91 results about "Odometer" patented technology

The application discloses a kind of tunnel construction dynamic measurement methods based on inertial navigation auxiliary, comprising: obtaining tunnel construction priori information and stage division parameter;With the measuring equipment of loaded inertial measurement unit, laser radar and odometer, it is installed on construction machinery, travels along tunnel axis, and real-time collection multi-source data;Based on point cloud feature density, the stage where it is located is automatically identified, and corresponding multi-level adaptive constraint model is dynamically switched, including adaptive nonholonomic kinematics constraint, environmental feature constraint based on point cloud data and design section, and absolute correction constraint based on template matching automatically identifying natural existing reserved reference point in tunnel;Extended Kalman filter is constructed to fuse multi-level constraint and dynamically adjust weight;Through loop detection and graph optimization, cumulative error is eliminated and control network is updated.The application realizes adaptive high-precision dynamic measurement of whole process of tunnel construction, does not need artificial layout beacon, effectively solves the problem of unstable positioning accuracy in complex environment.

The present application relates to the field of agricultural intelligent equipment and plant protection technology, and discloses a method for preventing and treating crop diseases based on precise ultraviolet irradiation and cross-ridge operation equipment, comprising the following steps: a main control module drives the chassis to cruise autonomously, and a self-adaptive adjustment module adjusts the damping force in real time according to the attitude data; a light supplementing module dynamically adjusts the light to cooperate with the identification module to obtain disease information; the main control module calculates the triggering condition based on the running speed and the installation distance between the identification module and the ultraviolet irradiation array, controls the corresponding ultraviolet unit to carry out pinpoint irradiation on the disease target; and an operation traceability module generates encrypted tamper-proof data and uploads it. The present application realizes precise irradiation through odometer pulse synchronization and speed dose compensation, maintains the operation attitude stable under complex terrain through active suspension control, and realizes reliable traceability of operation data through chain encryption technology, solving the problems of lack of precision in traditional prevention and treatment, unstable equipment attitude and difficult data traceability.

This invention discloses a USV berthing and unberthing pose estimation method based on ship-shore LiDAR collaboration, comprising: acquiring shipborne and shore-based point clouds and shore-based fixed reference poses; performing odometer calculations on continuous shipborne point clouds to generate LiDAR odometer factors; constructing a global descriptor for the ship-shore point clouds and performing scene matching to generate candidate ship-shore point cloud pairs; extracting key points and calculating FPFH descriptors, solving for high-precision ship-shore relative poses through coarse / fine registration, obtaining the final ship-shore relative poses through asynchronous reconstruction and generating correlation factors; constructing a collaborative factor graph model containing odometer and ship-shore correlation factors; solving the model and minimizing the overall residuals to obtain a globally consistent optimal pose, thereby achieving berthing and unberthing control. This invention improves ship-shore data transmission efficiency and matching accuracy through global descriptor matching and FPFH feature compression; and eliminates the cumulative error of single-ship odometers by leveraging shore-based global pose constraints, achieving high-precision continuous pose estimation under long-term navigation and ensuring the robustness of berthing and unberthing navigation.

The application discloses a heading information determination method, which comprises the following steps: utilizing a rotation mechanism built in a strapdown inertial navigation system, and according to a preset rotation sequence of a fixed period, performing rotation adjustment on a prism normal of an inertial measurement unit of the strapdown inertial navigation system; utilizing a real-time prism normal of the inertial measurement unit to correct a posture matrix collected by the strapdown inertial navigation system and a speed collected by an odometer; and taking the corrected posture matrix and the corrected speed as heading information. Thus, the rotation modulation of a gyroscope and an accelerometer can be realized by utilizing the rotation mechanism built in the strapdown inertial navigation system, a long-time heading keeping method of a vehicle-mounted inertial / odometer based on the rotation modulation is realized, the constant error source except for the axial (rotation axis direction) error can be automatically offset, the long-time heading precision of the system is improved, and the heading information precision is improved.

