43 results about "Robot hardware" patented technology

The invention provides a control method, which is applied to an artificial intelligence experiment system, at least comprises a robot hardware platform, a controller, a wheeled robot chassis and an industrial mechanical arm, and the method comprises the following steps: starting an operation system and executing hardware self-inspection; loading a large language model service and a traditional AI model, wherein the traditional AI model comprises a speech recognition model, a target detection model and a semantic analysis model; a natural language instruction of a user is received, voice is converted into a structured text through a voice recognition model, an instruction intention is analyzed through a large language model and a semantic analysis model, robot control parameters are generated, and a multi-modal interaction control signal is obtained; based on the multi-mode interaction control signal, a camera is called to collect image data, the position and category of a target object to be grabbed by the robot are recognized through a target detection model, the moving path of the robot and the grabbing track of the mechanical arm are calculated in combination with laser radar data, and the wheeled robot chassis and the industrial mechanical arm are controlled to execute coordinated actions.

According to the agriculture-like cultivated land and land humidity monitoring robot and the monitoring method, structurally, a soil turning device, a soil extraction device and a driving wheel device are connected with a driving motor, the driving motor, a camera device and the soil extraction device are in signal connection with a remote control terminal, and the soil turning device is arranged in a front-back telescopic and up-down movable mode relative to a machine body; the soil extraction device is movably arranged up and down relative to the machine body, the camera device is rotationally arranged relative to the machine body and used for monitoring the movement of the robot in real time, and by optimizing robot hardware configuration and sensor layout and integrating multiple sensors and a core control unit, comprehensive and accurate collection of farmland soil key parameters is achieved. And a wireless data transmission and cloud storage system is deployed to ensure real-time and accurate transmission and analysis of data. Artificial intelligence and computer vision technologies are integrated, soil images are processed, features are extracted and classified, and accurate evaluation and prediction of soil quality are achieved.

The invention provides a cluster robot triangular lattice formation control method based on extremely simple local sensing and a cluster aggregation algorithm. By simplifying a sensing and decision-making mechanism, the method can significantly improve the self-organization capability, adaptability and low-cost deployment capability of the cluster robot, and is suitable for large-scale deployment scenes such as environment monitoring, post-disaster rescue, automatic patrol and the like. Specifically, the robot acquires position and state information of other robots in a neighborhood by using a local sensing sensor, forms a uniform equilateral triangle lattice formation through self-organization position adjustment, and keeps formation stability by using a simple interaction rule. After the tasks are completed, the robot achieves self-organizing aggregation through a cluster aggregation algorithm, and tight recovery or subsequent task allocation can be completed without global information or complex communication. According to the method, the hardware complexity and the communication requirement of the robot can be reduced, the method is flexibly adapted to a dynamic complex environment, and the method has good performance and wide applicability.

The invention relates to a cluster robot platform based on implicit cooperation and multi-dimensional perception, and aims to improve the performance of a cluster robot in the aspects of measurement precision, perception capability and real-time performance. The platform is composed of a cluster robot hardware system and a software system. The hardware system comprises a bottom layer motion system, an optical cooperation beacon system and a main control board. A software system realizes cooperation and efficient behaviors among cluster robots through a multi-neighbor recognition and identity matching mechanism, a cooperative positioning mechanism, a relative speed direction measurement and calculation mechanism and a realistic differentiation information perception mechanism. The robot can accurately identify neighbors, perform calculation and adjust behaviors in real time so as to adapt to complex cluster tasks. The cluster robot platform supports verification and adaptation of various cluster algorithms, has high-precision perception, excellent real-time performance and wide algorithm adaptation capacity, and provides a solid foundation for research, development and application of the cluster robot technology.

The invention provides a hardware-level safety control device of an intelligent robot. The hardware-level safety control device of the intelligent robot comprises a power cut-off execution module, the network destroying module is provided with an isolation assembly capable of physically destroying a robot network communication unit; the independent security protection module integrates a hardware firewall and an encryption communication unit; the emergency control module is electrically connected with the power cut-off execution module and the network destroying module through a hard wire direct connection channel, an encryption communication channel with an external control terminal is established through the independent safety protection module, and a closed-loop safety control system independent of a robot main control system is formed. Through the synergistic effect of the power cut-off execution module and the network destroying module, power output and network communication of the robot are synchronously blocked, the double risks of out-of-control of hardware and data leakage are thoroughly eliminated, and comprehensive safety protection is formed.

