Disclosure of Invention
In order to overcome a series of defects existing in the prior art, the application aims at providing an ocean mapping system based on intelligent navigation of a deepwater detection robot, which comprises the following modules:
the data acquisition module synchronously acquires navigation positioning data and submarine topography data through a multi-source navigation device and a submarine measurement device carried by the deepwater detection robot;
The navigation fusion module is used for realizing self-adaptive fusion of the multi-source data and outputting initial pose estimation;
The quality control module is used for carrying out quality evaluation and anomaly identification on the real-time mapping data, generating local retest requirements according to the evaluation result and ensuring that the mapping precision meets the requirements;
The path planning module is used for planning an optimal mapping path in real time based on the current pose, task requirements and local terrain characteristics of the robot and carrying out online adjustment according to the retest requirements of the quality control module;
The fault-tolerant control module is used for realizing multi-level navigation source fault diagnosis, and starting a degradation navigation strategy when navigation is abnormal or invalid so as to ensure that basic navigation and mapping functions can be maintained;
And the terrain mapping module is used for compensating and splicing submarine terrain data in real time by using the pose information provided by the navigation fusion module.
Further, the data acquisition module comprises the following components:
the navigation positioning unit is used for acquiring original navigation data of the deepwater detection robot through an integrated GPS, an inertial navigation system and sonar positioning equipment;
The submarine topography detection unit is used for acquiring submarine topography data in real time by utilizing multi-beam sonar and side-scan sonar technologies, wherein the submarine topography data comprises the depth, the shape and the characteristics of a seabed;
the environment sensing unit acquires marine environment parameters through an environment sensor and provides real-time data support of environment change during mapping;
The attitude monitoring unit is used for acquiring attitude data of the robot in real time by using the gyroscope and the accelerometer;
And the data preprocessing unit is used for carrying out preliminary processing and formatting on the collected original data and providing standardized data input for a subsequent module.
Further, the navigation fusion module comprises the following components:
The credibility evaluation unit is used for evaluating the data reliability of each navigation source in real time based on a deep learning algorithm and outputting a credibility score of each navigation source;
the weight adjusting unit dynamically adjusts the weight of each navigation source in the fusion calculation according to the reliability score;
an error modeling unit modeling error characteristics of each navigation source using an improved kalman filter algorithm;
The data synchronization fusion unit is used for carrying out time synchronization and integration on navigation data from different sources and calculating a preliminary pose estimation result according to the dynamic weight and the error model;
And the pose output unit outputs the fused preliminary pose estimation result and provides high-precision navigation information for subsequent path planning and mapping compensation.
Further, the reliability score of the navigation source is formulated as: Wherein,Reliability scoring for the ith navigation source, wherein the value is between 0 and 1 to represent the reliability of the navigation source, Wi is a weight matrix corresponding to the ith navigation source, xi is a feature vector of the ith navigation source, and bi is a bias term;
the weight of each navigation source in the fusion calculation is dynamically adjusted by the following formula: wi is the weight of the ith navigation source, corresponding to the confidence scoreProportional to the ratio; the sum of the reliability scores of all navigation sources is ensured to be 1, and normalization processing is carried out.
Further, the preliminary pose estimation result is expressed as: Wherein,Pi(ti) represents navigation data of the ith navigation source at the time ti; Is the change rate of the ith navigation source at the time ti, ti represents the timestamp of the ith navigation source, tsync represents the target synchronization time, ei(tsync) represents the error estimation value of the error model of the ith navigation source at the synchronization time tsync, and ei represents the error model of the ith navigation source.
