1029 results about "Tree trunk" patented technology

The invention discloses a fine individual tree segmentation method fusing air-based laser radar point clouds and ground-based laser radar point clouds, and aims to solve the challenges that the data volume of combined point clouds is large, crown staggering is complex in a dense forest environment, and part of tree trunk point clouds are lost due to shielding, and the method combines an individual tree segmentation strategy from top to bottom and an individual tree segmentation strategy from bottom to top. A weighted height-density map is adopted to carry out individual tree positioning, a mark control watershed algorithm is combined to realize preliminary tree crown region extraction, then multi-label local graph segmentation optimization is introduced, segmentation boundaries are corrected, an under-segmentation unit is optimized by adopting a shortest path, and a complete segmentation framework is formed. On the basis of airborne and foundation point clouds of six actually measured sample plots, the method is verified after registration and combination. The advantages of airborne point cloud and foundation point cloud are combined, and the proposed individual tree segmentation method provides technical support for refined forest modeling and ecological monitoring.

The invention discloses a pest control management system and method for forest tree breeding, and relates to the technical field of forest tree breeding and seedling raising, and the method comprises the following steps: collecting environment data and soil rhizosphere data by collecting multi-mode images of leaves and trunks of a pinus massoniana young forest and videos that pinus massoniana caterpillars gnaw leaves and cluster diffusion; constructing a disease and insect pest recognition model by using a convolutional neural network, performing definite diagnosis on pine wood nematode disease, pine caterpillars and monochamus alternatus hope, and quantifying the disease and insect pest occurrence area and population density of each sub-region; calculating a pine wood nematode disease control demand index, comparing the pine wood nematode disease control demand index with a threshold value, and triggering a corresponding control strategy; the diffusion coefficient of the pine caterpillars is calculated and compared with a risk threshold value, and a prevention and control scheme is formulated; calculating the propagation coefficient of the monochamus alternatus hope, judging the propagation risk and giving a prevention and control strategy; according to the invention, intelligent pest identification, risk quantification and control strategy closed-loop control are realized, treatment and scientific prevention and control are facilitated, and the pest control efficiency of the pinus massoniana young forest and the forest stand health level are improved.

The invention discloses a three-dimensional tree modeling method and device based on 3DGS. The method comprises the steps of determining modeling components including a trunk, branches and leaves for a three-dimensional tree modeling task, obtaining tree-surrounding image data collected by an unmanned aerial vehicle, performing camera pose estimation and obtaining sparse point clouds through sparse reconstruction, inputting a three-dimensional Gaussian splash 3DGS model to generate point clouds of a three-dimensional tree, and obtaining three-dimensional tree point clouds according to the point clouds of the three-dimensional tree point clouds. Selecting a corresponding model according to a segmentation object to generate 2D masks of different visual angles, extracting a complete tree to select a segmentation arbitrary model SAM model, extracting a tree component to select a semantic segmentation network MFAANet model based on a coding and decoding structure, generating the corresponding 2D masks on input images of different visual angles, and based on a multi-visual-angle mask obtained through segmentation, extracting a multi-visual-angle mask based on the multi-visual-angle mask. And generating a final point cloud classification result through a multi-view mask voting mechanism. The method is obviously superior to the existing method in the aspects of tree rendering quality and component segmentation precision.

The invention discloses a single tree trunk structure extraction method and system based on deep learning, and the method comprises the steps: obtaining two-dimensional image data of a single tree, marking the two-dimensional image data, and constructing a single tree trunk segmentation data set; constructing a spatial domain and frequency domain double-branch network based on the segmented data set, respectively extracting spatial domain features and frequency domain features, and fusing the double-branch features to generate a coding feature map; the coding feature map is decoded, in the decoding process, coordinate convolution CoordConv is adopted to enhance position perception, and mask segmentation and semantic label extraction are respectively carried out through a dynamic mask reconstruction branch DRMask Branch and an instance branch Inst Branch; based on a bipartite graph matching strategy, associating results of the mask segmentation and instance branches, and realizing segmentation of a single tree trunk and matching of instance-level labels; according to the method, the boundary precision and the detail reconstruction capability of trunk segmentation are remarkably improved, and the problem of feature loss of a traditional method in a complex under-forest environment is solved.