The invention belongs to the field of robot path planning, and particularly relates to a method and system for dynamically covering a cleaning path of a cleaning robot, and the method comprises the following steps: sending a cleaning region task instruction to a task processing module; the task processing module controls the cleaning robot to sequentially execute cleaning tasks according to the task execution sequence; the motion control module sends corresponding cleaning path planning commands to the path planning module according to different cleaning task types; the sensing module processes the laser data and the camera data and then sends the laser data and the camera data to the path planning module; the path planning module updates the environmental information of the cleaning area, plans a cleaning path in real time based on the updated environmental information of the cleaning area, and sends the planned path to the motion control module; and the motion control module controls hardware of the cleaning robot to dynamically and intelligently execute cleaning operation. A dynamic path planning technology is used, so that the cleaning robot can dynamically plan the cleaning path in real time in a dynamic cleaning area, and the cleaning operation is efficiently completed.

The invention relates to the technical field of robots and network security, discloses a robot health consensus system and method based on ontology trust, and aims to solve the problems that an existing robot health declaration is easy in identity counterfeiting, one-sided in state evaluation, and easy in privacy leakage and replay attack in a certification process. The method comprises the following steps: generating an identity label strongly bound with robot hardware based on a physical unclonable function (PUF); fusing multi-source sensor data in a recursive updating mode to generate a state topology vector capable of reflecting a historical evolution trend; and obtaining a comprehensive health value in combination with the internal health value calculated by the vector and a network consensus factor. Through physical layer identity anchoring, dynamic and continuous state evaluation and an encryption certification mechanism, the authenticity and fraud resistance of the health declaration of the robot are remarkably improved, comprehensive evaluation of the health state of the robot is achieved, state privacy of the robot is effectively protected in interaction, and replay attacks are prevented.

The utility model discloses a modular joint motor set connecting structure for a quadruped robot, a shutdown motor set is connected with a joint motor set fixing cantilever of a robot trunk, and the joint motor set is mainly composed of a joint motor set fixing frame, a one-axis motor, a two-axis motor, a two-axis motor supporting frame, a three-axis motor supporting frame and a three-axis motor. The joint motor set is high in integration degree, and modularized and standardized design can be achieved; the joint motor set is more convenient to disassemble, assemble and maintain, dependence on manpower is reduced, and efficiency and maintainability are improved; the whole joint motor set is located outside the robot trunk, the influence of motor heat on the robot is reduced, heat dissipation is improved, and product reliability is improved; the joint motor is arranged outside the metal cavity of the robot trunk, electromagnetic interference of the motor to robot hardware is reduced, and product reliability is improved.

The invention discloses a fire-fighting reconnaissance robot system hardware reliability evaluation method, which comprises the steps of collecting hardware performance data and key environment stress data of a fire-fighting reconnaissance robot, and constructing basic evaluation data; extracting a core evaluation factor evolved along with the change of the fire scene condition, and determining reliability index data according to the core evaluation factor; determining a subjective weight and an objective weight according to the real-time reliability index data in the current environment; performing dynamic adjustment on the subjective weight and the objective weight of the fire-fighting reconnaissance robot index to obtain a self-adaptive comprehensive weight vector; and constructing a fuzzy relation matrix by using the core evaluation factor and the comprehensive weight vector, and performing synthesis operation to obtain a real-time updated system reliability level. According to the reliability evaluation method, real-time, dynamic and accurate evaluation of the reliability of the fire-fighting reconnaissance robot hardware can be realized, and intelligent decision support is provided for dynamic task decision and imminent maintenance of the robot.

