25results about How to "Avoid computational burden" patented technology

The invention relates to the technical field of obstacle avoidance early warning, in particular to an obstacle avoidance early warning method and system for a robot. The method comprises the following steps: acquiring motion data and moving target point data of a robot; robot motion mode classification is conducted on the robot motion data, and a robot motion mode and a robot static mode are generated; performing environment detection on the moving target point data by using a built-in laser radar of the robot according to the static mode to obtain standard static environment detection data; and planning the path according to the standard static environment detection data, and generating initial running path planning data of the robot. According to the invention, the obstacle avoidance early warning precision and real-time performance of the robot in a complex environment are improved by accurately distinguishing static and dynamic objects, reconstructing the motion trail and planning and adjusting the path.

The invention relates to the technical field of video processing, and discloses a livestock three-dimensional behavior detection method, which comprises the following steps: S1, establishing a livestock three-dimensional behavior detection system for shooting livestock behavior videos; s2, constructing a livestock three-dimensional behavior real-time detection model, namely an RT-DETRv3 model; and S3, according to a livestock shot video acquired by the behavior detection system, outputting different behaviors of the livestock by using the real-time detection model for classification and analysis, and generating a visual livestock behavior statistical report. According to the invention, through combination of the multi-view camera system and the deep learning algorithm, especially integration of the 3D convolutional neural network and the RT-DETRv3Transformer, three-dimensional behaviors of livestock can be captured in real time. Compared with the prior art which only depends on two-dimensional monitoring of a single visual angle, the method can achieve the precise recognition of the complex three-dimensional postures of standing, lying, walking and the like of the livestock, and greatly improves the detection accuracy and precision. In addition, attitude key point detection and a multi-target supervision mechanism are used.

The invention relates to the field of multi-robot cooperative scheduling, and discloses a multi-robot cooperative scheduling unmanned underground mechanical garage system and method, and the system comprises a hardware subsystem which comprises a transfer robot, a parking space frame, a vertical elevator, and an entrance and exit junction area; the control system is deployed in the central control server, and the control system comprises a multi-layer dynamic data grid generation module which is used for constructing a discretized data grid corresponding to a garage physical space; and the central data calculation module is used for continuously updating the data of each data layer in the multi-layer dynamic data grid according to the real-time state of the garage. In the regional congestion degree and traffic control layer, the congestion index is calculated according to the density, the speed variance and the dispersion in the movement direction of the transfer robot, and the task gravitation modulation coefficient is generated, so that the tendency of the robot to enter the crowded region is reduced, the adaptive dispersion of the traffic flow is promoted, and the continuity of garage operation is maintained.

The invention belongs to the technical field of laser Raman gas analysis and detection, and particularly relates to a zero drift suppression and ray interference resistance method for laser Raman gas detection. According to the method, the interference of noise and cosmic rays in the laser Raman gas quantitative detection process can be reduced, so that the precision of a laser Raman gas quantitative detection result is relatively high, a data abnormal phenomenon hardly exists, and the probability of zero drift can be reduced. According to the method, the influence caused by noise and cosmic rays can be eliminated in a large probability.

This invention discloses a UAV-GPR noise suppression and imaging method and apparatus based on Hankel trajectory matrix decomposition, belonging to the field of ground-penetrating radar technology. The method first preprocesses the radar echo data and constructs a Hankel-block-Hankel trajectory matrix, achieving noise suppression through singular value decomposition and adaptive truncation and reconstruction based on a quasi-local curvature metric. Then, the derivative threshold reference point method is used to accurately extract the arrival time of surface reflections from the noise-suppressed data, thereby estimating and correcting flight altitude variations. Next, singular value decomposition based on a distance threshold is applied to suppress surface reflections in the altitude-corrected data. Finally, a time-delay summation imaging method based on refraction point approximation in a two-layer medium model is derived, and the filtered data is focused and imaged using the average flight altitude. This invention achieves accurate detection and location of underground targets.

The application provides an environment monitoring method and device based on wireless sensor network joint self-learning, and relates to the field of Internet of Things monitoring.The method comprises the following steps: acquiring an original time sequence and determining the total iteration number of each edge node;at each edge node, the original time sequence is normalized, segmented, aggregated and approximately calculated, and symbolized according to the total iteration number, to generate a discrete finite symbol sequence;the discrete finite symbol sequence is subjected to abnormal probability calculation to generate a single local anomaly detection threshold;at the convergence node, the single local anomaly detection threshold is subjected to weighted fusion to obtain a weighted threshold;the single local anomaly detection threshold is updated through the weighted threshold until the iteration ends, the local anomaly detection threshold is output, and the anomaly detection of the environment is completed.The scheme effectively reduces the storage and calculation costs, realizes the communication between nodes, balances the energy consumption of the edge nodes, and improves the detection accuracy.

