13results about How to "Implement adaptive switching" patented technology

This invention belongs to the field of robot control, and particularly relates to a mobile robot path optimization system and method supporting track mode switching. The system uses an encoding module to integrate the mobile robot's operating scenario and real-time load status to generate a navigation task seed containing a path cost vector, navigation strategy identifier, and scenario transition point parameters. The central scheduler dynamically selects and calls the corresponding track navigation strategy sub-model or flat ground navigation strategy sub-model based on this navigation task seed. Combined with a spatiotemporal reservation table based on continuous value reservation weights to achieve multi-vehicle conflict avoidance and resource scheduling, executable trajectory control instructions are generated. This invention achieves load-adaptive path planning, efficient multi-vehicle collaboration, and smooth, seamless navigation switching between track and flat ground scenarios ensured by a switching prediction sub-model, significantly improving operational efficiency and system reliability in mixed and complex environments.

This invention discloses a fault handling method and apparatus based on main and distribution network coordination, belonging to the field of power system automation technology. The method includes: using a graph neural network to perform topology identification and fault location of the power grid, generating fault impact features; based on the spatial range and electrical connection relationships of the fault impact features, obtaining the switches at both ends of the faulty section of the power grid, and performing fault isolation through remote control tripping; analyzing the fault isolation information using topology impact analysis; and, combined with changes in the current topology of the power grid, generating a power grid protection strategy migration mechanism and initiating a fault recovery process to restore power supply. This invention achieves dynamic modeling of the power grid topology and accurate identification of faulty sections by using a graph neural network to embed the power grid graph and combining multi-channel time series information to generate node potential state vectors, thus improving the real-time performance and accuracy of fault location.

This invention discloses an automotive suspension assembly device, belonging to the field of automotive parts processing technology. It includes two bases, each with a support frame vertically mounted on its upper surface. Both support frames share a lifting mechanism. A vertical telescopic rod is vertically fixed to the side wall of one of the support frames. A horizontal telescopic rod is horizontally fixed to the output end of the vertical telescopic rod, and a partition is horizontally fixed to the output end of the horizontal telescopic rod. Two parallel first guide grooves are symmetrically formed on the top wall of the partition. During the lowering stage of the suspension parts, the V-shaped guide groove (wider at the top and narrower at the bottom) cooperates with the first ball bearings to create a "rolling friction" state between the edge of the heavy part and the slope, eliminating the risk of scratching the part surface and overcoming the technical pain point of traditional assembly requiring manual forceful pushing and pulling for positioning. Simultaneously, it achieves non-destructive and precise centering guidance of heavy parts towards the center baseline, improving positioning accuracy and operational efficiency in the initial assembly stage.

The invention discloses a cross-scene humanoid robot control method and related equipment, and the method comprises the steps: carrying out the analysis of obtained multi-modal perception data, and determining visual data; analyzing the visual data, and determining a target scene type; performing feature extraction on the multi-modal perception data, determining multi-modal feature data, and performing feature fusion according to a multi-modal feature model to obtain a global feature map; classifying according to the global feature map, and determining a semantic segmentation map; performing calculation according to the target scene type, the multi-modal sensing data and a preset weight set, and determining a degree-of-freedom weight set; performing hierarchical reinforcement learning according to the global feature map, the semantic segmentation map and the degree-of-freedom weight set to obtain a target joint control instruction; and controlling the robot according to the target joint control instruction. The embodiment of the invention can improve the control precision and the control stability. The method can be widely applied to the technical field of robot control.

The utility model discloses a high level bridge wind resistance noise reduction louver screen relates to high level bridge screen structure field, including steel structure frame, multiple rotatable windproof sound insulation flabellums are equipped with on steel structure frame along the height direction equidistance, two connecting rods are symmetrically equipped on multiple windproof sound insulation flabellums, two connecting rods are all hinged with multiple windproof sound insulation flabellums, and the top of steel structure frame is equipped with wind power drive arrangement, and wind power drive arrangement is connected with the top of two connecting rods all, and can drive connecting rod and move up, and the bottom of connecting rod is equipped with reset device, can utilize the angle regulation of wind power direct drive windproof sound insulation flabellum, realizes the automatic switching of low -speed noise reduction and high -speed wind resistance, and has reliability and environmental adaptability simultaneously, thereby solves the problem that existing fixed type wind screen needs manual regulation blade angle, and the effect of sealing noise reduction is not good, and the problem of response lag.

