481 results about "Omni directional" patented technology

A broadband omni-directional bicone antenna. The antenna can comprise conductive surfaces of conical voids provided within a solid dielectric structure. The outside surface of the solid structure can support a radio frequency (RF) lens geometry operable for beam forming. The beam forming can modify the elevation pattern of the electromagnetic radiation from the bicone antenna. The solid dielectric structure may be machined or molded from a single piece of material. The conical voids provided within the solid structure can be metallized to provide conductive bicone radiators. The outer surface beam shaping lenses can be zoned or continuous and can provide elevation patterns with increased gain, cosecant squared falloff, or various other patterns. The beam shaping lens may be formed from any low-loss dielectric. Alternatively, the lens may be formed from a less dense material such as dielectric foam that can support radial conductive beam forming vanes.

The invention discloses a re-parameterization unmanned aerial vehicle target detection method based on multi-core fusion and omnidirectional connection, and the method comprises the following steps: obtaining an unmanned aerial vehicle aerial image data set, and constructing a detection model comprising a backbone network, a feature fusion network and a deformable task decoupling detection head; the backbone network extracts multi-scale and multi-direction features by six parallel paths in a training stage through a wide-branch re-parameterization convolution module, and re-parameterization is carried out in a reasoning stage to obtain single-path convolution; the feature fusion network performs down-sampling through a spatial deep convolution module and reduces spatial information loss, and realizes cross-stage omnidirectional feature interaction and double attention enhancement in combination with an omnidirectional kernel cross-stage partial connection module; and the deformable task decoupling detection head decouples the classification and regression features, optimizes feature expression, weights the classification features and then performs aggregation decoding. The model is trained to be used for a test set to output a detection result, detection precision and reasoning efficiency can be balanced, and the model adapts to a complex aerial photography scene of an unmanned aerial vehicle.

The invention belongs to the technical field of unmanned aerial vehicles, and provides an omni-directional vision positioning system and method and an unmanned aerial vehicle, the omni-directional vision positioning system is installed on a fuselage of the unmanned aerial vehicle, and the omni-directional vision positioning system comprises an image acquisition device used for acquiring a real-time scene image in an omni-directional range in the flight process of the unmanned aerial vehicle; the inertial measurement sensor is connected with the image acquisition equipment and is used for synchronously acquiring motion state data of the unmanned aerial vehicle; the navigation system is connected with the image acquisition equipment and the inertial measurement sensor and is configured to execute at least one of distortion correction processing and feature extraction processing on the real-time scene image to obtain a processed image; and according to the motion state data and the processed image, determining current pose information of the unmanned aerial vehicle in a space coordinate system constructed by taking a take-off point of the unmanned aerial vehicle as an original point. The positioning precision and robustness of the unmanned aerial vehicle in a complex environment are improved, and the requirement of high-precision autonomous flight is met.

According to the mass center self-balancing chassis system for U-rib weld defect detection and the control method, the problem that when detection equipment moves on a U-rib curved surface, the precision is reduced due to mass center deviation and view field inclination can be solved. The system integrates a Mecanum wheel omni-directional moving unit, four groups of electric telescopic supporting legs, a two-axis holder and an ECU (electronic control unit), and has the innovation point that dynamic mass center balance and posture self-correction are realized through rigid-flexible combined supporting leg design and multi-modal sensing fusion. The ECU adopts an improved Canny algorithm to process the image, extracts a U rib edge point set and fits a reference center; in combination with pressure and tilt angle sensor data, a real-time mass center is solved through a moment balance formula, and then an improved self-adaptive PID algorithm is used for driving the supporting legs to stretch out and draw back, so that the mass center returns to a reference center; the two-axis holder compensates the view field angle according to the inclination angle data to ensure that the view field angle is perpendicular to the weld joint. Through cooperation of a mechanical structure and an algorithm, the system can control centroid offset to be smaller than or equal to 5 mm and view field inclination to be smaller than or equal to 1 degree, the detection stability and precision are remarkably improved, and the algorithm is easy to deploy.

