6824results about How to "Increase success rate" patented technology

The invention discloses an exhibition hall robot visual language navigation method based on a large model, belongs to the field of exhibition hall robots, and aims at solving the problems existing in an existing exhibition hall robot. According to the method, an exhibition hall robot multi-mode scene sensing system senses a current scene, and candidate target objects are obtained through fusion features of RGB images and laser point clouds; extracting image features of all candidate target objects in the scene; according to the exhibition hall robot multi-mode environment understanding system, an exhibition hall robot receives a voice instruction, a visual language multi-mode large model is used for understanding a target object and a user intention, then high-dimensional semantic features of a target object text are extracted from the system, and the target object is found out from candidate target objects; according to the exhibition hall robot intelligent navigation system based on the multi-modal large model, an exhibition hall robot carries out path planning and autonomous movement according to a pose sequence generated by an intention, and self behaviors are dynamically adjusted according to a real-time changing scene provided by the exhibition hall robot multi-modal scene sensing system.

The invention provides a space field game behavior dynamic evolution and strategy deduction optimization method and system, and relates to the technical field of space field game behavior optimization, and the method comprises the steps: analyzing historical behavior data through a deep neural network, constructing a dynamic relation graph of participants, inputting a multi-agent reinforcement learning model, and outputting a game behavior model. Then, a multi-layer game scene evolution tree is constructed, the instant income and the long-term risk are calculated in combination with a dual neural network, dynamic updating is performed by using a search algorithm of a hierarchical attention mechanism, and candidate evolution paths are generated; and finally, evaluating the candidate paths by using a strategy evaluation model constructed by a space-time diagram neural network, selecting an optimal strategy path, and generating a strategy deployment scheme implemented by stages in combination with a knowledge inference engine. According to the method, the evolution trend of the game behavior in the space field can be effectively predicted, a feasible strategy deployment scheme is provided, and the scientificity and effectiveness of strategy decision are improved.

The invention relates to the technical field of autonomous robot decision-making systems, in particular to an autonomous robot decision-making system and method based on multi-modal sensing fusion, and the system comprises a multi-modal data collection module which is used for obtaining multi-modal sensing data of an environment; the data preprocessing module is connected with the multi-modal data acquisition module and is used for preprocessing the multi-modal sensing data; the feature extraction module is connected with the data preprocessing module and used for extracting features from the preprocessed multi-modal sensing data; the dynamic space-time diagram construction module is connected with the feature extraction module and is used for constructing a dynamic space-time diagram based on the features; the multi-modal data fusion module is connected with the dynamic space-time diagram construction module and is used for fusing the multi-modal data in the dynamic space-time diagram; through adaptive time alignment and a cross-modal space attention mechanism, the system can effectively process space-time inconsistency of data of different sensors, and centimeter-level positioning precision is realized.

The invention relates to a multi-channel firmware upgrading method and system supporting breakpoint resume, and the method comprises the steps: segmenting a firmware file into a plurality of firmware data blocks, and generating a unique identifier and a checksum for each firmware data block; acquiring a network transmission channel, performing task matching with the network transmission channel according to the unique identifier to obtain a corresponding distribution transmission task, and establishing a corresponding initial record document; transmitting the firmware data block through the network transmission channel according to the distribution transmission task, monitoring the transmission state in real time, and carrying out breakpoint record updating on the initial record document to obtain a corresponding updated document; when it is detected that transmission is completed, verifying the firmware data block according to the checksum, and recombining the firmware data block to obtain the firmware file; and generating a corresponding upgrading report based on the updating document and the upgrading information. The upgrading strategy can be continuously optimized, and the system management efficiency can be improved.

