21results about How to "Solve inaccurate" patented technology

This application provides an action determination method and apparatus, a robot system, a storage medium, and an electronic device, relating to the field of robotics. The method includes: acquiring task instructions and multi-view image data, wherein the multi-view image data includes image data collected by the robot from M views, where M is an integer greater than or equal to 2; determining a feature set based on the multi-view image data, wherein the feature set includes a semantic feature subset, a geometric feature subset, and a three-dimensional spatial feature subset corresponding to the M views; and determining the robot's execution action based on the feature set and the task instructions. This solves the problem of inaccurate robot execution actions in related technologies, enabling action generation based on accurate three-dimensional spatial understanding and improving the accuracy of execution actions.

The application provides a filter rod manual arranging method based on WMS, which comprises the following steps: reading the RFID tag of the filter rod by the RFID data reading and writing device to obtain the label information of the filter rod; judging whether the label information corresponding to the scanned filter rod exists in the filter rod high-bay warehouse WHS; if yes, returning the tray number from the WHS system; if no, storing the physical information in the feedback label according to the set coding, and performing the flat warehouse warehousing in the WHS system according to the physical information, wherein the physical information comprises the brand, the production machine, the production time, the shift and the shift. The application can improve the traceability efficiency of the cigarette filter rod quality and reduce the traceability cost.

The application discloses a kind of line laser 3D camera and mechanical arm's eye calibration system and method, belong to industrial manufacturing and automation control technical field, including calibration board, line laser 3D camera and mechanical arm, the line laser 3D camera is connected with the mechanical arm, the calibration board is composed of calibration ball, connecting rod and mounting plate, single frame data of the line laser 3D camera shooting is solved by single frame data solution calibration board pose algorithm.The application uses above-mentioned line laser 3D camera and mechanical arm's eye calibration system and method, calibration board and single frame data solution calibration board pose algorithm can solve the attitude of calibration board in camera coordinate system according to line laser 3D camera single frame data, and then by shooting multiple frame data and combining the posture information of mechanical arm under each frame image, the conversion relationship between camera coordinate system and mechanical arm coordinate system can be solved without camera size and installation information, to solve the inaccurate problem caused by machining and installation error.

The application relates to a finger automatic polishing method and device, and relates to the technical field of locomotive maintenance. The finger is placed in a 3D visual identification area, 3D visual identification is carried out on the finger, the shape characteristics of the finger are fitted according to the 3D visual identification result, the state of the finger is judged according to the shape characteristics, the polishing amount is calculated according to the shape characteristics and the state of the finger, the polishing amount information is transmitted to a polishing system for polishing, the problems that exist in the current finger polishing process, such as inaccurate judgment, high work intensity, time and labor consuming and difficult polishing effect guarantee, are solved, the automatic operation of the finger polishing is realized, the shape characteristics of the finger can be accurately identified through the 3D visual detection system, the problem of inaccuracy caused by the human eye observation is solved, the detection accuracy is effectively improved, and the polishing effect is guaranteed.

The application provides a kind of unmanned ship consistency control method and equipment, it is related to unmanned system cooperative control technical field.The method comprises: according to the real-time state of the unmanned ship and the state of the unmanned ship at last event trigger time, construct real-time local measurement error;According to the state of the unmanned ship at last event trigger time and the state of all adjacent unmanned ships of the unmanned ship, construct joint error;Based on the real-time local measurement error and joint error of each unmanned ship, judge whether event trigger condition is satisfied, when any unmanned ship satisfies event trigger condition, obtain and according to the state of the unmanned ship at event trigger condition satisfying time, the state of the adjacent unmanned ship of the unmanned ship and unmanned ship controller gain matrix, update the controller output of the unmanned ship and the adjacent unmanned ship of the unmanned ship, carry out consistency control to each unmanned ship according to controller output;The application can accurately and safely carry out multi-unmanned ship consistency control under the condition that system model is unknown.