The invention discloses a Mecanum wheel omnidirectional robot anti-interference control method and system, and relates to the technical field of robot anti-interference control, and the method comprises the steps: collecting multi-source original data, and carrying out the preprocessing; multi-source data fusion is carried out, a current operation condition is identified based on a fusion result, and an augmented state is generated; calculating a baseline control instruction; inputting to the reinforcement learning agent, and outputting a corresponding compensation instruction; adding the baseline control instruction and the compensation instruction to obtain a candidate control instruction; inputting to the security constraint projector, and correcting to obtain a final control instruction; issuing to a motor driver of the Mecanum wheel for execution; and actual state data of the executed robot is collected as execution feedback data, and the execution feedback data and a new target instruction are used for generating standardized observation data at the next moment. According to the invention, cross-working-condition high-precision, fast-recovery, low-energy-consumption and compliant-constraint omnidirectional chassis anti-interference control is realized, and the tail robustness and the system safety and reliability are improved.

The invention discloses a multi-frequency omnidirectional unmanned aerial vehicle active defense system and method, and the method comprises the steps: obtaining defense region parameters, and constructing a navigation signal generation task library based on the coverage requirements of GPSL1 and Beidou B1 / B2 / B3 frequency bands; resonance interference points are determined through carrier locking resonance analysis, Doppler frequency shift traps are designed, frequency shift fragility features are extracted in combination with C / A code recombination processing, and a targeted interference enhancement domain is constructed; the method comprises the following steps: respectively carrying out constellation topology reconstruction on a GPS (Global Positioning System) and a Beidou system, generating false orbit parameters through ephemeris data tampering, and forming a mixed defense signal domain by utilizing carrier phase interleaving processing; analyzing signal power to determine strong and weak interference areas, and setting graded attenuation parameters to construct a graded signal attenuation field; the attenuation field and the interference enhancement domain are fused to generate dynamic simulation parameters, false orbit parameters are adjusted in real time to generate a coordinate input mode, a high-dynamic navigation signal is reconstructed, a satellite working mode is selected, an L1 / B1 / B2 / B3 multi-band radio frequency output instruction is finally generated, and all-dimensional intelligent deception jamming of the unmanned aerial vehicle navigation system is achieved.

The invention relates to a multi-user holographic sand table collaborative resolving system and method based on 360-degree omni-directional perception gesture recognition. The method comprises the following steps: acquiring continuous multi-frame multi-view hand image data; hand instance separation and key point extraction are carried out on the multi-view hand image data of each frame, and spatial 3D coordinates of 21 key points are independently solved for each hand instance; based on the space 3D coordinate sequence of the continuous frames, space-time action features are extracted, and a gesture template library is matched to recognize gesture types and motion parameters; analyzing the gesture type, matching an interaction intention in combination with a predefined gesture-instruction mapping table, and generating an interaction control instruction according to a matching result; generating a three-dimensional holographic sand table image cooperatively operated by multiple persons in real time according to the interaction control instruction; through a multi-hand parallel processing technology, different hand instance feature extraction paths are separated, the problem of multi-hand ID jump is solved, occlusion area coordinate prediction is optimized, and interactive use experience is improved.

The invention discloses a conference sound reinforcement system based on an AI intelligent algorithm and 360-degree omni-directional noise reduction, and relates to the technical field of audio signal processing, the conference sound reinforcement system comprises a conference management center, the conference management center is in communication connection with the following modules: a multi-sound-source sensing module used for capturing 360-degree omni-directional sound field information in a conference environment and constructing a sound source space distribution model; according to the invention, the omnidirectional microphone array unit covers all directions of a conference space, synchronously collects audio data streams, eliminates the limitation of a conventional unidirectional microphone, combines the sound source positioning space mapping unit, constructs a sound source space distribution model based on the time difference of arrival and the phase difference through a deep learning model, and improves the sound source positioning accuracy. The azimuth angle, pitch angle and distance parameters of the sound source are accurately analyzed, the position of the sound source is mapped to a virtual space coordinate system, a dynamically updated 3D sound source distribution diagram is generated, the position change and intensity distribution of the sound source are reflected in real time, and the positioning precision in a complex acoustic environment is remarkably improved.

A communication method including emitting laser pulses as a signal using one or more transmitters at an atmospheric point. Plasma is generated at the atmospheric point to radiate omnidirectional light. The plasma from the atmospheric point is detected using one or more receivers. The one or more receivers have a direct line-of-sight to the atmospheric point to receive the signal from the one or more transmitters via the omnidirectional light generated by the plasma. The signal is demodulated from the transmitted plasma at each receiver, thereby transmitting information encoded on the signal.

Omni-directional, horizontally polarized antennas are provided. According to one aspect, an omni-directional, horizontally polarized antenna includes a first reflector and a second reflector separated from the first reflector. Disposed between the first and second reflectors are a planar array of radially directed horizontally polarized antenna elements and a planar feed network configured to feed the antenna elements of the planar array. A coaxial feed structure configured to communicate electrical signals to or from the feed network is provided.