The invention relates to the fields of robot technologies, intelligent hardware and the like, and aims to provide a design scheme of the modular hardware architecture and the hot plug functional component. The core design content is as follows: a modular hardware main architecture adopts a layered and partitioned design, comprises a core control module, a sensing module and the like, and realizes plug and play through a standardized interface; according to the hot plug function assembly, a camera, an emotion sensor and other modules support safe plugging in the operation process, and automatic recognition and drive loading can be achieved. And module management and monitoring: monitoring the state of the module in real time, and supporting remote upgrading, fault isolation and hot backup. And safety protection: multi-stage electrical protection, a hot plug protection circuit and a physical locking mechanism are provided. The method has the advantages that the robot hardware flexibility, maintainability and expansibility are improved, the upgrading and maintaining difficulty is lowered, the downtime is shortened, application of the robot in multiple fields is promoted, the cost is lowered, and the life cycle is prolonged.

The invention relates to a robot hardware-in-the-loop simulation method and system, and the method comprises the steps: controlling a preset robot controller to operate in a preset real machine, and obtaining camera data, IMU data and whole body joint data corresponding to the real machine through a preset DDS data channel; inputting the camera data, the IMU data and the whole body joint data into preset sensing, decision and motion control algorithms through a robot controller to generate a corresponding execution instruction; and inputting the execution instruction into a preset synchronous signal generator, and generating corresponding action instruction data, so that a robot controller controls each joint actuator of the target robot to execute corresponding action according to the action instruction data. Therefore, the problems that in the prior art, the integrated full-link simulation function of sensing, decision making and control is difficult to achieve, and simulation rhythm synchronization of sensing, decision making and control cannot be achieved are solved.

The invention discloses a hidden agricultural product detection method based on a progressive aggregation network, and belongs to the technical field of computer vision and agricultural robots. Aiming at the problems of low detection precision and poor robustness of hidden agricultural products in a complex agricultural environment, the method comprises the following steps: firstly, constructing a hidden agricultural product visual detection training sample data set suitable for a complex agricultural scene; the invention further provides an end-to-end real-time encoder-decoder structure, namely a progressive aggregation network, the network comprises an image pyramid encoder, a progressive internal aggregation module and a progressive guide aggregation decoder, and multi-level feature extraction and fusion are achieved through cooperation of the image pyramid encoder, the progressive internal aggregation module and the progressive guide aggregation decoder. And then, training the network model on a GPU cluster, and finally completing deployment and application on a picking robot hardware platform. The method is high in systematicness and detection efficiency and can be directly integrated in an existing agricultural product picking robot system, and the detection precision and picking operation efficiency of hidden agricultural products are effectively improved.

The invention provides a robot task scheduling method and device, equipment, a medium and a product. The method comprises the steps of obtaining a target task; constructing a robot atomic action library for a robot hardware module of the target robot according to the target task to form an atomic action layer; obtaining a logic step layer according to the atomic action and predefined action logic; according to the target task and the logic steps, obtaining an executable task layer of the target robot for the target task; constructing a multi-level priority queue scheduling structure according to the atomic action layer, the logic step layer, the executable task layer and a preset priority rule; and dynamically scheduling the robot according to the multi-level priority queue scheduling structure. A multi-level priority queue structure is constructed, a task dynamic scheduling mechanism is established, planning and scheduling of the robot are carried out at a finer-grained atomic action level, and execution coherence and scheduling instantaneity of the robot to long-time-sequence tasks are improved.

A method and a device for planning and controlling a leg robot, as well as a robot and a storage medium, wherein the method comprises: determining a foot-end reference state of each leg based on the motion state and swing parameter of the leg of a leg robot in the next instant; planning a collision-aware swing leg trajectory based on the foot-end dynamics model and the discrete collision model for the leg that will begin to swing in the next instant; performing whole-body control of the leg robot based on the motion state, the reference state sequence, and the foot-end reference state, wherein for the leg that was intended to be a support leg during planning and that did not touch the ground, collision-aware whole-body control is performed based on a whole-body dynamics model and a discrete collision model.This solves the problems of limited passive flexible buffering of collisions in leg robots in related technology, the large computational volume of the control algorithm, the low control accuracy, the high wear of the robot hardware, the short lifespan and the inability to meet the needs of practical applications.

