WO2021134217A1

WO2021134217A1 - Vehicle control method and apparatus based on remote takeover, and computer device

Info

Publication number: WO2021134217A1
Application number: PCT/CN2019/129923
Authority: WO
Inventors: 庄尚芸
Original assignee: 深圳元戎启行科技有限公司
Priority date: 2019-12-30
Filing date: 2019-12-30
Publication date: 2021-07-08
Also published as: CN113366399A; CN113366399B

Abstract

Provided is a vehicle control method based on remote takeover, the method comprising: receiving remote takeover requests sent by a plurality of vehicles, wherein the remote takeover request carries anomaly information; calculating a remote takeover weight of the remote takeover request according to the anomaly information; acquiring current load information of a plurality of control nodes (106), and calculating a node idle resource according to the current load information; allocating a corresponding target control node (106) to the remote takeover request according to the remote takeover weight and the node idle resource; and generating a remote takeover task according to the remote takeover request, and allocating the remote takeover task to the target control node (106), such that the target control node (106) performs remote takeover on a corresponding vehicle. The method can effectively improve the resource allocation efficiency and the processing efficiency of a remote takeover request.

Description

Vehicle control method, device and computer equipment based on remote takeover

Technical field

This application relates to a vehicle control method, device and computer equipment based on remote control.

Background technique

An unmanned car is a smart car that senses the road environment and the surrounding environment of the vehicle through the on-board sensing system, and controls the steering and speed of the vehicle to reach the predetermined target based on the road, vehicle position and obstacle information obtained by the perception, so that the vehicle can Drive safely on the road.

When an unmanned car will inevitably have an abnormal situation during the automatic driving process, it is usually necessary for the controller of the remote control platform to handle the vehicle when an abnormal situation or emergency occurs in the vehicle to ensure the safety of the vehicle. The remote takeover platform usually needs to supervise a large number of unmanned vehicles. When multiple vehicles encounter abnormal conditions, the pressure on the remote takeover platform is relatively high, and the processing efficiency of abnormal vehicles is low.

Summary of the invention

According to various embodiments disclosed in the present application, a remote control-based vehicle control method, device, and computer equipment are provided.

A vehicle control method based on remote takeover, including:

Receiving remote takeover requests sent by multiple vehicles, where the remote takeover request carries abnormal information;

Calculating the remote takeover weight of the remote takeover request according to the abnormal information;

Acquiring current load information of multiple control nodes, and calculating node idle resources according to the current load information;

Allocating a corresponding target control node for the remote takeover request according to the remote takeover weight and the node idle resources; and

A remote takeover task is generated according to the remote takeover request, and the remote takeover task is allocated to the target control node, so that the target control node remotely takes over the corresponding vehicle.

A vehicle control method based on remote takeover, including:

Acquire driving state information of the vehicle, and perform abnormality detection based on the driving state information;

When the driving state information is abnormal, control the vehicle to stop, and generate a remote takeover request based on the abnormal driving state information;

Sending the remote takeover request to a central server; causing the central server to generate a remote takeover task according to the remote takeover request, and assign the remote takeover task to a corresponding target control node;

Acquire the remote takeover instruction sent by the target control node, and send vehicle backhaul information to the target control node according to the remote takeover instruction; and

The receiving target control node issues a remote control instruction according to the vehicle return information, and controls the vehicle according to the remote control instruction.

A vehicle control device based on remote takeover, including:

The request receiving module is configured to receive remote takeover requests sent by multiple vehicles, and the remote takeover requests carry abnormal information;

The resource calculation module is configured to calculate the remote takeover weight of the remote takeover request according to the abnormal information; obtain the current load information of multiple control nodes, and calculate the idle resources of the node according to the current load information; and according to the remote takeover weight sum The node idle resource allocates a corresponding target control node for the remote takeover request; and

The task allocation module is configured to generate a remote takeover task according to the remote takeover request, and allocate the remote takeover task to the target control node, so that the target control node remotely takes over the corresponding vehicle.

A vehicle control device based on remote takeover, including:

The abnormality detection module is used to obtain the driving state information of the vehicle, and perform abnormality detection based on the driving state information; when the driving state information is abnormal, control the vehicle to stop, and generate a remote takeover request based on the abnormal driving state information;

A request sending module, configured to send the remote takeover request to a central server; enable the central server to generate a remote takeover task according to the remote takeover request, and allocate the remote takeover task to a corresponding target control node;

An information upload module, configured to obtain a remote takeover instruction sent by the target control node, and send vehicle backhaul information to the target control node according to the remote takeover instruction; and

The remote takeover module is configured to receive a remote control instruction issued by the target control node according to the information returned by the vehicle, and control the vehicle according to the remote control instruction.

A computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the remote takeover-based vehicle control method provided in any one of the embodiments of the present application when the computer program is executed.

One or more non-volatile computer-readable storage media storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, one or more processors execute the readable storage medium to realize the present invention. Apply for the steps of the vehicle control method based on remote takeover provided in any of the embodiments.

The details of one or more embodiments of the present application are set forth in the following drawings and description. Other features and advantages of this application will become apparent from the description, drawings and claims.

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. A person of ordinary skill in the art can obtain other drawings based on these drawings without creative work.

Fig. 1 is an application scenario diagram of a vehicle control method based on remote takeover according to one or more embodiments.

Fig. 2 is a schematic flowchart of a vehicle control method based on remote takeover according to one or more embodiments.

Fig. 3 is a schematic flow chart of a step of calculating idle resources of a node according to one or more embodiments.

Fig. 4 is a schematic flow diagram of a remote takeover queue allocation step according to one or more embodiments.

Fig. 5 is a schematic flowchart of a vehicle control method based on remote takeover in another embodiment.

Fig. 6 is a schematic flowchart of a step of remotely identifying abnormal information according to one or more embodiments.

Fig. 7 is a block diagram of a vehicle control device based on remote takeover according to one or more embodiments.

Fig. 8 is a block diagram of a vehicle control device based on remote takeover in another embodiment.

Figure 9 is a block diagram of a computer device according to one or more embodiments.

Fig. 10 is a block diagram of a computer device in another embodiment.

Detailed ways

In order to make the technical solutions and advantages of the present application clearer, the following further describes the present application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

In one of the embodiments, the vehicle control method based on network monitoring provided by the present application can be specifically applied to the field of automatic driving, for example, can be applied to the application environment as shown in FIG. 1. The vehicle 102 communicates with the central server 104 through the network, and the central server 104 communicates with the control node 106 through the network. The central server 104 receives remote takeover requests sent by multiple vehicles 102, and the central server 104 calculates the remote takeover weight of the remote takeover request according to the abnormal information carried in the remote takeover request. The central server 104 obtains the current load information of multiple control nodes, calculates node idle resources according to the current load information, and allocates the corresponding target control node 106 to the remote takeover request according to the remote takeover weight and node idle resources. The central server 104 generates a remote takeover task according to the remote takeover request, and assigns the remote takeover task to the target control node 106 so that the target control node 106 remotely takes over the corresponding vehicle 102. The central server 104 and the control node 106 can be implemented by independent servers or a server cluster composed of multiple servers.