The invention relates to the field of robot attitude control, in particular to a robot attitude control method and system based on intelligent recognition. The method comprises the following steps: continuously acquiring a real-time sandbox scene video stream based on a high-definition camera; and carrying out time sequence frame segmentation on the real-time sandbox scene video stream, and carrying out global contrast enhancement processing to obtain a global contrast optimization image sequence. Carrying out dynamic object identification on the global contrast optimization image sequence, and carrying out continuous frame optical flow motion tracking to generate a dynamic object optical flow change track; optimizing the image sequence according to the global contrast to extract scene ground and scene obstacle data; performing dynamic trajectory rendering on the scene ground and scene obstacle data based on the dynamic object optical flow change trajectory so as to construct a dynamic scene twinborn model; robot position posture parameters and user operation instructions are recognized. According to the invention, timely attitude adjustment response and accurate movement position control are realized.

The invention provides a multi-task monitoring and scheduling method and system, and relates to the technical field of task scheduling, and the method comprises the steps: receiving a task submission request, and constructing a task execution time prediction model based on a historical execution time sequence and a bidirectional gating recurrent neural network; monitoring a task execution process, and identifying an abnormal task by using a hierarchical Markov decision and a spectral clustering algorithm; for an abnormal task, constructing a task dependency graph based on the resource competition degree, calculating a critical path, determining an optimal calculation node, and performing re-execution; according to the method, the task execution time can be accurately predicted, abnormal tasks can be found in time, resource optimization scheduling is carried out, and the system operation efficiency is improved.

The invention discloses a multi-unmanned aerial vehicle maritime search and rescue path planning method based on multi-target particle swarm optimization. The method comprises the steps that S1, an area to be searched and rescued is converted into a two-dimensional search and rescue area; s2, performing grid decomposition on the two-dimensional search and rescue area; s3, obtaining a probability map through a Gaussian mixture model; s4, establishing an unmanned aerial vehicle search and rescue multi-target collaborative optimization function; s5, solving the function based on a hybrid particle swarm optimization algorithm to obtain an optimal solution; and generating a specific search and rescue path of the unmanned aerial vehicle according to the solved optimal solution. According to the method, an unmanned aerial vehicle search and rescue multi-target collaborative optimization function is established, the unmanned aerial vehicle is guided to execute the search and rescue task by taking resource balanced allocation, maximum path safety, preferential coverage of a high-risk area and minimum energy consumption as targets, task allocation, path planning and energy management are improved, multi-target optimization of multi-unmanned aerial vehicle search and rescue path planning is realized, and the search and rescue efficiency is improved. An accurate flight path is formulated for each unmanned aerial vehicle, comprehensive coverage of search and rescue tasks is ensured, and the search and rescue efficiency and success rate are remarkably improved.

The invention relates to the technical field of training strategy optimization, in particular to an optimization method and system based on a rehabilitation training strategy. The method comprises the following steps: carrying out real-time monitoring on physiology and training performance of a patient in a rehabilitation training process through wearable equipment, and carrying out evaluation analysis on an emotional state and athletic ability to obtain emotional feeling state data and athletic ability data of the patient; performing rehabilitation target prediction analysis according to the emotion feeling state data and the exercise ability data to obtain a prediction target of rehabilitation training of the patient; performing target feedback difference comparative analysis on the predicted target and an expected target to obtain a target difference feedback result; performing strategy optimization analysis on the current rehabilitation training strategy to generate a rehabilitation training strategy optimization scheme; and performing strategy dynamic optimization adjustment on the current rehabilitation training strategy to obtain an optimized rehabilitation training strategy. The rehabilitation training efficiency and effect of the patient can be improved.

The invention discloses a multi-underwater-robot cooperative control system and method based on predictive control, and relates to the technical field of cooperative control. During operation of the system, underwater environment data are collected through sensor equipment, an underwater three-dimensional dynamic environment model is generated by using a data fusion technology, and the underwater three-dimensional dynamic environment model is established based on task requirements; tasks are decomposed and distributed to all the robots, through an optimization algorithm, optimal paths of the robots and the mechanical arms are planned, the motion trails of the underwater robot mechanical arms are dynamically optimized based on model predictive control MPC, a real-time control strategy is provided for task division among the multiple mechanical arms, and the real-time control strategy of the underwater robot mechanical arms is provided based on environment modeling and the real-time state. Future motion states of the robot and the mechanical arm of the robot are predicted, an optimization objective function including task completion time, energy consumption and mechanical arm operation precision is constructed, a rolling optimization method is utilized, a control instruction is updated in real time, and real-time data sharing among multiple underwater robots is supported based on an acoustic communication technology.

