CN-121091894-B - Dynamic task allocation and path planning method for unmanned ship cluster

CN121091894BCN 121091894 BCN121091894 BCN 121091894BCN-121091894-B

Abstract

The invention discloses a dynamic task allocation and path planning method of unmanned ships and light boats, which comprises the steps of establishing a virtual data field representing an operation environment, defining task targets as virtual information sources capable of generating scalar concentration gradients, enabling each unmanned ship in the cluster to independently calculate motion vectors based on the gradients perceived locally, and dynamically adjusting weights of communication links connecting the unmanned ships and light boats according to information interaction frequencies among the unmanned ships and light boats. According to the invention, the aquatic mycelium network core constructed based on the mycelium network growth and information transfer model is introduced, task allocation is converted into pheromone gradient optimizing problem, and communication interaction history is used as the basis of network topology self-adaptive growth, so that communication uncertainty is converted into the key dimension of driving clusters to realize self-organization, self-healing and emerging intelligence, and the task execution efficiency, system toughness and autonomous coordination capability of unmanned ship clusters under the conditions of limited communication and high dynamic environment are remarkably improved.

Inventors

DONG CHAO
ZHENG BING
CHEN YANKUN
CHEN FENG

Assignees

自然资源部南海调查中心

Dates

Publication Date: 20260512
Application Date: 20251029

Claims (7)

1. The method for dynamic task allocation and path planning of the unmanned ship cluster is characterized by comprising the following steps of: establishing, in at least one computing processor, a virtual data field representing an operating environment; Defining one or more task targets as virtual information sources capable of generating scalar concentration gradients in the virtual data field; Independently calculating a motion vector for each unmanned boat based on the scalar concentration gradient perceived locally by the unmanned boat; dynamically adjusting the weight of a communication link connecting any two unmanned boats according to the information interaction frequency between the unmanned boats, wherein the weight adjustment is independent of or combined with the physical signal quality of the link; The method further comprises the steps of realizing the whole path planning of the cluster and the self-organizing construction of a network topological structure by executing independent motion of each unmanned ship based on local gradient calculation and combining with dynamic adjustment of the weight of the communication link; Modeling each unmanned ship as a hypha network tip, modeling the communication link as hypha connected with the tip, and modeling the task target as a nutrition source, thereby converting the problems of dynamic task allocation and path planning of unmanned ship clusters into a problem of simulating biological network growth and resource optimization.
2. The method of claim 1, wherein dynamically adjusting the weight of the communication link comprises increasing the weight of the link when the information interaction frequency exceeds a predetermined reinforcement threshold, and decreasing the weight of the link when the information interaction frequency is below a predetermined weakness threshold.
3. The method of dynamic task allocation and path planning for an unmanned aerial vehicle cluster of claim 2, wherein said reducing the weight of said link comprises at least one of reducing the priority of said link in a data routing protocol or logically disconnecting said link.
4. The method of claim 1, further comprising receiving an operator command and setting a location, intensity, or type of the virtual information source in the virtual data field according to the command, wherein the type includes at least an attraction type information source and a repulsion type information source.
5. A system for dynamic task allocation and path planning for an unmanned ship cluster, the system comprising: a plurality of unmanned boats, each unmanned boat equipped with a communication module and a boat-mounted processor, wherein the boat-mounted processor is configured to: Receiving information representing an operating environment and comprising a virtual data field of one or more virtual information sources; generating a driving instruction to control the movement of the unmanned ship based on the local gradient of the virtual data field of the unmanned ship self position; Dynamically adjusting state parameters of communication links with other unmanned vessels based on information interaction historical data between the unmanned vessels and the other unmanned vessels, wherein the adjustment of the state parameters takes precedence over or is combined with physical layer signal indexes; The system is configured to implement the cluster global path planning and the self-organizing construction of a network topology by performing independent motions of each unmanned aerial vehicle based on local gradient calculations in combination with dynamic adjustment of the communication link state parameters; The system logically models each unmanned ship as a hypha network tip, models the communication link as hypha connected with the tip, and models the virtual information source as a nutrition source, thereby converting the problems of dynamic task allocation and path planning of unmanned ship clusters into a problem of simulating biological network growth and resource optimizing.
6. The system of claim 5, wherein the on-board processor is configured to dynamically adjust the status parameters by increasing the priority of the corresponding communication link in the routing table when the information interaction frequency with another unmanned ship satisfies a growth condition, and decreasing the priority of the corresponding communication link or marking it as inactive when the information interaction frequency satisfies a decay condition.
7. The system of claim 5, further comprising a remote control terminal configured to allow an operator to create, delete or modify the virtual information sources in the virtual data field to macroscopically direct the overall behavior of the unmanned ship cluster, rather than directly controlling individual unmanned ships.