The invention relates to an intelligent climbing coconut tree flexible picking and transporting integrated robot which comprises two open-close type circular rails, a connecting frame, a climbing mechanism, two open-close type clamping arm assemblies, a clamping and picking mechanism and a pruning mechanism, the two open-close type circular rails are distributed up and down in a spaced mode, and the two open-close type circular rails are both automatically opened or closed to form a whole ring; the connecting frame is installed between the non-opening-closing sides of the two sets of opening-closing type circular rails, the climbing mechanism is arranged on the inner side of the connecting frame and used for making contact with a tree trunk and climbing along the tree trunk, and the two sets of opening-closing type clamping arm assemblies are installed on the connecting frame and can be automatically opened or closed to hold the tree trunk. The lower end of the lower opening-closing type clamping arm assembly and the lower end of the connecting frame are jointly provided with a ground walking mechanism, and the clamping picking mechanism and the pruning mechanism are installed on the two sets of opening-closing type annular rails through a set of moving assemblies respectively and can move in the annular direction of the opening-closing type annular rails through the moving assemblies respectively. The robot has the advantages that automatic climbing, picking and transporting walking can be achieved, potential safety hazards are reduced, and picking and transporting efficiency is improved.

The invention discloses a rotor unmanned aerial vehicle visual navigation method and system for an unknown forest region, and the method comprises the steps: firstly, combining a depth camera with a Sobel gradient filtering mechanism, and effectively eliminating interference information; secondly, pixel-level segmentation and three-dimensional positioning of the trunk are realized by segmenting a cascade model of a convolutional network and a residual network, visual information is accurately mapped to a world coordinate system, and the problem of positioning drift caused by similar textures in a forest scene is solved; based on starting point-target point-terminal point minimum jerk trajectory planning, a smooth trajectory is generated through polynomial optimization and quadratic programming, flight jitter is inhibited fundamentally, and the mechanical loss is reduced while the image acquisition quality is guaranteed; and finally, through real-time target point updating and trajectory dynamic correction of the depth camera, a sensing-planning-execution closed loop is formed, so that the unmanned aerial vehicle has dynamic obstacle avoidance and path online optimization capabilities, autonomous crossing is finally realized, and navigation safety and stability in a complex environment are significantly improved.

The invention relates to a system for remote and / or autonomous holding at least a portion of a tree trunk, said system comprising: a remotely and / or autonomously controlled Unmanned Aerial Vehicle (100), UAV, at least one remotely and / or autonomously controlled means (105) for holding at least a portion of a tree trunk attachable to said UAV (100), means for detecting at least a portion of a tree, means for detecting at least one tree parameter of at least a portion of a tree and / or at least one growing condition of at least a portion of a tree, a base station for communication with said UAV (100), means configured for directing said means (105) for holding at least a portion of a tree trunk to a particular position of said tree trunk depending on said at least one detected tree parameter and / or said at least one detected growing condition.

The invention discloses a tree climbing device with axial climbing and circumferential steering functions. The tree climbing device comprises a vertically symmetrical main trunk mechanism, an upper clamping and climbing mechanism and a lower clamping and climbing mechanism, wherein the upper clamping and climbing mechanism and the lower clamping and climbing mechanism are identical and bilaterally symmetrical. The main trunk mechanism is arranged in the middle of the robot, telescopic mechanisms are placed in an upper operation platform and a lower operation platform of the main trunk mechanism to control clamping and loosening of the clamping mechanisms, the clamping and climbing mechanisms are arranged on the upper portion and the lower portion of the main trunk and each composed of two grabbing clamps of the completely same structure, and universal wheels are carried in the clamping and climbing mechanisms to serve as execution ends. The rotor wings are carried outside to serve as a power mechanism, and climbing and circumferential movement of the robot on trees are achieved through driving of the auxiliary effect of the rotor wings. A turnover mechanism is arranged in the middle of the main trunk mechanism, the robot can be controlled to turn over through steering engine driving, and conversion of the robot from the trunk to the side trunk is achieved. Under the movement action of the main trunk mechanism, the upper clamping and climbing mechanism and the lower clamping and climbing mechanism, the tree climbing device is simple in structure, good in obstacle crossing ability and good in stability, and clamping force can be actively adjusted to adapt to continuous changes of the diameter of a tree; the self-locking device can move axially and circumferentially along a trunk, has a self-locking function and can stop at any position.

