Vehicle remote driving method and system based on 5G communication

Abstract

The invention relates to the technical field of automatic driving, and provides a vehicle remote driving method and system based on 5G communication. The method comprises the following steps: deploying a 5G base station, and generating a remote communication link; uploading the vehicle state and environment sensing data to a remote driving center in real time; the remote center plans a path based on a driving strategy library of a preset traffic rule and generates a safety control instruction; and issuing an instruction to the vehicle through the 5G base station to execute remote driving. According to the invention, the technical problems of environment perception lagging, insufficient control instruction real-time performance and difficulty in guaranteeing the driving safety in a complex traffic scene caused by communication delay and data transmission bandwidth limitation in an existing remote driving system are solved, and the purposes of realizing real-time data transmission and instruction issuing through 5G low-delay communication and improving the driving safety are achieved. And the preset driving strategy library is combined to ensure that the control instruction accords with the traffic rule, so that the response speed and the safety of remote driving are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, in particular to a vehicle remote driving method and system based on 5G communication. BACKGROUND

[0002] With the rapid development of automatic driving technology, vehicle remote driving as an important extension direction, has gradually become a research hotspot in the field of intelligent transportation. However, the traditional remote driving system is limited by the communication technology bottleneck, and there are problems such as high data transmission delay and poor link stability, which leads to control instruction response lag, and it is difficult to meet the real-time requirement of driving scene demand. Especially in complex traffic environment, vehicles need to process a large amount of environmental perception data and dynamic driving state information, and the defects of insufficient bandwidth and time delay fluctuation of traditional 4G network are more prominent, which directly restricts the safety and reliability of the remote driving system. At the same time, the traffic management department has increasingly strict requirements for the compliance of remote driving, and the system must be built-in with decision logic that meets the regional traffic rules, which puts higher requirements on data processing efficiency and control instruction generation speed. SUMMARY

[0003] The present application provides a vehicle remote driving method and system based on 5G communication, aiming to solve the technical problems of environmental perception lag, insufficient real-time of control instruction, and difficulty in ensuring driving safety in complex traffic scenarios caused by communication delay and data transmission bandwidth limitation in existing remote driving systems.

[0004] The first aspect of the present application provides a vehicle remote driving method based on 5G communication, the method comprising: setting a 5G communication base station, the 5G communication base station being used to connect a remote driving center and a target vehicle, and generating a remote communication link; uploading driving state data and environmental perception data of the target vehicle to the remote driving center based on the remote communication link, wherein the driving state data includes vehicle speed and steering angle, and the environmental perception data includes laser radar point cloud and image acquisition data; the remote driving center determines a safe control instruction based on a driving strategy library that meets a preset traffic rule, and plans a driving path for the target vehicle; and the safe control instruction is sent to the target vehicle for vehicle remote driving by using the 5G communication base station.

[0005] In another aspect of the present application, a vehicle remote driving system based on 5G communication is provided, which comprises: a communication link generation module: a 5G communication base station is set up, which is used to connect a remote driving center and a target vehicle, and generate a remote communication link; a data uploading module: based on the remote communication link, driving state data and environment perception data of the target vehicle are uploaded to the remote driving center, wherein the driving state data comprises vehicle speed and steering angle, and the environment perception data comprises laser radar point cloud and image acquisition data; a path planning module: the remote driving center plans a driving path for the target vehicle based on a driving strategy library that meets preset traffic rules, and determines a safety control instruction; and a remote driving module: the safety control instruction is sent to the target vehicle for vehicle remote driving using the 5G communication base station.

[0006] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0007] The above-mentioned vehicle remote driving method based on 5G communication establishes a communication link between a remote driving center and a target vehicle by setting up a 5G communication base station, and uploads driving state data (such as speed and steering angle) and environment perception data (such as laser radar point cloud and image data) of the vehicle to the remote driving center. The remote driving center plans a driving path and generates a safety control instruction according to a preset traffic rule and a driving strategy library, ensuring the safety and efficiency of vehicle driving. Finally, the safety control instruction is transmitted to the target vehicle in real time using the 5G communication base station for remote driving control, thereby improving the real-time response capability and safety of remote driving, and ensuring precise control and stable operation of the vehicle in complex environments.