Further, the path planning module includes the following components:
The task demand analysis unit analyzes task demands according to the targets and constraints of the mapping task and provides task-oriented constraint conditions for path planning;
the environment constraint analysis unit is used for analyzing local topographic features based on real-time submarine topography data and providing environment constraint information for path planning by combining environment perception data;
the path generation unit is used for learning and generating an optimal mapping path in real time based on the current pose, task requirements and terrain characteristics through a reinforcement learning algorithm;
the path optimizing unit dynamically adjusts the current path planning according to the retest demand or feedback information generated by the quality control module, and ensures that the path meets the mapping precision and task requirements;
The safety monitoring unit is used for adjusting the path in real time according to the terrain analysis result and the real-time environment perception, avoiding collision or entering a dangerous area and ensuring the safety of the robot;
and the path output unit converts the optimized path into a control instruction sequence and controls the robot to execute.
Further, the optimal mapping path is generated gradually through a reinforcement learning strategy pi*(sT), and is expressed as follows:,
Wherein aT is an action at the current time T, namely a path planning decision selected by the robot, w1、w2、w3 is a weight coefficient of task demand, terrain adaptability and progress optimization respectively, sT is a state of the robot at the time T, Rtask(sT,aT) is a task reward after taking action aT and used for evaluating the completion degree of the current task demand in the state sT, Rterrain(sT,aT) is a terrain adaptability reward after taking action aT and used for evaluating the adaptation condition of the current path planning on the seabed terrain in the state sT, Rprogress(sT,aT) is a progress reward after taking action aT and used for evaluating the contribution of the current path planning to the task progress in the state sT, gamma is a discount factor and used for balancing the relation between the current reward and future rewards, and V (sT+1) is a value function after the robot takes the optimal strategy in the state sT+1 and represents the maximum long-term reward obtained from the state sT+1.
Further, the fault tolerant control module includes the following components:
the navigation state monitoring unit monitors stability, accuracy and consistency of each navigation source data in real time and captures potential abnormal signals or failure signs;
the fault diagnosis unit is used for analyzing the type and severity of the abnormality of the navigation source based on a multi-level diagnosis algorithm and determining the cause and the influence range of the fault;
the redundant strategy management unit is used for managing the standby navigation source and the redundant strategy library, and rapidly switching to the suboptimal navigation source when the main navigation source fails so as to maintain the navigation function;
The degraded navigation execution unit starts a standby navigation scheme under the condition that a navigation source fails or is abnormal, and maintains a basic navigation function;
The abnormal data processing unit is used for uniformly processing abnormal data from each module and comprises data rejection, abnormal correction and interference suppression;
And the navigation performance evaluation unit evaluates the performance in the degraded navigation mode in real time, ensures that the performance meets the minimum task requirement, and tries to recover the normal navigation mode when the condition allows.
Further, the topographic mapping module includes the following components:
The topographic data compensation unit is used for compensating the collected submarine topographic data in real time according to the current pose information of the robot and correcting data deviation caused by positioning errors or sensor deviation;
The data splicing unit splices submarine topography data acquired in different positions and time periods to form a continuous topography map, so that seamless connection between the data is ensured;
the SLAM algorithm implementation unit optimizes the relation between the pose of the robot and the topographic data in real time based on the SLAM algorithm, and establishes and updates an accurate map of the submarine topography;
The closed-loop optimization unit optimizes the accumulated error of the robot position through closed-loop detection and correction of the SLAM algorithm, and ensures continuous improvement of the map and positioning accuracy;
the topographic feature analysis unit is used for extracting and analyzing features of the submarine topographic data and providing topographic feature information for other modules;
And the precision evaluation and output unit is used for performing precision evaluation on the spliced topographic data, outputting a final submarine topography map and providing a positioning optimization result to other modules.
Further, the data bias due to the positioning error or sensor bias is corrected by the following formula: Wherein Dfinal is corrected submarine topography data, (xraw,yraw) is the coordinate of original submarine topography data, which represents the initial coordinate of submarine topography in the measured data, and is not corrected, ex,ey is a positioning error, which represents the horizontal positioning error and the vertical positioning error of the robot respectively, eθ is an attitude error, which represents the orientation error of the robot, θ is the robot orientation, R (eθ) is a rotation matrix for correcting topography data deviation caused by the attitude error eθ, sx,sy is a sensor deviation, which represents systematic deviation or error of the sensor itself in the horizontal direction and the vertical direction respectively.
