862 results about "Perception system" patented technology

Disclosed are a high-performance loosely-coupled multi-modal data fusion system for a smart driving environmental perception system and a vehicle-mounted device, comprising: a fusion detection model based on a modality-independent feature interaction strategy, which is configured for converting a LiDAR point cloud, a camera image, and a millimeter-wave radar point cloud into a unified bird's-eye view representation, and performing multi-modal fusion; and a fusion tracking model based on a motion-appearance feature cascaded coupling data association strategy, which is configured for performing subsequent trajectory tracking and matching according to multi-modal fusion feature information. A VoD data set and a K-Radar data set are selected for training, verifying, and testing the comprehensive performance of the models, and a TensorRT accelerated inference model is applied, then quantized, and deployed to a vehicle-mounted computational testing platform. The present invention is compatible with mainstream sensor deployment solutions, and achieves the efficient complementary fusion of multi-source heterogeneous sensor information, significantly improving the reliability, accuracy, and adaptability of vehicle-mounted perception systems, thereby effectively responding to extreme operating conditions such as complex traffic scenarios and inclement weather.

The invention relates to the technical field of target detection, in particular to a robot three-dimensional environment sensing method and device based on deep visual learning, and the method comprises the steps: collecting visual data based on a sensing system carried by a robot, carrying out the synchronous processing, extracting the spatial structure characteristics of a synchronous visual data stream and a preliminary semantic segmentation map, and carrying out the recognition of a target image; combining the spatial structure features with the preliminary semantic segmentation map to generate a geometric semantic feature map; extracting and optimizing local, regional and global features of the geometric semantic feature map, and performing three-dimensional modeling processing according to a multi-scale feature tensor to generate a three-dimensional geometric model; and carrying out fusion optimization on the structure information of the three-dimensional geometric model and the geometric semantic feature map, and carrying out verification processing based on a verification framework to obtain three-dimensional environment perception data. Through deep visual learning and visual data fusion, the defects of insufficient accuracy and limited deep semantic understanding ability in complex three-dimensional environment perception in the prior art are solved.

The invention relates to the technical field of intelligent power grids, in particular to an intelligent power distribution room sensing system and method based on data fusion. The multi-source data acquisition module is used for synchronously acquiring electrical parameters, mechanical vibration signals, temperature distribution data and environment monitoring data of power equipment in a power distribution room; the edge computing gateway is connected to the multi-source data acquisition module and is configured to perform time alignment, abnormal value elimination and feature extraction on the original sensing data; the data fusion analysis server is connected to the edge computing gateway through a network and comprises a space-time alignment unit used for unifying monitoring data of different sampling frequencies to the same time reference; the self-adaptive weight fusion unit is used for dynamically adjusting fusion weight according to the reliability of the data of each sensor; the state evaluation unit is used for generating an equipment health degree score and a fault early warning signal based on a fusion result; according to the scheme, monitoring blind areas and misjudgment risks caused by data islands can be fundamentally solved.

The embodiment of the invention provides an unhooking robot control method based on multi-source visual fusion, which is applied to the technical field of robot control and comprises the following steps: acquiring an RGB image, a depth image, an infrared image and IMU data through a multi-source sensing system mounted at the tail end of a robot; carrying out feature fusion identification by adopting a double-branch neural network, and outputting the boundary contour of the lifting hook and the three-dimensional coordinates of the optimal grabbing point; the visual coordinates are unified to a robot base coordinate system through a registration correction mechanism; a Transform prediction model is constructed based on the visual and inertial signals, and future pose changes of the lifting hook are estimated; a feedforward control track is generated to counteract swing of the lifting hook, and track correction is carried out in combination with visual servo feedback; and a joint instruction is generated through path planning and inverse kinematics solution, and the mechanical arm is driven to complete precise unhooking operation. According to the method, the recognition precision, the anti-interference capability and the operation success rate of unhooking operation in complex illumination and dynamic environments are effectively improved.

