An intelligent motion system based on cloud-edge collaborative control

Abstract

The present invention particularly relates to an intelligent motion system based on cloud-edge collaborative control. The intelligent motion system based on cloud-edge collaborative control includes a central data cloud module, a wireless network transmission module and at least one intelligent mobile device, wherein the intelligent mobile device is equipped with an edge computing device. The intelligent motion system based on cloud-edge collaborative control realizes multi-sensor data fusion, cloud-edge real-time status communication, edge-side data transmission, dynamic tuning of artificial intelligence algorithms and offline autonomy on the edge side, saves space for edge computing devices to store large amounts of data, realizes functional decoupling of cloud-edge terminals, improves the training speed of the overall artificial intelligence algorithm, and can ensure the stable operation of the mobile platform.

Description

Technical Field

[0001] The present invention relates to the field of cloud-edge collaborative control technology, and in particular to an intelligent motion system based on cloud-edge collaborative control. Background Art

[0002] With the continuous development of cloud-edge collaborative control and artificial intelligence algorithms, the scene perception ability of intelligent motion platforms has been greatly improved. In the field of intelligent perception and motion planning, the interactive mode combined with cloud-edge collaborative control has also received more and more attention. Due to the frequent instability of the network environment and the complex and frequently changing characteristics of the platform's motion scene, a good intelligent motion platform has put forward higher requirements in terms of real-time data communication, reasonable allocation of computing power, and storage space for data volume. Especially in research topics in the field of unmanned driving with a higher level of intelligence, more attention is paid to the stability of equipment operation and the general interaction framework. At the same time, for scenarios that are disconnected from cloud communication connections, it is particularly important for edge devices to have a certain degree of edge autonomy.

[0003] In order to solve the problems of large vehicle-mounted communication delay, high computing load and insufficient data storage space, the present invention proposes an intelligent motion system based on cloud-edge collaborative control. Summary of the invention

[0004] In order to remedy the deficiencies of the prior art, the present invention provides a simple and efficient intelligent motion system based on cloud-edge collaborative control.

[0005] The present invention is achieved through the following technical solutions:

[0006] An intelligent motion system based on cloud-edge collaborative control, characterized in that it includes a central data cloud module, a wireless network transmission module and at least one intelligent mobile device, wherein the intelligent mobile device is equipped with an edge computing device;

[0007] The central data cloud module is composed of a metadata storage unit A and an integrated server for training artificial intelligence algorithms, and is used to realize the storage of device metadata, training of artificial intelligence algorithms, remote control and status tracking;

[0008] The edge computing module is responsible for preliminary processing of sensor signals of smart mobile devices, including but not limited to data desensitization and analytical reasoning, controlling the movement of smart mobile devices, and realizing key data transmission between the central data cloud module and the edge computing device through the wireless network transmission module, achieving a balance between latency, computing volume and data volume;

[0009] The intelligent mobile device comprises a sensor unit, a perception and computing unit, a decision-making and planning unit and a bottom control unit;

[0010] The sensor unit includes but is not limited to laser radar, camera, odometer, IMU inertial measurement device and collision sensor, which is responsible for collecting the physical information of the position and posture of the smart mobile device and the surrounding environment, generating fusion data of multiple sensor devices, and uploading key information;

[0011] The perception computing unit is responsible for realizing the positioning and perception of the vehicle in the world coordinate system and performing a preliminary evaluation of the motion state of the intelligent mobile device;

[0012] The decision-making planning unit is responsible for analyzing and processing the motion decisions of the intelligent mobile device to realize path planning and physical obstacle avoidance functions;

[0013] The bottom control unit controls the movement of the intelligent mobile device by installing a servo control unit, a power supply unit and a motor unit to complete the driving route planned by the controller.