Compared with the prior art, the application has the following beneficial effects:
According to the application, through the integration of the multi-source navigation equipment and the submarine measurement equipment carried by the deepwater detection robot, high-efficiency and accurate submarine topography mapping is realized, and through intelligent navigation, path planning and fault-tolerant control, the continuity and reliability of mapping tasks are ensured, and meanwhile, the accuracy and the integrity of mapping data are improved.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention become more apparent, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiments of the present invention. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are some, but not all, embodiments of the invention.
All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiments described below, together with the words of orientation, are exemplary and intended to explain the invention and should not be taken as limiting the invention.
As shown in fig. 1, an ocean mapping system based on intelligent navigation of a deepwater detection robot comprises the following modules:
the data acquisition module synchronously acquires navigation positioning data and submarine topography data through a multi-source navigation device and a submarine measurement device carried by the deepwater detection robot;
The navigation fusion module is used for realizing self-adaptive fusion of the multi-source data and outputting initial pose estimation;
The quality control module is used for carrying out quality evaluation and anomaly identification on the real-time mapping data, generating local retest requirements according to the evaluation result and ensuring that the mapping precision meets the requirements;
The path planning module is used for planning an optimal mapping path in real time based on the current pose, task requirements and local terrain characteristics of the robot and carrying out online adjustment according to the retest requirements of the quality control module;
The fault-tolerant control module is used for realizing multi-level navigation source fault diagnosis, and starting a degradation navigation strategy when navigation is abnormal or invalid so as to ensure that basic navigation and mapping functions can be maintained;
And the terrain mapping module is used for compensating and splicing submarine terrain data in real time by using the pose information provided by the navigation fusion module.
The data acquisition module is a basic part of the system and is responsible for synchronously collecting various sensor data on the deepwater exploration robot. The robot performs real-time data acquisition by mounting multi-source navigation devices such as an Inertial Navigation System (INS), a sonar positioning system (SSBL), a depth gauge, and the like, and submarine measurement devices such as a sonar imaging system, a laser scanner, and the like. These devices are capable of providing important information including navigational positioning data (e.g., position, attitude, etc. of the robot) and seafloor topography data (e.g., seafloor topography, obstructions, etc.). The parallel acquisition of the multi-source data not only improves the integrity and the precision of the data, but also can effectively overcome the problems of errors and blind areas possibly caused by single equipment, and provides high-quality original data for subsequent navigation and mapping.
The navigation fusion module is one of core technologies in the system, and has the task of effectively fusing navigation data from different sensors and outputting initial pose estimation of the robot. The data of various sensors usually have certain errors, and how to comprehensively utilize the data through an intelligent algorithm and improve the positioning accuracy are key of the module. Common methods include Kalman filtering, particle filtering and the like, and the positioning accuracy after fusion is ensured by dynamically adjusting the weights of different sensors. The module not only can provide the position information of the robot in real time, but also can carry out self-adaptive adjustment on the weight of the sensor according to different environmental conditions, so that the robot can still be accurately positioned in a complex underwater environment, and an accurate basis is provided for subsequent path planning and topographic mapping.
The main function of the quality control module is to evaluate the quality of mapping data acquired in real time and identify abnormal data. Due to the complex submarine mapping environment, the acquired data may have errors or incomplete conditions under the influence of various factors such as water flow, robot motion state, sensor faults and the like. The module performs accuracy analysis and anomaly detection on each frame of data by designing a data quality evaluation algorithm, and can timely discover and mark potential error data. Meanwhile, the module generates local retest demands according to quality evaluation results, and when certain area mapping precision is found to be substandard, the robot is triggered to re-map the areas, so that the precision of final data is ensured. The technology effectively ensures the overall mapping precision of the system, and particularly ensures the reliability and safety of tasks in a deepwater environment.
