133 results about "Perceptual information" patented technology

The invention provides a multi-mode sensor fusion algorithm for multi-signal processing and a system thereof, and relates to the technical field of sensor data processing, and the system sequentially generates standardized features through Sp1, generates a fusion feature set through Sp2, optimizes fusion parameters through Sp3, generates compensation features through Sp4, outputs perception information through Sp5, and outputs abnormal conditions such as sensor failure and overload calculation. Through feedback mechanism processing, a termination condition of perception information output or new task triggering, and breakthrough in feature retention and data consistency, a fusion result can still keep high precision in a high noise or complex scene, through failure detection and feature association degree priority ranking, other sensor data with high association degree are used for reconstruction, and a fusion result is obtained. According to the method, the accuracy of feature reconstruction is greatly improved, the high real-time performance of edge equipment and the computing power of the cloud are fully utilized by an edge-cloud cooperation mechanism, the resource utilization rate is optimized, and the stability and adaptability of the system in practical application are greatly improved.

The invention provides an end-to-end automatic driving track generation method and system, computer equipment and a medium, and belongs to the field of path intelligent planning, and the method comprises the steps: obtaining traffic environment three-dimensional perception information and a natural language instruction; converting the traffic environment three-dimensional perception information to obtain a structured semantic mark, and adjusting the weight of the structured semantic mark through a natural language instruction; based on the weight of the structured semantic mark, visual flow features are extracted from the structured semantic mark, and language flow features are extracted from the natural language instruction; based on a cross-modal attention mechanism, the visual flow features and the language flow features are fused, fusion features are obtained, and a simulation track is generated; real-time collision time is calculated, when the real-time collision time is larger than a collision time threshold value, the simulation track is corrected based on physical constraints and a traffic regulation library, and a corrected driving track is obtained. The automatic driving simulation track constructed by the method conforms to the intention of a user, and the system reliability and the man-machine cooperation efficiency are improved.

The invention provides a multi-agent collaborative visual completion method and system based on semantic communication, and the method comprises the steps: a first agent collects an occluded overall original image through a visual sensor, generates an occluded region mask through a lightweight semantic segmentation network, extracts mask semantic information, and transmits the mask semantic information to a second agent; the second intelligent agent matches the global semantic feature of the second intelligent agent with the received mask semantic information, locates a sheltered area, then cuts the sheltered area to obtain a semantic feature vector of the area, and returns the semantic feature vector to the first intelligent agent; and after the first intelligent agent receives the image, performing high-fidelity reconstruction on the occlusion area by using a conditional diffusion model to obtain a complemented image with consistent semantics, and performing weighted fusion and local color matching and splicing on the complemented image and the original image to obtain a global perception result. According to the scheme, perceptual information sharing and occlusion area reconstruction with low bit rate, high robustness and high real-time performance can be realized under limited communication resources.

The invention relates to the field of audio detection, and provides an audio deep counterfeiting detection method and system fusing multi-source features and cross-scale modeling, which effectively improves the recognition capability of a hidden counterfeiting mode in a synthetic sample through a double-branch data enhancement strategy, and improves the recognition capability of the hidden counterfeiting mode while keeping the speech naturalness and auditory acceptability. A representative forgery disturbance space is introduced, so that generalization and robustness under multi-type deep forgery attacks are remarkably improved, multi-level deep audio features and physical acoustic features are obtained respectively and are subjected to feature fusion, the sensitivity to physical layer artifacts is improved, the detection accuracy is further improved, and the detection efficiency is improved. And through multi-scale attention enhancement, weak change characteristics across time and frequency scales in the forged audio are efficiently captured, through structural design of grouped convolution and a multi-branch attention path, local and global perception information is effectively integrated, and the accuracy, robustness and generalization ability of audio deep forging detection are improved.