According to the light and simple crawler-type vibration harvester suitable for the steep mountain land, the triangular crawler belt is adopted for driving, the contact area with the ground is increased, the climbing performance is improved, maneuverability, flexibility and reliability are high, flexible connection is formed through the elastic coupler, and vibration transmission is stable. The device comprises a vehicle body, a suspension platform and a vibration clamping mechanism, wherein the vehicle body comprises an H-shaped bottom formed by two longitudinal beams and a cross beam; triangular driving tracks are arranged at the rear ends of the longitudinal beams; auxiliary wheels are arranged at the front ends of the longitudinal beams; the suspension platform is located above the position between the front portions of the two longitudinal beams, the rear portion of the suspension platform is fixed to the sliding frame, and the sliding frame is arranged on the guide rods in an up-down sliding mode. The lifting driving device for driving the suspension platform or the sliding frame to lift up and down relative to the guide rod is connected with the suspension platform or the sliding frame; the vibration clamping mechanism used for clamping a tree trunk and conducting excitation vibration on the tree trunk is hung on the driving hanging platform through a flexible rope.

The utility model relates to the technical field of forestry maintenance, and discloses a tree-climbing pruning robot which comprises a supporting mechanism, the supporting mechanism comprises a machine box internally provided with a storage battery and a control module, the two sides of the machine box are each provided with a positioning connecting rod, and the whole machine box is in a clamping ring shape with a notch and is used for surrounding a tree; the tree climbing machine further comprises three tree climbing mechanisms, each tree climbing mechanism comprises a driving box with a driving shaft, the driving boxes are connected to the surface of the machine box and the ends of the two position adjusting connecting rods respectively, and the driving shafts of the driving boxes are sleeved with crawler belts attached to the surfaces of trees. The pruning device further comprises a pruning mechanism which is arranged at the top end of the driving box and used for pruning trees, the tree climbing mechanism can be attached to the surfaces of the trees and is matched with the crawler belt to move to drive the device body to move upwards, and therefore the pruning mechanism climbs the trees to the proper positions to prune branches. Manual tree climbing operation is avoided, and the safety of the pruning process is improved.

The invention discloses a pesticide compound composition of abamectin and dsRNA (double-stranded ribonucleic acid) and application of the pesticide compound composition in prevention and control of nematodes. According to the invention, dsRNA is respectively designed and synthesized for five Mcm genes of bursaphelenchus xylophilus, and nucleotide sequences of positive-sense strands of the dsRNA are respectively shown as SEQ ID No.1-SEQ ID No.5; the synthesized dsRNA and abamectin of different components are compounded to obtain the pesticide compound composition, the pesticide compound composition shows a remarkable synergistic effect in the aspect of preventing and treating pine wood nematode diseases, and the nematode killing effect of combined use of the dsRNA and the abamectin is superior to that of single use of any agent; the pesticide compound composition has the potential to become a novel trunk injection for preventing the pine wood nematode disease.

The hemp tree is a malviaceae hibiscus deciduous tree, the trunk and branches contain 65%-75% of cellulose, the hemp tree grows fast, the adaptability is high, the resource utilization rate is low, and the hemp tree powder can achieve the win-win situation of high value and environmental protection. The invention discloses a hemp tree powder bio-based degradable material and a preparation method thereof. The material is prepared from 42%-58% of hemp tree powder, 28%-38% of polylactic acid (PLA), 4%-7% of nano montmorillonite, 3%-5% of hexamethylene diisocyanate and 1.5%-2.5% of calcium stearate. In the preparation process, the hemp tree powder is subjected to alkaline leaching, acid neutralization and enzymolysis modification, nano-montmorillonite composite modification, polyethylene glycol and PLA graft modification; and carrying out segmented temperature control mixing, optimized feeding, warm water granulation and antioxidant addition to prepare the target material. The oxygen transmission rate of the material is smaller than or equal to 45 cm / (m24h 0.1 MPa), the bending strength is larger than or equal to 26 MPa, the tensile strength retention rate after thermal oxidative aging at 120 DEG C for 168 h is larger than or equal to 88%, and the material is excellent in performance and can be used for packaging and disposable products and efficiently utilizing hemp tree resources.

