CN-121477736-B - Multi-node singlechip automatic synchronous control method based on edge cooperation

CN121477736BCN 121477736 BCN121477736 BCN 121477736BCN-121477736-B

Abstract

The invention discloses an edge cooperation-based multi-node singlechip automatic synchronous control method, which relates to the technical field of distributed cooperative control and comprises the steps of calculating the norm of a cooperative synchronous error vector, presetting a trigger threshold, packaging local real-time State data of a corresponding singlechip node into a State data packet when the norm of the cooperative synchronous error vector is larger than the trigger threshold, sending the State data packet to an edge server, controlling the edge server to receive the State data packet, carrying out State estimation by combining a State-Space model, generating a global State estimation value, calculating an optimal cooperative control sequence for the singlechip node by utilizing a model predictive control algorithm according to the global State estimation value, extracting a first control instruction from the optimal cooperative control sequence, and sending the first control instruction to the corresponding singlechip node to drive an executing mechanism to act.

Inventors

ZHANG LIANG
MA LI
YE YITING
ZHOU JIANZHUANG
ZHANG XIN
LIU CHUANG
ZHANG XINYU
YANG SHUYI
ZHAO NING
WANG XINYONG

Assignees

渤海大学

Dates

Publication Date: 20260505
Application Date: 20251113

Claims (8)

1. The multi-node singlechip automatic synchronous control method based on edge cooperation is characterized by comprising the following steps of, An edge server is used for generating a desired state track for each singlechip node, specifically: after receiving the collaborative task, the edge server analyzes the collaborative task, decomposes a macroscopic task target, plans a motion path from a starting point to a terminal point of each singlechip node, and obtains an expected state track of each singlechip node; The method comprises the steps of constructing an initial cooperative communication topology, specifically: analyzing expected state tracks of all singlechip nodes by using an edge server, and identifying constraint relations among the singlechip nodes; determining singlechip nodes needing direct communication cooperation based on the identified constraint relation, and distributing connection weight for each pair of singlechip nodes to form an initial cooperation communication topology; each singlechip node performs local tracking control according to the expected state track to acquire local real-time state data; based on the local real-time state data of each singlechip node and the initial cooperative communication topology, calculating a cooperative synchronization error vector between adjacent singlechip nodes; Calculating the norm of the collaborative synchronization error vector, presetting a trigger threshold, and packing local real-time state data of the corresponding singlechip node into a state data packet and sending the state data packet to an edge server when the norm of the collaborative synchronization error vector is larger than the trigger threshold; The control edge server receives the State data packet, performs State estimation by combining with a State-Space model, and generates a global State estimation value; calculating an optimal cooperative control sequence for the singlechip node by using a model predictive control algorithm according to the global state estimation value; and extracting a first control instruction from the optimal cooperative control sequence, and transmitting the first control instruction to a corresponding singlechip node to drive an executing mechanism to act.
2. The multi-node singlechip automation synchronization control method based on edge cooperation as set forth in claim 1, wherein each singlechip node performs local tracking control according to an expected state track to obtain local real-time state data, specifically: Each singlechip node acquires the physical position and angular velocity value of the executing mechanism by using an inertia measurement sensor according to the expected state track; And performing data conversion and filtering processing on the physical position and the angular velocity value to form local real-time state data.
3. The multi-node singlechip automation synchronization control method based on edge cooperation as set forth in claim 2, wherein the calculating of the cooperative synchronization error vector between adjacent singlechip nodes based on the local real-time state data and the initial cooperative communication topology of each singlechip node is specifically as follows: Instructing each singlechip node to read adjacent singlechip nodes defined in the initial cooperative communication topology; According to the local real-time state data of each singlechip node, calculating the actual relative state vector between adjacent singlechip nodes; searching expected relative state vectors between adjacent singlechip nodes; A collaborative synchronization error vector is calculated based on the actual relative state vector and the expected relative state vector.
4. The method for automatically synchronizing and controlling the multi-node singlechip based on edge coordination as set forth in claim 3, wherein the method is characterized in that the norm of the coordination synchronization error vector is calculated, a trigger threshold is preset, and when the norm of the coordination synchronization error vector is larger than the trigger threshold, local real-time state data of the corresponding singlechip node is packed into a state data packet and sent to an edge server, and the method specifically comprises the steps of: after reading each component of the collaborative synchronization error vector, carrying out norm operation to obtain the norm of the collaborative synchronization error vector; Presetting a trigger threshold according to the norm of the collaborative synchronization error vector; And comparing the norm of the collaborative synchronization error vector with a trigger threshold, and when the norm of the collaborative synchronization error vector is larger than the trigger threshold, packaging the local real-time state data of the corresponding singlechip node into a state data packet and sending the state data packet to an edge server through a communication network.
5. The method for automatically and synchronously controlling the multi-node single-chip microcomputer based on the edge cooperation of claim 4, wherein the control edge server receives the State data packet and performs State estimation by combining with a State-Space model to generate a global State estimation value, specifically: Receiving a state data packet of the singlechip node by using a monitoring network port of the edge server, analyzing the state data packet, and extracting an identifier of the singlechip node and corresponding local real-time state data; And taking the received local real-time State data as an observed quantity to be input into a State-Space model for State prediction, and generating global State estimated values of all singlechip nodes.
6. The method for automatically and synchronously controlling the multi-node singlechip based on edge cooperation according to claim 5, wherein the optimal cooperative control sequence is calculated for the singlechip nodes by using a model predictive control algorithm according to the global state estimation value, and specifically comprises the following steps: Taking the global state estimation values of all the singlechip nodes as the initial state of a model predictive control algorithm, and constructing a constrained quadratic programming problem by the model predictive control algorithm by taking the collaborative tracking errors and control energy consumption of all the singlechip nodes as optimization targets; and solving the quadratic programming problem and outputting an optimal cooperative control sequence.
7. The method for automatically and synchronously controlling the multi-node singlechip based on edge synergy, as set forth in claim 6, wherein the solving the quadratic programming problem and outputting the optimal synergy control sequence means performing iterative computation on the quadratic programming problem by using a quadratic programming solver, searching for decision variables which meet all linear inequality constraints and minimize quadratic objective functions, taking all the found decision variables as optimal control amounts, and arranging the decision variables in time sequence to form the optimal synergy control sequence.
8. The method for automatically and synchronously controlling the multi-node single-chip microcomputer based on the edge cooperation of claim 7, wherein the method is characterized in that a first control instruction is extracted from an optimal cooperation control sequence and is issued to a corresponding single-chip microcomputer node to drive an executing mechanism to act, and the method is specifically as follows: extracting a first control instruction corresponding to a current control period from the optimal cooperative control sequence of each singlechip node; Packaging the first control instruction and an identifier of a corresponding singlechip node to generate an instant control instruction packet; The edge server is used for sending the instant control instruction packet to the singlechip node appointed by the identifier through the communication network, and the singlechip node converts the instant control instruction packet into a driving signal through the driving circuit, so that the executing mechanism executes corresponding actions.