This invention provides an IMU mounting angle estimation method based on gravity projection and displacement recursion, comprising: establishing a gyroscope zero-bias estimation model for the vehicle and an accelerometer-gravity projection correlation model in a stationary state; when the vehicle is stationary, verifying the validity of the accelerometer measurements through the consistency of gravity vector magnitude, and simultaneously establishing constraint equations with the gravity projection to obtain the roll and pitch angles; during the short-term straight-line movement phase of the vehicle, performing mechanical choreography to obtain the horizontal displacement in the IMU coordinate system, and solving for the heading mounting angle based on the geometric correlation between the horizontal displacement and the heading mounting angle. This invention achieves integrated estimation of the three mounting angles by solving the roll and pitch mounting angles in the stationary phase and the heading mounting angle after startup, adapting to the initial calibration scenario of vehicle inertial navigation, eliminating the need for external references such as GNSS and odometers, and significantly simplifying the implementation conditions of mounting angle calibration.

This application provides a robot localization method, a computer program product, and a storage medium. The robot localization method includes: acquiring image data and odometer data of a target cage during robot movement; acquiring coordinate information of feature structures on the target cage from the image data; and determining the robot's localization data based on the coordinate information and the odometer data. This eliminates the need for additional QR code tags, positioning base stations, etc., reducing costs and making the robot less susceptible to interference. Furthermore, by combining the coordinates of feature structures in the image with the odometer data, the robot's position can be comprehensively determined, improving the accuracy of robot localization.

A train positioning method, apparatus, and device based on multi-source information fusion, relating to the field of railway positioning technology, includes: acquiring lidar data, IMU data, and train image data; fusing the lidar data and IMU data to obtain a laser inertial odometer trajectory, and fusing the IMU data and train image data to obtain a visual inertial odometer trajectory; identifying kilometer markers in the train image data and calculating the distance between the kilometer markers and the train to obtain the train's absolute position; correcting the laser inertial odometer trajectory and the visual inertial odometer trajectory based on the train's absolute position; performing degradation detection, and fusing the corrected laser inertial odometer trajectory and the visual inertial odometer trajectory using a Kalman filter to calculate the train's real-time position and positioning information interval; this method, based on kilometer marker correction of the laser inertial odometer and visual inertial odometer trajectory and fusion of positioning results, achieves higher positioning accuracy than existing methods and exhibits good robustness.

This invention provides an IMU installation angle estimation method based on threshold-triggered velocity change recursion, comprising: establishing a gyroscope zero-bias estimation model for the vehicle and an accelerometer-gravity projection correlation model in a stationary state; constructing a horizontal reference coordinate system to eliminate roll and pitch tilt biases, transferring IMU measurements to the horizontal reference coordinate system, and setting a threshold to initiate the recursion of velocity change; substituting the direction cosine matrix containing only the heading angle and the vehicle system velocity change vector into the transformation formula to solve for the heading installation angle, thereby obtaining a complete estimate of the three-axis installation angles. This invention calculates the roll and pitch installation angles when stationary, and calculates the heading installation angle through threshold-triggered velocity change recursion, achieving integrated estimation of the three installation angles. It is suitable for initial calibration scenarios of vehicle-mounted inertial navigation, requires no external references such as GNSS or odometers, and significantly simplifies the implementation conditions of installation angle calibration.

The application provides a railway map matching method and system based on a Beidou navigation system and an odometer, relates to the technical field of heavy-haul railway transportation, and comprises the following steps: acquiring first state data of a train based on the Beidou navigation system, and acquiring second state data of the train based on the odometer; performing fusion processing on the first state data and the second state data, determining an observation state of the train at each moment and corresponding state transition probability and state emission probability; and determining an optimal matching path of the train at each moment based on the observation state of the train at each moment and the corresponding state transition probability and state emission probability. The method and system provided by the application improve the positioning accuracy and stability of heavy-haul railway vehicles in complex environments.

The application relates to the technical field of navigation positioning of pipe network inspection equipment, and particularly discloses a high-precision positioning system for pipe network inspection equipment based on multi-sensor fusion. In view of the problem that no GNSS signal exists in the pipe network and long-distance inspection leads to serious cumulative drift of positioning, the system integrates IMU, a wheeled odometer and a three-dimensional laser radar, realizes space-time synchronization at the hardware level through a preprocessing module, uses an ESKF fusion model and IMU pre-integration for high-frequency pose estimation at the front end, corrects global errors through a factor graph optimization algorithm in combination with key node features and a global topological prior map at the back end, and introduces measurement weight adaptive adjustment logic to cope with environmental degradation. The application effectively solves the problem that the positioning accuracy is difficult to maintain for a long time in the complex environment of the underground pipe network, and significantly improves the reliability and trajectory consistency of the inspection operation.

