Open type control platform and method for industrial six-axis robot

Abstract

The invention relates to the technical field of industrial robot control, in particular to an open type control platform and method for an industrial six-axis robot, the open type control platform comprises a hardware architecture and a software architecture, the hardware architecture adopts a modular design of combining a heterogeneous core module and a standardized extension module, and the software architecture is of a layered decoupling structure. The platform can switch two independent control paths of industrial-grade operation and scientific research-grade verification, and has hardware interface expansion capability and software algorithm self-definition capability. In order to solve the problem of system closeness, an open hardware design document is adopted, and a standardized software API and a layered decoupling software architecture are provided, so that a user is endowed with an access right to bottom hardware and a secondary development capability of a core algorithm, full-stack opening from hardware to software is realized, and the user experience is improved. The problem that a user of a traditional system cannot perform customized development and deep intervention is solved.

Description

Technical Field

[0001] This invention relates to the field of industrial robot control technology, and in particular to an open control platform and method for an industrial six-axis robot. Background Technology

[0002] As core equipment in modern intelligent manufacturing, the control system of industrial robots is crucial in determining their performance, flexibility, and level of intelligence. Currently, mainstream industrial robots generally employ dedicated, closed control architectures. While these architectures have undergone long-term industrial validation in terms of stability and reliability, with the manufacturing industry's transformation towards flexibility and intelligence, and the increasing demand for open experimental platforms in scientific research and education, existing control systems are increasingly revealing the following limitations: The system's underlying hardware and software are not open, making it difficult for users to perform customized development and algorithm intervention; operation heavily relies on dedicated teach pendants, resulting in a single and inflexible interaction method; hardware interfaces are fixed, making it difficult to expand with new sensors or intelligent computing units; program and data management methods are closed, hindering migration, collaboration, and version control; communication protocol support is limited, brand compatibility is poor, and modern network communication capabilities are lacking. These inherent defects severely restrict the application and development of robots in flexible production, cutting-edge algorithm verification, rapid secondary development, and cross-platform collaboration.

[0003] Therefore, this application proposes an open control platform and method for an industrial six-axis robot. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies and solve the technical problems existing in the background technology, this invention proposes an open control platform and method for an industrial six-axis robot.

[0005] This invention is achieved through the following technical solution: An open control platform for an industrial six-axis robot includes a hardware architecture and a software architecture. The hardware architecture adopts a modular design combining heterogeneous core modules and standardized extension modules. The software architecture is a layered and decoupled structure. The platform can switch between two independent control paths: industrial-grade operation and scientific research-grade verification. It also has hardware interface expansion capabilities and software algorithm customization capabilities.

[0006] Prior to this, the heterogeneous core module integrates an ARM processing unit and an FPGA processing unit. The ARM processing unit is equipped with a real-time optimized operating system and undertakes high-level logic decision-making, trajectory planning calculation, communication protocol parsing, and human-machine interaction management tasks. The FPGA processing unit is configured as an EtherCAT hardware master station to realize distributed clock synchronization control and multi-axis pulse signal hard timing generation functions, ensuring the real-time motion control requirements with nanosecond-level synchronization accuracy.

[0007] Preferred, the standardized expansion module integrates multiple hardware interfaces, including at least one industrial fieldbus interface compatible with EtherCAT, PROFINET or CANopen protocols, six or more pulse direction output interfaces and corresponding encoder feedback input interfaces, isolated digital input / output interfaces, at least one extended communication interface containing CAN bus or RS485, at least one wireless communication interface supporting Wi-Fi or 5G standards, and an AI acceleration expansion interface for accessing the coprocessor.

