14 results about "Location aware" patented technology

A multi-target tracking and monitoring method and system with high trajectory consistency, belonging to the field of computer vision, solves the problems of identity switching and trajectory fragmentation caused by long-term target occlusion and dense interaction in multi-target tracking, especially in UAV monitoring scenarios. The method includes the following steps: 1. Inputting a video sequence, performing target detection and feature acquisition for each frame; 2. Performing preliminary association through a basic tracker to obtain an initial trajectory; 3. Detecting the disappearance and appearance of abnormal trajectories using a position awareness submodule; 4. Storing abnormal trajectory features using a long-term memory submodule; 5. Performing secondary association using a multi-step cross-frame matching submodule; 6. Completing trajectory updates. This invention is applicable to UAV monitoring scenarios such as high-altitude inspection, low-altitude operations, and outdoor scene monitoring.

The application discloses a remote sensing image language description generation method based on position perception and reinforcement learning. First, a multi-scale feature extraction module is used to obtain global image embedding, CLIP image embedding containing language features and regional image embedding of the remote sensing image respectively. Second, position perception attention processing is performed, the global and regional features are compressed and matrix interaction is performed to generate regional enhanced features containing spatial topological relationship. Then, multi-scale feature fusion is performed, the attention mechanism with random drop strategy and the gating mechanism are used to deeply fuse the regional enhanced features and the CLIP image embedding. Finally, the multi-scale fused features are input into an improved Transformer remote sensing image language description model to generate accurate and spatially logical remote sensing image natural language description. In addition, the application adopts a supervised pre-training and reinforcement learning fine-tuning combined training mode to improve the generation accuracy of the model.

A location-aware search assist capability identifies location-aware search query suggestions using location information associated with the location-aware search query suggestions. A user's search query input and location and a location associated with each location-aware search query suggestion candidates may be used to identify a set of search query suggestions for presentation to the user. Location-aware search query suggestion candidates may be ranked in accordance with a closeness of each one's location to the user's location. The ranking may be performed using a score, such as a popularity score associated with each search query suggestion candidate. The location-aware search query suggestion candidates having a location closer to the user's location may be promoted by adjusting each candidate's popularity score upward, and the search query suggestion candidates that are farther away from the user's location may be demoted by adjusting each such candidate's popularity score downward.

This invention discloses an outdoor positioning method based on the supervision and fusion of multi-source trajectories of WiFi and Bluetooth, belonging to the field of outdoor positioning and location awareness technology. The method includes: collecting signal strength data from a target terminal at multiple time steps using WiFi access points and Bluetooth beacons to obtain initial WiFi and Bluetooth positioning trajectories; performing temporal alignment and preprocessing on the initial WiFi and Bluetooth positioning trajectories and the actual positioning trajectories to form WiFi, Bluetooth, and actual positioning trajectory sequences; dividing the WiFi, Bluetooth, and actual positioning trajectory sequences using a sliding window; for each sliding window's WiFi, Bluetooth, and actual positioning trajectory sub-sequences, the method includes: constructing multi-dimensional features; inputting the multi-dimensional features into a CNN-LSTM model, using the actual positioning trajectory sub-sequence as supervision, to generate optimized WiFi positioning sub-trajectories; and concatenating all WiFi positioning sub-trajectories to obtain the optimized WiFi positioning trajectory.

The application discloses a kind of open coal mine vehicle based on rtk post-difference vehicle high-precision positioning method, comprising: first query fusion multi-path heat map information digital twin scene map to obtain environmental feature label, according to this adaptive adjustment multi-source sensor fusion strategy, generate first positioning solution, then through 5G network and edge server cooperation, using network rtk enhancement correction number correction to obtain second high-precision positioning solution, finally calculate and output fusion multi-dimensional information real-time credibility score.The method effectively solves the problem of positioning failure under the complex environment of open pit mine, realizes continuous, accurate and reliable position perception, and provides reliable guarantee for unmanned operation.

This invention discloses an intelligent marketing method, system, and program product based on artificial intelligence, relating to the field of intelligent marketing technology. The method first performs real-time anonymization processing on visual perception data before uploading it to a cloud server along with location perception data. The cloud server then extracts static attributes and dynamic behavior sequence sets from the perception data, generates a dynamic-static fusion feature vector through a dynamic-static feature fusion network, inputs it into a real-time intent state machine to predict the intent state transition probability, and calculates the dynamic service urgency based on customer value scores, thereby generating a customer service waiting queue. Service tasks are then allocated and instructions are issued based on the working status of a shopping guide robot. The shopping guide robot executes the service and feeds back the results to the cloud server to update customer value scores and state transition parameters. This approach, while protecting customer privacy, achieves accurate perception of customer intent and optimized scheduling of service resources, significantly improving the accuracy and efficiency of intelligent marketing.