The utility model relates to the technical field of agricultural machinery, and provides a forest fruit picking and collecting all-in-one machine which comprises a machine body, a collecting mechanism, a vibrating mechanism and a material conveying mechanism, the collecting mechanism and the vibrating mechanism are both connected with the machine body, the collecting mechanism has a first unfolding state and a first folding state, and when the collecting mechanism is in the first folding state, the collecting mechanism is unfolded. The collecting mechanism is used for collecting the forest fruits which are vibrated off, the vibrating mechanism is arranged below the collecting mechanism, the vibrating mechanism comprises a vibrating machine and a clamping assembly, the clamping assembly is connected with the vibrating machine, the clamping assembly has a second unfolding state and a second folding state, and when the clamping assembly is in the second folding state, the clamping assembly is used for clamping tree trunks and vibrating off the forest fruits; and the conveying mechanism is used for outputting the forest fruits collected by the collecting mechanism. According to the forest fruit picking device, a traditional picking mode is changed, integrated operation of forest fruit picking, collecting and conveying is achieved, the picking efficiency is improved, and damage to tree trunks is reduced.

The invention discloses a handheld forest region sample plot calibration surveying and mapping device and method, and relates to the technical field of forest resource investigation, and the device comprises a laser radar, a visual camera and a global navigation satellite system / inertial orientation positioning navigation system integrated navigation unit; when a user carries the device to move, the processing unit constructs a three-dimensional point cloud map and calculates a six-degree-of-freedom pose; when the signal of the global navigation satellite system is unavailable, calculating a current position by combining the three-dimensional point cloud map, the six-degree-of-freedom pose and inertial data; and the processing unit calculates parameters of the stumpage and the candidate area and generates a sample plot calibration surveying and mapping report. According to the invention, the device achieves the synchronous capturing of the characteristics of the trunk base and the canopy at a handheld height, an operator only needs to click and select an initial position on a touch screen, and the system combines laser instant positioning and map construction with a global navigation satellite system / inertial orientation positioning navigation system for tight coupling positioning. And automatically generating sample plot boundaries and ecological parameters conforming to regulations.

Disclosed is a customizable decorative tree, a customizable decorative tree is provided, which includes a trunk, a base, and a top section. The trunk is formed from a plurality of individual tree sections that are designed for selective sectional assembly and disassembly. Adjacent tree sections are connected using a shared mechanical connector and a corresponding electrical connector. This integrated system facilitates both robust structural connection and the simultaneous transmission of electrical power through the assembled trunk and base. The tree sections may have varying heights and diameters, allowing a user to customize the trunk to achieve a natural appearance or a unique style. By integrating the electrical power transmission within the structure of the trunk, the design eliminates the need for external wiring, thereby enhancing safety and aesthetics.

The application relates to a plant protection unmanned aerial vehicle spraying control method, device, equipment and medium, the method comprising: acquiring pose transformation data of each trunk of a pose sensor in a fruit tree canopy in a preset time range, each flight parameter of a plant protection unmanned aerial vehicle and fruit tree canopy image data of a vision sensor in the plant protection unmanned aerial vehicle; determining a mist drop deposition amount in the fruit tree canopy under each flight parameter, constructing a plant protection unmanned aerial vehicle spraying control model according to the mist drop deposition amount and the canopy disturbance prediction model; inputting each flight parameter of the plant protection unmanned aerial vehicle into a pre-trained plant protection unmanned aerial vehicle spraying control model, determining a fruit tree canopy disturbance state corresponding to the flight parameter and a mist drop deposition amount under the fruit tree canopy disturbance state, so as to complete the control of the plant protection unmanned aerial vehicle spraying. The application can greatly avoid the great influence of the wind field generated by the unmanned aerial vehicle rotor on the deposition effect of the pesticide mist drop.

The invention discloses a single tree point cloud instance segmentation method, device and equipment and a storage medium, and the method comprises a scene semantic segmentation step, a Z-axis standardization step, a trunk height section interception and three-dimensional to two-dimensional conversion step, a section detection and root seed generation step, and a three-dimensional growth type aggregation segmentation step. By constructing a closed-loop process combination of semantic segmentation pre-filtering, ground reference standardization, trunk height section two-dimensional enhancement, root seed generation through section detection and three-dimensional aggregation segmentation, stable individual tree instance segmentation in a complex terrain and tree adjacent overlapping scene is realized. And reliable input data is provided for subsequent tree parameter calculation and greening asset statistics.

