Unmanned aerial vehicle path planning system based on environment dynamic updating
Through the path planning module and real-time environment acquisition subsystem combined with deep learning and potential field method, the drone can independently adjust the flight route in a dynamic environment, solving the problem that drones cannot avoid obstacles, and improving flight efficiency and safety.
Patent Information
- Application Number
- CN202510606584.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-12
- Publication Date
- 2025-08-12
AI Technical Summary
The existing drone flight route planning technology cannot avoid temporary obstacles independently and requires manual adjustment, resulting in low flight efficiency and easy damage.
The path planning module, real-time environment acquisition subsystem and path update subsystem are adopted to collect environmental data in real time through infrared sensing and imaging equipment, use deep learning technology to identify obstacles, and adjust the drone's flight route through the potential field method to achieve dynamic path update.
Synchronous adjustment of flight routes in real-time environmental changes has improved drone flight efficiency, reduced flight time and avoided obstacle collisions, ensuring flight smoothness and safety.
Smart Images
Figure BDA0005398389100000031 
Figure BDA0005398389100000041 
Figure BDA0005398389100000042
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) planning, and in particular to a UAV path planning system based on dynamic environment updating. Background Art
[0002] A drone is an unmanned aircraft controlled by radio remote control equipment and its own program control device. It has no cockpit but is equipped with autopilot, program control device and other equipment. Personnel on the ground, on ships or at the mother aircraft remote control station use radar and other equipment to track, locate, remotely control, telemeter and transmit digital data to it. It can take off like an ordinary aircraft under radio remote control or be launched into the air with a booster rocket, or it can be carried into the air by a mother aircraft and released for flight. When recovered, it can land automatically in the same way as an ordinary aircraft, or it can be recovered by remote control with a parachute or net, and can be used repeatedly. It is widely used in aerial reconnaissance, surveillance, communication, anti-submarine, electronic jamming, etc.
[0003] Unrestricted by ground transportation, drones can fly directly from their origin to their destination, enabling them to complete long-distance transport missions in record time. For example, a Hunan highway maintenance drone completes a round-trip in just three minutes. Yunnan power grid drones reduce delivery times from two to three hours to just two or three minutes. Mizhi tree-planting drones complete a round-trip in one minute. Drones are relatively low-cost to operate, operating on batteries, requiring no fuel and requiring minimal maintenance. Furthermore, their high degree of automation reduces labor costs, further enhancing economic efficiency. Drones can operate in a variety of environments, including urban, rural, and remote areas, quickly adapting to diverse transport needs and meeting diverse customer requirements. For example, they can be quickly deployed to transport emergency supplies after natural disasters. They are also effective in complex terrain, such as highways, highways, and power generation sites nestled in mountainous terrain. As a green transportation tool, drones offer environmental advantages due to their low carbon emissions, helping to reduce carbon footprints and promote the development of green logistics.
[0004] Currently, the flight route planning technology for drones is relatively mature. Once the starting and end points are determined, the drone can fly along a fixed path. However, during the flight, the drone cannot autonomously avoid temporary obstacles or change the flight path. Manual adjustments are required to the drone's position before it can continue flying. Human intervention is still required during the process, which greatly delays the drone's flight efficiency and can easily cause damage to the drone. Summary of the Invention
[0005] In view of the shortcomings of the existing technology, the present invention provides a UAV path planning system based on dynamic environment update, which solves the problems raised in the above background technology.
[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a UAV path planning system based on dynamic environment update, including a path planning module, a real-time environment acquisition subsystem, a path update subsystem, and a UAV control module;
[0007] The user sets the initial parameters of the drone's flight route through the path planning module. At the same time, the environmental data collected during the flight are collected and captured in real time through the implementation of the environmental acquisition subsystem, and the environmental data are processed and analyzed. The path update subsystem re-plans the drone's flight route based on the processed environmental data and transmits it to the drone's own control module via wireless signals to perform corresponding flight control operations on the drone. In this way, the drone's flight route can be adjusted synchronously according to the real-time updated environmental changes during the flight.