[0008] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the following specific embodiments of the present application can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0010] Figure 1 FIG. 1 is a flowchart of a vehicle remote driving method based on 5G communication in an embodiment.

[0011] Figure 2A vehicle remote driving system architecture based on 5G communication for an embodiment.

[0012] Reference signs: communication link generation module 11, data upload module 12, path planning module 13, remote driving module 14. DETAILED DESCRIPTION

[0013] The embodiment of the present application provides a vehicle remote driving method and system based on 5G communication, and solves the technical problem of insufficient real-time control instruction and delayed environment perception caused by communication delay and data transmission bandwidth limitation in the existing remote driving system, which is difficult to guarantee the driving safety in a complex traffic scene.

[0014] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0015] It should be noted that the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server comprising a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or modules that are not clearly listed or inherent to the process, method, product or device.

[0016] Embodiment one, as shown in the present application provides a vehicle remote driving method based on 5G communication, the method comprises: Figure 1

[0017] Setting up a 5G communication base station, which is used to connect the remote driving center and the target vehicle, and generate a remote communication link.

[0018] In the embodiment of the present application, by arranging 5G communication base stations in the target area, a stable remote communication link is created to connect the remote driving center and the target vehicle. This remote communication link can realize real-time data transmission, and ensure that the remote driving center can maintain stable communication with the vehicle, thereby providing technical support for subsequent vehicle control, data upload and instruction issuance. This remote link based on 5G communication has the advantages of low delay, high bandwidth and high stability, and can meet the high requirements of automatic driving for data real-time and accuracy.

[0019] ​uploading driving state data and environment perception data of the target vehicle to the remote driving center based on the remote communication link, wherein the driving state data includes vehicle speed and steering angle, and the environment perception data includes laser radar point cloud and image acquisition data.

[0020] In one embodiment, real-time data of the target vehicle is uploaded to the remote driving center using the previously established remote communication link. These data include the vehicle's driving state information such as current vehicle speed and steering angle, which helps the remote driving center monitor the vehicle's motion state in real time. At the same time, environment perception data is also uploaded, which is generated by sensors on the vehicle such as laser radar and camera. Laser radar point cloud provides spatial information of the surrounding environment, while image acquisition data contains visual images around the vehicle for identifying roads, obstacles, etc. Through the uploading of these data, the remote driving center can fully grasp the vehicle and its environment, providing accurate information support for subsequent path planning and decision-making.

[0021] Further, the application provides a method for uploading driving state data and environment perception data of the target vehicle to the remote driving center, which comprises:

[0022] dimensionality reduction processing is performed on the laser radar point cloud to convert it into a two-dimensional feature map; based on the image acquisition data, first key feature information associated with traffic signs and second key feature information associated with obstacles are identified; and a data priority queue is set according to the two-dimensional feature map in combination with the first and second key feature information.

[0023] Preferably, for the three-dimensional point cloud data collected by the laser radar, denoising and filtering processing are performed on the three-dimensional point cloud data to remove irrelevant noise, and then a principal component analysis (PCA) algorithm is applied to map the data from a three-dimensional space to a two-dimensional plane while retaining the most important spatial features by calculating the principal components of the point cloud data. In this way, the three-dimensional laser radar point cloud is successfully converted into a two-dimensional feature map, facilitating subsequent analysis and processing. For the image acquisition data collected by the camera, image preprocessing is performed to remove noise and perform edge detection to improve the accuracy of subsequent recognition. Subsequently, a convolutional neural network (CNN) is used to recognize traffic signs in the image, and a target detection algorithm (such as YOLO or Faster R-CNN) is used to identify the type and location of the traffic sign to obtain the first key feature information associated with the traffic sign. At the same time, an image segmentation technique (such as the U-Net network) is used to perform pixel-level segmentation of obstacles in the image to extract information such as the boundary, shape, and size of the obstacles, obtaining the second key feature information associated with the obstacles. These identified key feature information can help the system understand the surrounding environment and provide support for path planning. Then, the two-dimensional feature map and the first and second key feature information are combined to set up a data priority queue. In this process, each feature is assigned a corresponding weight based on its importance and urgency, for example, obstacles or traffic signs such as traffic lights that are closer in distance are given higher weights, while other information is given lower weights. Finally, the data is sorted according to these weights to generate a priority queue, ensuring that the most important information is processed first and timely feedback is provided to the remote driving center. This process ensures that the vehicle can respond to changes in the surrounding environment in real time and make accurate control decisions.

