CN-121995934-A - Underwater robot control algorithm based on bionic motion

CN121995934ACN 121995934 ACN121995934 ACN 121995934ACN-121995934-A

Abstract

The invention relates to the technical field of underwater robot control, and discloses an underwater robot ‌ control algorithm ‌ based on bionic motion. The algorithm obtains the environmental perception data of the underwater robot, derives an initial motion instruction through analysis and processing, and provides a basis for the initial action of the robot. And acquiring attitude monitoring data of the robot, and deciding whether to correct the initial motion instruction according to the data. When the correction is determined to be needed, water flow visual data around the robot is collected, water flow pattern features are extracted from the water flow visual data, feature description values are generated, and then the feature description values are compared with a historical feature library. If the history feature library lacks the same feature description value, the algorithm calculates the similarity index between the feature description value and the history feature, and corrects the initial motion instruction based on the index.

Inventors

HU XIAODONG
WU CHAO
XIONG XIHUI
ZHONG SHILIN
WU XIAOCHEN
YANG XUEJIE
ZHANG CHENWEI

Assignees

上海海达通信有限公司

Dates

Publication Date: 20260508
Application Date: 20260128

Claims (10)

1. The underwater robot control algorithm based on bionic motion is characterized by being realized through the following processing flow: Acquiring environment sensing data of the underwater robot and deducing an initial motion instruction according to the environment sensing data; Acquiring attitude monitoring data of the underwater robot, deciding whether to correct the initial motion instruction based on the attitude monitoring data, acquiring water flow visualization data around the underwater robot when deciding to correct, extracting water flow mode features from the water flow visualization data, generating feature description values according to the water flow mode features, comparing the feature description values with a historical feature library, correcting the initial motion instruction according to comparison results, calculating similarity indexes between the feature description values and the historical features when the historical feature library lacks the same feature description values, and correcting the initial motion instruction based on the similarity indexes.
2. The underwater robot control algorithm based on bionic motion according to claim 1, wherein the deriving of the initial motion command based on the environment-aware data includes that the environment-aware data includes water flow velocity information, water flow direction information and water depth information, the initial motion command is obtained by comparing the water flow velocity information with a standard velocity reference value to obtain a velocity adjustment factor, comparing the water flow direction information with a reference direction reference value to obtain a direction correction amount, comparing the water depth information with a target depth reference value to obtain a depth offset, integrating the velocity adjustment factor, the direction correction amount and the depth offset into a comprehensive parameter, and converting the comprehensive parameter into the initial motion command through a predefined mapping relation.
3. The underwater robot control algorithm based on the bionic motion according to claim 1, wherein the deciding whether to correct the initial motion command based on the attitude monitoring data includes extracting pitch angle data, roll angle data and yaw angle data from the attitude monitoring data, calculating a difference between the pitch angle data and a steady pitch threshold to obtain a pitch deviation, calculating a difference between the roll angle data and a steady roll threshold to obtain a roll deviation, calculating a difference between the yaw angle data and a steady yaw threshold to obtain a yaw deviation, comparing the pitch deviation, the roll deviation and the yaw deviation with an allowable deviation range, and deciding to correct the initial motion command if any deviation exceeds the allowable deviation range.
4. The underwater robot control algorithm based on bionic motion according to claim 1, wherein the extracting of the water flow pattern features from the water flow visualization data comprises analyzing the water flow velocity field image data and the water flow vortex image data by an image processing algorithm to obtain velocity distribution features including velocity gradient values and flow uniformity, analyzing the water flow vortex image data by a pattern recognition algorithm to obtain vortex features including vortex size and vortex strength, and the water flow pattern features are composed of the velocity distribution features and the vortex features.
5. The underwater robot control algorithm based on the bionic motion according to claim 4, wherein the generating of the characteristic description value according to the water flow pattern characteristic comprises performing quantization processing on the water flow pattern characteristic of each monitoring area, the characteristic description value is obtained by dividing the velocity gradient value by the area of the area to obtain a standardized velocity gradient, subtracting the flow uniformity from an ideal uniformity reference value to obtain a uniformity deviation, dividing the vortex size by a reference vortex size to obtain a size ratio, comparing the vortex strength with a strength threshold to obtain a strength factor, and weighting and fusing the standardized velocity gradient, the uniformity deviation, the size ratio and the strength factor into the characteristic description value.
6. The bionic motion-based underwater robot control algorithm according to claim 1, wherein the comparing the feature description value with a history feature library comprises traversing all history feature description values in the history feature library, searching a history record consistent with a current feature description value, searching a history motion instruction corresponding to the history record if the history record is consistent, taking the history motion instruction as a correction reference, and triggering a similarity index calculation flow if the history record is not consistent.
7. The underwater robot control algorithm based on the bionic motion according to claim 6, wherein when there is a consistent history, the modifying the initial motion command according to the comparison result includes extracting the history motion commands corresponding to all the consistent histories, calculating an average value of the history motion commands as an average motion command, mixing the average motion command with the initial motion command, the mixing ratio being determined based on the number of histories, and the mixed command being the modified motion command.
8. The underwater robot control algorithm based on bionic motion according to claim 6, wherein the calculating of the similarity index between the feature description value and the historical feature comprises calculating euclidean distance between each feature description value and each historical feature description value in the historical feature library, taking the reciprocal of the minimum euclidean distance as a similarity score, and normalizing the similarity scores of all feature description values to obtain the similarity index.
9. The underwater robot control algorithm based on the bionic motion according to claim 8, wherein the modifying the initial motion command based on the similarity index includes dividing an adjustment level according to the size of the similarity index, the adjustment level including a high adjustment level, a medium adjustment level, and a low adjustment level, presetting an adjustment coefficient for each adjustment level, and multiplying the adjustment coefficient with the initial motion command to obtain the modified motion command.
10. The underwater robot control algorithm based on bionic motion according to claim 1, wherein the algorithm further comprises a history feature library updating process, wherein after finishing motion instruction correction, the current feature description value and the corresponding corrected motion instruction are stored in the history feature library.