[0008] The path planning module is used to set the initial flight path of the drone and enter path information, including the input of parameters such as takeoff mark point, landing mark point, flight altitude limit, flight route, and flight time. The flight path is planned and limited based on multiple flight parameters, multiple flight routes are generated for comparison, and the corresponding route is selected as the flight path according to flight requirements.
[0009] The real-time environment acquisition subsystem uses infrared sensors and cameras to capture video of the external environment while the drone is flying along the planned path, collects video and image data, and analyzes the data. This data is used to build a model and serve as the basis for the path update subsystem to identify obstacles in the drone's flight path map.
[0010] The path update subsystem changes / adjusts the UAV's flight path based on the feature information and model data acquired by the real-time environment acquisition subsystem, preventing the UAV from stopping or colliding while maintaining smooth flight, thereby maintaining the UAV's optimal flight path in real time.
[0011] The drone control module, including the main control module, signal conditioning and interface module, servo drive module and signal interaction module, is used to control the flight and model of the drone, maintain the flight stability of the drone, receive remote control signals in real time, and feedback the data collected during the flight.
[0012] Optionally, the flight requirements include flight time cost, flight distance cost, flight energy consumption cost, flight threat cost, and flight turning angle cost.
[0013] Optionally, the obstacle threat area set in the flight environment is recorded as K, and the obstacle threat cost penalty coefficient is recorded as γ c, the cost function F2(X i ) is calculated as follows:
[0014]
[0015] Optionally, the real-time environment acquisition subsystem includes a video image processing module and a model generation module.
[0016] Optionally, the video image processing module uses deep learning technology to classify and identify videos, quickly identify the content in the video, and provide a basis for subsequent processing and analysis; by training the deep learning model, it is able to recognize objects, scenes, and facial data in the video, thereby achieving video classification and recognition; and analyzes and predicts the motion trajectory of objects in the video, while reconstructing low-resolution videos into high-resolution videos, thereby improving the resolution and clarity of the video;
[0017] The model generation module describes the content and structure of video data by extracting key features from the video data, including color, texture, shape, and dynamic features. Based on the feature extraction, the video data analysis model uses a deep learning algorithm to establish a convolutional neural network model, divides the data into training set, validation set, and test set, and uses the training data to train the model; after the model is established, the model is evaluated to determine the performance and generalization ability of the model.
[0018] Optionally, the specific process of the path updating subsystem is as follows:
[0019] (1) Get the current position of the UAV and reinitialize the starting point marker of the flight path;
[0020] (2) Synchronously obtain the surrounding environment information, including the location and range of obstacles;
[0021] (3) Calculate the potential field based on the positions of the target point and the obstacle, and find the gradient of the potential field;
[0022] (4) Determine whether the UAV has reached the landing mark point. If it has, end the path planning;
[0023] (5) If the drone encounters an obstacle, the potential field is adjusted according to the location and range of the obstacle to generate repulsive force to avoid collision between the drone and the obstacle;
[0024] (6) If the drone does not reach the landing mark, repeat the above steps until the drone reaches the landing mark.
[0025] Optionally, the path updating subsystem adopts a potential field method to regard the surrounding environment as a potential field, and uses the gradient of the potential field to plan the path of the UAV;
[0026] The drone's target point is marked as an attractor, and obstacles are marked as repulsors. The balance point is found between the attractor and repulsor.
[0027] The potential field of the target point is specifically:
[0028] U goal (x,y)=k*((xx goal ) 2 +(yy goal ) 2 )
[0029] Where k is a constant, (x goal ,y goal ) are the coordinates of the target point;
[0030] The potential field of the obstacle is specifically:
[0031]
[0032] Among them, k obs is a constant, d obs is the distance from the UAV to the obstacle, f(d obs ) is the obstacle correction function;
[0033]
[0034] Among them, d0 is the range of the obstacle.