The invention discloses a mechanical arm grabbing control method and system based on digital twinning, and relates to the technical field of industrial robot intelligent control, and the mechanical arm grabbing control method comprises the steps that target object data and grabbing scene data are collected, and a digital twinning model of a target object and a grabbing scene is constructed; dynamic changes of the target object and the grabbing scene are monitored in real time, the digital twin model is dynamically corrected, and a high-precision digital twin model is generated; based on a high-precision digital twinborn model, extracting characteristics of a target object, and identifying an optimal grabbing point; generating a grabbing strategy based on the optimal grabbing point and the high-precision digital twin model, and verifying the feasibility of the grabbing strategy; and according to a verification result, dynamically executing an intelligent grabbing strategy. According to the method, the grabbing precision and stability of the mechanical arm are improved by correcting the digital twin model in real time and recognizing the optimal grabbing point.

The invention provides an intelligent all-weather autonomous cruise unmanned rescue boat and a rescue control method thereof.The intelligent all-weather autonomous cruise unmanned rescue boat comprises a boat body and a multi-source energy supply system, an environment and target sensing system, an intelligent control system, a propelling system, a satellite positioning system and an autonomous navigation system which are arranged on the boat body. The rescue boat can autonomously acquire environment data and drowning person information, and dynamically plan an optimal cruise route through a deep reinforcement learning algorithm. The intelligent control system is responsible for controlling cruise, identifying drowners, generating rescue strategies and executing rescue tasks. After rescue is completed, the system can automatically re-plan a cruise route to continue the task. By means of the design, all-weather and full-autonomous water cruise and rescue are achieved, the rescue coverage range and efficiency are remarkably improved, manpower consumption and safety risks are reduced, the rescue response time is shortened, the rescue success rate is increased, and particularly the rescue capacity under the complex environment is remarkably improved.

The invention discloses an acousto-optic multi-view-field detection device and method for underwater defects of a hydraulic structure, and the method comprises the steps: obtaining original data, carrying out the filtering processing of the original data, and obtaining the filtered original data; based on the filtered original data, constructing a preprocessed multi-modal data tensor; performing spatial registration processing on the multi-modal data to obtain multi-modal data under a unified coordinate system; a self-adaptive tensor decomposition method is adopted for processing, and real-time position coordinates and movement track data of the robot are obtained; processing by adopting a multi-scale feature analysis method to obtain a target position coordinate and a target feature description; and processing by adopting a space-time optimization method and an uncertainty analysis method to obtain a final target positioning result and a confidence evaluation result. According to the method, the detection precision and the system robustness are improved, the detection accuracy is improved, the system positioning precision reaches the centimeter level, and the method is particularly suitable for intelligent defect detection in a complex hydraulic environment.

The invention relates to an activity scene arrangement scheme generation method and system, and the method comprises the following steps: obtaining activity scene data, and carrying out the analysis to obtain an activity theme and the characteristics of participants; objective condition parameters of the target site are obtained, and the candidate arrangement elements are screened; constructing a multi-dimensional attribute vector according to the screened arrangement elements; clustering the multi-dimensional attribute vectors, and performing spatial layout planning on layout elements; performing combined optimization on the generated spatial layout plan, color scheme and light effect setting; the time period of the activity is extracted, and the recommended arrangement scheme is adjusted according to the relevance; and performing visual rendering on the adjusted recommended arrangement scheme, and optimizing and determining an activity scene arrangement scheme. Intelligentization and individuation of activity scene arrangement are achieved, integrity, creativity and coordination of the arrangement effect are improved, meanwhile, influences of actual site conditions and activity time are considered, and a more optimized scene arrangement solution is provided for a user.