The present application relates to the field of ship unhooking, in particular to a quick unhooking wharf ship unhooking hook, comprising a base and a winding machine, the winding machine is fixedly installed above the base, a guide rope is connected to the winding wheel of the winding machine, a baffle is installed on one side of the base, and three groups of fixing seats are installed in sequence on the side of the baffle away from the winding machine. The present application uses a pure mechanical trigger type unhooking mechanism, the structure design of the limiting cylinder diameter being larger than the guide cavity is used, the spring reset and the detour groove inclined surface extrusion are matched, the cable and the guide rope are automatically separated, the connection does not need to be disassembled at close range, the problems of difficult disassembly of heavy cable and hidden danger of operation safety are solved. At the same time, the motor drives the gear and rack transmission, drives the guide cylinder to move horizontally accurately, automatically compensates the deviation of the ship berthing angle, makes the cable and the anchor hook accurately aligned, does not need manual adjustment, significantly reduces the labor intensity, and comprehensively improves the automation level, efficiency and safety of the wharf mooring operation.

This invention relates to the field of fluid-structure interaction (FSI) technology, and discloses a method, apparatus, device, and medium for simulating fluid-structure coupling. The invention establishes a first mapping relationship between SPH boundary particles and FEM surface nodes, and in the iterative calculation, utilizes the allocation weights of the FEM surface nodes to distribute fluid forces to the corresponding surface nodes, improving the accuracy of load transfer at the FSI interface. A second mapping relationship is established between SPH boundary particles and FEM element nodes, and in the iterative calculation, the interpolation weights of the FEM element nodes are used to update the physical quantities of the SPH boundary particles based on the updated node physical quantities, ensuring the accuracy of kinematic information transfer at the coupling interface. By sequentially performing fluid force calculation, load allocation, FEM solution, and boundary particle interpolation update in each iteration step, this invention can completely and accurately simulate the dynamic process of fluid-structure interaction.

The application discloses a kind of bone healing degree analysis method, system, equipment and medium, system includes CT image acquisition module, stress analysis judging module, wall thickness analysis judging module, callus analysis judging module, result comprehensive judging module and auxiliary judging module, the application obtains healthy side T0~Tx time image, affected T1~Tx time image and T2~Tx time simulation image in the fixed state under simulation of affected TM2~TMx in time axis of bone healing in sequence, then according to CT image, stress analysis, wall thickness analysis and callus analysis are carried out, three kinds of judging results are comprehensively, finally output accurate bone healing degree result and fixed efficacy analysis result, overcome the inaccurate problem of single factor determination.

Embodiments of the present application provide a content generation method and device based on an index document, equipment, medium and product. The method comprises: obtaining a structured index document; wherein the structured index document comprises a plurality of index entries, each index entry comprising a label code and semantic element information; the label code comprises the coding values of a plurality of label dimensions, and each coding value of the label dimension is used to indicate the attributes of the source content or the target content in the label dimension; the semantic element information comprises a plurality of semantic elements, and each semantic element is used to indicate the semantic characteristics of the source content or the target content; in response to a user instruction input by a user, generating target content based on a large model according to the structured index document; wherein the user instruction represents the related requirements for the target content to be generated. The method is used to achieve the global cognition of the large model to the system architecture and accurately generate the target content.

Provided in one scenario are an information processing method, device, equipment, and storage medium. The method comprises: in response to a preset event trigger, obtaining interface information of a graphical interaction interface, wherein the interface information comprises a first interaction object and an interaction focus, and the interaction focus represents an interaction position in the graphical interaction interface; based on spatial distribution information of the first interaction object relative to the interaction focus, obtaining context information of request information; and based on the context information and the request information, generating response information corresponding to the request information. The technical solution in one scenario accurately understands the request intention corresponding to a fuzzy instruction, solves the problem in the related art that the response information is inaccurate due to an inability to accurately understand the request intention, improves the accuracy of the response information, and improves the user experience.

