CN-121977533-A - Unmanned vehicle cluster fixed time distributed type cooperative positioning method under positioning fault

CN121977533ACN 121977533 ACN121977533 ACN 121977533ACN-121977533-A

Abstract

The invention relates to the technical field of unmanned vehicles, in particular to a fixed time distributed type cooperative positioning method for an unmanned vehicle cluster under a positioning fault. According to the method, an unmanned vehicle pose estimation model is built by combining with a kinematic model of an unmanned vehicle, and then the linear speed and the angular speed of the unmanned vehicle, which are measured by a sensor, are substituted into the unmanned vehicle pose estimation model to deduce pose estimation of each unmanned vehicle. According to the invention, a complete set of distributed positioning algorithm frame is designed through the relative pose information obtained through unmanned workshop communication and the measurement of the vehicle-mounted sensor, the single-point fault risk of the centralized architecture is effectively avoided, and a robust distributed positioning algorithm is provided for solving the measurement deviation problem caused by the accuracy reduction of the vehicle-mounted sensor or the environmental interference, so that the measurement error of the sensor can be effectively inhibited.

Inventors

DU HAIBO
Sun Xunhong
WEN GUANGHUI
YU LANLIN
XIAO JIAN

Assignees

合肥工业大学

Dates

Publication Date: 20260505
Application Date: 20260401

Claims (10)

1. A fixed-time distributed cooperative positioning method for an unmanned vehicle cluster under a positioning fault is characterized in that an unmanned vehicle pose estimation model is firstly built by combining a kinematic model of an unmanned vehicle, and then the linear speed and the angular speed of the unmanned vehicle, which are measured by a sensor, are substituted into the unmanned vehicle pose estimation model to deduce pose estimation of each unmanned vehicle ); The unmanned vehicle pose estimation model is as follows: Wherein, the 、 And Respectively the estimated values of the x-direction coordinate, the y-direction coordinate and the orientation of the unmanned vehicle i, 、 And Respectively is 、 And Is the first derivative of (a); And The linear velocity measurement value and the angular velocity measurement value of the unmanned vehicle i are respectively; sgn is a sign function, g is a reference parameter; 、 And Is a set coefficient; A weighted cumulative error calculated for unmanned vehicles i in the x-direction based on relative position measurements between unmanned vehicles and landmarks; a weighted cumulative error calculated for unmanned vehicles i in the y-direction based on relative position measurements between unmanned vehicles and landmarks; The error is accumulated for the weight of the drone i in the direction calculated based on the angular difference measurements between the drones and between the drone and the landmark.
2. The method for fixed time distributed co-location of an unmanned vehicle cluster under a location fault as claimed in claim 1, wherein the weighted cumulative error introduces a communication weight, and the communication weight is set as follows: constructing an adjacency matrix A based on the assumption that the system communication topology is strong communication and used for expressing the communication weight between unmanned vehicles, and constructing a landmark association matrix , For the communication weights of drone i and landmark k, The method is characterized in that the method indicates that the unmanned vehicles i can measure landmarks k, wherein n and m are the number of unmanned vehicles and the number of landmarks respectively; Solving an adjacency matrix A and a landmark association matrix B based on the following constraint conditions to obtain communication weights between the unmanned vehicles and landmarks; The constraint conditions include: ; Is symmetrically positive; Wherein L is Laplacian matrix corresponding to adjacency matrix A, and P is detail balance weight matrix constructed based on assumption that system communication topology is strong communication and detail balance diagram, namely , ; Representing the communication weights of the unmanned vehicles i to j, Representing the communication weights of the unmanned vehicles j to i, 、、 And The weights of corresponding unmanned vehicles 1, i, j and n in P are respectively equal to or greater than 1 and equal to or less than n, and equal to or greater than 1 and equal to or less than j and equal to or less than n.
3. The method for fixed time distributed co-location of a cluster of drones under a location fault of claim 2, wherein the coefficients satisfy the following constraints: Wherein, the The maximum error constant of the sensor; as a result of the transition coefficient(s), 。
4. The method for fixed time distributed co-location of a drone cluster in a location fault of claim 3, wherein the fixed time for the pose estimation error convergence of the drone in the drone cluster is: Wherein, T is a fixed time, In order to set the constant value of the constant, , And Are coefficients.
5. The method for fixed time distributed co-location of a drone cluster in a localized fault of claim 4, ; Is a matrix Is a minimum feature value of (a).
6. The method for locating a fixed time distributed co-location of an unmanned vehicle cluster in a fault of claim 4, wherein: n is the number of unmanned vehicles, Is a matrix Is a minimum feature value of (a).
7. The method for fixed time distributed co-location of a cluster of drones in a location fault of claim 6, wherein the error convergence range of the pose estimation of the drone is: Wherein P is a detail balance weight matrix, L is a Laplacian matrix, B is a landmark associated moment, e, f and g respectively represent estimated error vectors of n unmanned vehicles in x direction, y direction and orientation, V 1 、V 2 and V 3 are transition functions, And Is a coefficient; in order to set the constant value of the constant, 。
8. The method for fixed time distributed co-location of a cluster of drones in a location fault of claim 1, wherein the kinematic model of the ith drone is: Wherein, the Representing an unmanned vehicle cluster; The first derivatives of the x-direction coordinate, the y-direction coordinate and the direction of the unmanned vehicle i are respectively; 、 And The linear speed, angular speed and heading of the unmanned vehicle i, respectively.
9. A system comprising a memory and a processor, the memory having stored therein a computer program, the processor being coupled to the memory, the processor being configured to execute the computer program to implement the method for fixed time distributed co-location of an unmanned vehicle cluster in a location fault as claimed in any one of claims 1 to 8.
10. A storage medium, characterized in that a computer program is stored, which computer program, when executed, is adapted to implement the method of fixed time distributed co-localization of a drone cluster under localization fault as claimed in any one of claims 1-8.