[0035] Optionally, the main control module is used to complete high-precision acquisition of multiple analog signals, including gyro signals, heading signals, rudder angle signals, engine speed, cylinder temperature signals, dynamic and static pressure sensor signals, and power supply voltage signals; and utilize multiple communication channels to respectively realize communication with the airborne data terminal, GPS signals, and digital sensors;
[0036] The signal conditioning and interface module conditions the sensor input signal to make it meet the input requirements of the main control module and provides an interface with other devices;
[0037] The servo drive module drives the servo according to the control signal output by the main control module, changes the aircraft's wing surface or controls the speed of each axis blade, thereby controlling the attitude of the UAV;
[0038] The UAV control module collects flight status data measured by various sensors in real time, receives control commands and data transmitted by the radio measurement and control terminal from the uplink channel of the ground measurement and control station, and outputs control instructions to the actuator after calculation and processing, thereby realizing the control of various flight modes in the UAV and the management and control of mission equipment.
[0039] The present invention provides a UAV path planning system based on dynamic environment update, which has the following features:
[0040] Beneficial effects:
[0041] In this drone path planning system based on dynamic environmental updates, users set the initial relevant parameters of the drone's flight route through the path planning module. At the same time, the environmental data during the flight process is collected and captured in real time by implementing the environment acquisition subsystem, and the environmental data is processed and analyzed. The path update subsystem replans and designs the drone's flight route based on the processed environmental data content, and transmits it to the drone's own control module via wireless signals to perform corresponding flight control operations on the drone. In this way, the drone's flight route can be synchronously adjusted according to the real-time updated environmental changes during the flight of the drone, achieving more efficient drone flight operations, reducing the drone's flight time, and avoiding the drone being stranded due to obstacles during flight. DETAILED DESCRIPTION
[0042] The technical solutions in the embodiments of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.
[0043] The present invention provides a technical solution: a UAV path planning system based on dynamic environment update, comprising a path planning module, a real-time environment acquisition subsystem, a path update subsystem, and a UAV control module;
[0044] The user sets the initial parameters of the drone's flight route through the path planning module. At the same time, the environmental data collected during the flight are collected and captured in real time through the implementation of the environmental acquisition subsystem, and the environmental data are processed and analyzed. The path update subsystem re-plans the drone's flight route based on the processed environmental data and transmits it to the drone's own control module via wireless signals to perform corresponding flight control operations on the drone. In this way, the drone's flight route can be adjusted synchronously according to the real-time updated environmental changes during the flight.
[0045] The path planning module is used to set the initial flight path of the drone and enter path information, including the input of parameters such as takeoff mark point, landing mark point, flight altitude limit, flight route, and flight time. The flight path is planned and limited based on multiple flight parameters, multiple flight routes are generated for comparison, and the corresponding route is selected as the flight path according to flight requirements.
[0046] Flight requirements include flight time cost, flight distance cost, flight energy cost, flight threat cost, and flight turning angle cost.
[0047] In the flight threat cost, the obstacle threat area set in the flight environment is recorded as K, and the obstacle threat cost penalty coefficient is recorded as γ c , the cost function F2(X i ) is calculated as follows:
[0048]
[0049] The real-time environment acquisition subsystem uses infrared sensors and cameras to capture video of the external environment while the drone is flying along the planned path, collects video and image data, and analyzes the data. This data is used to build a model and serve as the basis for the path update subsystem to identify obstacles in the drone's flight path map.
[0050] The real-time environment acquisition subsystem includes a video image processing module and a model generation module;
[0051] The video image processing module uses deep learning technology to classify and identify videos, quickly identifying the content in the video and providing a basis for subsequent processing and analysis. By training the deep learning model, it can identify objects, scenes, and facial data in the video, thereby achieving video classification and recognition. It also analyzes and predicts the motion trajectory of objects in the video, and reconstructs low-resolution videos into high-resolution videos, improving the resolution and clarity of the video.
[0052] The model generation module describes the content and structure of video data by extracting key features from the video data, including color, texture, shape, and dynamic features. Based on the feature extraction, the video data analysis model uses a deep learning algorithm to build a convolutional neural network model, divides the data into training sets, validation sets, and test sets, and uses the training data to train the model. After the model is built, the model is evaluated to determine the performance and generalization ability of the model.
[0053] The path update subsystem changes / adjusts the UAV's flight path based on the feature information and model data acquired by the real-time environment acquisition subsystem, preventing the UAV from stopping or colliding while maintaining smooth flight, thereby maintaining the UAV's optimal flight path in real time.