This invention provides a helium extraction device for separating flash vapor from liquefied natural gas, comprising a support platform; the top of the support platform is equipped with a membrane separation unit, a catalytic dehydrogenation unit, a water removal unit, an adsorption unit, and a low-temperature purification unit; the catalytic dehydrogenation unit is connected to the membrane separation unit and the water removal unit on both sides respectively; the water removal unit is connected to the adsorption unit through a first connecting pipe, and the adsorption unit is connected to the low-temperature purification unit through a second connecting pipe; the catalytic dehydrogenation unit is used to convert hydrogen in the flash vapor into water vapor; the water removal unit includes a cylinder fixed to the top of the support platform, a condenser cylinder is disposed inside the cylinder, and a scraper ring is fitted onto the condenser cylinder, with the scraper ring connected to a lifting mechanism; the scraper ring consists of a first scraper and a second scraper, with the first and second scraper rings connected to an adjustment component. The lifting mechanism and the openable scraper ring structure can scrape down water droplets and water films adhering to the inner wall of the condenser cylinder, solving the problems of incomplete drainage and water droplets adhering to the condenser cylinder.

The application discloses a kind of two-way audio signal arbitration and adaptive output system and method, belong to digital conference system audio processing technical field, the normal operation of double redundancy structure is guaranteed by data acquisition module, the state information that first audio source and second audio source emit is carried out data processing by arbitration decision module, and current output signal channel information fed back by state feedback module is received by arbitration decision module, the current audio source corresponding to the current output signal channel information of the audio source that needs to be output obtained is compared, audio output switching signal is obtained, to ensure that no audio output state appears, improve the accuracy and reliability of audio output;In addition, the audio signal corresponding to the audio stream data emitted by first audio source and second audio source is selectively output by using adaptive output module, to ensure that the output audio signal does not exist mixing state, ensures conference tone, and realizes audio signal adaptive switching.

The invention discloses a cluster networking method based on an unmanned aerial vehicle task networking database. The cluster networking method specifically comprises the steps of performing stage division of various types of tasks based on task scenes; constructing a basic task scene networking database based on historical experiences; based on the divided task stages, performing similarity search with task scenes in a networking database; and carrying out consistent networking strategy transfer on the cluster based on a networking database similarity search result. According to the method, historical task stages are classified and analyzed through machine learning based on the constructed task scene networking database, reusable experience is formed, a new task directly calls a verified optimization strategy through scene matching, and particularly, when a large-scale unmanned aerial vehicle cluster executes a dynamic multi-stage task, the task efficiency is greatly improved. The optimal networking strategy matched with the current task scene can be quickly, adaptively, consistently and reliably selected and switched, so that the overall collaborative efficiency and robustness of the cluster in a complex task environment are improved.

The invention discloses a photovoltaic array adaptive global maximum power point tracking method and system. The method comprises the following steps: solving global maximum power point reference voltage and global maximum power point reference power output by a photovoltaic array based on current environment data by adopting a two-stage nested multi-strategy particle swarm algorithm; judging whether a preset accurate control switching condition is met or not according to the output power of the photovoltaic array and the global maximum power point reference power; if a precise control switching condition is met, switching to a self-adaptive super-spiral sliding mode control mode, otherwise, maintaining a PI control mode; driving a power switch tube in the DC-DC converter to act by using a PI control mode or a self-adaptive super-spiral sliding mode control mode; according to the method, reference parameters are solved through the two-stage nested multi-strategy optimization particle swarm optimization algorithm, a dual-mode switching mechanism of PI control and self-adaptive super-spiral sliding mode control is matched, accurate and rapid tracking of the global maximum power point in a complex environment is achieved, and steady-state power fluctuation is effectively restrained.