[0014] The central data cloud module records the status and operation logs of edge devices, and carries an integrated server to train and optimize artificial intelligence algorithms, and sends them to edge computing devices for operation;

[0015] The metadata storage unit A uses a DB (database) to store and record the status information and operation health log of the edge device, and also backs up and records the key data;

[0016] The integrated server is responsible for completing the training and tuning of the artificial intelligence algorithms required for navigation of smart mobile devices. It uses the data uploaded by the edge computing device to mark and clean it, trains the current intelligent algorithm, and after achieving the required accuracy, sends it to the edge device to complete the intelligent control of the platform.

[0017] The edge computing device is composed of a metadata storage unit B, a device twin unit, a data transmission unit and a control unit. It marks and cleans the sensor signals of the smart mobile devices, records the status information and operation logs of the smart mobile devices, and realizes the functions of data monitoring, load balancing and offline autonomy.

[0018] The metadata storage unit B is responsible for completing the recording and log backup of the device status, realizing cloud-edge collaborative control, and realizing the offline autonomous function when the network status of the wireless network transmission module is interrupted, thereby ensuring the operating status of the edge computing device;

[0019] The device twin unit communicates data with the underlying smart mobile device to complete data cleaning and marking processing and monitor the status of the edge computing device. At the same time, it models and controls the smart mobile device model according to the commands issued by the control unit.

[0020] The data transmission unit is responsible for realizing the communication between the smart mobile device and the edge computing device, being compatible with the communication protocol supported by the edge computing device, and realizing state management and data backup;

[0021] The control module is the control center of the edge computing device, responsible for realizing the functions of data monitoring, load balancing, state preservation and offline autonomy.

[0022] The perception computing unit obtains environmental information data around the intelligent mobile device from the sensor unit, calculates and analyzes the obtained data, constructs the position and posture data of the intelligent mobile device relative to the world coordinates, and transmits it to the decision-making planning unit;

[0023] The planning unit receives the analysis data from the perception and computing unit, plans the route and avoids obstacles, and completes the real-time navigation between the starting point and the destination; at the same time, the path planning information, the distance information and the real-time navigation information are sent to the bottom control unit;

[0024] The bottom control unit receives the command signal from the decision-making and planning unit, controls the movement of the intelligent mobile device, and completes the driving, starting, stopping and speed command operations of the servo control unit on the motor unit.

[0025] The intelligent motion system of the present invention is based on cloud-edge collaborative control, and the implementation steps are as follows:

[0026] Step S1: The central data cloud module communicates with the edge computing device through the wireless network transmission module to implement the issuance of instructions and algorithm functions;

[0027] Step S2: The edge computing device deploys the algorithm function issued by the central data cloud module. At the same time, the central data cloud module monitors the health status of the edge computing device in real time, records and issues data;

[0028] Step S3: further send the command data of the edge computing device to the smart mobile device, which collects the physical information of its posture and surrounding environment through the sensor unit, and obtains the information of the posture and surrounding environment of the smart mobile device in the world coordinate system through data modeling and processing of the perception computing unit;

[0029] Step S4: the data obtained by the perception and computing unit is further transmitted to the decision-making and planning unit for processing, and the path planning and dynamic obstacle avoidance functions are realized through the artificial intelligence algorithm, and the motion control instructions are sent to the intelligent mobile device;

[0030] Step S5: After receiving the control command signal, the intelligent mobile device operates the motor unit through the servo control unit to complete the movement and state feedback of the intelligent mobile device.

[0031] The present invention is an intelligent motion system based on cloud-edge collaborative control. The intelligent mobile device feeds back the current posture information to the device twin unit. After data cleaning and marking, it is stored in the database of the edge computing device and sent to the central data cloud module. The central data cloud module stores the current device metadata and sampled environmental data, extracts key information and sends it to the integrated server as a training set for the artificial intelligence algorithm for training, and sends the optimized results to the edge computing device.