This invention provides a laser SLAM inspection vehicle and method suitable for complex underground infrastructure environments. A multi-line lidar system is equipped with multiple laser emission and reception channels to generate high-density, large-area 3D point cloud data. A high-precision inertial measurement unit (IMU) provides the system with high-frequency attitude, angular velocity, and acceleration information. A wheeled odometer collects wheel speed data to provide short-term pose increment information, effectively supplementing the lidar and IMU. An edge computing device performs real-time data processing using SLAM algorithms. An embedded motion control unit runs an operating system to precisely control the chassis motors and transmit data with various subsystems. The software framework is built on the operating system and includes distributed nodes, a publish / subscribe communication model, and an open-source toolkit ecosystem. This achieves lightweight 3D perception capabilities, strong anti-interference performance, and supports fully automatic online repositioning, making it suitable for use in underground infrastructure such as subway tunnels, urban utility tunnels, and large warehousing centers.

This invention discloses an online cycle time reconstruction method for multi-process linkage equipment based on digital twins, specifically relating to the field of industrial automation control technology. The controller initializes the acquisition channel and performs anti-jitter screening, marking low-confidence events. Events such as material feeding, material discharge, cutter dead point, and weighing are solidified as odometer anchor point events. Anchor point segments are divided, and the quality of these events is evaluated. During material feeding, a workpiece identifier is generated and written to the work-in-process queue, and a twin workpiece instance is established and bound on the twin side. A displacement odometer is established, and anchor point predictions are compared with actual triggers to identify drift. Multi-source evidence is fused to obtain a comprehensive odometer drift correction quantity, outputting the displacement correction quantity and position uncertainty. The position is cumulatively corrected, and a comprehensive virtual workstation misalignment trigger gating quantity is generated by combining the virtual workstation window and the actuator readiness factor. The triggering strategy is adaptively adjusted to achieve cycle time reconstruction, thereby improving position tracking accuracy and reducing the risk of misalignment triggering.

The application discloses a kind of bridge bottom box girder double steel wire detection robot control system and operating method, for the problems of low efficiency, insufficient precision, weak environmental adaptability of existing bridge detection, the system includes host computer, lower computer and motion control system, closed loop network is constructed by TCP / UDP, serial port and other multi-protocol redundant communication, supports automatic and manual dual operation mode.Motion control system adopts PID closed-loop control and encoder-odometer fusion positioning, realizes accurate movement along double steel wire, multi-pose shooting and obstacle avoidance;Core module has double backup design, with power switching and communication link redundancy mechanism, to ensure stable operation.The application realizes automatic detection of box girder interior without dead angle, improves detection efficiency and data reliability, adapts to complex working environment, reduces manual risk, and has important engineering application value.

The application relates to the technical field of forestry maintenance automation equipment, and discloses a tree maintenance robot and a working method thereof, which comprises the following: a mobile platform, which is equipped with an encoder odometer to realize autonomous movement of the robot; a multi-sensor perception system, which is used for environment perception and self-positioning; a spraying execution mechanism, which is used for performing spraying work on trees; an automatic supply system, which comprises a charging interface and a feeding / discharging valve and is used for realizing automatic supply of energy and materials; a controller, which is used for data fusion and decision control; an environment map is constructed based on multi-sensor fusion SLAM technology, a semantic map is generated by identifying the trunk position and diameter, a full-coverage working path is planned, the spraying execution mechanism is controlled to perform adaptive spraying on target trees, and the automatic supply system is dispatched to complete autonomous recharging and feeding when resources are insufficient; and the tree maintenance robot effectively solves the problems of automation, intelligentization and continuity of tree maintenance work in a complex fruit forest environment.

The application provides a visual positioning scale compensation Kalman filtering method, which realizes online real-time estimation and dynamic compensation of scale error by fusing height information, kinematic constraint and visual observation through explicitly taking scale error factor as a state variable in a state space model of extended Kalman filtering (EKF). The technical scheme of the application is applied to solve the technical problems of unstable scale estimation and large accumulated error of pose in the prior art visual inertial odometer (VIO) system when facing severe motion, visual degradation or low-cost IMU zero bias.