In one of the embodiments, as shown in FIG. 2, a vehicle control method based on remote control is provided. Taking the method applied to the central server in FIG. 1 as an example for description, the method includes the following steps:

Step 202: Receive remote takeover requests sent by multiple vehicles, where the remote takeover requests carry abnormal information.

The vehicle can be an unmanned car, which is a smart car that senses the road environment through the on-board sensor system, automatically plans the driving route, and controls the vehicle to reach the predetermined goal. The vehicle is equipped with an on-board sensor system, an automatic driving system, and a remote takeover system.

During the automatic driving of the vehicle, the vehicle can communicate with the control server through the central server, so that the remote controller can remotely take over the vehicle through the control server. The central server can be used to manage remote takeover requests of multiple vehicles and multiple control servers. For example, when the vehicle is abnormal or has a malfunction, the vehicle needs to be remotely taken over to ensure the safety of the vehicle during driving.

The vehicle can perform anomaly detection on its own during the automatic driving process. When an abnormal situation is detected, such as a vehicle failure, a road failure, or an obstacle blocking and other abnormal conditions. The vehicle can send a remote takeover request to the control server of the remote control platform, and the control server issues a remote takeover instruction to the vehicle based on the remote takeover request sent by the vehicle to take over the vehicle remotely. The control server of the remote control platform can also directly issue a remote takeover instruction to the vehicle according to the remote takeover demand or when the automatic driving system fails, so as to remotely take over the vehicle.

Multiple autonomous vehicles can send remote control requests to the central server at the same time. The central server receives the remote takeover requests sent by multiple vehicles, dispatches each remote takeover request, and distributes it to the corresponding target control server to make the target control The server remotely takes over the corresponding remote takeover request and the vehicle. Among them, the control server may be a server corresponding to the control node.

Step 204: Calculate the remote takeover weight of the remote takeover request according to the abnormal information.

The remote takeover request sent by the vehicle carries the abnormal information of the vehicle. After receiving the remote takeover request sent by multiple vehicles, the central server calculates the remote takeover weight of the corresponding remote takeover request according to the abnormal information of each vehicle. The remote takeover weight can identify the importance and urgency of the current remote takeover request. For example, the central server may determine the vehicle abnormality degree value and the takeover urgency value according to the vehicle fault information and the abnormality type in the abnormal information, and calculate the remote takeover weight of the remote takeover request based on the vehicle abnormality value and the takeover urgency value.

For example, when the vehicle is parked on a dangerous road section and cannot move forward, or when the faulty part of the vehicle is serious and there is a greater danger, then it is necessary to remotely take over the vehicle as soon as possible to remove the abnormal condition of the vehicle in time.

Step 206: Obtain current load information of multiple control nodes, and calculate node idle resources according to the current load information.

The central server can manage the status of all vehicles and control nodes in real time. Each control node includes corresponding load information. The load information can indicate the current load working status information of the control server, and the idle resources of the nodes can indicate the current available control server. Available resources for processing remote takeover requests.

The central server obtains the current load information of multiple control nodes, and calculates the node idle resources of each control node according to the current load information of each control node. For example, the central server can use dynamic load balancing algorithms to control the server's real-time load status information to distribute multiple remote takeover requests, such as algorithms such as the least connection method and the weighted least connection method. The central server can also obtain the pending tasks of each control node, and then calculate the node idle resources of each control node according to the current load information of the control node and the pending tasks.

Step 208: Assign a corresponding target control node to the remote takeover request according to the remote takeover weight and node idle resources.

After obtaining node idle resources of multiple control nodes, the central server determines the resource consumption of the remote takeover request according to the remote takeover weight of the remote takeover request. The resource consumption can be expressed as the resource of the control node consumed by the remote takeover request. The central server then allocates a corresponding target control node for the remote takeover request according to the remote takeover weight and resource consumption of the remote takeover request.

For example, the central server can schedule multiple remote takeover requests through a resource allocation model. The central server can use the polling algorithm to poll the current load weights of multiple control nodes through the resource allocation model to obtain the current load balance of each control node, and select the corresponding control for the resource acquisition request according to the current load balance of each control node Node ID, smoothly process the current load weight corresponding to the selected control node ID, and use the smoothed result to select the control node ID corresponding to the next remote takeover request until the corresponding control node is selected for multiple remote takeover requests Logo. The selected control node identifier is determined as the target control node corresponding to the corresponding remote takeover request.

Step 210: Generate a remote takeover task according to the remote takeover request, and assign the remote takeover task to the target control node, so that the target control node can remotely take over the corresponding vehicle.

The central server can generate a remote takeover task immediately after receiving the remote takeover request, or can generate a corresponding remote takeover task after determining the target control node corresponding to the remote takeover request. After determining the target control node corresponding to the remote takeover request, the central server assigns the remote takeover task to the corresponding target control node. After the remote takeover task is successfully assigned, that is, after the target control node confirms the takeover, a communication channel between the vehicle that takes over the task remotely and the target control node is established, so that the control platform corresponding to the target control node can remotely take over the corresponding vehicle.

If the remote takeover task is not allocated successfully, the central server re-allocates the remote takeover task. If the remote takeover task is assigned successfully, the vehicle can also re-send the remote takeover request to the central server after a preset frequency interval.

In another embodiment, if the remote takeover task is not assigned successfully, the central server can also generate multiple remote takeover queues based on the current load information of multiple control nodes, and allocate the remote takeover request to the corresponding remote takeover request according to the remote takeover weight of the remote takeover request. The remote takes over the queue to queue up for idle control nodes.

After the target control node successfully takes over the remote takeover task, it will issue a remote control instruction to the vehicle corresponding to the remote takeover task. After the vehicle obtains the remote takeover instruction issued by the control server corresponding to the target control node, it uploads road image information and vehicle status information to the control server. Road image information can include video data around the vehicle and road screen video data, and road image information can be It is a continuous video frame. The vehicle status information may include vehicle status information such as vehicle positioning information, vehicle navigation information, in-vehicle temperature information, vehicle instrument information, and sensor information. Transmission of road image information and vehicle status information is used for remote takeover personnel to remotely take over and control the vehicle according to road conditions and vehicle status. As a result, an abnormal autonomous vehicle can be remotely taken over in a timely and effective manner, and dangers caused by disordered driving of the vehicle can be prevented, and the safety of the vehicle under abnormal conditions can be effectively guaranteed.

