20results about How to "Improve navigation accuracy" patented technology

The application discloses a visual language navigation method based on adaptive prompt information during testing, comprising the following steps: obtaining natural language-based navigation instructions and 360-degree panoramic visual observation information of an intelligent agent; extracting low-frequency visual features of sub-images and selecting valuable information from an image memory bank for prompting; extracting a text feature vector corresponding to the navigation instructions and an initial state vector of the current environment; selecting historical information from the visual feature memory bank using the state vector for prompting; inputting the state vector and the instruction features into a visual-language interaction module to generate navigation action selection; updating the normalization layer in the full connection layer according to the prediction score of the navigation action selection; and repeating the above steps until the intelligent agent stops selecting or reaches the maximum number of movement steps. The application can effectively enhance the navigation performance of the pre-trained intelligent agent model in unknown test environments.

This application provides a self-calibration method for the magnetic shielding parameters of a laser gyroscope IMU, relating to the fields of inertial navigation and precision instrument technology. The method includes the following stages: Offline initialization modeling stage: Applying a multi-dimensional standard excitation magnetic field to the IMU, simultaneously acquiring external excitation magnetic field data, residual magnetic field data inside the IMU, control parameter data of the magnetic shielding system, and zero-bias drift data of the laser gyroscope. This application provides a self-calibration method for the magnetic shielding parameters of a laser gyroscope IMU, abandoning the model of permanently fixing magnetic shielding parameters. It creatively proposes a data-driven online real-time parameter optimization mechanism, which can automatically adapt to wide temperature range changes and shielding performance degradation caused by material aging, ensuring that the IMU always operates in a state where the magnetic shielding effectiveness is approximately optimal.

The invention belongs to the technical field of positioning and navigation, and particularly discloses a GNSS / DR (Global Navigation Satellite System / Digital Radio) tight integration navigation method based on an SINS (Strapdown Inertial Navigation System). The method comprises the following steps of: inputting a difference value between a pseudo range calculated by a GNSS (Global Navigation Satellite System) and a position calculated by an SINS (Strapdown Inertial Navigation System) into a filter as an observed quantity, and performing state error estimation; and correcting a navigation positioning result by using a state error estimation result, and outputting the corrected positioning result as a tightly integrated navigation positioning result. According to the scheme, the technical problems that an existing SINS / GNSS integrated navigation system is weak in anti-interference capacity, poor in system observability and low in precision are solved.

This invention discloses a long-range underwater vehicle with retractable wings, belonging to the technical field of underwater exploration equipment. It includes a vehicle body with a shell and a propulsion system. The vehicle body is characterized by further including a rotating shaft, retractable wings, and a wing drive motor. The rotating shaft is symmetrically mounted on both sides of the shell, forming a rotating pair with the shell. The axis of the rotating shaft is neither parallel nor intersecting with the axis of the vehicle body. Two retractable wings are respectively mounted on the rotating shaft. The wing drive motor is connected to and drives the rotating shaft to rotate. This vehicle, through the deployment and retraction of the retractable wings, can, from a physical structure perspective, possess both the gliding motion mode of an underwater glider with a sawtooth trajectory and the autonomous high-speed, constant-depth navigation mode of an autonomous underwater vehicle. It combines endurance and maneuverability, and features a simple structure and low failure rate.

The invention provides carrier rocket autonomous integrated navigation equipment and a carrier rocket autonomous integrated navigation method based on a star sensor. The device is composed of an inertial measurement combination, a star sensor and an MEMS backup inertial navigation unit. The method comprises the following steps: finishing space-time alignment before emission; during flight, inertial navigation is used as a trunk, a star sensor is used for auxiliary correction, and in the atmosphere, through low-frequency observation, a least square method is used for fitting and correcting the attitude in a sliding section; and switching to a high-frequency observation mode at the tail stage of flight, and carrying out real-time optimal estimation and correction on navigation parameters and inertial device errors through Kalman filtering in combination with a GNSS (Global Navigation Satellite System) so as to ensure high-precision orbit injection. And when the main inertial navigation precision is abnormal, the system is automatically switched to the backup unit. According to the scheme, the reliability, the adaptability and the overall precision of the system are remarkably improved through the hardware redundancy design of the main / standby dual-inertial navigation system, the combination of front and later intelligent segmentation correction and inertial measurement, and the combined navigation fusion algorithm of the star sensor and the GNSS.

