43 results about "Robots actuators" patented technology

The invention is suitable for the technical field of artificial intelligence, and provides a robot visual navigation optimization system based on edge computing, the system comprises a visual sensor module, an edge computing unit, a navigation algorithm module, a communication module and a robot executor, the visual sensor module is used for collecting high-definition images or video streams, and the edge computing unit is used for processing the high-definition images or video streams; the edge calculation unit is responsible for locally executing a visual navigation algorithm, the navigation algorithm module is deployed on the edge calculation unit, and the navigation algorithm module is used for executing a visual odometer, synchronous positioning and map construction, environment perception, obstacle detection, local path planning and motion control; and the communication module is used for carrying out model updating, task scheduling and non-real-time data uploading with a cloud. Complex visual data processing and navigation decision are sunk to the robot body or near-end edge equipment, so that the navigation precision and the response speed of the robot in complex and diversified environments are remarkably improved, and the energy consumption efficiency is optimized.

The invention discloses a multi-robot path coordination system based on reinforcement learning, and the system comprises a state collection module which is used for collecting the operation state and environment perception information of each robot, and generating an original data set; the state modeling module is used for constructing local state representation for each robot in each decision period; the opponent perception module is used for executing opponent learning perception training and outputting opponent parameter estimation vectors; the improved LOLA module is used for generating a final estimation result on the basis of self-strategy updating of the traditional LOLA; the strategy updating module is used for correcting the updating direction of the reinforcement learning strategy of the local machine and generating a feasible action set; the arbitration decision module is used for executing arbitration solution in combination with the candidate path action distribution to generate a scheduling result; and the execution feedback module is used for issuing the control action to a robot executor for execution, generating a path coordination result and storing the path coordination result. According to the invention, path coordination of multiple robots is realized.

The invention provides a dynamic walking exoskeleton control method and system based on man-machine complementary optimization, and the method comprises the steps: defining the step height and normal contact force of a contact point, and constructing a nonlinear complementary problem constraint and model; defining quantitative targets of nine types of motion tasks including contact point centroid position tracking, centroid linear velocity tracking, foot yaw angle alignment, frame attitude tracking, base quaternion derivative regularization, contact force regularization, joint regularization, swing height control and contact control regularization, and establishing a multi-task target function; a static nonlinear complementary problem is coupled with exoskeleton continuous dynamics, and a coupling system is constructed; solving the optimization objective function by adopting a neurodynamics algorithm to obtain the optimal control input of the exoskeleton-human body system; and the control input mapping obtained through optimization is converted into a driving instruction of a robot actuator, and a motion control closed loop is completed. According to the method, the complex movement requirements under strong coupling constraint interaction of the humanoid and the robot can be more accurately met.

An artificial intelligence-based business management system for real-time decision optimization, consisting of: a data acquisition unit configured to ingest structured, semi-structured and unstructured data streams from enterprise resource planning (ERP) systems, customer relationship management (CRM) platforms, Internet of Things (IoT) devices, external market feeds and financial transaction systems, encrypting, timestamping and verifying the data prior to further processing; a graph processing unit that is communicatively connected to the said acquisition layer, wherein the unit is configured to encode heterogeneous data into dynamic graph structures comprising nodes representing business entities and edges representing transaction or relationship dependencies, wherein the unit is further configured to perform deduplication, metadata tagging and real-time data synchronization; a decision optimization unit operationally linked to the graph processing unit, wherein the decision optimization unit includes modules for reinforcement learning, modules for Bayesian optimization and multi-objective solvers configured to simulate multiple alternative decision paths and select an optimal path based on performance indicators such as cost efficiency, resource utilization, customer satisfaction and risk minimization; a real-time inference control unit with specialized hardware cores, including at least one graphics processing unit (GPU), a field-programmable gate array (FPGA) and an application-specific integrated circuit (ASIC), wherein the accelerator performs inference tasks of the decision optimization unit with a latency in the millisecond range; a diagnostic processing unit configured to generate causal diagrams, feature mapping maps, and interpretable result summaries according to the optimization outputs; and a control interface unit configured to transmit optimized decisions to process controls within the enterprise, robot actuators, planning systems or interactive dashboards, with the interface supporting bidirectional communication for higher-level interventions, error feedback and triggers for re-optimization.

