Target identification and positioning method for air-ground cooperative unmanned system

Abstract

The invention provides a target identification and positioning method for an air-ground cooperative unmanned system. The method comprises the following steps: step 1, carrying out time synchronization on a platform and a sensor; 2, identifying and positioning a high-altitude target by the unmanned aerial vehicle; step 3, ground target identification and positioning of the unmanned vehicle, and mutual transmission of positioning results by an air-ground unmanned system; 4, asynchronous data fusion is carried out based on Kalman filtering, and an accurate target coordinate position is output; asynchronous data fusion of air and ground view angle identification and positioning results of the target is realized by using the Kalman filtering algorithm, the accurate coordinate position of the target can be output, the air-ground cooperative accurate target identification and positioning function is achieved, the problem that the accurate coordinate position of the target is difficult to output by a single unmanned system is solved, and the accuracy of the target is improved. The technical effect of cooperative stable and accurate target identification and positioning of the air-ground unmanned system is achieved, cooperative stable target identification and positioning of the unmanned aerial vehicle and the unmanned vehicle are achieved, and a more accurate target position is output by fusing a plurality of coordinate results.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of target identification, and particularly relates to a target identification and positioning method for an air-ground cooperative unmanned system. BACKGROUND

[0002] In the field of target identification and positioning of an unmanned system, the unmanned system needs to realize identification and positioning of a target, output accurate coordinate positions of the target, and have robustness in different environments. Current schemes mostly adopt a single system to independently work, mainly acquire a global image through a top-down view of a UAV to perform detection and positioning, or acquire a local image through a ground view of an unmanned vehicle to perform detection and positioning. These methods all have the shortcoming that global perception and local precision cannot be achieved simultaneously, resulting in technical defects such as target missed detection, false detection and positioning error accumulation when the unmanned system performs a target identification and positioning task.

[0003] The existing target identification and positioning scheme of an unmanned system mostly adopts a single unmanned system to perform a task. The UAV is limited by the flight height and can usually only provide a top-down or oblique view, resulting in missing of bottom or side information. The unmanned vehicle is limited by the ground height and the field of view is easily blocked, and cannot perceive high targets. Therefore, the single unmanned system is not stable in performing the target identification and positioning task, especially for the blocked target, the identification and positioning precision will decrease or even missed detection occurs. In addition, the existing target identification and positioning scheme mostly adopts sensors such as laser radar which are relatively high in price, not only the purchase cost, but also the load of the unmanned system is significantly increased, and the computing capacity requirement of the system is also relatively high.

[0004] Therefore, the application provides a target identification and positioning method for an air-ground cooperative unmanned system to solve the above problems. SUMMARY

[0005] In order to solve the above technical problems, the application provides a target identification and positioning method for an air-ground cooperative unmanned system to solve the problems in the prior art that the target identification and positioning scheme mostly adopts sensors such as laser radar which are relatively high in price, not only the purchase cost, but also the load of the unmanned system is significantly increased, and the computing capacity requirement of the system is also relatively high.

[0006] A target identification and positioning method for an air-ground cooperative unmanned system, comprising the following steps:

[0007] Step 1: time synchronization of a platform and a sensor;

[0008] Step 2: high-altitude target identification and positioning of a UAV;

[0009] Step 3: ground target identification and positioning of an unmanned vehicle, and mutual transmission of positioning results of the air-ground unmanned system;

[0010] Step four: asynchronous data fusion based on Kalman filtering, output accurate target coordinate position.

[0011] Further, in step one, before the target identification and positioning of the unmanned system, the local clocks of all platforms and sensors are unified to the same global time reference.

[0012] Further, in step two, the unmanned aerial vehicle quickly identifies and positions the ground target based on the global image from the overhead perspective at high altitude, and sends the calculated coordinate position to the ground unmanned vehicle.

[0013] Further, in step three, the unmanned vehicle reaches the designated position according to the coordinates identified by the unmanned aerial vehicle, and then performs relatively accurate identification and positioning of the target based on the local image from the ground perspective.