[0024] The remote driving center plans a driving path for the target vehicle based on a driving strategy library that complies with preset traffic rules to determine a safety control instruction.

[0025] In one embodiment, the remote driving center utilizes a preset driving strategy library in compliance with traffic rules to plan a driving path for the target vehicle. Specifically, the remote driving center considers factors such as traffic signs, road conditions, real-time traffic conditions, etc. based on the current driving state of the target vehicle, environmental perception data, and preset traffic rules, and utilizes a multi-objective path planning model to plan a driving path that is more suitable for the current traffic conditions, wherein the multi-objective path planning model is continuously optimized and trained using data in the driving strategy library. Subsequently, the remote driving center simulates and verifies the multiple paths planned by the multi-objective path planning model to determine the optimal driving path, which not only complies with traffic regulations but also ensures safe and smooth driving. Then, according to the planned optimal driving path, the remote driving center further generates safety control instructions to instruct the target vehicle to perform specific driving actions such as acceleration, deceleration, steering, etc., which will ensure that the vehicle safely drives along the planned path and can respond to possible emergency situations.

[0026] Further, the present application provides a remote driving center based on a driving strategy library in compliance with preset traffic rules, and the method comprises:

[0027] Through historical driving data and real-time traffic conditions, a reinforcement learning algorithm is used to dynamically update the driving strategy library; at the same time, a multi-objective path planning model is constructed to generate candidate paths; the candidate paths are virtually simulated and verified to evaluate the path feasibility and path risk level, and the optimal driving path is selected.

[0028] Preferably, the remote driving center updates the data in the driving strategy library by synchronizing historical driving data and real-time traffic conditions, and continuously dynamically updates the path strategy mapping model bound to the driving strategy library based on the data in the driving strategy library through a reinforcement learning algorithm. The path strategy mapping model is based on a neural network. The dynamic updating steps include forward propagation, loss calculation, back propagation, parameter optimization, etc. Specifically, the reinforcement learning algorithm automatically optimizes the driving strategy by continuously learning historical driving experience and real-time traffic changes, so that the most suitable safety control instructions can be matched according to the traffic conditions corresponding to the optimal driving path of the vehicle, thereby improving the overall driving efficiency and safety. At the same time, the remote driving center also constructs a multi-objective path planning model based on a spatio-temporal convolution network (ST-ConvNet), which considers multiple objectives and constraints to generate several candidate driving paths. These candidate paths are planned according to different traffic demands and safety requirements, such as the shortest path, the lowest risk path, etc. Then, the remote driving center will verify these candidate paths through virtual simulation, and evaluate the feasibility and potential risk level of each path through simulation software, including the influence of traffic congestion, road obstacles, etc. Finally, the remote driving center will select the optimal driving path according to the simulation results to ensure that the target vehicle can drive on the safest and most efficient path.

[0029] Further, the present application provides a method for constructing a multi-objective path planning model, comprising:

[0030] Predicting the traffic congestion of the target road section in the future period through a spatio-temporal convolution network to determine traffic flow prediction data; dynamically adjusting the weight parameters of each road section in the path planning according to the traffic flow prediction data, and setting a safety redundancy condition.

[0031] Optionally, the remote driving center uses a multi-objective path planning model based on a spatio-temporal convolution network, which is constructed through steps such as forward propagation, loss calculation, back propagation, and parameter optimization. According to the current traffic conditions of the target road section, the traffic congestion in the future period is predicted. This multi-objective path planning model combines the information of time and space dimensions, and predicts the traffic flow of each road section in the future period by analyzing historical traffic flow data and real-time traffic conditions. Subsequently, according to these traffic flow prediction data, the remote driving center dynamically adjusts the weight parameters of each road section in the path planning. For example, if the traffic flow of a road section is large in the predicted period, it may cause congestion, which will reduce the weight of the road section, thereby avoiding planning the vehicle to pass through this road section. Conversely, if the traffic flow of the predicted road section is small, the weight of the road section may be increased, making it a preferred choice for vehicle travel. In addition, safety redundancy conditions are set according to traffic flow prediction data, which means that if the traffic flow prediction of a road section shows that congestion may occur, an alternative path will be provided for vehicle planning, and the path will be adjusted if necessary to ensure that the vehicle can safely travel to the destination in the shortest time. These redundancy conditions enhance the flexibility of path planning and the ability to respond to sudden traffic conditions, thereby improving the efficiency and safety of remote driving.