The invention discloses an ancient tree health state monitoring method based on digital twinning, and relates to the field of ancient tree health monitoring, and the method comprises the steps: collecting point cloud data and texture data of an ancient tree, and constructing an initial three-dimensional model of the ancient tree; a micro ion selective electrode array is deployed in an ancient tree root distribution area, Ca2 and K + ion flow velocity changes are detected through a constant potential rectifier, and the mycorrhizal symbiosis index is calculated in combination with soil temperature and humidity data; uploading the mycorrhizal symbiosis index to a digital twinborn body, and displaying the active state of a root system area in an initial three-dimensional model of the ancient tree in a three-dimensional thermodynamic diagram form; a 16-channel MOX gas sensor electronic nose array is installed on the surface of a trunk of an ancient tree, and volatile organic compound (VOC) concentration data of the ancient tree are collected. According to the invention, a micro electrode array with a 45-degree inclination angle is adopted to non-destructively monitor the ion flow velocity of a root system through the mycorrhizal symbiosis index detection unit, and a three-dimensional thermodynamic diagram is generated in combination with an inverse distance weighted interpolation algorithm, so that dynamic visualization of underground mycorrhizal activity is realized.

The provided is a hanging pod, including a hanging pod body, a light-emitting device, and a soft cushion. The hanging pod body is hollow and conical, allowing a user to lie or sit inside the hanging pod body for rest or entertainment. A top of the hanging pod body is provided with a connector, through which the hanging pod is connected to an indoor hanging point or an outdoor tree trunk, railing, etc., such that the hanging pod is vertically hung as a whole. Featuring convenient setup and small footprint, the hanging pod can serve as a lightweight and easy-to-carry relaxation tool for the user during outdoor activities, and can be set up in a small space indoors for the user to rest or entertain. The light-emitting device is detachably connected to the hanging pod body. The light-emitting device can provide light indication and illumination in dark environments.

The invention discloses a method for improving soil fertility of an orchard in a Yellow River old lane area. The orchard is covered with green grass; after the fruits are harvested and before leaves fall, a first fertilization ditch is dug at a crown water drip, and smashed straw, base fertilizer and surface soil are sequentially laid in the first fertilization ditch from bottom to top; meanwhile, digging 3-4 radial second fertilization ditches according to a water drip from a trunk to a crown, sequentially laying dry branches and fallen leaves, wheat straws and livestock and poultry manure in the second fertilization ditches from bottom to top, and putting earthworms into surface straws of the second fertilization ditches; a humic acid-containing water-soluble fertilizer is applied in a drip irrigation manner during topdressing. Therefore, through the synergistic effect of multiple technical measures such as orchard grass covering, organic fertilizer layer application, earthworm density increasing and bio-fertilizer application increasing, a benign soil ecological system is constructed, and therefore the problems that the content of organic matter in orchard soil in the Yellow River old road area is low, and the water and fertilizer retention capacity is weak are effectively solved, the soil fertility is remarkably improved, and the yield of the orchard is increased. And the fruit yield and quality are improved.

A marking element for tree trunks, with a base plate that has an striking surface and a cutting surface facing away from the striking surface. A cutting edge protrudes from the cutting surface and an electronic information carrier is attached to the base plate, which has a memory module and an antenna that is electrically connected to the memory module. The antenna has an antenna section which protrudes laterally beyond the striking surface and beyond the cutting surface.

The invention discloses a forest LiDAR point cloud crown-trunk segmentation method and system based on soft semantic grouping, and the method comprises the steps: constructing a soft semantic grouping SSGM module, carrying out the processing of input forest LiDAR point cloud data, and carrying out the soft semantic annotation of points in a crown-trunk transition region through a probability multi-class distribution mechanism; constructing a feature fusion clustering (FCM) module, receiving the point cloud features processed by the SSGM module, and performing instance-level clustering through a density-invariant clustering algorithm; constructing a deformable convolution up-sampling DCUM module, performing up-sampling on the clustered point cloud features, and adapting to a non-rigid geometric structure of the forest point cloud through learnable spatial offset; performing end-to-end joint training and parameter optimization on the SSGM module, the FCM module and the DCUM module to complete panoramic segmentation of the crown and the trunk in the forest LiDAR point cloud; the method adapts to point cloud density and distribution difference of different LiDAR scanning platforms, overcomes dependence of a traditional method on a single data source, and realizes cross-platform and cross-season stability.