The path planning module is responsible for planning an optimal mapping path in real time according to the current pose, task requirements and local terrain features of the robot. The module aims to ensure that the robot performs mapping in a complex submarine environment in an optimal path, so that task requirements can be met, obstacles can be avoided, and mapping time can be reduced. Meanwhile, the path planning module can dynamically adjust according to the local retest demand provided by the quality control module. For example, when mapping accuracy of a certain area is problematic, the path planning module automatically plans a retest path passing through the area, so as to improve accuracy and integrity of data. The design of the module can greatly improve the operation efficiency of the robot, reduce unnecessary path repetition and improve the overall benefit of mapping operation.
The fault-tolerant control module is a key guarantee module of the system and mainly realizes multi-level navigation source fault diagnosis and emergency response. When any navigation equipment has faults or anomalies, the fault-tolerant control module can detect and start corresponding fault emergency mechanisms in real time. For example, when the sonar positioning system fails, the module can automatically switch to the inertial navigation system, and the position of the robot is corrected through a complementary algorithm so as to maintain the navigation continuity. The module is designed through multi-level redundancy, so that the robot can maintain basic navigation and mapping functions even in a complex environment, the task interruption risk caused by equipment faults is reduced, and the high availability of the system is ensured.
The topographic mapping module compensates and splices the submarine topography data in real time by utilizing the pose information provided by the navigation fusion module. Due to the complexity of the subsea environment, the robot may be affected by water currents or obstacles during the movement, resulting in displacements or deviations during the data acquisition. Therefore, after the module collects the topographic data, the module can carry out space compensation according to the real-time pose of the robot, so that the collected topographic data can accurately reflect the real situation of the seabed. Meanwhile, the splicing algorithm can seamlessly splice the measurement data from different positions to generate a complete submarine topography. The high efficiency of the module ensures the continuity and accuracy of the submarine topography map, and provides a reliable basis for subsequent data analysis and decision making.
In a comprehensive view, the ocean mapping system based on intelligent navigation of the deepwater detection robot can realize high-precision and high-efficiency submarine topography mapping in a complex submarine environment through cooperative work of a plurality of technical modules. The data acquisition module provides rich original data, the navigation fusion module ensures accurate positioning information, the quality control module improves the reliability of the data, the path planning module optimizes the efficiency of mapping operation, the fault-tolerant control module ensures the stability of the system, and the topography mapping module finally generates an accurate submarine topography. The organic combination of the technical modules enables the system to show strong adaptability and reliability in deepwater exploration tasks, and has important significance for marine mapping operation.
Further, the data acquisition module comprises the following components:
the navigation positioning unit is used for acquiring original navigation data of the deepwater detection robot through an integrated GPS, an inertial navigation system and sonar positioning equipment;
The submarine topography detection unit is used for acquiring submarine topography data in real time by utilizing multi-beam sonar and side-scan sonar technologies, wherein the submarine topography data comprises the depth, the shape and the characteristics of a seabed;
the environment sensing unit acquires marine environment parameters through an environment sensor and provides real-time data support of environment change during mapping;
The attitude monitoring unit is used for acquiring attitude data of the robot in real time by using the gyroscope and the accelerometer;
And the data preprocessing unit is used for carrying out preliminary processing and formatting on the collected original data and providing standardized data input for a subsequent module.
In summary, through analysis of each component in the data acquisition module, it can be seen that the technical effect of each component plays a critical role in intelligent navigation of the deepwater detection robot. From the accurate position information provided by the navigation positioning unit, to the high-precision terrain data provided by the submarine topography detection unit, and to the real-time monitoring of the environment sensing unit and the attitude monitoring unit, all components together ensure that tasks can be efficiently and stably executed in complex and changeable marine environments. The data preprocessing unit provides optimized data input for the subsequent modules, and ensures the data fluency and accuracy of the whole system. The organic cooperation of each component ensures that the data acquisition module can provide comprehensive and reliable data support for the deepwater exploration robot, ensures the high efficiency and accuracy of the system and promotes the smooth completion of the submarine surveying and mapping task.