[0032] Using the 5G communication base station, the safety control instructions are sent to the target vehicle for vehicle remote driving.

[0033] In one embodiment, the remote driving center transmits the generated safety control instructions to the target vehicle in real time through the established 5G communication base station. Through the 5G network, the instructions can be transmitted from the remote driving center to the vehicle in a very short time, ensuring that the vehicle can respond quickly and execute the instructions. These safety control instructions usually include the driving behavior of the vehicle, such as acceleration, deceleration, steering, or parking operations, which aim to ensure that the vehicle travels along the predetermined path and responds to sudden traffic conditions or obstacles in real time. Through the low delay and high bandwidth characteristics of 5G communication, the transmission of instructions has almost no delay, ensuring efficient and safe remote driving of the vehicle in complex environments.

[0034] Further, the present application provides a method for sending the safety control instructions to the target vehicle for vehicle remote driving, which comprises:

[0035] The safety control instructions are encrypted and processed, and an instruction transmission confirmation mechanism is set up. The target vehicle feeds back confirmation information after receiving the safety control instructions. If no confirmation information is received from the target vehicle, the instruction is triggered to be retransmitted.

[0036] Optionally, the remote driving center first encrypts the generated safety control instructions to ensure that the instructions are not tampered with or stolen during transmission. Encryption ensures the security of communication, so that the integrity and confidentiality of the instructions are protected. The safety control instruction encryption can use symmetric encryption (such as AES) or asymmetric encryption (such as RSA). Subsequently, an instruction transmission confirmation mechanism is set up. When the target vehicle receives these encrypted safety control instructions, it will immediately send confirmation information to the remote driving center, indicating that the instructions have been successfully received and are ready for execution. If the remote driving center does not receive the confirmation information from the target vehicle, it will consider that the instructions may not have been successfully transmitted, or the vehicle may not have been able to process it in time. Therefore, an automatic retransmission mechanism will be triggered to send the same safety control instructions to the target vehicle again to ensure that the instructions can be correctly conveyed and executed, thereby ensuring the reliability and safety of remote driving.

[0037] Further, the present application comprises:

[0038] According to the time stamp of the safety control instruction and the time stamp of the confirmation information, the communication delay of the 5G communication base station is determined; it is judged whether the communication delay of the 5G communication base station exceeds a first preset delay threshold, and if it exceeds, it is switched to a redundant communication link.

[0039] Optionally, the remote driving center first records the time stamp of the safety control instruction and the time stamp of the confirmation information fed back by the target vehicle. These two time stamps are used to calculate the time required from instruction sending to confirmation receiving, i.e. the communication delay. By comparing the two time stamps, the communication delay of the 5G communication base station can be accurately calculated. Subsequently, the calculated communication delay is compared with a preset first delay threshold. If the communication delay exceeds this threshold, it indicates that the performance of the 5G communication link has problems and may not meet the needs of real-time driving. In order to ensure the safety and stability of the vehicle, it will automatically switch to a redundant communication link, such as a backup 5G remote communication link, a 4G remote communication link or other low-delay communication methods, to ensure that the transmission of instructions and the remote driving operation of the vehicle are not affected. This mechanism dynamically monitors the communication delay to ensure that even when the 5G network quality decreases, the vehicle can still obtain timely instructions through other reliable communication links, ensuring the continuity and safety of driving.

[0040] Further, the present application provides a judgment on whether the communication delay of the 5G communication base station exceeds a second preset delay threshold, and if it exceeds, a delay reminder is sent to request local driving to take over; if the local driving takeover has not been processed within a dynamic threshold period, the signal strength is written into the safety redundancy conditions as an additional safety redundancy condition.