The application discloses an adaptive error state fusion-based optimization method for a laser radar inertial odometer, and belongs to the technical field of robot positioning and navigation. Firstly, the input point cloud is ground segmented, and is divided into ground and non-ground feature points to distinguish geometric characteristics and constraint capabilities. Subsequently, a point-to-plane error model is constructed for the two types of features, and an error state vector is calculated. The key step is to adopt an adaptive weighted fusion strategy, dynamically allocate weights according to the distribution and reliability of the two types of features in the current environment, fuse the error vectors, and balance the contributions in the optimization. Finally, the system pose is updated by using the fusion result. By introducing the segmentation and adaptive fusion mechanism, the application effectively prevents the problem that a single feature excessively dominates the optimization process when the feature distribution is uneven, ensures the balanced optimization of parameters such as translation and rotation, and significantly improves the robustness and positioning accuracy of the system in complex dynamic and feature degradation environments.

Embodiments of the present application provide a method and device for checking repositioning pose, electronic equipment and storage medium. The method comprises: obtaining a current odometer pose corresponding to a current positioning frame of the smart mobile device each time a current repositioning pose corresponding to the current positioning frame is obtained; calculating a relative positioning pose between the current repositioning pose and a previous repositioning pose in a first window; calculating a relative odometer pose between the current odometer pose and a previous odometer pose in a second window; determining a difference between the relative positioning pose and the relative odometer pose as a pose difference corresponding to the current repositioning pose; and checking the current repositioning pose based on the pose differences corresponding to a first preset number of repositioning poses included in the first window to obtain a checking result. By checking the repositioning pose, the influence of false positioning on the stable operation of the smart mobile device is reduced.

This invention relates to the field of vehicle control technology, and provides a method, apparatus, device, and medium for detecting stall faults in autonomous vehicles. It can collect data from multiple sensors in real time and confirm whether stall fault detection conditions are met based on this data, avoiding the problem of false alarms caused by data from a single sensor. The judgment based on stall fault detection conditions also effectively improves reliability. Based on an odometer-command speed integrator, an odometer-wheel speed integrator, a wheel-motor speed integrator, and a wheel slippage detection integrator, it can effectively prevent misjudgments. It can detect whether the target autonomous vehicle is in a transient protection state, effectively distinguishing between speed changes caused by normal operation and actual hardware faults. Based on fault mode recognition using the odometer-wheel speed integrator, wheel-motor speed integrator, and wheel slippage detection integrator, it can accurately locate the fault after detecting vehicle stall.

The application discloses a scene perception and reconstruction system based on geometric semantic information fusion, and belongs to the technical field of computer vision.The system comprises a video panoramic segmentation (VPS) module, a visual odometer (VO) module and a scene rendering (SR) module, the VPS module acquires panoramic segmentation results by fusing the VO module and adaptively fusing geometric information; the VO module estimates camera pose, depth and optical flow, and improves the estimation performance by acquiring semantic information from the VPS module; the two modules promote each other through recursive interaction; and the SR module aims to synthesize a realistic scene through a sparse point cloud map.The core motivation of the technology lies in strengthening the correlation between the VO and VPS tasks, so as to realize more fine 3D scene understanding.The application is a unified perception, positioning and reconstruction method, and compared with a direct convolution mode, the adaptive geometric information fusion of the application can support more real application scenarios and has strong expandability.

The application relates to a monocular vision vehicle positioning system fusing UWB ranging information and a method thereof, the system comprising: a measurement preprocessing module for carrying out distortion removal and photometric correction on a monocular camera original image, and interpolating and acquiring a synchronous distance measurement according to UWB label ranging information; a monocular vision odometer module for incrementally estimating a camera pose and constructing two kinds of visual relative pose measurements; a non-line-of-sight identification and elimination module for eliminating non-line-of-sight measurements from the synchronous distance measurement; an initialization module for estimating an initial global heading angle and a visual odometer initial scale; and a factor graph optimization module combining output data of the monocular vision odometer module, the non-line-of-sight identification and elimination module and the initialization module to estimate vehicle global pose information. Compared with the prior art, the application can simultaneously solve the problems existing in the monocular vision odometer, greatly reduce the required number of UWB base stations, and accurately estimate the vehicle global pose in real time under various complex working conditions.