The invention relates to the technical field of animal physiology research, and discloses a yak living gastric juice and intestinal juice sampling system and method capable of achieving segmented positioning, and the system comprises a negative pressure sampling unit, a plurality of positioning units, a physiological parameter sensing unit, a control unit and a multi-connected independent sample collection device. According to the method, real-time navigation is carried out through the sensor, the positioning unit is sequentially controlled to anchor different target parts of the digestive tract of the yak, and pure digestive juice samples are extracted respectively. The device solves the problems of inaccurate positioning, large trauma and easy sample pollution in the prior art, realizes noninvasive, accurate and repeatable collection of the gastrointestinal tract microorganism sample of the living yak, and is especially suitable for gastrointestinal tract microbiomics research.

The present application relates to the technical field of dredging equipment maintenance, and particularly relates to a mud pump life modeling method based on multi-source residual collaborative learning, which comprises the following steps: firstly, acquiring simulation model output results simulating the work of a mud pump; then, collecting multi-source data of a mud pump work site; then, performing residual calculation according to the multi-source data and the simulation model output results to obtain multi-source residuals; and finally, correcting a preset mud pump life model based on the multi-source residuals to obtain an actual mud pump life model. The present application comprehensively covers the physical nature of the multi-factor coupling damage of the mud pump through multi-source data fusion, accurately identifies the deviation between the model assumption and the actual environment by using the residual-driven model correction mechanism, continuously corrects the mud pump life model based on real-time residuals, has dynamic iterative evolution capability, provides core support with mechanism depth and engineering practicability for the intelligent operation and maintenance of the dredging mud pump, and solves the problem of inaccurate mud pump life modeling in the prior art.

The invention discloses a commodity oral playing video generation method and device and a storage medium, and relates to the technical field of computer vision, and the method comprises the steps: constructing a corresponding generation condition data packet for each to-be-generated target video clip; determining a generation condition feature vector and a plurality of noise-added feature vectors, and inputting the generation condition feature vector and each noise-added feature vector into a lightweight video generation model to generate a target video clip and an end frame through iterative diffusion according to a preset number of diffusion times; taking the tail frame of the target video clip as a context frame in the generation condition data corresponding to the next to-be-generated target video clip, and returning to execute the step of constructing the corresponding generation condition data packet for each to-be-generated target video clip until all the target video clips are generated; and splicing all the generated target video clips to obtain a commodity oral playing target video, thereby solving the problem of inaccurate commodity oral playing video in the prior art, and generating a high-quality long video.

The invention relates to the technical field of charging correction, in particular to a vehicle charging correction method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining historical charging data that the vehicle quits charging under the condition that the vehicle is not fully charged; determining a calibration point of the power battery according to the historical charging data, identifying a calibration voltage value of the calibration point, and determining a calibration charge state of the power battery according to the calibration voltage value of the calibration point; in the charging process of the vehicle, if it is recognized that the current voltage value of the power battery reaches the calibration voltage value, the current charge state displayed by the vehicle is corrected into the calibration charge state. Therefore, the problem that the SOC displayed by the vehicle is inaccurate due to the fact that the SOC displayed by the vehicle cannot be fully charged and corrected when the vehicle is not fully charged and quits the charging scene in the related technology is solved.

This utility model belongs to the field of marking fixture technology, specifically disclosing a blind hole gear shaft marking fixture, including a marking fixture body and a positioning pin. The marking fixture body has a central insertion hole and a pin hole on its end face, the pin hole being located outside the insertion hole. One end of the workpiece can be inserted into the insertion hole on the marking fixture body, and the pin hole and a weight-reducing hole on the workpiece are directly opposite each other. The positioning pin is sequentially inserted into the weight-reducing hole on the workpiece and the pin hole on the marking fixture body. This utility model can solve the problem of inaccurate marking positions caused by manual positioning of the workpiece.

This utility model discloses an adjustable placement device for a plate freezer, including a freezing box with a placement plate installed inside. The adjustable placement device also includes an adjustment device installed on the outer wall of the placement plate; a friction structure is installed on the top of the freezing box. This utility model relates to the field of plate freezer technology. Through the cooperation of an adjustment block, a lead screw, and a fixed shell, when adjusting the distance between the placement plates, a handle drives the lead screw to rotate. The threaded connection between the lead screw and the adjustment block moves the placement plate inside the freezing box, allowing for rapid adjustment of the placement plate's position. This solves the problems of existing fixed placement structures, which are difficult to adjust flexibly according to actual needs, resulting in significant limitations in use. Furthermore, some placement devices with adjustment functions have complex adjustment methods, are inconvenient to operate, and have low adjustment accuracy, failing to meet diverse usage requirements.