[0014] Further, in the process of target identification and positioning of the unmanned vehicle, the unmanned aerial vehicle and the unmanned vehicle adjust their observation positions of the target to achieve stable identification and positioning of the target, and the unmanned aerial vehicle and the unmanned vehicle transmit each other's real-time coordinate positions calculated through target identification and positioning.

[0015] Further, in step four, based on Kalman filtering, asynchronous data fusion is performed, the state estimation theory is used to dynamically weight the observation data of the two, an optimal estimation output with minimum variance and no bias is output, a more accurate coordinate result is output, and the rapid and accurate identification and positioning of the target by the air-ground unmanned system is achieved.

[0016] Further, the Kalman filtering algorithm is based on linear system model and Gaussian noise assumption, and realizes optimal estimation of the target state through prediction and update process of system state, and represents the motion state of the target as a state vector, which contains the position, velocity, acceleration and other information of the target.

[0017] Further, the Kalman filtering algorithm defines state transition matrix and observation matrix, which are used to describe the change rule of target state with time and the relationship between observation value and target state respectively, in the prediction stage, according to the state estimation of last time and state transition matrix, combined with process noise covariance matrix, the target state and covariance matrix at current time are predicted, in the update stage, when the new observation value is obtained, the Kalman gain is calculated by using observation matrix and measurement noise covariance matrix, and the predicted value and the observation value are fused through the Kalman gain, so that more accurate state estimation is obtained.

[0018] Compared with the prior art, the present application has the following beneficial effects:

[0019] 1、The present application significantly improves the accuracy and efficiency of target identification and positioning through the efficient cooperation of air-ground collaborative unmanned systems. The unmanned aerial vehicle uses high-altitude perspective to quickly scan and preliminarily locate the target, and transmits the rough coordinates to the unmanned vehicle, which then performs fine identification and positioning based on ground perspective. Through real-time data interaction and adjustment of observation position, the system ensures continuous and stable tracking of the target. Subsequently, the system uses Kalman filtering algorithm to fuse asynchronous data, dynamically weights the observation results of the unmanned aerial vehicle and the unmanned vehicle, and outputs the optimal target coordinate estimation value. This method not only overcomes the limitations of single platform perspective, but also effectively reduces the influence of environmental interference on positioning accuracy, providing reliable technical support for air-ground collaborative operation.

[0020] 2、The present application greatly reduces the overall cost and load demand of the system by combining low-cost sensors and intelligent algorithms. It reduces the equipment purchase cost and significantly reduces the weight of the unmanned system, improving its endurance and flexibility. At the same time, by optimizing the calculation process of the Kalman filtering algorithm, the requirement for hardware computing power is reduced, enabling the system to run efficiently under limited resources. In addition, the present application designs a flexible time synchronization mechanism to ensure that the unmanned aerial vehicle and the unmanned vehicle can work stably and reliably in complex environments, thus achieving high-precision target identification and positioning while considering economy and practicality, providing a feasible solution for multi-scenario applications.

[0021] 3、The present application uses Kalman filtering algorithm to realize asynchronous data fusion of air and ground perspective identification and positioning results of the target, which can output the accurate coordinate position of the target, achieve the function of air-ground collaborative precise target identification and positioning, solve the problem that single unmanned system cannot output the accurate coordinate position of the target, and obtain the technical effect of air-ground collaborative stable and accurate target identification and positioning, realize stable identification and positioning of the target by the unmanned aerial vehicle and the unmanned vehicle, and output more accurate target position by fusing multiple coordinate results. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is a flowchart of the present application. DETAILED DESCRIPTION

[0023] The embodiments of the present application will be further described in detail below in conjunction with the drawings and examples. The following examples are used to illustrate the present application, but cannot be used to limit the scope of the present application.

