An embodied robot communication device based on ethercat multi-master station

Abstract

The application is suitable for the technical field of robot control communication, and provides a somatic robot communication device based on an EtherCAT multi-master station, which comprises an EtherCAT multi-master station chip, a main processor and a plurality of EtherCAT slave stations. The multi-master station chip divides all the slave stations into a plurality of independent EtherCAT sub-networks which are isolated from each other in the physical layer and the data link layer; a plurality of independent message transceiving control units are integrated in the multi-master station chip, each of which drives a sub-network independently; a distributed clock synchronization unit is integrated in the multi-master station chip, which realizes the hardware level synchronization of each sub-network; and a memory management unit is integrated in the multi-master station chip, which remaps the scattered memory addresses into continuous addresses and realizes high-speed communication with the main processor through a PCIE interface. Through the parallel of multiple networks and the isolation of faults, the communication cycle is significantly shortened, the system reliability is enhanced, the motion coordination is ensured, and the problems of the performance bottleneck and poor reliability of the existing single master station architecture are solved.

Description

Technical Field

[0001] This invention belongs to the field of robot control and communication technology, and particularly relates to a communication device for embodied robots based on EtherCAT multi-master stations, which is used to realize regional independent real-time control and communication of the joint motors of the limbs of embodied robots. Background Technology

[0002] Embossed robots are integrated intelligent equipment that combines mechanical structure, motion control, network communication, and sensing monitoring technologies. They are widely used in service, industrial inspection, and special operations scenarios. The robot's limbs have numerous moving joints, each independently controlled by a drive motor. EtherCAT, due to its advantages of high real-time performance, high bandwidth, and low latency, has become the mainstream industrial Ethernet communication protocol for joint motion control in embossed robots.

[0003] In existing technologies, embodied robots generally use a single EtherCAT soft master station. The master station module is centrally located on the chest of the embodied robot, and EtherCAT slave station chips are deployed in each joint of the limbs. The entire communication network is set up with only one master station system, and the network topology is mainly divided into two types: linear topology and star topology.

[0004] In a linear topology, the master station connects all the slave stations in series. Each joint node requires two communication cables. This results in complex wiring for the robot's limbs, with significant line bending and losses, which is detrimental to the robot's flexible movement and subsequent maintenance. The topology diagram is shown below. Figure 1 As shown; the star topology uses a splitter to achieve one main station with multiple outputs. Although this simplifies cabling to some extent, it does not change the core architecture of a single main station and a single network. Its topology diagram is as follows. Figure 2 As shown.

[0005] Currently, mainstream unibody robots have around 40 EtherCAT slave stations configured in their limb joints. All slave nodes are connected to the same EtherCAT communication network. The existing single soft master architecture exposes two major flaws: 1. Poor real-time communication: The number of communication nodes in the network is huge, and a single master station needs to poll and process all slave station data, which significantly extends the data transmission and reception cycle and increases communication latency, making it difficult to meet the real-time control requirements of high dynamic and high precision robot joints; 2. Weak system fault tolerance: All slave nodes operate on the same physical layer and data link layer network, with interconnected networks and no isolation mechanism. When any slave node fails, disconnects, or experiences signal abnormalities, the fault will affect the entire EtherCAT communication network, causing sudden vibrations in other nodes, resulting in extremely low equipment safety and stability.

[0006] As androids evolve towards higher degrees of freedom, higher motion precision, and higher reliability, the single EtherCAT soft master architecture can no longer meet the requirements of multi-joint, high real-time, and high fault tolerance applications. Therefore, developing an EtherCAT multi-master communication device that can split the communication network, shorten the communication cycle, achieve fault isolation, and ensure the synchronization of multi-joint movements has become an urgent technical problem to be solved in this field. Summary of the Invention

[0007] The purpose of this invention is to provide an embodied robot communication device based on EtherCAT multi-master station, which aims to solve the problems of high communication latency, easy spread of node failures across the network, and poor system fault tolerance in the existing single EtherCAT soft master station architecture.

