CN-121979239-A - Automatic-throwing search and rescue system and method for amphibious unmanned aerial vehicle

CN121979239ACN 121979239 ACN121979239 ACN 121979239ACN-121979239-A

Abstract

The invention relates to an air-water amphibious unmanned aerial vehicle self-throwing search and rescue system and method, which belong to the technical field of on-water emergency rescue, and comprise an unmanned aerial vehicle self-throwing search and rescue module and an intelligent emergency control module, wherein the unmanned aerial vehicle self-throwing search and rescue module is used for collecting image data and point cloud data of a search and rescue area in real time according to a search and rescue task and transmitting the image data and the point cloud data to the intelligent emergency control module, and the intelligent emergency control module is used for identifying a search and rescue target based on the image data and the point cloud data, tracking the identified search and rescue target to obtain a target pose, planning a flight track of the unmanned aerial vehicle self-throwing search and rescue module based on the target pose and the acquired pose, controlling the unmanned aerial vehicle self-throwing search and rescue module to drop to the periphery of the search and rescue target based on the planned flight track, so that the timeliness and reliability of the air-water amphibious unmanned aerial vehicle self-throwing search and rescue are improved.

Inventors

MA QUANDANG
Cheng Cunxi
ZHAO ZIQI
WU MINGHAN
JIANG XINYI
BAI ZHONGDA
YANG YUHANG
CHEN GUOHAO
FAN YIXUAN
QI JIAYUE
LIU SENHAN

Assignees

武汉理工大学

Dates

Publication Date: 20260505
Application Date: 20260116

Claims (10)