The invention relates to the technical field of space positioning and path planning, in particular to a three-dimensional space positioning and path planning method for an unmanned aerial vehicle tower. The method comprises the following steps of performing accurate longitude and latitude height data acquisition based on a laser range finder to generate accurate longitude and latitude height data; obtaining tower equipment parameters; performing inspection point three-dimensional coordinate calculation according to the tower equipment parameters and the tower three-dimensional coordinate system to generate inspection point three-dimensional coordinate data; cruise point comprehensive constraint range data are obtained according to the unmanned aerial vehicle flight association data and the polling area terrain data, and cruise point comprehensive constraint range data are generated; performing cruise node three-dimensional coordinate design on the three-dimensional coordinate data of the inspection point according to the cruise point comprehensive constraint range data to generate cruise point three-dimensional coordinate data; and performing optimized cruise path planning processing according to the cruise point three-dimensional coordinate data to generate optimized cruise path data. According to the invention, more accurate unmanned aerial vehicle tower space positioning and path planning are realized.

The invention discloses a robot visual identification, positioning and grabbing method based on RGBD point cloud, and the method comprises the following steps: 1, carrying out the camera point cloud data processing, synthesizing an RGB image and a depth image into a point cloud image, and carrying out the filtering and denoising processing of the point cloud data; step 2, target detection and capture attitude generation: performing target detection and instance segmentation on the RGB image, identifying a target object and generating a corresponding mask frame, and applying a mask to the depth image to extract depth point cloud data corresponding to the target object; obtaining the depth point cloud data, generating a plurality of groups of grabbing postures, and screening out an optimal grabbing posture; and thirdly, the mechanical arm grabbing execution module receives the optimal grabbing posture information output by the target detection and grabbing posture generation module and controls the robot to complete the grabbing action.

The invention discloses an unmanned ship navigation path intelligent optimization system and method based on ocean data, and particularly relates to the technical field of unmanned ship navigation. The accuracy of monitoring data of the sensor in a marine environment is continuously evaluated through a series of steps of preliminary calibration, data acquisition, filtering processing, abnormal marking, feature extraction and the like of the unmanned ship sensor, and real-time evaluation and time period division of the quality of the monitoring data are realized through comprehensive analysis of abnormal features of sound velocity and conductivity. The abnormal degree is predicted in the abnormal monitoring time period, and for the condition of high abnormal degree, a safety avoidance strategy is automatically set through a path planning system, and the route and the navigational speed of the unmanned ship are dynamically optimized, so that a risk area is avoided, the navigation precision and safety are improved, the reliability of sensor data is remarkably improved, and the safety of the unmanned ship is improved. The influence of data deviation on path planning is reduced, and the task success rate and the navigation safety of the unmanned ship in a complex marine environment are enhanced.

The invention discloses a mechanical arm control method, device and equipment and a storage medium, and relates to the technical field of intelligent control, and the mechanical arm control method comprises the steps that voice information and image information are obtained; the voice information and the image information are analyzed based on a large language model, and grabbing task target information of the mechanical arm is obtained; the grabbing pose of the mechanical arm is obtained according to the grabbing task target information; a mechanical arm is controlled to complete a grabbing task based on a mechanical arm grabbing pose controller, according to the method, external voice information and environment images are obtained through analysis of a large language model, matched targets in the voice information and the environment images are understood, and the grabbing task is automatically generated according to the positions of the matched targets in the images; the large language model is fully utilized to process the task input information, the task target is identified to generate the capture task information, the success rate of the capture task is improved, and the task scene is widened.