[0008] The present invention is implemented as follows: a communication device for an embodied robot based on EtherCAT multi-master station, the embodied robot communication device including an EtherCAT multi-master station chip, a main processor and several EtherCAT slave stations deployed at the motors of each joint of the embodied robot's limbs. The EtherCAT multi-master chip communicates with all the EtherCAT slave stations respectively, and divides all the EtherCAT slave stations into multiple independent EtherCAT communication sub-networks that are isolated from each other at the physical layer and data link layer. Different EtherCAT communication sub-networks do not affect each other. The EtherCAT multi-master chip integrates a multi-channel EtherCAT message transmission and reception control unit, a distributed clock synchronization unit, a memory management unit, and a high-speed interface unit. Each of the EtherCAT message transmission and reception control units drives an EtherCAT communication subnetwork independently, and independently completes the reception and transmission control of synchronous and asynchronous messages within the corresponding subnetwork; The distributed clock synchronization unit is electrically connected to each EtherCAT message transceiver control unit and is used to perform hardware-level clock synchronization on multiple independent EtherCAT communication sub-networks, so that each sub-network outputs synchronized control commands. The memory management unit is connected to each EtherCAT message transceiver control unit and is used to map the memory addresses corresponding to each EtherCAT message transceiver control unit to consecutive memory addresses through address remapping, so that the main processor can access them continuously. One end of the high-speed interface unit is connected to the memory management unit, and the other end of the high-speed interface unit communicates with the main processor to realize high-speed data transmission between the EtherCAT multi-master chip and the main processor.

[0009] A further technical solution of the present invention is that the number of EtherCAT slave stations configured in each EtherCAT communication sub-network is ≥1. This splits the original single network of 40 nodes into multiple smaller networks of 8 nodes, balancing the effect of chip complexity and shortened communication cycle.

[0010] A further technical solution of the present invention is that: after the distributed clock synchronization unit completes the clock synchronization of multiple EtherCAT communication sub-networks, the overall clock synchronization error is ±100ns. This ensures that the limb joints controlled by different sub-networks achieve nanosecond-level synchronization at the hardware level, guaranteeing a high degree of coordination and consistency in the robot's overall movements.

[0011] A further technical solution of the present invention is that the main processor adopts a high-performance processor based on Intel x86 architecture, ARM architecture, or RISC-V architecture, and the high-speed interface unit is adapted to the communication protocol specifications of the Intel x86 architecture, ARM architecture, or RISC-V architecture processor. This fully utilizes the mature high-performance computing ecosystem to provide powerful computing support for complex embodied intelligence algorithms.

[0012] A further technical solution of the present invention is that the multiple EtherCAT message transceiver control units are configured to operate independently. When a communication failure or node abnormality occurs in the EtherCAT communication sub-network driven by one of the EtherCAT message transceiver control units, all other EtherCAT communication sub-networks maintain normal communication and control operations without interrupting the communication of the remaining sub-networks. This achieves physical isolation of network failures, greatly improving the risk resistance and operational reliability of the entire robot system.

[0013] A further technical solution of the present invention is as follows: the EtherCAT multi-master chip is fixedly installed inside the torso of the android; each EtherCAT slave station is mounted one-to-one on the drive motor of the android's limb joints, used to collect joint motor operating status data and receive motion control commands from the EtherCAT communication subnetwork. This clarifies the physical layout of the device, shortens the communication distance, and facilitates the collection of joint motor operating status data and the reception of motion control commands.

[0014] A further technical solution of the present invention is as follows: the message transmission and reception control unit sets up independent message buffers for synchronous messages and asynchronous messages. Synchronous messages are used to transmit real-time motion control commands for robot joints, while asynchronous messages are used to transmit robot joint status monitoring data. Separating data streams with different real-time requirements avoids non-real-time data blocking real-time control commands, further ensuring the real-time performance of the control.

[0015] A further technical solution of the present invention is that the memory management unit is configured with an address mapping table, which converts scattered memory addresses into contiguous memory addresses, enabling the main processor to perform fast addressing and data reading / writing. This provides the main processor with a unified and contiguous address space for access, simplifies software design, and improves data reading / writing efficiency.