In the above-mentioned vehicle control method based on remote takeover, after the central server receives the remote takeover request sent by multiple vehicles, it calculates the remote takeover weight of the remote takeover request based on the abnormal information carried in the remote takeover request, thereby effectively identifying the importance of the remote takeover request The degree and urgency of the remote takeover request for the importance and urgency of the emergency shall be handled as soon as possible to prevent dangerous accidents. The central server obtains the current load information of multiple control nodes, calculates node idle resources based on the current load information, and allocates the corresponding target control node for the remote takeover request according to the remote takeover weight and node idle resources, thereby effectively according to the remote takeover weight and The node's idle resources are reasonably allocated to the remote takeover request. The central server then generates a remote takeover task according to the remote takeover request, and assigns the remote takeover task to the target control node, so that the target control node remotely takes over the corresponding vehicle. By identifying the remote takeover weight of the remote takeover request, and assigning the corresponding control node to the remote takeover request based on the remote takeover weight and node idle resources, it can effectively balance the load of multiple control nodes, which can effectively improve the efficiency of resource allocation and remote control. Take over the processing efficiency of the request.

In one of the embodiments, calculating the remote takeover weight of the remote takeover request based on the abnormality information includes: extracting the abnormality type and vehicle failure information in the abnormality information; determining the vehicle abnormality degree value and the takeover urgency value according to the abnormality type and the vehicle failure information ; Calculate the remote takeover weight of the remote takeover request based on the vehicle abnormality value and the takeover urgency value.

The abnormal information includes vehicle fault information. After the vehicle sends a remote takeover request carrying the abnormal information to the central server, the central server extracts the abnormal type and vehicle fault information in the abnormal information. Vehicle failure information includes vehicle state failure information and road state abnormal information. Among them, the vehicle status fault information may include sensor faults, such as lidar, camera, IMU (Inertial Measurement Unit), GPS, etc. faults, and vehicle-mounted PC faults, and vehicle status fault information such as abnormal automatic driving programs. The abnormal road state information includes abnormal road conditions such as road obstacles and road blockages. For example, when the vehicle is parked on a dangerous road section and cannot move forward, or when the faulty part of the vehicle is serious and there is a greater danger, then it is necessary to remotely take over the vehicle as soon as possible to remove the abnormal condition of the vehicle in time.

The central server can be pre-configured with a vehicle abnormal configuration table, and the vehicle abnormal configuration table is configured with a mapping relationship between abnormal types and vehicle fault information and abnormal degree values and takeover urgency values. The central server extracts the vehicle failure information in the abnormal information and determines the corresponding abnormal type, and matches the abnormal high configuration table according to the abnormal type and vehicle failure information requested by the remote takeover, and determines the vehicle abnormality value and the takeover emergency according to the matching result. Degree value, and then calculate the remote takeover weight of the remote takeover request according to the vehicle abnormality value and the takeover urgency value. As a result, the importance and urgency of the remote takeover request can be effectively identified, and the remote takeover request of the urgent importance and urgency can be processed as soon as possible to prevent dangerous accidents and improve the safety of abnormal vehicles.

In one of the embodiments, as shown in FIG. 3, the step of acquiring current load information of multiple control nodes and calculating node idle resources according to the current load information specifically includes the following content:

Step 302: Invoke the resource allocation model, analyze the current load information and node performance parameters of the multiple control nodes through the resource allocation model, and obtain the current load weights of the multiple control nodes.

Step 304: Obtain the tasks to be processed by the control node, and calculate the node resource utilization rate of each control node according to the current load weight and the tasks to be processed.

Step 306: Calculate node idle resources of multiple control nodes according to the current load weight and node resource utilization.

The resource allocation model may be a neural network-based decision model, and the resource allocation model may be a neural network model obtained by pre-training, which is used to calculate control node resources and allocate resources to remote takeover requests.

After the central server receives the remote takeover request sent by multiple vehicles, it obtains the current load information of the multiple control nodes bound to the central server, and the central server can also obtain the node performance parameters of each control node. The node performance parameter may represent various performance index parameters of the control node, and the node performance parameter may be the current performance parameter of the control node. The central server can also obtain the pending tasks of each control node. Among them, some control nodes may still have assigned remote takeover tasks in a pending state. At this time, the expected load of these nodes is relatively high. For example, node performance parameters may include index parameters such as the number of CPUs, CPU frequency, memory capacity, disk speed, and network throughput. The current load information of the control node may include indicators such as CPU occupancy rate, memory occupancy rate, and network bandwidth occupancy rate.

The central server calls the resource allocation model, analyzes the current load information and node performance parameters of multiple control nodes through the resource allocation model, and obtains the current load weights of the multiple control nodes. Specifically, the current load weight may be the ratio of the current load information to the node performance parameter.

The central server further calculates the node resource utilization of each control node according to the current load weight of the control node and the tasks to be processed, and the central server further calculates the node idle resources of multiple control nodes according to the current load weight and the node resource utilization. For example, the central server can also predict the load of the control node based on the current load weight of the control node and the belt processing task. The central server can also use the pending tasks of each control node to calculate the load increment of the control node, and calculate the node idle resources of the control node according to the current load weight and load increment of the control node. By calculating the node idle resources of the control node according to the remote takeover weight and the node status of the control node, resource calculation and resource allocation can be effectively performed, and the remote takeover request can be effectively allocated based on the remote takeover weight and node idle resources.

In one of the embodiments, assigning the corresponding target control node to the remote takeover request according to the remote takeover weight and node idle resources includes: using the resource allocation model to predict the corresponding resource consumption value according to the remote takeover weight; and remote takeover according to the remote takeover request The weight and the resource consumption value are calculated to match the idle resources of the node; the target control node corresponding to the remote takeover request is determined according to the matched node idle resource, and the remote takeover request is allocated to the corresponding target control node.

After obtaining the current load information of the multiple control nodes, the central server calls the resource allocation model, and analyzes the current load information and node performance parameters of the multiple control nodes through the resource allocation model to obtain the current load weights of the multiple control nodes. The central server can also use the resource allocation model to predict the corresponding resource consumption value according to the remote takeover weight of the remote takeover request. Among them, the resource consumption value may indicate the control node resources that the remote takeover request is expected to consume, such as resource consumption information such as the time, duration, and difficulty of the remote takeover.

The central server can also predict the load of the control node based on the current load weight of the control node and the belt processing task. The central server can also use the pending tasks of each control node to calculate the load increment of the control node, and calculate the node idle resources of the control node according to the current load weight and resource consumption value of the control node and the load increment.

The central server then determines the target control node corresponding to the remote takeover request according to the matching node idle resources, and allocates the remote takeover request to the corresponding target control node, so that the target control node remotely takes over the corresponding vehicle. By assigning the corresponding target control node to the remote takeover request according to the remote takeover weight and node idle resources, it can effectively allocate the remote takeover request according to the remote takeover weight and node idle resources, thereby effectively improving the resource allocation efficiency and remote takeover The efficiency of request processing.

In one of the embodiments, as shown in FIG. 4, the method further includes a step of remotely taking over the queue allocation, and this step specifically includes the following content:

Step 402: If there is no node idle resource currently, generate multiple remote takeover queues according to multiple remote takeover requests and current load information of multiple control nodes.

In step 404, the remote takeover request is allocated to the corresponding remote takeover queue according to the remote takeover weight and the resource consumption value of the remote takeover request.