The invention relates to the field of mobile robots, in particular to a four-wheel omni-directional robot chassis navigation control device and method.The four-wheel omni-directional robot chassis navigation control device comprises a chassis structure and a driving system, a radar and a mechanical arm are arranged on the chassis structure, and omni-directional wheels are installed in the centers of the four side walls of the chassis structure; the axes of the omni-directional wheels are parallel to the edge of the chassis structure, the axes of the adjacent omni-directional wheels are perpendicular to each other, the diagonal direction of the chassis structure serves as the advancing direction of the robot, the advancing direction of the omni-directional wheels and the overall advancing direction of the chassis structure form an included angle of 45 degrees, and the driving system comprises four motors with Hall encoders. The four motors independently drive the independent omnidirectional wheels, the upper computer and the lower computer conduct instruction issuing and state feedback through UART serial port communication, and the problems that at the present stage, steering is conducted by depending on steering engine driving or positioning is conducted through a single sensor, the structure is complex, the anti-interference capacity is limited, and operation is complex are solved. And high precision is difficult to maintain in a dynamic environment.

The invention discloses a robot environment sensing and positioning system based on an ROS. Effective data of multiple sensors can be efficiently processed, and the timestamps are unified for data fusion. According to the invention, an NUC11 industrial personal computer receives sensor data of an inertial sensor, a laser radar and an industrial camera and transmits the sensor data to a corresponding algorithm module for processing; a grid map is established after the surrounding environment is sensed; the speed and the direction are sent to the STM32 single-chip microcomputer through serial port communication, the STM32 single-chip microcomputer calculates the speed and the direction and then converts the speed and the direction into current values, current value data are sent to the motor through CAN communication, the Mecanum wheels are driven to move in all directions, and the navigation robot capable of achieving self-positioning and omni-directional movement is achieved. According to the omni-directional moving efficient wheeled robot, Mecanum wheels are innovatively combined to achieve omni-directional moving, a universal positioning and navigation robot chassis frame system is constructed, and the omni-directional moving efficient wheeled robot can become a universal solution for a cleaning robot, a meal delivery robot, an express delivery robot and the like in combination with different application scenes.

A modular omnidirectional robotic system for navigating nonplanar structures including pipes, poles, chimneys, and trees. The system comprises multiple modules connected by hinges in an open C-shape configuration. Each module contains an omnidirectional drive mechanism with compliant treads, motors, batteries, and controllers. Adjustable tensioning mechanisms adapt to different structure diameters and enable both external and internal operation. The open configuration allows navigation around complex geometries including T-junctions, branches, and sharp bends by aligning the open section with obstructions. Sharp turns are negotiated using the leading edge of the drive mechanism as a pivot point. A control system coordinates module movement using sensors. The system operates on ferrous and non-ferrous materials, features an adaptable payload area for tools and sensors, and can be scaled by adding modules for larger structures.

The invention discloses a steel plate quality detection operation mobile robot platform and an operation method thereof, and the platform comprises an omnidirectional mobile robot chassis which is used for detecting walking and carrying configuration hardware on a to-be-detected thick steel plate; the quality inspection sensor module is arranged at the side part of the chassis of the omnidirectional mobile robot and is used for acquiring quality inspection data of the to-be-detected thick steel plate; the industrial personal computer module is arranged on the chassis of the omni-directional mobile robot and is used for storing and processing quality inspection data; the wireless transceiving module is arranged on the chassis of the omni-directional mobile robot and is used for transmitting and interacting quality inspection data and instructions; the navigation positioning sensor module is arranged at the upper part of the chassis of the omni-directional mobile robot and is used for scanning the surrounding environment; the safety module is arranged on the chassis of the omni-directional mobile robot and is used for anti-falling detection; and the external auxiliary device is used for assisting the detection operation of the steel plate quality detection operation mobile robot platform. The invention solves the problems of heavy manual off-line quality detection task and low efficiency and reliability in the large-scale production period of the existing thick steel plate.