[0016] A further technical solution of the present invention is: the high-speed interface unit adopts the standard PCIe bus protocol or other high-speed communication interface protocols to provide a high-speed transmission channel between the EtherCAT multi-master chip and the main processor. This meets the high-speed exchange requirements of massive motion control data and status feedback data between the main processor and the multi-master chip.

[0017] A further technical solution of the present invention is that: the multiple EtherCAT communication sub-networks respectively correspond to the limb regions of the left arm joint group, right arm joint group, left leg joint group, and right leg joint group of the embodied robot. Each joint in the joint group of each limb region communicates with the EtherCAT multi-master chip through its EtherCAT communication sub-network. This provides an intuitive and efficient sub-network partitioning method that conforms to the robot's shape and structure, facilitating system-level design, debugging, and maintenance.

[0018] The beneficial effects of this invention are as follows: Compared with the prior art, the embodied robot communication device based on EtherCAT multi-master station provided by this invention has the following beneficial effects: 1. Improve real-time communication and shorten communication cycle. This application splits the approximately 40 EtherCAT slave stations of the whole machine into multiple independent communication sub-networks, with the number of slave stations in a single sub-network controlled to 8. This significantly reduces the number of communication nodes in a single network, reduces the computational overhead of message polling and data forwarding, effectively shortens the EtherCAT communication cycle, improves the real-time performance of joint control, and meets the requirements of high-precision and high-dynamic robot motion control.

[0019] 2. Achieve physical isolation of faults and improve system fault tolerance. Each EtherCAT communication subnetwork is completely isolated at the physical layer and data link layer, and each message sending and receiving control unit operates independently. When a joint or subnetwork experiences a disconnection, hardware failure, or communication anomaly, the fault is limited to the current subnetwork and will not spread to other subnetworks. The remaining limbs of the robot can continue to function normally, completely solving the problem of "one-point failure, whole-network paralysis" in existing single-master-station networks, and significantly improving the safety and stability of equipment operation.

[0020] 3. Hardware-level high-precision clock synchronization ensures coordinated operation. This application integrates a distributed clock synchronization unit, using hardware to synchronize the clocks of all independent sub-networks, with synchronization errors controlled within ±100ns. Even if the limbs belong to different communication sub-networks, each joint can still receive synchronization control commands, ensuring coordinated and continuous limb movements and avoiding problems such as misalignment or asynchrony in limb movements.

[0021] 4. Optimize memory access efficiency and improve overall system performance. The memory management unit remaps the discrete memory addresses of the multi-channel message transceiver control unit to continuous memory addresses. In conjunction with the address mapping table, it enables fast addressing. The main processor can continuously and efficiently read and write data, reducing the performance loss caused by address jumps and further improving the overall data interaction efficiency.

[0022] 5. High-speed bus adaptation, strong compatibility It adopts a standard PCIe high-speed interface to realize data interaction between the chip and the main processor, providing gigabit-level transmission bandwidth. At the same time, it is compatible with mainstream Intel x86 architecture, ARM architecture, or RISC-V architecture high-performance processors, with good hardware compatibility. It can be directly applied to existing mainstream robotic main control platforms, reducing equipment modification costs and implementation difficulties.

[0023] 6. The network is rationally zoned, and cabling and maintenance are convenient. The robot's communication sub-networks are divided according to the four limb regions: left arm, right arm, left leg, and right leg. The network partitions match the robot's physical structure, and the wiring is neat and clear. At the same time, the partitioned network facilitates fault location and single-point repair, making subsequent equipment maintenance more convenient. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the linear topology used in the EtherCAT communication network for embodied robots in existing technologies.

[0025] Figure 2 This is a schematic diagram of the star topology structure of the EtherCAT communication network for embodied robots in the existing technology.

[0026] Figure 3 This is a block diagram of the internal modules of the EtherCAT multi-master chip provided in an embodiment of the present invention.

