36results about How to "Stable tracking" patented technology

The invention discloses a low-altitude unmanned aerial vehicle countering monitoring system based on state feature decoding, relates to the technical field of unmanned aerial vehicle prevention and control, and is used for solving the problem of poor low-altitude target detection and countering. According to the invention, through closed-loop management and control of heterogeneous radar networking detection, signal level fusion processing, threat assessment and resource scheduling and intelligent countering decision, accurate identification and directional countering of the low-altitude unmanned aerial vehicle are realized; according to the method, the detection precision and the micro-Doppler feature extraction capability of a low-speed small target are improved by combining a virtual aperture synthesis technology, then unbalanced resource allocation is implemented based on a threat level, the system efficiency is optimized, and finally seamless migration of an interfered node task is realized by constructing a signal interference situation map. And continuous tracking and effect evaluation on the target are maintained in the active radio frequency countering process, so that the overall efficiency and reliability of low-altitude security and protection in a complex environment are effectively improved.

The invention discloses a pulse signal source tracking method based on a robust Kalman filter. The method comprises the following steps: reading a target state vector and error covariance matrix at a previous moment, and observing a noise covariance matrix; a variational Bayesian parameter before iteration, a target state vector and an error covariance matrix are initialized; starting iteration, and calculating the quadratic form sufficient statistics of the observation error and the forecast error; updating variational Bayesian parameters; updating the forecast error covariance matrix and the observation noise covariance matrix; calculating a target state vector and an error covariance matrix of the iteration; when the variation of the target state vector of the current iteration and the target state vector of the last iteration is smaller than the tolerance, ending the iteration, and outputting the target state vector at the current moment, the error covariance matrix and the observation noise covariance matrix; compared with a deterministic integral sampling method, the method provided by the invention adopts adaptive volume sampling to realize higher tracking precision under lower calculation cost.

The invention relates to the technical field of machine vision and intelligent monitoring, and discloses an automatic camera tracking method and device based on machine vision, and the method comprises the steps: obtaining a real-time video frame sequence, and carrying out the preprocessing of the real-time video frame sequence, and obtaining an initial image set; identifying features of the initial image set, judging an abnormal state, and fusing the data to generate behavior change trend data; a smooth trajectory is predicted based on the data, and a parameter adjustment data set is obtained by combining camera position calculation deviation; adjusting camera parameters according to the generation instruction, detecting shielding and fusing trajectory data to obtain real-time tracking parameters; and applying parameters, complementing missing frame verification, updating a process and reextracting features, and performing time sequence analysis in case of abnormality to obtain a final tracking logic sequence. According to the method, accurate target locking can be realized, and the stability and accuracy of abnormal emotion tracking in a crowded environment are improved.

The application discloses a stable domain constraint trajectory planning and control method for a modular distributed electric drive heavy load vehicle, comprising the following steps: constructing a nonlinear dynamics model of the modular distributed electric drive heavy load vehicle, fitting a tire lateral force model through a rational polynomial, and establishing a vehicle system state space expression based on the rational polynomial form; adopting a sum of squares planning method to estimate the stable domain of the vehicle under different vehicle speeds, road adhesion coefficients and front, middle and rear wheel turning angles, and analyzing the influence of each state quantity on the shape and boundary of the stable domain; embedding the stable domain constraint into a trajectory planning layer based on a quintic polynomial to generate an obstacle avoidance trajectory meeting the dynamics stability; designing a multi-axle proportional steering strategy based on a steady state assumption, and combining a nonlinear model predictive controller to realize trajectory tracking control, so that the stability and control feasibility of the vehicle in a high-speed dynamic obstacle avoidance process are ensured. Through construction of a high-speed dynamic obstacle avoidance simulation scene, the effectiveness, stability and control feasibility of the stable domain constraint trajectory planning and control framework are verified.

The present application provides a kind of offshore area navigation safety control and optimization method, comprising: step 1, the multi-source observation data of offshore navigation target is collected, and pre-processing is carried out, to obtain the space-time observation correlation graph;Step 2, the dynamic credibility of each observation node is evaluated using multi-modal consistency graph neural network;According to the obtained dynamic credibility, an adaptive weighted factor graph model is constructed, and the system state is solved to obtain the optimal fusion track, the corrected environmental parameter and the posterior error covariance matrix;Step 3, based on the optimal fusion track, the corrected environmental parameter and the posterior error covariance matrix, a regional risk field is constructed;Step 4, based on the regional risk field, a multi-objective collaborative optimization model is constructed and solved to obtain the navigation control parameter;Step 5, execute navigation control parameter, and adopt multilevel closed loop feedback to optimize, complete offshore area navigation safety control and optimization.