Description

Unmanned vehicle cluster fixed time distributed type cooperative positioning method under positioning fault Technical Field The invention relates to the technical field of unmanned vehicles, in particular to a fixed time distributed type cooperative positioning method for an unmanned vehicle cluster under a positioning fault. Background With the deep integration of artificial intelligence and automation technology, multi-vehicle systems have become a key paradigm in performing distributed tasks (e.g., collaborative detection, formation control, object tracking) in dynamic environments. The traditional positioning method depends on external infrastructure (such as GPS and UWB), but has obvious limitations that GPS signals are easy to be blocked in indoor, underground or dense urban areas, and in complex scenes such as indoor, underground or dense cities, the traditional positioning method which depends on the external infrastructure is basically invalid, and UWB deployment cost is high and is difficult to adapt to dynamic environment changes. Furthermore, centralized information processing architectures face single point of failure and communication bandwidth bottleneck risks. In recent years, the distributed co-location technology provides a location solution without global information by integrating relative measurement of an unmanned workshop, local environment perception data and multi-sensor filtering fusion, and relieves the fault risk of a centralized architecture to a certain extent. However, the existing method mostly ignores the influence of the sensor measurement error (such as noise, deviation or drift), and the convergence speed is mostly asymptotically stable, so that the real-time requirement of the high dynamic scene is difficult to be satisfied. The fixed time control strategy has the advantages of presettable convergence time, strong anti-interference performance and the like, but the application in the positioning of multiple unmanned vehicles is insufficient, and especially a targeted fault tolerance mechanism for common faults (such as faults of a positioning sensor and a measuring sensor) is lacking. Disclosure of Invention In order to solve the problem that the unmanned vehicle cluster co-location in the prior art lacks a fault tolerance mechanism for the unmanned vehicle self-carried sensor fault, the invention provides a unmanned vehicle cluster fixed time distributed co-location method under the fault location, which can realize accurate pose estimation in fixed time through unmanned workshop communication and on-site sensor measurement and effectively challenges the measurement error. According to the unmanned vehicle cluster fixed time distributed cooperative positioning method under the positioning fault, an unmanned vehicle pose estimation model is firstly built by combining with a kinematic model of an unmanned vehicle, and then the linear speed and the angular speed of the unmanned vehicle, which are measured by a sensor, are substituted into the unmanned vehicle pose estimation model to deduce pose estimation of each unmanned vehicle); The unmanned vehicle pose estimation model is as follows: Wherein, the 、AndRespectively the estimated values of the x-direction coordinate, the y-direction coordinate and the orientation of the unmanned vehicle i,、AndRespectively is、AndIs the first derivative of (a); And The linear velocity measurement value and the angular velocity measurement value of the unmanned vehicle i are respectively; sgn is a sign function, g is a reference parameter; 、 And Is a set coefficient; A weighted cumulative error calculated for unmanned vehicles i in the x-direction based on relative position measurements between unmanned vehicles and landmarks; a weighted cumulative error calculated for unmanned vehicles i in the y-direction based on relative position measurements between unmanned vehicles and landmarks; The error is accumulated for the weight of the drone i in the direction calculated based on the angular difference measurements between the drones and between the drone and the landmark. Preferably, the weighted cumulative error introduces a communication weight, and the communication weight is set as follows: constructing an adjacency matrix A based on the assumption that the system communication topology is strong communication and used for expressing the communication weight between unmanned vehicles, and constructing a landmark association matrix ,For the communication weights of drone i and landmark k,The method is characterized in that the method indicates that the unmanned vehicles i can measure landmarks k, wherein n and m are the number of unmanned vehicles and the number of landmarks respectively; Solving an adjacency matrix A and a landmark association matrix B based on the following constraint conditions to obtain communication weights between the unmanned vehicles and landmarks; The constraint conditions include: ; Is symmetrically positive;