Description

Underwater robot control algorithm based on bionic motion Technical Field The invention relates to the technical field of underwater robot control, in particular to an underwater robot ‌ control algorithm ‌ based on bionic motion. Background The ocean occupies about 71% of the surface of the earth, contains rich resources such as petroleum, natural gas, mineral products, biological resources and the like, and is also an important field of numerous scientific researches, and plays a key role from the research of an ocean ecosystem to the exploration of the climate change of the earth. In military aspect, the strategic significance of ocean space is increasingly prominent, and the tasks of underwater monitoring, investigation and the like are vital to national defense safety. The underwater robot is used as an important tool for exploring the ocean by human beings, can break through the limit of the operation of the human beings under the water, such as bearing high pressure, adapting to darkness, complex water flow environment and the like, so that the human beings can know the ocean more deeply and comprehensively, and perform activities such as resource exploration, scientific research, military application and the like. In deep sea oil and gas exploration, an underwater robot can carry various sensors and equipment to detect submarine topography and geological structures in detail, determine the distribution condition of oil and gas resources and provide key data for subsequent exploitation work. In marine science research they can be used to collect marine life samples, monitor marine environmental parameters, help scientists reveal the mystery of the marine ecosystem. The traditional underwater robot control algorithm presents obvious defects when facing complex and changeable underwater environments. In the area of the surging of the hidden stream, the speed and the direction of the water flow are continuously changed, the traditional algorithm is difficult to accurately sense the changes in real time and adjust the movement of the robot, so that the navigation track of the robot is deviated, and even the robot is possibly flushed away by the hidden stream, and the preset task cannot be completed. In the reef area, the robot needs to accurately control the position and the posture of the robot to avoid collision, but the traditional algorithm is difficult to flexibly shuttle in complex reef terrain due to limited sensing and analysis capability of surrounding environment, and collision accidents are easy to occur to damage the robot equipment. The conventional algorithm is not flexible enough to modify the instruction after generating the initial motion instruction. When the underwater environment is dynamically changed, such as when a sudden obstacle or a sudden water flow mutation occurs, the conventional algorithm cannot effectively adjust the initial motion instruction in time. This is because conventional algorithms tend to make decisions based on preset rules and models, lacking the ability to quickly respond to real-time changing conditions. In practical application, when the underwater robot advances according to the initial instruction, if a large obstacle suddenly appears in front, the conventional algorithm may not be able to quickly make avoidance actions, resulting in collision between the robot and the obstacle. The underwater environment sensing data contains a large amount of information such as sonar data, visual image data, water flow speed and direction data and the like. The traditional control algorithm has defects in processing the mass data, and can not extract key information from the data rapidly and accurately and perform effective analysis and decision. Traditional sonar data processing algorithms may not be able to identify tiny targets in complex seafloor terrain in time, or when processing visual image data, marine organisms or other objects of interest may not be able to be accurately identified due to algorithm limitations. The underwater robot is difficult to make accurate judgment and decision when facing complex environments, and the operation efficiency and accuracy of the underwater robot are affected. Disclosure of Invention The invention aims to provide an underwater robot ‌ control algorithm ‌ based on bionic motion so as to solve the problems in the background art. To achieve the above object, the present invention provides an underwater robot ‌ control algorithm ‌ based on bionic motion, the algorithm comprising: Acquiring environment sensing data of the underwater robot and deducing an initial motion instruction according to the environment sensing data; Acquiring attitude monitoring data of the underwater robot, deciding whether to correct the initial motion instruction based on the attitude monitoring data, acquiring water flow visualization data around the underwater robot when deciding to correct, extracting water flow mode features from the water flow vis