[0041] Optionally, the communication delay of the 5G communication base station is further monitored. If the previously calculated communication delay exceeds a more stringent second preset delay threshold, it means that the communication delay has reached a critical value, which may affect the real-time performance and safety of remote driving. Therefore, a delay reminder is immediately sent to notify the remote driving center and request local driving to take over to prevent the vehicle from being unable to respond to instructions in time. If the local driving fails to successfully handle or respond to the delay reminder within the set dynamic threshold period, further preventive measures are taken. At this time, the signal strength between the vehicle and the communication base station is monitored as an additional safety redundancy condition. A low signal strength may indicate poor communication quality, and the vehicle may face the risk of being unable to obtain instructions in time. The system writes the signal strength as a safety redundancy condition into the path planning and control decision-making, ensuring that in the case of unstable signal, the path can be automatically adjusted or appropriate safety measures can be taken to improve the safety and reliability of the vehicle. Through this mechanism, sufficient safety redundancy can be ensured throughout the remote driving process according to the real-time changes in communication quality, avoiding potential risks caused by network delay or communication failure.

[0042] In summary, the embodiments of the present application have at least the following technical effects:

[0043] The embodiments of the present application first set up a 5G communication base station, which is used to connect the remote driving center and the target vehicle to generate a remote communication link. Then, based on the remote communication link, the driving state data and environmental perception data of the target vehicle are uploaded to the remote driving center, wherein the driving state data includes vehicle speed and steering angle, and the environmental perception data includes laser radar point cloud and image acquisition data. Subsequently, the remote driving center determines the safety control instructions based on the driving strategy library that complies with the preset traffic rules. Finally, the 5G communication base station is used to issue the safety control instructions to the target vehicle for remote driving of the vehicle. These technical effects collectively solve the technical problems of environmental perception lag, insufficient real-time performance of control instructions, and difficulty in ensuring driving safety in complex traffic scenarios caused by communication delay and data transmission bandwidth limitations in existing remote driving systems. The technical effects of achieving real-time data transmission and instruction issuance through 5G low-delay communication, ensuring that the control instructions comply with traffic rules by combining a preset driving strategy library, and improving the response speed and safety of remote driving are achieved.

[0044] Embodiment two, based on the same inventive concept as the vehicle remote driving method based on 5G communication in the foregoing embodiments, such as Figure 2As shown, the present application provides a vehicle remote driving system based on 5G communication, which comprises: a communication link generation module 11: setting a 5G communication base station for connecting a remote driving center and a target vehicle, generating a remote communication link; a data uploading module 12: uploading the driving state data and environmental perception data of the target vehicle to the remote driving center based on the remote communication link, wherein the driving state data includes vehicle speed and steering angle, and the environmental perception data includes laser radar point cloud and image acquisition data; a path planning module 13: the remote driving center plans the driving path of the target vehicle based on a driving strategy library conforming to the preset traffic rules, and determines a safety control instruction; a remote driving module 14: using the 5G communication base station, the safety control instruction is sent to the target vehicle for vehicle remote driving.

[0045] Further, the data uploading module 12 is also used to execute the following method:

[0046] The laser radar point cloud is processed by dimension reduction to convert into a two-dimensional feature map; based on the image acquisition data, a first key feature information associated with a traffic sign and a second key feature information associated with an obstacle are identified; and according to the two-dimensional feature map, the first key feature information and the second key feature information are combined to set a data priority queue.

[0047] Further, the path planning module 13 is also used to execute the following method:

[0048] Through historical driving data and real-time traffic conditions, a reinforcement learning algorithm is used to dynamically update the driving strategy library; at the same time, a multi-objective path planning model is constructed to generate a candidate path; the candidate path is virtually simulated and verified to evaluate the path feasibility and path risk level, and the optimal driving path is selected.

[0049] Further, the path planning module 13 is also used to execute the following method:

[0050] The traffic congestion situation of the target road section in the future period is predicted by a space-time convolution network to determine traffic flow prediction data; according to the traffic flow prediction data, the weight parameters corresponding to each road section in the path planning are dynamically adjusted, and a safety redundancy condition is set.

[0051] Further, the remote driving module 14 is also used to execute the following method:

[0052] The safety control instruction is encrypted, an instruction transmission confirmation mechanism is set, and the target vehicle feeds back confirmation information after receiving the safety control instruction; if no confirmation information is received from the target vehicle, the instruction is triggered to be retransmitted.

[0053] Further, the remote driving module 14 is further configured to execute the following method:

[0054] According to the time stamp of the safety control instruction and the time stamp of the confirmation information, a communication delay of the 5G communication base station is determined; whether the communication delay of the 5G communication base station exceeds a first preset delay threshold is judged, and if so, a redundant communication link is switched to.