The invention relates to an airborne end'low-slow-small 'target identification tracking method and system based on improved YoV5 and ECO, belongs to the technical field of unmanned aerial vehicle target identification and tracking, and solves the problems that'low-slow-small' target identification accuracy is low and tracking is easy to lose in the prior art. Adding a small-scale detection head P2 at a detection end to obtain the detection end of the low-slow-small target detection model; adding a feature fusion layer F2 corresponding to the detection head P2 to a neck network of the YoV5 model, and adding a CBAM model to each feature fusion layer of the neck network to obtain the neck network; training a low-slow-small target detection model based on the training set, performing low-slow-small target detection identification on a target image acquired in real time through the trained low-slow-small target detection model, and outputting a target identification result; and determining a tracking target based on the target identification result, and continuously obtaining continuous frame images of the tracking target based on a preset tracking algorithm.

A method for planning an autonomous landing safe trajectory suitable for UAV inspection is proposed. After the UAV takes off, it acquires the current pose of the UAV, the point cloud map information of the surrounding environment, and the spatial orientation of the target point. By establishing the spatial transformation relationship between the camera coordinate system and the body coordinate system, the orientation vector relative to the ground target is calculated, and the target point is generated within the radar-sensible area. Based on the generated target point, an optimization function that minimizes the fourth derivative is constructed based on multiple high-order polynomials to generate a smooth trajectory with continuous position, velocity, and acceleration. The initial trajectory is transformed into a B-spline curve, and local trajectory segments with collision risk are optimized. A segmented penalty mechanism is designed based on the boundary threshold to guide the trajectory away from the boundary area. The time span is adjusted by calculating the trajectory derivative over-limit ratio, and a trajectory similarity cost function is introduced to ensure that the final optimized trajectory maintains the original geometric structure, thereby achieving trajectory control optimization and safe landing of the UAV.

This invention provides a multi-channel image sensor data transmission system and optoelectronic system based on FPGA-PCIe, belonging to the technical field of video processing and acquisition systems. The multi-channel image sensor data transmission system includes an FPGA circuit board, external memory, and a back-end image processing platform. The FPGA circuit board includes an integrated image processing module, a bus control module, and a PCIe interface control module. Utilizing an FPGA to process and acquire multiple sensor video streams saves hardware costs and circuit board area. It also ports various operation operators to the FPGA, enabling real-time processing and acquisition of the currently transmitted video stream, improving the real-time performance of the operation. Simultaneously, it achieves control and transmission of multiple video streams, meeting the requirements for data transmission agility and high speed. This solves the problem in existing methods where inconsistent video resolutions and frame rates from different image sensors lead to insufficient data bandwidth to meet protocol requirements, resulting in the inability of PCIe devices at both ends of the FPGA to communicate normally.

This invention discloses a spatiotemporally integrated UAV image target tracking method, belonging to the field of aerial image processing technology. The method includes the following steps: Step 1, constructing a backbone network based on the VIT model; Step 2, constructing a network framework for a single-stream VIT model; Step 3, constructing a loss function based on contrastive learning; Step 4, adding a dynamic template update mechanism; Step 5, constructing two sequence decoder structures; Step 6, inputting the image and past sequences into the network to obtain the final tracking result. This invention provides a spatiotemporally integrated UAV image target tracking method. In the air domain, it designs a single-stream VIT-based backbone network and a contrastive learning-based loss function for small targets; in the temporal domain, it designs a dynamic template update mechanism and a trajectory prediction-based decoder to address background interference. These two structures together form a spatiotemporally integrated tracking algorithm, effectively improving the success rate and accuracy of target tracking from the UAV's perspective.