[0032] The beneficial effects of the present invention are: the intelligent motion system based on cloud-edge collaborative control realizes data fusion of multiple sensors, real-time status communication between cloud and edge, data transmission on the edge side, dynamic tuning of artificial intelligence algorithms and offline autonomy on the edge side, saving space for edge computing devices to store large amounts of data, realizing functional decoupling of cloud and edge, improving the training speed of the overall artificial intelligence algorithm, and being able to ensure the stable operation of the mobile platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0034] Attached Figure 1 Schematic diagram of the intelligent motion system based on cloud-edge collaborative control of the present invention. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0036] The intelligent motion system based on cloud-edge collaborative control includes a central data cloud module, a wireless network transmission module and at least one intelligent mobile device, wherein the intelligent mobile device is equipped with an edge computing device;

[0037] The central data cloud module is composed of a metadata storage unit A and an integrated server for training artificial intelligence algorithms, and is used to realize the storage of device metadata, training of artificial intelligence algorithms, remote control and status tracking;

[0038] The edge computing module is responsible for preliminary processing of sensor signals of smart mobile devices, including but not limited to data desensitization and analytical reasoning, controlling the movement of smart mobile devices, and realizing key data transmission between the central data cloud module and the edge computing device through the wireless network transmission module, achieving a balance between latency, computing volume and data volume;

[0039] The intelligent mobile device comprises a sensor unit, a perception and computing unit, a decision-making and planning unit and a bottom control unit;

[0040] The sensor unit includes but is not limited to laser radar, camera, odometer, IMU inertial measurement device and collision sensor, which is responsible for collecting the physical information of the position and posture of the smart mobile device and the surrounding environment, generating fusion data of multiple sensor devices, and uploading key information;

[0041] The perception computing unit is responsible for realizing the positioning and perception of the vehicle in the world coordinate system and performing a preliminary evaluation of the motion state of the intelligent mobile device;

[0042] The decision-making planning unit is responsible for analyzing and processing the motion decisions of the intelligent mobile device to realize path planning and physical obstacle avoidance functions;

[0043] The bottom control unit controls the movement of the intelligent mobile device by installing a servo control unit, a power supply unit and a motor unit to complete the driving route planned by the controller.

[0044] The central data cloud module records the status and operation logs of edge devices, and carries an integrated server to train and optimize artificial intelligence algorithms, and sends them to edge computing devices for operation;

[0045] The metadata storage unit A uses a DB (database) to store and record the status information and operation health log of the edge device, and also backs up and records the key data;

[0046] The integrated server is responsible for completing the training and tuning of the artificial intelligence algorithms required for navigation of smart mobile devices. It uses the data uploaded by the edge computing device to mark and clean it, trains the current intelligent algorithm, and after achieving the required accuracy, sends it to the edge device to complete the intelligent control of the platform.

[0047] The edge computing device is composed of a metadata storage unit B, a device twin unit, a data transmission unit and a control unit. It marks and cleans the sensor signals of the smart mobile devices, records the status information and operation logs of the smart mobile devices, and realizes the functions of data monitoring, load balancing and offline autonomy.

[0048] The metadata storage unit B is responsible for completing the recording and log backup of the device status, realizing cloud-edge collaborative control, and realizing the offline autonomous function when the network status of the wireless network transmission module is interrupted, thereby ensuring the operating status of the edge computing device;

[0049] The device twin unit communicates data with the underlying smart mobile device to complete data cleaning and marking processing and monitor the status of the edge computing device. At the same time, it models and controls the smart mobile device model according to the commands issued by the control unit.

[0050] The data transmission unit is responsible for realizing the communication between the smart mobile device and the edge computing device, being compatible with the communication protocol supported by the edge computing device, and realizing state management and data backup;

[0051] The control module is the control center of the edge computing device, responsible for realizing the functions of data monitoring, load balancing, state preservation and offline autonomy.

[0052] The perception computing unit obtains environmental information data around the intelligent mobile device from the sensor unit, calculates and analyzes the obtained data, constructs the position and posture data of the intelligent mobile device relative to the world coordinates, and transmits it to the decision-making planning unit;

[0053] The planning unit receives the analysis data from the perception and computing unit, plans the route and avoids obstacles, and completes the real-time navigation between the starting point and the destination; at the same time, the path planning information, the distance information and the real-time navigation information are sent to the bottom control unit;

[0054] The bottom control unit receives the command signal from the decision-making and planning unit, controls the movement of the intelligent mobile device, and completes the driving, starting, stopping and speed command operations of the servo control unit on the motor unit.