Further, the navigation fusion module comprises the following components:
The credibility evaluation unit is used for evaluating the data reliability of each navigation source in real time based on a deep learning algorithm and outputting a credibility score of each navigation source;
the weight adjusting unit dynamically adjusts the weight of each navigation source in the fusion calculation according to the reliability score;
an error modeling unit modeling error characteristics of each navigation source using an improved kalman filter algorithm;
The data synchronization fusion unit is used for carrying out time synchronization and integration on navigation data from different sources and calculating a preliminary pose estimation result according to the dynamic weight and the error model;
And the pose output unit outputs the fused preliminary pose estimation result and provides high-precision navigation information for subsequent path planning and mapping compensation.
The navigation fusion module realizes the efficient fusion of different navigation source data through the cooperative work of a plurality of key components such as reliability assessment, weight adjustment, error modeling, data synchronous fusion, pose output and the like, and provides accurate pose estimation for the deepwater exploration robot. By dynamically adjusting the weight of each navigation source, evaluating its reliability in real time, and accurately modeling the error characteristics of the sensor, the module is able to maintain high-precision, high-reliability navigation in complex and diverse marine environments. The module not only improves the adaptability of the robot in a complex submarine environment, but also provides powerful data support for subsequent path planning and topographic mapping.
Further, the reliability score of the navigation source is formulated as: Wherein,Reliability scoring for the ith navigation source, wherein the value is between 0 and 1 to represent the reliability of the navigation source, Wi is a weight matrix corresponding to the ith navigation source, xi is a feature vector of the ith navigation source, and bi is a bias term;
the weight of each navigation source in the fusion calculation is dynamically adjusted by the following formula: wi is the weight of the ith navigation source, corresponding to the confidence scoreProportional to the ratio; the sum of the reliability scores of all navigation sources is ensured to be 1, and normalization processing is carried out.
The core idea of the above formula is to dynamically evaluate the credibility of each navigation source and adjust its weight in the fusion calculation according to its reliability. The strategy based on credibility scoring and weight adjustment can flexibly cope with the changes of navigation sources under different environments and conditions, and ensures that the fused pose estimation is more accurate and reliable. By means of the method, the navigation fusion module can maintain high-precision navigation in a complex marine environment, and overall operation efficiency and precision of the deepwater detection robot are greatly improved.
Further, the preliminary pose estimation result is expressed as: Wherein,Pi(ti) represents navigation data of the ith navigation source at the time ti; Is the change rate of the ith navigation source at the time ti, ti represents the timestamp of the ith navigation source, tsync represents the target synchronization time, ei(tsync) represents the error estimation value of the error model of the ith navigation source at the synchronization time tsync, and ei represents the error model of the ith navigation source.
The calculation of the initial pose estimation result is realized by fusing the data of a plurality of navigation sources, and factors such as time deviation, change rate, error model and the like of each navigation source are considered. According to the method, the weight of each navigation source is dynamically adjusted, and the error is corrected, so that a high-precision pose estimation result can be provided for the deepwater exploration robot in a complex marine environment. The estimation result provides reliable navigation support for subsequent path planning and submarine mapping tasks, and improves the working efficiency and mapping accuracy of the whole system.
Further, the path planning module includes the following components:
The task demand analysis unit analyzes task demands according to the targets and constraints of the mapping task and provides task-oriented constraint conditions for path planning;
the environment constraint analysis unit is used for analyzing local topographic features based on real-time submarine topography data and providing environment constraint information for path planning by combining environment perception data;
the path generation unit is used for learning and generating an optimal mapping path in real time based on the current pose, task requirements and terrain characteristics through a reinforcement learning algorithm;
the path optimizing unit dynamically adjusts the current path planning according to the retest demand or feedback information generated by the quality control module, and ensures that the path meets the mapping precision and task requirements;
The safety monitoring unit is used for adjusting the path in real time according to the terrain analysis result and the real-time environment perception, avoiding collision or entering a dangerous area and ensuring the safety of the robot;
and the path output unit converts the optimized path into a control instruction sequence and controls the robot to execute.