[0027] Figure 4 This is a schematic diagram of the network topology of the embodied robot communication device based on EtherCAT multi-master station provided in an embodiment of the present invention. Detailed Implementation

[0028] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0029] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0030] like Figure 3 and Figure 4 As shown, the embodied robot communication device based on EtherCAT multi-master station provided by the present invention includes: an EtherCAT multi-master station chip, a main processor, and multiple EtherCAT slave stations.

[0031] Physically, the EtherCAT multi-master chip is fixedly installed inside the body of the robot, serving as the communication hub. The main processor uses a high-performance processor based on Intel x86, ARM, or RISC-V architecture, and connects to the EtherCAT multi-master chip via a PCIe bus. Multiple EtherCAT slave stations are mounted one-to-one on the drive motors of each joint in the robot's limbs.

[0032] In terms of logical architecture, the EtherCAT multi-master chip divides all EtherCAT slave stations 300 into multiple EtherCAT communication sub-networks. For example, ... Figure 4 As shown, the robot's left arm joint group can be divided into sub-network one, the right arm joint group into sub-network two, the left leg joint group into sub-network three, and the right leg joint group into sub-network four. Each sub-network is isolated from each other at both the physical layer (e.g., independent Ethernet ports) and the data link layer (e.g., independent MAC addresses), so their communication does not interfere with each other.

[0033] like Figure 3 As shown, the EtherCAT multi-master chip integrates several key functional units: Multi-channel EtherCAT message transceiver control unit: The chip integrates N (N≥4) completely independent transceiver control units. Each transceiver control unit drives one EtherCAT communication subnetwork. For example, the first transceiver control unit is specifically responsible for transmitting and receiving synchronous and asynchronous messages with all slave stations in subnetwork one (left arm). Its function is to independently manage the communication timing and data exchange of its subnetwork, thus completely decoupling the communication process of each subnetwork in hardware.

[0034] Distributed Clock Synchronization Unit: This unit is connected to each transceiver control unit via hardware circuitry. Its core function is to achieve high-precision synchronization of all independently operating EtherCAT sub-networks. During operation, the distributed clock synchronization unit selects a reference clock (e.g., the clock of the first sub-network) and dynamically compensates for transmission delays and drift between other sub-networks and the reference clock through hardware mechanisms, ultimately ensuring that the control commands output by all sub-networks are strictly aligned in time. Experiments show that this unit can achieve an overall clock synchronization error within ±100ns. Its role is that, although the left arm and right leg are controlled by different networks and different transceiver units, the time difference between the command reaching the joint motors is negligible when performing coordinated actions such as "waving" and "walking," ensuring the naturalness and coordination of the robot's movements.

[0035] Memory Management Unit: This unit internally configures an address mapping table. Since each of the multiplexer control units has its own independent memory region (such as FIFO or RAM) for buffering transmitted and received data, these memory addresses are physically dispersed. To facilitate the main processor's fast access to data in all subnets, the memory management unit uses address remapping technology to map these dispersed physical memory addresses to a contiguous logical memory address space visible to the main processor. Its function is to allow the main processor to efficiently read and write control instructions and status data of all nodes through the PCIe interface, just like accessing a contiguous array, greatly simplifying the design of the software driver layer.

[0036] High-speed interface unit: One end of this high-speed interface unit connects to the memory management unit, and the other end connects to the main processor's PCIe controller. It conforms to the standard PCIe bus protocol or other high-speed communication interface protocols, providing a gigabit-level (or even higher) bidirectional high-speed transmission channel. Its function is to enable the main processor to quickly send complex embodied intelligence algorithms and generate large amounts of motion planning data to the EtherCAT multi-master chip through this interface, while also uploading massive amounts of status data from joint sensors in real time, eliminating data bottlenecks; its high-speed interface unit is a PCIe high-speed interface.

[0037] As a preferred implementation, each EtherCAT communication subnetwork is configured with approximately eight EtherCAT slave stations. For example, a typical robotic arm (containing multiple degrees of freedom such as shoulder, elbow, and wrist) typically has 6-8 joint motors. By limiting the number of nodes in each subnetwork to 8, the communication cycle can be effectively compressed to 1 / 5 or even less of the original single large network (40 nodes), while the hardware resource overhead within the chip remains within a reasonable range.

