Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

28results about How to "Improve navigation performance" patented technology

Robot attitude control method and system based on intelligent identification

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.
Owner:SHENZHEN GJS TECHNOLOGY CO LTD

Robot control instruction analysis system based on multi-algorithm fusion

The invention relates to the technical field of robot control instruction analysis, in particular to a robot control instruction analysis system based on multi-algorithm fusion, which comprises a data collection module, an environment map analysis module, a global path analysis module and a moving trajectory analysis module, wherein the data collection module collects state data of a robot, surrounding environment data of the robot and a target position; the environment map analysis module determines an environment occupation grid map according to the environment data; the global path analysis module determines a global path according to an environment occupation grid map, a target position, self state data and robot motion constraint parameters; the running track analysis module determines a smooth track according to the global path and the state data of the running track analysis module and detects an environment occupation grid map, if a dynamic obstacle avoidance object exists in the environment occupation grid map, the environment occupation grid map is updated, the smooth track is modified according to the environment occupation grid map updated in real time, and the dynamic obstacle avoidance object is obtained. And the final robot moving trajectory is obtained.
Owner:XIAMEN XIANGYU TECHNOLOGY CO LTD

Multi-mode navigation control method and system of electric wheelchair controller

The invention relates to the technical field of electric wheelchair control, in particular to a multi-mode navigation control method and system of an electric wheelchair controller. The method comprises the following steps: acquiring a real-time environment monitoring video based on an electric wheelchair high-definition camera; video jitter distortion compensation and dynamic object frame-by-frame continuous optical flow tracking are carried out on the real-time environment monitoring video, and a moving track of each dynamic object is generated; performing environment static obstacle accurate identification on the real-time environment monitoring video, and performing dynamic environment perception modeling based on a moving track of each dynamic object so as to generate a dynamic environment map; a navigation mode switching decision is made based on a real-time wheelchair operation instruction of the user to obtain a current navigation mode of the user, and the current navigation mode of the user comprises a rocker control mode and an intelligent navigation mode. According to the invention, efficient and intelligent multi-mode control of the electric wheelchair is realized, so that the electric wheelchair can more flexibly and safely cope with a complex environment.
Owner:GUANGDONG K SILVER IND CO LTD

Optical system, projection display device, and transport vehicle

The utility model relates to an optical system, a projection display device and a transport carrier. The optical system is applied to the projection display device with an image source module. The optical system comprises a light splitting module and a beam expanding module. The light splitting module comprises a first light splitting unit and a second light splitting unit which are arranged in parallel. The image source light beam is reflected to the first light splitting unit after being expanded by the beam expanding module; the first light splitting unit transmits part of the image source light beam to form a first virtual image and reflects part of the image source light beam to form a first reflected light beam, and the first reflected light beam can be conducted to the second light splitting unit; the second light splitting unit reflects the first reflected light beam and forms a second virtual image, and the imaging distances of the first virtual image and the second virtual image are different. The optical system in the embodiment can simultaneously form at least two virtual images with different imaging distances, so that the at least two virtual images can be presented by using a relatively simple structure, and the presented image layers are relatively abundant.
Owner:GUANGZHOU AUTOMOBILE GROUP CO LTD

Novel automatic loading and unloading system of unmanned forklift in van environment

The invention relates to the technical field of automatic loading and unloading of logistics forklifts, in particular to a novel automatic loading and unloading system of an unmanned forklift in a van environment, and the system comprises a positioning map binding module which is used for binding platform areas where a plurality of trucks are parked on a positioning map; and the task issuing module is configured to receive a task issued by a driver through a pad system when a truck arrives and transmit the task to a warehouse management system (WMS), and the WMS issues a cargo taking and placing task of a corresponding map point position according to a platform number. And the visual servo system adopting the solid-state laser radar provides high-precision and high-stability navigation capability for the AGV.
Owner:MULTIWAY ROBOTICS TECH (SHENZHEN) CO LTD