The invention discloses a time delay constrained Internet of Vehicles multi-mode cooperative sensing method and system. The method comprises the following steps: collecting point cloud data and image data, and carrying out feature extraction, fusion and space projection; constructing a two-dimensional convolution neural network, and extracting a position-level importance probability graph and a mask used for screening key information; constructing a benefit estimation neural network and a constraint evaluation neural network, and predicting expected values of system benefits and constraint conditions; meanwhile, an overall strategy selection module is established; the motorcade carries out target identification and calculates rewards based on information interaction so as to obtain a perception experience sample; and updating network parameters based on the perception experience sample to obtain a final overall strategy selection module. According to the method, the transmitting power, the transmission channel and the sensing data area of the vehicle are optimized based on reinforcement learning, a strategy correction mechanism is introduced, a risk strategy of high time delay or high packet loss is dynamically corrected, and target identification deviation and decision errors caused by sensing information delay and wrong transmission are avoided.

The invention relates to the technical field of intelligent sorting control. The multi-modal sensing sorting control method comprises the steps of obtaining multi-modal original sensing information, processing the multi-modal original sensing information, extracting category information and three-dimensional space pose information of a to-be-sorted target, generating a target recognition result, and recognizing the to-be-sorted target on the basis of the target recognition result. The method comprises the following steps: executing a path planning operation to obtain sorting path information containing a grabbing point location and a movement track, obtaining preset target clamping pose information, and generating a clamping control instruction based on the sorting path information and the preset target clamping pose information. The method has the effect of improving the sorting efficiency in a complex operation scene.

The invention relates to the technical field of medical image processing, and discloses an ultrasonic image edge segmentation method and system, a terminal and a medium, and the ultrasonic image edge segmentation method comprises the steps: constructing an initial model, obtaining a training set, inputting the training set into the initial model, and obtaining a prediction probability graph; performing data preprocessing on the training set to obtain boundary supervision information; boundary sensing information is obtained according to the prediction probability graph and the boundary supervision information; obtaining a target model according to the boundary perception information, the training set and the initial model; and acquiring an echocardiogram of a user, and inputting the echocardiogram into the target model to obtain a segmentation probability graph. According to the method, the boundary monitoring information is provided by preprocessing the training set to solve the problem of boundary blur, the perceptual ability of the model to the boundary is enhanced through the boundary monitoring information, and the boundary segmentation precision of the ultrasonic image can be improved.

The invention relates to a multi-vehicle cooperative sensing method based on an attention mechanism, and belongs to the technical field of mobile communication. A multi-vehicle collaborative perception scene lacks selective attention to perception information of different values, so that perception characteristics of each vehicle are difficult to effectively utilize. At the same time, in a high mobility scene, it is difficult to align sensing information time delay and asynchronous features, resulting in low dynamic target sensing precision. In order to solve the problem, the method comprises the following steps: firstly, filtering redundant sensing information of cooperative vehicles, and constructing a screening model based on aerial view feature intersection-to-union ratio so as to select high-quality cooperative vehicles; spatial calibration is completed in combination with sparse consensus foreground features and geometric consistency verification, and feature fusion is realized through an attention mechanism; and finally, finishing time calibration of the dynamic target by adopting a time alignment strategy based on distillation learning. The method is suitable for multi-vehicle cooperative perception in an Internet of Vehicles scene, the focus on high-quality perception information is improved, and meanwhile, the real-time perception precision and robustness are enhanced.

The invention relates to the technical field of vehicles, in particular to a multi-mode out-of-vehicle interaction method and related equipment based on environment self-adaptation, and the method comprises the steps: obtaining perception information in real time through an environment perception module, the perception information comprising environment conditions, interaction object information and scene semantic understanding information; based on the perception information, utilizing an adaptive decision engine to dynamically analyze an optimized interaction mode and parameters, and generating interaction content; outputting interaction information according to the interaction content generated by the self-adaptive decision engine through a multi-modal output module; the interaction strategy is dynamically adjusted by sensing the environment state in real time, and the safety and the interaction efficiency can be improved.