[0008] Prior to this, the software architecture is divided into four layers from bottom to top: a hardware abstraction layer, which realizes the mapping and matching of logical control axes and physical drive units, the binding of process data objects, and integrates real-time safety monitoring functions; a control kernel layer, which incorporates an open-source EtherCAT master station protocol stack, a robot kinematics calculation library, and a servo drive state machine management module compliant with the CIA402 standard; an algorithm plug-in layer, which constructs standardized algorithm interface specifications, is equipped with a plug-in management unit, and supports dynamic loading of algorithm function modules such as trajectory planning, real-time interpolation, and dynamics compensation; and an application layer, which provides instruction parsing tools, data application programming interfaces, remote human-machine interaction interfaces, and file preloading management functions.

[0009] Preferred, the implementation of the scientific research-grade verification control path is as follows: it supports importing N×6-dimensional trajectory matrix files generated by external simulation tools, characterized by joint spatial position or pulse number; during the system initialization phase, the trajectory matrix data is preloaded into the controller's global memory array at once; during the control loop, the trajectory point data is read directly from the memory array according to a preset period, skipping the planning and interpolation logic of the algorithm plug-in layer, and directly outputting it to the servo drive network through the hardware abstraction layer.

[0010] Preferred, the industrial-grade operation control path is implemented as follows: receiving high-level control commands sent by a host computer or teaching device, converting them into structured data containing target pose and motion constraints through an application layer command parsing tool; calling a trajectory planning algorithm loaded by an algorithm plugin layer to generate a coarse interpolation path; within the real-time control cycle, refining the coarse interpolation path into target position commands for each control cycle through a real-time interpolation algorithm, and outputting them through an industrial bus or pulse interface via a hardware abstraction layer to complete the closed-loop control process.

[0011] An industrial six-axis robot control method applicable to the open control platform described above includes the following steps: Step 1: Select the control path according to the actual application scenario. If it is an industrial production scenario, start the industrial-grade operation control path and execute steps 2-4; if it is a scientific research and verification scenario, start the scientific research and verification control path and execute steps 5-7. Step 2: Receive high-level control commands from the host computer or teaching device, and convert them into structured data containing the target pose and motion constraints using an application-layer command parsing tool; Step 3: Call the trajectory planning algorithm already loaded in the algorithm plugin layer to generate a coarse interpolation path based on the structured data; Step 4: During the real-time control cycle, the real-time interpolation algorithm of the algorithm plug-in layer is called to refine the coarse interpolation path into the target position instructions of each servo drive unit. The instructions are then output through the industrial bus or pulse interface via the hardware abstraction layer. At the same time, the operating status is monitored in real time through the safety monitoring module to complete the closed-loop control. Step 5: In scientific research verification scenarios, the trajectory matrix file generated by external simulation tools is preloaded into the controller memory; Step 6: Start the high-fidelity playback task, read trajectory point data from memory at fixed intervals, and output it directly to the driver network through the hardware abstraction layer; Step 7: Synchronously record the feedback data of the servo drive unit for subsequent algorithm verification and analysis.

[0012] Prior to this, during the execution of steps 4 and 6, the safety monitoring module of the hardware abstraction layer detects the following error, torque parameters and limit status in real time. When the detected data exceeds the preset threshold, the safety protection mechanism is immediately triggered to cut off the power output and lock the safety relay.

[0013] Prioritizes the use of a remote human-machine interface at the application layer to enable online modification of control parameters, real-time monitoring of system status, cloud synchronization of control programs, and version management.

[0014] Prior to this, in the industrial-grade operation control path, the control kernel layer manages each slave driver through the integrated EtherCAT master protocol stack in a periodic synchronous data exchange manner, and with the help of the distributed clock mechanism implemented by FPGA, ensures that the control commands of all axes are executed synchronously within microseconds.

[0015] The beneficial effects of this invention are: 1. This invention addresses the problem of system closure by adopting open hardware design documents, providing standardized software APIs, and a layered and decoupled software architecture. This grants users access to the underlying hardware and the ability to perform secondary development of core algorithms, achieving full-stack openness from hardware to software. This solves the problem that users of traditional systems cannot perform customized development and in-depth intervention.