The invention belongs to the technical field of robot control, and discloses a foot-arm multi-task control method and system based on time-varying terminal cost, and the method comprises the steps: generating a whole-body reference trajectory of a robot offline, and pre-calculating a time-varying terminal cost matrix sequence based on the reference trajectory; based on the time-varying terminal cost matrix sequence, constructing and solving a nonlinear model predictive control optimization problem to obtain an optimization sequence of a system state and control input; and the optimization sequence serves as a reference instruction and is input into a layered whole-body controller for multi-task priority optimization, and a joint control instruction is generated and output to a robot execution mechanism. According to the method, the problem of short vision of a traditional NMPC in a long-time task is effectively relieved, and the operation precision of the robot is remarkably improved.

The invention relates to the technical field of robot motion control, and discloses a robot adaptive motion control method and system, and the method comprises the steps: obtaining the environment parameters and motion state parameters of a robot; constructing a terrain fitness evaluation factor based on the terrain feature data, calculating a dynamic compensation coefficient based on the load variation, determining an environmental disturbance error according to the deviation between the external disturbance force information and a preset disturbance threshold, and determining a motion trajectory error according to the offset between the end effector pose and the target trajectory; inputting the joint angle, the centroid acceleration, the terrain fitness evaluation factor, the dynamic compensation coefficient, the environmental disturbance error and the motion trail error into an adaptive strategy model, and generating a gain adjustment instruction of a PID controller; and according to the gain adjustment instruction, the proportional term coefficient, the integral term coefficient and the differential term coefficient of the PID controller are updated, a self-adaptive control instruction is generated, and a robot execution mechanism is driven to move. According to the invention, the motion stability, precision and efficiency of the robot in changeable scenes can be improved.

The invention discloses a liquid crystal elastomer structure multi-objective optimization method based on depth evolution mapping. According to the method, by constructing a mapping model from individual features to operator parameters, an optimization algorithm can adaptively adjust the intensity of crossover and mutation operations according to the position, fitness ranking and neighborhood features of each candidate solution in a search space, so that the local refining effect is enhanced while the global exploration capability is kept. Compared with an existing parameter adjusting method based on a unified parameter strategy or population statistics, the method has the advantage that self-adaptive optimization of an individual level is realized. The method belongs to the cross technical field of computational material science, intelligent optimization algorithms and structural design, can be applied to the performance optimization problem of intelligent material systems such as soft robot drivers, artificial muscles and flexible sensors, and can also be popularized to other engineering design scenes involving high-dimensional parameter space and multi-target tradeoff.

The invention relates to the technical field of robot actuators, and particularly discloses an integrated joint module with torque sensing and energy self-recovery functions and a design method. The design method comprises the steps that a physical framework of the joint module is constructed; a nonlinear elastic torsion model and a dynamic friction model of the harmonic reducer are constructed, and torque is calculated and derived; calculating a physical sampling torque; fusing the derived torque and the physical sampling torque by adopting an adaptive Kalman filtering mode to obtain a final torque signal; an energy feedback model in a motor braking mode is constructed, and when the joint module decelerates or is dragged by external force, the instantaneous flow of recyclable electric energy is calculated in real time based on the energy feedback model; a super capacitor and a DC-DC converter are introduced for energy recovery; and when the joint module is started or enters an acceleration stage, the energy stored by the super capacitor is released to the direct current bus. Nonlinear transmission interference caused by a harmonic reducer can be effectively eliminated, and in-situ recovery and peak power assistance of braking kinetic energy can be achieved.

The invention relates to the technical field of robot hand control, and discloses a multi-camera-based robot hand positioning and musical instrument playing control method and system, and the method comprises the steps: obtaining a global image through a head camera, and providing a hand rough position; the fingertip camera obtains a local image and provides fine positioning of a target; fusing coarse and fine positioning, calculating a relative position, and planning a motion track; a robot execution mechanism is controlled to execute the track to complete playing; monitoring an actual position, calculating an error, and correcting a joint when the error exceeds a threshold value; the system comprises a head camera module, a fingertip camera module, a visual fusion module, a control module, a robot execution mechanism and a feedback and correction module. According to the invention, by acquiring the high-resolution local image and combining the global rough position information provided by the head camera module, effective combination of rough positioning and fine positioning is realized, and the problem that single global vision is easily blocked by hands or the precision is insufficient due to long distance is effectively overcome.