The invention discloses an electromechanical equipment health assessment and early warning method based on multi-mode dynamic perception, and belongs to the field of intelligent operation and maintenance of electromechanical equipment. The problems that in the prior art, a single physical quantity cannot comprehensively reflect the equipment state and a traditional signal processing algorithm cannot adapt to the equipment degradation mode change are solved, a panoramic sensing system covering multiple physical fields such as vibration, temperature and noise is constructed through a multi-mode sensor network and a dynamic weight fusion algorithm, and the multi-physical-field multi-physical-field panoramic sensing method is applied to the multi-physical-field multi-physical-field panoramic sensing system. The problem of isolated island of traditional single-dimensional monitoring information is solved; through a physical-depth mixed feature extraction architecture, combining interpretable engineering features with abstract features extracted by a deep neural network to form a health assessment model with mechanism transparency and mode generalization ability; through deep integration of the digital twin platform and the RPA technology, the manual inspection frequency and workload are reduced, the fault recognition accuracy is promoted to increase year by year, and continuously optimized intelligent operation and maintenance ecology is formed.

The embodiment of the invention provides a multi-modal fusion deep learning analysis method and system. The method is applied to the technical field of multi-modal learning, and comprises the following steps: obtaining multi-modal original data, sequentially processing image, text, audio and video data, and extracting visual features of the image, semantic features of the text, frequency spectrum and time sequence features of the audio, image features and time sequence features of a video frame and time domain features of an audio sequence; and then, according to the complementary information of the multi-source features, fusion processing is carried out to form a unified multi-modal feature representation, the unified multi-modal feature representation is input to a preset deep learning analysis model, and finally a multi-modal analysis result of comprehensive expression is obtained. According to the scheme, information complementarity and robustness are enhanced through multi-modal feature fusion, the comprehensive analysis capability of the model on semantic understanding, behavior recognition and state judgment in a complex scene is remarkably improved, and a more accurate, efficient and stable decision basis is provided for a multi-modal intelligent sensing system.

The invention provides a scene adaptive projection vehicle lamp system based on deep reinforcement learning and a control method, and relates to the technical field of intelligent vehicle lamps and automatic driving perception systems. Comprising a multi-modal sensing module, a feature fusion module, a strategy generation module and an execution module. The multi-mode sensing module is used for collecting environment state data and performing primary processing to form an environment data information flow; the feature fusion module is used for generating a unified environment feature vector for the environment data information flow; the strategy generation module is used for receiving the environment feature vector and generating a vehicle lamp adjustment strategy through multi-layer neural network calculation; evaluating a result obtained by executing the vehicle lamp adjustment strategy based on the vehicle lamp, and optimizing strategy parameters of the strategy network based on a PPO algorithm; the execution module is used for controlling the vehicle lamp according to the vehicle lamp adjustment strategy output by the strategy network. The intelligent level of the vehicle lamp is remarkably improved, and the system is widely applied to night driving assistance, urban interaction prompt and low-visibility driving scenes.

The invention relates to the technical field of body-equipped intelligent robots and intelligent equipment, in particular to a self-adaptive flexible tightening body-equipped intelligent robot which comprises a body-equipped moving platform, the top of the body-equipped moving platform is provided with a body-equipped lifting mechanism, and the top of the body-equipped moving platform is provided with a body-equipped articulated arm module. And a tightening actuator and a clamping jaw actuator are arranged at the tail end of the outer part of the body articulated arm module. According to the self-adaptive flexible tightening intelligent robot and the tightening process method, sensing and self-adaptive adjustment of external environment changes are achieved through multi-module integration, and the tightening process method based on a multi-mode sensing system and an intelligent control system is constructed; vision, force sense and touch sense data are fused, real-time monitoring and dynamic adjustment of the tightening process are achieved through a machine learning algorithm, and the intelligent level and quality stability of tightening operation are effectively improved.