The application discloses a self-adaptive gravity compensation trajectory correction method based on a lead angle and belongs to the technical field of aircraft guidance control. The application constructs a dynamic mapping relation between a trajectory deviation and a gravity compensation amount by using the real-time measured lead angle information of the aircraft, adaptively adjusts a gravity compensation coefficient to correct the trajectory, reduces the compensation amount to inhibit the trajectory lifting when attacking a short-distance target, enhances the compensation to offset the falling effect when attacking a long-distance target, improves the problem that the gravity compensation is fixed and does not affect the guidance effect in the traditional method when attacking targets with different distances, reduces the dependence on the difficult-to-measure parameters such as the remaining time in other methods, and improves the striking precision and the shooting range of the aircraft on different targets.

According to the vehicle-mounted augmented reality multi-layer display device provided by the invention, a vehicle-mounted HUD image is divided into a plurality of image layers with different image characteristics through different imaging technologies for respective imaging, and the multi-layer images are combined for unified display based on a specific sight line direction, so that a user with the sight line in the specific direction is enabled to have a user experience. When the AR-HUD display effect is accurate, the HUD images are displayed in the corresponding position or direction of the real environment in a superposed mode, when the sight line of the user moves, the superposed images are separated, and the user is prompted to adjust the sight line direction through the change of the relative position relation between the multiple layers of images so as to see the accurate AR-HUD display effect. The user eyes do not need to be tracked and monitored constantly to adjust the image display position, the monitoring and operation burden caused by human eye position change is avoided, and the structure is simple.

The invention relates to the technical field of geographic information system and seismological observation application, in particular to a data resource pool building system which comprises a data acquisition module, a non-seismic event exclusion module, a physical performance detection module and an abnormal reason tracing module. According to the method, seismic waveform original data are periodically collected, storage duration comparison is carried out, seismic observation data categories are judged based on multi-dimensional parameters, physical performance fault categories are identified through hardware, software and architecture state parameters, and the method specifically comprises the steps of extracting and analyzing data geographic positions to judge scheduling paths, identifying and clearing repeated observation contents, and determining the physical performance fault categories. Calculating waveform stacking pressure and signal writing skew, judging frequency domain sorting delay, calculating a time adjustment factor based on frequency drift and correcting a storage time threshold value; through multi-dimensional data verification and load optimization, dynamic adjustment of a data storage time threshold is realized, and data storage efficiency and system stability are improved.

The invention discloses a Mama enhanced industrial product surface defect detection method based on YOLOv11, and belongs to the technical field of defect detection.The method comprises the steps that preprocessed industrial product surface image data are input into an industrial product surface defect detection model; according to industrial product surface image data, feature vectors are extracted through a plurality of C3k2 modules in sequence, and the feature vectors are input into an SPPF module to obtain the feature vectors; according to the SPPF module, obtaining a feature vector, splicing the feature vector with a feature vector in any C3k2 module so as to obtain a feature vector through the Mama module, splicing the feature vector obtained by the Mama module with any other feature vector so as to obtain a feature vector through the Mama module, and repeating the steps; and randomly selecting the feature vectors obtained by a plurality of Mamba modules, and inputting the feature vectors into a plurality of detection heads in a one-to-one correspondence manner for detecting surface defects. According to the method, both the detection precision and the reasoning efficiency are considered, and particularly, the detection rate and the positioning accuracy of surface defects of complex, tiny and long-strip-shaped industrial products are remarkably improved.

The present application belongs to the technical field of radio frequency fingerprint identification, and aims at the problem of decline of recognition rate of existing methods in low signal-to-noise ratio environment, and proposes a radio frequency fingerprint identification method and system based on feature latent space diffusion model. The method comprises the following steps: after receiving the radio frequency signal, firstly, the system is classified and pre-trained, and the parameters of the feature extraction module and the classification module are updated; then, latent space diffusion pre-training is carried out, the noise time step prediction module and the latent space diffusion noise prediction module are trained based on the feature extraction and classification module, and the parameters are optimized in combination with feature reconstruction, time step prediction and classification loss; finally, systematic fine-tuning is carried out, and all modules are optimized as a whole. Repeat the training until the neural network converges, and output the optimal network parameters for testing. Through the combination of feature extraction and latent space noise suppression, the robustness and accuracy of radio frequency equipment identification under low signal-to-noise ratio conditions are effectively improved.