[0054] The specific process of the path update subsystem is as follows:
[0055] (1) Get the current position of the UAV and reinitialize the starting point marker of the flight path;
[0056] (2) Synchronously obtain the surrounding environment information, including the location and range of obstacles;
[0057] (3) Calculate the potential field based on the positions of the target point and the obstacle, and find the gradient of the potential field;
[0058] (4) Determine whether the UAV has reached the landing mark point. If it has, end the path planning;
[0059] (5) If the drone encounters an obstacle, the potential field is adjusted according to the location and range of the obstacle to generate repulsive force to avoid collision between the drone and the obstacle;
[0060] (6) If the drone does not reach the landing mark, repeat the above steps until the drone reaches the landing mark;
[0061] The path update subsystem uses the potential field method to treat the surrounding environment as a potential field and uses the gradient of the potential field to plan the path of the UAV;
[0062] The drone's target point is marked as an attractor, and obstacles are marked as repulsors. The balance point is found between the attractor and repulsor.
[0063] The potential field of the target point is specifically:
[0064] U goal (x,y)=k*((xx goal ) 2 +(yy goal ) 2 )
[0065] Where k is a constant, (x goal ,y goal ) are the coordinates of the target point;
[0066] The potential field of the obstacle is specifically:
[0067]
[0068] Among them, k obs is a constant, d obs is the distance from the UAV to the obstacle, f(d obs ) is the obstacle correction function;
[0069]
[0070] Among them, d0 is the range of the obstacle;
[0071] The drone control module, including the main control module, signal conditioning and interface module, servo drive module, and signal interaction module, is used to control the drone's flight and model, maintain the drone's flight stability, receive remote control signals in real time, and provide feedback on data collected during flight.
[0072] The main control module is used to complete the high-precision acquisition of multiple analog signals, including gyro signals, heading signals, rudder angle signals, engine speed, cylinder temperature signals, dynamic and static pressure sensor signals, and power supply voltage signals; it uses multiple communication channels to respectively communicate with the airborne data terminal, GPS signals, and digital sensors;
[0073] The signal conditioning and interface module conditions the sensor input signal to make it meet the input requirements of the main control module and provides an interface with other devices;
[0074] The servo drive module drives the servo according to the control signal output by the main control module, changes the aircraft's wing surface or controls the speed of each axis blade, thereby controlling the attitude of the UAV;
[0075] The UAV control module collects flight status data measured by various sensors in real time, receives control commands and data transmitted by the radio measurement and control terminal from the uplink channel of the ground measurement and control station, and outputs control instructions to the actuator after calculation and processing, thereby realizing the control of various flight modes in the UAV and the management and control of mission equipment.
[0076] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.
Claims
1. A UAV path planning system based on dynamic environment update, characterized in that: Including path planning module, real-time environment acquisition subsystem, path update subsystem, and UAV control module; The user sets the initial parameters of the drone's flight route through the path planning module. At the same time, the environmental data collected during the flight are collected and captured in real time through the implementation of the environmental acquisition subsystem, and the environmental data are processed and analyzed. The path update subsystem re-plans the drone's flight route based on the processed environmental data and transmits it to the drone's own control module via wireless signals to perform corresponding flight control operations on the drone. In this way, the drone's flight route can be adjusted synchronously according to the real-time updated environmental changes during the flight. The path planning module is used to set the initial flight path of the drone and enter path information, including the input of parameters such as takeoff mark point, landing mark point, flight altitude limit, flight route, and flight time. The flight path is planned and limited based on multiple flight parameters, multiple flight routes are generated for comparison, and the corresponding route is selected as the flight path according to flight requirements. The real-time environment acquisition subsystem uses infrared sensors and cameras to capture video of the external environment while the drone is flying along the planned path, collects video and image data, and analyzes the data. This data is used to build a model and serve as the basis for the path update subsystem to identify obstacles in the drone's flight path map. The path update subsystem changes / adjusts the UAV's flight path based on the feature information and model data acquired by the real-time environment acquisition subsystem, preventing the UAV from stopping or colliding while maintaining smooth flight, thereby maintaining the UAV's optimal flight path in real time. The drone control module, including the main control module, signal conditioning and interface module, servo drive module and signal interaction module, is used to control the flight and model of the drone, maintain the flight stability of the drone, receive remote control signals in real time, and feedback the data collected during the flight.
