12results about How to "Efficient perception" patented technology

The invention belongs to the technical field of dynamic traffic, and particularly discloses a multi-source data fusion traffic dynamic induction method, which comprises the following steps: S1, traffic information is acquired from multi-source heterogeneous data, and the data comprises real-time traffic information, navigation map data, a road network topological structure and a public traffic timetable; s2, carrying out data fusion processing on the traffic information obtained in the step S1, and generating fused traffic state estimation; s3, on the basis of travel historical records and preferences of the user, constructing a personalized travel guide model, and providing an optimal travel suggestion; and S4, monitoring the traffic condition in real time, performing dynamic adjustment based on user feedback and emergencies, and optimizing the travel path. By integrating real-time information from various data sources, comprehensive perception of road traffic conditions is ensured, traffic changes in space and time can be processed at the same time, the weight is dynamically adjusted according to the reliability of the data sources, and the influence of data islands and information loss is reduced.

The invention belongs to the technical field of wireless communication, and discloses a fluid antenna assisted communication, perception and calculation integrated (ISCC) method, which comprises the following steps of: unloading communication and perception tasks of a terminal to a mobile edge calculation (MEC) server at a base station end, and constructing an efficient ISCC system by utilizing the strong calculation capability of the server; a fluid antenna is integrated to a base station end, a fluid antenna assisted ISCC system is provided, and a fluid antenna assisted communication and perception model is established; the time delay of communication and sensing tasks is calculated, a time delay optimization problem is constructed, and the total time delay of the system is minimized by optimizing the position of a fluid antenna. According to the method, the flexibility of the fluid antenna is fully utilized, and the position of the antenna at the base station end is dynamically adjusted, so that the channel condition is adapted and optimized, and efficient and low-delay communication, sensing and calculation between the terminal and the base station are realized.

The invention discloses an underwater image quality enhancement method based on coupling degradation processing. The method comprises the following steps: acquiring an underwater image data set; constructing a combined degradation processing model, and inputting the underwater image data set into the combined degradation processing model for training to obtain a trained combined degradation processing model; and obtaining a to-be-detected underwater image, and inputting the to-be-detected underwater image into the trained joint degradation processing model for underwater image enhancement to obtain a target underwater quality enhanced image. According to the method, the problem that the image quality cannot be effectively recovered when a traditional image processing method faces a complex coupling degradation problem is solved, efficient and accurate perception of the underwater robot to the environment is achieved, and therefore the operation safety of the underwater robot is improved.

The invention discloses a self-powered flexible sensor and a preparation method thereof, and belongs to the field of electronic skin, man-machine interaction, flexible electronic devices and sensors. The invention relates to a cellulose acetate-based ionic gel self-energized flexible sensor which is prepared from cellulose acetate, ionic liquid, a polymer monomer, a cross-linking agent and an initiator and has good mechanical property and antibacterial property, the cellulose acetate enhances the mechanical property of the gel through non-covalent crosslinking, and the ionic liquid plays a conductive role. The flexible sensor has good conductivity, high sensitivity and stability, and can be stretched, twisted and bent. Meanwhile, the high transparency and excellent biocompatibility can be realized, so that the electronic skin can be used as the electronic skin for multifunctional perception and integrated application. The material is environment-friendly, the preparation method is simple, the operability is high, the application field is wide, and the prospect is wide.

The precise meal delivery robot based on the laser radar comprises a mounting table, connecting plates are fixedly mounted on the two sides of the mounting table, a driving wheel is rotationally mounted at one end of each connecting plate, a motor is fixedly mounted at the other end of each connecting plate, and an output shaft of each motor penetrates through the corresponding connecting plate to be fixedly mounted at one end of the corresponding driving wheel; a stabilizing frame is fixedly mounted on the upper surface of the mounting table, a linear module is fixedly mounted on the upper surface of the stabilizing frame, connecting arms are fixedly mounted on the two sides of a movable block of the linear module, and bearing mechanisms are slidably mounted on the outer surfaces of the connecting arms. Through the design of the linear module, the three-dimensional camera and the bearing mechanism, the food delivery tray can be moved to the corresponding accurate height position according to the arm position of a guest, the guest has better experience in the food taking process, the hanging type food delivery tray has the function of being conveniently taken down, and the service life of the food delivery tray is prolonged. And a convenient food taking function of the guest can be increased.

The present application discloses a remote sensing image clustering method and system, which relates to the field of image processing. The method includes: acquiring a remote sensing image; performing superpixel segmentation on the remote sensing image to obtain a superpixel map; obtaining superpixel features and an adjacency matrix based on the superpixel map; constructing an image clustering model, and inputting the superpixel features and the adjacency matrix into the image clustering model to obtain clustering results; and completing the classification of objects in the remote sensing image based on the clustering results. Among them, the image clustering model includes a cross-attention map diffusion module, a discriminative spectral feature extraction module, a map self-contrast loss function, and a clustering output module. The present application can well realize efficient neighbor perception, adaptive extraction of advanced spatial features, and effective identification of discriminative information, thereby realizing accurate clustering of remote sensing images.