The invention discloses an equipment operation monitoring and fault diagnosis system based on multi-perception information fusion, and belongs to the technical field of equipment monitoring, and the system comprises the steps: collecting monitoring data in an equipment operation process, and carrying out the preprocessing of the monitoring data, so as to obtain monitoring processing data; obtaining an equipment fault label and corresponding monitoring processing data according to the historical equipment fault record; performing feature extraction on the monitoring processing data, and extracting feature parameters for reflecting the operation state of the equipment; calculating a mutual information value of each feature parameter and the fault tag by using a mutual information method, obtaining key features of which the mutual information values are greater than or equal to a key threshold value, and performing fusion through weighted summation according to a preset weight to form a multi-perception fusion feature; based on the multi-perception fusion features, fault diagnosis is carried out on the equipment by adopting a support vector machine classification algorithm, a diagnosis result is output, and finally the diagnosis result is displayed through a display and early warning module and an early warning signal is sent out.

The invention discloses a space robot compliance control method and device based on multi-dimensional visual constraints, and belongs to the technical field of space robots and autonomous control. The method comprises the following steps: collecting and fusing data through a heterogeneous visual sensor group, and carrying out uncertainty quantification to generate a perception confidence index; constructing and dynamically adjusting a visual constraint function set based on the index, solving a motion feasible region according to the visual constraint function set, and generating an expected motion track; inputting the track into an impedance controller, and adjusting the rigidity and damping parameters in real time according to the sensing confidence so as to output an instruction; the instruction is verified through digital twin simulation, and if the instruction does not pass the verification, a fault-tolerant mechanism is triggered for optimization; and finally, updating the system parameters on line according to the deviation between the actual execution data and the simulation data. According to the method, the technical problems of poor operation safety, low control precision and insufficient autonomy of the space robot in a complex environment caused by single sensing information dimension, lack of uncertainty quantization and fixed control strategy are solved.

According to the driving control method, the cooperative driving method and the related devices, in the method, communication between a vehicle and a roadside sensing device is achieved through a cloud device, so that under the condition that the vehicle cannot sense a first object, sensing information of the first object from the roadside sensing device can be obtained. And fusing the obtained sensing information with the sensing information obtained by the sensor of the vehicle so as to enlarge the sensing range, thereby realizing the driving control of the vehicle based on the fused sensing information. The scheme can effectively reduce potential safety hazards of vehicle driving, and is low in implementation cost and easy to popularize and use.

The invention discloses a traffic subject collaborative cognition method based on semantic representation in the technical field of traffic transportation. The method comprises the following steps: S1, collecting original data including state information of a traffic scene and a traffic subject, integrating the original data into perception information through a data fusion technology, and converting the perception information into structured semantic representation information; s2, establishing a multi-mode collaborative interaction architecture for different traffic scenes, dynamically selecting the collaborative interaction architecture according to a real-time traffic state, and performing ordered information interaction according to tasks; and S3, the edge traffic subjects autonomously generate decisions, and the central traffic subject performs overall optimization to realize multi-traffic subject collaborative cognition and behavior consistency. According to the method, semantic mutual understanding among different traffic subjects is realized, and an information barrier is broken; the method realizes the balance between local decision and global optimization, improves the decision efficiency, and has good expandability and adaptability.

The invention discloses a target detection system and method based on a multi-modal large language model, and the method employs a DETR to generate a proposal box as an object candidate, and achieves the precise detection of complex targets, such as long-tail classes, combined objects, and the like, through the powerful semantic reasoning capability of an MLLM. The method comprises the following steps: constructing a dual-encoder architecture in which a main visual encoder and an auxiliary visual encoder cooperate, enabling the main visual encoder to continue MLLM global semantic comprehension ability, enabling the auxiliary visual encoder to specially extract local features of an object, and achieving deep complementation of semantic and perceptual information on a dimension level through multi-layer and multi-scale feature splicing; and an object index and coordinate prediction dual-output mode is innovatively designed to form a retrieval-generation dual-guarantee system, so that the target detection precision and robustness are effectively improved.

The invention provides a multi-mode sensing inspection method based on terminal equipment. The multi-mode sensing inspection method comprises the following steps: collecting video data, audio data, geographic position data and IMU (Inertial Measurement Unit) data; video frames are extracted based on language trigger thinning, pose trigger thinning and visual trigger thinning; and finally, identifying hidden dangers based on the extracted video frames. According to the method, significant thinning processing of the video frames is realized through various kinds of perception information, the input load of the large model is greatly reduced on the premise that key semantic information is not missed, the problems of frame input limitation and low analysis efficiency of the multi-mode large model in analysis of the long-time industrial inspection video are solved, and the method is suitable for large-scale industrial inspection. Therefore, the efficiency and accuracy of large model analysis are improved. And the frame extraction process does not depend on a fixed frame rate, but is based on semantic driving of human behaviors and equipment states, so that the extracted key frame is more effective.

