CN-121984566-A - Robust clock synchronization parameter estimation method for unmanned aerial vehicle cluster network

CN121984566ACN 121984566 ACN121984566 ACN 121984566ACN-121984566-A

Abstract

The invention relates to a robust clock synchronization parameter estimation method for an unmanned aerial vehicle cluster network, and belongs to the technical field of unmanned aerial vehicles. The method comprises the steps of obtaining sending and receiving time stamp information in a multi-round synchronization process through bidirectional synchronization message interaction between a master node unmanned aerial vehicle and a slave node unmanned aerial vehicle, constructing a master-slave node clock synchronization model containing random time delay influence aiming at the problem of synchronization precision reduction caused by multi-type random time delay distribution in an unmanned aerial vehicle cluster network scene based on the time stamp information, jointly estimating clock frequency deviation and clock phase deviation of the slave node relative to the master node by adopting an optimal invariable estimation method, and correcting or compensating a local clock of the slave node according to an estimation result to realize time synchronization of all nodes in the unmanned aerial vehicle cluster network. And the robustness and stability of time synchronization are improved under the unmanned aerial vehicle cluster network environment with high dynamic and random time delay uncertainty, and the requirement of multi-unmanned aerial vehicle collaborative perception, formation control and task scheduling on high-precision time synchronization is met.

Inventors

WANG HENG
DING DACHUAN
LI MIN

Assignees

重庆邮电大学

Dates

Publication Date: 20260505
Application Date: 20260123

Claims (8)

1. The robust clock synchronization parameter estimation method for the unmanned aerial vehicle cluster network is characterized by being applied to the unmanned aerial vehicle cluster network with limited or invalid Global Navigation Satellite System (GNSS) signals, wherein the unmanned aerial vehicle cluster network comprises at least one master node unmanned aerial vehicle and a plurality of slave node unmanned aerial vehicles, and the method comprises the following steps of: S1, acquiring time stamp information of a synchronous message in the sending and receiving processes through bidirectional synchronous message interaction between a master node unmanned aerial vehicle and a slave node unmanned aerial vehicle; S2, based on the timestamp information, constructing a master-slave node clock synchronization model containing random delay influence aiming at asymmetric propagation influence of propagation delay uncertainty of a synchronization message in a slave-to-master node direction and a master-to-slave node direction caused by relative motion among unmanned aerial vehicle nodes, wherein the random delay comprises random delay from the slave node to the master node direction and random delay from the master node to the slave node direction; S3, under the condition that the statistical distribution of the random time delay is not given a specific assumption, based on the master-slave node clock synchronization model, adopting an optimal invariant estimation method to carry out joint estimation on clock frequency offset and clock phase offset of the slave node relative to the master node; And S4, correcting or compensating the local clock of the slave node according to the joint estimation result of the clock frequency offset and the clock phase offset so as to finish time synchronization between the master node and the slave node in the unmanned aerial vehicle cluster network.
2. The robust clock synchronization parameter estimation method for an unmanned aerial vehicle cluster network according to claim 1, wherein the S1 specifically comprises the following steps: S11, in the first place Secondary synchronization period, slave node At the position of Transmitting a synchronization request to a master node at a moment Wherein the transmission time is carried ; S12 once the master node Receiving the synchronous request and recording the current time ; S13, after a predefined time interval, the master node At the position of Time reply slave node And contains a time stamp ; S14, slave node At the position of Receiving master node at moment And sending reply information.
3. The robust clock synchronization parameter estimation method for an unmanned aerial vehicle cluster network according to claim 1, wherein the step S2 specifically comprises the steps of: S21, completion of After the round message interaction, the slave node An observation set is collected, expressed as the following formula: Wherein N represents the total communication round between the master node and the slave node; Judging whether the synchronous turn reaches a set value, if so, continuing to carry out the next step S22, otherwise, returning to the step S11, and setting N as N+1; s22, according to affine clock model, master node And slave node Can be modeled as the following formula: Wherein, the And Respectively represent slave nodes And a master node A clock phase offset and a frequency offset between them, And Representing the speed of light and the slave node, respectively Relative master node Is used for the speed of the (c) in the (c), Representing slave nodes And a master node An initial fixed time delay between the two, And Respectively represent slave nodes To the master node And a master node To slave node In addition, for convenience, define And 。
4. The robust clock synchronization parameter estimation method for an unmanned aerial vehicle cluster network according to claim 1, wherein the step S3 comprises: And stacking timestamp data obtained from the nodes in the process of multi-round synchronous message interaction, constructing a clock synchronization model in a matrix form, and further converting the clock synchronization model into a vector form to obtain a vectorization observation model containing clock frequency deviation, clock phase deviation and random time delay terms, wherein the vectorization observation model represents the linear relation between each timestamp and the clock parameter through a preset structure matrix.
5. The robust clock synchronization parameter estimation method for unmanned aerial vehicle cluster network according to claim 4, wherein the step S3 further comprises: And further transforming the observed data and the unknown parameters under the action of a position scale transformation group, so that the probability density function of the observed data is kept unchanged under the transformation group, thereby obtaining an induced parameter transformation relation matched with the transformation group.
6. The robust clock synchronization parameter estimation method for an unmanned aerial vehicle cluster network according to claim 5, wherein the step S3 further comprises: and selecting a loss function which is kept unchanged under the position scale transformation group as an estimation criterion, constructing a conditional risk function aiming at clock frequency offset and clock phase offset based on the loss function, and solving to obtain the optimal unchanged estimated value of the clock frequency offset and the clock phase offset of the slave node relative to the master node under the meaning of minimizing the conditional risk.
7. The method for estimating robust clock synchronization parameters for an unmanned aerial vehicle cluster network of claim 4, wherein the vectorized observation model is expressed as the following formula: Wherein, the , And, in addition, the method comprises, 。
8. The robust clock synchronization parameter estimation method for an unmanned aerial vehicle cluster network according to claim 6, wherein the optimal invariant estimated value is expressed as the following formula: Wherein the method comprises the steps of , 。