The invention relates to the technical field of space telescopes, in particular to a method for keeping the surface type of a primary mirror of a spliced telescope based on multi-agent cooperative control, and the method comprises the steps: setting an agent on each sub-mirror in the primary mirror, and obtaining the current attitude information of each sub-mirror; each sub-mirror is connected with other sub-mirrors by utilizing an intelligent agent through wireless communication, and the current attitude information of each sub-mirror is shared to the connected sub-mirror connected with the sub-mirror in real time, so that the current shared attitude information of each sub-mirror is obtained; according to the current attitude information and the current shared attitude information of each sub-mirror, determining a current kinetic model of each sub-mirror, and further obtaining the attitude information of each sub-mirror at the next moment; and the above contents are repeated until the current attitude information of each sub-mirror converges to the ideal attitude information of each sub-mirror, so that information sharing between the sub-mirrors and collaborative decision between the connected sub-mirrors are realized, the dynamic adaptive capacity is improved, and the requirements of adaptive real-time adjustment of the sub-mirror attitude and the main mirror surface shape are met.

This invention discloses a rapid probabilistic conflict detection method based on track uncertainty. The method first acquires the aircraft's current state, reference track information, thunderstorm area prediction information, and rolling prediction parameters to construct a rolling prediction scenario. Then, based on the overall system error framework, a track uncertainty model is established, decomposing navigation system errors and flight technical errors into components along the track and perpendicular to the track. Covariance matrices in the global coordinate system are obtained through coordinate rotation. Furthermore, conflict models between aircraft and between aircraft and thunderstorms are constructed, unifying the two types of conflicts as a probability problem of a two-dimensional Gaussian random variable falling into the conflict area. Finally, Cholesky decomposition and Gauss-Hermite numerical quadrature are used to calculate approximate conflict probabilities at each moment within the rolling window, and the maximum probability is taken as the detection result for the current rolling step. This invention improves the conflict detection efficiency under track uncertainty and thunderstorm constraints, enhancing the level of collaboration, automation, and intelligence during aircraft tactical operations.

The present invention discloses a real-time basketball goal detection and recognition method based on video input, and specifically relates to the technical field of basketball goal detection. A fixed frame length impact assessment model is constructed according to basketball flight trajectory information, player pitching behavior information and image processing delay information. The system can adjust the initial fixed frame length in real time to ensure that the key picture of the goal moment is captured, and will not cause missed judgment or misjudgment of the goal moment due to the inability of the fixed frame length to adapt to the different durations of fast shots or slow shots, avoid misjudgment caused by the inability of the fixed frame window to adjust according to the actual duration of the shooting action, and improve the detection accuracy of the basketball goal status; through adaptive regulation, the system can be dynamically adjusted according to the real-time video input, reduce unnecessary waste of computing resources, avoid redundant calculations, ensure that it can quickly respond and accurately identify the basketball goal status in different scenarios, and adapt to the diverse actions and complex backgrounds in basketball games.

This invention discloses an online scalable configuration method and system for dynamically changing time-sensitive networks (TSNs), relating to the Internet of Things (IoT) field. The invention proposes an efficient inter-flow conflict detection mechanism based on multi-level flow grouping. By using a correlation analysis of flow period and offset, it avoids traditional conflict detection methods based on link maximal cliques and link hyperperiods, enabling rapid response to large-scale data flow conflict detection needs. It designs an online incremental method based on decoupling routing and scheduling, as well as offline pre-routing, accelerating the computation speed of routing and scheduling schemes while ensuring scheduling space. This invention reduces the computational complexity of data flow set hyperperiod sensitivity and the maximum link slot occupancy in TSNs, accelerating online scheduling and providing efficient conflict detection support for online scalable configuration.