The present invention relates to an embodied intelligence agent based on physical intelligence and active perception, and a control method therefor. The method controls the embodied intelligence agent on the basis of a multimodal embodiment large model, and the multimodal embodiment large model comprises a data processing module, a perception and planning module, a physical intelligence module, an information transmission module, an environment interaction module, and a motion control execution module. The method comprises: the perception and planning module receives data processed by the data processing module and, according to the data, outputs perception information and a task planning instruction; the physical intelligence module receives the perception information and the task planning instruction, and outputs a motion action planning instruction; the environment interaction module collects environment information in real time and compares same with the perception information; and the motion control module receives the motion action planning instruction and incorporates the comparison result to perform action correction so as to implement control of the embodied intelligence agent. Compared with the prior art, the present invention has the advantages of high generalisation performance and high task execution accuracy.

An electronic device comprising circuity configured to determine, for a track selected from a set of tracks, a perceptual information loss related to removing the selected track from the set of tracks.

A controller may obtain perception information from a perception sensor system of the machine that includes a plurality of sensors, wherein the perception information includes respective perception data captured by the plurality of sensors. The controller may determine validity information associated with the perception information. The controller may select, based on the validity information, a portion of the perception information. The controller may determine, based on the portion of the perception information, a current distance to a target.

The invention provides an AI intelligent agent system, and the system comprises a perception processing module which is used for receiving external perception information and processing the external perception information through a corresponding model, and obtaining corresponding perception data; the cache middleware module is used for storing process data, and the process data comprises the perception data and / or context data; the agent scheduling module is used for acquiring a user instruction, acquiring the stored process data from the cache middleware module, filling the instruction and the process data into a cue word template, and delivering the cue word template to the LLM after fine adjustment; the LLM after fine tuning is used for reasoning according to the content of the cue word template to generate a reasoning result. The invention further provides a corresponding working method of the AI agent system. According to the method, the interpretability, the maintainability, the higher generalization ability and the higher response speed of modular design can be achieved at the same time.

The embodiment of the invention provides a sensing method and a related device. The method comprises the following steps: acquiring first sensing information of a plurality of communication devices for a first area; the multiple pieces of first perception information are input into a perception model, second perception information is obtained, and the perception model is used for predicting new perception information according to the input perception information; and optimizing the perception model according to the second perception information and third perception information of the first area, the third perception information being a reference value of the perception information of the first area. According to the embodiment of the invention, the accuracy of sensing information can be improved, so that the sensing performance of a communication network is improved.

A control unit (114) can receive perceptual information from a perception sensor system (112) of the machine (100), which comprises a plurality of sensors (126). The perceptual information includes respective perception data acquired by the plurality of sensors (126). The control unit (114) can determine validity information associated with the perceptual information. Based on the validity information, the control unit (114) can select a portion of the perceptual information. Based on this portion of the perceptual information, the control unit (114) can determine a current distance to a target.

The application discloses a kind of memory hub methods and systems for multi-agent cognitive decision-making, comprising: obtaining the perception information, behavior result or decision context generated in the running process of cognitive system;The perception information is structured modeling, and memory unit with time attribute, access attribute and evolution attribute is generated;The memory unit is stored and managed in life cycle;Before or in the process of cognitive decision generation, according to the current decision context, select relevant historical experience from the memory unit to participate in decision formation;Based on the use effect of memory unit in the decision process, update the hierarchical state or call strategy of the memory unit, and construct the continuously evolving cognitive decision process based on historical experience feedback. Unified management and monitoring can be carried out on memory access behavior, permission configuration, life cycle state and abnormal conditions, and parameter configuration and state viewing can be supported during system operation.