Tool (108) for fixing an energy direction sensor (100) for joining thermoplastic components (104) by ultrasonic welding, wherein the tool (108) has an ultrasonic input, a contact wheel holder (112) and at least one contact wheel (116) rotatably arranged on the contact wheel holder (112) with a structured contact surface (118); tool arrangement (102) for positioning, applying and / or fixing an energy direction sensor (100) for joining thermoplastic components (104) by ultrasonic welding, wherein the tool arrangement (102) has at least one receptacle (106) for a ribbon and / or film-shaped energy direction sensor (100) and such a tool (108); effector for a robot, wherein the effector has such a tool (108) and / or such a tool arrangement (102); and use of such a tool (108) and such a tool arrangement. (102) and / or such an effector,wherein, using the tool (108), the tool arrangement (102) and / or the effector, an energy direction transmitter (100) for joining thermoplastic components (104) by ultrasonic welding is positioned, applied and / or fixed.

A magnet mount for a robotic actuator is disclosed. According to one aspect of the present invention, a magnet mounting base for a robot actuator, which is assembled on the output side of the robot actuator and coupled to a rotating plate, comprises: a mounting member which is coupled to the lower part of the inner surface of the plate and rotates, and which has a hollow cylindrical shape; the magnet part is attached to the outer periphery of the lower part of the mounting part and is made of a magnetic material; wherein an attachment area to which the magnet component is attached is formed on the outer side of the mounting component, and a pattern part with a certain pattern is formed on the attachment area.

The application provides a dynamic walking exoskeleton control method and system based on human-computer complementary optimization, comprising the following steps: defining the step height of the contact point, the normal contact force, constructing the nonlinear complementary problem constraint and model; defining the centroid position tracking of the contact point, the centroid linear velocity tracking, the foot yaw angle alignment, the frame attitude tracking, the base quaternion derivative regularization, the contact force regularization, the joint regularization, the swing height control, the contact control regularization, 9 types of motion task quantization targets, and establishing a multi-task objective function; coupling the static nonlinear complementary problem with the exoskeleton continuous dynamics to construct a coupled system; solving the optimization objective function by using a neural dynamics algorithm to obtain the optimal control input of the exoskeleton-human system; and mapping and converting the control input obtained by optimization into the driving instructions of the robot executor to complete the motion control closed loop. The application can more accurately adapt to the complex motion requirements under the human-robot strongly coupled constraint interaction.

To provide a robot that can improve calibration accuracy of a rotation sensor for detecting movement of an actuator of the robot.SOLUTION: A robot has a connection frame (63) which supports a rolling actuator. The connection frame (63) has a first fitted part (63g) which is fitted to a rotary output part (12c) of the actuator (12), and a remainder (a first arm part (63b), a support part (63a), a second arm part (63c)) that is connected to the first fitted part (63g). a sensor rotation part (16a) of a rotation sensor (16) is fitted to the first fitted part (63g). In the state that the first fitted part (63g) is fitted to the rotation output part (12c) and the remainder of the connection frame (63) is removed from the first fitted part (63g), the first fitted part (63g) and the rotation output part (12c) can rotate through an angle greater than 360 degrees.SELECTED DRAWING: Figure 8

This invention discloses a robot actuator with autonomous feedback and rapid deformation. It includes actuators mounted on the legs at the bottom of the robot. When the robot lands on a landing platform, the actuators adjust the tilt and shorten / extend of the legs to achieve a smooth landing. The actuator comprises a housing, tilt adjustment structures, shortening adjustment structures, extension adjustment structures, and front and rear transmission rods and adjustment plates. The transmission rods contact and output applied pressure and sense the shape of the external contact surface, transmitting pressure changes to the tilt, shortening, and extension adjustment structures. This invention can automatically sense and feedback the shape and pressure of the external contact surface, automatically deform, and adapt to the tilt of the external contact surface, solving the problems of lack of sensing and difficulty in adaptation between the robot and the external contact surface.

