliability of blind spot identification, and finally determine the spatial position, range and formation cause of each main lighting blind spot.

[0052] Step 210, performing fill light analysis on each main lighting blind spot according to each main lighting blind spot and environmental perception data, and obtaining lighting power density distribution of each blind spot and lighting angle of each blind spot.

[0053] Among them, fill light analysis can be the process of calculating the fill light demand based on environmental perception data after identifying the main lighting blind spot, and formulating the best fill light plan.

[0054] The blind spot lighting power density distribution can be the light intensity distribution of the fill light device in the blind spot, which is used to evaluate whether the fill light effect meets the lighting requirements. The distribution is determined by the fill light source power, beam angle, light attenuation model, ambient light intensity and other parameters, and is calculated through lighting simulation.

[0055] The blind spot lighting angle can be the irradiation direction of the fill light source, including the internal fill light angle and the external fill light angle. The internal fill light angle is adjusted by the gimbal control of the drone, while the external fill light angle is determined by the position and rotation angle of the ground light source or other drone light sources.

[0056] Specifically, after the main lighting blind spot is identified, it is necessary to perform fill light analysis in combination with environmental perception data (including real-time light level, obstacle distribution, meteorological conditions, etc.) to determine the blind spot lighting power density distribution and blind spot lighting angle of each main lighting blind spot to ensure the uniformity of lighting in the target area.

[0057] First, the fill light power density distribution and angle of the internal fill light (the drone's own fill light) and the external fill light (ground auxiliary light source or other drone fill light) are calculated separately. For the internal fill light, the Monte Carlo path tracing algorithm is used to calculate the additional fill light power density distribution required under the contribution distribution of the main lighting source in the blind spot combined with the contribution distribution of the ambient light, based on the posture, light source characteristics and illumination range of any path point of the drone in the flight path planning data. At the same time, the gimbal control algorithm (combined with the IMU inertial measurement unit, ambient light sensor and real-time positioning data) is used to automatically calculate and adjust the light source posture, determine the optimal internal fill light angle, and accurately align the fill light direction with the blind spot, while avoiding excessive light affecting the surrounding area.

[0058] If the built-in fill light cannot meet the fill light demand, the external fill light demand is further calculated, and the photon mapping algorithm is used to calculate the light source power density distribution of different external light sources (such as ground lamps or other drones) in the blind area under the effect of the main lighting contribution distribution + ambient light contribution distribution + built-in fill light contribution distribution. In addition, the calculation of the lighting angle of the external fill light needs to consider the position of the light source, the shape of the blind area, and the light diffusion characteristics. The geometric optics method is used to calculate the optimal incident angle so that the light beam covers the blind area and reduces overlapping lighting. The formula is tan (θ) = blind area height difference / blind area horizontal distance, and combined with the beam diffusion angle calculation formula beam opening angle ≈ 2 × arctan (blind area width / 2 / fill light distance), to ensure a reasonable light diffusion range.

[0059] Finally, the power density distribution and lighting angle of the internal fill light and the external fill light are integrated, and the light source output is optimized through the weighted average method to avoid excessive brightness or insufficient illumination in local areas. At the same time, the direction of the light source is adjusted to ensure that the light evenly covers the blind spots. Finally, the blind spot lighting power density distribution and the blind spot lighting angle of each main lighting blind spot are output respectively. The fill light strategy is dynamically optimized in combination with real-time environmental perception data, so that the UAV and the external light source can work together to ensure the lighting uniformity of the target area and improve the energy efficiency and adaptability of the lighting system.

[0060] Step 212, generating drone lighting control information according to the main lighting power density distribution, the main lighting angle, the blind spot lighting power density distribution, and the blind spot lighting angle.

[0061] Among them, the UAV lighting control information can be a control instruction generated based on the main lighting power density distribution, the main lighting angle, the blind spot lighting power density distribution and the blind spot lighting angle, which is used to adjust the lighting behavior of the UAV in real time.

[0062] Specifically, based on the main lighting power density distribution and the main lighting angle, the UAV flight lighting control instructions are generated to adjust the UAV's flight attitude (position, altitude, heading) and the power output of the light source to ensure that the main lighting source can accurately cover the target area while avoiding excessive lighting or energy waste. Then, combined with the blind spot lighting power density distribution and blind spot lighting angle, the system analyzes the lighting needs of the area not covered by the main lighting, generates fill light scheduling instructions, and decides whether to use the UAV's own fill light (adjust the beam angle or power) or dispatch other fill light means (such as additional UAV fill light or ground lighthouse support).

[0063] During the actual flight of the drone, the system combines environmental perception data to monitor changes in light intensity and target areas in real time, and generates dynamic lighting adjustment instructions to dynamically adjust lighting power, lighting angle and flight path to ensure uniform light distribution and adapt to the external environment (such as pedestrian flow and weather changes). In addition, if the fill light demand changes, a blind spot lighting distribution optimization instruction is generated to calculate the new blind spot lighting distribution, and the fill light task allocation of the drone and ground light sources is optimized in real time to ensure that the overall lighting coverage is maximized. Ultimately, all execution instructions are compiled into the lighting control instructions for the entire drone, which are used as drone lighting control information to automatically execute main lighting adjustment, fill light drone scheduling, ground lighting optimization and flight obstacle avoidance strategies.

[0064] In the above-mentioned UAV lighting control method based on three-dimensional model data, the spatial information and environmental characteristics of the target area are accurately constructed by acquiring the three-dimensional model data of the target scene, the environmental perception data and the flight positioning data of the UAV, and the flight trajectory planning of the lighting UAV is combined to realize intelligent path optimization, so that the UAV can perform lighting tasks efficiently and safely. Based on the flight path planning data and environmental perception data, the main lighting end of the UAV is accurately analyzed for lighting, and the main lighting power density distribution and lighting angle are optimized to ensure the lighting uniformity and effectiveness of the main lighting area. In addition, the main lighting blind area in the target scene is automatically detected, and the fill light analysis is performed based on the environmental perception data to calculate the optimal blind area lighting power density distribution and lighting angle, so as to effectively reduce the lighting dead angle and improve the overall lighting coverage and quality. Finally, by integrating the data of the main lighting and the blind area fill light, the UAV lighting control information is generated, so that the UAV can intelligently adjust the lighting strategy under different environmental conditions, which can not only effectively improve the lighting effect, lighting accuracy and lighting uniformity of the UAV, but also improve the energy utilization efficiency, reduce unnecessary energy consumption, and significantly enhance the visibility and safety at night or in low light environments.

[0065] In an exemplary embodiment, Figure 3As shown, according to the flight path planning data and the environmental perception data, the main lighting end of the UAV is subjected to lighting analysis to obtain the main lighting power density distribution and the main lighting angle, including steps 302 to 306. Among them:

[0066] Step 302: Determine scene lighting requirement information of the target scene based on the flight path planning data and the environment perception data.