Step 406: Calculate the resource consumption status of the multiple control nodes, assign the multiple control nodes to the corresponding remote takeover queue according to the resource consumption status of the multiple control nodes, and assign the remote takeover request to the corresponding target control in the corresponding remote takeover queue. node.

After the central server obtains the current load information of multiple control nodes, the central server calls the resource allocation model, uses the resource allocation model to predict the corresponding resource consumption value according to the remote takeover weight, and calculates the matching between the remote takeover weight and the resource consumption value according to the remote takeover request The nodes have idle resources. According to the remote takeover weight and node idle resources, the corresponding target control node is allocated to the remote takeover request.

Further, if there is no idle resource of the node currently, that is, each control node is in a load state, and there is no idle control node. At this time, the central server generates multiple remote takeover queues based on multiple remote takeover requests and the current load information of multiple control nodes. The central server can generate multiple corresponding queues based on the number of remote takeover requests and the current load information of multiple control nodes. The number of remote takeover queues. The central server may further allocate the remote takeover request to the corresponding remote takeover queue according to the abnormal type of the remote takeover request, the remote takeover weight and the resource consumption value. For example, remote takeover requests with a more urgent exception type or a higher remote takeover weight can be assigned to a remote takeover queue with a higher processing priority, so that the remote takeover request with a higher remote takeover weight can be allocated and processed in a timely and effective manner.

The central server then calculates the resource consumption status of multiple control nodes, assigns the multiple control nodes to the corresponding remote takeover queue according to the resource consumption status of the multiple control nodes, and distributes the remote takeover request to the corresponding target control in the corresponding remote takeover queue. Node, so that the target control node takes over the corresponding vehicle remotely. As a result, even when there are currently no idle node resources, the remote takeover request can be allocated to the corresponding remote takeover queue according to the remote takeover weight and resource consumption value, and multiple control nodes can be allocated to the corresponding remote takeover queue according to the resource consumption status of the multiple control nodes The corresponding remote takeover queue. When there are idle node resources, the remote takeover request can be allocated to the corresponding target control node in the remote takeover queue in a timely and effective manner. As a result, the remote takeover request can be reasonably allocated based on the remote takeover weight and node idle resources, thereby effectively improving the resource allocation efficiency and the processing efficiency of the remote takeover request.

In one of the embodiments, as shown in FIG. 5, a vehicle control method based on remote control is provided. Taking the method applied to the vehicle in FIG. 1 as an example, the method includes the following steps:

Step 502: Acquire driving state information of the vehicle, and perform abnormality detection based on the driving state information.

The vehicle can perform anomaly detection on its own during the automatic driving process. When an abnormal situation is detected, such as a vehicle breakdown, a road breakdown or an obstacle blocking and other abnormal situations. The vehicle can send a remote takeover request to the central server of the central control platform, and the vehicle can communicate with the control server through the central server, so that the remote controller can remotely take over the vehicle through the control server. Among them, the control server may be a server corresponding to the control node. The control server issues a remote takeover instruction to the vehicle based on the remote takeover request sent by the vehicle to remotely take over the vehicle.

In the process of driving, the vehicle's driving state information is obtained. The driving state information of the vehicle may include state information of the vehicle such as vehicle state and vehicle state parameters. The vehicle status can include the status of the vehicle being online, the vehicle being offline, normal automatic driving, and abnormal takeover. Vehicle state parameters may include parameter information such as current position information, current vehicle speed, lights, gears, acceleration information, vehicle navigation information, vehicle temperature information, vehicle instrument information, sensor information, and so on. The driving state information may also include log information generated by the automatic driving system of the vehicle in automatic driving, and may also include road state information and the like. During the process of automatic driving, the vehicle continuously detects abnormalities in the driving state information.

The vehicle can detect vehicle status fault information and road status abnormal information. Among them, the vehicle status fault information may include sensor faults, such as lidar, camera, IMU (Inertial Measurement Unit), GPS and other faults, and vehicle-mounted PC faults, and vehicle status fault information such as abnormal automatic driving programs. The abnormal road state information includes abnormal road conditions such as road obstacles and road blockages.

In step 504, when the driving state information is abnormal, the vehicle is controlled to stop, and a remote takeover request is generated according to the abnormal driving state information.

When the vehicle detects that there is an abnormality in the driving state through the driving state information, the vehicle generates a vehicle control instruction for parking, and controls the vehicle to park according to the vehicle control instruction. The vehicle generates a remote takeover request based on the abnormal driving status information to send the remote takeover request carrying the abnormal information to the central server, so that the central server distributes the remote takeover request to the corresponding target control node. For example, when the vehicle is parked on a dangerous road section and cannot move forward, or when the faulty part of the vehicle is serious and there is a greater danger, it is necessary to remotely take over the vehicle as soon as possible to remove the abnormality of the vehicle in time.

Step 506: Send the remote takeover request to the central server; make the central server generate a remote takeover task according to the remote takeover request, and assign the remote takeover task to the corresponding target control node.

The vehicle generates detailed abnormal information based on the abnormal driving status information, and generates a remote takeover request carrying the abnormal information, and then sends the remote takeover request to the central server, so that the central server generates remote takeover tasks according to the remote takeover request, and distributes the remote takeover tasks To the corresponding target control node, so that the target control node can take over the vehicle remotely.

Further, multiple autonomous vehicles can send remote control requests to the central server at the same time. The central server receives remote takeover requests sent by multiple vehicles, and dispatches each remote takeover request to the corresponding target control server. Make the target control server corresponding to the target control node remotely take over the corresponding remote takeover request and the vehicle.

Step 508: Obtain the remote takeover instruction sent by the target control node, and send vehicle backhaul information to the target control node according to the remote takeover instruction.

After the central server receives the remote takeover request sent by multiple vehicles, it can also calculate the remote takeover weight of the remote takeover request based on the abnormal information carried in the remote takeover request, which can effectively identify the importance and urgency of the remote takeover request, and provide timely response. Remote takeover requests of the importance and urgency of the urgency are handled as soon as possible to prevent dangerous accidents. The central server obtains the current load information of multiple control nodes, calculates node idle resources based on the current load information, and allocates the corresponding target control node for the remote takeover request according to the remote takeover weight and node idle resources, thereby effectively according to the remote takeover weight and The node's idle resources are reasonably allocated to the remote takeover request. The central server then generates a remote takeover task according to the remote takeover request, assigns the remote takeover task to the target control node, so that the target control node takes over the corresponding vehicle remotely, and issues a remote control instruction to the vehicle.

After the vehicle receives the instruction of the target control node, it can switch the automatic driving mode to the control mode corresponding to the remote takeover mode, and upload the vehicle back information to the target control node according to the remote control instruction. The vehicle return information can include vehicle status information and road image information. The vehicle status information may include vehicle status information such as vehicle positioning information, vehicle navigation information, in-vehicle temperature information, and vehicle instrument information. The road image information may include video data around the vehicle and road screen video data, and the road image information may be continuous video frames. Transmission of road image information and vehicle status information is used for remote takeover personnel to remotely take over the vehicle according to road conditions and vehicle status.