[0016] 2. This invention addresses the issues of poor scalability and compatibility by employing a modular hardware design with a core board and expansion board, as well as a hardware abstraction layer that supports multi-protocol conversion. This enables the convenient integration of various sensors, intelligent computing units, and driver devices from different brands and with different interfaces, achieving broad compatibility between hardware interfaces and communication protocols. This solves the problems of rigid system hardware, difficulty in expansion, and high brand barriers.

[0017] 3. This invention addresses the problem of inflexible human-computer interaction by supporting multi-mode interaction based on PC, mobile terminals, and the Web at the application layer and integrating a wireless communication module. This provides multiple control and teaching methods in addition to dedicated teach pendants, achieving flexibility in interaction methods and remote operation capabilities, and solving the problem of excessive reliance on a single dedicated teach pendant.

[0018] 4. This invention addresses the problem of insufficient control precision and real-time performance by using FPGA hardware to implement EtherCAT master station and distributed clock synchronization. This physically controls the timing jitter of communication and control to the nanosecond level, achieving synchronization precision and control real-time performance far superior to traditional software solutions. It also solves the high-precision control bottleneck caused by software uncertainties such as operating system scheduling.

[0019] 5. This invention addresses the problem of the disconnect between scientific research verification and industrial application. By innovatively designing a dual-path control method for industrial production and scientific research verification, the same platform can meet the reliability and automation requirements of industrial sites while providing a high-fidelity, interference-free physical verification environment for algorithm research. This achieves seamless integration from scientific research to industrial application and solves the pain points of inconsistent simulation and physical experiment results and long verification cycles. Attached Figure Description

[0020] Figure 1 This is a block diagram of the overall system structure of the present invention; Figure 2 This is a schematic diagram of the hardware system architecture of the present invention; Figure 3 This is a block diagram of the software layered architecture of the present invention; Figure 4 This is a flowchart of the industrial production mode control method of the present invention; Figure 5 This is a flowchart of the scientific research simulation mode control method of the present invention. Detailed Implementation

[0021] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Experimental methods in the following embodiments, unless otherwise specified, are generally performed under conventional conditions or as recommended by the manufacturer.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of skill in the art. The reagents and raw materials used in this invention are readily available through conventional means, and unless otherwise specified, they shall be used in accordance with conventional methods in the art or as per the product instructions. Furthermore, any methods and materials similar to or equivalent to those described herein may be applied to the methods of this invention. The invention will now be further described with reference to the accompanying drawings and specific embodiments. The preferred embodiments and materials described herein are for illustrative purposes only.

[0023] The industrial six-axis robot open control platform provided in this embodiment of the invention has the following overall structure: Figure 1 As shown, the platform is based on an open embedded control platform, connecting and controlling the six-axis robot body through a heterogeneous servo drive network. Users can operate it remotely or locally through various user interaction terminals (such as PC software, mobile APP, and web browser). Simultaneously, the platform supports data exchange with cloud-edge collaboration and edge computing systems, enabling cloud-based program management, data analysis, AI model deployment, and digital twin simulation.

[0024] 1. Specific implementation of the hardware system The core architecture of the hardware system, such as Figure 2 As shown, a modular design using a core board and an expansion baseboard is adopted.

[0025] The core board can be a heterogeneous computing chip that integrates an ARM processor and an FPGA, such as the Xilinx Zynq-7000 series SoC. This chip integrates a dual-core ARM Cortex-A9 processor (Processing System, PS) and a programmable logic unit (PL).

[0026] PS end: Runs an operating system based on the Linux kernel with the PREEMPT_RT real-time patch, mainly responsible for non-real-time or soft real-time tasks, including: high-level trajectory planning algorithm - the trajectory planning can be implemented with the help of the function library provided by existing open source robot development frameworks (such as Orocos, MoveIt, etc.), Ethernet communication protocol stack (TCP / IP), instruction parser, system status management, and human-computer interaction service for connecting to the host computer.

[0027] PL (FPGA): Programmed to implement the following hard real-time functions: EtherCAT hardware master IP core: Uses open source or self-developed IP core to realize EtherCAT frame packet assembly, parsing and transmission, and strictly follows the ETG.1100 specification.