[0067] The scene lighting requirement information may be the result of analyzing the lighting requirements of the target area, including the minimum required lighting intensity, lighting uniformity requirements, key lighting requirements for specific areas, and possible lighting blind spots.

[0068] Specifically, the system uses 3D scene modeling data to identify buildings, roads, key target points and obstructions that may affect lighting in the area, and clearly identifies areas that require key lighting; at the same time, it combines environmental perception data, such as current ambient light intensity, weather conditions, air quality, etc., to evaluate existing lighting conditions to determine the lighting needs of different areas. For example, if the current ambient light is strong, the system can reduce the lighting power, while in low-light environments or bad weather, the lighting intensity needs to be increased. In addition, the system must determine the minimum lighting threshold, lighting uniformity requirements, and special lighting accuracy requirements for specific scenes (such as construction sites, airports, patrol routes, etc.) in accordance with the set illumination standards. By comprehensively analyzing scene characteristics and real-time ambient light levels, the system ultimately generates complete scene lighting demand information.

[0069] Step 304: Perform lighting analysis on the main lighting end of the drone according to the scene lighting requirement information to obtain the main lighting power density distribution.

[0070] The main lighting end can be the main lighting device on the drone, which is responsible for providing the main light source to illuminate the target area. The lighting device can be a high-power LED lamp, a laser lighting system, or a multi-beam projection lamp, and the specific choice depends on the application scenario. The core parameters of the main lighting end include luminous flux (brightness), beam angle, light attenuation characteristics, and adjustability.

[0071] Specifically, the system obtains the position, altitude, heading and other attitude information of the drone at the path point from the flight path planning data, and establishes a lighting model in combination with the characteristics of the light source (such as luminous flux, beam angle, and light attenuation characteristics) to analyze the lighting effect of the current light source on the target area; then the system simulates the propagation path of the light, analyzes the distribution of light intensity in different areas, and calculates the initial light power density distribution of each area. In order to ensure the stability of the lighting, the system will adjust the initial light power density distribution of each area in combination with the scene lighting demand information. For example, the power density distribution is reduced under high ambient light conditions to avoid light pollution, and the power density distribution is enhanced under low light conditions to meet the brightness requirements; in addition, the system will dynamically adjust the power output through real-time lighting feedback to avoid excessive or insufficient lighting and ensure lighting uniformity; the system output data after double optimization is ultimately used as the main lighting power density distribution of the main lighting end.

[0072] Step 306, according to the scene lighting requirement information and the drone posture information, the servo offset of the main lighting end is calculated to obtain the main lighting angle.

[0073] Among them, the drone's posture information can be the drone's spatial position in the air (X, Y, Z coordinates), attitude (pitch angle, yaw angle, roll angle) and flight speed and other parameters. These data are used to accurately determine the state of the drone and assist in adjusting its lighting direction. This information is usually obtained by GPS (Global Positioning System), IMU (Inertial Measurement Unit), visual SLAM (Simultaneous Localization and Mapping) or radar sensors, and updated in real time.

[0074] The servo offset can be the servo rotation angle change value calculated to adjust the lighting direction of the drone, which is used to control the pitch or yaw angle of the lighting beam to ensure that the light covers the target area.

[0075] Specifically, in order to accurately control the direction of illumination, the system needs to calculate the optimal illumination angle of the main illumination end according to the scene illumination requirements and the real-time attitude information of the drone, and determine the offset of the servo to adjust the direction of the light beam. In the specific implementation process, the system analyzes the current position of the drone, the flight altitude, and the position of the target area, and determines the optimal projection direction of the light beam, so that the light can cover the entire target area or the specified position of the target area as much as possible, while avoiding the light from scattering to the non-target area. The system then calculates the deviation between the current beam direction and the optimal illumination angle of the target area, and obtains the servo rotation angle that needs to be adjusted. If there is a deviation value, the system will adjust the direction of the light source through the servo control or the pan-tilt to accurately align it with the target area. In addition, in order to further optimize the lighting effect, the system also needs to calculate the appropriate beam opening angle and further adjust the light source direction through the servo control or the pan-tilt to meet the beam opening angle, so as to ensure that the light is neither overly concentrated in a small area nor diffused too much to cause insufficient illumination. Finally, the system outputs the main illumination angle and the servo offset, and continuously adjusts the light source direction during the flight of the drone to adapt to environmental changes and ensure lighting stability and uniformity of coverage.

[0076] In this embodiment, the lighting requirements of the target scene are accurately determined through flight path planning data and environmental perception data, thereby optimizing the configuration of the main lighting end of the drone and improving the uniformity and energy efficiency of the lighting coverage. Based on the scene lighting demand information, the system calculates the main lighting power density distribution to ensure that the lighting intensity matches the actual needs and avoid insufficient lighting or energy waste. At the same time, combined with the drone's posture information, the servo offset of the main lighting end is dynamically adjusted to optimize the main lighting angle to ensure that the light beam is accurately projected into the target area, thereby improving the stability and adaptability of the lighting. It can effectively improve the accuracy and automation level of the drone's intelligent lighting, enabling it to adapt to complex environmental changes and achieve efficient lighting control in tasks such as night patrols, emergency rescue, and construction lighting.

[0077] In an exemplary embodiment, Figure 4 As shown, according to the scene lighting requirement information, the main lighting end of the drone is subjected to lighting analysis to obtain the main lighting power density distribution, including steps 402 to 406. Among them:

[0078] Step 402: Detect ambient light compensation information of the target scene according to the environmental perception data.

[0079] The ambient light compensation information may be the lighting compensation requirement calculated after evaluating the natural light level in the target scene and the influence of other external light sources.

[0080] Specifically, after the drone enters the target scene, it first needs to detect the ambient light compensation information of the current scene through environmental perception data, that is, the contribution of natural light or other external light sources to the scene lighting. This process mainly relies on the light sensors carried by the drone (such as photodiodes, HDR cameras, spectrometers) to collect light intensity data at different locations in real time, and combines weather sensors (such as temperature and humidity sensors, cloud density detectors) to evaluate the impact of current weather on lighting. For example, on sunny days, the natural light provided by the sun is higher, while on cloudy days, foggy days or at night, the ambient light is lower, requiring higher artificial lighting compensation.

[0081] In order to accurately assess the impact of ambient light, the system will measure the lighting of multiple key areas (such as work areas, roads, and building facades) during the flight of the drone or in advance, and establish an ambient lighting distribution map to analyze the lighting differences in different areas. For example, on the backlit side of a building, the ambient light illumination is low, while in open areas there may be strong natural light. The system will also use historical environmental data and weather forecast information to predict future trends in ambient light changes. For example, changes in the angle of the sun may cause certain areas to have different lighting at different times. After completing data collection, the system will filter and normalize the light sensor data to eliminate noise interference (such as reflections, instantaneous brightness changes, etc.), and finally obtain accurate ambient light compensation information.

[0082] Step 404: determining actual scene lighting information of the target scene according to the difference between the scene lighting requirement information and the ambient light compensation information.