Step 510: The receiving target control node issues a remote control instruction according to the information returned by the vehicle, and controls the vehicle according to the remote control instruction.

After the vehicle uploads road image information and vehicle status information to the control server, the remote takeover platform sends remote control instructions to the vehicle through the control server corresponding to the target control node according to the road image information and vehicle status information. The remote control instructions are used to control the vehicle. After the vehicle obtains the remote control instruction, it executes the remote control instruction, and controls the vehicle to perform corresponding control operations according to the remote control instruction, so as to realize the remote control of the vehicle. For example, the remote control instructions may include instructions to control the driving of the vehicle, such as accelerating, decelerating, turning on lights, and stopping. As a result, an abnormal autonomous vehicle can be remotely taken over in a timely and effective manner, and dangers caused by disordered driving of the vehicle can be prevented, and the safety of the vehicle under abnormal conditions can be effectively guaranteed.

In this embodiment, the vehicle obtains the driving state information of the vehicle, and performs abnormality detection based on the driving state information. When an abnormality in the driving state information is detected, the vehicle is controlled to stop, and a remote takeover request is generated based on the abnormal driving state information. The vehicle sends the remote takeover request to the central server, so that the central server generates remote takeover tasks according to the remote takeover request, and allocates the remote takeover tasks to the corresponding target control node, which can effectively allocate remote takeover requests based on node idle resources. . The vehicle then obtains the remote takeover instruction sent by the target control node, and sends the vehicle backhaul information to the target control node according to the remote takeover instruction. After the vehicle receives the target control node and issues a remote control instruction according to the vehicle's return information, it controls the vehicle according to the remote control instruction. The resource allocation of control nodes for remote takeover requests through the central server can effectively allocate corresponding control nodes for remote takeover requests in a timely and effective manner, thereby effectively improving resource allocation efficiency and processing efficiency of remote takeover requests.

In one of the embodiments, the driving state information includes vehicle state information, and abnormal detection based on the driving state information includes: sending a fault monitoring instruction to the automatic driving system according to a preset frequency; detecting multiple sensors based on the fault monitoring instruction and vehicle state information If the abnormal state and driving abnormal state are detected, the detection result is obtained; if the detection result includes abnormal information, the vehicle fault information in the abnormal information is determined, and the abnormal information is determined to identify the abnormal type and abnormal level.

When the vehicle is running, it can continuously obtain the driving state information of the vehicle, and continuously detect the abnormality of the driving state information. Specifically, the vehicle can also send fault monitoring instructions to the automatic driving system according to the preset frequency during the automatic driving process, and the vehicle detects the abnormal state and driving abnormal state of multiple sensors according to the fault monitoring instruction and the vehicle status information, and feeds back the corresponding Test results. The vehicle recognizes whether there is an abnormality in the vehicle state based on the feedback detection result.

If there is abnormal information in the detection result, the vehicle fault information is determined based on the detected abnormal index parameters, and corresponding abnormal information is generated. The abnormal information can include multiple vehicle current state information such as abnormal index parameters, vehicle fault location and current location information. . The vehicle can also determine the corresponding abnormality type and abnormality level according to the abnormality index parameters. Among them, the abnormality level can be used to determine the abnormality degree value and the takeover urgency value corresponding to the abnormality information, and indicate the current abnormal state degree and urgency degree of the vehicle. By identifying and determining the abnormality type and abnormality level of the vehicle, the importance and urgency of the remote takeover request can be identified, so that the central server can promptly process the urgent importance and urgency of the remote takeover request as soon as possible to prevent dangerous accidents.

In one of the embodiments, the driving state information includes road image information, and the detecting anomaly according to the driving state information includes: performing road recognition and obstacle recognition on the road image information through a road recognition model to obtain a road recognition result; When there is a road abnormality or a road obstacle in the road recognition result, the vehicle is controlled to stop, and the road abnormality information is generated according to the recognition result.

During the process of automatic driving, the vehicle monitors road status information in real time to detect road abnormalities. For example, when it is recognized that the road ahead is unrecognizable, there are road obstacles, and the staying time is too long, it means that the current road state is abnormal. At this time, the vehicle needs to be controlled to stop. The vehicle can perform road recognition and obstacle recognition on the road image information collected by the vehicle through the road recognition model that has been trained. Specifically, the vehicle can recognize the road information and information in the road image information through the road recognition model. For example, it can recognize the detailed road information in the point cloud image by acquiring the point cloud image of the current frame and the point cloud image of the previous frame in the road image information. , And identify the key point cloud and motion trajectory in the point cloud image, identify the obstacle in the road image information according to the abnormal key point cloud and the motion trajectory, and perform anomaly detection on the obstacle. When the obstacle is abnormal, abnormal information including obstacle information is generated.

When there is a road abnormality or a road obstacle in the road recognition result, the vehicle is controlled to stop and the road abnormality information is generated according to the recognition result. The vehicle generates a remote takeover request based on the road anomaly information, and sends the remote takeover request to the central server, so that the central server generates a remote takeover task according to the remote takeover request, and assigns the remote takeover task to the corresponding target control node, so that the target control node pairs The vehicle takes over remotely. As a result, an abnormal autonomous vehicle can be remotely taken over in a timely and effective manner, and dangers caused by disordered driving of the vehicle can be prevented, and the safety of the vehicle under abnormal conditions can be effectively guaranteed.

In one of the embodiments, as shown in FIG. 6, after the remote takeover task is assigned to the corresponding target control node, it further includes a step of remotely identifying abnormal information, which specifically includes the following content:

Step 602: Receive an abnormal information identification result and operation prompt information sent by the target control node.

Step 604: Control the automatic driving of the vehicle according to the recognition result of the abnormal information and the operation prompt information.

Step 606: Continue to detect the driving state information of the vehicle until it is detected that the abnormal information is removed and the driving state of the vehicle is in a normal state, and the remote takeover task is ended.

The vehicle performs anomaly detection based on the driving state information. When an abnormality is detected in the driving state information, the vehicle is controlled to stop, and a remote takeover request is generated based on the abnormal driving state information. The vehicle sends a remote takeover request to the central server, so that the central server generates a remote takeover task according to the remote takeover request, and assigns the remote takeover task to the corresponding target control node, so that the target control node can take over the vehicle remotely.

The vehicle then obtains the remote takeover instruction sent by the target control node, and sends the vehicle backhaul information to the target control node according to the remote takeover instruction. Further, after the target control node obtains the abnormal information of the vehicle and the return information of the vehicle, it analyzes the abnormality according to the abnormal information and the return information of the vehicle, and determines whether the abnormal situation of the vehicle needs to be remotely taken over. For example, when there is an obstacle in front of the vehicle or the vehicle cannot identify the road autonomously, the controller corresponding to the target control node may not directly control the vehicle remotely. The controller can mark the type of obstacle and the processing method according to the information returned by the vehicle. The correct road information can be marked, and the vehicle performs autonomous identification and automatic driving processing based on the marked abnormal information recognition result and operation prompt information. The target control node generates the abnormal information recognition result and operation prompt information according to the abnormal information and the vehicle return information, and the vehicle controls the automatic driving of the vehicle according to the abnormal information recognition result and the operation prompt information.