[0028] Distributed Clock Management Unit: Responsible for generating and synchronizing the master clock for the entire EtherCAT network. Through the precise timing control capabilities of the FPGA, the synchronization accuracy of the slave clocks is controlled at the nanosecond level, significantly reducing or even eliminating microsecond-level jitter caused by operating system scheduling in traditional PC software master stations.

[0029] Multi-axis pulse generator: generates 6 independent pulse signals (PULSE / DIR) with direction control. The pulse frequency and number are calculated in real time by the target position written by the PS terminal through the AXI bus, and the absolute accuracy of the output timing is ensured by the FPGA hardware timer.

[0030] The expansion baseplate is designed around the core board and provides a full range of standardized industrial interfaces: Communication interface module: EtherCAT interface: The TIDP83822 or similar industrial-grade PHY chip can be used to convert the RMII signal output by the FPGA into a standard 100BASE-TX Ethernet signal.

[0031] Universal Ethernet interface: Equipped with a Gigabit Ethernet PHY chip for debugging, file transfer, and connecting to non-real-time networks.

[0032] USB-UART debug port: used for system startup, kernel log output, and low-level debugging.

[0033] Industrial I / O and Expansion Modules: Isolated digital input / output: Optical or magnetic isolation devices are used. The input interface is used to connect limit, zero, emergency stop and other switch signals, and the output interface is used to control the end effector or brake.

[0034] Expansion bus: Provides one isolated CAN bus interface and one RS485 interface for connecting six-dimensional force sensors, barcode scanners or other field devices.

[0035] Wireless communication and expansion modules: It can integrate communication modules that comply with Wi-Fi 6, Bluetooth 5.0 and 5G standards, such as commercial modules based on PCIe or USB interfaces.

[0036] It provides a standard M.2 KeyM interface, which can be plugged into AI accelerator cards such as NVIDIA Jetson Nano or Intel Neural Compute Stick 2 for deploying AI algorithms such as visual servoing and object recognition.

[0037] 2. Specific implementation of the software system Layered architecture of software systems, such as Figure 3 As shown, a modular design is adopted, and the layers are called through clear API interfaces.

[0038] Hardware Abstraction Layer: Axis mapper: Maintains a configuration file (such as XML or YAML) that maps application-layer defined logical axes (such as "Joint 1") to specific physical resources. For example, it can map to a specific PDO entry of EtherCAT slave station 1, or to a pulse generator channel of an FPGA.

[0039] PDO Data Binding Manager: During system initialization, it establishes a two-way binding relationship between variables in the control kernel layer (such as "target position_joint1") and the corresponding process data objects of the EtherCAT slave station according to the configuration, so as to realize automatic periodic data exchange.

[0040] DC Synchronization Manager: Works in conjunction with the DC unit within the FPGA to ensure that all slave stations are clock-aligned at the start of each control cycle, and that sampling and instruction execution are strictly synchronized.

[0041] Safety Sentinel: Running as a high-priority interrupt service routine, it checks the follow-up error and actual torque fed back from the slave station at the end of each control cycle and compares them with the soft limits set in the software. Any exceeding of these limits will immediately trigger the safety state machine, cutting off power output and locking the safety relays.

[0042] Control kernel layer: Kinematics Library: Implements the DH parameter model for a six-axis robot, providing the forward_kinematics and inverse_kinematics functions to perform position transformations between joint space and Cartesian space.

[0043] CIA402 State Machine: Maintains a state machine (such as "Disabled", "Enabled", "Running", "Fault") that conforms to the IEC61800-7-201 standard for each drive axis, strictly managing the power-on, enable, operation and fault reset process of the servo.

[0044] Open source EtherCAT master stack: Integrates and optimizes open source protocol stacks such as IgHEtherCATMaster or SOEM, and is responsible for managing network initialization, slave scanning, SDO configuration, and periodic PDO data exchange loops.