[0083] The actual scene lighting information may be the current lighting level and actual lighting effect of the target area calculated by combining the ambient light compensation information and the scene lighting requirement information.

[0084] Specifically, after obtaining the ambient light compensation information, the system needs to combine the scene lighting demand information to calculate the actual lighting information of the current scene, that is, the difference between the existing lighting level and the ideal lighting demand. First, the system analyzes the scene lighting demand information, including the minimum illumination requirements of each area, the lighting uniformity standard, and the lighting requirements of special functional areas. For example, at a construction site, the work area may require at least 500 lux of illumination, while the security patrol area may require a basic lighting of 200 lux. The system then compares the ambient light compensation information with the lighting demand standard to calculate the lighting gap or excess lighting situation in each area. If the ambient light illumination of an area has met or exceeded the requirement, no additional lighting is required; but if the lighting level is lower than the standard, the lighting power that needs to be supplemented is calculated. For example, assuming that the target illumination of an area is 500 lux, and the actual measured illumination after ambient light compensation is only 200 lux, the system determines that the lighting power gap in the area is 300 lux, which needs to be compensated by the main lighting end.

[0085] In addition, the system will combine the 3D scene modeling data to analyze the propagation of light, including light reflection, refraction, and occlusion, and further correct the calculation results. For example, in highly reflective areas (such as glass curtain walls), there may be a problem of local over-brightness, and the system needs to reduce the light output in this area; in shadow areas blocked by buildings, additional lighting intensity needs to be increased. Through comprehensive analysis, the system ultimately generates the actual scene lighting information of the target scene, providing accurate data input for the calculation of the main lighting power distribution.

[0086] Step 406: Perform ray tracing analysis on the target scene based on the three-dimensional model data and the actual lighting information of the scene to obtain the main lighting power density distribution.

[0087] Specifically, after determining the actual lighting information of the target scene, it is necessary to combine the 3D model data and use ray tracing technology to simulate the propagation path of light to calculate the main lighting power density distribution. Ray tracing analysis is an optical simulation method that can accurately simulate the propagation, reflection, refraction and occlusion of light emitted by a light source in space, thereby calculating the light intensity distribution of each area.

[0088] In the specific implementation process, the system establishes a complete three-dimensional scene lighting simulation environment based on the three-dimensional model data and the actual lighting information of the scene, including the position of the drone, the light source characteristics of the main lighting end (luminous flux, illumination angle, beam distribution), and the surface characteristics of objects in the scene (such as reflectivity, diffuse reflection or specular reflection properties). The system then emits a large number of virtual light rays from the main lighting end, calculates the propagation trajectory of the light in the scene, and analyzes the lighting coverage of different areas. For example, when light shines on a building, the system calculates the contribution of its reflected light to other areas and identifies lighting blind spots caused by occlusion.

[0089] During the simulation, the system will combine the energy attenuation characteristics of light to calculate the changes in light intensity at different distances. For example, as the distance of light propagation increases, the light intensity will decay according to the inverse square law, so the light power in long-distance areas needs to be increased accordingly. At the same time, the system will also detect possible highlight areas (such as areas where multiple light sources overlap) and avoid excessive lighting through power adjustment. The final ray tracing analysis outputs data on the main lighting power density distribution, which accurately indicates the light intensity of each area and provides targeted power adjustment suggestions. For example, in areas with insufficient light, the system will automatically increase the drone's light power or adjust the beam angle; in areas with excessive light, the system may reduce the light source power or adjust the flight path to optimize the uniformity of light.

[0090] Among them, the calculation formula for the main lighting power density distribution is:

[0091]

[0092] Among them, E actual (P) is the main lighting power density distribution, E env (P) is the ambient light compensation information, P is the coordinate in the target scene where the illumination needs to be calculated, l is the lth main lighting coordinate, I l (θ l +Δθ comp (P)) is the main lighting angle correction data Δθ comp (P) Corrected light intensity, θ l The direction angle defined by the main lighting's own luminosity curve, d l is the distance between coordinate point P and coordinate point l, φ l is the angle between the surface normal at coordinate point P and the main lighting direction, Δφ adj (P) is the additional angle correction of the drone’s posture information, V(P, L l ) is the visibility function between coordinate point P and coordinate point l, α(R env (P)) is the comprehensive attenuation or amplification factor, P env (P) is the measurement parameter of the local environment of the coordinate point P, and α(·) is the dynamic adjustment function.

[0093] In this embodiment, the ambient light compensation information of the target scene is detected through environmental perception data, and the influence of natural light and other external light sources is accurately evaluated to achieve intelligent lighting adjustment. Combined with the scene lighting demand information, the system calculates the actual lighting demand to ensure uniform lighting coverage and reduce energy waste and light pollution. In addition, the three-dimensional model data and ray tracing analysis are used to accurately simulate the light propagation path, optimize the main lighting power density distribution, ensure the reasonable distribution of the lighting beam, minimize blind spots, and improve lighting quality and energy efficiency. It can effectively enhance the adaptive ability of the drone lighting system, enabling it to adjust the lighting strategy according to real-time environmental changes and realize intelligent and efficient lighting control.

[0094] In an exemplary embodiment, Figure 5 As shown, according to the scene lighting requirement information and the drone posture information, the servo offset of the main lighting end is calculated to obtain the main lighting angle, including steps 502 to 508. Among them:

[0095] Step 502: Determine the lighting space orientation angle data of the main lighting end according to the scene lighting requirement information and the drone posture information.

[0096] Among them, the lighting space orientation angle data can be a core parameter set of the direction of the drone lighting beam without environmental compensation.

[0097] Specifically, before calculating the angle of the main lighting end, the system analyzes the scene lighting demand information of the target scene, including the position, height, light intensity requirements and lighting coverage of the target area. At the same time, the real-time posture information of the drone (including its position, flight altitude, heading, pitch angle, yaw angle, and roll angle) is obtained, and the optimal projection direction of the light source is calculated in combination with the relative coordinate relationship of the drone relative to the target area. Then, based on the satisfaction of the scene lighting demand information, the system analyzes the physical characteristics of the main lighting end (such as beam angle, gimbal rotation range, and light source power distribution) to determine the optimal initial orientation of the light source under the current drone posture. For example, when the drone is in a higher position, the light beam needs to be tilted downward at a certain angle to ensure that the light effectively covers the target area on the ground. The system ultimately generates lighting space orientation angle data.

[0098] Step 504: Compensate the lighting space orientation angle data according to the environment perception data of the target scene to obtain the lighting space orientation angle.

[0099] Among them, the lighting space orientation angle can be the light beam direction of the main lighting end of the drone after environmental compensation, including the horizontal rotation angle (yaw angle) and the vertical pitch angle (pitch angle).

[0100] Specifically, after determining the lighting space orientation angle data, the system needs to combine the environmental perception data of the target scene to compensate for the light source direction to adapt to dynamic environmental changes and optimize lighting accuracy. Environmental perception data includes factors such as wind speed, wind direction, temperature and humidity, obstacle distribution, terrain undulations, and light reflection characteristics of the target area, all of which will affect the final lighting effect.