The path planning module ensures that the deepwater detection robot can efficiently and safely execute mapping tasks in a complex submarine environment through cooperation of a plurality of components such as task demand analysis, environment constraint analysis, path generation, path optimization, safety monitoring, path output and the like. The task demand analysis unit provides constraint conditions for task guidance for path planning, the environment constraint analysis unit provides necessary environment constraints in combination with terrain and environment data, the path generation unit generates an optimal path by using a reinforcement learning algorithm, the path optimization unit carries out dynamic adjustment according to quality control feedback, the safety monitoring unit ensures the safety of the robot, and the path output unit ensures that the path is accurately executed. Through the coordination work of the modules, the path planning module can maximally improve the working efficiency and the task completion degree of the robot while guaranteeing the mapping precision and the task requirements.
Further, the optimal mapping path is generated gradually through a reinforcement learning strategy pi*(sT), and is expressed as follows:,
Wherein aT is an action at the current time T, namely a path planning decision selected by the robot, w1、w2、w3 is a weight coefficient of task demand, terrain adaptability and progress optimization respectively, sT is a state of the robot at the time T, Rtask(sT,aT) is a task reward after taking action aT and used for evaluating the completion degree of the current task demand in the state sT, Rterrain(sT,aT) is a terrain adaptability reward after taking action aT and used for evaluating the adaptation condition of the current path planning on the seabed terrain in the state sT, Rprogress(sT,aT) is a progress reward after taking action aT and used for evaluating the contribution of the current path planning to the task progress in the state sT, gamma is a discount factor and used for balancing the relation between the current reward and future rewards, and V (sT+1) is a value function after the robot takes the optimal strategy in the state sT+1 and represents the maximum long-term reward obtained from the state sT+1.
The generation of the optimal mapping path is realized step by step through a reinforcement learning strategy, and path selection is performed under each state according to a plurality of factors such as task requirements, terrain adaptability, progress and the like. By setting reasonable reward functions and weight coefficients and combining dynamic adjustment of discount factors, the reinforcement learning algorithm can help the robot to generate an optimal path meeting task requirements. The method not only can process complex submarine topography environment, but also can flexibly adjust the path according to real-time task demands and progress demands, and ensures that the mapping task can be efficiently and safely completed.
Further, the fault tolerant control module includes the following components:
the navigation state monitoring unit monitors stability, accuracy and consistency of each navigation source data in real time and captures potential abnormal signals or failure signs;
the fault diagnosis unit is used for analyzing the type and severity of the abnormality of the navigation source based on a multi-level diagnosis algorithm and determining the cause and the influence range of the fault;
the redundant strategy management unit is used for managing the standby navigation source and the redundant strategy library, and rapidly switching to the suboptimal navigation source when the main navigation source fails so as to maintain the navigation function;
The degraded navigation execution unit starts a standby navigation scheme under the condition that a navigation source fails or is abnormal, and maintains a basic navigation function;
The abnormal data processing unit is used for uniformly processing abnormal data from each module and comprises data rejection, abnormal correction and interference suppression;
And the navigation performance evaluation unit evaluates the performance in the degraded navigation mode in real time, ensures that the performance meets the minimum task requirement, and tries to recover the normal navigation mode when the condition allows.
