84results about How to "Reduce cumulative error" patented technology

The invention provides a new energy automobile endurance prediction method based on working condition identification and prediction, and belongs to the technical field of new energy automobile endurance mileage estimation. According to the method, sliding window statistical characteristics are constructed through speed, current, voltage and other signals, typical working conditions of cities / suburbs / highways and the like are identified, the working condition of a next time window is predicted, and energy consumption priori which can be updated in real time along with changes of road conditions is formed. Secondly, a rolling capacity estimation model is provided, the available capacity of the battery is continuously evaluated based on historical charging and discharging data, and dynamic changes of the SOH of the battery along with time and use conditions are recorded; according to the method, working condition prior, rolling SOH and time sequence energy consumption characteristics are fused, a TCN-BiLSTM multi-model cooperation framework is constructed, short-term dynamic and long-term attenuation trends are considered, high-precision and generalizable endurance mileage prediction under complex working conditions and battery state fluctuation is realized, reliable support is provided for intelligent energy consumption management and journey planning of vehicles, and the method is suitable for popularization and application. The method can also be applied to an intelligent power distribution system with a high reliability requirement, and provides a core solution for technology promotion service in related fields.

The invention discloses an uplink and downlink combined wireless communication method and system based on channel semantic feature feedback. The method constructs an uplink and downlink closed-loop semantic communication architecture: a terminal side obtains a downlink channel matrix, generates a channel semantic feature vector through a channel semantic feature extraction network, inputs the channel semantic feature vector into a compression coding network to obtain a low-dimensional feedback code word, and sends the low-dimensional feedback code word to a base station side through an uplink limited feedback link; the base station side decodes and restores the low-dimensional feedback code word to obtain a reconstructed channel semantic feature vector, inputs the reconstructed channel semantic feature vector as prior information into a channel sensing sequence coding and modulation network, and generates a complex modulation signal under the condition of meeting the transmission power constraint; and the terminal side is combined with local channel information, and original data is recovered through a channel sensing demodulation network. Through joint optimization of the channel semantic feedback and the transceiver end-to-end transmission process, the channel state information feedback overhead is reduced under the constraint of limited feedback resources, and the transmission reliability and robustness in a complex channel environment are improved.

The invention belongs to the field of multi-sensor fusion systems, and discloses a multi-sensor fusion system simultaneous positioning and mapping method and a related device, and the method comprises the steps: obtaining an environment image, a laser point cloud and IMU data; scene semantic information is extracted through a multi-modal large model; dynamically generating a weight matrix of vision and laser radar relative to an IMU (Inertial Measurement Unit) by utilizing a vision and point cloud condition deep neural network in combination with scene information and IMU observation; and performing local and global optimization by combining each sensor speedometer factor, an IMU pre-integration factor and the weight matrix, and finally outputting robot state estimation and map point coordinates. Through semantic understanding and a self-adaptive weight mechanism driven by a conditional deep neural network model, a fusion strategy can be intelligently adjusted before environment change or sensor degradation, the robustness, precision and consistency of a system in an unstructured scene are improved, and the problems of positioning drift and map failure caused by fixed weight in a traditional method are solved.

This utility model discloses a novel switch structure, including a panel assembly, a PCB board, a power board, and a base assembly. The panel assembly includes an outer frame, a button assembly, and a back cover arranged sequentially from front to back. A cavity is formed on the surface of the back cover. A receiving cavity is provided inside the outer frame, and a button hole communicating with the receiving cavity is formed on the surface of the outer frame. An assembly groove is formed on the inner surface of the outer frame. The PCB board is located between the button assembly and the back cover assembly, and at least one button is provided on the PCB board. The button assembly includes a button plate and a button glass. The button plate has button areas corresponding to each button, and the button plate is connected to the outer frame via the assembly groove. The button areas protrude from the button plate. The button glass is disposed within the button hole and is fixedly connected to the button area. The button hole and button glass are arranged in a one-to-one correspondence. After the back cover is connected to the outer frame via a locking member, the PCB board and the button assembly are fixed within the receiving cavity. This utility model features high assembly precision, low cost, and multiple control functions.