The invention belongs to the technical field of path planning, and more specifically relates to a limited space unmanned aerial vehicle bridge simulation inspection path planning method. The method comprises the following steps: sensing and subscribing point cloud information of an environment by an unmanned aerial vehicle, wherein the point cloud information comes from a laser radar, a depth camera or a fusion point cloud; filtering: carrying out point cloud noisy point self-adaptive multi-level filtering on the point cloud information, executing nine-step self-adaptive noisy point filtering, and outputting filtered point cloud and a free space F; planning and bridge-imitating inspection path generation are carried out, an initial path is generated through a narrow channel RRT *, a safe flight inspection corridor SFIC is constructed, and a middle bridge-imitating inspection path is generated in the SFIC in an optimized mode. The problems of insufficient environmental perception precision, low safety and efficiency of path planning, lack of dynamic adaptability of path planning and low path quality during the inspection process of an existing unmanned aerial vehicle in a limited space in a bridge are solved.

This invention discloses a multi-target real-time tracking method based on deep learning, comprising: Step 1, acquiring a real-time image sequence through a video acquisition device and standardizing the image sequence; Step 2, inputting the standardized image sequence into a neural network model to output target bounding boxes containing human features; Step 3, inputting the target bounding boxes into a tracker, wherein the tracker calculates the feature vector of each human based on the OSNet appearance feature extraction network, and uses the Hungarian algorithm to solve for the optimal association, assigning a unique identifier to each tracked human, and forming and updating a dynamic tracking list of multiple targets.

The invention provides a flight operation robot contact force tracking control method based on variable stiffness admittance. The method comprises the following steps: S1, establishing a dynamic model of a flight operation robot system; s2, designing a self-adaptive variable stiffness admittance model, generating an expected reference trajectory by dynamically adjusting a stiffness coefficient, and compensating the influence of unknown environmental parameters; s3, designing a nonlinear disturbance observer to estimate system disturbance, and providing disturbance compensation for the pose controller; s4, a pose controller is designed based on the obstacle Lyapunov function, reference trajectory tracking is achieved, and the position error is strictly restrained to be within a safe range; s5, the rotating speed of each rotor wing of the flying operation robot is calculated through control input; the flying operation robot is controlled to realize stable contact force tracking; according to the method, uncertainty and contact disturbance in an unknown dynamic environment can be effectively dealt with, and the contact force tracking precision and the operation safety of the flying operation robot are improved.

This invention provides a substation equipment installation guidance and control method and system based on semantic perception and EKF algorithm, belonging to the field of substation equipment installation guidance and control technology. To address the technical problems of current substation equipment installation guidance methods relying on manual experience, resulting in poor installation accuracy, dependence on stringent auxiliary markers, and inability to adapt to unstructured installation environments, this invention preprocesses the raw point cloud data of the acquired equipment, including voxel downsampling and statistical outlier removal, filtering out dust and noise from the construction site. The unordered point cloud is classified into base, equipment, ground, or background, and then subjected to perception processing and semantic segmentation. In the segmented point cloud data, the flange plane is extracted using a random sampling consistency algorithm, and the bolt hole center coordinates are identified using a clustering algorithm. Coarse and fine registration strategies are employed for virtual-real registration operations. This invention is applied to substation equipment installation guidance.

The application discloses a time cooperative guidance control method considering speed limitation of an unmanned aerial vehicle, and comprises the following steps: obtaining the motion state information of a target in real time through an optoelectronic pod carried on the unmanned aerial vehicle; obtaining the speed and acceleration of the target in real time according to the motion state information of the target; obtaining the acceleration instruction of the unmanned aerial vehicle in real time according to the expected attack time, the relative distance, the motion information of the target and the speed limitation of the unmanned aerial vehicle itself; and controlling the unmanned aerial vehicle to fly according to the obtained acceleration instruction of the unmanned aerial vehicle. The time cooperative guidance control method considering the speed limitation of the unmanned aerial vehicle can realize omnidirectional saturated attack on an arbitrary motion target under the speed constraint condition of the unmanned aerial vehicle.

This invention discloses a method and system for risk assessment and graded early warning of offshore wind farm vessels. The method involves collecting and fusing multi-source sensing data; performing spatiotemporal alignment and fusion processing on the fused multi-source sensing data to generate a bird's-eye view feature representation containing spatiotemporal characteristic information of the vessel; based on the bird's-eye view feature representation, employing a fusion prediction model combining a vessel motion physics model and a time series prediction algorithm, calculating the motion reachability set of the target vessel within the predicted time period; conducting a risk assessment based on the spatial relationship between the motion reachability set and a preset safe area of ​​the wind farm, and triggering graded early warnings based on the assessment results. This invention solves the problems of blind spots in single-source monitoring and high false alarm rates in traditional early warning systems, achieving accurate and dynamic assessment and proactive early warning of vessel collision risks in offshore wind farm waters, significantly improving regulatory safety and efficiency.