The vehicle continues to detect the driving state information of the vehicle until it detects that the abnormal information is removed and the driving state of the vehicle is normal. For example, the vehicle has driven through an obstacle or the vehicle has left its current position and can autonomously identify the road. It can indicate that the abnormal information is removed or The driving state of the vehicle is normal, and the vehicle can end the remote takeover task at this time. Through the control node, identify the abnormal situation of the vehicle in time and analyze the corresponding abnormal information recognition result and operation prompt information to enable the vehicle to perform autonomous identification and automatic driving. When the abnormality is resolved and the driving state returns to normal, the remote takeover will be terminated, which can be effective It saves resources for remote takeover and also effectively improves the processing efficiency of remote takeover requests.

In one of the embodiments, the method further includes: sending the vehicle return information to the target control node according to the remote takeover instruction, so that the target control node performs anomaly analysis based on the abnormal information and the vehicle return information; if the abnormal condition of the vehicle is not required When remotely intervening, it receives an end takeover instruction issued by the target control node, and the end takeover instruction includes prompt information; the remote takeover mode is switched to the automatic driving mode, and the vehicle is controlled to drive automatically according to the driving state information and prompt information.

The vehicle obtains the driving state information of the vehicle, and performs abnormality detection based on the driving state information. When an abnormality in the driving state information is detected, the vehicle is controlled to stop, and a remote takeover request is generated based on the abnormal driving state information. The vehicle sends a remote takeover request to the central server, so that the central server generates a remote takeover task according to the remote takeover request, and allocates the remote takeover task to the corresponding target control node, so that the target control node can take over the vehicle remotely.

The vehicle then obtains the remote takeover instruction sent by the target control node, and sends the vehicle backhaul information to the target control node according to the remote takeover instruction. Further, after the target control node obtains the abnormal information of the vehicle and the return information of the vehicle, it analyzes the abnormality according to the abnormal information and the return information of the vehicle, and determines whether the abnormal situation of the vehicle needs to be remotely taken over. For example, if the abnormal situation of the vehicle driving state is that there is a small obstacle in front of the road, if the target control node analyzes that the obstacle can be driven directly without affecting the vehicle and driving state, it can be determined that the abnormal situation is No intervention of remote takeover is required. At this time, the target control node can send prompt information directly to the vehicle, and the prompt information can include information such as the obstacle type and processing method to prompt the vehicle to continue driving along the original route or the obstacle does not constitute an impact. The target control node issues an end takeover instruction to the vehicle to end the remote takeover.

After the vehicle receives the end takeover instruction issued by the target control node, it switches the remote takeover mode to the automatic driving mode, continues to recognize the road screen, and controls the automatic driving of the vehicle through the automatic driving system according to the driving status information and prompt information. By recognizing the abnormal condition of the vehicle in time, if the abnormal condition of the vehicle is recognized as not requiring remote intervention, prompt information is sent in time and the remote takeover is ended, so as to save remote takeover resources and effectively improve the processing efficiency of remote takeover requests.

In another embodiment, if the abnormal situation of the vehicle driving state is that there is an obstacle ahead, and the obstacle needs to be bypassed, the target control node can analyze the prompt information of bypassing the obstacle or stopping and avoiding, so as to make The vehicle performs autonomous road recognition according to the prompt information. Until the vehicle can automatically recognize road conditions and normal driving through the automatic driving system, prompt information can be sent to the target control node, so that the target control node will issue an end takeover instruction to the vehicle to end the remote takeover. This can effectively improve the processing efficiency of remote takeover requests.

It should be understood that although the various steps in the flowcharts of FIGS. 2-6 are displayed in sequence as indicated by the arrows, these steps are not necessarily performed in sequence in the order indicated by the arrows. Unless specifically stated in this article, there is no strict order for the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in Figures 2-6 may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. These sub-steps or stages The execution order of is not necessarily performed sequentially, but may be performed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

In one of the embodiments, as shown in FIG. 7, a vehicle control device based on remote takeover is provided, including: a request receiving module 702, a resource calculation module 704, and a task allocation module 706, in which:

The request receiving module 702 is configured to receive remote takeover requests sent by multiple vehicles, and the remote takeover requests carry abnormal information;

The resource calculation module 704 is used to calculate the remote takeover weight of the remote takeover request according to the abnormal information; obtain the current load information of multiple control nodes, and calculate the node idle resources according to the current load information; and make the remote takeover request according to the remote takeover weight and node idle resources Assign the corresponding target control node; and

The task allocation module 706 is configured to generate a remote takeover task according to the remote takeover request, and allocate the remote takeover task to the target control node, so that the target control node remotely takes over the corresponding vehicle.

In one of the embodiments, the resource calculation module 704 is also used to extract the abnormality type and vehicle failure information in the abnormality information; determine the vehicle abnormality degree value and the takeover urgency value according to the abnormality type and the vehicle failure information; and according to the vehicle abnormality degree value And the takeover urgency value to calculate the remote takeover weight of the remote takeover request.

In one of the embodiments, the resource calculation module 704 is also used to call a resource allocation model, and analyze the current load information and node performance parameters of multiple control nodes through the resource allocation model to obtain the current load weights of the multiple control nodes; Obtain the to-be-processed tasks of the control node, calculate the node resource utilization of each control node according to the current load weight and the to-be-processed tasks; and calculate the node idle resources of multiple control nodes according to the current load weight and the node resource utilization.

In one of the embodiments, the resource calculation module 704 is further configured to use the resource allocation model to predict the corresponding resource consumption value according to the remote takeover weight; calculate the matching node idle resources according to the remote takeover weight and the resource consumption value of the remote takeover request; and The task allocation module 706 is further configured to determine the target control node corresponding to the remote takeover request according to the matched node idle resources, and allocate the remote takeover request to the corresponding target control node.

In one of the embodiments, the resource calculation module 704 is further configured to generate a remote takeover queue according to multiple remote takeover requests and current load information of multiple control nodes if there are no node idle resources; according to the remote takeover weight of the remote takeover request And the resource consumption value to assign the remote takeover request to the corresponding remote takeover queue; and calculate the resource consumption status of multiple control nodes, and assign multiple control nodes to the corresponding remote takeover queue according to the resource consumption status of the multiple control nodes. The remote takeover request is allocated to the corresponding target control node in the corresponding remote takeover queue.