The application relates to the technical field of laser SLAM, and discloses a laser mapping method and system based on ISAM2, which comprises the following steps: acquiring lidar, inertial measurement unit and wheel odometer data, obtaining a point cloud sequence through tight coupling pre-integration motion distortion correction, adaptive voxel down-sampling and space-time feature fusion dynamic object marking; extracting track lines, tunnel profiles and column features from the point cloud sequence to construct a structured line feature descriptor sequence; calculating continuous frame relative poses and uncertainty covariance by using a multi-constraint fusion registration cost function; performing loop detection by using multi-level feature fusion scene descriptor and sequence consistency analysis multi-hypothesis tracking; eliminating accumulated errors by using ISAM2 incremental optimization and gradual loop factor activation; and outputting a three-dimensional point cloud map and an optimized trajectory by fusing point cloud confidence weighting. The application can realize quasi-real-time and high-precision laser mapping in a highly repetitive subway tunnel environment.

The application discloses a tunnel three-dimensional center line extraction method based on octree occupancy map and distance transformation, which takes tunnel scene point cloud sequences registered to the same global coordinate system as input, incrementally constructs a global octree occupancy map to obtain an occupancy leaf voxel center point set, extracts a candidate occupancy voxel center point set and performs three-dimensional principal component analysis to obtain a stable axial direction, constructs a cross-section plane along the stable axial direction, extracts a point set within the slice thickness range, projects the point set to a two-dimensional coordinate system, rasterizes a cross-section obstacle grid, performs two-dimensional distance transformation, takes a point corresponding to a maximum value of a distance field as a maximum inscribed circle center, and back-projects the point to obtain a slice center point and a clear radius, accumulates continuous effective center points, applies a maximum turning angle constraint, and outputs a tunnel three-dimensional center line. The application does not require strict alignment of the input tunnel coordinate system Z-axis with the gravity direction, and does not depend on an odometer trajectory or a platform attitude estimation axial direction, and is suitable for straight or curved tunnel structures with longitudinal slopes.

The application discloses a pipeline robot operation mechanism and a virtual reality and force feedback control method, and relates to the technical field of robots, which comprises a mobile chassis, an operation mechanism and a control system, the operation mechanism comprises a lifting mechanism, a holder mechanism and a dredging mechanism, the control system comprises a virtual reality unit and a force feedback hand controller, the virtual reality unit comprises a sensor module, a controller module, a network cable and a computer, the sensor module comprises a laser radar, an odometer, an inertial unit and a current sensor, the operation mechanism, the sensor module and the controller module are installed on the mobile chassis, the controller module and the computer are connected through the network cable, and the force feedback hand controller is connected with the computer through another network cable. The three-dimensional environment reconstruction method is used for three-dimensional reconstruction of a pipeline environment to realize virtual reality visual feedback, and the force feedback hand controller is used for controlling the mobile chassis and the operation mechanism. The application provides real-time feedback of vision and force touch, and improves the safety of robot operation.

The application discloses a wearable human-machine operation data acquisition system and method based on relative visual inertial measurement, which is used for robot imitation learning. The system comprises a chest reference node and a hand operation node, each node comprising a camera and an inertial measurement unit, and the poses of the trunk and the hand in a global coordinate system are obtained through a visual inertial odometer respectively. A central processing unit calculates the relative pose of the hand relative to the chest in real time, effectively decouples the overall movement of the human body and the fine action of the hand, and introduces a rotation compensation mechanism based on the angular velocity of the chest reference node, so as to eliminate the false displacement caused by body rotation. The system outputs a strictly time-synchronized, standardized format relative pose sequence, multi-view images and inertial data, which can be directly used for imitation learning algorithm training. The application solves the problems of poor mobility, motion coupling and complex deployment of the traditional teaching system, and realizes high-precision and high-robustness data acquisition in a free-moving scene.