The utility model relates to a kind of low-pressure four-axis robot hardware circuit, comprising: core board, mainboard, drive board, safety relay, switching power supply and AC contactor;The core board is connected with the mainboard, the mainboard is connected with the drive board, and the mainboard is connected with safety relay by relay control line, and the safety relay is connected with the AC contactor by contact control line, and the AC contactor is connected with the drive board by switching power supply;The core board, mainboard and drive board constitute electronic circuit part, optimize the controller multi-core heterogeneous computing architecture, realize software and hardware collaborative processing, provide powerful computing power for controller;With modularization thought design relevant PCB, guarantee the multiplicity of PCB;Using safety relay and AC contactor designs double-loop structure, through hardware redundancy, real-time monitoring and high-level safety certification, complete fault tolerance to single point is realized, while giving consideration to quick response and maintainability.

The embodiment of the invention provides a robot control method, device and equipment, the architecture design of a robot at least comprises a data layer and a decision-making layer, the data layer is configured with various task background data, task types supported by the robot and various meta-actions supported by the robot, and the decision-making layer deploys a large language model. The robot control method comprises the following steps: acquiring real-time sensing information of a task scene; in the decision-making layer, a meta-action sequence is generated through a large language model on the basis of cue words including data, real-time perception information, task types and various meta-actions, and the data is obtained by searching from various task background data according to initial perception information of a task instruction and a task scene; and controlling the robot to execute the meta-action sequence. Therefore, the expandability and compatibility of a robot control mode to various task scenes and various robot hardware systems are effectively improved by utilizing personalized information configuration of the data layer and the model capability of a large language model in the decision-making layer.

This application discloses a companion robot interaction method and system, relating to the field of multimodal interaction technology and applied to a cloud service platform. The method includes: acquiring interaction events reported by a user through a companion robot hardware terminal or mobile terminal; converting the interaction events into interaction commands; and sending the interaction commands to the companion robot hardware terminal, so that the companion robot hardware terminal can trigger game plots or develop personality attributes according to the interaction commands, and interact with the user according to the game plots or personality attributes. Through the interwoven personality attribute development and game plot-triggered interaction process, the playability and personalization of human-computer interaction are both enhanced.

The invention relates to a multi-mode synchronous control method for driving a robot to narrate, which comprises the following steps: receiving to-be-described text data, and standardizing the to-be-described text data to obtain preprocessed text data; performing hierarchical decoding on the preprocessed text data based on a large language model to obtain structured feature annotation data; generating a multi-modal expression sequence according to the structured feature annotation data, and performing timestamp alignment synchronization on the multi-modal expression sequence to obtain a collaborative multi-modal instruction; and carrying out bandwidth allocation and execution scheduling on the collaborative multi-modal instruction by adopting a dynamic priority strategy, and driving robot hardware to execute a corresponding multi-modal expression action. The diversity of robot emotion expression content is improved, and interaction delay is reduced.

The application provides a trajectory error diagnosis method and system for a robot hardware interface, comprising: offline recording of planning trajectory data and original actual feedback data containing original time stamps during robot movement; replaying the planning trajectory data according to the original time stamps and control cycles to reproduce the original instruction sequence; in each control cycle of the replay, synchronously collecting the replay instruction values; based on the replay instruction values and the original actual feedback data, calculating the joint space error curve cycle by cycle; finally, based on the time sequence characteristics of the joint space error curve, automatically diagnosing and outputting the candidate error causes. The application improves the error reproduction capability through offline recording and replay, realizes accurate positioning of the error through cycle-by-cycle monitoring, and realizes standardized and efficient problem positioning through automatic diagnosis based on error characteristics, which can effectively improve the stability and precision of robot motion control.