The application belongs to the technical field of computer vision, and discloses a three-dimensional point cloud registration method based on point-line feature fusion. The three-dimensional point cloud registration method firstly uses SAC-IA to complete static registration of three-dimensional point clouds on the surface of a target object, and establishes an initial motion transformation relationship from the three-dimensional point clouds of the 3D model of the target object to the three-dimensional point clouds on the surface of the target object. Then, in the dynamic tracking stage, point features extracted by ORB are matched with line segment features extracted by LSD-LBD, coarse registration is realized by combining an EPnPL algorithm, and the robustness of pose estimation is improved by eliminating false matching through RANSAC. Finally, in the fine registration stage, point-to-plane ICP is used to further optimize pose estimation. The method does not need to rely on artificial marking, and can complete stable three-dimensional point cloud registration based on natural features, and can be widely applied to complex scenes such as mechanical product maintenance and industrial assembly.

The application discloses a BWPT system power control method suitable for a variable receiving end of an electric vehicle. The system is composed of a network side and a vehicle side DC bus, a filter capacitor, a bidirectional full-bridge converter, a resonance compensation network, a magnetic coupling mechanism and a sampling and digital control module. The receiving end receiving coil structure is variable, and LCC or S compensation can be adopted. In the forward charging and reverse power feedback mode, the control module is based on the DC side voltage sampling and the power analysis model, and the rated output power is the primary target, and the system driving frequency and the external phase angle are adjusted. The method can realize stable and practical power control under the condition of "fixed primary side and variable secondary side".

The application discloses a soft manipulator preset time sliding mode tracking control method and system based on PSO-BP, belongs to the soft robot control technical field, and relates to the following technical scheme: a nonlinear model of a soft manipulator is established, a controlled object model is obtained by rewriting the soft manipulator, and robust right coprime factorization processing is performed to obtain an equivalent control object; a tracking error is constructed, an error integral term is introduced into a sliding mode surface, and an integral sliding mode surface is established; a compound sliding mode control law is designed based on the integral sliding mode surface; a plurality of control gains corresponding to system total convergence time is obtained through simulation; a sample data set of the control gain and the total convergence time is constructed; the sample data set is trained to establish an inverse mapping model of the expected convergence time to the control gain; a target convergence time is inputted to obtain a corresponding control gain, which is substituted into the compound sliding mode control law to generate a driving air pressure control amount. The application has the beneficial effects that a soft manipulator preset time sliding mode tracking control method based on PSO-BP is provided.

The invention relates to an unmanned ship prediction tracking control method and system based on a GRU disturbance observer, and belongs to the field of ocean intelligent equipment control. The method comprises the following steps: constructing an unmanned ship kinematics model and an unmanned ship dynamics model; transmitting sensor data to the unmanned ship kinematics model and the unmanned ship dynamics model, and calculating an actual disturbance value; and obtaining an unmanned ship linearization model according to the constructed GRU disturbance observer model, and obtaining a target control quantity for the unmanned ship linearization model through a model prediction control strategy. The method solves the problems that the accuracy of tracking the expected path by the unmanned ship is reduced and the course error is increased due to low prediction precision of model prediction control.

This invention discloses a pneumatic pitch drive system and method for an underwater semi-active flapping fin power generation device, belonging to the field of fluid energy conversion technology. The system includes a mounting frame, an air supply module, and a motion module. The motion module includes flapping fins and a pitch drive assembly, which comprises dual pneumatic chambers, a piston-rack assembly, gears, and an angle detection element. The air supply module is separate from the motion module and includes an air source and valve group for independently controlling the air supply and exhaust of the dual pneumatic chambers. The controller adjusts the pressure difference between the two chambers according to the deviation between the actual pitch angle and the target pitch angle, driving the piston-rack assembly to reciprocate, and driving the flapping fins to pitch motion via the gears, forming a closed-loop control. This invention reduces the use of underwater electrical components, lowers the risk of seal failure and corrosion, and overcomes the response lag and motion overshoot problems caused by the compressibility of gas through closed-loop control, improving the stability of pitch control and energy harvesting efficiency.