The invention relates to a surgical robot navigation system, method and medium. The surgical robot comprises a patient surgical trolley and a mechanical arm, the axes of all moving joints of the mechanical arm intersect at a mechanical fixed point, and the patient surgical trolley comprises an adjusting component used for adjusting the position of the surgical robot. The system comprises a fixing device and at least two groups of light emitters, the fixing device is arranged at the mounting position of the surgical robot, so that the ray path of the light emitted by the at least two groups of light emitters passes through the mechanical fixed point without being shielded; the at least two groups of light emitters are fixed on the surgical robot through a fixing device and are configured to project light rays with a target graph; the position of the surgical robot is adjusted through an adjusting component until the light rays emitted by the at least two sets of light emitters intersect at the mechanical fixed point and form a closed projection graph, and the mechanical fixed point coincides with the target surgical position. The method can improve the navigation accuracy.

The invention discloses an SINS / LBL tightly integrated navigation method and device based on double equidistant transformation groups SE2 (3), and relates to the technical field of autonomous navigation of underwater vehicles. Comprising the following steps: acquiring navigation parameters of an inertial navigation system, and converting inertial navigation mechanical arrangement under an earth coordinate system through auxiliary gravity and auxiliary speed to obtain converted earth coordinate system mechanical arrangement; taking navigation parameters as group elements of SE2 (3), and defining an invariant error state; error modeling is carried out on the SE2 (3) through logarithm mapping; deriving a system matrix and a noise distribution matrix to obtain an invariant error state space model; deducing a conversion relation between the measurement information and an invariant error state to obtain an SINS / LBL tight combination measurement model; generating a sampling point in the error Lie algebra so as to map the sampling point to an SE2 (3) group and execute nonlinear measurement updating to obtain an updated error state; and feeding back to a navigation calculation result of the inertial navigation system to obtain a parameter estimation result. The starting efficiency and the navigation calculation precision of the integrated navigation system are improved.

The application discloses a SINS / DVL installation error angle calibration method based on optimization alignment, and the method comprises the following steps: (1) constructing a SINS / DVL combined navigation motion model containing installation error angles; (2) designing an installation error angle parameter optimization target function; and (3) designing a SINS / DVL installation error angle optimization estimation algorithm. The application provides a real-time online installation angle error calibration algorithm which only relies on SINS and DVL output data, and effectively estimates the installation error between the SINS and the DVL in a complex underwater environment without GNSS signals, thereby improving the navigation performance of the SINS / DVL combined navigation.

This invention belongs to the field of inertial astronomical integrated navigation technology, specifically relating to a multi-source cooperative integrated navigation method based on inertial navigation system state awareness, comprising: S1, setting a main inertial navigation system and at least one sub-inertial navigation system in the integrated navigation system; S2, performing main-sub-inertial navigation system communication state detection, and adopting different processing strategies according to the length of the interruption when the main-sub-inertial navigation system communication is detected to be interrupted; S3, performing main-sub-inertial navigation system communication state detection, and performing fault self-reconstruction after the main-sub-inertial navigation system communication is restored; S4, performing main inertial navigation system navigation state detection within the sub-inertial navigation system, and designing filters within the sub-inertial navigation system based on the main inertial navigation system state, thereby enabling a multi-source cooperative integrated navigation method that can dynamically respond to the main-sub-inertial navigation system communication state, possesses fault self-reconstruction capability, and adaptability to state switching.

This invention discloses a zero-shot navigation method based on dynamic navigation chains and multi-dimensional value maps, comprising the following steps: acquiring natural language commands of different navigation types and collecting multimodal perception data; converting the natural language commands into a unified chain-like planning paradigm, and updating the action sequence, landmark, and target object relationship co-occurrence probability list of the navigation chain; generating scene point clouds and navigable maps, and constructing semantic value maps, action value maps, trajectory value maps, intuitive value maps, and relationship co-occurrence value maps; performing pattern matching on the action sequences of the navigation chain, triggering a swing suppression mechanism when a swing state is detected, and performing differentiated action constraints and action value map assignment updates according to the swing sequence type; generating a global decision value map by weighted linear superposition of various value maps; obtaining the next navigation target point by maximizing the value of the global decision value map, and planning a feasible trajectory until a preset termination navigation condition is reached. This invention can improve navigation efficiency and accuracy.