The invention relates to the technical field of torque ripple suppression, in particular to a permanent magnet synchronous motor torque ripple suppression method based on feedforward compensation. The method comprises the following steps: S1, establishing a mathematical model of the permanent magnet synchronous motor; s2, analyzing factors causing torque pulsation through electromagnetic parameters and mechanical parameters based on the mathematical model of the permanent magnet synchronous motor; s3, according to an analysis result, designing a fluctuation model based on discrete Fourier transform; s4, generating a compensation signal based on Fourier series expansion by using the designed fluctuation model; and S5, combining the generated compensation signal with an existing control system in a superposition mode, and adding the superposed compensation signal into a control instruction of the permanent magnet synchronous motor to smooth torque output. Through the design and application of the feedforward compensation signal, the method can effectively counteract the periodic torque ripple caused by the cogging effect, the magnetic field nonuniformity, the inverter switching frequency and other factors.

The invention discloses a multi-scale solving method for cavitation flow field numerical simulation, and relates to the technical field of cavitation flow field numerical simulation. In order to solve the technical problems that in the prior art, an existing double-scale method still has limitation when being used for treating conversion of different-scale steam structures, it is difficult to ensure reasonable conversion of steam bubbles among different frames, and consequently simulation continuity and accuracy are affected, the invention provides a steam bubble simulation method. According to the technical scheme provided by the invention, the multi-scale solving method for numerical simulation of the cavitation flow field comprises the following steps of: judging the scale of a cavitation bubble; processing the large-scale cavitation bubbles in the Euler framework, namely identifying a steam structure, capturing a steam-liquid interface and solving a macroscopic flow field; and processing the small-scale cavitation bubbles in the Lagrange framework, wherein the step comprises solving the motion trail, the speed and the pulsation. The method is suitable for being applied to multi-scale solving work of cavitation flow field numerical simulation.

The invention relates to the technical field of nuclear research, in particular to a stable and efficient global coupling iteration method for physical and thermal coupling simulation of a reactor. According to the invention, stable, rapid, accurate and high-resolution simulation of high-fidelity physical-thermal coupling of the whole reactor core can be realized by adopting the MFNK method, and the MFNK method provided by the invention can be further expanded and applied to the field of reactor multi-physical field coupling simulation, such as coupling for increasing fuel performance and the like.

The invention provides a spectral absorption three-dimensional imaging system and method based on parallel light beams and area array detection, relates to the technical field of optical precision measurement and tomography, and aims to solve the problems that high time and high spatial resolution, low system complexity and low cost cannot be compatible at the same time and three-dimensional field distribution cannot be quickly reconstructed with high precision in the prior art. The system comprises a laser emission and modulation module, a beam shaping module, a to-be-measured area, an area array sensor and a data processing module. The light beam shaping module, the to-be-detected area and the area array sensor are arranged on the same plane axis, the light beam shaping module comprises a first collimating mirror and an imaging module which are sequentially connected along a light path, the first collimating mirror forms parallel light beams, and the imaging module performs multi-angle rear two-dimensional imaging; and the area array sensor is electrically connected with the data processing module, is processed and controlled by the upper computer, and reconstructs a two-dimensional image into a three-dimensional field. According to the invention, problems in the prior art are solved, and rapid and high-precision three-dimensional imaging is realized.

The application discloses a wind farm droop and inertia support coefficient feasible region state space mapping evaluation method, wherein historical database operation data participating in primary frequency modulation is utilized on the wind farm side to construct a state space mapping linear function relationship formula among the wind farm droop coefficient, the wind farm inertia support coefficient, the wind speed, the power system droop coefficient and the power system inertia coefficient in the process of primary frequency modulation and inertia support, and the final speed of the wind turbine and the maximum power of the wind turbine; real-time wind speed of the wind farm and received power system droop and inertia parameters are measured, the linear function relationship formula among the above variables is utilized, the wind turbine final speed limit and the wind turbine power limit are considered, the wind farm droop and inertia support coefficient feasible region boundary is evaluated online, and the power system operation side is reported; the power system operation side updates the system coefficient and issues each wind farm, and the optimal droop-inertia frequency modulation coefficient feasible region of each wind farm and the power system where the wind farm is located is solved.

The invention provides a digital twinning rapid modeling and simulation system oriented to a small medical scene, and relates to the technical field of image processing, a local digital twinning model is constructed based on multi-source sensing data and a local space reconstruction mechanism, and through real-time mapping of key parameters such as a patient body position, an instrument posture and propulsion depth, a digital twinning model is constructed. And rapid modeling and dynamic updating of the medical operation scene are realized. The system introduces an operation path stability evaluation mechanism, an instant operation risk response evaluation mechanism and a twinborn model effectiveness judgment mechanism, performs joint calculation on multi-dimensional parameters by adopting a weighted fusion algorithm, and realizes real-time identification and intervention on a path offset risk, an instrument tissue action risk and a model failure risk in combination with threshold judgment and strategy triggering. According to the invention, rapid construction, adaptive correction and simulation support of the digital twinborn model can be realized in a small medical scene, and the safety, stability and system applicability of interventional operation are improved.