The invention belongs to the technical field of artificial intelligence and intelligent sensing, particularly relates to a multi-source sensor fusion sensing system based on adaptive noise suppression, and aims to solve the problems of noise interference, modal mismatch and insufficient robustness in multi-source sensor fusion in a complex dynamic environment. The system comprises a front-end preprocessing module, an adaptive noise suppression engine, a multi-modal feature alignment unit, a credibility-driven fusion reasoning core and a closed-loop feedback optimization mechanism. Through real-time noise modeling and dynamic weight adjustment, high-precision alignment and fusion of multi-source signals are realized, and the sensing stability and real-time performance in an extreme scene are significantly improved.

The invention relates to the field of industrial automation, in particular to a multi-camera vision stacking pose planning method and system for an intelligent loading and unloading robot. The method comprises the following steps: constructing a collaborative sensing system through a plurality of industrial cameras deployed in a target space, and collecting dynamic environment data of a stacking area in real time; generating a self-adaptive stacking pose based on a hierarchical pose planning mechanism: analyzing an initial pose parameter through a theme control layer and generating a preliminary placement instruction, and verifying the feasibility of the preliminary placement instruction based on environmental data of the collaborative awareness system through a visual verification layer, and a final placement coordinate is optimized by combining the accurate positioning layer with a virtual box matching algorithm. The problems that in the prior art, real-time environment sensing capacity is lacked, pose planning is disjointed with the actual environment, and dynamic change scenes cannot be processed are solved.

The invention relates to the technical field of battery fault detection, and particularly discloses a new energy automobile battery fault intelligent diagnosis method and system, and the method comprises the steps: collecting the three-dimensional time sequence data flow of the voltage, temperature and internal resistance of a battery pack through a distributed sensor array; the environment temperature, the number of charge and discharge cycles and the vehicle operation condition parameters are synchronously integrated as auxiliary diagnosis dimensions, and a multi-source heterogeneous data sensing system is constructed; the method comprises the following steps: preprocessing original data by adopting a dual-channel hybrid filtering architecture based on a multi-source heterogeneous data sensing system, and establishing an adaptive filtering parameter adjustment mechanism to generate multi-modal data; according to the method, multi-dimensional time sequence data and auxiliary diagnosis parameters of the battery pack are comprehensively collected through a constructed multi-source heterogeneous data sensing system, a rich and accurate data basis is provided for fault diagnosis, noise interference is reduced through a two-channel hybrid filtering architecture and a self-adaptive filtering parameter adjustment mechanism, and the fault diagnosis accuracy is improved. And a solid foundation is laid for subsequent feature extraction and model training.

The invention discloses a unified framework for solving automatic driving track prediction and planning consistency based on a world model. According to the method, through cooperative work of the automatic driving domain controller and the vehicle-mounted sensing system, end-to-end joint optimization of track prediction and planning in a complex traffic scene is realized, time sequence dependence and interaction dynamics among intelligent agents are accurately captured, and the prediction capability and robustness of a model are remarkably improved. The method comprises the following specific steps: firstly, constructing a generative world model, and generating potential future state representation by utilizing a behavior conditional and backtracking expansion technology; secondly, in combination with global modeling and a local convolutional network, multi-scale features are extracted, adaptive fusion is carried out, and a multi-modal prediction trajectory is generated; then, a multi-target planning model is adopted to integrate various driving indexes, and a track with the minimum loss function is generated; finally, path planning parameters are dynamically optimized through real-time environment perception and decision feedback, and the problems of prediction uncertainty and planning consistency of the automatic driving track are effectively solved.

The invention discloses a driver state sensing system based on physiological indexes and external behavior analysis, which relates to the technical field of vehicle active safety control, and comprises a sensor group integrated in a driver contact or close area and a vehicle control part and used for collecting physiological signals and behavior characteristic signals of a driver in real time, the physiological signals at least comprise heart rate and skin electric signals, and the behavior characteristic signals at least comprise eyelid movement, head posture and steering wheel operation signals. According to the driver state sensing system, the comprehensiveness and accuracy of driver state sensing are remarkably improved by integrating physiological indexes and external behavior analysis, the limitation of single signal judgment is effectively overcome through a multi-source signal fusion mechanism, the reliability of state evaluation is ensured, and the advanced signal preprocessing technology adopted by the system is high in reliability. The data anti-interference capability is enhanced, the feature extraction precision is improved, and the robustness of the model is enhanced.