[0101] In the specific implementation process, the system analyzes the wind speed and direction of the environmental perception data, predicts the possible yaw or attitude changes of the drone during flight, and adjusts the direction of the light source in advance to ensure that the light can always be aimed at the target area. For example, in a strong wind environment, the drone may have a slight lateral drift. The system will compensate the projection angle of the light beam in advance to keep the lighting direction stable. At the same time, the system considers the light reflection of the target area. For example, if the ground reflectivity is high, it may cause excessive light in the local area. At this time, the system will adjust the beam angle to reduce direct illumination of the high-reflection area to avoid glare problems. In addition, if there are dynamic obstructions (such as pedestrians and vehicles) in the target area, the system will adjust the direction of the light source according to the height and movement speed of the obstacle, so that the lighting can flexibly avoid the obstruction and improve the lighting efficiency. Finally, the system optimizes the direction of the light source according to all environmental compensation factors, and obtains the optimized lighting space orientation angle, ensuring that the main lighting end can still accurately project to the target area in a complex environment, improving lighting stability and energy efficiency.

[0102] Step 506, when an abnormal flying object appears in the environmental perception data, predict the flight path of the abnormal flying object based on the flight status information of the abnormal flying object.

[0103] Among them, abnormal flying objects can be flying targets that enter the target scene range during the execution of the UAV lighting mission and may affect the flight safety or lighting stability of the UAV.

[0104] Among them, the flight status information can be a description of the dynamic behavior and motion parameters of the abnormal flying object in the air, including position coordinates (X, Y, Z), speed (Vx, Vy, Vz), heading angle, pitch angle, roll angle, climb / descent rate, etc.

[0105] Among them, the flight route of the flying object can be the future flight path calculated by the trajectory prediction algorithm based on the flight status information of the abnormal flying object.

[0106] Specifically, when an abnormal flying object (such as a drone, a bird, or a low-flying obstacle) is detected in the environmental perception data, the system needs to predict the movement trajectory of the flying object to prevent the lighting beam from being disturbed or affecting flight safety. First, the system uses a radar sensor, an optical camera, or an infrared detection system to track abnormal flying objects in real time and obtain their flight status information such as position, altitude, speed, and direction. Then the system combines machine learning trajectory prediction models (such as Kalman filtering, long short-term memory network (LSTM), and Bayesian trajectory prediction) to analyze the historical movement trajectory of abnormal flying objects and infer their flight routes. For example, if a drone is detected approaching an illuminated area at a fixed speed, the system can predict its flight position in the next few seconds and calculate the illuminated area that it may block. In addition, the system will analyze the behavior patterns of flying objects. If a bird is detected hovering at high altitude, the system may determine that there is a large uncertainty in its flight path. For low-altitude drones, the system can predict their possible flight targets and obstacle avoidance strategies, and then formulate response plans. The precise trajectory prediction system can optimize the direction of the lighting beam in advance to ensure the continuity of the lighting task, while reducing the risk of flight safety affected by beam obstruction or reflection.

[0107] Step 508, optimizing the lighting space orientation angle according to the flight path of the flying object to obtain the main lighting angle.

[0108] Specifically, after predicting the flight path of an abnormal flying object, the system needs to optimize the lighting space orientation angle to ensure that the lighting beam can maintain accurate coverage of the target area while avoiding interference or safety risks to the abnormal flying object. First, the system analyzes the movement trend of the abnormal flying object in the future, and according to the area that it may block, uses an adaptive lighting optimization algorithm, combined with the priority of the lighting task, to calculate the angle or method of optimizing the lighting space orientation angle to obtain the main lighting angle. For example: if the abnormal flying object only passes through the lighting area for a short time, the system can reduce the lighting power or adjust the beam angle in a short time to reduce beam interference; if the abnormal flying object stays for a long time or frequently crosses the lighting area, the system can re-optimize the flight position of the drone so that the lighting angle is projected from a new orientation to keep the lighting stable. In addition, in some special cases, such as detecting that the abnormal flying object is an emergency avoidance object (such as a rescue drone or a flock of low-flying birds), the system can combine the obstacle avoidance strategy to actively adjust the lighting angle so that the beam avoids the critical flight area, and temporarily shut down or change the lighting mode when necessary to ensure safety.

[0109] Among them, the calculation formula of the main lighting angle is:

[0110]

[0111] in, For the main lighting angle, is the target angle vector, θ drone (t) is the attitude angle of the UAV itself, β avoid (t) is the avoidance angle compensation, Δ SLAM (t) is the fine-tuning amount generated by visual feedback, W(t) is the lighting equipment coordination weight, and F(·) is the multidimensional mapping control strategy function.

[0112] In this embodiment, the lighting space orientation angle data is calculated through the scene lighting demand information and the drone posture information to ensure that the main lighting end light beam is accurately aimed at the target area. At the same time, dynamic compensation is performed in combination with environmental perception data to adapt the lighting angle to external lighting and environmental changes, thereby improving lighting stability and uniformity. When an abnormal flying object is detected, the system predicts the flight trajectory based on its flight status information and optimizes the lighting angle to avoid lighting interference or safety risks. Ultimately, it can effectively improve the intelligence level of the drone lighting system, enabling it to have adaptive adjustment, obstacle avoidance compensation and precise lighting capabilities, ensuring efficient and stable lighting coverage in complex environments.

[0113] In an exemplary embodiment, Figure 6As shown, according to each main lighting blind area and environmental perception data, fill light analysis is performed on each main lighting blind area to obtain the lighting power density distribution of each blind area and the lighting angle of each blind area, including steps 602 to 606. Among them:

[0114] Step 602, calculating the fill light requirement information of each blind spot according to each main lighting blind spot and the environmental perception data.

[0115] The blind spot fill light requirement information may be additional lighting requirement data determined after calculating the illumination deficiency of the main lighting blind spot.

[0116] Specifically, after identifying the main lighting blind spots, the system needs to combine the environmental perception data to calculate the fill light requirements of each main lighting blind spot to ensure overall lighting uniformity. First, the system analyzes the spatial distribution characteristics of the main lighting blind spots, including its location, area, height, reasons for obstruction (such as buildings, terrain undulations), and other information. Then, the ambient light level of the current blind spot is measured using a light sensor and a ray tracing algorithm to calculate the illumination difference between it and the standard lighting requirement. For example, if the minimum illumination requirement for a blind spot is 300 lux, and the actual measured ambient light is only 100 lux, the fill light requirement for the blind spot is 200 lux.

[0117] In addition, the system also considers factors such as weather conditions, reflectivity, and light attenuation to optimize fill light calculations. For example, in high humidity or haze environments, light may be more severely scattered and attenuated, so the fill light power needs to be increased. At the same time, if the main lighting blind spot is on a highly reflective surface (such as near a glass curtain wall), the fill light demand may be reduced to prevent excessive lighting or glare problems. Finally, based on the above factors, the system generates blind spot fill light demand information for each main lighting blind spot, including the required light intensity, fill light range, and light source direction, to provide data support for the subsequent formulation of fill light strategies.