Valve leaflet cutting device and system

The invention relates to a valve leaflet cutting device and system. The valve leaflet cutting device at least comprises a far-end cutting part, and the far-end cutting part comprises a main body piece, a tool bit, a clamping arm and a moving piece; the tool bit is arranged in the main body part and can axially move between the far end and the near end of the main body part, and the tool bit is provided with double-side cutting edges used for cutting valve leaflets; the near end of the clamping arm is rotatably connected with the main body piece, and the far end is used for clamping and fixing a valve leaflet; the moving part is in sliding connection with the clamping arm, can axially move relative to the main body part and is used for driving the clamping arm to rotate around the main body part to realize opening and closing of the clamping arm; when the clamping arms are closed and the valve leaflet is fixed, the fixed valve leaflet is cut through the axial movement of the tool bit.
Owner:CHENGDU JINJIAO MEDTECH CO LTD

Unmanned salvage ship capable of recognizing images of tiny foreign matters on water surface and image capturing and processing method of unmanned salvage ship

The embodiment of the invention provides a salvage unmanned ship for water surface tiny foreign matter image recognition and an image capturing and processing method thereof, and belongs to the technical field of water surface cleaning and water quality monitoring. The salvage unmanned ship comprises a ship body; the image acquisition and processing module is used for capturing an image of the water surface and identifying tiny foreign matters through a target detection algorithm; the garbage collecting device is mounted on the ship body and can collect tiny foreign matters on the water surface according to an identification result so as to clean garbage floating on the water surface; wherein the target detection algorithm is a Cascade R-CNN target detection algorithm, and the Cascade R-CNN target detection algorithm is combined with a feature pyramid network to process a multi-scale target through feature extraction from bottom to top and feature fusion from top to bottom. The Cascade R-CNN target detection algorithm is combined with the feature pyramid network to effectively solve the problems of low water surface garbage cleaning efficiency, difficulty in small foreign matter detection and the like in the prior art.
Owner:NANTONG UNIV

Robot navigation method based on multi-mode inverse reinforcement learning

The invention provides a robot navigation method based on multi-mode inverse reinforcement learning, which comprises the following steps: acquiring robot pose data, acquiring laser radar data, and generating a grid map; the laser radar data, the raster map and the robot pose information generate fusion observation representation through a multi-modal fusion network; initializing network parameters, taking the fusion observation representation as expert demonstration track data, and pre-training a generator and a discriminator in a generative adversarial inverse reinforcement learning framework through an expert demonstration track to obtain a pre-training model; standardizing a reward function through expert trajectory data; and constructing a generative adversarial reinforcement learning overall framework, training to obtain an intelligent control model, and deploying the model to realize intelligent control of the robot. According to the method, multi-modal data fusion and generative adversarial reinforcement learning can be realized, and the performance of the robot in the aspect of navigation is effectively improved.
Owner:NORTHWESTERN POLYTECHNICAL UNIV

Map main direction detection method and system of robot and application

The invention provides a map main direction detection method and system of a robot and application. The method comprises the steps of obtaining a probability grid and converting the probability grid into a grayscale image; extracting all contours in the grayscale image, and screening the contours according to the area to obtain a contour set; polygonal approximation is carried out on the contour; calculating to obtain a first main direction based on the side length and angle of each side of the polygon, and calculating to obtain a second main direction based on the minimum enclosing rectangle of the polygon; and integrating the first main direction and the second main direction to obtain a main direction deviation angle of the robot. The probability grid is used for predicting the main direction of the robot, and the first main direction based on the side length and angle of the polygon and the second main direction based on the minimum enclosing rectangle are integrated, so that the accuracy and robustness of direction identification can be effectively improved, the calculation efficiency is optimized, the positioning precision is improved, and the accuracy of direction identification is improved. And the navigation capability and adaptability of the robot in a complex environment are improved.
Owner:PARTICLE EVOLUTION ROBOT (ZHUHAI) CO LTD