A perception system may be used to generate bounding boxes for objects in a vehicle scene. The perception system may receive images and feature maps corresponding to the received images. The perception system may correlate object queries from previous time steps with object queries from the current time step.

The invention discloses a bus and station interactive scheduling method and system based on V2X and deep reinforcement learning decision, a medium and equipment, and the method comprises the steps: collecting bus operation dynamic data, station passenger flow and environment data in real time, and combining V2X network interaction information to construct a full-dimension perception system; a deep reinforcement learning model is adopted for dynamic decision making, intelligent closed-loop control of vehicle scheduling, station service and traffic signal cooperation is achieved, the bus punctuality rate is remarkably increased, the waiting time of passengers is shortened, operation safety is guaranteed through active early warning of abnormal conditions, and intelligent upgrading of a bus system from passive response to active prevention is achieved.

The invention is applicable to the field of intelligent audio adjustment, and provides an immersive audio-video follow-up adjustment method and system, and the method comprises the steps: constructing a multi-dimensional perception system, and collecting multi-source information in real time; carrying out fusion processing on the collected multi-source information based on a deep neural network model, mining a dynamic mapping relation between a user state and the video content through space-time correlation analysis, and identifying a user interaction intention and an emotional tone and a space scene attribute of the video content; according to a fusion processing result, calling a dynamic parameter adjustment engine, and generating an audio parameter adjustment scheme in real time; a user experience feedback closed loop is constructed, a visual attention area of a user is collected through eye movement tracking equipment, and personalized adjustment preference parameters are generated; according to the method and the device, the audio effect is accurately matched with the user state and the audio and video content, the naturalness and the adaptability of immersive experience are remarkably improved, universality and individual differences are considered, and the audio experience which is more suitable for scenes and needs of the user is brought to the user.

The invention discloses an airport hub construction intelligent regulation and control method and system based on BIM + 5G, and relates to the technical field of construction scheduling optimization, and the method comprises the steps: collecting on-site real-time data to construct a path construction sequence model, constructing a dynamic construction state set with consistent space and time sequence through component coding and semantic attribute mapping, and constructing a dynamic construction state set with consistent time sequence; and constructing a minimum disturbance optimization algorithm of four-dimensional disturbance cost based on the construction disturbance mapping graph, outputting procedure sequence adjustment, resource rearrangement, path decoupling and environment avoidance intervention suggestions, and tracking a construction response state in real time based on an intervention execution feedback mechanism. According to the method, unified modeling of construction plans, resource allocation and green indexes is achieved by constructing a ternary structure and a construction disturbance mapping graph, the field state is dynamically collected in combination with a 5G and edge sensing system, process abnormity and resource conflicts are accurately recognized, an efficient and executable regulation and control strategy is generated based on a multi-target disturbance optimization algorithm, and the efficiency and the reliability of the system are improved. And the intelligence, responsiveness and energy-saving level of the construction process are obviously improved.

The invention provides a high-frequency welded pipe roller wear state monitoring method and device, and relates to the technical field of artificial intelligence and data processing, a holographic sensing system of working condition parameters and physical signals is constructed through synchronous monitoring of a multi-source sensor, and noise spectrum offset is tracked in real time based on time-frequency expression of data in different modes, so that the real-time monitoring of the wear state of a high-frequency welded pipe roller is realized. And an anti-error filtering protection channel is constructed, so that the inherent characteristics of the damage can be effectively captured. In addition, multi-physical field damage characterization information is fused from multiple complementary dimensions of transient impact (instantaneous frequency vector), spectrum complexity (spectral entropy) and amplitude distribution deflection (skewness), more comprehensive and richer state description can be provided, accurate quantitative evaluation of the roller wear level is achieved, and the capacity of real-time monitoring and accurate recognition of the roller wear state is remarkably improved.