A magnet mount for a robot actuator is disclosed. According to an aspect of the present disclosure, the magnet mount includes a cylindrical mount member that is coupled to a lower inner side surface of a plate to rotate and has a hollow; and a magnet member that is attached along a lower outer circumference of the mount member and made of a magnetic material, in which an attachment area to which the magnet member is attached is formed on an outer side of the mount member, and a pattern part having a predetermined pattern is provided in the attachment area.

This invention relates to the field of magnetic code disk materials technology, specifically disclosing a magnetic code disk material for humanoid robot actuators, its preparation method, and its applications. The magnetic code disk material for humanoid robot actuators comprises a substrate, a Bi layer, an AlNiFeCo thin film, an Nb layer, and a Ta layer stacked sequentially. The novel structural design provided by this invention introduces a dual-intercalation cladding structure of Bi and Nb layers, which synergistically act on the AlNiFeCo thin film, meeting the application requirements of high-precision magnetic encoder code disk materials for high coercivity and high remanence. Simultaneously, the element diffusion is a diffuse pinned network distribution, which is beneficial for obtaining uniform and consistent high-quality sine wave signal output performance during code disk writing. The magnetic code disk material prepared by this invention is suitable for the preparation of high-performance, high-uniformity magnetic code disk materials and high-precision, miniaturized magnetic encoder technology, and has excellent application prospects in humanoid robot actuators.

The application discloses a multi-object-oriented action sequence generation system and method and a robot control method, relates to the technical field of artificial intelligence, and solves the technical problems of target selection deviation and operation failure caused by perception attention drift and strategy lazy learning in a multi-object scene.The technical solution points are that a perception grounding expert module performs explicit space coordinate regression on the extracted fusion feature map, and outputs the normalized boundary box coordinates of the target object;then the coordinates are encoded into independent anchor token, and a decoupled input sequence of visual token, state token, anchor token and action token is constructed;the decoupled input sequence is modulated, and finally the anchor token is forced to be queried by the action expert module in each denoising step, so that a short-time action block sequence of the precise anchor target is generated and delivered to the robot executor to drive the current operation step.The target selection accuracy in the multi-object interference task is improved, and the instruction following robustness and operation success rate are enhanced.

The invention discloses an artificial pneumatic muscle based on electromagnetic driving and a robot actuator. The artificial pneumatic muscle comprises an electromagnetic head, a compression device and a woven mesh body. One end of the electromagnetic head is fixedly connected with one end of the compression device; the compression device is of a rigid body structure and comprises a piston cylinder and a piston arranged in the piston cylinder and driven by electromagnetism. The woven mesh body covers the outer side wall of the compression device, and forms a closed air cavity together with the piston cylinder; compressed air with pre-tightening pressure is pre-filled in the closed air cavity; when the electromagnetic head generates electromagnetism, the piston is driven to move in the piston cylinder in a reciprocating mode so as to change air pressure in the closed air cavity and drive the woven mesh body to contract or stretch in the axial direction. The problems that a pneumatic muscle system is heavy and poor in portability are solved, and a new-generation flexible driver is achieved.

This invention belongs to the field of collaborative robot automation control, specifically a collision detection method for collaborative robots based on decision trees. It includes the following steps: constructing a kinematic model for the robot; performing singularity analysis on the kinematic model to remove cases where the robot is located at singular points; obtaining the robot's theoretical torque based on kinematic gravity compensation theory; training the decision tree model using robot actuator data and the robot's theoretical torque; and inputting real-time robot actuator data and theoretical torque into the trained decision tree model to determine whether a collision has occurred at the current moment. This invention's machine learning algorithm based on decision trees uses velocity, acceleration, real-time torque, theoretical torque, and the derivative of real-time torque as judgment criteria, resulting in more comprehensive and accurate data consideration.