2. The UAV path planning system based on dynamic environment update according to claim 1, characterized in that: The flight requirements include flight time cost, flight distance cost, flight energy consumption cost, flight threat cost and flight turning angle cost.
3. The UAV path planning system based on dynamic environment update according to claim 2, characterized in that: In the flight threat cost, the obstacle threat area set in the flight environment is recorded as K, and the obstacle threat cost penalty coefficient is recorded as γ c , the cost function F2(X i ) is calculated as follows:
4. The UAV path planning system based on dynamic environment update according to claim 1, characterized in that: The real-time environment acquisition subsystem includes a video image processing module and a model generation module.
5. The UAV path planning system based on dynamic environment update according to claim 4, characterized in that: The video image processing module uses deep learning technology to classify and identify videos, quickly identifying the content in the video and providing a basis for subsequent processing and analysis. By training the deep learning model, it can identify objects, scenes, and facial data in the video, thereby achieving video classification and recognition. It also analyzes and predicts the motion trajectory of objects in the video, and reconstructs low-resolution videos into high-resolution videos, improving the resolution and clarity of the video. The model generation module extracts key features from video data, including color, texture, shape, and dynamic features, to describe the content and structure of the video data. Based on the feature extraction, the video data analysis model uses a deep learning algorithm to build a convolutional neural network model, divides the data into training, validation, and test sets, and uses the training data to train the model. After the model is built, it is evaluated to determine the performance and generalization ability of the model.
6. The UAV path planning system based on dynamic environment update according to claim 1, characterized in that: The specific process of the path update subsystem is as follows: (1) Get the current position of the UAV and reinitialize the starting point marker of the flight path; (2) Synchronously obtain the surrounding environment information, including the location and range of obstacles; (3) Calculate the potential field based on the positions of the target point and the obstacle, and find the gradient of the potential field; (4) Determine whether the UAV has reached the landing mark point. If it has, end the path planning; (5) If the drone encounters an obstacle, the potential field is adjusted according to the location and range of the obstacle to generate repulsive force to avoid collision between the drone and the obstacle; (6) If the drone does not reach the landing mark, repeat the above steps until the drone reaches the landing mark.
7. The UAV path planning system based on dynamic environment update according to claim 1, characterized in that: The path update subsystem adopts the potential field method to regard the surrounding environment as a potential field, and uses the gradient of the potential field to plan the path of the UAV; The drone's target point is marked as an attractor, and obstacles are marked as repulsors. The balance point is found between the attractor and repulsor. The potential field of the target point is specifically: OR goal (x,y)=k*((xx goal ) 2 +(yy goal ) 2 ) Where k is a constant, (x goal ,y goal ) are the coordinates of the target point; The potential field of the obstacle is specifically: Among them, k obs is a constant, d obs is the distance from the UAV to the obstacle, f(d obs ) is the obstacle correction function; Among them, d0 is the range of the obstacle.
8. The UAV path planning system based on dynamic environment update according to claim 1, characterized in that: The main control module is used to complete the high-precision acquisition of multiple analog signals, including gyro signals, heading signals, rudder angle signals, engine speed, cylinder temperature signals, dynamic and static pressure sensor signals, and power supply voltage signals; and uses multiple communication channels to respectively communicate with the airborne data terminal, GPS signals, and digital sensors; The signal conditioning and interface module conditions the sensor input signal to make it meet the input requirements of the main control module and provides an interface with other devices; The servo drive module drives the servo according to the control signal output by the main control module, changes the aircraft's wing surface or controls the speed of each axis blade, thereby controlling the attitude of the UAV; The UAV control module collects flight status data measured by various sensors in real time, receives control commands and data transmitted by the radio measurement and control terminal from the uplink channel of the ground measurement and control station, and outputs control instructions to the actuator after calculation and processing, thereby realizing the control of various flight modes in the UAV and the management and control of mission equipment.