The application discloses a white oil block stripe improvement method and device based on a jet printer and a medium. The method comprises the following steps: acquiring a production image data template to be jet printed in real time, and acquiring current jet printing times of the jet printer and image processing description information corresponding to the production image data template; performing white oil block and character analysis processing on the production image data template by a preset raster image processor and according to image resolution parameters and pixel sampling point proportion parameters, to obtain a white oil block sampling point graph and a character sampling point graph; acquiring jet printing times corresponding to the current jet printing times of the jet printer; and performing jet printing processing according to the white oil block sampling point graph and the character sampling point graph, to obtain a production image data jet printing image. The method solves the problem that the jet printing of the jet printer cannot improve white oil block stripes, thereby improving the accuracy of the jet printing image, reducing white oil block stripes in the production image data jet printing image, and improving the jet printing quality and accuracy of the production image data jet printing image.

Embodiments of the present application disclose a recommendation model training method, a recommendation method and related devices. A user behavior sequence sample marked with a real result is input into a recommendation model to obtain an overall value vector. Then, the overall value vector is calibrated according to a pricing value vector to obtain a calibrated value vector. The calibrated value vector and previous user behavior features are input into the recommendation model again to obtain a prediction result. The calibrated value vector can help the recommendation model obtain a more accurate prediction result. A loss function is determined based on the behavior features, the prediction result and the real result. The recommendation model is updated based on the loss function until the loss function converges.

The present application belongs to the technical field of patent recommendation based on specific calculation model, and more particularly relates to a patent recommendation method based on a Transformer encoder and a regularization strategy. The method comprises: establishing a certain number of patent data sets according to a plurality of search texts, and classifying according to search statements; performing data preprocessing on the search texts and the patent texts in the data sets; establishing a word and sentence double-processing model based on the Transformer encoder and the regularization strategy, and optimizing the model structure and output results using a regularization strategy based on Dropout and symmetric JS divergence; processing the word granularity level at the sentence granularity level; linearly weighting the sentence granularity level results and the word granularity level results, and outputting classification results. The present application solves the problems of inaccurate information matching caused by inconsistent lengths of patent texts and search texts in the prior art and low accuracy of traditional patent recommendation.

The application discloses a kind of sensor positioning exception determination method and device, storage medium and electronic device, wherein the above-mentioned method includes: by radar sensor determines the first position change of mobile robot in preset time interval;And by inertia measuring unit and wheeled odometry determines the second position change of mobile robot in preset time interval;And by wheel speed sensor determines the third position change of mobile robot in preset time interval;According to first position change, second position change and third position change, determine whether radar sensor exists positioning exception.Using the above technical solution, it is solved in the process that mobile robot uses radar sensor to construct map, since it is unclear whether radar sensor exists positioning exception, the problem that the map constructed is inaccurate.

The application relates to the technical field of satellite communication, and discloses a satellite searching system and method of a ground terminal, the system comprising: a master device with a first antenna and a second antenna, the first antenna being used for communication with a slave device, and the second antenna being used for receiving a low-orbit satellite beacon signal; the slave device with a first antenna and a second antenna, the first antenna being used for communication with the master device, and the second antenna being used for receiving a satellite beacon signal; a synchronization module used for synchronizing the clocks of the master device and the slave device; and a master device calculation module used for calculating the position of a satellite in a master device coordinate system in combination with the position of the slave device. According to the application, the first antenna is used to realize the intercommunication between the master device and the slave device, the time stamp data of the slave device is used, the master device calculates the accurate position of the satellite according to the time difference and the position, and the positioning problem when ephemeris information is lost or cannot be updated in real time is solved.