This invention provides a motion state recognition adaptive vehicle integrated navigation method and device. The method includes: acquiring data from the vehicle's inertial measurement unit and odometer; determining the probability of the vehicle being in each preset motion state based on a pre-trained classifier; and determining the current motion state of the vehicle based on each state probability and a preset threshold; constructing an error state model and an observation model based on the vehicle's satellite navigation information, inertial navigation information in the inertial measurement unit, and preset deviation constraints, wherein the observation model includes noise covariance matrices corresponding to nonholonomic constraints and zero-speed update constraints; adjusting the noise covariance matrices corresponding to nonholonomic constraints and zero-speed update constraints according to the vehicle's motion state and the probability of being in each preset motion state; performing measurement updates on the observation model based on the adjusted noise covariance matrix to correct the error states in the error state model, and outputting the vehicle's navigation and positioning results.

The application provides a fruit and vegetable picking robot based on UWB positioning and YOLOv5 vision fusion and a control method thereof, and relates to the technical field of agricultural robots.The robot comprises a tracked or wheel-track combined mobile chassis, a six-degree-of-freedom mechanical arm and a flexible end effector for fruit picking, a UWB high-precision positioning module for global positioning, a binocular depth camera for target three-dimensional positioning, a laser radar obstacle avoidance module for environmental obstacle avoidance, an embedded cooperative control system, a communication module and a power supply module for power supply.The application adopts a dynamic fusion architecture of UWB global positioning and YOLOv5 visual local recognition.Compared with a traditional single sensing scheme, through the cooperative mode of global positioning bottom navigation and local visual accurate calibration, the problem of single visual positioning drift and the cumulative error of the odometer in long-distance driving is solved, and the accuracy of the robot orchard inter-row cruise, path correction and fixed-point parking is greatly improved.

The application relates to a mowing robot laser radar odometer rotation error compensation method and device and a mowing robot. The method comprises the following steps: acquiring current pose information of a mowing robot based on laser radar odometer output, wherein the current pose information at least comprises a current position and a current heading angle; when a preset straight line motion judgment condition is met, determining a current time pose as a reference pose, and calculating an actual motion direction angle of the robot based on a position change between a subsequent time pose and the reference pose; determining an instantaneous rotation error according to an angle difference between the actual motion direction angle and the current heading angle; accumulating instantaneous rotation errors of multiple time points to obtain a cumulative rotation error; compensating the current heading angle according to the cumulative rotation error to obtain a compensated heading angle, and outputting compensated pose information. By adopting the method, path drift of the robot in long-time operation can be effectively reduced, and the precision and stability of robot navigation are significantly improved.

This invention proposes a cross-view geolocation method based on kernel dynamic attention to achieve registration between ground-based lidar point clouds and high-resolution remote sensing images. First, a feature pyramid network is used to extract multi-scale features from two modalities. In the coarse matching stage, a matching module with alternating stacked kernel dynamic self-attention layers and cross-attention layers processes the features of both modalities. The kernel dynamic self-attention layer performs convolutional post-modulation on the global attention map, expanding the spatial scope layer by layer to obtain coarse matching candidates. In the fine matching stage, using the coarse matching results as seeds, local image patches and corresponding point cloud subsets are extracted from the high-resolution feature map. The kernel dynamic attention module refines the correspondence, obtaining two-dimensional to three-dimensional matching point pairs. Finally, a robust pose solving algorithm is used to recover the translation and heading angles between the lidar and remote sensing images, achieving cross-view geolocation. Combining this with an onboard odometer can effectively suppress cumulative drift and improve long-term positioning accuracy.

The application relates to a multi-robot cooperative mapping system in the field of simultaneous localization and mapping, which comprises a plurality of mobile platforms and a central control center; each mobile platform is loaded with a laser radar, a laser odometer and a computing host; the laser radar is used for acquiring point cloud data of a surrounding environment; the laser odometer is used for acquiring pose data of the mobile platform; the computing host processes the point cloud data and the pose data to obtain key frame point cloud and a key pose graph, and sends the key frame point cloud and the key pose graph to the central control center through a wireless network card; the central control center performs loop detection based on the key frame point cloud and the key pose graph to find a loop candidate pair, performs pose graph aggregation and optimization based on a loop detection result, the key frame point cloud and the key frame point cloud to obtain a global point cloud map of a region to be mapped. The application solves the problems of low mapping efficiency and long mapping time consumption in the prior art.