This invention discloses a single-base station UWB / inertial fusion navigation method based on an antenna array, belonging to the field of integrated navigation. This invention collects... k The system uses distance data between the UWB base station and the tag, carrier phase data received by the UWB base station array antenna, accelerometer data, and gyroscope data. It determines whether the navigation system is initialized; if so, it uses the carrier phase data received by the array antenna to calculate the tag's azimuth and elevation angles in the UWB coordinate system, and performs pre-integration between adjacent UWB measurement frames using inertial sensor data. It then optimizes the carrier navigation information by combining UWB ranging, angle measurement errors, and inertial pre-integration errors, and outputs the carrier's navigation information. This invention overcomes the shortcomings of multi-base station UWB / inertial fusion methods in terms of base station installation and deployment, and time synchronization, and improves navigation accuracy by performing global optimization of the carrier's pose through tight coupling.

This invention relates to a meta-learning-based intelligent modeling and compensation method for gyroscope temperature drift, belonging to the field of inertial sensing and intelligent navigation. Addressing the complex temperature drift characteristics of fiber optic gyroscopes, this invention first constructs a database of gyroscope temperature drift under different temperatures and input angular velocities. Secondly, in the offline stage, by combining least-squares analytical solutions with neural network learning residuals, gyroscope temperature drift is modeled as a structured function driven by both angular velocity features and environmental parameters. Then, a meta-learning strategy is used to train the feature extraction network and the parameterization network to obtain an initial model with generalization capabilities across temperature conditions. Finally, in the online stage, the model is rapidly adapted using a small sample size, and Kalman filtering is used to fine-tune the model parameters, achieving real-time estimation and compensation of gyroscope temperature drift. This invention improves the accuracy of gyroscope error estimation and compensation in complex environments, effectively enhancing the navigation performance of inertial navigation systems.

The application is suitable for the technical field of dynamic path planning, and particularly relates to an industrial robot dynamic path planning learning method and system.The method comprises the following steps: S100: demarcating a working area of an industrial robot, selecting a deployment position of a plurality of marker lights in each industrial robot, setting the color of each marker light, and generating a combined code, wherein the combined code of each industrial robot is different; collecting video monitoring data containing all industrial robots by using a video monitoring device pre-deployed in the working area; according to a preset step, a plurality of snapshots are intercepted, and the position of the industrial robot is located through the combined code; and S200: grid division is performed on the working area, and a reference point is selected.The application can realize incremental planning of the industrial robot by defining a conflict line, can dynamically adjust the moving route, is convenient for quickly adapting to changes in obstacles, and greatly improves the obstacle avoidance efficiency of the industrial robot.

The invention discloses an intelligent fusion navigation filtering method and system for complex environment transition. Navigation data and environment data are collected by a sensor arranged on ocean equipment, preliminary navigation state estimation is obtained after the navigation data passes through a strapdown inertial navigation module, and local error state estimation and filter health degree are obtained after the navigation data and the navigation data are input into a parallel sub-filter; combining the navigation data and the environment data with the health degree of the filter to serve as fusion data, inputting the fusion data into a reinforcement learning agent to obtain a dynamic fusion weight, performing weighted fusion on the plurality of local error state estimations by using the dynamic fusion weight, and superposing the fusion data on the initial navigation state estimation to generate a final navigation result. According to the method, the optimization mapping relation between the environment characteristics and the filtering weight is directly learned through the reinforcement learning agent, integrated self-adaptive adjustment from environment perception to fusion decision is achieved, and the estimation precision and robustness of a navigation system in dynamic complex environments such as a bay-ocean transition section are effectively improved.

The utility model discloses intelligent blind person navigation glasses, and belongs to the technical field of glasses. The glasses mainly comprise a glasses frame, lenses and glasses legs, and further comprise an intelligent auxiliary system module arranged on the glasses frame, and the intelligent auxiliary system module comprises a main control chip, a visual environment sensing module, a motion detection path navigation module, a language interaction module, a wireless communication remote connection module and a real-time feedback module. According to the intelligent blind person navigation glasses, all the modules are coordinated through the main control chip, visual perception, motion detection, language interaction, wireless communication and implementation feedback are combined, the effects of precise navigation, environment perception and real-time interaction are achieved, and higher navigation precision, safety and user experience are brought.