The fault-tolerant control module ensures that the deepwater detection robot can continue to perform tasks and maintain basic functions when encountering navigation source faults or anomalies through multi-level monitoring, diagnosis, redundant switching and data processing. The navigation state monitoring unit captures abnormal signals of the navigation source in real time, the fault diagnosis unit provides accurate fault analysis, and the redundancy strategy management unit and the degradation navigation execution unit ensure that the system can be quickly switched to the standby navigation source and maintain task execution. The abnormal data processing unit and the navigation performance evaluation unit ensure the validity of the data and the stability of the system performance. Through cooperation of the components, the fault-tolerant control module enhances the robustness of the system, so that the robot can always maintain high-efficiency navigation capability in complex and variable environments, and smooth performance of submarine mapping tasks is ensured.
Further, the topographic mapping module includes the following components:
The topographic data compensation unit is used for compensating the collected submarine topographic data in real time according to the current pose information of the robot and correcting data deviation caused by positioning errors or sensor deviation;
The data splicing unit splices submarine topography data acquired in different positions and time periods to form a continuous topography map, so that seamless connection between the data is ensured;
the SLAM algorithm implementation unit optimizes the relation between the pose of the robot and the topographic data in real time based on the SLAM algorithm, and establishes and updates an accurate map of the submarine topography;
The closed-loop optimization unit optimizes the accumulated error of the robot position through closed-loop detection and correction of the SLAM algorithm, and ensures continuous improvement of the map and positioning accuracy;
the topographic feature analysis unit is used for extracting and analyzing features of the submarine topographic data and providing topographic feature information for other modules;
And the precision evaluation and output unit is used for performing precision evaluation on the spliced topographic data, outputting a final submarine topography map and providing a positioning optimization result to other modules.
The SLAM (Simultaneous Localization AND MAPPING) algorithm is a technique that builds maps in real time by sensors and simultaneously self-locates in an unknown environment. It is widely used in the field of robot navigation, autopilot, unmanned aerial vehicle flight, and other applications where tasks need to be performed in an unknown or dynamic environment. The core task of SLAM is to enable the robot to locate its own position and build a map of the surrounding environment without external positioning support (such as GPS).
The topography mapping module cooperates through a plurality of precision components, so that high-precision submarine topography mapping is realized. The terrain data compensation unit and the data splicing unit ensure the accuracy and seamless connection of data, and the SLAM algorithm realization unit and the closed-loop optimization unit ensure the continuous accuracy of the map by optimizing the pose and the map in real time. The topographic feature analysis unit provides valuable environmental information for the system, and the precision evaluation and output unit ensures that the mapping result meets the task requirement. By the technical means, the topography mapping module can provide a high-quality and real-time updated submarine topography map in a complex submarine environment, and a firm guarantee is provided for the deepwater exploration robot to finish an accurate submarine mapping task.
Further, the data bias due to the positioning error or sensor bias is corrected by the following formula: Wherein Dfinal is corrected submarine topography data, (xraw,yraw) is the coordinate of original submarine topography data, which represents the initial coordinate of submarine topography in the measured data, and is not corrected, ex,ey is a positioning error, which represents the horizontal positioning error and the vertical positioning error of the robot respectively, eθ is an attitude error, which represents the orientation error of the robot, θ is the robot orientation, R (eθ) is a rotation matrix for correcting topography data deviation caused by the attitude error eθ, sx,sy is a sensor deviation, which represents systematic deviation or error of the sensor itself in the horizontal direction and the vertical direction respectively.
Through the correction formula and the flow, the deepwater detection robot can effectively correct submarine topography data deviation caused by positioning errors, attitude errors and sensor deviation. The correction measures ensure the accuracy of the acquired submarine topography data, and provide a reliable basis for subsequent path planning, map construction and accuracy evaluation, so that the performance and operability of the whole ocean mapping system are improved.
Finally, it should be pointed out that the above embodiments are only intended to illustrate the technical solution of the invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will understand that modifications may be made to the technical solutions described in the foregoing embodiments or equivalents may be substituted for some of the technical features thereof, and that such modifications or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention in essence of the corresponding technical solutions.