1. The automatic-throwing search and rescue system of the amphibious unmanned aerial vehicle is characterized by comprising an unmanned aerial vehicle automatic-throwing search and rescue module and an intelligent emergency control module; the unmanned aerial vehicle self-throwing search and rescue module is used for collecting image data and point cloud data of a search and rescue area in real time according to a search and rescue task and transmitting the image data and the point cloud data to the intelligent emergency control module; The intelligent emergency control module is used for identifying a search and rescue target based on the image data and the point cloud data, tracking the identified search and rescue target to obtain a target pose, planning the flight track of the unmanned aerial vehicle self-casting search and rescue module based on the target pose and the acquired pose of the unmanned aerial vehicle self-casting search and rescue module, controlling the unmanned aerial vehicle self-casting search and rescue module to land to the periphery of the search and rescue target based on the planned flight track, and realizing the unmanned aerial vehicle self-casting search and rescue of the amphibious unmanned aerial vehicle.
2. The air-water amphibious unmanned aerial vehicle self-throwing and rescuing system according to claim 1, wherein the unmanned aerial vehicle self-throwing and rescuing module comprises a multi-mode camera unit and a communication unit; the multi-mode camera unit is used for collecting image data and point cloud data of a search and rescue area in real time according to a search and rescue task; the communication unit is used for transmitting the image data and the point cloud data to the intelligent emergency control module.
3. The air-water amphibious unmanned aerial vehicle self-throwing and rescuing system according to claim 2, wherein the unmanned aerial vehicle self-throwing and rescuing module further comprises a composite light engine body structure, a hybrid power propulsion unit, an energy and management unit and a propeller protection net; The composite light-weight machine body structure is a concentric circle type life buoy appearance, and the edge of the concentric circle type life buoy appearance comprises an ergonomic concave plate surface for supporting a search and rescue target and an annular functional handle for grasping the search and rescue target; The hybrid propulsion unit is used for switching the aerial high-speed cruising of the unmanned aerial vehicle and flexible maneuvering on the water surface; The energy and management unit is used for providing energy supply for the multi-mode camera unit, the communication unit and the hybrid propulsion unit; The screw protection network is used for providing safety protection for the screw of unmanned aerial vehicle.
4. The air-water amphibious unmanned aerial vehicle self-throwing and rescuing system according to claim 2, wherein the intelligent emergency control module comprises an intelligent analysis unit, a track planning unit and a control unit; The intelligent analysis unit is used for detecting the search and rescue target of the image data by adopting a deep learning model, obtaining the category, the detection frame and the image texture characteristics of the search and rescue target, extracting the point cloud geometric characteristics of the search and rescue target based on the point cloud data, matching the point cloud geometric characteristics with the image texture characteristics, determining the search and rescue target and the search and rescue target pose, and tracking the search and rescue target by adopting Kalman filtering to obtain the target pose, wherein the network architecture of the deep learning model is YOLOv n model; the track planning unit is used for planning the flight track of the unmanned aerial vehicle self-throwing search and rescue module based on the target pose and the acquired pose of the unmanned aerial vehicle self-throwing search and rescue module; the control unit is used for controlling the unmanned aerial vehicle self-throwing search and rescue module to drop to the periphery of the search and rescue target based on the planned flight track, so that the unmanned aerial vehicle self-throwing search and rescue of the amphibious unmanned aerial vehicle is realized.
5. The air-water amphibious unmanned aerial vehicle self-throwing and rescuing system according to claim 4, wherein the YOLOv n model comprises a backbone network, a neck network, dyHead layers and a detection head, and the intelligent analysis unit is specifically configured to: performing multi-scale feature extraction on the image data through the backbone network to obtain a first scale feature, a second scale feature and a third scale feature; Fusing the first scale feature, the second scale feature and the third scale feature through the neck network to obtain fused image features; Enhancing the fused image features through DyHead layers of attention mechanisms, wherein the attention mechanisms comprise a scale-aware attention mechanism, a space-aware attention mechanism and a task-aware attention mechanism; and carrying out category prediction, bounding box regression and targeting prediction on the enhanced image features through a detection head to obtain the category of the target, the detection box and the image texture features.
6. The empty water amphibious unmanned aerial vehicle self-casting search and rescue system of claim 4, wherein the loss function of the deep learning model is: , , , , , , , Wherein, the As a loss function of the deep learning model, For the cross-correlation of the predicted frame with the real frame, In order to be able to achieve a distance loss, In order for the angle to be lost, In order to be the channel height, For the width of the channel it is, Is the first The number of the detection frames is equal to the number of the detection frames, Is the first The number of the detection frames is equal to the number of the detection frames, For predicting frame and true frame The normalized distance in the direction is chosen to be the same, For predicting frame and true frame The normalized distance in the direction is chosen to be the same, As the correlation coefficient for the angle loss, As the correlation coefficient for the angle loss, Is a loss of shape.
7. The empty water amphibious unmanned aerial vehicle self-throwing search and rescue system of claim 4, wherein the intelligent analysis unit is further configured to: Calculating cosine similarity between the point cloud geometric features and the image texture features; when the cosine similarity is greater than or equal to a preset similarity threshold, establishing a corresponding relation between the point cloud geometric features and the image texture features; and fusing the point cloud geometric features and the image texture features of the corresponding relation to determine the search and rescue target and the pose of the search and rescue target.
8. The air-water amphibious unmanned aerial vehicle self-throwing search and rescue system according to claim 4, wherein the trajectory planning unit is specifically configured to: determining a starting point and an ending point of the unmanned aerial vehicle automatic search and rescue module based on the target pose and the pose of the unmanned aerial vehicle automatic search and rescue module; and planning the flight track of the unmanned aerial vehicle self-casting search and rescue module by adopting a cubic B spline smooth curve based on the starting point and the ending point to obtain the planned flight track.
9. The air-water amphibious unmanned aerial vehicle self-casting and rescuing system according to claim 4, wherein the intelligent emergency control module further comprises a communication navigation unit, a wireless charging unit, a multidimensional monitoring unit and a charging pile interface unit; The communication navigation unit is used for carrying out real-time data interaction and centimeter-level positioning with the unmanned aerial vehicle automatic search and rescue module; the wireless charging unit is used for providing wireless charging for the unmanned aerial vehicle self-throwing search and rescue module; the multidimensional monitoring unit is used for monitoring the temperature, the humidity and the wind speed of the wireless charging unit; and the charging pile interface unit is used for providing an installation interface for the communication navigation unit, the wireless charging unit and the multidimensional monitoring unit.
10. An air-water amphibious unmanned aerial vehicle self-throwing search and rescue method, which is characterized by being realized by the air-water amphibious unmanned aerial vehicle self-throwing search and rescue system as claimed in any one of claims 1 to 9, comprising: based on the unmanned aerial vehicle self-rescue module, acquiring image data and point cloud data of a search and rescue area in real time according to a search and rescue task, and transmitting the image data and the point cloud data to an intelligent emergency control module; Based on the intelligent emergency control module is based on image data and point cloud data are discerned to search and rescue target to track the search and rescue target after the discernment, obtain target position appearance, based on target position appearance and the position appearance of unmanned aerial vehicle self-throwing search and rescue module of acquireing are right unmanned aerial vehicle self-throwing search and rescue module's flight trajectory plans, based on the flight trajectory control of planning unmanned aerial vehicle self-throwing search and rescue module drops to search and rescue target periphery, realizes the unmanned aerial vehicle self-throwing search and rescue of empty water amphibious.