A robot, a robot actuator, and an actuator frame are disclosed. The actuator frame includes a base frame, a fixed base, an outer shaft, and an inner shaft. The base frame has motor mounting plates on its left and right sides, each with a fixed workstation for storage. An inverted U-shaped groove is formed between the two motor mounting plates. The inner shaft faces the Y-axis and is mounted in the inverted U-shaped groove. The fixed base has a rotating hole. One end of the outer shaft is rotatably connected to the rotating hole around the X-axis. The other end of the outer shaft is rotatably fitted onto the inner shaft around the Y-axis. The left and right sidewalls of the inverted U-shaped groove have sidewall openings, and the inner shaft has a hollow structure inside. The sidewall openings expose the hollow structure, forming a wiring channel between them. This solution enables the robot actuator to move in multiple directions (front, back, left, right), and also allows the wires to be internally placed within the wiring channel of the inverted U-shaped groove, avoiding the problem of wires becoming tangled around the shaft during multi-angle movement of the actuator.

The application provides a body intelligent control system, and belongs to the technical field of artificial intelligence. The system comprises: an external input module, which is used for receiving external input data in real time; a three-dimensional Gaussian field construction and maintenance module, which is used for constructing and maintaining a dynamic semantic three-dimensional Gaussian field in real time; a spatial language analysis module, which is used for analyzing natural language instructions, extracting operation entities, spatial relationship words and action words; a spatial relationship geometry anchoring module, which is used for determining a target three-dimensional area corresponding to the operation entities in the dynamic semantic three-dimensional Gaussian field; a three-dimensional action planning module, which is used for carrying out end-to-end action planning in the dynamic semantic three-dimensional Gaussian field, and generating a collision-free action trajectory; and an execution control module, which is used for converting the collision-free action trajectory into a control instruction sequence, and driving a robot executor to complete the operation corresponding to the action words.

The application discloses a robot actuator and a robot, wherein the robot actuator comprises a shell and a reducer assembly. The reducer assembly comprises an inner ring gear, a first bearing, a planetary wheel assembly and a limiting piece. The inner ring gear is in limiting connection with the inner wall of the shell. The first bearing comprises relatively rotating first and second parts, the first part is in limiting connection with the inner wall of the inner ring gear, the first part is in limiting connection with the inner wall of the shell, and the second part is arranged in space with the inner ring gear. The planetary wheel assembly comprises a planetary wheel carrier and a planetary gear set, the planetary gear set is connected with the planetary wheel carrier and is rotatably connected with the inner ring gear; the planetary wheel carrier is connected with the second part. The limiting piece is in threaded connection with the planetary wheel carrier, the limiting piece is in limiting connection with the axial end surface of the second part, and the radial end surface of the limiting piece is in rotating connection with the shell. The robot actuator provided by the application is not prone to displacement of at least part of the reducer assembly in the shell under the action of impact load, and the actuator can stably work.

In an industrial robot, an external high-precision metrology tracking system, such as a laser tracker system, is used to directly measure robot kinematic errors and corrections are implemented during processing so that the end effector of the robot may be accurately positioned so that a tool or other object carried by the robot effector can carry out a designated function, such as machining a workpiece or other operation requiring that the effector be accurately positioned with respect to a workpiece.

The invention provides an action generation model reasoning acceleration method and system based on space-time consistency prediction and a terminal, and the method comprises the steps: precisely judging whether a historical action sequence exists at present or not when each control period is started, and flexibly selecting according to a judgment result; a predicted action sequence is obtained by means of a space-time consistency action sequence prediction model; and if not, generating an initialized action sequence by utilizing an action generation model based on a diffusion model. And then, on the basis of the action generation model, diffusion reverse denoising is carried out according to the obtained sequence, a final action sequence is generated and sent to a robot actuator to be executed, meanwhile, the sequence serves as a historical sequence of the next period, and the steps are repeated till all control operations are completed. In the reasoning process, the space-time consistency action sequence prediction model and the action generation model cooperatively work in a cascade mode, the reasoning efficiency can be improved while the performance is guaranteed with few denoising steps, the action generation speed can be greatly improved, and the robustness, stability and practicability of a diffusion strategy in a real physical system are enhanced.