The present invention discloses a near-zone anti-interference waveform design method and system for unmanned aerial vehicle-borne radar, and relates to the technical field of radar waveform design. The technical points of the present invention include: for unmanned aerial vehicle-borne radar systems, construct a set of pre-coded waveforms based on orthogonal frequency division multiplexing technology; construct waveform optimization constraints; construct waveform optimization objective functions and reference radar waveforms; use the MM algorithm to transform the waveform optimization objective function, and transform the non-convex constraints in the waveform optimization constraints into convex constraints; use standard numerical tools to iteratively solve the transformed waveform optimization objective function to obtain the optimal waveform sequence. The present invention avoids the problem that the common low peak-to-average ratio constraint fails to fully control the uniformity of the transmitting antenna power, and prevents the situation of high-power dynamic range fluctuations and low transmitting power efficiency. The present invention has significant improvements in symbol error rate and average achievable rate, and improves anti-interception performance, with a lower level of correlated side lobes.

The invention discloses a mantis shrimp eye imitating polarization flexible navigation system, and belongs to the technical field of underwater vehicle navigation. The system comprises a horizontal rotating base assembly, a support assembly, a spherical polarization sensing assembly and an eyestalk-imitating manifold protective cover. An underwater vehicle butt-joint base plate of the horizontal rotating base assembly is fixedly connected with the surface of the underwater vehicle, and a driving mechanism is integrated in the horizontal rotating base assembly; a vertical shaft of a Y-shaped bracket of the bracket assembly penetrates through a mounting hole in the upper surface of the horizontal rotating base assembly and is axially limited; bearing seats are arranged at the tail ends of two horizontal arms at the top of the Y-shaped bracket, and the bearing seats are matched with the spherical polarization sensing component to form rotatable mechanical connection, so that the spherical polarization sensing component is endowed with pitching motion capability in a vertical plane; the whole eyestalk-like manifold protective cover wraps the support assembly and the spherical polarization sensing assembly and is fixedly connected with the horizontal rotating base assembly. According to the invention, efficient sensing and rapid tracking of the underwater polarized light field are realized.

This invention discloses a multi-task teacher-student network collaborative training method for underwater scattering environments, aimed at improving the scene perception and understanding capabilities of autonomous underwater vehicles. The invention trains high-performance teacher models using datasets with single depth labels and single semantic labels, respectively, and designs a lightweight multi-task student model framework. Through feature extraction and fusion, the parameter size of the model is reduced and inference speed is improved. Through the design of the loss function in the knowledge distillation process, it is ensured that the student model can inherit the depth perception and semantic understanding capabilities of the teacher model. This invention can be widely applied in fields such as marine resource exploration, underwater infrastructure maintenance, and environmental monitoring, providing accurate environmental understanding and target recognition capabilities for underwater work, and has significant application prospects and economic benefits.

The invention provides an automatic driving vehicle dynamic pick-up management system based on a multi-modal large model, and belongs to the technical field of automatic driving. A multi-modal large model processing module; a personalized pick-up point recommendation module; a dynamic path planning and pick-up point adjusting module; an intelligent passenger interaction and feedback module; the system aims at providing intelligent pick-up and drop-off point selection and dynamic adjustment capability for an automatic driving vehicle. According to the system, efficient perception and real-time response to a complex environment are realized by fusing multi-modal data, and an optimal path and a pick-up point are generated in real time; the system can automatically adjust pick-up points and path planning according to real-time traffic, environment changes and passenger demands, and provides personalized passenger experience.

The embodiment of the invention provides a content recommendation method and device, a server side and a storage medium, and relates to the technical field of data processing, the method is applied to the server side, and the method comprises the steps of obtaining a first position of a client side in response to a content recommendation request sent by the client side; determining a first geographic area where the client is located according to the first position; determining a target activity circle with an intersection between the located geographic area and the first geographic area; content to be recommended to the client is recalled from the content associated with the target activity circle; and recommending the recalled content to the client. By applying the technical scheme provided by the embodiment of the invention, the content conforming to the interest and preference of the user can be recommended to the user, so that the satisfaction of the user is improved.

The invention relates to the technical field of blind person intelligent navigation, and particularly discloses a blind person intelligent navigation system based on image understanding, which comprises an intelligent navigation suit and an intelligent blind person walking stick, the intelligent navigation suit comprises a protective suit unit, a first sensor unit, a camera unit, a first data transmitter unit and an earphone unit, the blind person walking stick comprises an intelligent walking stick unit, a second sensor unit and a second data transmitter unit, and the intelligent walking stick unit obtains surrounding environment information through the second sensor unit arranged in the intelligent walking stick unit; the obstacle avoidance navigation system of the intelligent navigation suit and the intelligent blind person walking stick provides the functions of sensing the surrounding environment in real time and giving out warning for the blind person, the blind person is helped to be prevented from colliding with obstacles, the navigation systems are arranged in the intelligent navigation suit and the intelligent blind person walking stick, route planning and navigation guidance are provided, and the navigation effect is good. And the system has the functions of article identification, audio prompt, data recording and the like.