Method for determining auxiliary identification of laparoscope

The invention relates to the field of medical imaging, and discloses a method for determining a laparoscope auxiliary identifier. The method comprises the following steps: determining a fluorescence labeling area according to a first image; determining an identifiable area according to the second image and the fluorescence labeling area; generating a differential surface associated with the fluorescence intensity interval based on the identifiable region; determining an auxiliary identifier based on the difference surface; the first image is obtained based on near-infrared fluorescence intensity, a green image signal of a pixel in the first image is in positive correlation with the fluorescence intensity, when the fluorescence intensity is lower than a first fluorescence intensity threshold value, the green image signal of the pixel is 0, and when the fluorescence intensity is higher than the first fluorescence intensity threshold value, the value of the green image signal of the pixel is 0-255; the values of the red image signal and the blue image signal are 0; the second image is a visible light image; the first image and the second image are collected through a laparoscope, and the collection positions are consistent. According to the method, the image navigation capability in the laparoscopic surgery is improved, and the accuracy of tissue identification is enhanced.
Owner:THE THIRD AFFILIATED HOSPITAL OF PLA NAVAL MEDICAL UNIVERSITY

Laser forklift pickup path re-planning and self-adaptive optimization method thereof

The invention relates to the technical field of path planning, in particular to a laser forklift goods taking path re-planning and self-adaptive optimization method. The method comprises the following steps: setting all parameter values required in calculation; the cargo tray position and the inclination angle recognized by the camera are obtained; according to the position and angle information of the tray, a third-order Bezier curve type goods taking route with the current position of the AGV as a starting point is planned; and according to the length and width of the fork of the forklift and the width of the tray, the planned route is adaptively optimized and adjusted, so that the forklift does not collide with the tray when running on the route. By comprehensively considering the current position and orientation of the AGV and the conversion relation among multiple coordinate systems, an accurate third-order Bezier curve type goods taking route is planned, adaptive optimization of the route is carried out, and the distance between control points is flexibly adjusted so as to balance the collision risk and the route curvature change, so that the accuracy of goods taking is improved. The safe distance between the forklift and the tray in the driving process is guaranteed, and collision is avoided.
Owner:临沂临工智能信息科技有限公司

Navigation processing method, device and electronic equipment

The present disclosure provides a navigation processing method, device and electronic device, which relate to the field of data processing technology, and specifically to the field of autonomous driving technology. The specific implementation scheme is: obtaining N first links located at a target intersection in a first map, the target intersection is an intersection in a first navigation route of a driving object, and the first navigation route is a navigation route planned on a second map; when the driving object travels to a first target position based on the first navigation route, based on the starting node of the second link, a target link matching the second link is screened out from the N first links, the second link is a link located at the target intersection in the first navigation route and has the least mileage from the driving object, and the distance of the first target position relative to the starting node of the second link is within a preset range; when the driving object travels to a second target position based on the first navigation route, the driving object is navigated based on the target link.
Owner:APOLLO INTELLIGENT CONNECTIVITY (BEIJING) TECH CO LTD

A pseudo-observation-based heterogeneous flight cluster inertial navigation error cooperative correction method

ActiveCN116086494BSolve the problem of inertial navigation positioning error divergenceImprove navigation performanceMeasurement devicesObservation dataClassical mechanics
The application discloses a heterogeneous flight cluster inertial navigation error cooperative correction method based on pseudo observation, which is composed of multiple unmanned aerial vehicles (UAVs) and the UAVs observe each other, and comprises the following steps: S1, acquiring inertial navigation positions of the UAVs at a current time and inertial navigation position estimation noises; S2, selecting a reference anchor point, and obtaining relative distance observation, relative azimuth observation and relative pitch angle observation among the UAVs in the cluster; S3, establishing an observation equation of the heterogeneous cluster UAVs according to the observation data obtained in the step S2; S4, constructing a pseudo observation equation according to whether the observation equation is sufficient; S5, classifying according to the rank of an observation matrix, and solving the observation equation; and S6, updating the inertial navigation position, attitude and speed of each UAV. The inertial navigation position information after correction is combined with the inertial navigation information to correct and update the navigation position, attitude and speed, and better overall positioning precision of the cluster can be obtained when the available observation quantity is insufficient.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS

Real-time imaging dynamic navigation system for breast tumor interventional operation

The invention discloses a real-time imaging dynamic navigation system for a breast tumor interventional operation, which comprises an acousto-optic scanning device, a Redis client, a Redis server, a database, a modeling platform, a Three.js dynamic driving engine and a display terminal, and is characterized in that the acousto-optic scanning device is arranged above a breast tumor and performs scanning and data acquisition in a non-contact manner during the operation; the Redis client, the Redis server and the database respectively acquire, distribute and store the acquired data, and the modeling platform performs model construction on the received data. A Three.js dynamic driving engine composed of a breast tumor data input end, an initial data driving unit, an incremental data driving unit, a rendering unit and a dynamic image output end carries out data processing on the model in a data driving mode. According to the dynamic navigation system, real-time data acquisition, transmission and data processing are realized, the generated navigation image is visual and dynamic, high-quality image guidance is provided for an operation, and the success rate and accuracy of the operation are improved.
Owner:THE SECOND AFFILIATED HOSPITAL ARMY MEDICAL UNIV

Indoor robot sensing method based on point cloud multi-sensor fusion mapping

The invention discloses an indoor robot sensing method based on point cloud multi-sensor fusion mapping, and relates to an indoor intelligent robot sensing method, which comprises the following steps of: obtaining multi-frame point cloud images of surrounding environment and obstacles by using a binocular camera and a laser radar; processing and fusing the three-dimensional point cloud images respectively acquired by the two three-dimensional point cloud images; in order to improve the robustness of perception, the three-dimensional point cloud is preprocessed before fusion, and interference of point cloud noise on subsequent matching is avoided; based on the target point cloud data corresponding to each frame of image, the point cloud map is accurately registered, so that the intelligent robot can specifically understand the external environment. According to the method, fusion mapping is carried out based on the point cloud data obtained by different sensors, the environment information of the intelligent equipment is determined, the wrong point cloud determined according to the depth corresponding to each point cloud is filtered out from the target point cloud data, and the mapping effect is optimized by filtering out the wrong point cloud influencing the accuracy of the volume measurement result, so that the accuracy of the volume measurement result is improved. The mapping error is obviously reduced, the sensing precision of the robot is improved, and the robustness of the system is improved.
Owner:SHENYANG INSTITUTE OF CHEMICAL TECHNOLOGY

A visual language navigation method integrating grid map and topological map

ActiveCN119164385BImprove navigation performanceImprove utilizationNavigational calculation instruments
The present invention discloses a visual language navigation method integrating a grid map and a topological map, belonging to the field of navigation technology. The method receives RGB and depth information through a waypoint predictor and linearly fuses them into RGBD information, thereby generating useful reachable nodes; in short-term reasoning, a grid map is used to model local historical observations during navigation, and time and space information is used to describe the environment of local visits; in long-term reasoning, a topological map is used to spontaneously update the topological map through the waypoints generated by the waypoint predictor, and the waypoints are effectively added to the dynamic topological map; then, a path is planned across modalities through visual and semantic information, and long-term and short-term navigation scores are dynamically integrated to make navigation actions. The present invention improves the accuracy of navigation.
Owner:XIANGTAN UNIV

Water surface three-dimensional collaborative navigation method and system and electronic equipment