The invention relates to the technical field of wireless communication perception, in particular to a user perception method based on a multi-layer intelligent omnidirectional metasurface, and proposes to realize accurate perception of reflection and refraction user orientations by utilizing heterogeneous characteristics of propagation paths on two sides of a multi-layer intelligent omnidirectional metasurface structure. Based on the architecture, a user reflection and refraction type judgment algorithm based on uplink pilot frequency information is provided, and the user reflection and refraction types are determined based on the heterogeneous characteristics of user channels on the two sides of the multi-layer intelligent omni-directional metasurface and based on least square optimization problem solving and target function size comparison. The propagation path angle estimation algorithm specifically comprises the following steps: firstly, based on uplink pilot frequency information, designing a channel estimation algorithm comprising alternating optimization, constant-rank manifold optimization, convex relaxation and other methods to estimate and obtain a channel matrix by utilizing channel constant-rank and angular domain sparse characteristics; and finally, based on a user type determination result, obtaining user angle information based on each channel matrix by using an orthogonal matching pursuit algorithm.

The invention discloses an omnidirectional space group target detection method based on micro-nano satellite group cooperation. The method comprises the following steps: carrying out observation configuration initialization on a micro-nano satellite group; establishing an intelligent detection model, and initializing observation star and decision star detection parameters and the intelligent detection model; the observation star performs target detection through the intelligent detection model to form a single-frame detection result; the decision-making star receives the single-frame detection result of each observation star to obtain summarized single-frame detection information; the decision star performs cross-view collaborative target association according to the summarized single-frame detection information, and updates a potential target list; and the decision star performs multi-view multi-frame collaborative target detection according to a single-frame detection result and the potential target list, updates a suspicious target list, and finally obtains suspicious target information. The method has the capability of autonomous information collaboration of seven micro-nano satellites, realizes omnibearing coverage target detection of a 4pi space above 200km, and has the capability of cross-view target association and multi-angle group target distinguishing during collaborative detection.

The invention specifically discloses an omni-directional human body activity recognition method and device based on an HDMR, a computer and a medium, and relates to the technical field of radar signal processing. According to the method, an HDMR-based synthetic data generation algorithm is utilized, data expansion can be carried out on collected samples in a radar zero-degree observation angle direction and a small number of samples in other non-zero-degree observation angle directions, and high-quality training data containing all observation angle directions are generated; the objective of the invention is to solve the problem of angle sensitivity of a monostatic radar in omni-directional human body activity recognition. In addition, according to the invention, the dynamic time warping distance DTWD is adopted to measure the similarity between the synthetic sample and the real sample, so that the quality of the synthetic sample is evaluated. And finally, inputting the synthesized sample data in different observation angle directions into a CNN classifier based on ResNet50 for training so as to realize omnidirectional human body activity recognition.

The invention provides a double-frequency navigation high-gain omnidirectional receiving antenna system, and relates to the technical field of communication navigation, and the system comprises an array module which is used for constructing a multi-dimensional three-dimensional antenna array in multilayer staggered arrangement, and forming a three-dimensional structure with omnidirectional coverage capability, so as to obtain a double-frequency-band signal adaptive to a navigation device; the separation module is used for identifying a satellite direct signal and a multipath signal formed by ground reflection by utilizing different polarization states through combination of a spatial domain and a polarization domain aiming at the dual-band signal so as to obtain a stable direct signal; and the control module is used for dynamically adjusting the signal weight of each unit in the array based on the stable direct signal, focusing energy in the incoming wave direction of the satellite signal to form a high-gain main beam, and realizing omnidirectional coverage of the array signal coverage capability. According to the invention, the dual-frequency navigation signal is received, high-gain omnidirectional coverage is realized, the signal receiving strength and stability are improved, and the positioning error is reduced.

The invention discloses a wind power plant arrangement optimization method and device, equipment and a storage medium, and relates to the technical field of wind power plant design and optimization, and the method comprises the steps: obtaining environment turbulence intensity and a fan thrust coefficient, and constructing an omnidirectional wind resource map; inputting the environmental turbulence intensity and the fan thrust coefficient into a TurbOPark wake flow model, calculating wake flow loss and obtaining the inflow wind speed of each fan under each wind direction and wind speed combination; taking the wind frequency in the omnidirectional wind resource map as a weight, and carrying out weighted summation processing on single-machine power obtained by converting the inflow wind speed through a power curve to obtain a target function of maximizing the total power generation of the whole field; and by taking the boundary polygon coordinates and the minimum spacing requirement as constraints, randomly disturbing the fan coordinates and comparing the target function before and after disturbance to iteratively update the fan position, the target fan coordinates and the whole-field total generating capacity corresponding to the target fan coordinates are obtained, and the optimization precision and the engineering efficiency are remarkably improved.