Description

Dynamic task allocation and path planning method for unmanned ship cluster Technical Field The invention relates to the technical field of unmanned carrier cluster control, in particular to a dynamic task allocation and self-adaptive path planning method and system for an unmanned ship cluster in a water area environment with limited communication, high dynamic performance and large scale. Background Unmanned ship cluster, because of its possesses advantages such as wide coverage, low operation cost, risk evasion ability reinforce, demonstrate huge application potential in fields such as marine environment monitoring, waters safety patrol, emergent search and rescue and topography survey under water. However, the realization of efficient and robust collaborative operation of large-scale unmanned boat clusters is always faced with a series of well-established technical challenges, wherein three core contradictions between autonomous adaptability and task certainty of the clusters, individual efficiency and system toughness, and high intelligence of the clusters and unreliability of underwater communication channels are particularly prominent. In the prior art, for example, in the patent application with publication number CN117519163a published by the chinese patent office, a water surface unmanned ship cluster control system is disclosed, this patent is a scheme of uploading massive terminal data to a cloud platform for centralized analysis and then issuing an optimization instruction by a central controller, but such a centralized architecture has the inherent technical defects of delay of control response caused by data link delay, bandwidth bottleneck and central processing bottleneck in the application of unmanned cluster control field, especially in the environments with severe communication conditions such as water area, the loss of central node or communication interruption will directly cause paralysis of the whole cluster system. In order to overcome the defects of the centralized architecture, a plurality of distributed control algorithms are developed by a person skilled in the art, such as a negotiation algorithm based on an auction mechanism, a formation control algorithm based on a consistency theory and the like, however, the existing distributed scheme generally has the technical problems of decision failure or system breakdown when communication is limited due to excessive dependence on global information consistency, for example, the efficiency and success rate of task allocation are seriously dependent on reliable transmission of key information such as broadcasting, bidding, winning bid and the like based on an auction protocol, and under the underwater acoustic communication environment with high packet loss rate, the protocol may fall into endless retransmission and confirmation circulation, even task allocation failure is caused due to failure to achieve consensus, and combat efficacy cannot be maintained in a dynamic and antagonistic water area environment. The root of the above dilemma is that those skilled in the art, under the theoretical framework of classical control theory and computer network protocols, form a deep technological prejudice that cluster intelligence must be designed and implemented by accurate mathematical modeling and reliable, explicit information exchange, which prejudice results in a focus of research and development on how to overcome the unreliability of communication, such as developing more robust physical layer modem technologies, designing more complex network layer routing protocols, or constructing more accurate environmental predictive models to reduce communication requirements. These approaches remain essentially improved within the framework of instruction-execution and explicit communication, regarding the dynamic uncertainty of the communication link as a system noise and control barrier that needs to be suppressed or eliminated, without fundamentally exploiting the endogenous nature of the communication link. Disclosure of Invention In order to overcome the defects in the prior art, the invention provides the aquatic mycelium network cluster control system and the aquatic mycelium network cluster control method which can realize large-scale unmanned cluster toughness, high efficiency and controllable cooperation under an unreliable communication environment so as to solve the problems of control response lag, decision failure and insufficient system toughness in a high-dynamic and communication limited environment in the prior art. In order to achieve the above purpose, the present invention provides the following technical solutions: a dynamic task allocation and path planning method for unmanned ship clusters comprises the following steps: Step S1, establishing a virtual data field representing an operation environment, and defining a task target or an environment event as a virtual information source capable of generating a scalar concentra