The invention relates to the technical field of water surface three-dimensional collaborative navigation scheme design, in particular to a water surface three-dimensional collaborative navigation method and system and electronic equipment. The underwater electromagnetic wave communication frequency automatic modulation model is trained, and after model training is completed and submarine position data of a target submarine are obtained through a sonar detector arranged on a water surface platform, the distance between the submarine position data and GPS position information of an underwater subsurface buoy is calculated; and inputting the distance into an underwater electromagnetic wave communication frequency automatic modulation model, and controlling the underwater subsurface buoy to communicate with the target submarine through electromagnetic waves according to a predicted electromagnetic wave frequency output by the model. The underwater electromagnetic wave communication frequency is automatically modulated, the technical problems that the influence on signal transmission is also increased due to the increase of the transmission distance, and the transmission frequency is difficult to be reasonably modulated in a threshold value calculation mode are effectively solved, and the positioning and navigation effects of the navigation system of the submarine are further improved to a great extent.
Owner:QINGDAO AUTOMATIC RES INST

UAV visual navigation method and device based on deep learning

The present invention relates to a UAV visual navigation method and device based on deep learning, including: obtaining a visual image collected by a UAV visual system, extracting features from the visual image, and obtaining an image feature representation; performing type analysis on the image feature representation to obtain scene data, depth information, and continuous image frames; performing scene recognition on the scene data to obtain recognition results of obstacles and passable areas; performing optical flow estimation on the continuous image frames and the depth information through a preset optical flow algorithm to obtain a pose estimation result; obtaining the mission requirements of the UAV to obtain a navigation control strategy with environmental adaptability; optimizing the adaptive hybrid learning model based on the feedback information combined with a preset attention mechanism and reward shaping conditions to obtain the optimized adaptive hybrid learning model. The present invention can improve the adaptability of UAVs in complex and dynamic environments.
Owner:SHENZHEN TIANYING BROTHERS UAV INNOVATION CO LTD

A spatio-temporal graph object navigation method based on target correlation

PendingCN122730023AAchieve integrationImprove navigation decision-making capabilities
This invention belongs to the field of intelligent robotics and visual navigation technology, and discloses a spatiotemporal graph-based object navigation method based on target correlation. It constructs a target-related object relationship graph and combines local relationship modeling with a global serialized information propagation mechanism to effectively model object relationships in the environment. Simultaneously, by introducing target correlation to control the information propagation order, node features can be updated progressively according to target correlation, thereby enhancing the modeling ability of long-distance dependencies. Combining temporal feature modeling with action decision-making processes, it achieves efficient robot navigation in complex environments. This invention can utilize the relationship information between objects in target object navigation tasks to improve target orientation capabilities; enhance long-distance dependency modeling capabilities through a serialized information propagation mechanism to improve navigation success rate; and exhibit stronger decision-making stability and environmental adaptability in complex environments and long-path tasks, thus improving overall navigation performance.
Owner:WUXI UNIV

Binocular vision navigation method for unmanned vehicles based on improved artificial potential field method

The invention belongs to the field of artificial intelligence technology, and specifically relates to a binocular vision navigation method for an unmanned vehicle based on an improved artificial potential field method. Firstly, a semi-global stereo matching algorithm in binocular vision is used to realize real-time perception of depth information of a scene, an FPGA implementation framework of the semi-global stereo matching algorithm based on block comparison is designed, the semi-global stereo matching algorithm is modularized, parallel operation is adopted within the module, and pipeline operation is adopted between modules to accelerate it, thereby reducing resource usage and improving the real-time performance of the stereo matching algorithm. Then, binarization, median filtering, connected domain detection and other operations are performed on the current scene, and the obstacle distance of the current scene is output in combination with a disparity map. Finally, in view of the local minimum value problem in the traditional artificial potential field method, the invention effectively solves the local minimum value problem by autonomously establishing a virtual target point, and realizes the path planning of the unmanned vehicle. The method of the invention is verified on a built unmanned vehicle system, and experimental results show the feasibility and superiority of the method of the invention.
Owner:XIAN MODERN CONTROL TECH RES INST

Mobile robot navigation method and system based on rotary imaging radar and satellite positioning