Embodiments of the application disclose a virtual world navigation method, device, equipment, medium and program product, and belong to the field of intelligent navigation. The method comprises displaying a virtual map of a virtual world; wherein the virtual world comprises at least one closed area, and the closed area comprises a virtual entrance available for passing; displaying a starting position identifier for indicating a starting position of a virtual role and a target position identifier for indicating a target position on the virtual map; in response to a first inside-outside position relationship of the starting position relative to the closed area being different from a second inside-outside position relationship of the target position relative to the closed area, displaying a first route passing through the starting position identifier and a virtual entrance identifier of the virtual entrance, and a second route passing through the virtual entrance identifier and the target position identifier on the virtual map; wherein the virtual entrance is a necessary node of a segmented navigation path; and the application can improve the navigation efficiency in the closed area scenario.

The present disclosure relates to a navigation method, a navigation device and a mobile carrier, and the navigation method is executed by the mobile carrier, comprising: obtaining current position information of the mobile carrier, moving along a preset guiding track according to the current position information and target position information obtained; and sending the current position information to a server. The scheme provided by the present disclosure makes the position information of the mobile carrier obtained by the server more accurate, so as to improve the navigation accuracy.

This invention discloses a SINS / LBL compact integrated navigation method and device based on a dual isometric transformation group SE2(3), which relates to the field of autonomous navigation technology for underwater vehicles. The method includes: acquiring navigation parameters of the inertial navigation system; transforming the inertial navigation mechanical arrangement in the Earth coordinate system using auxiliary gravity and auxiliary velocity to obtain a transformed Earth coordinate system mechanical arrangement; using the navigation parameters as group elements of SE2(3) and defining an invariant error state; performing error modeling on SE2(3) through logarithmic mapping; deriving the system matrix and noise allocation matrix to obtain an invariant error state space model; deriving the transformation relationship between measurement information and the invariant error state to obtain a SINS / LBL compact integrated measurement model; generating sampling points in the error Lie algebra to map to the SE2(3) group and performing nonlinear measurement updates to obtain the updated error state; and feeding this back to the navigation solution results of the inertial navigation system to obtain parameter estimation results. This invention improves the startup efficiency and navigation calculation accuracy of the integrated navigation system.

This invention belongs to the field of autonomous driving positioning and navigation technology, specifically a method for autonomous vehicle trajectory and navigation image projection calibration. The main steps include: acquiring map data of the actual scene via satellite to establish a high-precision navigation map and image; controlling the vehicle to drive on the road and obtaining the vehicle's driving trajectory via GPS; extracting the pixel corner trajectory shape of the actual road in the navigation image, calculating the GPS corner trajectory shape based on the trajectory recorded in GPS, and generating point-pair data by connecting the discrete points of the GPS trajectory with the image pixels; performing coordinate system transformation on the point-pair data through similarity transformation, and projecting the discrete points of the GPS trajectory onto the high-precision navigation image through calibration transformation to complete the projection calibration. This method solves the problems of geometric distortion and computational efficiency in existing affine transformation models, reduces the dependence on high-precision sensors, and adapts to the navigation needs of autonomous vehicles in road scenarios.

This invention discloses an automatic projection focusing intraoperative navigation system for fluorescence imaging, relating to the field of fluorescence imaging navigation technology. A focusing drive unit drives a near-infrared camera lens assembly and a projection module for optical focusing. The near-infrared camera lens assembly receives fluorescence signals and forms a focal plane image. The projection module projects the processed fluorescence signal onto the tissue surface in real time. A transmission calibration module is coupled to the focusing drive unit for synchronous focusing, ensuring that the projection position of the fluorescence signal on the tissue surface matches the imaging position. This invention achieves synchronous focusing of the near-infrared camera and projection module through a unified linkage focusing mechanism and introduces adjustable calibration coefficients based on their respective optical characteristics. This solves the problems of difficulty in matching the imaging focal plane and the projection focal plane, cumbersome focusing operations, and repeated intraoperative calibration in existing systems. It improves the spatial consistency between intraoperative fluorescence imaging and projection, optimizing surgical navigation efficiency and accuracy.