The invention discloses a texture fusion method and device for a multi-source heterogeneous tree three-dimensional model, a terminal and a storage medium, and belongs to the field of three-dimensional model processing. The method comprises the following steps: acquiring an oblique photography tree model of a trunk part of a target tree, and reconstructing a texture map of the target tree to obtain a trunk texture map; constructing a laser tree model of the target tree, and performing texture mapping on a trunk part of the laser tree model based on the trunk texture map to obtain a target trunk model; obtaining a tree crown simulation texture map, and performing texture mapping on a tree crown part of the laser tree model based on the tree crown simulation texture map to obtain a target tree crown model; and obtaining a target transition region, and carrying out boundary weighted fusion on the target transition region to obtain a target three-dimensional tree model. According to the method, geometric and texture collaborative optimization processing of the multi-source heterogeneous tree model is realized, and the tree model with more excellent material performance and scene adaptability can be generated.

The invention relates to the field of rubber tapping of rubber trees, in particular to an unmanned aerial vehicle rubber tapping device which comprises an unmanned aerial vehicle, a clamping supporting device is arranged on the unmanned aerial vehicle and comprises a clamping device, a flexible clamping rubber tapping device is arranged below the clamping device, and the flexible clamping rubber tapping device comprises a multi-joint rail support formed by combining a plurality of joints. A flexible clamping rail is installed above the multi-joint rail support in parallel, the multi-joint rail support and the flexible clamping rail deform synchronously, a rubber tapping walking mechanism is connected to the flexible clamping rail in a sliding and clamping mode and comprises a tool base, a tool mechanism used for rubber tapping is arranged on the tool base, a rail clamping device is arranged in the tool base, and the rail clamping device is connected with the flexible clamping rail in a sliding and clamping mode. The rail clamping device is rotationally connected with the cutter base; and the flexible clamping rubber tapping device further comprises a first driving device. When the rubber tapping device is used, the problem of different rubber tapping depths caused by irregular tree trunks in the traditional rubber tapping operation can be solved.

The invention belongs to the technical field of fruit picking, and particularly relates to tea oil fruit picking and selecting integrated equipment which comprises a moving frame, a mounting plate is connected to the top of the moving frame, two arc-shaped plates are rotationally connected to the mounting plate, baffles are connected to the left ends of the arc-shaped plates, and a funnel-shaped fruit receiving bag is connected between the two baffles. The fruit receiving bag has an open state and a closed state, a round opening is formed in the middle of the fruit receiving bag in the closed state, a material passing opening is formed in the right portion of the fruit receiving bag, and two arc-shaped soft blocks are connected to the round opening. The two baffles are opened, the fruit receiving bag is opened, the trunk is located between the two arc-shaped soft blocks, the two baffles are closed, the fruit receiving bag is closed, the trunk extrudes the V-shaped rods, the two V-shaped rods are opened, the two V-shaped rods surround the trunk through the volute spiral spring, the pusher is controlled to make the trunk shake, the tea oil fruits fall into the fruit receiving bag, and the tea oil fruits fall into the material passing frame through the material passing opening. The tea oil fruits and leaves are separated through the material guide rod, the fruit receiving bag is automatically installed, and the picking and selecting efficiency is improved.