The application provides a distortion and perceptual code rate adaptive variable code rate depth image compression method, encoder model construction is carried out, including dividing the image to be encoded into image blocks of different granularities, extracting hierarchical features, and obtaining non-repeated hierarchical features by filtering using an image mask; hierarchical vector quantization codebooks are used to quantize the hierarchical features, discrete index representations are generated, and code stream files are encoded; decoder model construction, including decoding the code stream files into discrete index representations, recovering the hierarchical features through the hierarchical vector quantization codebooks, fusing and decoding using a hybrid conditional decoder, and generating a reconstructed image; the encoder model, the decoder model and the hierarchical vector quantization codebook are jointly trained; the input image is encoded into a code stream file through the trained encoder model, and the code stream file is decoded into a reconstructed image through the trained decoder model. The application can realize fine bit rate control and adaptive switching between perceptual and distortion optimization.

This invention relates to the field of ground source heat pump drilling technology, and particularly to an auxiliary device for ground source heat pump drilling, comprising a rigid support unit, an elastic support unit, a conversion unit, and an adjustment unit. Through the coordinated operation of the rigid support unit, elastic support unit, conversion unit, and adjustment unit, this invention achieves adaptive switching of working state and contact area under varying well diameter. In the wide diameter section, the inner and outer sleeves rotate relative to each other, and the main and auxiliary support components jointly contact the well wall to form stable support, effectively avoiding excessive scraping of loose formations that could cause diameter expansion, while also reducing frictional resistance with the well wall. In the narrow diameter section, the conversion unit switches the inner and outer sleeves to a synchronous rotation state, and the adjustment unit drives the auxiliary support component to retract, leaving only the main support component in contact with the well wall and rotating to scrape it, significantly reducing frictional resistance while efficiently removing mud cake and eliminating the risk of blockage.

The invention belongs to the technical field of unmanned vehicle environment perception and obstacle avoidance, and discloses an unmanned vehicle transparent obstacle visual touch fusion perception and obstacle avoidance method, and the method comprises the steps: carrying out the transparent object edge detection of an image collected by a vehicle-mounted around-view camera through an improved Transform semantic segmentation network, and outputting an initial position coordinate and a credibility score; a tactile verification and enhancement module is built, flexible multi-mode tactile sensors fusing triboelectric sensing and visual tactile sensing are deployed on the edges of a bumper and a hub of the unmanned vehicle, and close tactile features are generated in the non-contact stage; when the visual credibility is relatively high, the visual data is dominated, the tactile data assists correction, and when the visual credibility is relatively low, a tactile signal dominant mode is triggered; and generating and executing an obstacle avoidance path. According to the invention, the detection precision of the unmanned vehicle on transparent obstacles in extreme scenes such as urban roads, indoor venues, rainy days and nights is obviously improved, the collision risk is reduced, and the unmanned vehicle is adapted to complex moving working condition requirements.

This invention discloses a metal gasket with a high-temperature protection mechanism and its processing method, relating to the field of sealing technology. It comprises a gasket body and cover plates covering the top and bottom of the gasket body. Interlocking protection components are arranged at equal angles along the circumference of the gasket body. Each interlocking protection component includes: a displacement drive element made of a unidirectional shape memory alloy; a guide slider, one end of which abuts against the displacement drive element, and the other end extending to the vicinity of the sealing surface of the gasket body; and a breakable protective plate, which is fixedly covered on the outer end face of the guide slider and flush with the overall sealing surface of the gasket. This metal gasket with a high-temperature protection mechanism and its processing method actively generate sealing protrusions under abnormal high-temperature conditions, dynamically compensating for sealing defects, effectively solving the problems of gasket creep relaxation and irreversible seal deterioration under high temperatures, and significantly improving sealing reliability under high-temperature conditions.