The invention provides an unmanned aerial vehicle full-time perception image reconstruction method based on multi-mode cooperative reinforcement learning and degeneration decoupling. A frequency perception feature modulation model, a dual-mode dual-domain transformation module and a dynamic bidirectional guide mechanism are included. According to the system, firstly, feature information of different frequency bands is adaptively separated and modulated through a frequency sensing feature modulation model, and decoupling and compensation of composite unknown degradation are achieved; realizing cross-domain interaction and information fusion of visible light and infrared characteristics in a spatial domain and a channel domain by using a bimodal dual-domain transformation module; and finally, realizing collaborative enhancement of cross-modal degradation perception through a bidirectional dynamic guide mechanism, and generating an unmanned aerial vehicle visible light reconstruction image and an infrared super-resolution image with higher structural consistency and texture fidelity. According to the method, deep fusion and degeneration decoupling of multi-modal information can be realized in a complex degeneration environment, and the imaging quality and the environmental adaptability of an unmanned aerial vehicle full-time sensing system are remarkably improved.

The invention discloses a sensor-based agricultural information data acquisition system and method, and belongs to the technical field of agricultural information acquisition. Multi-type sensor nodes are arranged in an agricultural target area in a heterogeneous manner; the method comprises the following steps: constructing a multi-dimensional influence factor model according to crop growth stages and environmental historical fluctuation data, dividing initial sensing sub-regions, and configuring a sensor cluster; dynamically updating the sensing boundary based on the historical change rate and the spatial gradient information; fusing the heterogeneous data by adopting a multi-channel time synchronization mechanism to generate a standardized environment vector set W; calculating a sampling priority matrix according to a parameter change trend in the W, and adaptively adjusting a node state; integrating an energy consumption estimation model, and executing low-power-consumption scheduling; when any parameter exceeds the threshold, high-density sampling and remote early warning are triggered; according to the method, high-precision, low-power-consumption and dynamic-response data acquisition and intelligent early warning can be realized, and the efficiency and reliability of an agricultural sensing system are improved.

The invention discloses an obstacle avoidance control method and system for unmanned aerial vehicle inspection, and relates to the technical field of unmanned aerial vehicle inspection path planning, the surrounding environment is sensed through a sensing system carried by an unmanned aerial vehicle, and navigation points falling into a sensing range are dynamically obtained; if multiple navigation points exist in the sensing range, the multiple navigation points serve as local targets at the same time to apply a gravitational potential field, the gravitational force of the navigation points is automatically distributed according to the distance, then the gravitational force of all the navigation points is summed, the total gravitational force is obtained, and multi-target resultant force guiding is achieved; when obstacle repulsive force calculation is carried out, a speed vector of an obstacle is extracted, future position changes of the moving obstacle are predicted, a time factor and an obstacle volume parameter are introduced into a potential field model, and self-adaptive adjustment is carried out on a repulsive force function; the problems that the unmanned aerial vehicle is prone to deviating from the global direction during obstacle avoidance, the path is discontinuous or the unmanned aerial vehicle detours too far, and an active avoidance mechanism still lacks when a high-speed dynamic obstacle occurs are effectively solved.

The invention relates to an unstructured environment-oriented heterogeneous data fusion sensing system for an agent with a body, which can be applied to the technical field of intelligent control. The system comprises a sensing module and a control module, the sensing module is used for acquiring multi-dimensional physical attribute data, three-dimensional space form data and dynamic mechanical response data of the intelligent body in real time; the control module is used for performing timestamp synchronous marking on the multi-dimensional physical attribute data, the three-dimensional space form data and the dynamic mechanical response data, and realizing spatial consistency mapping of the multi-dimensional physical attribute data, the three-dimensional space form data and the dynamic mechanical response data through coordinate system conversion to obtain heterogeneous data; performing feature matching on the heterogeneous data to generate a fusion perception vector; based on the fusion perception vector, generating an autonomous decision instruction corresponding to the body agent; and controlling the body agent to execute a target action corresponding to the autonomous decision instruction. By adopting the system, the control accuracy of the intelligent body can be improved.