The utility model discloses a low and small-sized steering wheel structure which comprises a bearing plate, a containing groove is formed in the upper surface of the bearing plate, and a driving wheel capable of intruding into the containing groove and a walking motor controlling the driving wheel are arranged on the lower surface of the bearing plate. The driving wheel part is arranged in the containing groove of the bearing plate, meanwhile, the steering motor is installed on the bottom face of the bearing plate, so that the whole steering wheel structure is more compact in the height direction, the overall height of the steering wheel is reduced, and the steering wheel is suitable for occasions with strict limitation on the height, such as a submerged AGV; moreover, even if large bearing capacity needs to be provided, the structure can guarantee the lower height of the AGV steering wheel driving device, and the characteristics of low height and compactness of the AGV steering wheel driving device are not affected.

This application belongs to the field of computer vision technology and discloses a cross-source point cloud registration method based on adversarial neural networks. The method includes: inputting a multi-frame LiDAR point cloud sequence and a UAV video sequence; segmenting each frame of the LiDAR point cloud into ground points and non-ground points and registering them to obtain a global LiDAR point cloud; performing motion recovery structure reconstruction on the UAV video sequence to generate a dense color point cloud; inputting the global LiDAR point cloud and the SfM dense color point cloud into an MMAlignNet network to construct a shared latent feature space; solving for the rigid transformation matrix from the LiDAR point cloud to the SfM dense color point cloud; and aligning the cross-modal point clouds according to the rigid transformation matrix to obtain a multimodal 3D reconstruction result. This application improves the registration accuracy, completeness, and robustness of large-scale outdoor environment 3D reconstruction by fusing UAV imagery and mobile LiDAR data, combined with cross-modal feature alignment and geometric constraint optimization methods.

The application discloses a multiscale target tracking method based on scale and position separation, comprising the following steps: inputting a template image and a current frame search image into a multiscale network respectively for feature extraction, and obtaining multiscale template features and target features correspondingly; inputting the template features and the target features into a position estimation detection head respectively for processing, and performing convolution on two output results to obtain a preliminary position estimation result; inputting the target features into a scale estimation detection head for processing to obtain a scale estimation result; clustering the preliminary position estimation result to obtain a final position estimation result; splicing the final position estimation result and the scale estimation result to obtain a prediction result; when judging that the number of processed search image frames reaches a preset number, updating the template features, and performing prediction on a next frame search image. The method improves the stability and robustness of the whole tracking process, and has high tracking accuracy.

The application discloses a kind of based on track keeping's unmanned aerial vehicle thermal infrared image splicing method and system, it is related to image splicing technical field, including steps: constructing image set;Obtain the matching feature point of any pair of to-be-spliced unmanned aerial vehicle's thermal infrared image respectively, and the re-projection error of matching feature point is generated alignment item by the similar transformation matrix of to-be-spliced thermal infrared image;The actual distance between the center of reference image and the thermal infrared image of any unmanned aerial vehicle is calculated, and the track trajectory protection item is obtained between the center point of similar transformation matrix and reference image by pixel distance;Alignment item and track trajectory protection item are weighted and summed to construct global optimization objective function, and global optimization objective function is converted into linear least square model with similar transformation constraint to solve, and generate spliced thermal infrared panoramic image.The application minimizes the cumulative error of splicing process, while ensuring the integrity of global structure.

The application discloses a stator assembly and a motor with the same. The stator assembly comprises a stator body, a first bearing support arranged on one end of the stator body and formed with a first bearing mounting position, and a second bearing support arranged on the other end of the stator body and formed with a second bearing mounting position. When a rear bearing is mounted on the second bearing support through the second bearing mounting position, an installation gap is left between the rear bearing and the second bearing support, which serves as a mounting position of a wave-shaped gasket. The structure of the application changes the mounting position of the wave-shaped gasket on the motor, effectively reduces the parts affecting the precision of the distance between the shaft extension end face and the flange end face of the motor, reduces the tolerance accumulation and assembly error of the parts, and further improves the precision of the distance between the shaft extension end face and the flange end face of the motor, and facilitates the assembly of the motor.

This invention provides a method and system for generating motion data based on first-person perspective video. The method involves acquiring video image information and determining the camera's internal parameters; extracting feature points from adjacent video frames to estimate camera motion; determining whether the camera has returned to a previously visited environmental area based on the acquired image features, and establishing loop closure constraints; detecting camera shake or stillness in the video to provide position commands for generating special effects; optimizing the camera pose measured at different times and using loop closure detection information to obtain globally consistent camera motion trajectory data; mapping the generated globally consistent camera motion trajectory data onto the motion range of a motion seat as the main driving signal, superimposing the position commands generated by special effects, and generating corresponding motion data. This invention offers high processing efficiency and provides users with a more realistic experience.