Description

Robust clock synchronization parameter estimation method for unmanned aerial vehicle cluster network Technical Field The invention belongs to the technical field of unmanned aerial vehicles, and relates to a robust clock synchronization parameter estimation method for an unmanned aerial vehicle cluster network. Background Unmanned aerial vehicles are paid attention to in the fields of academic research and engineering application in recent years by virtue of core advantages of convenient deployment, simple structure and flexible operation. By means of the characteristics, the unmanned aerial vehicle network achieves multi-scene large-scale landing, and covers key fields such as environment monitoring, industrial Internet of things, disaster emergency response, collaborative perception and the like. The core complaint of the diversified application scenes is the collaborative operation of a plurality of unmanned aerial vehicles, and the unified time reference is a precondition for guaranteeing the data interaction, task scheduling and action coordination among the nodes. Conventionally, unmanned aerial vehicles rely on the global navigation satellite system (Global Navigation SATELLITE SYSTEM, GNSS) to acquire high-precision reference times. However, complex environments such as urban dense buildings, mountain canyons, indoor storage or underground engineering can shield, shelter or interfere with GNSS signals, resulting in failure of their timing functions, so that the unmanned aerial vehicle must rely on an autonomous time synchronization mechanism to maintain a uniform time scale between network nodes. In a wireless network, each node transmits time stamp information through two-way message interaction, and clock offset and phase offset are estimated by utilizing the collected observed values, so that time synchronization among the nodes is realized. However, the existing method is mainly oriented to industrial network or low dynamic wireless network scenes, and the influence of high mobility and topology rapid change of the unmanned aerial vehicle cluster on time synchronization is not considered. Specifically, under the condition of dynamic change of a network topology structure, the synchronous message is affected by various factors such as channel quality fluctuation, medium access competition, interference, retransmission and the like, so that the statistical characteristic of the end-to-end time delay presents uncertainty. In addition, the existing time synchronization method generally depends on specific time delay distribution or approximately stable statistical information, lacks consideration on multi-type time delay distribution, can face the problem of reduced clock parameter estimation performance among unmanned aerial vehicles in a high-dynamic unmanned aerial vehicle cluster network scene, and cannot meet the high-precision requirement of multi-unmanned aerial vehicle collaborative operation on time synchronization. Based on the shortcomings of the prior art, it is highly desirable to provide a high-precision time synchronization method capable of realizing joint estimation of clock frequency offset and clock phase offset under the condition of not depending on specific random time delay distribution assumption, so as to solve the problem of synchronization performance degradation caused by random time delay uncertainty in an unmanned aerial vehicle cluster network under a GNSS failure environment, and ensure the reliability and stability of collaborative operation. Disclosure of Invention Therefore, the invention aims to provide the robust clock synchronization parameter estimation method for the unmanned aerial vehicle cluster network, so as to solve the problem of clock synchronization precision reduction caused by multi-type random time delay uncertainty under the condition of GNSS signal limitation or failure. According to the method, time stamp information is acquired through bidirectional data exchange between the master unmanned aerial vehicle and the slave unmanned aerial vehicle, and the clock frequency offset and the clock phase offset are jointly estimated by adopting an optimal invariable estimation method, so that high-precision time synchronization among multiple unmanned aerial vehicles is realized. In order to achieve the above purpose, the present invention provides the following technical solutions: The robust clock synchronization parameter estimation method for the unmanned aerial vehicle cluster network is applied to the unmanned aerial vehicle cluster network with limited or invalid Global Navigation Satellite System (GNSS) signals, wherein the unmanned aerial vehicle cluster network comprises at least one master node unmanned aerial vehicle and a plurality of slave node unmanned aerial vehicles, and the method comprises the following steps: S1, acquiring time stamp information of a synchronous message in the sending and receiving processes through bidirectional synch