Description

Automatic-throwing search and rescue system and method for amphibious unmanned aerial vehicle Technical Field The invention relates to the technical field of water emergency rescue, in particular to an automatic-throwing search and rescue system and method of an air-water amphibious unmanned aerial vehicle. Background Along with the increasing frequency of inland and offshore traffic activities, the risk of water safety accidents is continuously increased, and the traditional search and rescue means mainly depend on rescue ships and manual visual observation, so that the problems of long response delay, limited search range, difficult approaching to a target under severe sea conditions, efficiency dip at night or in a low-visibility environment and the like exist. The existing unmanned aerial vehicle visual recognition detection is easy to be interfered under a complex water surface environment, the target pixel of a person falling into water is small in duty ratio under the high-altitude aerial photographing visual angle of the unmanned aerial vehicle, serious background interference is easy to generate under the condition of interference of sea wave lines, floaters and the like under extreme conditions such as multi-sea waves, strong light waves and the like, a high-light area is easy to form by the strong light reflection and the waves on the water surface, the image quality is seriously distorted, the person falling into water can be submerged by light reflection or is confused with floating sundries, the human body characteristics are covered, the person falling into water is difficult to accurately recognize the person falling into water and the obstacle on the water, and after the unmanned aerial vehicle recognizes the person falling into water, the information of the person falling into water is transmitted to a rescue ship, so that the person falling into water cannot be separated from danger quickly, and the timeliness and reliability of offshore emergency rescue are reduced. Disclosure of Invention In view of the foregoing, it is necessary to provide a system and a method for automatically rescuing an unmanned aerial vehicle with amphibious functions, which are used for solving the technical problems of insufficient target detection precision and low timeliness and reliability of emergency rescue in a complex water surface environment. In order to solve the problems, in a first aspect, the invention provides an air-water amphibious unmanned aerial vehicle self-throwing search and rescue system, which comprises an unmanned aerial vehicle self-throwing search and rescue module and an intelligent emergency control module; the unmanned aerial vehicle self-throwing search and rescue module is used for collecting image data and point cloud data of a search and rescue area in real time according to a search and rescue task and transmitting the image data and the point cloud data to the intelligent emergency control module; The intelligent emergency control module is used for identifying a search and rescue target based on the image data and the point cloud data, tracking the identified search and rescue target to obtain a target pose, planning the flight track of the unmanned aerial vehicle self-casting search and rescue module based on the target pose and the acquired pose of the unmanned aerial vehicle self-casting search and rescue module, controlling the unmanned aerial vehicle self-casting search and rescue module to land to the periphery of the search and rescue target based on the planned flight track, and realizing the unmanned aerial vehicle self-casting search and rescue of the amphibious unmanned aerial vehicle. In one possible implementation manner, the unmanned aerial vehicle self-casting search and rescue module comprises a multi-mode camera unit and a communication unit; the multi-mode camera unit is used for collecting image data and point cloud data of a search and rescue area in real time according to a search and rescue task; the communication unit is used for transmitting the image data and the point cloud data to the intelligent emergency control module. In one possible implementation manner, the unmanned aerial vehicle self-casting search and rescue module further comprises a composite lightweight engine body structure, a hybrid power propulsion unit, an energy and management unit and a propeller protection net; The composite light-weight machine body structure is a concentric circle type life buoy appearance, and the edge of the concentric circle type life buoy appearance comprises an ergonomic concave plate surface for supporting a search and rescue target and an annular functional handle for grasping the search and rescue target; The hybrid propulsion unit is used for switching the aerial high-speed cruising of the unmanned aerial vehicle and flexible maneuvering on the water surface; The energy and management unit is used for providing energy supply for the multi-mode camera un