[0045] Algorithm plugin layer: Define a unified C++ abstract base class, such as `class TrajectoryPlanner{public:virtualboolplan(Constrain&c,Trajectory&t)=0;}`. Users can inherit from this class to implement their own algorithms (such as time-optimal trajectory planning or dynamics-based trajectory optimization), and compile them into dynamic link libraries (.so files). At platform runtime, the configuration file specifies which plugin to load, enabling "hot-plugging" of algorithms.

[0046] Application layer: Instruction parser: Parses standard G code or custom simple instruction sets (such as MovL, MovJ).

[0047] Scientific data interface: Provides a RESTful API or file system interface, allowing external programs (such as MATLAB, Python scripts) to upload CSV files containing N×6 dimensional trajectory point data.

[0048] Remote HMI: The control interface is developed based on web technologies (such as Vue.js + WebSocket), allowing users to monitor the robot's status, modify parameters, and manually teach it in real time through a browser.

[0049] File Preload Manager: In "Scientific Research Verification Mode", the specified CSV trajectory file is read into a contiguous array in memory at once when the system starts, avoiding access to the file system in the real-time control loop.

[0050] 3. Specific implementation of control methods The control method of the present invention provides two implementation paths: (1) Industrial production mode (online closed-loop control), such as Figure 4 As shown ① Command input: The operator sends a motion command through the web interface or handheld terminal, such as "MovJ, target pose {X,Y,Z,Rx,Ry,Rz}, velocity 50%, acceleration 20%".

[0051] ② Instruction parsing: The application layer instruction parser converts the text instruction into an internal structure, which includes constraints such as target pose, maximum velocity, and acceleration.

[0052] ③ Trajectory planning: The control kernel calls the currently loaded trajectory planning plugin (e.g., a cubic spline planner). The plugin calculates a coarse interpolation path for the time-position sequence based on the current pose, target pose, and constraints.

[0053] ④ Real-time interpolation: In a 1ms real-time task cycle, the control kernel calls a real-time interpolation plugin (such as an S-curve speed planner) to refine the coarse interpolation path into specific position commands (pulse count or EtherCAT target position value) that need to be sent to the 6 servo axes in the current cycle.

[0054] ⑤ HAL and Bus Output: The axis mapper in the hardware abstraction layer writes the position commands of the six logical axes into the transmit buffer of the EtherCAT master stack via PDO data binding; or writes them into the pulse generator register of the FPGA via the AXI bus. The EtherCAT IP core or pulse generator of the FPGA sends the commands to the servo driver at the next precise hardware synchronization moment.

[0055] ⑥ Feedback and Monitoring: The driver executes and reports the actual location, and the data returns via the same path. The security sentinel monitors errors in real time and controls the kernel to update the state machine.

[0056] (2) Scientific research verification mode (offline open-loop transparent transmission), such as Figure 5 As shown: ① Data generation: The algorithm simulation is completed in MATLAB / Simulink using the Robotics Toolbox, generating a high-density joint space trajectory matrix Data[N*6], where N is the number of points.

[0057] ② Data preparation: Using the tools and scripts provided by the platform, the joint angle (radians) is converted into the target position value (pulse count or EtherCAT position command unit) of the corresponding servo drive and packaged into a standard format CSV file.

[0058] ③ Data preloading: Upload the CSV file via the platform's data API or file management interface. When the system starts the "Scientific Research Verification" task, the entire file content is loaded into a specified array in the PS terminal's DDR memory at once.

[0059] ④ High-fidelity playback: The system creates a high-priority real-time task that executes at a fixed interval of 1ms (or less). This task bypasses all trajectory planning and interpolation algorithm logic and directly reads the next set (6) of target position values ​​from a memory array pointed to by a global index pointer.

[0060] ⑤ Pass-through execution: The read position value is directly output to the servo driver through the hardware abstraction layer in the same way as step 5 of the "industrial production mode". The whole process is like playing a pre-recorded high-fidelity audio, minimizing interference from the controller's own algorithm.