The invention relates to the technical field of medical instruments, and particularly discloses a high-precision automatic vein blood sampling system and method, and the system comprises an optimal puncture blood vessel decision module which determines an optimal puncture blood vessel data set; the dynamic ultrasonic data acquisition module is used for guiding an ultrasonic probe on a mechanical arm to perform skin fitting ultrasonic imaging based on the optimal puncture blood vessel coordinate under the camera coordinate system to obtain dynamic ultrasonic data of the optimal puncture area; the ultrasonic blood vessel data acquisition module performs enhancement and ROI extraction on blood vessel signals in the dynamic ultrasonic data; an ultrasonic blood vessel data analysis module extracts all potential blood vessel form distribution data sets from the time sequence blood vessel ROI candidate set, and performs puncture optimization degree and credibility analysis on all the potential blood vessel form distribution data sets to determine a final needle insertion point and a final puncture angle; the vein blood sampling puncture control module is used for controlling a puncture needle to perform puncture blood sampling at a final puncture position at a final puncture angle; the device is used for realizing high-precision automatic control of vein blood sampling.

The invention relates to the technical field of force feedback handles, in particular to a remote surgical robot control system based on a force feedback handle and a control method thereof. The method comprises the following steps: carrying out multi-dimensional force data acquisition on an end effector of the surgical robot, and carrying out contact force size, direction and time sequence feature extraction to obtain multi-dimensional force feature data; performing surgical operation safety baseline analysis according to the multi-dimensional force feature data to generate surgical operation safety baseline data; performing dynamic force feedback threshold calculation according to the surgical safety operation baseline data to obtain force feedback threshold range data; and performing adaptive compression processing on different force feedback signals according to the force feedback threshold range data to obtain dynamic force feedback control parameters. Through real-time control instruction optimization and a closed-loop feedback strategy, the system can quickly respond to an operation intention, early warning is carried out on potential risks, and it is ensured that the operation process is safer and more reliable.

The invention relates to the technical field of pipeline welding, in particular to a pipeline welding construction control method, system and equipment and a medium. Firstly, the material type, size parameters and environment temperature and humidity data of a to-be-welded pipeline are obtained, the optimal temperature curve and the welding ending moment in the welding process are predicted, then image information of a welding part is collected through a camera, the welding depth and welding width parameters are recognized, meanwhile, real-time temperature data are collected through a temperature sensor, and the welding precision is improved. Whether the welding temperature needs to be adjusted or not is judged by combining the predicted optimal temperature curve and the recognized fusion depth and fusion width parameters, if the welding temperature needs to be adjusted, a temperature adjustment value and the control power of the heating device are calculated, a corresponding temperature control instruction is generated and sent to the heating device to be executed, and automatic control and quality monitoring of the welding process are achieved. The dependence on artificial experience is reduced, and the welding efficiency and the success rate are improved.

The invention discloses a project execution risk assessment system and method based on data mining, and relates to the technical field of project execution risk assessment, and the method specifically comprises the following steps: obtaining comprehensive data information related to project execution, and storing the comprehensive data information after obtaining; analyzing the stored comprehensive data information, evaluating an accumulated risk of dependency of each task chain in a project execution process, classifying each task chain in the project according to an evaluation result, identifying high-risk, medium-risk and low-risk task chains, and generating a corresponding risk evaluation report; according to the risk assessment report, advising a manager to take corresponding risk mitigation measures for task chains of different risk levels; and updating comprehensive data information in real time, periodically recalculating an accumulated risk assessment coefficient, and dynamically adjusting a risk assessment result. According to the method, the problem of accumulative risk assessment of a complex dependent task chain is solved, and the effects of accurate classification management and dynamic risk adjustment are achieved.

The invention discloses a wireless communication network data secure transmission method and system, and relates to the technical field of secure transmission. During system operation, agricultural monitoring data are collected in real time through various sensor devices arranged in an agricultural environment, encryption processing is performed by using a symmetric encryption algorithm, and a dynamic key length adjustment mechanism is supported; a random number generator is integrated, unified management is performed on key generation, distribution, updating and destruction in a system, a timestamp or a random serial number is embedded into each data packet, the uniqueness of the timestamp or the serial number is verified at a receiving end, so that replay attacks are resisted, and the data are comprehensively analyzed and evaluated according to a set algorithm, so that the replay attack is prevented. The method comprises the following steps: acquiring a security data transmission performance index SDTP through calculation, generating a monitoring report, detecting abnormal conditions in a data transmission process, including channel quality reduction, data loss or verification failure, and providing a visual monitoring result.