[0055] Further, the remote driving module 14 is further configured to execute the following method:

[0056] Whether the communication delay of the 5G communication base station exceeds a second preset delay threshold is judged, and if so, a delay reminder is sent to request local driving to take over; if the local driving does not handle within a dynamic threshold period, a signal strength is written into the safety redundancy condition as an additional safety redundancy condition.

[0057] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The above describes a specific embodiment of the present application. The processes depicted in the drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

[0058] The above only describes the preferred embodiments of the present application and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0059] The specification and drawings are merely exemplary of the present application, and any and all modifications, variations, combinations or equivalents that fall within the scope of the present application are considered to be covered by the present application. Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A vehicle remote driving method based on 5G communication, characterized in that: The method comprises: Setting up a 5G communication base station, wherein the 5G communication base station is used to connect the remote driving center and the target vehicle to generate a remote communication link; Uploading the target vehicle's driving status data and environmental perception data to the remote driving center based on the remote communication link, wherein the driving status data includes vehicle speed and steering angle, and the environmental perception data includes lidar point cloud and image acquisition data; The remote driving center plans the driving path of the target vehicle based on a driving strategy library that complies with preset traffic rules and determines safety control instructions; Using the 5G communication base station, the safety control instruction is sent to the target vehicle for remote vehicle driving.

2. The method according to claim 1, wherein Uploading the driving status data and environmental perception data of the target vehicle to the remote driving center, the method includes: Performing dimensionality reduction processing on the laser radar point cloud to convert it into a two-dimensional feature map; Based on the image acquisition data, identifying first key feature information associated with the traffic sign and second key feature information associated with the obstacle; A data priority queue is set according to the two-dimensional feature map in combination with the first key feature information and the second key feature information.

3. The method according to claim 1, wherein The remote driving center is based on a driving strategy library that complies with preset traffic rules. The method includes: Dynamically updating the driving strategy library using reinforcement learning algorithms based on historical driving data and real-time traffic conditions; At the same time, a multi-objective path planning model is constructed to generate candidate paths; Perform virtual simulation verification on the candidate paths, evaluate the path feasibility and path risk level, and select the optimal driving path.

4. The method according to claim 3, wherein Constructing a multi-objective path planning model, the method includes: Use spatiotemporal convolutional networks to predict traffic congestion on the target road section in the future and determine traffic flow forecast data; According to the traffic flow prediction data, the weight parameters corresponding to each road section in the path planning are dynamically adjusted to set safety redundancy conditions.

5. The method according to claim 4, wherein The method includes sending the safety control instruction to the target vehicle to perform remote driving of the vehicle, and comprising: Encrypting the safety control command, setting a command transmission confirmation mechanism, and causing the target vehicle to feed back confirmation information after receiving the safety control command; If no confirmation information is received from the target vehicle, a resend instruction is triggered.

6. The method according to claim 5, wherein The method comprises: Determining a communication delay of the 5G communication base station according to a timestamp of the security control instruction and a timestamp of the confirmation information; Determine whether the communication delay of the 5G communication base station exceeds a first preset delay threshold; if so, switch to a redundant communication link.

7. The method according to claim 6, wherein determining whether the communication delay of the 5G communication base station exceeds a second preset delay threshold, and if so, sending a delay reminder and requesting local driving to take over; If the local driving takeover is not processed within the dynamic threshold period, the signal strength is written into the safety redundancy condition as an additional safety redundancy condition.

8. The vehicle remote driving system based on 5G communication is characterized by: The system is used to execute the vehicle remote driving method based on 5G communication according to any one of claims 1 to 7, comprising: Communication link generation module: setting up a 5G communication base station, which is used to connect the remote driving center and the target vehicle to generate a remote communication link; Data upload module: Uploads the target vehicle's driving status data and environmental perception data to the remote driving center based on the remote communication link, wherein the driving status data includes vehicle speed and steering angle, and the environmental perception data includes lidar point cloud and image acquisition data; Path planning module: The remote driving center plans the driving path of the target vehicle based on the driving strategy library that complies with preset traffic rules and determines safety control instructions; Remote driving module: Use the 5G communication base station to send the safety control instructions to the target vehicle for remote driving of the vehicle.

Images