The invention relates to a tree climbing and picking robot with branch crossing and self-adaptive grasping functions. The tree climbing and picking robot comprises a scissor module, a self-adaptive clamping jaw, a telescopic mechanism and a holder. The scissor module is used for pruning to realize picking operation; the upper portion and the lower portion of the robot are each provided with a self-adaptive clamping jaw, and the robot is used for grabbing a trunk in a self-adaptive mode through opening and closing of the self-adaptive clamping jaws. The telescopic mechanism is used for achieving stretching and retracting of the robot, the upper end of the telescopic mechanism is connected with the self-adaptive clamping jaw on the upper portion, the lower end of the telescopic mechanism is connected with the self-adaptive clamping jaw on the lower portion through the holder, the relative angle between the two self-adaptive clamping jaws is changed through the holder, and posture adjustment of the robot is achieved. Through cooperative work of the self-adaptive clamping jaw, the telescopic mechanism and the holder, the robot can climb along a tree trunk, cross multiple branches and adjust the tail end operation posture. According to the invention, stable grasping can be realized, high rigidity can be maintained under a large stroke, the flexible branch crossing capability is realized, and high-efficiency and high-safety automatic picking operation can be realized.

The invention discloses an airborne laser radar point cloud inverted tree detection method and device and electronic equipment, and the method employs a deep learning semantic segmentation model to process a first point cloud, can precisely distinguish an inverted tree trunk from ground cover objects, rocks and other sundries, remarkably reduces the false detection rate, and guarantees the accuracy of target recognition. And then, a two-dimensional plane projection image is generated based on the second point cloud, so that the problem of discontinuity caused by bending and fracture of the three-dimensional point cloud can be avoided, structured input is provided for subsequent target detection, and the continuity of the detection process is ensured. Afterwards, target detection is carried out on the two-dimensional plane projection image, fragmented missing detection caused by point cloud discontinuity can be avoided, and the overall recognition capability (high recall rate) of the inverted tree is greatly improved while high detection accuracy (low false positive) is kept. And finally, mapping a two-dimensional detection result back to a three-dimensional space, so that three-dimensional point cloud data of a complete inverted tree can be obtained, and a high-density point cloud environment can be adapted more effectively.

The invention discloses a robot orchard inter-row positioning method based on multi-modal semantic graph optimization, and aims to solve the problems of inter-row leakage and inter-row jump in positioning in orchard moving operation. Time synchronization and external parameter calibration are carried out on multi-sensor data, semantic analysis and geometric preprocessing are carried out, and the positioning accuracy of the orchard inter-row positioning method is improved. The method comprises the following steps: generating tree trunk, pile body, tree wall and ground boundary observation, constructing a semantic factor graph containing row identifiers and continuous variables, adopting a graph converter to perform adaptive weighting and covariance calibration, combining soft and hard two-stage increment optimization of anti-fact mutual exclusion gating and discrete continuous combination, outputting the row identifiers, robot trajectories, transverse deviation and course angles, and constructing a semantic factor graph containing the row identifiers and the continuous variables. The technical effects of stably inhibiting the inter-row jump and improving the row identification judgment accuracy and the transverse deviation and course angle estimation precision are achieved.

A mobile machine movable between multiple rows of trees includes at least two LiDAR sensors to output two-dimensional or three-dimensional point clouds data indicating a distribution of objects in a surrounding environment of the mobile machine, a storage to store environment map data indicating a distribution of trunks of the multiple rows of trees, a localization processor, and a controller to control movement of the mobile machine in accordance with a position of the mobile machine as estimated by the localization processor. The localization processor is configured or programmed to detect the trunks of the rows of trees in the surrounding environment of the mobile machine based on the point clouds data that is repeatedly output from the at least two LiDAR sensors while the mobile machine is moving, and to perform matching between the detected trunks of the rows of trees and the environment map data to estimate a position of the mobile machine.

The invention discloses a tree positioning calculation method based on unmanned aerial vehicle remote sensing, and relates to the technical field of agriculture and forestry vegetation positioning. According to the method, firstly, a digital orthoimage and a digital earth surface model (DSM) are generated through aerial photography of an unmanned aerial vehicle, a crown mask is obtained by using an instance segmentation technology, and a trunk position is accurately positioned by calculating a fusion score in the mask by using a joint resolving mechanism of a DSM local peak domain and an iterative corrosion morphological core. The method improves the positioning precision and efficiency through a DSM highest point * iterative corrosion combined optimization strategy, and improves the positioning stability of dense crowns through the corrosion depth constraint space position. According to the method, the positioning precision and efficiency of a complex forest environment are remarkably improved, and the problems of insufficient positioning precision, difficulty in positioning trunks under dense crowns and the like in the existing unmanned aerial vehicle remote sensing technology are solved.