The invention discloses a four-door two-cover flexible trimming die switching system, and relates to the technical field of automobile manufacturing equipment.The four-door two-cover flexible trimming die switching system comprises a lightweight trimming die unit and a switching mounting base, and the lightweight trimming die unit is detachably mounted on the switching mounting base through a robot grabbing device; the bottom of the light-weight edge rolling die unit is provided with a gun changing disc butt joint part, and the switching installation base is provided with an electric quick-plug device and a positioning locking device which are matched with the gun changing disc butt joint part. According to the invention, a split type framework of the lightweight trimming die unit and the switching mounting base is adopted, and a full-automatic die switching system is constructed through a robot automatic grabbing and gun changing disc rapid butt joint technology. A gun replacing disc butt joint part arranged at the bottom of the die body, an electric quick-plug device of the base and a positioning locking device form a standard interface, and a robot is matched for accurate grabbing and positioning, so that the die replacing process is optimized from traditional manual hoisting calibration to machine autonomous recognition butt joint.

The invention provides an object pose estimation method. The object pose estimation method comprises the steps of obtaining object structure features of a target object, hand pose features of a robot hand and contact area features of the hand in contact with the target object; fusing the object structure feature and the contact area feature to obtain a first fusion feature; fusing the hand posture feature with the contact area feature to obtain a second fused feature; fusing the first fusion feature and the second fusion feature to obtain a global fusion feature; and performing pose estimation based on the global fusion features to obtain a target pose of the target object. According to the embodiment of the invention, the accuracy of object pose estimation can be improved.

The application discloses a kind of food-grade silica gel triple-sealed high-precision bearings, it is related to bearing technical field, including bearing inner ring, bearing outer ring and the sealing ring of being installed in the inner side of bearing outer ring, the outer raceway both sides of bearing inner ring are equipped with one straight angle sealing groove with the cross section of 90 ° at two ends of shoulder or baffle respectively;The sealing ring includes metal framework and food-grade silica gel rubber coating integrally formed on the metal framework;The food-grade silica gel rubber coating extends towards the inner side of bearing inner ring and forms integrated three sealing lips, including two radial sealing lips with the contact of two radial side walls of the straight angle sealing groove, and one axial sealing lip with the contact of the axial bottom surface of the straight angle sealing groove.The application has the beneficial effects of material safety, no pollution, excellent sealing performance, high processing efficiency, the processing cycle is shortened by more than 40%, the leakage rate is reduced by more than 95%, and has wide application value.

The invention relates to the technical field of digital data processing, in particular to an energy efficiency service knowledge graph construction method based on an artificial intelligence large model. The knowledge graph construction method comprises the following steps: preprocessing data; knowledge modeling; labeling the data; constructing an entity relationship extraction model; model training; and knowledge extraction: deploying the trained model on line, and inputting to-be-extracted data into the model to carry out combined extraction of entities and relationships so as to obtain entities and relationships predicted by the model. Compared with the prior art, the entity relation joint extraction method provided by the invention has the advantages that data annotation and model training only need to be performed once, and the workload of data annotation and model training can be effectively reduced. By sharing the model parameters of the entity identification and relation extraction tasks, the model can automatically learn related information of the two tasks, and meanwhile, accumulative errors caused by assembly line work can be effectively reduced.

The present application belongs to the field of lithium battery SOC calculation, and particularly relates to a lithium battery SOC estimation method based on high-precision ampere-hour integral algorithm. The method comprises establishing an ampere-hour integral mathematical model and solving, and performing Taylor expansion on the solution; the average slope is obtained by means of linear combination of three-point slope by investigating the slope of two points in the interval, so as to ensure that the precision of the third-order Taylor expansion formula and the equivalent formula is the same. When the algorithm is executed, the charge state and the capacity value are initialized first, then the sampling time is obtained according to the sampling frequency, and then the relevant parameters are calculated according to the current sampling value, and finally the current SOC is calculated according to the specific value and the three slopes and is iterated in a loop. This method can improve the ampere-hour integral precision, and weaken the cumulative error and SOC jump.