[0061] ⑥ Data Recording and Analysis: During transparent transmission execution, the system synchronously records the actual position feedback value of the servo driver for each cycle through the same channel and stores it as another data file. After the experiment, this feedback data can be compared with the theoretical trajectory in MATLAB to accurately evaluate the algorithm's performance in a real physical system.

[0062] further: The hardware system adopts a modular architecture with a heterogeneous core board and a standardized expansion board. The core board integrates an ARM processor as a high-level computing unit and a field-programmable gate array (FPGA) as a hard real-time control unit. The FPGA is configured as an EtherCAT hardware master station to achieve high-precision synchronization and multi-axis pulse generation. The expansion board provides a rich set of standardized interfaces, covering various industrial buses such as EtherCAT, PROFINET, and CANopen, and is compatible with traditional drive interfaces such as pulse / direction and encoders. It also integrates a wireless communication module and an AI accelerator card expansion interface, achieving broad compatibility and plug-and-play functionality from traditional devices to intelligent peripherals.

[0063] The software system adopts a four-layer decoupled architecture, which includes, from bottom to top: 1. Hardware Abstraction Layer: Responsible for unified mapping of physical interfaces and driver adaptation, and integrates security monitoring functions.

[0064] 2. Control kernel layer: Integrates open-source EtherCAT master stack, robot kinematics library and standard driver state machine to form a real-time control center.

[0065] 3. Algorithm Plug-in Layer: Standardized interfaces are defined through C++ abstract classes, allowing users to customize or replace core algorithms such as trajectory planning, real-time interpolation, and dynamics compensation in the form of plug-ins, forming an "algorithm sandbox".

[0066] 4. Application Layer: Provides multi-mode human-computer interaction, instruction parsing, data management and communication services.

[0067] This application implements a dual-path control method: Industrial production path: Complete the entire closed loop from high-level instruction parsing and pluggable algorithm planning to real-time bus / pulse control, meeting the autonomy and stability requirements of actual operation.

[0068] Scientific research verification path: It supports preloading trajectory data matrices generated by external simulation tools (such as MATLAB) into memory and performing high-fidelity, data pass-through execution without intermediate algorithm intervention in real time, ensuring a high degree of consistency between simulation and physical experiment results and greatly improving the rigor of scientific research verification.

[0069] Any aspects not covered in this invention are applicable to existing technologies.

Claims

1. An open control platform for an industrial six-axis robot, characterized in that, It includes hardware architecture and software architecture. The hardware architecture adopts a modular design that combines heterogeneous core modules and standardized extension modules. The software architecture is a layered and decoupled structure. The platform can switch between two independent control paths: industrial-grade operation and scientific research-grade verification. It also has hardware interface expansion capabilities and software algorithm customization capabilities.

2. The open control platform for an industrial six-axis robot according to claim 1, characterized in that, The heterogeneous core module integrates an ARM processing unit and an FPGA processing unit. The ARM processing unit is equipped with a real-time optimized operating system and undertakes high-level logic decision-making, trajectory planning calculation, communication protocol parsing, and human-machine interaction management tasks. The FPGA processing unit is configured as an EtherCAT hardware master station to realize distributed clock synchronization control and multi-axis pulse signal hard timing generation functions, ensuring the real-time motion control requirements with nanosecond-level synchronization accuracy.

3. The open control platform for an industrial six-axis robot according to claim 1, characterized in that, The standardized expansion module integrates multiple hardware interfaces, including at least one industrial fieldbus interface compatible with EtherCAT, PROFINET or CANopen protocols, six or more pulse direction output interfaces and corresponding encoder feedback input interfaces, isolated digital input / output interfaces, at least one extended communication interface containing CAN bus or RS485, at least one wireless communication interface supporting Wi-Fi or 5G standards, and an AI acceleration expansion interface for accessing the coprocessor.