[0118] Step 604 , determining information of each internal fill light device and information of each external fill light device respectively according to the fill light demand information of each blind spot.

[0119] Among them, the information of the internal fill light equipment can be the detailed parameters of the fill light device carried by the drone itself, including the fill light source type (such as LED fill light, spotlight), luminous flux (lm), beam angle, rotation range, power adjustment capability, gimbal controllable angle, etc.

[0120] Among them, the external fill-in-light equipment information can be the parameter data of the auxiliary light source outside the drone, including ground fixed light sources (such as overhead lighthouses, mobile light sources) and fill-in-light devices provided by other drones. This information usually includes the light source position, height, luminous flux, beam angle, power adjustment range, mobility (fixed or adjustable), etc.

[0121] Specifically, after the blind spot fill lighting demand information is determined, the system needs to decide how to implement the fill lighting based on the blind spot fill lighting demand information, that is, to determine the fill lighting configuration scheme of the internal fill lighting equipment (the drone’s own fill lighting) and the external fill lighting equipment (other drones or ground auxiliary lighting equipment).

[0122] To determine the information of the internal fill light equipment, the system first checks the auxiliary lighting capabilities of the drone itself, including parameters such as the auxiliary light source power, beam angle, and adjustable range, to determine whether it can effectively cover the blind spot. If the current position and posture of the drone allow its fill light to directly cover the blind spot, the internal fill light is used first to obtain the information of each internal fill light equipment. At the same time, the system will calculate the required auxiliary light source direction adjustment amount, and adjust the lighting direction in combination with the drone's gimbal control system so that the light beam is accurately directed to the blind spot. In addition, if the blind spot is small or the lighting demand is low, the system may only need to increase the power of the main light source in the machine without starting the auxiliary light source or adjusting the angle of the auxiliary light source.

[0123] Regarding the external fill-light equipment information, if the internal fill-light equipment cannot completely cover the blind spot (such as the auxiliary light source angle is limited, the power is insufficient, and the lighting range is insufficient), the system will dispatch the external fill-light equipment. The external fill-light equipment may include ground fixed lighting equipment (such as elevated lighthouses, mobile light sources), or other drone formation fill-light. The system analyzes the available external light sources, including their location, light source parameters (power, angle, coverage range), mobility, etc., and selects the optimal fill-light method. For example, if the blind spot is in the shadow of a building and the drone's flight altitude is limited, the system may choose a ground lighthouse for fill-light; if the blind spot is in an area where the drone cannot directly project, the system may dispatch another drone to the appropriate position to provide fill-light, and obtain the information of each external fill-light equipment.

[0124] Step 606, calculating the blind spot lighting power density distribution and the blind spot lighting angle corresponding to each main lighting blind spot according to the information of each internal fill-light device and the information of each external fill-light device.

[0125] Specifically, after determining the internal and external fill-light equipment, the system needs to calculate the blind spot lighting power density distribution and blind spot lighting angle of each main lighting blind spot to ensure that the fill-light can accurately and efficiently cover the blind spot.

[0126] For calculating the blind spot lighting power density distribution, the system first calculates the lighting intensity of each blind spot according to the power output capacity of the internal and external fill light equipment; and then analyzes the distribution of the lighting intensity of each blind spot in the blind spot by simulating the light propagation path. If the distribution fails to meet the blind spot fill light demand information, the lighting intensity of the internal fill light and the external fill light are adjusted respectively to obtain the internal blind spot lighting power density distribution and the external blind spot lighting power density distribution. For example, if the internal fill light can cover 70% of the blind spot, but the power density is insufficient, the system will increase the power of the external fill light to ensure that the overall lighting reaches the set illumination standard. In addition, the system also takes into account the energy attenuation and environmental reflection of the light, and dynamically optimizes the fill light power to avoid local over-brightness or over-darkness.

[0127] For the calculation of blind spot lighting angles (inside and outside the aircraft), first of all, for the fill light of the auxiliary light source inside the aircraft, the system calculates the pitch angle and yaw angle of the auxiliary light source based on the real-time posture information of the drone, and obtains the lighting angle of the blind spot inside the aircraft, so that the light can accurately cover the blind spot. The gimbal control system of the drone will receive the calculation results of the lighting angle of the blind spot inside the aircraft, and adjust the direction of the light source in real time to ensure the stability of the lighting. For the fill light of the auxiliary light source outside the aircraft, the system calculates the optimal projection angle according to the position of the ground light source or other drones, and obtains the lighting angle of the blind spot outside the aircraft, so that the light beam of the external light source can effectively fill the blind spot, and work together with the fill light inside the aircraft to form a complete lighting coverage.

[0128] Finally, the system will use optimization algorithms (such as illumination balance optimization and minimum energy consumption allocation strategy) to optimize the internal blind spot lighting power density distribution, the external blind spot lighting power density distribution, the internal blind spot lighting angle, and the external blind spot lighting angle, and then perform virtual equivalent replacement of the power density distribution and the lighting angle, and merge them into a comprehensive blind spot lighting power density distribution and blind spot lighting angle to achieve accurate fill light power output and beam direction, ensure uniform distribution of light in the entire blind spot, and minimize energy consumption. For example, if the internal fill light can complete 80% of the fill light task, the system will instruct the external light source to only provide the remaining 20% ​​of the fill light power to avoid unnecessary energy consumption.

[0129] In this embodiment, the blind spot fill light demand information is calculated through the main lighting blind spot analysis and environmental perception data to ensure that all lighting blind spots are accurately identified and compensated. The system further distinguishes between internal fill light equipment (drone's own fill light) and external fill light equipment (ground light source or other drones), optimizes the fill light resource configuration, and improves lighting efficiency. Based on the fill light equipment information, the blind spot lighting power density distribution and blind spot lighting angle (inside and outside the machine) are calculated to ensure that the blind spot is evenly covered by light to avoid excessive lighting or insufficient illumination. It can effectively improve the adaptive ability of the drone intelligent lighting system, realize efficient and accurate dynamic fill light in complex environments, and improve lighting uniformity and energy utilization.

[0130] In an exemplary embodiment, Figure 7 As shown, according to the information of each internal fill-light device and each external fill-light device, the blind spot lighting power density distribution and the blind spot lighting angle corresponding to each main lighting blind spot are calculated, including steps 702 to 710. Among them:

[0131] Step 702: Input the information of each internal fill-light device and each external fill-light device into the fill-light heuristic model to obtain the blind spot fill-light sequence and the blind spot fill-light time.

[0132] Among them, the fill light heuristic model can be an intelligent algorithm framework for optimizing the blind spot fill light solution, which calculates the optimal fill light strategy (time and sequence) by analyzing the power, angle, position, response time and coverage of multiple fill light devices. This model usually combines greedy algorithms, genetic algorithms, ant colony optimization algorithms or reinforcement learning methods to quickly find the optimal solution among multiple fill light solutions.