The invention provides a mobile robot navigation method and system based on a rotary imaging radar and satellite positioning, and relates to the technical field of mobile robot navigation. Synchronously acquiring surrounding environment three-dimensional point cloud data, satellite positioning data and inertial measurement data; feature information is extracted and integrated to generate positioning attitude data; distinguishing static and dynamic characteristics according to the three-dimensional point cloud data, and generating a dynamic local environment map; when the satellite positioning signal fails, dead reckoning is carried out by combining the three-dimensional point cloud data, the inertial measurement data and the positioning attitude data before failure, and the positioning attitude data is updated; and combining the new positioning attitude data and the dynamic local environment map with the target driving data to generate an initial navigation control instruction, continuously collecting a data dynamic adjustment instruction, and controlling the robot to complete navigation. The method can adapt to a complex dynamic environment and realize high-precision reliable navigation.
Owner:SHANGHAI ENZUO TECH CO LTD

A navigation wheel angle controller for agricultural machinery based on model predictive optimization

The present invention relates to an agricultural machinery navigation wheel angle controller based on model prediction optimization, and the control method executed by the controller includes: step 1, the agricultural machinery navigation wheel angle controller obtains the position information and motion parameters of the agricultural machinery in real time through the satellite navigation positioning system and attitude sensor installed on the agricultural machinery, and the path tracking control algorithm in the navigation controller solves the expected front wheel steering angle; step 2, the expected wheel angle is input into the pre-built agricultural machinery steering system model prediction to obtain the compensation wheel angle; step 3, based on the expected wheel angle and the compensation wheel angle, the optimized navigation wheel angle is calculated. The controller is mainly used in the electronically controlled steering system of the agricultural machinery navigation system, and the agricultural machinery can be driven stably and accurately in a complex farmland environment by compensating and controlling the steering wheel angle of the agricultural machinery, thereby improving the automation level and operation efficiency of agricultural production.
Owner:HAINAN UNIV

Illumination condition adaptive soil humidity estimation method based on multi-modal data fusion

The invention relates to the field of robot terrain perception, and provides an illumination condition adaptive soil humidity estimation method based on multi-modal data fusion. The method comprises the following steps: firstly, providing an optimal exposure time calculation method for constructing a multi-modal soil humidity data set; secondly, soil image shadows are removed through an SS-GAN algorithm, a channel attention mechanism ECA and a standard convolutional layer are improved in combination with environmental information, ResNet-50 is embedded as a network backbone, and soil humidity estimation sub-models adapted to different illumination conditions are designed; then, training each sub-model by using an improved loss function to enable the sub-model to adapt to a specific illumination condition; finally, a sub-model fusion and switching method is provided, and a soil humidity estimation map is generated through gray value mapping. According to the invention, multi-modal data fusion is carried out through the improved neural network model, and a fusion and switching algorithm of the sub-models is designed, so that the soil humidity estimation precision of the robot under a complex illumination condition can be improved, and thus the terrain perception capability is improved.
Owner:HARBIN UNIV OF SCI & TECH

A polarization smearing error estimation method based on meta-learning in a high dynamic environment

The application provides a polarization tailing error estimation method based on meta-learning in a high dynamic environment, and belongs to the field of bionic polarization navigation, and comprises the following steps: a polarization angle error database caused by tailing interference under different rotating speeds is established; in an offline stage, a rotating speed related network and a rotating speed independent network are trained based on a meta-learning strategy, polarization angle errors are decoupled into rotating speed related features and rotating speed independent features; in an online stage, when performing linear motion, a polarization / inertial navigation correction dynamic model three-dimensional attitude error is used; in turning motion, the rotating speed related features are taken as states, the polarization angle inverted by a carrier dynamic model is taken as a measurement, the rotating speed independent features are taken as a measurement matrix, the rotating speed related features are estimated, the rotating speed related features and the rotating speed independent features are fused, and the estimation of the polarization tailing error is completed. The application combines offline and online learning to complete adaptive estimation and compensation of the polarization tailing error under different rotating speeds in a high dynamic environment, and improves the navigation performance of a bionic polarization integrated navigation system.
Owner:BEIHANG UNIV