In one of the embodiments, as shown in FIG. 8, a vehicle control device based on remote takeover is provided, including: an abnormality detection module 802, a request sending module 804, an information upload module 806, and a remote takeover module 808, in which:

The abnormality detection module 802 is used to obtain the driving state information of the vehicle, and perform abnormality detection based on the driving state information; when the driving state information is abnormal, control the vehicle to stop, and generate a remote takeover request based on the abnormal driving state information;

The request sending module 804 is configured to send a remote takeover request to the central server; make the central server generate a remote takeover task according to the remote takeover request, and allocate the remote takeover task to the corresponding target control node;

The information upload module 806 is used to obtain the remote takeover instruction sent by the target control node, and send the vehicle backhaul information to the target control node according to the remote takeover instruction; and

The remote takeover module 808 is used to receive a remote control instruction issued by the target control node according to the information returned by the vehicle, and to control the vehicle according to the remote control instruction.

In one of the embodiments, the driving status information includes vehicle status information, and the abnormality detection module 802 is also used to send fault monitoring instructions to the automatic driving system according to a preset frequency; detect abnormal states of multiple sensors according to the fault monitoring instructions and vehicle status information And the abnormal driving state to obtain the detection result; and if the detection result includes abnormal information, determine the vehicle fault information in the abnormal information, and determine the abnormal information to identify the abnormal type and the abnormal level.

In one of the embodiments, the driving state information includes road image information, and the anomaly detection module 802 is also used to perform road recognition and obstacle recognition on the road image information through the road recognition model to obtain the road recognition result; and when there is a road recognition result When the road is abnormal or road obstacle, the vehicle is controlled to stop, and the road abnormal information is generated according to the recognition result.

In one of the embodiments, the remote takeover module 808 is also used to receive the abnormal information recognition result and operation prompt information sent by the target control node; control the automatic driving of the vehicle according to the abnormal information recognition result and the operation prompt information; and continuously detect the driving state of the vehicle Information, until the abnormal information is detected to be removed and the vehicle's driving state is normal, the remote takeover task is ended.

In one of the embodiments, the remote takeover module 808 is also used to send vehicle return information to the target control node according to the remote takeover instruction, so that the target control node performs anomaly analysis based on the abnormal information and the vehicle return information; if the abnormal condition of the vehicle is When remote intervention is not required, receive the end takeover instruction issued by the target control node, the end takeover instruction includes prompt information; and switch the remote takeover mode to automatic driving mode, and control the vehicle to automatically drive according to the driving status information and prompt information.

For the specific limitation of the vehicle control device based on remote takeover, please refer to the above limitation on the vehicle control method based on remote takeover, which will not be repeated here. The various modules in the vehicle control device based on remote takeover can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules may be embedded in the form of hardware or independent of the processor in the computer equipment, or may be stored in the memory of the computer equipment in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

In one of the embodiments, a computer device is provided. The computer device may be a central server, and its internal structure diagram may be as shown in FIG. 9. The computer equipment includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer readable instructions, and a database. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the non-volatile storage medium. The computer equipment database is used to store data such as remote takeover requests, driving status information, and control node information. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer-readable instructions are executed by the processor to realize a vehicle control method based on remote takeover.

In one of the embodiments, a computer device is provided. The computer device may be a vehicle, and its internal structure diagram may be as shown in FIG. 10. The computer equipment includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer readable instructions, and a database. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the non-volatile storage medium. The database of the computer equipment is used to store data such as driving status information and road image information. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer-readable instructions are executed by the processor to realize a vehicle control method based on remote takeover.

Those skilled in the art can understand that the structures shown in Figures 9 and 10 are only block diagrams of part of the structure related to the solution of the present application, and do not constitute a limitation on the computer equipment to which the solution of the present application is applied. The specific computer The device may include more or fewer parts than shown in the figures, or combine certain parts, or have a different arrangement of parts.

One or more non-volatile computer-readable storage media storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute A step of.

A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing relevant hardware through computer-readable instructions. The computer-readable instructions can be stored in a non-volatile computer. In a readable storage medium, when the computer-readable instructions are executed, they may include the processes of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database, or other media used in the embodiments provided in this application may include non-volatile and/or volatile memory. Non-volatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not a limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Channel (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be It is considered as the range described in this specification.

The above-mentioned embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the scope of the invention patent. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can be made, and these all fall within the protection scope of this application. Therefore, the scope of protection of the patent of this application shall be subject to the appended claims.

Claims

A vehicle control method based on remote takeover, including:

Receiving remote takeover requests sent by multiple vehicles, where the remote takeover request carries abnormal information;

Calculating the remote takeover weight of the remote takeover request according to the abnormal information;

Acquiring current load information of multiple control nodes, and calculating node idle resources according to the current load information;

Allocating a corresponding target control node for the remote takeover request according to the remote takeover weight and the node idle resources; and

A remote takeover task is generated according to the remote takeover request, and the remote takeover task is allocated to the target control node, so that the target control node remotely takes over the corresponding vehicle.
The method according to claim 1, wherein the calculating the remote takeover weight of the remote takeover request according to the abnormal information comprises:

Extracting the abnormal type and vehicle fault information in the abnormal information;

Determine the value of the abnormal degree of the vehicle and the value of the takeover urgency according to the abnormal type and the vehicle failure information; and

Calculate the remote takeover weight of the remote takeover request according to the vehicle abnormality value and the takeover urgency value.
The method according to claim 1, wherein said acquiring current load information of multiple control nodes and calculating node idle resources according to said current load information comprises:

Invoking a resource allocation model, and analyzing the current load information and node performance parameters of the multiple control nodes through the resource allocation model, to obtain the current load weights of the multiple control nodes;

Obtain the pending tasks of the control node, and calculate the node resource utilization rate of each control node according to the current load weight and the pending tasks; and

Calculate node idle resources of multiple control nodes according to the current load weight and node resource utilization.
The method according to claim 3, wherein the allocating a corresponding target control node for the remote takeover request according to the remote takeover weight and the node idle resource comprises:

Using the resource allocation model to predict the corresponding resource consumption value according to the remote takeover weight;

Calculate the matching node idle resources according to the remote takeover weight of the remote takeover request and the resource consumption value; and determine the target control node corresponding to the remote takeover request according to the matched node idle resources, and allocate the remote takeover request to The corresponding target control node.
The method according to claim 4, wherein the method further comprises:

If there is no node idle resource currently, generate a remote takeover queue according to multiple remote takeover requests and current load information of multiple control nodes;

Allocating the remote takeover request to the corresponding remote takeover queue according to the remote takeover weight and resource consumption value of the remote takeover request; and

Calculate the resource consumption status of multiple control nodes, assign the multiple control nodes to the corresponding remote takeover queue according to the resource consumption status of the multiple control nodes, and assign the remote takeover request to the corresponding target control node in the remote takeover queue. .
A vehicle control method based on remote takeover, including:

Acquire driving state information of the vehicle, and perform abnormality detection based on the driving state information;

When the driving state information is abnormal, control the vehicle to stop, and generate a remote takeover request based on the abnormal driving state information;