This invention discloses a method and system for fatigue life analysis in multi-objective optimization design of pressure vessels. By constructing a parameterized model including structural parameters, material parameters, and cyclic load boundary parameters, stress response data at key locations are calculated, and a stress power spectral density function is generated. Based on the power spectral density function, spectral moment parameters and frequency domain characteristic parameters are extracted, and a probabilistic model is used to calculate the cyclic stress distribution, fatigue damage rate, and fatigue life. A multi-objective evaluation vector is constructed by combining structural design variables, stress response, and life results to uniformly evaluate structural quality, strength utilization, and fatigue life. An evolutionary optimization algorithm is used to introduce life contribution weights for iterative screening and updating of design variables. Through iterative control, model updates and optimization convergence are achieved, ultimately outputting a structural design scheme that meets multi-objective constraints and corresponding fatigue life analysis results, realizing the collaborative optimization of pressure vessel structural design and life assessment.

The present invention relates to the technical field of monitoring images, a method and system for improving the resolution of monitoring images based on a high-definition camera, comprising: accessing a storage system for storing monitoring images of a high-definition camera, wherein the storage system comprises a cloud storage system and a local storage system, searching and obtaining cloud monitoring images and local monitoring images from the cloud storage system and the local storage system, calculating the image similarity between the real-time monitoring image and the cloud monitoring image, and between the real-time monitoring image and the local monitoring image, obtaining a cloud image similarity value and a local image similarity value, and if the cloud image similarity value is less than or equal to the local image similarity value, performing a resolution improvement operation on the real-time monitoring image to obtain a high-quality monitoring image, storing the high-quality monitoring image in the cloud storage system, and generating a feedback instruction that the resolution has been improved. The present invention can solve the problem of improving the resolution of monitoring images while smoothly realizing monitoring by a high-definition camera.

The invention relates to the field of logistics distribution environment perception, in particular to an intelligent logistics distribution method based on a multi-modal logistics knowledge graph. The method comprises the following steps: firstly, acquiring a road image through a visual sensor, and constructing a visual variation observation range; furthermore, a visual characteristic spectrum is selected to extract corresponding visual characteristics in a visual variation observation range, the extracted visual characteristics construct a time domain change curve of the visual characteristics, and a sensitive time domain and a non-sensitive time domain in the driving environment are dynamically captured according to the real-time change condition of the curve. And in a sensitive time domain, image abnormity is determined by comparing the change trend consistency of new energy logistics vehicle vision and radar geometric features. And in a non-sensitive time domain, whether a visual detection target has a corresponding point cloud cluster or not is determined in an interval mode. Therefore, the perception reliability of the new energy logistics vehicle in a complex high-risk period is enhanced, and the system perception reliability, the overall energy efficiency and the resource utilization rate are improved.

The invention relates to a photochromic method and a photochromic device for an object, and the photochromic method is used for carrying out light processing on all-black photochromic dye on the surface of the object, and comprises the following steps: separating a design pattern into monochromatic images of three channels of red, green and blue; tCyan, TMagenta and TYellow are obtained, and when light with the RGB values of (255, 0, 0), (0, 255, 0) and (0, 0, 255) irradiates the all-black photochromic dye, the time when the blue, the magenta and the yellow on the all-black photochromic dye are completely eliminated is TCyan, TMagenta and TYellow respectively, and when light with the RGB values of (255, 0, 0), (0, 255, 0) and (0, 0, 255) are irradiated on the all-black photochromic dye, the time when the blue, the magenta and the yellow on the all-black photochromic dye are completely eliminated is TCyan, TMagenta and TYellow; and projecting a single-color image of a red channel obtained by the S410 for Tcyan duration to the all-black photochromic dye, projecting a single-color image of a green channel obtained by the S410 for Tmagenta duration to the all-black photochromic dye, and projecting a single-color image of a blue channel obtained by the S410 for Tyellow duration to the all-black photochromic dye. Through the setting, the complexity of calculation is reduced, the time is saved, and the method can be conveniently operated on equipment with limited processing capacity.