[0038] As another preferred implementation, to further improve system reliability, the multi-channel EtherCAT message transceiver control unit is configured to operate in a completely independent mode. Suppose that during robot operation, the transceiver control unit responsible for the "right leg" sub-network detects a disconnection of a slave node within its network. At this time, the control unit will attempt to reconnect or report an error, but its hardware failure will not affect the normal timing of the other control units. Therefore, the robot's arms and left leg can still maintain normal communication and control. Based on this capability, the robot can take safety actions such as "single-leg standing alarm" or "slow squatting," avoiding overall "paralysis" caused by a single point of failure. Its function is to provide hardware-level fault isolation capabilities unmatched by existing technologies.

[0039] Furthermore, the message transmission and reception control unit of this application has separate message buffers for synchronous and asynchronous messages. Synchronous messages (such as periodic position / speed commands) enter a high-priority queue to ensure real-time performance; asynchronous messages (such as non-periodic parameter configurations and status monitoring data) enter a normal queue. This differentiated processing mechanism further ensures that the real-time transmission of motion control commands is not interfered with by monitoring data.

[0040] The embodied robot communication device based on EtherCAT multi-master station provided in this application solves the core problems of long communication cycle and poor reliability in the prior art from the hardware level through innovative chip architecture and network partitioning method, and ensures high-precision synchronization between multiple networks, providing a key communication foundation for high-performance and high-reliability embodied robot systems. Example 1

[0041] This embodiment discloses a holographic robot communication device based on EtherCAT multi-master stations, which is the basic implementation method of this application.

[0042] The main body of this device consists of an EtherCAT multi-master chip, a main processor, and several EtherCAT slave stations distributed on the motors of each joint of the robot's limbs. The EtherCAT multi-master chip is fixedly installed inside the robot's torso, serving as the core control unit of the entire communication system. Each drive motor of each limb joint is equipped with a one-to-one EtherCAT slave station. The slave station collects real-time operating status data such as motor speed, current, and position, and receives motion control commands from the communication network to drive the motors to complete the actions.

[0043] This device utilizes an EtherCAT multi-master chip to divide all EtherCAT slave stations into multiple independent EtherCAT communication sub-networks that are completely isolated at the physical layer and data link layer. Data cannot flow between these sub-networks, and interference is mutually isolated.

[0044] The EtherCAT multi-master chip integrates four core functional modules: a multi-channel EtherCAT message transmission and reception control unit, a distributed clock synchronization unit, a memory management unit, and a PCIe high-speed interface.

[0045] The EtherCAT message transceiver control unit integrates multiple independent message transceiver control units within this chip. Each unit drives one EtherCAT communication sub-network, and each unit operates independently without interference. Each unit has an independent buffer area to store synchronous and asynchronous messages: synchronous messages carry real-time motion control commands for robot joints, with the highest priority to ensure immediate delivery of action commands; asynchronous messages carry joint motor status monitoring and fault reporting data for equipment status inspection. This unit independently completes the reception, parsing, forwarding, and transmission of all messages within its sub-network, achieving autonomous communication management within a single network. It implements distributed management of the communication network, splits the communication load, and differentiates message types for buffer management, ensuring real-time delivery of control commands and stable uploading of status data.

[0046] The distributed clock synchronization unit is electrically connected to all EtherCAT message transmission and reception control units within the chip, serving as a global clock reference to perform hardware-level clock synchronization for all independent EtherCAT communication sub-networks. After synchronization calibration, the clock synchronization error of the entire network is stably controlled within ±100ns. All sub-networks output control commands based on the same clock reference, ensuring that the timing of joint movements of different limbs such as the left arm, right arm, left leg, and right leg of the robot is completely consistent. This solves the problem of asynchronous clocks among multiple independent networks, achieving nanosecond-level high-precision synchronization in hardware, ensuring coordinated movements and uniform posture of the robot's limbs, and avoiding movement misalignment and stuttering.