The application discloses a water meter core insertion device with visual correction function, a pneumatic gripper adopts multi-finger, the inner side of the finger and the contact surface of the core adopt arc-shaped profiling design, the finger clamping pressure is adjustable, and the deformation of the core or the falling of the core caused by improper clamping force in the grabbing process is avoided. The shell clamp device adopts right positioning cylinder, clamping cylinder, centering cylinder, left positioning cylinder, or front and rear positioning block clamping positioning, and adopts centering shaft correction clamping mode, which further improves the positioning accuracy of the shell, and ensures that the core is smoothly inserted in place. The visual camera takes pictures to calculate and correct the angle operation process, which is to calculate the correct rotation correction angle, and the finger clamping position avoids the protruding positioning surface on the core. Each functional device adopts compact modular design, which not only realizes single machine application and multi-station production line online use, but also realizes compact layout, reduces equipment motion beats, reduces equipment system cumulative error and the like.

The invention discloses a complex dynamic industrial visual positioning method and system based on multi-agent collaboration, and relates to the technical field of machine vision, distributed visual agent nodes are deployed in an industrial dynamic operation scene, and node microsecond-level time synchronization, space joint calibration and communication networking are completed; the nodes synchronously collect industrial images according to a preset time sequence, and a standardized data set is generated through preprocessing; dynamic target feature extraction is completed based on an improved deep convolutional neural network, and cross-node feature matching is completed in combination with epipolar constraints; a multi-view geometric constraint model is constructed to solve the three-dimensional pose of the target, and a positioning result is generated through dynamic interference compensation optimization; and outputting a positioning result, completing closed-loop feedback, and dynamically adjusting node parameters to realize multi-agent collaborative global optimization. According to the invention, the precision and the anti-interference capability of visual positioning in a complex dynamic industrial scene are improved, the multi-node load balancing is optimized, and the visual positioning operation requirements of various industrial dynamic scenes can be met.

The invention relates to an integrated camera control structure in the technical field of camera lens control. The integrated camera control structure comprises a lens gear, a driving motor and an axial limiting ring, the lens gear is of an annular gear structure and is suitable for being meshed with a camera lens. The driving motor is located below the lens gear, a motor gear is arranged above the driving motor, the motor gear is meshed with the lens gear, and the driving motor drives the motor gear and the lens gear to rotate; the axial limiting ring is of an annular structure and located above the lens gear, the axial limiting ring and the lens gear are located on the same axis, and the axial limiting ring is fixed to the camera lens and can do up-down telescopic motion along with the camera lens. The camera integrated control structure is close in cooperation, and the lens adjusting precision and the transmission efficiency are high.

This invention relates to the field of robotics, and in particular to a multi-segment parallel crawling robot for multi-curvature pipe corridors. The robot comprises at least three segments connected in series, with adjacent segments sharing a platform. Each platform has two symmetrically arranged mechanical legs for gripping pipes. Each segment includes a parallel mechanism connecting two platforms, which drives relative movement between the two connected platforms. This invention employs a multi-segment parallel connection method, increasing the overall structural freedom and reliability. It can flexibly bend according to changes in pipe curvature, thus smoothly passing through continuous curved pipe sections, enabling efficient passage and stable operation of the robot in complex pipeline environments.

The invention relates to a small four-direction angle adjusting mechanism, belongs to the technical field of laser control and detection, and solves one of the problems that an existing laser emitting device detection device is tedious and complex in operation process, large in equipment size or large in weight, the adjusting precision seriously depends on the hand feeling of an operator, or the detection efficiency is low and the like. The mechanism comprises an orientation driving mechanism, a high-low driving mechanism, a supporting rotating mechanism, a zero position self-checking mechanism and an adaptive assembly, the orientation driving mechanism, the high-low driving mechanism and the supporting rotating mechanism achieve four-direction angle adjustment of the mechanism, and the adaptive assembly rapidly clamps detectors of different sizes. According to the invention, rapid and repeatable fine adjustment of four-direction angles of the optical system of the detector of the laser emitting device can be realized, the structure is compact and portable, the detection efficiency is high, and the adaptability is strong.

The application discloses an image processing method, device, equipment, storage medium and computer program product. The method comprises the following steps: acquiring a rolling shutter (RS) frame collected by an event camera and event data corresponding to the RS frame; encoding the RS frame and the event data through an encoding unit of an image processing model to obtain space-time implicit representation (STR) data corresponding to the RS frame; embedding an exposure time of a global shutter (GS) frame corresponding to the RS frame into the STR data through a time embedding unit of the image processing model to obtain a time tensor corresponding to the RS frame; and performing pixel-by-pixel decoding on the time tensor corresponding to the RS frame and the STR data corresponding to the RS frame through a decoding unit of the image processing model to generate a GS frame. The scheme disclosed by the application can improve the definition of the GS frame recovered from the RS frame.