This invention relates to the field of heterogeneous robot technology, specifically to a method and system for adaptive dynamic computing power allocation for heterogeneous robots. The method includes: collecting typical task information of heterogeneous robots; classifying tasks into different computing power requirement levels based on task complexity, real-time requirements, and data processing volume; extracting key information to construct a preset task computing power library; receiving robot task instructions, parsing the tasks, setting task real-time weights, collecting robot hardware status parameters in real time, and calculating the appropriate computing power; calculating the matching degree between the appropriate computing power and the computing power currently provided by the hardware; when the matching degree is lower than a preset threshold, performing structured channel pruning and fine-tuning the model through knowledge distillation; when the computing power requirement still cannot be met after pruning, identifying non-core tasks and initiating a cloud-based task allocation process. This technical solution can improve the task execution efficiency and hardware adaptability of heterogeneous robots.

The application discloses a kind of robot simulation scene and trajectory automatic generation method and system, receive natural language operation instruction, and through semantic analysis, visual asset generation, three-dimensional reconstruction and physical credible layout generation three-dimensional simulation scene that can be used for robot operation training;Utilize video generation model to synthesize the continuous image sequence of target operation task, and extract two-dimensional visual motion trajectory in target object area;By depth estimation, three-dimensional motion fusion, grasping posture generation and inverse kinematics mapping, two-dimensional visual motion trajectory is converted into robot end effector trajectory and corresponding robot joint trajectory, the simulation scene and expert trajectory generated are paired, form training data set for robot imitation learning strategy training, and support the strategy obtained by training is deployed to simulation environment or real robot hardware.The application can reduce the cost of robot training data acquisition and artificial trajectory design, improve the degree of automation of simulation scene and operation trajectory generation.

The present application provides a hardware-level safety control device for an intelligent robot, comprising a power cut-off execution module including a triggerable circuit breaker component arranged in the main power circuit of the robot; a network destruction module configured with an isolation component that can physically destroy the network communication unit of the robot; an independent safety protection module integrating a hardware firewall and an encryption communication unit; an emergency control module electrically connected to the power cut-off execution module and the network destruction module respectively through a hard-wired direct connection channel, and establishing an encrypted communication channel with an external control terminal through the independent safety protection module, forming a closed-loop safety control system independent of the robot's main control system, and through the synergistic effect of the power cut-off execution module and the network destruction module, the robot's power output and network communication are synchronously blocked, completely eliminating the dual risks of hardware loss of control and data leakage, and forming comprehensive safety protection.

The present invention discloses a hardware-accelerated positioning system and a robot under the cooperation of software and hardware. The hardware-accelerated positioning system includes a CPU and a correlation scanning and matching hardware circuit; the correlation scanning and matching hardware circuit is used to search each layer of the first map to be searched in a hardware parallel processing manner and update the first upper bound value of the currently searched node; the CPU is used to, whenever the correlation scanning and matching hardware circuit searches for a node corresponding to a grid point of a map at a preset depth, search each layer of the second map to be searched by executing the branch and bound algorithm, and perform scanning and matching of the point cloud collected by the lidar on the corresponding layer of the second map to be searched by executing the correlation scanning and matching algorithm, and then calculate and update the second upper bound value of the currently searched node by using the matched result; the CPU is further used to configure the pose corresponding to the latest obtained second upper bound value or the pose corresponding to the latest obtained first upper bound value as the pose of the robot.

The application provides a kind of automatic inspection robot traffic inspection method based on urban traffic scene, it is related to intelligent traffic technology field, the method comprises: obtaining original multi-modal sensor data from inspection robot hardware platform and pre-processing, standard multi-modal sensor data is constructed BEV space representation learning module, simultaneously constructs VLA model module, constructs traffic event detection and anomaly identification module, and constructs end-to-end motion planning and navigation module and man-machine interaction and remote management module, define model training and deployment strategy to carry out model training and deployment control automatic inspection robot traffic inspection to multiple modules.Solve the technical problems that the prior art has insufficient perception, semantic understanding and navigation planning capability, which leads to the difficulty of efficient and reliable autonomous inspection of the inspection robot in complex traffic environment.The technical effects of autonomous perception, intelligent decision-making and efficient inspection of the inspection robot in complex traffic environment are realized.