[0047] The memory management unit connects to all EtherCAT message transmission and reception control units and is configured with a dedicated address mapping table. Since the original memory addresses of the various message transmission and reception control units are scattered and disordered, the memory management unit uses address remapping technology to convert all discrete memory addresses into a unified contiguous memory address segment. When the main processor accesses data, it can perform batch reads and writes and fast addressing according to contiguous addresses, eliminating the need for repeated jumps to discrete addresses. This optimizes the memory layout, reduces the main processor's data access overhead, and improves the overall data interaction speed and computational efficiency.

[0048] The PCIe high-speed interface connects to the memory management unit on one end and the main processor on the other. The interface uses the standard PCIe bus protocol to build a gigabit-level high-speed data transmission channel. In this embodiment, the main processor is an Intel x86 architecture high-performance processor. The PCIe high-speed interface is fully compatible with the communication protocol of this architecture, enabling high-volume, low-latency data interaction between the EtherCAT multi-master chips and the main controller. This high-speed communication channel ensures bidirectional transmission of massive amounts of data across multiple networks, while also maintaining compatibility with mainstream x86 main controller platforms, improving hardware versatility.

[0049] The overall working logic of this embodiment is as follows: The main processor issues motion commands through the PCIe high-speed interface, which are then forwarded by the memory management unit to the corresponding EtherCAT message transceiver control unit. The message transceiver control unit sends the commands to the EtherCAT slave stations within the corresponding sub-network, driving the joint motors to move. The status data collected by each slave station is transmitted back in reverse, and after being aggregated by the chip, it is uploaded to the main processor. The distributed clock synchronization unit calibrates the clocks of all sub-networks throughout the process to ensure synchronized movements. When any node in a sub-network fails or communication is interrupted, only that sub-network stops working, while the other sub-networks continue to operate normally. Example 2

[0050] This embodiment further defines the preferred implementation method based on Embodiment 1.

[0051] Each EtherCAT communication subnetwork in this device is configured with eight EtherCAT slave stations. The robot has approximately forty joint slave stations in total across its four limbs, which are grouped into multiple subnetworks of eight slave stations each. By reducing the number of communication nodes in a single network to eight, the data processing load of a single master station unit is minimized, the message polling cycle is shortened, and the real-time communication performance of the EtherCAT network is significantly improved, making it suitable for the control requirements of high-speed moving joints. Example 3

[0052] This embodiment further defines the preferred implementation method based on Embodiment 1.

[0053] Multiple EtherCAT message transceiver control units operate independently, possessing fault isolation capabilities: when a message transceiver control unit and its driven EtherCAT communication subnetwork experience communication interruptions, slave node hardware malfunctions, signal interference, or other faults, the fault is limited to the current subnetwork. All other message transceiver control units and their corresponding communication subnetworks continue to operate normally, and the communication process does not need to be interrupted. This achieves horizontal isolation of network faults, preventing single-point failures from causing network-wide paralysis, significantly improving the fault tolerance and operational reliability of the embodied robot control system, and ensuring that the equipment can continue to operate even in scenarios with localized faults. Example 4

[0054] This embodiment further defines the preferred implementation method based on Embodiment 1.

[0055] The communication subnetwork is regionalized based on the robot's physical structure: multiple EtherCAT communication subnetworks correspond one-to-one with the four limb regions of the robot: left arm joint group, right arm joint group, left leg joint group, and right leg joint group. All joints within each limb region are connected to a dedicated, independent EtherCAT communication subnetwork, and each limb interacts with the EtherCAT multi-master chip through this dedicated subnetwork. The network partitioning corresponds one-to-one with the robot's physical structure, resulting in neat and orderly wiring, reducing losses from line crossings and bends. Simultaneously, faults can be directly located to the corresponding limb region, simplifying subsequent maintenance and troubleshooting processes and improving equipment operation and maintenance efficiency.