[0133] The blind spot fill-in order can be the priority arrangement of the fill-in tasks for multiple blind spots, which determines which areas need to be filled in first and which areas can be processed later. The order is optimized based on factors such as the size of the blind spot, the degree of illumination loss, the urgency of the target task, and the endurance of the drone.

[0134] The blind spot fill-in time can be the time when and how long the fill-in light for a specific blind spot starts during the mission to ensure the maximum lighting energy efficiency. This time is calculated based on factors such as the flight speed of the drone, the mission execution time window, and the lighting changes in the blind spot.

[0135] Specifically, the system inputs all the information of internal fill lighting equipment (such as the drone's light source power, gimbal rotation range, and beam angle) and external fill lighting equipment (such as the lighting capacity of the ground light source, installation position, rotation angle, and additional drone support) into the fill lighting heuristic model, which performs optimization calculations based on factors such as blind spot lighting requirements, light source distribution, and equipment energy efficiency.

[0136] During the calculation process, the system first evaluates the availability and response speed of each fill-light device. For example, if the drone's internal fill-light can be quickly adjusted and cover the blind spot, the internal fill-light will be given priority; if the blind spot is close to the ground light source, the external fill-light will be given priority to reduce the drone's energy consumption. Then the system determines the order of fill-light tasks based on task priorities (such as construction lighting needs, security patrol areas, etc.), and uses optimization algorithms (such as genetic algorithms, ant colony optimization algorithms) to calculate the optimal fill-light time for each blind spot to ensure the continuity of lighting coverage and avoid energy waste or light pollution caused by multiple light sources being turned on at the same time.

[0137] Step 704, calculating the blind spot lighting power density distribution corresponding to each main lighting blind spot according to the blind spot fill light sequence and each blind spot fill light time.

[0138] Specifically, the system analyzes the lighting needs of each blind spot in order of priority based on the blind spot fill light order, and evaluates the impact of light sources on the blind spot at different time points in combination with the blind spot fill light time. For example, at a certain time point, only the internal fill light is working, and the light density at that moment is mainly provided by the drone lighting. In the case of overlapping fill light time periods, the composite illumination of the internal and external fill lights needs to be calculated to ensure that the overall light is evenly distributed.

[0139] The system then simulates the light propagation path, combines the light source power output, light attenuation model and light reflectivity analysis, and calculates the actual blind spot lighting power density distribution of different blind spots at different times. For example, in open areas, the light energy decays quickly, and higher power density fill light is required; in areas close to buildings or highly reflective surfaces, the fill light power may need to be reduced to avoid excessive lighting or glare problems. In addition, the system also needs to consider the lighting range and beam diffusion angle of the light source to ensure that the light is evenly distributed throughout the blind spot, without causing local overbrightness or darkness due to improper light focusing. The system outputs the lighting power density distribution of each main lighting blind spot.

[0140] Among them, the calculation formula for the lighting power density distribution of each blind spot is:

[0141]

[0142] Among them, PD i (t) is the blind spot lighting power density distribution of the ith main lighting blind spot at time t, S in A collection of in-camera fill light equipment. ex A collection of external fill light equipment, ω k，i (t) is the opening coefficient of the internal fill light device k to the i-th main lighting blind spot, ω m，i (t) is the opening coefficient of the external fill light device m to the i-th main lighting blind spot, ψ k is the light intensity reference value of the internal fill light device k, η m ψ k is the light intensity reference value of the external fill light device m, f(θ k，i (t), d k，i (t)) is the light attenuation and angle distribution function of the internal fill light device k, g(φ m，i (t), h m，i (t)) is the light attenuation and angle distribution function of the built-in fill light device m, θ k，i (t) is the initial fill light angle of the built-in fill light device k to the i-th main lighting blind spot, φ m，i(t) is the initial fill light angle of the external fill light device m to the i-th main lighting blind spot, d k，i (t) is the distance from the in-camera fill light device k to the center of the blind spot of the i-th main lighting blind spot, φ m，i (t) is the distance from the i-th main lighting blind spot to the center of the blind spot of the external fill light device m.

[0143] Step 706, according to the blind spot fill-in order, the fill-in time of each blind spot and the drone posture information, calculate the optimal internal fill-in angles and the optimal external fill-in angles corresponding to any main lighting blind spot.

[0144] Among them, the optimal internal fill light angle can be the optimal beam direction of the fill light device carried by the drone itself without the fill light overlap detection, ensuring that the light can accurately illuminate the blind spot and form the best lighting coordination with other light sources. When calculating this angle, the system will analyze the current posture of the drone (altitude, pitch angle, yaw angle, roll angle), the location and size of the blind spot, the beam diffusion characteristics and other parameters.

[0145] Among them, the optimal external fill light angle can be the best projection direction of the fill light beam provided by the ground light source or other drones that have not undergone fill light overlap detection, ensuring the synergy between multiple light sources, making the lighting coverage uniform and without blind spots. When calculating this angle, the system will consider factors such as the fixity of the fill light source, the rotation range, the beam diffusion angle, and the location of the target blind spot.

[0146] Specifically, the system ensures that the light source is adjusted to the optimal illumination angle at the appropriate time based on the order of fill-in lighting in the blind spot and the fill-in lighting time. For example, if a drone needs to provide fill-in lighting for a specific blind spot at time T1, and another drone or ground light source needs to intervene to provide fill-in lighting at time T2, the system needs to calculate the independent fill-in lighting angle of the drone at time T1, and recalculate the coordinated angle of all fill-in lighting sources at time T2.

[0147] On this basis, the system combines the drone's posture information (including real-time position, altitude, heading, pitch angle, yaw angle, and roll angle) to calculate the optimal rotation angle of each in-flight fill-in device. For example, if the drone's flight altitude is higher than the target blind spot, the fill-in angle needs to be tilted downward; if the drone needs to fill in the blind spot during lateral flight, the system needs to calculate the composite adjustment of the pitch angle and the yaw angle so that the light beam can always stably cover the blind spot and obtain the optimal in-flight fill-in angle.

[0148] For external fill-in lighting equipment (such as ground light sources or other drone fill-in lighting equipment), the system needs to calculate the optimal illumination direction based on the installation position, height, beam angle, rotation range and other parameters of each fill-in lighting equipment. For example, if there is an obstruction between the ground light source and the blind spot, the system may need to adjust the incident angle of the beam or the position of the light source to optimize the lighting effect and obtain the optimal external fill-in lighting angles.

[0149] Step 708: In any main lighting blind spot, fill light overlap detection is performed on the corresponding main lighting blind spot according to each optimal internal fill light angle and each optimal external fill light angle to obtain fill light overlap data.

[0150] Among them, fill light overlap detection can be the process of analyzing whether there is excessive overlap between the lighting areas of different light sources, ensuring uniform light distribution while avoiding energy waste or light pollution. The system will spatially model the beam direction, power density, and projection range of all fill light sources, and calculate the lighting overlap area using ray tracing algorithms or radiosity analysis.