The application provides a high-precision low-drift wide-range three-dimensional point cloud map construction and repositioning method, in which point cloud information and IMU information collected based on a three-dimensional laser radar are used for inter-frame point cloud registration to obtain the relative pose between key frames; considering the robot GPS information constraint, the closed loop detection constraint and the laser radar odometry constraint, the relative pose between the key frames is obtained and optimized to complete the three-dimensional point cloud map construction of the robot; meanwhile, the key frame descriptor is constructed by considering the key frame point cloud feature information, so that the inter-frame similarity is judged and the repositioning of the robot in the three-dimensional map is completed; and the cumulative error caused by the inter-frame registration of the point cloud is fully considered when the three-dimensional point cloud map construction proposition is constructed, the error caused by the non-linear optimization of the laser radar odometry is constrained by the GPS information and the closed loop detection constraint, and the globally consistent map is constructed, so that the quality of the three-dimensional point cloud map construction of the mobile robot can be greatly improved.

This invention discloses a pre-distribution assembly, comprising: an inlet diffuser, which includes: a sleeve consisting of an inner cylinder and an outer cylinder arranged coaxially, with a liquid-holding zone between the inner and outer cylinders, and multiple strip-shaped holes on the upper sidewall of the inner cylinder; an annular bottom plate coaxially connected to the bottom ends of the inner and outer cylinders, the annular bottom plate having multiple through holes and / or multiple slits; a top cover, with a sealing cap located at the upper end of the inner cylinder; a rotating shaft coaxially passing through the inner cylinder, the lower end of the rotating shaft passing through the annular bottom plate; a spiral blade disposed on the rotating shaft and positioned opposite to the strip-shaped holes; a conical splash plate linked to the lower end of the rotating shaft; and a stepped distribution plate coaxially disposed below the inlet diffuser, in a stepped shape with a lower center and higher periphery. This invention also discloses a hydrogenation reactor. The pre-distribution assembly of this invention, through the cooperation of the inlet diffuser and the stepped distribution plate, achieves uniform distribution of gas and liquid phase materials across the entire cross-section of an ultra-large-scale hydrogenation reactor.

The embodiment of the invention provides a signal processing device and method and a chip, and the device comprises a signal generation module, a counting module, a register module, a control module and an audio processing module, and the signal generation module is used for generating a first clock signal according to a first frequency control word; the counting module is used for counting a first number of first clock signals in N preset clock periods by adopting a circular windowing mode, and storing the first number into the register module; the control module is used for acquiring the first number from the register module, comparing the first number with a preset clock period to obtain a comparison result, and determining an adjustment strategy corresponding to the first frequency control word according to the comparison result, so that an input clock signal and an output clock signal of the audio processing module are kept in a fixed phase; the problem that the deviation of the output clock signal in one direction becomes larger and larger is avoided, and the problem of sound interruption is solved, so that high-quality signal transmission of an audio system is realized.

The invention relates to the technical field of building curtain wall construction, and discloses a multi-direction adjustable aluminum-based metal curtain wall installation method which comprises the following steps: determining the plate division size and the reinforcing rib layout based on wind pressure load and finite element analysis, designing a multi-dimensional adjustable hanging system, generating design parameter data, and setting the design parameter data as the design parameter data; on the basis of the design parameter data, aluminum-based metal plates and reinforcing ribs are machined through numerical control equipment, and machined curtain wall plate data are generated; design parameter data, processed curtain wall plate data and on-site measurement reference data are systematically integrated, so that a uniform and accurate data reference is provided for keel installation and plate hanging, and the definition and consistency of a whole-process construction basis from surveying and setting-out to final fastening are ensured; the initial in-position accuracy of integral installation of the curtain wall is improved, and accumulative errors caused by data disjunction are reduced.