The present application discloses an operating parameter adjustment method, apparatus, electronic device, and storage medium. The operating parameter adjustment method includes: collecting user perception information; wherein the perception information includes at least one of the following: tactile-related information, auditory-related information, and visual-related information; determining a target perception intensity based on the perception information; and determining a target operating parameter of the target component based on a relevant value of the target component corresponding to the target perception intensity, wherein the target operating parameter of the target component includes any of the following: a target frequency of the CPU, a target frequency of the GPU, and a target bandwidth of the memory. Implementing this method is beneficial for optimizing the operating performance and power consumption of the target component.

The present invention provides an intelligent construction control method, apparatus, and system for splitting jet grouting mixing piles, relating to geotechnical engineering. The method includes receiving real-time perception information from a construction information self-perception system, which includes drilling depth, drilling rig output power, drill bit torque, drill rod axial force, and pore water pressure of the drill bit. This information is input into a trained machine-learning model for stratum information, establishing a nonlinear implicit correspondence relationship between the perception information and construction stratum conditions, and outputting real-time stratum state information. The real-time stratum state information is then input into a trained self-matching machine-learning model for optimal construction parameters, establishing a nonlinear implicit correspondence relationship between the stratum state information and optimal construction parameters, and outputting current optimal construction parameters. This invention enables real-time intelligent inversion of stratum information and dynamic self-optimization control of construction parameters.

The invention provides a target object sensing method, a communication device and a program product, relates to the technical field of communication, and can at least solve the problem of low target sensing accuracy in related technologies. The method comprises the following steps: acquiring perception information of a target object perceived by a plurality of perception devices at different positions; and selecting at least two pieces of target perception information from the multiple pieces of perception information, and confirming the behavior or position of the target object according to the at least two pieces of target perception information. The target sensing accuracy can be improved.

The invention provides a multi-scene state machine fused four-wheel-foot robot motion control method and system, and relates to the technical field of intelligent control, and the method comprises the steps: obtaining the perception information of a robot body; according to the perception information, constructing an observation space vector by fusing the current control instruction; inputting the observation space vector into a deep reinforcement learning control network based on an actor-commentator architecture, and generating an action instruction; training a deep reinforcement learning control network through a near-end strategy optimization algorithm according to the action instruction and a preset multi-dimensional reward function, and optimizing a motion control strategy; based on the optimized motion control strategy, constructing a finite-state machine framework, and defining a plurality of motion states; and according to the current topographic features and task requirements, the four-wheel-foot robot is controlled to be gradually switched among all the motion states through a preset state transition condition.

The present application relates to the field of intelligent automatic driving, and particularly relates to a knowledge-driven end-to-end automatic driving method based on a sparse expert mechanism and a diffusion model. The method comprises the following steps: S1: perception information processing and state coding; S2: sparse expert module construction and multi-task training; a sparse expert module composed of multiple experts is constructed, and reusable driving skills are obtained through multi-task behavior cloning training; S3: diffusion strategy network and action sequence generation; based on the diffusion model, a future multi-step control action sequence is generated from the current state condition, and a continuous and stable driving decision is formed; and S4: continuous learning and task migration mechanism. The present application constructs a modular driving knowledge structure that is combinable and interpretable, significantly improves the strategy modeling capability; the diffusion generation mechanism effectively improves the smoothness and stability of the decision-making process; the continuous learning and task migration mechanism with structural decoupling improves the long-term adaptability and deployment efficiency of the system.

This invention discloses a multi-agent cooperative anti-interference method based on spectrum data aggregation. First, a multi-agent cooperative anti-interference model based on spectrum data aggregation is established. Second, a joint action is selected based on a greedy strategy. Third, the joint action is executed to obtain a reward, the current spectrum is perceived, the state transitions to the next state, and the experience is stored in the agent's experience pool. Then, random batch sampling is performed from the experience pool, the target Q-value is calculated, the gradient of the loss function is calculated, and the weight values ​​are trained and updated. The algorithm terminates when the maximum number of iterations is reached. This invention features a complete model with clear physical meaning, avoids malicious interference and mutual interference between users, significantly improves the anti-interference performance of ultra-dense networks with incomplete perceptual information, and provides a fairer frequency usage strategy.