An integrated robot cerebellar controller

Abstract

The utility model relates to a kind of integrated robot cerebellum controller in the field of robot cerebellum controller, including equipment shell and built-in PCB board.Equipment shell adopts aluminium alloy material, is composed of upper cover plate, bottom cover plate, front cover plate and rear cover plate, two side inner walls are provided with slot for PCB board sliding joint, outside is opened heat dissipation hole, and copper column is installed in the upper cover plate inside and contacts with PCB board to dissipate heat.PCB board is integrated high-density mainboard, carries AI chip, industrial communication interface, neural network acceleration chip, industrial multi-protocol exchange chip and wide-temperature memory grain, supports high-speed data processing and control execution, and PCB board is inserted with slot, and it is convenient for fault IO module to be directly plugged in and replaced, improve operation and maintenance efficiency, effectively solve the heat dissipation problem, enhance data processing and control execution capability.

Description

Technical Field

[0001] This utility model relates to the field of robot cerebellar controllers, specifically to an integrated robot cerebellar controller. Background Technology

[0002] The robot cerebellum controller plays a crucial role in ensuring the robot's stable operation and efficient work by undertaking key functions such as data processing, command transmission, and equipment control during the robot's operation.

[0003] Current robot cerebellar controllers generally employ general-purpose industrial PCs or multi-module discrete designs, resulting in excessive size and weight. The stacking of discrete hardware leads to a large space requirement for the controller, making integration into confined spaces such as the chest cavity or joints of small robots difficult, thus limiting lightweight robot design. Furthermore, PCB board installation is typically complex; when an I / O module fails, due to design flaws in the existing structure, it often requires system shutdown and complete disassembly to replace the faulty I / O module, reducing maintenance efficiency. Additionally, the PCB board generates significant heat during operation, causing excessively high internal temperatures in the controller, affecting its performance and stability. In terms of data processing and control execution, inter-module data interaction relies on an external bus, causing response delays and making it difficult to meet high-speed dynamic balancing requirements. The fixed hardware architecture also results in poor scalability, making it difficult to quickly adapt algorithms for different robot configurations such as legged or wheeled robots. Therefore, those skilled in the art have proposed an integrated robot cerebellar controller to address the problems mentioned in the background. Utility Model Content

[0004] The purpose of this invention is to address the above-mentioned shortcomings and provide an integrated robot cerebellum controller. This invention solves the technical problems of existing technologies, such as excessive device size and weight, complex PCB board installation, excessive heat generated by the PCB board during operation, data interaction between modules relying on external buses resulting in response delays, fixed hardware architecture leading to poor scalability, and difficulty in quickly adapting algorithms to different robot configurations such as legged or wheeled robots.

[0005] The objective of this utility model is achieved through the following means:

[0006] An integrated robot cerebellum controller includes a device housing, which comprises an upper cover plate, a bottom cover plate, a front cover plate, and a rear cover plate. Slots are formed on the inner sides of the side walls of the upper and bottom cover plates, and heat dissipation holes are formed on the outer sides of the side walls of the upper and bottom cover plates. Copper pillars are installed on the inner side of the upper cover plate. A PCB board is disposed inside the device housing, and a neural network acceleration chip, an industrial multi-protocol switching chip, and a wide-temperature memory chip are integrated on the PCB board. A magnetic I / O module is installed on the inner wall of the upper cover plate, and the magnetic I / O module is located at the end of the slot.

[0007] Furthermore, the device housing is made of aluminum alloy and covered with a nano-ceramic thermally conductive coating to optimize heat conduction. This architecture integrates heat dissipation, computing, and expansion capabilities in a very small volume through spatial stacking and functional layering, significantly solving the problem of excessive space occupation of traditional controllers. The PCB board is in contact with the copper pillars, which can conduct the heat of the PCB board to the top cover made of aluminum alloy. In addition, with the help of heat dissipation holes and external air convection, effective heat dissipation can be achieved.