4. The open control platform for an industrial six-axis robot according to claim 1, characterized in that, The software architecture is divided into four layers from bottom to top: the hardware abstraction layer, which realizes the mapping and matching of logical control axes and physical drive units, the binding of process data objects, and integrates real-time safety monitoring functions; the control kernel layer, which has built-in open-source EtherCAT master station protocol stack, robot kinematics calculation library, and servo driver state machine management module conforming to CIA402 standard; and the algorithm plug-in layer, which builds standardized algorithm interface specifications, is equipped with a plug-in management unit, and supports dynamic loading of algorithm function modules such as trajectory planning, real-time interpolation, and dynamics compensation. The application layer provides instruction parsing tools, data application programming interfaces, remote human-computer interaction interfaces, and file preloading management functions.

5. The open control platform for an industrial six-axis robot according to claim 4, characterized in that, The implementation method of the scientific research-grade verification control path is as follows: it supports the import of N×6-dimensional trajectory matrix files generated by external simulation tools, characterized by joint spatial position or pulse number; during the system initialization phase, the trajectory matrix data is preloaded into the controller's global memory array at once; during the control loop, the trajectory point data is read directly from the memory array according to a preset period, skipping the planning and interpolation logic of the algorithm plug-in layer, and outputting it directly to the servo drive network through the hardware abstraction layer.

6. The open control platform for an industrial six-axis robot according to claim 1, characterized in that, The industrial-grade operation control path is implemented as follows: receiving high-level control commands sent by a host computer or teaching device, converting them into structured data containing target pose and motion constraints through an application layer command parsing tool; calling the trajectory planning algorithm loaded by the algorithm plugin layer to generate a coarse interpolation path; within the real-time control cycle, refining the coarse interpolation path into target position commands for each control cycle through a real-time interpolation algorithm, and outputting them through an industrial bus or pulse interface via a hardware abstraction layer to complete the closed-loop control process.

7. An industrial six-axis robot control method applicable to the open control platform described in any one of claims 1-6, characterized in that, Includes the following steps: Step 1: Select the control path according to the actual application scenario. If it is an industrial production scenario, start the industrial-grade operation control path and execute steps 2-4; if it is a scientific research and verification scenario, start the scientific research and verification control path and execute steps 5-7. Step 2: Receive high-level control commands from the host computer or teaching device, and convert them into structured data containing the target pose and motion constraints using an application-layer command parsing tool; Step 3: Call the trajectory planning algorithm already loaded in the algorithm plugin layer to generate a coarse interpolation path based on the structured data; Step 4: During the real-time control cycle, the real-time interpolation algorithm of the algorithm plug-in layer is called to refine the coarse interpolation path into the target position instructions of each servo drive unit. The instructions are then output through the industrial bus or pulse interface via the hardware abstraction layer. At the same time, the operating status is monitored in real time through the safety monitoring module to complete the closed-loop control. Step 5: In scientific research verification scenarios, the trajectory matrix file generated by external simulation tools is preloaded into the controller memory; Step 6: Start the high-fidelity playback task, read trajectory point data from memory at fixed intervals, and output it directly to the driver network through the hardware abstraction layer; Step 7: Synchronously record the feedback data of the servo drive unit for subsequent algorithm verification and analysis.

8. The industrial six-axis robot control method according to claim 7, characterized in that, During the execution of steps 4 and 6, the safety monitoring module of the hardware abstraction layer detects the following error, torque parameters and limit status in real time. When the detected data exceeds the preset threshold, the safety protection mechanism is immediately triggered, cutting off the power output and locking the safety relay.

9. The industrial six-axis robot control method according to claim 7, characterized in that, It supports remote human-machine interaction at the application layer, enabling online modification of control parameters, real-time monitoring of system status, cloud synchronization of control programs, and version management.

10. The industrial six-axis robot control method according to claim 7, characterized in that, In the industrial-grade operation control path, the control kernel layer manages each slave driver through the integrated EtherCAT master protocol stack in a periodic synchronous data exchange manner. With the help of the distributed clock mechanism implemented by FPGA, it ensures that the control commands of all axes are executed synchronously within microseconds.

Images