An intelligent, standalone, and modular secure parcel management platform integrates Internet of Things (IoT) architecture with artificial intelligence (AI) and machine learning (ML) to autonomously receive, authenticate, monitor, and secure delivery, return, and exchange of parcels. The system includes a parcel receptacle equipped with a removable and upgradeable IoT enclosure housing cameras, sensors, processors, communication modules, and power components, wherein AI and ML models analyze images, video, sensor data, contextual signals, and external data sources to identify delivery events, authenticate delivery personnel, detect anomalies, predict delivery windows, and dynamically control access mechanisms. The platform supports edge and cloud-based intelligence, authenticated integrations with carrier and e-commerce systems, location awareness, and real-time user interaction via networked devices. By combining modular hardware design with adaptive AI-driven decision making, the invention provides a scalable, future-proof secure delivery infrastructure that improves reliability, automation, and security of unattended parcel handling across residential, commercial, and multi-unit environments.

The present disclosure provides a position perception method and device of a target object, electronic equipment, storage medium and program product, relates to the technical field of artificial intelligence, in particular to the technical field of machine vision, and can be applied to an automatic driving scene. The specific implementation scheme is as follows: determining a target object in an acquired roadside perception image through a target detection network; predicting three-dimensional attribute information of the target object through a three-dimensional attribute prediction network; determining a position determination method corresponding to the target object according to the three-dimensional attribute information; and determining physical position information of the target object according to the three-dimensional attribute information by using the position determination method. The present disclosure adopts different position determination methods to determine the physical position information of different target objects, thereby improving the accuracy of the determined physical position information.

A system having a system backend having a backend processor configured to define a patient clinical team associated with a patient clinical process comprising of a series of patient clinical tasks effected within the healthcare facility, each patient clinical task with an associated patient clinical task location within the healthcare facility, a group of mobile devices communicably connected via a network to the system backend, each mobile device being associated with and configured for use by at least one care member of the patient clinical team, a low energy (LE) beacon array arranged in a predetermined relationship with and differentiating between different patient clinical task locations, and communicably connected to each of the mobile devices that is configured to resolve its proximity, from the beacon array, relative to the at least one patient clinical task location. The backend processor is configured to dynamically enhance the patient clinical team by dynamically changing in real-time a predetermined characteristic of the patient clinical team based on the resolved proximity of each mobile device associated with each care member of the patient clinical team.

This specification relates to adapting machine-learning models to respond to location-based queries. According to a first aspect of this specification, there is described a computer implemented method comprising receiving, at a user device, a user query, and determining location information indicating a location related to the query. The method comprises retrieving one or more location profiles based on the location information. A location profile is associated with a respective geographic area and comprises a set of parameter modifications to parameters of a base machine-learning model. The method comprises applying the set of parameter modifications of the one or more retrieved location profiles to the base machine-learning model to obtain a modified machine-learning model and processing the user query using the modified machine-learning model to generate a response to the user query. The response to the user query is output via the user device.

A sensing processing method and a system. The method comprises: a sensing network element receives a first request, the first request being used for requesting to obtain a sensing result of a first area; the sensing network element obtains first information, the first information being used for determining a first position under sensing blocking in the first area; and the sensing network element obtains a first sensing result on the basis of the first position and the first request, the first sensing result not comprising a second sensing result corresponding to the first position. In the technical solution, when responding to a sensing request, a sensing result about a blocked object can be deleted on the basis of blocking setting information, and users can send the blocking setting information to a sensing processing network element according to actual blocking requirements, thereby improving the flexibility of sensing.

The application discloses a MES-oriented heterogeneous device networking fusion closed-loop management and control method and system, and relates to the technical field of intelligent manufacturing. The method comprises the following steps: resource identity recognition, location sensing and networking to generate networking basic data; a multi-protocol gateway collects heterogeneous data and converts the data into data packets containing identity, time and parameters; filtering and extraction, time window is determined according to identity, direct and boundary offset data are distinguished; boundary offset establishes a candidate in adjacent windows, and target attribution is obtained according to identity, sequence and state consistency; attribution is corrected and data is fused to generate process state data; multi-dimensional monitoring comparison is performed to generate abnormal identity and type level; instructions are generated according to type level and are sent to a board, MES and equipment, and results are updated and re-judged. Through unified access and standardization of heterogeneous resources, correction of process boundary data attribution and state fusion closed-loop verification, the MES-oriented networking fusion management and control are realized, and the accuracy and reliability of data correlation and closed-loop disposal are improved.

The present application relates to the field of laser measurement and calibration technology, and particularly relates to an intelligent deviation correction laser dotting and control system and a use method, wherein the lower computer comprises: a posture sensor arranged in a positioning assembly and used for detecting a placement posture of the positioning assembly; a position sensor used for detecting a relative position between a main body and the positioning assembly; a motion sensor arranged in the positioning assembly and used for detecting a displacement trajectory of the positioning assembly in a moving process; a control sensor arranged in the main body and used for monitoring a state of the main body; and a communication device used for data transmission; the upper computer comprises: an analysis unit used for receiving and analyzing posture information and position information; in the present scheme, through organic combination of posture recognition, position sensing, dynamic following and automatic control, intelligent automatic alignment and accurate deviation correction of a laser beam are realized, operation steps are greatly simplified, calibration efficiency and precision are improved, and the intelligent level and on-site adaptability of the equipment are enhanced.