[0055] The intelligent motion system based on cloud-edge collaborative control is implemented in the following steps:

[0056] Step S1: The central data cloud module communicates with the edge computing device through the wireless network transmission module to implement the issuance of instructions and algorithm functions;

[0057] Step S2: The edge computing device deploys the algorithm function issued by the central data cloud module. At the same time, the central data cloud module monitors the health status of the edge computing device in real time, records and issues data;

[0058] Step S3: further send the command data of the edge computing device to the smart mobile device, which collects the physical information of its posture and surrounding environment through the sensor unit, and obtains the information of the posture and surrounding environment of the smart mobile device in the world coordinate system through data modeling and processing of the perception computing unit;

[0059] Step S4: the data obtained by the perception and computing unit is further transmitted to the decision-making and planning unit for processing, and the path planning and dynamic obstacle avoidance functions are realized through the artificial intelligence algorithm, and the motion control instructions are sent to the intelligent mobile device;

[0060] Step S5: After receiving the control command signal, the intelligent mobile device operates the motor unit through the servo control unit to complete the movement and state feedback of the intelligent mobile device.

[0061] In this intelligent motion system based on cloud-edge collaborative control, the intelligent mobile device feeds back the current posture information to the device twin unit. After data cleaning and labeling, it is stored in the database of the edge computing device and sent to the central data cloud module. The central data cloud module stores the current device metadata and sampled environmental data, extracts key information and sends it to the integrated server as a training set for the artificial intelligence algorithm, and sends the optimized results to the edge computing device.

[0062] Compared with the existing technology, this intelligent motion system based on cloud-edge collaborative control has the following characteristics:

[0063] First, it realizes functions such as multi-sensor data fusion, cloud-edge real-time status communication, edge-side data transmission, dynamic tuning of artificial intelligence algorithms, and offline autonomy on the edge side.

[0064] Second, the sensor unit is deployed to the smart mobile device, and the data collected by the sensor unit is sent to the edge computing device for status storage and environmental data upload, and then sent to the central data cloud module through the wireless network transmission module, saving space for edge computing devices to store large amounts of data.

[0065] Third, after uploading the data, training is performed on a remote server, which realizes the functional decoupling of cloud and edge, and improves the training speed of the overall artificial intelligence algorithm.

[0066] Fourth, it has the function of real-time data synchronization, which reduces the dependence on wireless network modules and realizes the offline autonomy of edge computing devices, which can ensure the stable operation of mobile platforms.