A dynamic micro-service migration and route redirection method based on topology awareness comprises the following steps: S1, acquiring cloud computing data center and edge server resource information, an application request stream set and an initialized mirror image deployment matrix in a dynamic scene, and dividing time slot-based input request stream features; s2, in each time slot, a multi-mirror-image preliminary deployment scheme of the micro-service is obtained in a marginal gain optimal mode, and a node routing scheme is redirected according to a mirror image deployment matrix of a physical server node; and S3, according to the resource cost, the multi-mirror-image preliminary deployment scheme and the node routing scheme, screening out topology dependence pairs with abnormal bandwidths in a request stream, and based on topology perception, executing migration operation for micro-service mirror images on server nodes of links with abnormal bandwidths to obtain an optimal joint decision of micro-service deployment and request routing. According to the method, topology-aware micro-service mirror image migration and online redirection probability routing are realized in a dynamic cloud computing data center scene.

The invention relates to the technical field of intelligent water affair management, and discloses a method and system for dynamically adjusting the reporting moment of an NB water meter, and the system comprises NB water meter terminal equipment which is responsible for collecting water meter data in real time and uploading the data to a data management platform through an NB-IoT network; the data management platform is responsible for receiving, processing and storing data from the plurality of NB water meter terminal devices; the reporting moment adjusting module is used for dynamically adjusting the reporting moment of the NB water meter terminal equipment according to the strategy generated by the data management platform; the communication management module is responsible for communication between the NB water meter terminal equipment and the data management platform; the intelligent alarm and event processing module is responsible for monitoring the running state of the system in real time; and an equipment remote control module. By dynamically adjusting the reporting time, the system can reduce unnecessary frequent reporting, especially in a time period when the water consumption is relatively small or relatively stable, so that the power consumption of the equipment is remarkably reduced. Therefore, the battery life of the NB water meter terminal equipment can be prolonged, and the maintenance cost is reduced.

The screen display glass film tearing equipment comprises a screen display glass loading unit, a film tearing unit and an adhesive supply unit, the screen display glass loading unit comprises a carrying table and a moving mechanism, the carrying table is used for positioning screen display glass downwards from a to-be-torn film, the film tearing unit is located below the to-be-torn film and comprises a film tearing head and a power device, and the film tearing head comprises a film tearing seat, a clamping jaw and a roller; the power device comprises a first power piece and a second power piece. On one hand, on the basis of movement cooperation of the clamping jaw and the rolling wheel, aligned gradual stripping of the rolling position and the stripping position of the to-be-torn film is kept, and on the other hand, the overturning angle is increased based on follow-up overturning of the extending section, so that the success rate of forming the stripping angle by inward overturning of the to-be-torn film is greatly increased; the probability that the glue is separated from the to-be-torn film or the corner of the to-be-torn film is torn is effectively reduced; on the other hand, diagonal film tearing is formed based on the relative movement of the screen display glass and the clamping jaw, the risk that the protective film is torn or falls off is reduced, and the whole film is torn off more efficiently in a more labor-saving mode.