The invention relates to the technical field of environment perception, and discloses a fusion perception system and method based on a multispectral sensor. According to the method, asynchronous heterogeneous environment data streams collected by a visible light sensor, a thermal imaging sensor, a millimeter-wave radar and a laser radar are obtained, environment information of multispectral bands is covered, comprehensive capture of target appearance, temperature, motion and geometric characteristics is achieved, space-time registration operation is executed on asynchronous heterogeneous data, space-time deviation between the sensors is eliminated, and the accuracy of target tracking is improved. And a multi-source sensing data cube with time-space synchronization is generated, and a consistent data basis is provided for subsequent fusion. Based on the data cube, a hierarchical feature fusion architecture is adopted to extract cross-spectrum joint features, a multi-dimensional feature tensor is generated, and internal association and complementary information of multi-source data is deeply mined. The multi-dimensional feature tensor is processed through a dynamic weighting decision mechanism, a three-dimensional situation awareness map of an environment target is output, environment dynamic change and sensor data fluctuation are adapted, and the reliability of environment awareness in a complex scene is improved.

The invention relates to the technical field of agricultural facility management, and discloses an AI-based agricultural scene omnibearing perception system and an implementation method, and the system comprises an order analysis module which is used for receiving order data and generating decision variables based on the order data; the collaborative optimization module is used for obtaining an optimization scheme based on a multi-objective function; the resource arrangement module is used for calculating a resource configuration scheme based on multi-target scheduling optimization; the sensing correction module is used for updating model parameters of the prediction model; the settlement loop module is used for performing delivery settlement based on the actual execution data; according to the invention, through adoption of a collaborative architecture, full-chain intelligent decision-making from market order analysis to agricultural product delivery is realized, through full-chain optimization and accurate decision-making, the net income of agricultural production is improved, and the production cost is reduced; through multi-objective function construction and a robust optimization algorithm, an optimal decision scheme can be found under complex constraint conditions, and the overall benefit of agricultural production is significantly improved.

The invention discloses a spine tactile sensing system and method based on a multi-modal sensor, and the method comprises the steps: obtaining an original tactile signal, carrying out the preprocessing of the original tactile signal, carrying out the feature extraction based on the preprocessed original tactile signal, and obtaining a spatial distribution feature and a time dynamic feature; on the basis of tail end coordinates collected by forward kinematics of the tail end of the mechanical arm, space mapping of the tactile signals and the spine anatomical structure is constructed, and anatomical position features are obtained on the basis of the preprocessed original tactile signals and the space mapping; constructing task-oriented feature vectors based on the spatial distribution features, the anatomical position features and the time dynamic features, wherein the task-oriented feature vectors comprise a sliding state recognition feature vector and a spine ordinal number positioning feature vector; and constructing a multi-stage adaptive evolution hidden Markov model, and realizing spine ordinal number positioning in combination with the task-oriented feature vector. According to the method, through multi-dimensional feature fusion and dynamic model adjustment, anatomical prior knowledge and real-time tactile signals are combined, and spine segments are accurately recognized.

The invention provides an intelligent micro-grid panoramic sensing system based on digital twinning, which relates to the technical field of micro-grid panoramic sensing and comprises a data acquisition layer, a data processing layer, a digital twinning model layer, an application service layer and a user interaction layer. A sensor is divided into a core layer, key equipment / nodes, an edge layer and general equipment / non-key nodes, a self-adaptive sampling frequency adjustment mechanism is introduced, the sampling frequency is dynamically adjusted according to the running state of the micro-grid, the frequency is reduced when the micro-grid is stable, and the frequency is improved when the micro-grid is abnormal, so that the sampling efficiency is improved. The problems of data redundancy and insufficient real-time performance of a traditional sensor network are solved, high-precision collection of key data, efficiency optimization of non-key data, collaborative data processing of edge calculation and federated learning are considered, a lightweight deep learning model is deployed at edge calculation nodes, and real-time data preprocessing, filtering, denoising and feature extraction are achieved.