A cart transport apparatus according to an embodiment includes: a driving part including a plurality of omnidirectional wheels; a driven part including a fixing part configured to fix a cart, the driven part being coupled to the driving part; an information acquisition part configured to read information regarding the cart from an information storage part provided in the cart; and a controller configured to calculate, from the information regarding the cart read by the information acquisition part, a turning center of the cart transport apparatus in a state in which the cart is fixed.

The invention discloses an omni-directional mobile robot pose adjustment method based on a laser radar, and the method comprises the steps: collecting and processing multi-frame laser radar data through a system to extract environment features in a working point creation stage after a robot is precisely guided to a target pose, and carrying out the correlation storage of a working point identifier, a global pose and standard feature data after verification. And when the robot navigates to a position near a working point, triggering an accurate adjustment process: calling pre-stored standard features by a system, collecting laser radar data in real time to perform same filtering and feature extraction processing, and performing iterative calculation through a point cloud matching algorithm to obtain X and Y direction and course angle deviations between a current position and a standard position. And finally, carrying out pose correction, and carrying out real-time cyclic feedback until the deviation reaches a millimeter-level precision threshold value. According to the invention, the dependence of a traditional two-dimensional code accurate positioning scheme on illumination, cleanliness and entity marking is overcome, and the method has the advantages of high precision, strong environmental adaptability and high reliability.

The invention provides an outdoor cross-modal robust positioning and navigation method for extreme severe weather, and relates to the field of online positioning and navigation of mobile robots. The method comprises the following steps: acquiring and preprocessing data; generating a path point set based on a CMR network, perfecting the path point set through local trajectory fitting and error optimization, and constructing a path point relative pose measurement-topology hybrid map; motion prior estimation is carried out based on Doppler velocity, effective matching prior estimation is generated in combination with LiDAR odometer information, cross-modal pose estimation is carried out through a CMR network, and a pure tracking strategy and PD controller path point tracking are adopted; and geometric and intensity alignment is carried out on the forward 4D millimeter wave radar point cloud and the omnidirectional 3D laser radar point cloud through a CMR network, a rotation matrix and a translation vector are output, and accurate positioning is completed. According to the invention, the sensing and positioning defects in extreme weather in the prior art are overcome, and the positioning precision and robustness are improved.

The invention discloses an underwater vehicle wireless electric energy transmission system with an omnidirectional anti-offset function. The transmission system comprises a magnetic coupler receiving assembly and a magnetic coupler sending assembly. The sending assembly is used for providing a charging magnetic field for the receiving assembly. The magnetic coupler transmitting assembly comprises a stepped coil and an anti-series coil, and the stepped coil is located in the anti-series coil; the magnetic coupler receiving assembly comprises a plurality of receiving coils, magnetic cores are arranged in the receiving coils, and the receiving coils are of solenoid structures and are wound around the magnetic cores; the number of the receiving coils is five, one receiving coil is arranged in the middle of the H shape, and the other four receiving coils are fixed to the ends of the H shape. According to the invention, the magnetic field intensity of the surrounding area of the wireless electric energy transmission system is improved, and the mutual inductance between the transmitting coil and the receiving coil is ensured to be stable when the omnidirectional position offset of the aircraft occurs.

The invention relates to the technical field of automobile chassis dynamics control, and discloses a hydrogen energy heavy truck omnidirectional degree-of-freedom optimization method, which comprises the following steps: acquiring multi-source sensor data and hydrogen fuel cell system state data of a hydrogen energy heavy truck, and executing time synchronization and noise preprocessing; carrying out real-time identification on the key state parameters of the vehicle by utilizing a nineteen-degree-of-freedom vehicle dynamics model and a self-adaptive filtering algorithm; a layered omnidirectional coordination control architecture is constructed and comprises a main control layer, a scheduling layer and an execution layer, and the main control layer generates a global motion control target; carrying out multi-objective optimization solution on the scheduling layer, and calculating to obtain an expected acting force distribution instruction; and the execution layer converts the instruction into a bottom layer execution control signal and drives the subsystems to cooperatively act so as to realize omni-directional degree-of-freedom optimization control. According to the invention, the limitations of poor coordination and difficulty in multi-target optimization caused by independent control of subsystems of a traditional chassis can be broken through, and accurate control of the hydrogen energy heavy truck under complex working conditions is realized.