Sending the remote takeover request to a central server; causing the central server to generate a remote takeover task according to the remote takeover request, and assign the remote takeover task to a corresponding target control node;

Acquire the remote takeover instruction sent by the target control node, and send vehicle backhaul information to the target control node according to the remote takeover instruction; and

Receive the target control node to issue a remote control instruction according to the vehicle return information, and control the vehicle according to the remote control instruction.
The method according to claim 6, wherein the driving state information includes vehicle state information, and the performing abnormality detection according to the driving state information includes:

Send fault monitoring instructions to the autopilot system according to the preset frequency;

Detect the abnormal state and abnormal driving state of a plurality of sensors according to the fault monitoring instruction and the vehicle state information, and obtain the detection result; and

If the detection result includes abnormal information, the vehicle fault information in the abnormal information is determined, and the abnormal information is determined to identify the abnormal type and the abnormal level.
The method according to claim 6, wherein the driving state information includes road image information, and the detecting abnormality according to the driving state information includes:

Carry out road recognition and obstacle recognition on the road image information through the road recognition model, and obtain the road recognition result;

And when there is a road abnormality or a road obstacle in the road recognition result, the vehicle is controlled to stop, and road abnormality information is generated according to the recognition result.
The method according to claim 6, characterized in that, after allocating the remote takeover task to the corresponding target control node, the method further comprises:

Receiving an abnormal information identification result and operation prompt information sent by the target control node;

Controlling the automatic driving of the vehicle according to the recognition result of the abnormal information and the operation prompt information; and

Continue to detect the driving state information of the vehicle until it is detected that the abnormal information is removed and the driving state of the vehicle is in a normal state, and the remote takeover task is ended.
The method according to claim 6, wherein the method further comprises:

Sending vehicle return information to the target control node according to the remote takeover instruction, so that the target control node performs an abnormality analysis based on the abnormal information and the vehicle return information;

If the abnormal situation of the vehicle is that no remote intervention is required, receive an end takeover instruction issued by the target control node, where the end takeover instruction includes prompt information; and

The remote takeover mode is switched to the automatic driving mode, and the automatic driving of the vehicle is controlled according to the driving state information and the prompt information.
A vehicle control device based on remote takeover, including:

The request receiving module is configured to receive remote takeover requests sent by multiple vehicles, and the remote takeover requests carry abnormal information;

The resource calculation module is configured to calculate the remote takeover weight of the remote takeover request according to the abnormal information; obtain the current load information of multiple control nodes, and calculate the idle resources of the node according to the current load information; and according to the remote takeover weight sum The node idle resource allocates a corresponding target control node for the remote takeover request; and

The task allocation module is configured to generate a remote takeover task according to the remote takeover request, and allocate the remote takeover task to the target control node, so that the target control node remotely takes over the corresponding vehicle.
A vehicle control device based on remote takeover, including:

The abnormality detection module is used to obtain the driving state information of the vehicle, and perform abnormality detection based on the driving state information; when the driving state information is abnormal, control the vehicle to stop, and generate a remote takeover request based on the abnormal driving state information;

A request sending module, configured to send the remote takeover request to a central server; enable the central server to generate a remote takeover task according to the remote takeover request, and allocate the remote takeover task to a corresponding target control node;

An information upload module, configured to obtain a remote takeover instruction sent by the target control node, and send vehicle backhaul information to the target control node according to the remote takeover instruction; and

The remote takeover module is configured to receive a remote control instruction issued by the target control node according to the information returned by the vehicle, and control the vehicle according to the remote control instruction.
A computer device includes a memory and one or more processors. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the one or more Each processor performs the following steps:

Receiving remote takeover requests sent by multiple vehicles, where the remote takeover request carries abnormal information;

Calculating the remote takeover weight of the remote takeover request according to the abnormal information;

Acquiring current load information of multiple control nodes, and calculating node idle resources according to the current load information;

Allocating a corresponding target control node for the remote takeover request according to the remote takeover weight and the node idle resources; and

A remote takeover task is generated according to the remote takeover request, and the remote takeover task is allocated to the target control node, so that the target control node remotely takes over the corresponding vehicle.
The computer device according to claim 13, wherein the processor further executes the following step when executing the computer-readable instruction: using the resource allocation model to predict the corresponding resource consumption value according to the remote takeover weight; Calculate the matching node idle resources according to the remote takeover weight of the remote takeover request and the resource consumption value; and determine the target control node corresponding to the remote takeover request according to the matched node idle resources, and allocate the remote takeover request to The corresponding target control node.
The computer device according to claim 14, wherein the processor further executes the following steps when executing the computer-readable instructions: if there is no node idle resource currently, according to multiple remote takeover requests and multiple control nodes Generate a remote takeover queue based on the current load information of the remote takeover request; allocate the remote takeover request to the corresponding remote takeover queue according to the remote takeover weight and resource consumption value of the remote takeover request; and calculate the resource consumption status of multiple control nodes, according to the number The resource consumption status of each control node allocates multiple control nodes to the corresponding remote takeover queue, and allocates the remote takeover request to the corresponding target control node in the corresponding remote takeover queue.
A computer device includes a memory and one or more processors. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the one or more Each processor performs the following steps:

Acquire driving state information of the vehicle, and perform abnormality detection based on the driving state information;

When the driving state information is abnormal, control the vehicle to stop, and generate a remote takeover request based on the abnormal driving state information;

Sending the remote takeover request to a central server; causing the central server to generate a remote takeover task according to the remote takeover request, and assign the remote takeover task to a corresponding target control node;

Acquire the remote takeover instruction sent by the target control node, and send vehicle backhaul information to the target control node according to the remote takeover instruction; and

Receive the target control node to issue a remote control instruction according to the vehicle return information, and control the vehicle according to the remote control instruction.
The computer device according to claim 16, wherein the processor further executes the following steps when executing the computer-readable instruction: receiving the abnormal information recognition result and operation prompt information sent by the target control node; The abnormal information recognition result and the operation prompt information control the automatic driving of the vehicle; and continue to detect the driving state information of the vehicle until it is detected that the abnormal information is removed and the driving state of the vehicle is normal, the end of the Take over tasks remotely.
The computer device according to claim 16, wherein the processor further executes the following steps when executing the computer-readable instruction: sending vehicle back information to the target control node according to the remote takeover instruction, so that The target control node performs an abnormality analysis based on the abnormal information and the vehicle return information; if the abnormal condition of the vehicle does not require remote intervention, receiving an end takeover instruction issued by the target control node, the The command for terminating the takeover includes prompt information; and the remote takeover mode is switched to the automatic driving mode, and the vehicle is controlled to drive automatically according to the driving state information and the prompt information.
One or more non-volatile computer-readable storage media storing computer-readable instructions, which when executed by one or more processors, cause the one or more processors to execute claim 1 To any of the steps described in 5.
One or more non-volatile computer-readable storage media storing computer-readable instructions, which when executed by one or more processors, cause the one or more processors to execute claim 6 To any of the steps described in 10.