The application relates to the mechanical field and discloses a robot joint module, which comprises a first joint driving unit, a second joint driving unit and a joint output unit, the first joint driving unit comprises a first shift fork structure, a first hob, a first driving part, a second driving part and a first shell assembly, the second joint driving unit comprises a second shift fork structure, a second hob, a third driving part, a fourth driving part and a second shell assembly, and the joint output unit comprises a third shell assembly and a spherical gear. The application can realize high integration and three-dimensional posture output, the outputs of the first joint driving unit and the second joint driving unit are relatively independent and do not interfere with each other, but driving superposition and cooperative driving are carried out in different directions, continuous controllable output of multiple degrees of freedom is finally realized, synchronous constraint and mutual coordination among the components can be realized through cooperative matching, meshing impact caused by single driving in advance is further reduced, and stable tracking of the multiple degrees of freedom posture is realized.

The invention discloses a multi-agent trajectory tracking method and system fusing strategy learning and rolling optimization control, and the method comprises the steps: introducing a leader-follower control framework, building a first-order multi-agent follower model, and representing a target trajectory through a kinetic equation of a leader; an optimal control problem of trajectory tracking is constructed based on a control framework of rolling time domain optimization, a state error penalty term and a control input penalty term are introduced into a cost function, and a distributed strategy learning algorithm is used for solving and rolling optimization to obtain an optimal control strategy; an actor-commentator network is used for carrying out online strategy learning, step-by-step execution is carried out in the form of time increment in each prediction time domain, an optimal value function and a control strategy are obtained through iterative updating, the calculation complexity is remarkably reduced, and meanwhile the online learning and instant deployment capacity is achieved; and using the optimized strategy to act on the system through a rolling optimization mechanism to complete the tracking of the time-varying target trajectory.

This invention provides an automatic target region tracking method for surgical videos, comprising: real-time updating of a tracking template; monitoring the reliability of the tracking template using a confidence parameter; extracting features from the image of the tracking template and the search image respectively; processing the extracted features based on the current tracking template to obtain classification results for the foreground and background, and regression distances from the four edges of the search box to sampling points in the classification feature image; obtaining multiple predicted boxes based on their distances to the bounding boxes; and obtaining the optimal predicted box by applying translation and scale penalties, thereby obtaining the tracking result. This invention introduces an online tracking template update mechanism based on the SiamBAN target tracking method, ensuring that the tracking template is always associated with the target region of the current image, and maintaining stable tracking even when the features of the target region change; and introduces a PSR confidence parameter to correct erroneous updates during template updates in cases of large-scale occlusion, preventing tracking drift.

In order to improve the tracking performance of underwater low-speed small targets in a complex dynamic time-varying environment, the invention provides an improved potential equilibrium multi-Bernoulli smooth tracking method in the complex dynamic time-varying environment. According to the method, the state of a new target is adaptively estimated by measuring and driving the new target, and a prediction result is updated by using a survival target and a new target measurement set, so that the purpose of improving tracking performance is achieved; meanwhile, a forward-backward smoothing algorithm is used in a combined mode, target state estimation precision is improved by means of measurement information at multiple moments, the tracking effect of the method is improved by changing the smoothing step length, and the problem of target number error estimation during target extinction is solved; and finally, stable and high-precision target tracking in a complex dynamic time-varying environment is realized under the condition of not increasing overall observation data. Compared with a traditional potential equalization multi-Bernoulli filtering method, the method provided by the invention has more robust tracking performance and higher tracking precision in clutter dense and dynamic time-varying scenes.

The application discloses a multi-target tracking method based on a "digital model + video data", which is applied to the field of target tracking, and aims at the fact that existing tracking algorithms are more likely to lose targets in complex scenes or dense areas of aircrafts, leading to tracking interruption or incorrect target association. The appearance prior information of the digital model is used to guide the segmentation and tracking results of multi-aircraft targets, aircraft targets of specific angles and specific sizes are matched with aircraft models in a digital model library through an effective matching mode, so that stable target appearance representation is obtained in the tracking process. Finally, a better multi-aircraft tracking result is obtained through a complete segmentation result and a feature matching method.