The invention relates to an electrochromic window multi-scene control method, system and device based on indoor human body thermal comfort and a medium. The method comprises the steps that a perception data set is constructed; based on the sensing data set, performing multi-parameter coupling calculation of solar incidence azimuth deviation constraint and dynamic projection overlapping to obtain a direct radiation component, performing window space coupling analysis and double space compensation calculation to obtain a scattered radiation component, and performing fusion to obtain an individual thermal radiation influence value; determining a scene category based on the perception data set, and establishing a thermal radiation evaluation benchmark by adopting a mode matched with the scene category; and if the thermal radiation evaluation benchmark exceeds a preset threshold value, the required transmissivity of the window faces is obtained through a full-transmissivity radiation distribution method, the corresponding window faces are adjusted to the target transmission gradient according to the required transmissivity, and coupling control over the wind direction and / or power of the air conditioning system is selectively triggered. According to the invention, by constructing a multi-dimensional dynamic sensing system and an intelligent control framework, collaborative optimization of human body thermal comfort and building energy efficiency is realized.

The invention discloses a numerical control machine tool dynamic error cooperative compensation control method adaptive to high-speed machining, and relates to the technical field of numerical control machine tool error control. According to the method, a multi-source dynamic error sensing system comprising a grating displacement sensor, a six-dimensional force sensor and the like is constructed to acquire data; after wavelet threshold denoising and Kalman filtering preprocessing, inputting a three-layer LSTM error coupling prediction model combined with an attention mechanism, embedding a servo motor load characteristic curve in the model, and outputting three types of error compensation amounts; through servo-level compensation and machining-level compensation, the position of a feed shaft, the rotating speed of a main shaft, the cutting feed rate and the behavior of a micro-displacement actuator are corrected, and machining errors caused by deflection and vibration conduction of the main shaft are counteracted. And iteratively updating model parameters by using a gradient descent algorithm. According to the method, through multi-source error synchronous sensing, error coupling modeling and hierarchical cooperative compensation, dynamic error cooperative control more adaptive to a high-speed processing scene is realized, and the method has a wide application value.

The invention provides an AI human shape recognition perception system and method based on binocular vision, and the method comprises the steps: carrying out the real-time collection through a binocular camera when a doorbell key is triggered, and carrying out the preprocessing of an original image collected in real time; performing coarse parallax estimation on the real-time image rectification to obtain a full-field coarse depth map, determining a human shape candidate region list by using the full-field coarse depth map and combining the heat source region of interest, and performing fine parallax estimation on the human shape candidate region list to obtain a fine depth patch; converting the corresponding fine depth patch into a three-dimensional point cloud set according to the pose information, and performing scale prior screening based on the corresponding three-dimensional point cloud set to obtain a plurality of human shape candidate reserved areas; and performing fusion identification according to the extracted multi-modal features to obtain a human shape identification result. According to the technical scheme provided by the invention, layered parallax estimation and three-dimensional scale prior screening can be carried out on the real-time image to realize high-reliability human shape recognition under the condition of low power consumption, so that the recognition reliability of a sensing system is improved.

The invention relates to an automatic driving intelligent safety multi-task dynamic test system based on a Carla simulation platform, and belongs to the field of artificial intelligence safety and automatic driving simulation test. The system comprises an automatic driving perception model integration and training module which comprises a multi-task data collection module, a fine tuning perception model and a compilation perception model Carla API interface; the attack and defense algorithm integration and training module is used for finely adjusting an attack and defense algorithm, writing an attack and defense algorithm Carla API (Application Program Interface) and replacing Carla object textures with an adversarial patch; and the real-time attack and defense deduction and evaluation recording module comprises a sensor for acquiring Carla real-time data, replacing textures, starting a perception model and starting an evaluation record of the success rate and the accuracy rate of defense and attack. According to the method, the robustness of the automatic driving perception model in a complex attack scene can be comprehensively evaluated, security holes of a perception system can be found in time, and the reliability and the security of an automatic driving technology in a complex environment are improved.