The invention discloses a double-AGV (Automatic Guided Vehicle) linkage omnidirectional dynamic path planning method based on a dynamic window method, and the method designs a real-time dynamic path planning scheme based on an omnidirectional differential composite motion model by utilizing the omnidirectional movement characteristic of Mecanum wheels. The method comprises the following steps: firstly, introducing a Y-direction linear velocity in a velocity sampling stage, and constructing an omnidirectional motion model to improve the omnidirectional motion capability; and meanwhile, by combining angular velocity sampling, dynamic adjustment of the attitude of the AGV is realized, an omnidirectional differential composite motion model is constructed, and the omnidirectional motion flexibility is enhanced. And secondly, fusing an artificial potential field method, designing an omnidirectional motion evaluation sub-function, carrying out multi-dimensional comprehensive evaluation on candidate trajectories, screening out an optimal trajectory facing a target point, and effectively improving the efficiency and safety of path planning. And finally, in combination with the contour parameters of the double-AGV linkage system, ensuring that the pilot AGV and the following AGV can be prevented from colliding with obstacles in the environment in the navigation process. According to the method, the transportation efficiency and the system stability are remarkably improved.

The invention discloses an omni-directional transfer device for vertical recovery of a rocket body, and relates to the technical field of rocket recovery, the omni-directional transfer device for vertical recovery of the rocket body comprises a plurality of automatic positioning and locking mechanisms, and the automatic positioning and locking mechanisms are arranged on the circumferential outer side of the rocket body; the automatic positioning and locking mechanism comprises a vehicle body, wherein a sliding block is slidably arranged on the top surface of the vehicle body; the vertical electric cylinder is rotatably arranged on the top surface of the vehicle body, and the upper end part of the vertical electric cylinder is connected with the locking hook through a first piston rod; and the first end of the overturning electric cylinder is rotatably connected with the sliding block, and the second end of the overturning electric cylinder is rotatably connected with the locking hook through a second piston rod. The omni-directional transferring device for vertical recovery of the rocket body solves the technical problems that an existing rocket body locking process is tedious in operation, needs manual intervention, has high requirements for operation space, is high in potential safety hazard, is long in consumed time, cannot be flexibly adjusted according to the diameter of the rocket body, and is small in application range.

The invention discloses a safe and hidden quasi-omnidirectional full-duplex laser communication device, system and method, and solves the technical problems that the position and orientation of an existing laser communication system have a leakage risk, and the existing laser communication system is difficult to adapt to a wide-angle communication application scene due to a small receiving and transmitting view field. The laser communication system comprises at least one active communication device and at least one passive communication device which are paired with each other, and the active communication device comprises a rotary table unit and a body arranged on the rotary table unit; the body is provided with a first main control unit, an optical signal generation unit, a divergence angle adjustment unit, an optical field regulation and control unit, a first optical receiving unit and a first acquisition processing unit; the passive communication device comprises a second optical receiving unit, a second collecting and processing unit and a second main control unit. According to the invention, the position, orientation and data leakage risks of two communication parties are reduced, and double-end full-duplex communication is realized at the same time.

1. The name of the design product: loudspeaker. 2. The use of the design product: a 360-degree omnidirectional loudspeaker. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: design 1 perspective view. 5. Design 1 is designated as the basic design.

The invention provides a trajectory tracking control method and system for a high-load omni-directional AGV, the omni-directional AGV is driven by a plurality of steering wheel mechanisms which are independently steered and driven, and the trajectory tracking control method is characterized by comprising the following steps: acquiring vehicle state information and expected trajectory information of the omni-directional AGV in real time; on the basis of the vehicle state information and the expected track information, core parameters of a model prediction controller are optimized online through a chaos simulated annealing algorithm, and the core parameters comprise a state weight matrix, a control increment weight matrix, a prediction time domain and a control time domain; wherein the model prediction controller performs state prediction based on a longitudinal-transverse decoupling dynamic model adaptive to a high-load working condition; performing rolling horizon optimization through the model prediction controller by utilizing the optimized core parameters, and solving to obtain an optimal control increment; and converting the optimal control increment into a driving instruction for a multi-steering-wheel mechanism, and controlling the omnidirectional AGV to track an expected trajectory.

A method and device for performing a perception task are disclosed. The method and device incorporate a dense regression model. The dense regression model advantageously incorporates a distortion-free convolution technique that is designed to accommodate and appropriately handle the varying levels of distortion in omnidirectional images across different regions. In addition to distortion-free convolution, the dense regression model further utilizes a transformer that incorporates an spherical self-attention that use distortion-free image embedding to compute an appearance attention and uses spherical distance to compute a positional attention.