The invention provides a spatial two-rotation and one-movement three-degree-of-freedom parallel mechanism. The spatial two-rotation and one-movement three-degree-of-freedom parallel mechanism comprises a fixed platform, a movable platform and three main branch chains. Each main branch chain supports the movable platform to realize two-rotation and one-movement three-degree-of-freedom movement and comprises a first branch chain and a second branch chain; the first end of the first branch chain and the first end of the second branch chain are in spherical connection with the movable platform through equivalent spherical hinges, the second end of the first branch chain is rotationally connected with the fixed platform, and the second end of the second branch chain is rotationally connected with the fixed platform through a rotating pair. The parallel mechanism is high in bearing capacity, high in rigidity, high in fault-tolerant capability, high in reliability and high in stability, and stable supporting and stable movement of the movable platform can be achieved; spherical surface connection is achieved through equivalent spherical hinges, two-rotation and one-movement three-degree-of-freedom flexible movement is achieved only through a rotary hinge mode, and the parallel mechanism has the advantages of being large in working space, small in accumulative error, timely in motion response and high in motion output efficiency and precision.

A multi-circle angular displacement measuring device based on a single gear structure comprises a shell, a rotating shaft, a driving wheel, a driven wheel, magnetic steel, a deep groove ball bearing, a circuit board and a Hall chip. Three deep groove ball bearings are respectively mounted in holes in the upper and lower end faces of the bottom of the shell; the small-section end of the rotating shaft penetrates through the first deep groove ball bearing and the second deep groove ball bearing, and the root of the rotating shaft is inserted into a mounting hole of the driving wheel. The driving wheel is fixed with the inner ring of the deep groove ball bearing II; the first magnetic steel is installed in a groove of the driving wheel. A driven wheel is fixed with an inner ring of the deep groove ball bearing III, and teeth of the driven wheel are meshed with teeth of the driving wheel; the second magnetic steel is installed in a groove of the driven wheel. The circuit board is arranged on a boss in the shell; the first Hall chip and the second Hall chip are installed on the circuit board and axially correspond to the first magnetic steel and the second magnetic steel respectively. The device is high in vibration resistance, small in accumulative error, high in positioning precision, absolute in mechanical measurement, capable of achieving real absolute multi-circle angle measurement and capable of being used for application needing to track multiple rotation periods.

This invention relates to the field of robot intelligent control technology, and more particularly to a robot operation method and system. Relying on a high-level visual language semantic planner to output sub-task labels, low-level action strategies can continuously call semantic labels to output actions without waiting for the completion of a single high-level inference, thus improving the continuity of action execution. This invention injects sub-task semantics into the visual feature space, enabling the robot to focus on corresponding visual targets and spatial relationships at different operation stages, improving the accuracy of multi-stage task action matching. This invention sets up a semantic perception temporal integration mechanism, weightedly fusing old and new action blocks during semantic label switching and semantic rollback, mitigating abrupt changes in cross-sub-task actions and improving the stability of contact operation tasks. When detecting robot operation anomalies or sub-task state reversals, this invention can redetermine and restore the planned actions for the sub-task stage, enhancing the autonomous recovery capability of long-term tasks.

The application provides a new energy battery base rod piece machining equipment, which comprises a rack, a clamping and positioning mechanism, a machining main shaft and a tool magazine. The clamping and positioning mechanism comprises a clamping support table, two end positioning assemblies arranged on the clamping support table, a plurality of intermediate positioning assemblies arranged between the two end positioning assemblies and a rotating mechanism for driving the clamping support table to rotate. The two end positioning assemblies are used for clamping two ends of a workpiece to achieve axial fixation, and the intermediate positioning assemblies are used for clamping the workpiece to achieve radial fixation. The clamping and positioning mechanism is provided with a plurality of clamping and positioning mechanisms and is arranged side by side along a second direction. The machining main shaft is movably arranged on the rack along a first direction, a second direction and a third direction. The tool magazine comprises a tool storage rack and a plurality of tool storage sub-racks arranged on the tool storage rack. The tool storage rack is movably arranged on the rack along the first direction. The new energy battery base rod piece machining equipment can be used for batch efficient machining of metal long rod pieces of different specifications.

The application discloses a machining tool and method for positioning and machining dovetail grooves of a large POD propeller rotor shaft. The tool comprises a first and a second rotary bracket, a hydraulic pressure assembly, a mechanical indexing positioning assembly and an angle encoder detection assembly, which are arranged along the rotor shaft in an axial direction. During machining, two rotary reference outer circles are formed on the rotor shaft and supported on the two rotary brackets, and then the theoretical angle of one end is determined by a dividing disc and a positioning pin, and the actual angle of the other end is detected by an angle encoder. After the angle error meets the requirements, the dovetail grooves are pressed and milled. The scheme can reduce the cumulative error between grooves of a long shaft, and improve the circumferential distribution accuracy and the consistency of the pressing.