The invention relates to the technical field of robot control, in particular to a self-adaptive control method and system for an upper limb rehabilitation robot. Comprising the steps that motion data of upper limb joints of a patient are collected through a sensor array and fused to generate comprehensive motion vector representation, motion pattern classification is conducted through a classification model, and whether coordination deviation exists between the joints or not is judged; if deviation exists, joint deviation distribution information is obtained, robot control parameters are optimized through an optimization model, and an adjustment instruction sequence is generated to drive a robot actuator to conduct intervention; after intervention, the intervention effect is evaluated through feedback data, the personalized demand model is updated according to the effect, and a subsequent intervention scheme is optimized. According to the invention, the defects of an existing rehabilitation robot in coping with the complex motion coordination problem are overcome, the self-adaptive control based on real-time data and feedback is realized, and the rehabilitation treatment effect and the personalized level are remarkably improved.

This invention discloses a design method and device for a wheat field weeding actuator based on an operational power consumption model. The weeding actuator includes a rotary blade and a roller, with multiple rotary blades connected to the roller. The initial position is taken as the tip of the rotary blade contacting the soil. The entire weeding process is divided into four working conditions: contacting the soil, cutting the soil, chopping the grass, and lifting the soil. The rotary blade weeding actuator then employs a power consumption model during operation: W = ∑ i 1(W1+W2+W3+W4); The wheat field weeding device mainly consists of a rotary blade, rollers, a drive motor, a controller, and auxiliary mechanisms. The actuator of this invention uses a weeding wheel structure. By analyzing the robot-soil-weed interaction during wheat field weeding, a power consumption model of the rotary blade weeding robot actuator is constructed, and the structural parameters with the lowest power consumption during operation are determined. This not only results in a wide working area, low operating resistance, and high weeding efficiency, but also effectively solves the problem of high seedling damage rate when weeding crops with small row spacing.

The invention relates to the technical field of underwater robot collaborative operation, in particular to a double-underwater robot collaborative operation integrated coordination control method, which comprises the following steps: S1, establishing an overall kinetic equation for describing an underwater robot actuator system based on a Lagrange method; s2, a kinematics mapping relation of the system is constructed in a memory of a processor, and the coordinate transformation operand between the underwater robot body and the mechanical arm is eliminated; s3, providing a joint space velocity observer independent of an acceleration sensor; and S4, designing an underwater robot-actuator integrated coordination controller on the basis of the observer, and designing an adaptive law to cope with the model uncertainty of the underwater complex environment. According to the invention, a hydrodynamic simplified model is adopted, the accuracy and complexity of the model are balanced, a novel double-underwater-robot integrated coordination control method is provided, and under the condition that the calculation complexity is reduced, the stabilization of a robot body and the cooperation of the tail end of an actuator are realized at the same time.

The application provides a robot imitation learning method based on intention understanding and subconscious execution, comprising the following steps: collecting a plurality of robot demonstration trajectory data to form a data set, calculating an intention influence measure according to the robot demonstration trajectory data; smoothing the intention influence measure by using a smoothing filter, calculating a dynamic down-sampling frequency, and down-sampling the robot demonstration trajectory data to generate an awareness trajectory data set; establishing a deep learning-based action prediction model, and training the deep learning-based action prediction model by using the awareness trajectory data set; inputting current observation data, sensor data and action data into the trained action prediction model to output a predicted future action sequence, and predicting an execution action at a next moment through a cognitive unloading mechanism and a weighted cumulative prediction strategy; feeding back the predicted execution action at the next moment to a robot executor to obtain new observation data, and repeating the above steps until a target task is completed.

A robotic vehicle-treatment system includes a robotic platform configured to autonomously position within a constrained vehicle service environment. A sensing system having at least one camera and depth sensor generates a three-dimensional digital model of a vehicle region and surrounding operational space. An artificial-intelligence control system processes the digital model to determine adaptive control commands for robotic motion and tool interaction. A controller coupled to one or more robotic actuators executes the adaptive control commands to manipulate at least one vehicle component associated with a vehicle lighting system. The adaptive control commands dynamically adjust robotic posture, motion path, and operational parameters in response to sensed environmental conditions. The system enables autonomous removal, installation, alignment, restoration, or inspection of vehicle lighting components within interior or exterior vehicle regions under varying environmental conditions.