Description

Multi-node singlechip automatic synchronous control method based on edge cooperation Technical Field The invention relates to the technical field of distributed cooperative control, in particular to an automatic synchronous control method of a multi-node singlechip based on edge cooperation. Background Along with the rapid development of industrial automation and intelligent manufacturing, multi-node cooperative control is widely applied to scenes such as robot clusters, flexible production lines and intelligent logistics, wherein a plurality of singlechips are used as bottom execution units, high-precision time and space synchronization is needed to be realized, so that the coordination and stability of an overall system are ensured, the traditional centralized control architecture depends on a central controller to uniformly schedule all nodes, the problems of large communication delay, concentrated computing load and the like are solved, the dynamic cooperative task requirement with high real-time requirement is difficult to meet, in recent years, the rising of an edge computing method provides a new solution for distributed control, and the response speed is improved while the network transmission cost is reduced by sinking computing capacity to an edge server close to an equipment end, so that a layered control architecture of cloud-edge-end cooperation is realized. However, the existing cooperative control method still has improvement, firstly, in terms of a communication mechanism, a fixed-period state data uploading mode is adopted, so that unnecessary communication load is generated under the condition of limited network resources, and secondly, in terms of control optimization, the existing method mostly adopts independent tracking control or weak coupling cooperative strategy, and the motion constraint relation and dynamic interaction influence among nodes cannot be fully considered, so that cooperative synchronization errors are accumulated, and accurate synchronous control is difficult. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides an edge-cooperation-based multi-node singlechip automatic synchronous control method which solves the problem that accurate synchronous control is difficult. In order to solve the technical problems, the invention provides the following technical scheme: the invention provides an edge-cooperation-based multi-node singlechip automatic synchronous control method which comprises the following steps of, Generating a desired state track for each singlechip node by using an edge server, and constructing an initial collaborative communication topology; each singlechip node performs local tracking control according to the expected state track to acquire local real-time state data; based on the local real-time state data of each singlechip node and the initial cooperative communication topology, calculating a cooperative synchronization error vector between adjacent singlechip nodes; Calculating the norm of the collaborative synchronization error vector, presetting a trigger threshold, and packing local real-time state data of the corresponding singlechip node into a state data packet and sending the state data packet to an edge server when the norm of the collaborative synchronization error vector is larger than the trigger threshold; The control edge server receives the State data packet, performs State estimation by combining with a State-Space model, and generates a global State estimation value; calculating an optimal cooperative control sequence for the singlechip node by using a model predictive control algorithm according to the global state estimation value; and extracting a first control instruction from the optimal cooperative control sequence, and transmitting the first control instruction to a corresponding singlechip node to drive an executing mechanism to act. The method for automatically and synchronously controlling the multi-node singlechip based on the edge coordination is characterized in that the step of generating the expected state track for each singlechip node by using the edge server means that after the edge server receives the coordination task, the coordination task is analyzed, a macroscopic task target is decomposed, a motion path from a starting point to an end point of each singlechip node is planned, and the expected state track of each singlechip node is obtained. As a preferable scheme of the multi-node singlechip automation synchronous control method based on edge cooperation, the invention comprises the following steps of: analyzing expected state tracks of all singlechip nodes by using an edge server, and identifying constraint relations among the singlechip nodes; Based on the identified constraint relation, the singlechip nodes needing direct communication cooperation are determined, and connection weights are distributed