The invention relates to a protective film uncovering process based on corner lifting and rolling cooperation type screen display glass. The protective film uncovering process comprises the following steps: S1, corner lifting; s2, tearing off; and S3, collecting the film. On one hand, through cooperation of the film tearing clamping jaw and the rolling wheel, step-by-step peeling with the rolling position and the peeling position aligned is kept, meanwhile, the angle between the connecting section and the screen display glass is gradually increased on the basis that the connecting section is turned inwards under the condition that the adhesion section is not stressed, the success rate of forming an inwards-turned peeling angle is greatly increased, and the production efficiency is improved. The probability that the film-tearing adhesive tape is separated from the protective film or the corner of the protective film is torn is reduced; on the other hand, the complete tearing rate of the protective film is increased based on the relative movement of the glass and the film tearing clamping jaw, the tearing risk of the protective film is reduced, more labor-saving and faster diagonal tearing is achieved, meanwhile, based on pulling formed by the film collecting clamping jaw and the film tearing clamping jaw and airflow blowing cooperation formed by the film collecting clamping jaw and the film tearing clamping jaw, the completeness of film collecting is improved, and the production efficiency is improved. And the problem that the film tearing adhesive tape is adhered to the film tearing clamping jaw is solved, so that the continuity of peeling off the protective film is improved.

The invention belongs to the technical field of artificial intelligence and natural language processing, and discloses a persuasion skill-based automatic generation large model jailbreak prompt method. According to the method, multiple persuasion skills are combined with multiple persuasion skills such as logic persuasion, emotional persuasion and authoritative persuasion, so that the attack success rate is increased; and dynamically adjusting the attack strategy: adjusting the attack strategy in real time according to the attack progress and target model feedback, continuously optimizing the attack prompt, and adopting different jailbreak methods for different target large models. And finally generating a prompt which enables various large models to break prisons with high probability based on various persuasion skills and feedback of the target large model, and providing a new data set for security testing of the large models.

The invention discloses a recommendation agent system, a recommendation method and a recommendation device, and relates to the technical field of agents, a sensing module obtains first dialogue data and / or multi-modal environment data, and a memory module obtains historical man-machine interaction data related to user intention and a target object to be recommended; the planning module retrieves a reference workflow matched with the intention of the user, generates an initial agent workflow according to the data acquired by the sensing module, the historical man-machine interaction data, the target object to be recommended and the reference workflow, and calls the execution module to execute the initial agent workflow; judging whether the initial agent workflow is updated or not based on second dialogue data and / or return data generated in the execution process, if yes, updating the initial agent workflow, and executing the updated agent workflow from the updating position to complete recommendation. According to the method, the recommendation tasks can be dynamically planned in combination with multiple types of data, and the planning accuracy and recommendation accuracy are improved.

The invention belongs to the technical field of robot path planning, discloses an autonomous robot path planning method based on deep reinforcement learning, and aims at the problem that an agent is difficult to fully sense the environment in a complex dynamic environment, the State space of SAC is expanded and adjusted to enhance comprehensive sensing of an unknown environment and more flexible omnibearing obstacle avoidance. A heuristic reward function is designed, dynamic adjustment of reward values is achieved, and therefore the learning process of the robot is optimized. In order to solve the problems that the calculation complexity of an algorithm is increased and the convergence speed is reduced in a complex environment, the sample utilization rate is optimized in combination with a prior experience playback (PER) technology and a SumTree data structure. The weight of a high-value sample is increased through the PER technology, so that an intelligent agent can learn an effective strategy more quickly in the training process; and a SumTree data structure is introduced, so that the PER efficiency is further improved.

The invention relates to the technical field of sensor upgrading, in particular to a remote upgrading method and system for a tire pressure sensor. The method comprises the following steps: acquiring real-time state data of a tire pressure sensor; carrying out outlier parameter elimination on the real-time state data of the tire pressure sensor, and carrying out state diagnosis mining, so as to generate a state diagnosis report of the tire pressure sensor; performing upgrade demand prediction based on the tire pressure sensor state diagnosis report, and performing priority upgrade analysis to extract a plurality of upgradable sensors; performing byte-by-byte difference identification on the plurality of upgradable sensors based on preset latest version information, and performing incremental data encapsulation to generate a plurality of incremental upgrading data blocks; acquiring a historical running log of the vehicle; performing dynamic vehicle using behavior analysis and vehicle idle time period prediction on the vehicle historical operation log to generate a plurality of idle prediction time periods; the upgrading efficiency and the safety performance of the tire pressure sensor are improved.