[0056] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A communication device for an embodied robot based on EtherCAT multi-master stations, characterized in that, The communication device for the android includes an EtherCAT multi-master chip, a main processor, and several EtherCAT slave stations located at the motors of each joint of the android's limbs. The EtherCAT multi-master chip communicates with all the EtherCAT slave stations respectively, and divides all the EtherCAT slave stations into multiple independent EtherCAT communication sub-networks that are isolated from each other at the physical layer and data link layer. Different EtherCAT communication sub-networks do not affect each other. The EtherCAT multi-master chip integrates a multi-channel EtherCAT message transmission and reception control unit, a distributed clock synchronization unit, a memory management unit, and a high-speed interface unit. Each of the EtherCAT message transmission and reception control units drives an EtherCAT communication subnetwork independently, and independently completes the reception and transmission control of synchronous and asynchronous messages within the corresponding subnetwork; The distributed clock synchronization unit is electrically connected to each EtherCAT message transceiver control unit and is used to perform hardware-level clock synchronization on multiple independent EtherCAT communication sub-networks, so that each sub-network outputs synchronized control commands. The memory management unit connects to each EtherCAT message transceiver control unit and is used to map the memory addresses corresponding to each EtherCAT message transceiver control unit to consecutive memory addresses through address remapping, so that the main processor can access them continuously. The memory management unit configures an address mapping table and completes the conversion of scattered memory addresses to consecutive memory addresses through the address mapping table, so that the main processor can perform fast addressing and data reading and writing. One end of the high-speed interface unit is connected to the memory management unit, and the other end of the high-speed interface unit communicates with the main processor to realize high-speed data transmission between the EtherCAT multi-master chip and the main processor.

2. The embodied robot communication device based on EtherCAT multi-master station according to claim 1, characterized in that, The number of EtherCAT slaves configured in each of the aforementioned EtherCAT communication subnetworks is ≥1.

3. The embodied robot communication device based on EtherCAT multi-master station according to claim 1, characterized in that, After the distributed clock synchronization unit completes the clock synchronization of multiple EtherCAT communication sub-networks, the overall clock synchronization error is ±100ns.

4. The embodied robot communication device based on EtherCAT multi-master station according to claim 1, characterized in that, The main processor adopts a high-performance processor with Intel x86 architecture, ARM architecture, or RISC-V architecture, and the high-speed interface unit is adapted to the communication protocol specifications of Intel x86 architecture, ARM architecture, or RISC-V architecture processors.

5. The embodied robot communication device based on EtherCAT multi-master station according to claim 1, characterized in that, The multiple EtherCAT message transceiver control units are configured to operate independently. When a communication failure or node abnormality occurs in the EtherCAT communication subnetwork driven by one of the EtherCAT message transceiver control units, all other EtherCAT communication subnetworks maintain normal communication and control operations without interrupting the communication of the other subnetworks.

6. The embodied robot communication device based on EtherCAT multi-master station according to claim 1, characterized in that, The EtherCAT multi-master chip is fixedly installed inside the body of the robot; each EtherCAT slave station is assembled one-to-one with the drive motor of the joint of the robot's limbs, used to collect joint motor operating status data and receive motion control commands from the EtherCAT communication sub-network.

7. The embodied robot communication device based on EtherCAT multi-master station according to claim 6, characterized in that, The message transmission and reception control unit sets up independent message buffers for synchronous messages and asynchronous messages. Synchronous messages are used to transmit real-time motion control commands for robot joints, while asynchronous messages are used to transmit robot joint status monitoring data.

8. The embodied robot communication device based on EtherCAT multi-master station according to claim 1, characterized in that, The high-speed interface unit adopts the standard PCIE bus protocol or other high-speed communication interface protocols to provide a high-speed transmission channel between the EtherCAT multi-master chip and the main processor.

9. The embodied robot communication device based on EtherCAT multi-master station according to claim 1, characterized in that, The multiple EtherCAT communication sub-networks correspond to the limb regions of the left arm joint group, right arm joint group, left leg joint group, and right leg joint group of the embodied robot, respectively. Each joint in the joint group in each limb region communicates with the EtherCAT multi-master chip through its EtherCAT communication sub-network.

Images