[0151] Among them, the fill light overlap data can be a quantitative analysis result after the illumination overlap area is detected, including information such as illumination intensity distribution, the area of ​​the light beam overlap area, and the identification of areas with excessive brightness.

[0152] Specifically, for any main lighting blind spot, the system simulates the lighting distribution based on each optimal internal fill light angle and each optimal external fill light angle, that is, according to each optimal internal fill light angle and each optimal external fill light angle, the illumination area of ​​all internal and external light sources is spatially mapped, and the overlapping area between different light sources is analyzed. For example, if a blind spot is filled with light from two light sources at the same time, and the total light intensity exceeds the set standard, it may cause the area to be too bright, and the fill light angle needs to be adjusted.

[0153] At the same time, the system analyzes the cross-projection of light sources to ensure that the light beams do not interfere with each other. For example, if a blind spot is illuminated by two light sources from different angles, it may cause a shadow area. Therefore, the system needs to optimize the direction of the light beams so that the two beams can evenly cover the blind spot without causing obvious cross-interference of light. The fill light overlap data is generated based on the analysis data of the above two.

[0154] Step 710, adjusting each optimal internal fill light angle and each optimal external fill light angle according to the fill light overlap data, to obtain each internal blind spot lighting angle and each external blind spot lighting angle.

[0155] Specifically, for any main lighting blind spot, the system analyzes the illumination intensity of the fill light overlapping data to determine whether it is necessary to reduce the power of certain light sources or adjust the direction of the light beam. For example, if the light density in a certain area is too high, the system may reduce the brightness of the fill light source inside the machine, and appropriately adjust the direction of the fill light source outside the machine to make the light evenly distributed. The system further optimizes the beam diffusion angle by adjusting the fill light angle of different devices to ensure uniform coverage of the illuminated area. For example, in some areas, the light may be too concentrated, resulting in excessive brightness. The system will adjust the opening angle of the light beam to make the light spread more evenly; in other areas, the light beam may need to be narrower to ensure long-distance lighting effects, and the system will appropriately reduce the beam diffusion angle.

[0156] Finally, according to the above analysis results, the optimal internal fill light angles and the optimal external fill light angles are adjusted respectively to meet the requirements of the above analysis results. The adjusted data are used as the internal blind spot lighting angles and the external blind spot lighting angles, and are sent to the UAV and ground light source as the final control command, so that all lighting equipment can work together to achieve efficient and accurate lighting fill light, ensuring the uniformity of lighting in the target scene and the optimal energy utilization rate.

[0157] Among them, the calculation formula for each blind spot lighting angle is:

[0158]

[0159] in, is the optimal fill light angle of the internal fill light device k, is the optimal fill light angle of the external fill light device m, PD i (θ k，i (t)) is the initial fill light angle θ of the internal fill light device k k，i The power density distribution of the i-th main lighting blind area under (t), PD i (φ m，i (t)) is the initial fill light angle φ of the external fill light device m m，i (t) is the power density distribution of the ith main lighting blind area, Λ(·) is the power density-benefit mapping function, γ drone (Δ pos (t)) is the offset of the drone's posture information Δ pos (t) penalty, γ drone is the adjustment cost coefficient of the UAV posture information, γ external (Δ deploy (t)) is the offset of the external device posture information Δ deploy (t) penalty, γ external is the adjustment cost coefficient of the external fill light equipment, Overlap(·) is the overlap detection function, ΔΘ k，iis the correction value of the internal fill light device k, ΔΘ m，i is the correction value of the external fill light device m, is the in-camera blind spot lighting angle of the in-camera fill light device k, is the external blind spot lighting angle of the external fill light device m, and R[·] is the overlap correction function.

[0160] In this embodiment, by utilizing the fill light heuristic model and combining the information of the internal fill light equipment and the external fill light equipment, the order and time of fill light in the blind spot are intelligently calculated to ensure the efficient execution of the fill light task. By optimizing the blind spot lighting power density distribution, the system accurately matches the lighting needs of each area to avoid energy waste. Combined with the drone posture information, the optimal internal fill light angle and the optimal external fill light angle are calculated to ensure that the light beam accurately covers the blind spot. The system further performs fill light overlap detection, analyzes the overlap of light beams, avoids excessive illumination or light pollution, and dynamically adjusts the lighting angle according to the fill light overlap data to make the fill light more intelligent. It can effectively improve the accuracy, adaptability and energy efficiency of the drone lighting system, and ensure efficient and uniform light coverage in complex environments.

[0161] It should be understood that, although the steps in the flowcharts involved in the embodiments described above are displayed sequentially as indicated by the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows.

[0162] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 8 The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface.

[0163] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 8 The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device.

[0164] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0165] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0166] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the above-mentioned method embodiments.

[0167] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A UAV lighting control method based on three-dimensional model data, characterized in that: The method comprises: Acquire three-dimensional model data, environmental perception data, and flight positioning data of a target scene; wherein a lighting drone is flying in the target scene; Planning a flight path of the UAV in the target scene according to the three-dimensional model data, the environmental perception data, and the flight positioning data to obtain flight path planning data; According to the flight path planning data and the environmental perception data, a lighting analysis is performed on the main lighting end of the UAV to obtain a main lighting power density distribution and a main lighting angle; Determining each main lighting blind area in the target scene according to the main lighting power density distribution and the main lighting angle; According to each of the main lighting blind spots and the environmental perception data, a fill light analysis is performed on each of the main lighting blind spots to obtain a lighting power density distribution of each blind spot and a lighting angle of each blind spot; UAV lighting control information is generated according to the main lighting power density distribution, the main lighting angle, the blind spot lighting power density distribution and the blind spot lighting angle.

2. The method according to claim 1, characterized in that The step of performing lighting analysis on the main lighting end of the drone according to the flight path planning data and the environmental perception data to obtain the main lighting power density distribution and the main lighting angle includes: Determining scene lighting requirement information of the target scene according to the flight path planning data and the environment perception data; According to the scene lighting demand information, performing lighting analysis on the main lighting end of the drone to obtain the main lighting power density distribution; And, according to the scene lighting requirement information and the drone posture information, the servo offset of the main lighting end is calculated to obtain the main lighting angle.

3. The method according to claim 2, characterized in that The performing lighting analysis on the main lighting end of the drone according to the scene lighting demand information to obtain the main lighting power density distribution includes: Detecting ambient light compensation information of the target scene according to the environmental perception data; Determining actual scene lighting information of the target scene according to a difference between the scene lighting requirement information and the ambient light compensation information; According to the three-dimensional model data and the actual lighting information of the scene, a ray tracing analysis is performed on the target scene to obtain the main lighting power density distribution.