The invention discloses a robot cluster scheduling system and method based on cloud side-end collaboration and an electronic device, and the system comprises a cloud platform, an edge cloud server and a plurality of robot hardware terminals, and the robot hardware terminals are used for determining a corresponding decision level according to the time delay demand and the computing power demand of a current robot task. Local decision making is carried out on the end-side hierarchical task according to the end-side AI chip, and scheduling control is carried out on the corresponding robot body according to a task decision making result; the edge cloud server is used for performing near-field collaborative decision-making on the edge level task and returning a task decision-making result to the robot hardware terminal; and the cloud platform is used for performing global task decision-making on the cloud level task based on the global digital twinborn model, and issuing a task decision-making result to the robot hardware terminal through the edge cloud server. The real-time performance and reliability of robot cluster scheduling are improved, and the method can be widely applied to the technical field of communication.

The invention provides a ubiquitous humanoid robot motion controller capable of being quickly deployed and a deployment method, and aims to solve the problem of quickly realizing a walking function of a humanoid robot, the motion controller comprises controller hardware and control software, the control software comprises a model training system, a motor parameter self-adaptive setting part, a real-time communication part and a control software programming interface, and can be conveniently butted with third-party simulation and training software through the designed interface to perform reinforcement learning training and simulation of a robot decision model; and the acquisition of related control parameters, the tuning of the related control parameters and the adaptation of humanoid robot hardware can be automatically carried out, the rapid deployment basically irrelevant to the robot hardware body is realized, and the software adaptation efficiency is greatly improved.

The invention relates to the technical field of computer vision and robot environment perception, in particular to a wheeled robot trafficability judgment algorithm under a pavement defect condition, which comprises the following steps: S1, designing a wheeled robot hardware system which comprises a carried camera for collecting pavement images in real time; s2, constructing an image data set containing a plurality of pavement defects, and marking the image data set; s3, based on a YOLO11n-seg model, developing a YOLO11n-RVD target detection segmentation model, and carrying out pavement defect identification and segmentation on the input image; s4, tracking the identified defects by using a multi-target tracking algorithm, and outputting a trafficability risk level in combination with a risk judgment algorithm; the problems of light weight of a target detection segmentation model and small target detection and positioning in the prior art are solved.

The invention belongs to the field of robot positioning, and particularly relates to a multi-sensor fusion cross-modal matching positioning method and system based on depth estimation. Pixel-level depth estimation is carried out based on a deep learning technology in a mode of collaborative fusion of a camera and a millimeter-wave radar to obtain a pixel point cloud; through matching the pixel point cloud with the high-precision map point cloud, cross-modal matching positioning is completed, accurate positioning of the robot in an operation environment is realized, and a positioning result with relatively high accuracy can be obtained while the hardware cost of the working robot is greatly reduced.

Embodiments include systems and methods for processing sensor data and generating operational instructions for ego (e.g., autonomous vehicle, robot) hardware. The ego includes any number of machine learning architectures, often neural network architectures, for processing the sensor data, perceiving the ego's surrounding environment, and making decisions regarding the ego's behavior. The ego's neural network architecture ingests the sensor data and uses the sensor data to perform any number of operations related to a particular domain or task, such as object recognition or path planning. A graph partitioner is trained to assign functions in the neural network's software and sensor data to specific hardware processing units. Some compilers are used to generate instructions based on the assigned type of processing unit.

According to the task planning and executing method and system of the heterogeneous robot, the storage medium and the computer equipment, the method is applied to the universal control system, the universal control system is in communication connection with different types of robots through the adapter, and unified management of the different types of robots can be achieved. Specifically, by introducing a meta-action / meta-skill library decoupled from robot hardware and a standardized task instruction, abstraction of robot hardware types by task planning and execution logic is realized, and repeated work of independently developing control codes for different robots is avoided; and the development cost and the maintenance difficulty of the heterogeneous robot system are obviously reduced. Meanwhile, according to the target planning scheme, the reasonability of task logic is ensured through manual verification, the same set of standardized instructions can adapt to different types of target robots through a hardware-instruction conversion mechanism of the adapter, and the compatibility of the system to heterogeneous hardware and the flexibility of task execution are greatly improved.