[0008] Furthermore, the PCB board adopts an integrated high-density main PCB board, equipped with AI chip and industrial communication interface core components. The PCB board integrates neural network acceleration chip, industrial multi-protocol switching chip and wide-temperature memory chip. When the device is running, sensor data is input via isolated RS-485, and the NPU completes inference tasks such as defect detection in real time. The results are controlled by EtherCAT high-speed control actuator.

[0009] Furthermore, the side of the PCB board slides into the slot, and the insertion of the PCB board, together with the magnetic IO module, achieves automatic alignment and power-on. Therefore, during maintenance, faulty IO modules can be directly plugged in and replaced without stopping the entire machine, greatly improving system availability and maintenance efficiency.

[0010] The beneficial effects of this invention are as follows: The integrated robot cerebellum controller simplifies PCB board installation by creating slots on the inner walls of the upper and lower cover plates. Simultaneously, the PCB board, in conjunction with the magnetic IO module 9, automatically aligns and powers on. During maintenance, faulty IO modules can be directly plugged in and replaced without stopping the entire machine, greatly improving system availability and maintenance efficiency. Heat dissipation holes are created on the outer walls of the upper and lower cover plates, and copper pillars are installed on the inner side of the upper cover plate to contact the PCB board, conducting heat from the PCB board to the upper cover plate. This, combined with air convection through the heat dissipation holes, achieves effective heat dissipation. The PCB board is an integrated high-density main PCB board, integrating a neural network acceleration chip, an industrial multi-protocol switching chip, and wide-temperature memory chips, enhancing data processing and control execution capabilities. Attached Figure Description

[0011] Figure 1 This is a three-dimensional structural diagram of an integrated robot cerebellum controller according to the present invention;

[0012] Figure 2 This is a magnified view of a partial structure of an integrated robot cerebellum controller according to this utility model;

[0013] Figure 3 This is a schematic diagram showing the location of the magnetic I / O module in an integrated robot cerebellum controller according to this utility model;

[0014] In the diagram: 100, device housing; 1, top cover; 2, bottom cover; 3, front cover; 4, rear cover; 5, PCB board; 6, heat dissipation holes; 7, slot; 8, copper pillar; 9, magnetic I / O module. Detailed Implementation

[0015] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0016] In this embodiment, refer to Figure 1 - Figure 3 The integrated robot cerebellum controller, specifically implemented therein, includes a device housing 100. The housing 100 includes an upper cover plate 1, a bottom cover plate 2, a front cover plate 3, and a rear cover plate 4. Slots 7 are formed on the inner sides of the side walls of the upper cover plate 1 and the bottom cover plate 2, and heat dissipation holes 6 are formed on the outer sides of the side walls of the upper cover plate 1 and the bottom cover plate 2. Copper pillars 8 are installed on the inner side of the upper cover plate 1. A PCB board 5 is disposed inside the device housing 100. A neural network acceleration chip, an industrial multi-protocol switching chip, and wide-temperature memory chips are integrated on the PCB board 5. A magnetic... The magnetic IO module 9 is located at the end of the slot 7. The device housing 100 is made of aluminum alloy and covered with a nano-ceramic thermal conductive coating to optimize heat conduction. This architecture integrates heat dissipation, computing and expansion capabilities in a very small volume through spatial stacking and functional layering, which significantly solves the problem of excessive space occupation of traditional controllers. The PCB board 5 is in contact with the copper pillar 8. The copper pillar 8 can conduct the heat of the PCB board 5 to the upper cover plate 1 made of aluminum alloy. In addition, with the heat dissipation holes 6, it can effectively dissipate heat by convection with the outside air.

[0017] PCB board 5 adopts an integrated high-density main PCB board, equipped with AI chip and industrial communication interface core components. PCB board 5 integrates neural network acceleration chip, industrial multi-protocol switching chip and wide-temperature memory chip. When the device is running, sensor data is input via isolated RS-485, and the NPU completes inference tasks such as defect detection in real time. The results are controlled by EtherCAT high-speed control actuator. The side of PCB board 5 slides and engages with slot 7. The insertion of PCB board 5, together with magnetic IO module 9, achieves automatic alignment and power-on effect. Therefore, during maintenance, faulty IO module can be directly plugged in and replaced without stopping the machine and disassembling the whole machine, which greatly improves system availability and maintenance efficiency.