4. The method according to claim 2, characterized in that: The calculating the steering gear offset of the main lighting end according to the scene lighting requirement information and the drone posture information to obtain the main lighting angle includes: Determine the lighting space orientation angle data of the main lighting end according to the scene lighting demand information and the drone posture information; Compensating the lighting space orientation angle data according to the environmental perception data of the target scene to obtain the lighting space orientation angle; In the case where an abnormal flying object appears in the environmental perception data, predicting a flight path of the abnormal flying object according to the flight state information of the abnormal flying object; According to the flight path of the flying object, the orientation angle of the lighting space is optimized to obtain the main lighting angle.

5. The method according to claim 2, characterized in that: The calculation formula of the main lighting power density distribution is: Among them, E actual (P) is the main lighting power density distribution, E env (P) is the ambient light compensation information, P is the coordinate in the target scene where the illumination needs to be calculated, l is the lth main lighting coordinate, I l (θ l +Δθ comp (P)) is the main lighting angle correction data Δθ comp (P) Corrected light intensity, θ l The direction angle defined by the main lighting's own luminosity curve, d l is the distance between coordinate point P and coordinate point l, φ l is the angle between the surface normal at coordinate point P and the main lighting direction, Δφ adj (P) is the additional angle correction of the drone’s posture information, V(P, L l ) is the visibility function between coordinate point P and coordinate point k, α(R env (P)) is the comprehensive attenuation or amplification factor, R env (P) is the measurement parameter of the local environment of the coordinate point P, and α(·) is the dynamic adjustment function; The calculation formula of the main lighting angle is: in, For the main lighting angle, is the target angle vector, θ drone (t) is the attitude angle of the UAV itself, β avoid (t) is the avoidance angle compensation, Δ SLAM (t) is the fine-tuning amount generated by visual feedback, W(t) is the lighting equipment coordination weight, and F(·) is the multidimensional mapping control strategy function.

6. The method according to claim 1, characterized in that The method of performing fill light analysis on each of the main lighting blind areas according to each of the main lighting blind areas and the environmental perception data to obtain lighting power density distribution of each blind area and lighting angle of each blind area includes: Calculating fill light requirement information of each blind spot according to each of the main lighting blind spots and the environmental perception data; According to the fill light demand information of each blind area, respectively determine the fill light equipment information of each internal machine and the fill light equipment information of each external machine; According to the information of each of the internal fill-light devices and the information of each of the external fill-light devices, the blind spot lighting power density distribution and the blind spot lighting angle corresponding to each of the main lighting blind spots are calculated; the blind spot lighting angle includes the internal blind spot lighting angle and the external blind spot lighting angle.

7. The method according to claim 6, characterized in that The calculating, according to each of the internal fill-light equipment information and each of the external fill-light equipment information, the blind spot lighting power density distribution corresponding to each of the main lighting blind spots and the blind spot lighting angle, comprises: Inputting each of the internal fill-light device information and each of the external fill-light device information into the fill-light heuristic model to obtain a blind spot fill-light sequence and a blind spot fill-light time; Calculating the blind spot lighting power density distribution corresponding to each of the main lighting blind spots according to the blind spot fill light sequence and each of the blind spot fill light time; Calculate each optimal internal fill light angle and each optimal external fill light angle corresponding to any of the main lighting blind spots according to the blind spot fill light sequence, each blind spot fill light time and the drone posture information; In any of the main lighting blind spots, according to each of the optimal internal fill light angles and each of the optimal external fill light angles, fill light overlap detection is performed on the corresponding main lighting blind spot to obtain fill light overlap data; According to the fill light overlap data, the optimal internal fill light angles and the optimal external fill light angles are adjusted to obtain the internal blind spot lighting angles and the external blind spot lighting angles.

8. The method according to claim 7, characterized in that The calculation formula for the blind spot lighting power density distribution is: Among them, PD i (t) is the blind spot lighting power density distribution of the ith main lighting blind spot at time t, S in It is a collection of in-camera fill light equipment. ex A collection of external fill light equipment, ω k，i (t) is the opening coefficient of the internal fill light device k to the i-th main lighting blind spot, ω m，i (t) is the opening coefficient of the external fill light device m to the i-th main lighting blind spot, ψ k is the light intensity reference value of the internal fill light device k, η m ψ k is the light intensity reference value of the external fill light device m, f(θ k，i (t), d k，i (t)) is the light attenuation and angle distribution function of the internal fill light device k, g(φ m，i (t), h m，i (t)) is the light attenuation and angle distribution function of the built-in fill light device m, θ k，i (t) is the initial fill light angle of the built-in fill light device k to the i-th main lighting blind spot, φ m，i (t) is the initial fill light angle of the external fill light device m to the i-th main lighting blind spot, d k，i (t) is the distance from the in-camera fill light device k to the center of the blind spot of the ith main lighting blind spot, φ m，i (t) is the distance from the i-th main lighting blind spot to the center of the blind spot of the external fill light device m; The calculation formula for each blind spot lighting angle is: in, is the optimal fill light angle of the internal fill light device k, is the optimal fill light angle of the external fill light device m, PD i (θ k，i (t)) is the initial fill light angle θ of the internal fill light device k k，i The power density distribution of the i-th main lighting blind area under (t), PD i (φ m，i (t)) is the initial fill light angle φ of the external fill light device m m，i (t) is the power density distribution of the ith main lighting blind area, Λ(·) is the power density-benefit mapping function, γ drone (Δ pos (t)) is the offset of the drone's posture information Δ pos (t) penalty, γ drone is the adjustment cost coefficient of the UAV posture information, γ external (Δ deploy (t)) is the offset of the external device posture information Δ deploy (t) penalty, γ external is the adjustment cost coefficient of the external fill light equipment, Overlap(·) is the overlap detection function, ΔΘ k，i is the correction value of the internal fill light device k, ΔΘ m，i is the correction value of the external fill light device m, is the in-camera blind spot lighting angle of the in-camera fill light device k, is the external blind spot lighting angle of the external fill light device m, and R[·] is the overlap correction function.

9. A UAV lighting control system based on three-dimensional model data, characterized in that: The system includes a computer, a drone and a data interaction terminal; The drone is used to obtain three-dimensional model data, environmental perception data and flight positioning data of the target scene; there is a lighting drone flying in the target scene; The computer is used to plan a flight path of the UAV in the target scene according to the three-dimensional model data, the environmental perception data and the flight positioning data to obtain flight path planning data; The computer is used to perform lighting analysis on the main lighting end of the UAV according to the flight path planning data and the environmental perception data to obtain the main lighting power density distribution and the main lighting angle; The computer is used to determine each main lighting blind area in the target scene according to the main lighting power density distribution and the main lighting angle; The computer is used to perform fill light analysis on each of the main lighting blind areas according to each of the main lighting blind areas and the environmental perception data, so as to obtain the lighting power density distribution of each blind area and the lighting angle of each blind area; The computer is used to generate the UAV lighting control information according to the main lighting power density distribution, the main lighting angle, the blind spot lighting power density distribution and the blind spot lighting angle; The data interaction terminal is used for data transmission between the computer and the drone.

10. The system according to claim 9, characterized in that The computer further comprises a processor, and when the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images