This invention relates to the field of autonomous UAV inspection technology, and discloses an autonomous obstacle avoidance method for bridge inspection using UAVs. The method includes: structuring prior bridge information to generate a 3D bridge model; calculating the working distance and performing multi-constraint viewpoint planning based on camera parameters and the 3D model to generate an initial viewpoint set; performing global path planning for the initial viewpoints through multi-objective greedy search; combining the 3D model to perform viewpoint connection obstacle avoidance, convex polyhedron feasible region construction, and trajectory optimization on the initial path; registering the optimized path with real-time SLAM point clouds and converting it to the world coordinate system; performing online semantic segmentation, viewpoint generation, and local optimization based on real-time perception to generate sub-flight trajectories; performing dynamic obstacle detection and observation gain-guided replanning based on a semantic map; and finally achieving high-precision tracking of each trajectory through feedforward-feedback control. This invention achieves fully autonomous bridge inspection with complete coverage, clear imaging, and adaptability to dynamic environments.

The invention relates to the technical field of computer vision, in particular to a pet continuous tracking method and device, electronic equipment and a storage medium. Stable apparent characteristics extracted by a re-identification model based on data enhancement training, real-time position information and an expected position predicted based on a historical track are fused and matched; according to the invention, a globally unique and coherent identity label can be generated and maintained for each pet (whether a newly appearing first pet or a tracked second pet) appearing in a video, so that continuous and stable tracking of a specific pet individual at different times in the same scene is realized; and on the basis of the accurate and reliable continuous tracking trajectory, a crucial data support is provided for subsequent accurate behavior analysis, individualized feeding linkage and cross-camera global monitoring, and the core problem of separation of pet detection and tracking links in the prior art is effectively solved.

The embodiment of the application provides a kind of based on multi-scale time window adaptive injection extreme value search control method, for the multi-time scale thermal dynamic characteristics (such as TEC fast response and crystal overall thermal balance slow response) existing in solid laser crystal temperature control system, by simultaneously injecting N frequency geometric progression distribution, fully separated sinusoidal disturbance signal in frequency domain, different scale thermal dynamic process is matched respectively;For the efficiency gradient estimate obtained by demodulating each scale, the signal-to-noise ratio based on real-time calculation is adaptively weighted and fused, so that the gradient information with high signal-to-noise ratio dominates in the fusion, so as to effectively suppress measurement noise;At the same time, according to the variance of the fused efficiency gradient, the frequency of each scale disturbance signal is dynamically scheduled, realizing multi-stage adaptive switching from coarse search to fine search to lock.

The application discloses a control method and system of a high-pressure oxygen cabin and a computer storage medium, and relates to the technical field of intelligent control. The control method of the high-pressure oxygen cabin comprises the following steps: acquiring a current quasi-steady pressure and a current temperature; determining a pressure deviation, and combining an adiabatic pressure change relation to determine a gas mass that needs to be injected in a discrete pulse form; combining a valve flow characteristic and an isothermal steady-state pressure change relation to determine a valve full opening duration and a target steady-state pressure; completely closing after being opened for the full opening duration, executing the discrete pulse, and recording an adiabatic pressure extreme value; after the valve is closed, combining a pre-labeled thermal relaxation characteristic time to construct a prediction model, continuously monitoring the cabin pressure and judging whether the quasi-steady state is reached, and updating the quasi-steady pressure after completion; and taking the updated quasi-steady pressure as feedback, again determining the next discrete pulse, forming a closed-loop control, and thereby realizing smooth and accurate control of the cabin pressure.

The application discloses a model training method, a beam tracking method, a device, equipment and a medium. The model training method comprises the following steps: generating a millimeter wave beam tracking sample according to a historical millimeter wave beam index sequence; when a current beam tracking model needs to be trained at a sending end or a receiving end, performing beam tracking training on the current beam tracking model according to the millimeter wave beam tracking sample to obtain a target beam tracking model; and the target beam tracking model is used for tracking a millimeter wave beam. The technical scheme of the embodiment of the application reduces the time cost of millimeter wave beam search, keeps stable tracking of the millimeter wave beam sent by a terminal in fast motion, and can effectively adapt to a high dynamic communication environment.