[0024] As Figure 1 shown, the present application provides an air-ground collaborative unmanned system target identification and positioning method, comprising the following steps:

[0025] Step 1: Time synchronization of platform and sensor;

[0026] Step two: high-altitude target recognition and positioning by the UAV;

[0027] Step three: ground target recognition and positioning by the unmanned vehicle, and mutual transmission of positioning results by the air-ground unmanned system;

[0028] Step four: asynchronous data fusion based on Kalman filtering, and output of accurate target coordinate position.

[0029] As an embodiment of the present application, in step one, before target recognition and positioning by the unmanned system, the local clocks of all platforms and sensors are unified to the same global time reference.

[0030] As an embodiment of the present application, in step two, the UAV quickly recognizes and positions the ground target based on the global image from the overhead perspective at high altitude, and sends the calculated coordinate position to the ground unmanned vehicle.

[0031] As an embodiment of the present application, in step three, the unmanned vehicle reaches the designated position according to the coordinates recognized by the UAV, and then relatively accurately recognizes and positions the target based on the local image from the ground perspective.

[0032] As an embodiment of the present application, during the target recognition and positioning process of the unmanned vehicle, the UAV and the unmanned vehicle adjust their observation positions of the target to achieve stable recognition and positioning of the target, and the UAV and the unmanned vehicle mutually transmit the real-time coordinate positions calculated by each through target recognition and positioning.

[0033] As an embodiment of the present application, in step four, asynchronous data fusion is performed based on Kalman filtering, the observation data of both is dynamically weighted using state estimation theory, an optimal estimation output with minimum variance and no bias is output, a more accurate coordinate result is output, and fast and accurate recognition and positioning of the target by the air-ground unmanned system is achieved.

[0034] As an embodiment of the present application, the Kalman filtering algorithm is based on linear system model and Gaussian noise assumption, and through the prediction and update processes of system state, optimal estimation of the target state is achieved, and the motion state of the target is represented as a state vector, which contains the position, velocity, acceleration and other information of the target.

[0035] As an embodiment of the present application, the Kalman filtering algorithm defines a state transition matrix and an observation matrix, respectively describing the change rule of the target state over time and the relationship between the observation value and the target state. In the prediction stage, according to the state estimation at the last moment and the state transition matrix, combined with the process noise covariance matrix, the target state and the covariance matrix at the current moment are predicted. In the update stage, after obtaining the new observation value, the Kalman gain is calculated using the observation matrix and the measurement noise covariance matrix, and the predicted value and the observation value are fused through the Kalman gain, so as to obtain more accurate state estimation.

[0036] In practical applications, this method can adapt to the target recognition and positioning needs in various complex environments. In urban search and rescue scenarios, unmanned aerial vehicles can quickly cover large areas, identify potential target locations, and pass information to unmanned vehicles. Unmanned vehicles can then conduct detailed searches in narrow streets or inside buildings to further confirm the precise location of the target. Through this collaborative working mode, the system not only improves search efficiency but also reduces the risk of misjudgment due to single perspective limitations.

[0037] In addition, this method also shows broad application prospects in fields such as agricultural monitoring, disaster assessment, and military reconnaissance. In agricultural scenarios, unmanned aerial vehicles can conduct high-altitude scanning of farmland to mark abnormal areas, while unmanned vehicles can conduct detailed detection in the field to provide more accurate data support. In disaster assessment, the air-ground collaborative system can efficiently complete the target recognition and positioning tasks in disaster areas under road blockage or complex terrain conditions, providing scientific basis for rescue decision-making.

[0038] To further enhance the robustness of the system, this method also designs a dynamic adjustment mechanism. When environmental conditions change, such as deteriorating weather conditions or target moving speed increasing, the system can automatically adjust the task allocation and observation strategy of unmanned aerial vehicles and unmanned vehicles according to real-time data. This flexibility ensures that the system can maintain efficient and stable working state when facing unexpected situations, thus meeting diverse task requirements.