[0067] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent motion system based on cloud-edge collaborative control, characterized by: It includes a central data cloud module, a wireless network transmission module and at least one smart mobile device, wherein the smart mobile device is equipped with an edge computing device; The central data cloud module is composed of a metadata storage unit A and an integrated server for training artificial intelligence algorithms, and is used to realize the storage of device metadata, training of artificial intelligence algorithms, remote control and status tracking; The edge computing module is responsible for preliminary processing of sensor signals of smart mobile devices, including but not limited to data desensitization and analytical reasoning, controlling the movement of smart mobile devices, and realizing key data transmission between the central data cloud module and the edge computing device through the wireless network transmission module, achieving a balance between latency, computing volume and data volume; The edge computing device is composed of a metadata storage unit B, a device twin unit, a data transmission unit and a control unit. It records the status information and operation logs of the smart mobile devices by marking and cleaning the sensor signals of the smart mobile devices, and realizes the functions of data monitoring, load balancing and offline autonomy. The metadata storage unit B is responsible for completing the recording and log backup of the device status, realizing cloud-edge collaborative control, and realizing the offline autonomous function when the network status of the wireless network transmission module is interrupted, thereby ensuring the operating status of the edge computing device; The device twin unit communicates data with the underlying smart mobile device to complete data cleaning and marking processing and monitor the status of the edge computing device. At the same time, it models and controls the smart mobile device model according to the commands issued by the control unit. The data transmission unit is responsible for realizing the communication between the smart mobile device and the edge computing device, being compatible with the communication protocol supported by the edge computing device, and realizing state management and data backup; The control unit is the control center of the edge computing device, responsible for realizing the functions of data monitoring, load balancing, state keeping alive and offline autonomy; The intelligent mobile device comprises a sensor unit, a perception and computing unit, a decision-making and planning unit and a bottom control unit; The sensor unit is responsible for collecting the physical information of the position and posture of the intelligent mobile device and the surrounding environment, generating fusion data of multiple sensor devices, and uploading key information; The perception computing unit is responsible for realizing the positioning and perception of the vehicle in the world coordinate system and performing a preliminary evaluation of the motion state of the intelligent mobile device; The perception computing unit obtains environmental information data around the intelligent mobile device from the sensor unit, calculates and analyzes the obtained data, constructs the position and posture data of the intelligent mobile device relative to the world coordinates, and transmits it to the decision-making planning unit; The decision-making planning unit receives the analysis data from the perception computing unit, plans the route and avoids obstacles, and completes the real-time navigation between the starting point and the destination; at the same time, the path planning information, the distance information and the real-time navigation information are sent to the bottom control unit; The bottom control unit receives the command signal from the decision-making and planning unit, controls the movement of the intelligent mobile device, and completes the driving, starting, stopping and speed command operations of the servo control unit on the motor unit; The decision-making planning unit is responsible for analyzing and processing the motion decisions of the intelligent mobile device to realize path planning and physical obstacle avoidance functions; The bottom control unit controls the movement of the intelligent mobile device by installing a servo control unit, a power supply unit and a motor unit to complete the driving route planned by the controller.

2. The intelligent motion system based on cloud-edge collaborative control according to claim 1 is characterized in that: The devices of the sensor unit include but are not limited to lidar, camera, odometer, IMU inertial measurement unit and collision sensor.

3. The intelligent motion system based on cloud-edge collaborative control according to claim 1, characterized in that: The central data cloud module records the status and operation logs of edge devices, and carries an integrated server to train and optimize artificial intelligence algorithms, and sends them to edge computing devices for operation; The metadata storage unit A uses a DB to store and record the status information and operation health log of the edge device, and also backs up and records the key data; The integrated server is responsible for completing the training and tuning of the artificial intelligence algorithms required for navigation of smart mobile devices. It uses the data uploaded by the edge computing device to mark and clean it, trains the current intelligent algorithm, and after achieving the required accuracy, sends it to the edge device to complete the intelligent control of the platform.

4. The intelligent motion system based on cloud-edge collaborative control according to any one of claims 1 to 3, characterized in that: The implementation steps are as follows: Step S1: The central data cloud module communicates with the edge computing device through the wireless network transmission module to implement the issuance of instructions and algorithm functions; Step S2: The edge computing device deploys the algorithm function issued by the central data cloud module. At the same time, the central data cloud module monitors the health status of the edge computing device in real time, records and issues data; Step S3: further send the command data of the edge computing device to the smart mobile device, which collects the physical information of its posture and surrounding environment through the sensor unit, and obtains the information of the posture and surrounding environment of the smart mobile device in the world coordinate system through data modeling and processing of the perception computing unit; Step S4: the data obtained by the perception and computing unit is further transmitted to the decision-making and planning unit for processing, and the path planning and dynamic obstacle avoidance functions are realized through the artificial intelligence algorithm, and the motion control instructions are sent to the intelligent mobile device; Step S5: After receiving the control command signal, the intelligent mobile device operates the motor unit through the servo control unit to complete the movement and state feedback of the intelligent mobile device.

5. The intelligent motion system based on cloud-edge collaborative control according to claim 4 is characterized in that: The intelligent mobile device feeds back the current posture information to the device twin unit. After data cleaning and labeling, it is stored in the database of the edge computing device and sent to the central data cloud module. The central data cloud module stores the current device metadata and sampled environmental data, extracts key information and sends it to the integrated server as a training set for the artificial intelligence algorithm, and sends the optimized results to the edge computing device.

Images