A grid mapping optimization method and system based on adaptive control factors

The application discloses a grid mapping optimization method and system based on an adaptive control factor, and belongs to the technical field of artificial intelligence.The application comprises a semantic map establishing module, a point cloud map establishing module, a semantic database, a map gridding module and a post-rendering module.The application carries out image preprocessing after camera image data is collected, and establishes a semantic map; carries out filtering after laser radar point cloud data is collected, and establishes a point cloud map; according to the data information of the corresponding importance score Score, the distance Dis from the central observation position and the multi-material composition Mat in the semantic database in the semantic model for establishing the point cloud map, the adaptive control factor delta of each individual model in the point cloud map is determined; according to the determined adaptive control factor, the connection between spatial points is realized in the map gridding module, and a gridded map is formed; according to pixel normal interpolation calculation of a light source, the map is restored and reconstructed through a rendering module, and the authenticity and integrity of the reconstructed map are improved.
Owner:BEIJING CHANGCHENG INST OF METROLOGY & MEASUREMENT AVIATION IND CORP OF CHINA

A mobile robot navigation method based on deep reinforcement learning

The application provides a kind of mobile robot navigation method based on deep reinforcement learning, realizes guiding to accelerate convergence before the training of deep reinforcement learning strategy, so that mobile robot can complete safe and efficient autonomous navigation only relying on on-board sensor in complex dynamic environment.The application comprises the following steps: S1, collecting data online, comprising the following steps: S11, training navigation model in simulation environment;Define the state space of mobile robot in simulation environment, and form state vector;S12, send state vector into graph attention network, obtain state representation;S13, policy network takes state representation as input, and outputs control action;The reference action of linear velocity and angular velocity is calculated by artificial potential field method;S2, during the training phase of navigation model, train critic network, policy network and graph attention network;S3, after the training of navigation model is completed, deploy navigation model on mobile robot.
Owner:ZHEJIANG GONGSHANG UNIVERSITY

Unmanned navigation method and system applied to forage harvesting

The application provides an unmanned navigation method and system applied to forage harvesting, relates to the technical field of agricultural machinery automation, and first captures forage growth state signals, terrain interaction feedback signals and operation track residual signals in a work area, constructs a track residual dynamic linkage model; generates track evolution rules containing track bending amplitude adjustment, elevation compensation and offset avoidance modes based on the track residual dynamic linkage model; collects real-time signals and current operation track data of equipment in the operation process through a global perception component; inputs the same into the track residual dynamic linkage model to generate track deviation correction parameters; generates dynamic track adjustment instructions to drive equipment to update the operation track according to the track deviation correction parameters and the track evolution rules, and inputs the updated data into the model in reverse to complete cyclic optimization. The application improves the precision and adaptability of the forage harvesting unmanned navigation, and improves the harvesting efficiency and quality.
Owner:INSTITUTE OF GRASSLAND RESEARCH OF CAAS +1

A generalizable end-to-end autonomous navigation method based on imitation-reinforcement learning.

ActiveCN122083939AImprove navigation performanceRealize real-time obstacle avoidanceNavigational calculation instrumentsBiological modelsData setNetwork architecture
This invention discloses a scene-generalizable end-to-end autonomous navigation method based on imitation-reinforcement learning, belonging to the field of robotics technology. Addressing the problems of unstable action output and susceptibility to local optima in existing robot navigation methods, this invention constructs a TD3 network architecture, continuously interacting with the environment in a Gazebo simulation to acquire experience. The network parameters are iteratively optimized and trained using an experience replay buffer. The trained neural network outputs continuous linear and angular velocity commands for the robot based on the input state, which, after scaling, drive the robot's movement, achieving obstacle avoidance and target point navigation. Target approach rewards, collision penalties, and motion process rewards are designed to guide the policy network to improve its stable navigation ability to the target point while ensuring safe obstacle avoidance. Imitation learning is introduced to optimize the robot's motion posture; the Actor network is fine-tuned by collecting expert datasets to ensure the robot maintains posture stability during movement.
Owner:ZHONGBEI UNIV