[0039] When tracking vehicles driving on fixed routes, the motion pattern of the vehicle is relatively regular, basically conforming to the uniform linear motion or uniform acceleration linear motion model, and the Kalman filter can accurately predict the next position of the vehicle to achieve stable tracking. In the agricultural plant protection scene, when the unmanned aerial vehicle needs to track the irrigation equipment or crop growth area at a fixed position in the farmland, the model-based tracking algorithm can effectively track the target according to the initial position and motion model of the equipment or area, providing support for precise plant protection.

[0040] In a dynamic environment, the motion state of the target can be influenced by various factors such as wind disturbance, terrain changes, or human factors. Through real-time correction of the target motion model and dynamic updating of the noise parameters, the system can maintain high tracking accuracy in complex scenarios. For example, when a UAV tracks a high-speed moving target, the system automatically adjusts the process noise covariance matrix based on the historical trajectory of the target and the current observation data, thereby improving the accuracy of the prediction.

[0041] During the cooperative positioning process, the communication quality between the UAV and the unmanned vehicle has an important influence on the overall performance. This method not only realizes high-precision identification and positioning of static and dynamic targets, but also significantly enhances the environmental adaptability and task flexibility of the system. Whether it is facing unexpected situations or long-term operation needs, the air-ground cooperative unmanned system can provide stable and efficient target tracking capability, providing strong technical support for multi-field application scenarios.

[0042] The embodiments of the present application are given for example and description, although the embodiments of the present application have been shown and described above, it is understood that the above-mentioned embodiments are exemplary and cannot be understood as limiting the present application, and those skilled in the art can make changes, modifications, replacements and modifications to the above-mentioned embodiments within the scope of the present application.

Claims

1. An air-ground cooperative unmanned system target identification positioning method, characterized in that, The method comprises the following steps: Step one: time synchronization of platform and sensor; Step two: high-altitude target recognition and positioning of UAV; Step three: ground target recognition and positioning of unmanned vehicle, and mutual transmission of positioning results of air-ground unmanned system; Step four: asynchronous data fusion based on Kalman filtering, and output of accurate target coordinate position.

2. The method of claim 1, wherein, In step one, before target recognition and positioning of unmanned system, local clocks of all platforms and sensors are unified to the same global time reference.

3. The method of claim 1, wherein, In step two, the UAV quickly recognizes and positions the ground target based on the global image from the overhead perspective, and sends the calculated coordinate position to the ground unmanned vehicle.

4. The method of claim 1, wherein, In step three, the unmanned vehicle reaches the specified position according to the coordinate recognized by the UAV, and then relatively accurately recognizes and positions the target based on the local image from the ground perspective.

5. The method of claim 1, wherein, In the process of target recognition and positioning of the unmanned vehicle, the UAV and the unmanned vehicle adjust their observation positions of the target to realize stable recognition and positioning of the target, and the UAV and the unmanned vehicle mutually transmit the real-time coordinate positions calculated by target recognition and positioning.

6. The method of claim 1, wherein, In step four, asynchronous data fusion based on Kalman filtering is performed, the observation data of both is dynamically weighted by using state estimation theory, an optimal estimation output with minimum variance and no bias is output, a more accurate coordinate result is output, and fast and accurate recognition and positioning of the target by the air-ground unmanned system are achieved.

7. The method of claim 1, wherein, The Kalman filtering algorithm is based on linear system model and Gaussian noise assumption, and realizes optimal estimation of target state through prediction and update processes of system state, and represents the motion state of the target as a state vector, which contains the position, velocity, acceleration and other information of the target.

8. The method of claim 1, wherein, The Kalman filtering algorithm defines state transition matrix and observation matrix, which are used to describe the change rule of target state with time and the relationship between observation value and target state respectively, in the prediction stage, the target state and covariance matrix at the current time are predicted according to the state estimation at the last time, state transition matrix and process noise covariance matrix, in the update stage, when the new observation value is obtained, the Kalman gain is calculated by using observation matrix and measurement noise covariance matrix, and the predicted value and the observation value are fused by using the Kalman gain, so that more accurate state estimation is obtained.

Images