[0018] The working principle of this device: The integrated robot cerebellum controller includes a device housing 100 and a PCB board 5. The device housing 100 is composed of an upper cover plate 1, a bottom cover plate 2, a front cover plate 3, and a rear cover plate 4. Slots 7 are provided on the inner walls of both sides of the upper cover plate 1 and the bottom cover plate 2. The slots 7 allow the PCB board 5 to be inserted, making the installation of the PCB board 5 simpler. The insertion of the PCB board 5, together with the magnetic IO module 9, achieves the effect of automatic alignment and power-on. Therefore, during the maintenance phase, the faulty IO module can be directly plugged in and replaced without stopping the machine and disassembling the whole machine, which greatly improves the availability and maintenance efficiency of the system. Heat dissipation holes 6 are provided on the outer walls of both sides of the upper cover plate 1 and the bottom cover plate 2. Copper pillars 8 are installed on the inner side of the upper cover plate 1. The copper pillars 8 can contact the PCB board 5. The heat of the PCB board 5 can be conducted to the upper cover plate 1 made of aluminum alloy through the copper pillars 8. In conjunction with the heat dissipation holes 6 and the convection with the outside air, effective heat dissipation can be achieved.

[0019] Furthermore, the device housing 100 is made of aluminum alloy and covered with a nano-ceramic thermal conductive coating to optimize heat conduction. Moreover, this architecture integrates heat dissipation, computing and expansion capabilities in a very small volume through spatial stacking and functional layering, which significantly solves the problem of excessive space occupation of traditional controllers.

[0020] The PCB board 5 is an integrated high-density main PCB board 5, equipped with AI chip and industrial communication interface core components. The PCB board 5 integrates a neural network acceleration chip (NPU, computing power 4TOPS), an industrial multi-protocol switching chip (supporting Profinet / EtherCAT) and wide-temperature memory chips (-40℃~85℃). When the device is running, sensor data is input via isolated RS-485, and the NPU completes inference tasks such as defect detection in real time. The results are used to control the actuator at high speed via EtherCAT.

[0021] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. An integrated robotic cerebellum controller comprising a device housing (100), characterized in that: The device housing (100) includes an upper cover plate (1), a bottom cover plate (2), a front cover plate (3), and a rear cover plate (4). Slots (7) are provided on the inner side of the two side walls of the upper cover plate (1) and the bottom cover plate (2). Heat dissipation holes (6) are provided on the outer side of the two side walls of the upper cover plate (1) and the bottom cover plate (2). Copper pillars (8) are installed on the inner side of the upper cover plate (1). A PCB board (5) is provided inside the device housing (100). A neural network acceleration chip, an industrial multi-protocol switching chip, and a wide-temperature memory chip are integrated on the PCB board (5). A magnetic IO module (9) is installed on the inner wall of the upper cover plate (1). The magnetic IO module (9) is located at the end of the slot (7).

2. The integrated robot cerebellum controller of claim 1, wherein: The device housing (100) is made of aluminum alloy and its surface is covered with a nano-ceramic thermally conductive coating.

3. The integrated robot cerebellum controller of claim 1, wherein: The PCB board (5) adopts an integrated high-density main PCB board, equipped with an AI chip and industrial communication interface core components.

4. The integrated robot cerebellum controller of claim 1, wherein: The sensor data of the PCB board (5) is input via isolated RS-485, and the NPU completes the defect detection in real time. The result is executed by EtherCAT high-speed control.

5. The integrated robot cerebellum controller of claim 1, wherein: The side of the PCB board (5) is slidably engaged with the slot (7).

6. The integrated robot cerebellum controller of claim 1, wherein: The PCB board (5) is in contact with and bonded to the copper pillar (8).

Images