A task scheduling method and related device

Abstract

The application discloses a task scheduling method and device, electronic equipment and computer readable storage medium, which are applied to a vehicle-road cooperation system and aim to guarantee that a vehicle maintains continuous and efficient computing power service capability in a coverage blind area of a roadside device. The method comprises the following steps: the roadside device is used to monitor a target vehicle in real time, and a first driving time of the target vehicle in a monitoring coverage range and a second driving time of the target vehicle in a monitoring blind area range are obtained; a target function about an optimized preset index parameter is constructed by using the first driving time, the second driving time, system parameters of the vehicle-road cooperation system and a total task quantity, and a target parameter is solved to obtain a vehicle-side task quantity and a roadside task quantity; the roadside device is used to process a roadside task corresponding to the roadside task quantity, and a roadside processing result is obtained; the roadside processing result is used as a roadside computing service, and a vehicle-side task corresponding to the vehicle-side task quantity is deployed to the target vehicle as a vehicle-side computing service, so that the computing service is provided for the target vehicle when the target vehicle is in the monitoring blind area range.

Description

Technical Field

[0001] This application relates to the field of information and communication technology, and in particular to a task scheduling method, a task scheduling device, an electronic device, and a computer-readable storage medium. Background Technology

[0002] With the deep integration of information and communication technologies and the transportation sector, the Internet of Vehicles (IoV) industry has achieved a new leap forward. Relying on technologies such as cloud computing and edge computing, large-scale deployment of roadside units (RSUs) for IoV has been realized, and vehicle-road cooperative technology has emerged.

[0003] By applying vehicle-to-infrastructure (V2I) technology, vehicles can connect to nearby roadside unit nodes to achieve real-time information sharing between vehicles and the road, ensuring vehicle driving safety and efficient road traffic operation. However, in the current V2I V2I environment, due to deployment costs and road infrastructure conditions, there are certain coverage blind spots between adjacent roadside unit nodes in some road sections or spatial areas. This prevents vehicles entering these areas from obtaining computing power support from roadside unit nodes, while relying solely on the limited computing power of a single vehicle is insufficient to meet the high resource requirements and low latency constraints of autonomous driving computing tasks.

[0004] Therefore, how to effectively ensure that vehicles have continuous and efficient computing power services in the blind spots of roadside equipment coverage, and thus ensure the continuous safe and stable operation of vehicles, is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] The purpose of this application is to provide a task scheduling method that can effectively ensure that vehicles have continuous and efficient computing power service capabilities in the blind spots of roadside equipment coverage, thereby ensuring the continuous, safe and stable operation of vehicles. Another purpose of this application is to provide a task scheduling device, electronic device and computer-readable storage medium, all of which have the above-mentioned beneficial effects.

[0006] Firstly, this application provides a task scheduling method applied to a vehicle-road cooperative system, comprising:

[0007] The target vehicle is monitored in real time using roadside equipment to obtain the first travel time of the target vehicle within the monitoring coverage area and the second travel time within the monitoring blind zone.

[0008] A target function for optimizing preset index parameters is constructed using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task load.

[0009] Solve the objective function to obtain the vehicle-side task quantity and the roadside task quantity;

[0010] The roadside equipment is used to process the roadside tasks corresponding to the roadside task volume to obtain the roadside processing results;

[0011] The roadside processing results are used as roadside computing services, and the vehicle-side tasks corresponding to the vehicle-side task volume are used as vehicle-side computing services, which are deployed to the target vehicle to provide computing services when the target vehicle is driving within the monitoring blind zone.

[0012] Optionally, the roadside equipment is used to monitor the target vehicle in real time to obtain the first travel time of the target vehicle within the monitoring coverage area, including:

[0013] The roadside equipment is used to monitor the target vehicle in real time to obtain the target vehicle's current location, speed, and planned trajectory, and to determine the monitoring coverage area of ​​the roadside equipment.

[0014] The first travel time of the target vehicle within the monitoring coverage area is calculated using the current location, the driving speed, and the planned trajectory.

[0015] Optionally, the step of using the roadside equipment to monitor the target vehicle in real time and obtain the target vehicle's current position, speed, and planned trajectory includes:

[0016] The roadside equipment is used to communicate wirelessly with the target vehicle to obtain the target vehicle's current location, driving speed, and planned trajectory.

[0017] Optionally, calculating the first travel time of the target vehicle within the monitoring coverage area using the current location, the vehicle speed, and the planned trajectory includes:

[0018] Determine the intersection point between the planned trajectory and the monitoring coverage area to obtain the location of the first intersection point;

[0019] Calculate the first distance between the current position and the first intersection point position;

[0020] The first travel time is calculated based on the first distance and the driving speed.

[0021] Optionally, the roadside equipment is used to monitor the target vehicle in real time to obtain the second travel time of the target vehicle within the monitoring blind zone, including:

[0022] Determine the next monitoring coverage area of ​​the next roadside device of the roadside device, and determine the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area;

[0023] The second travel time of the target vehicle within the monitoring blind zone is calculated using the current location, the driving speed, and the planned trajectory.

[0024] Optionally, determining the next monitoring coverage area of ​​the next roadside device of the roadside device includes:

[0025] Based on the planned trajectory, route prediction is performed to determine the next roadside device for the roadside equipment.

[0026] Data communication is performed with the next-side device to obtain the next monitoring coverage area of ​​the next-side device.

[0027] Optionally, calculating the second travel time of the target vehicle within the monitoring blind zone using the current location, the vehicle speed, and the planned trajectory includes:

[0028] Determine the intersection point between the planned trajectory and the monitoring coverage area to obtain the location of the first intersection point;

[0029] Determine the intersection point between the planned trajectory and the next monitoring coverage area to obtain the location of the second intersection point;

[0030] Calculate the second distance between the first intersection point and the second intersection point;

[0031] The second travel time is calculated based on the second distance and the vehicle speed.

[0032] Optionally, before determining the next monitoring coverage area of ​​the next roadside device of the roadside device, and determining the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area, the method further includes:

[0033] Determine whether the next-side device is in working condition;

[0034] If so, then the steps of determining the next monitoring coverage area of ​​the next roadside device of the roadside device and determining the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area are executed;

[0035] If not, then determine the start time of the next roadside device, and calculate the second travel time based on the start time and the first travel time.

[0036] Optionally, calculating the second driving time based on the start time and the first driving time includes:

[0037] Determine the current time point;

[0038] The departure time of the target vehicle from the monitoring coverage area is calculated based on the current time and the first driving time.

[0039] The second travel time is calculated based on the start time and the departure time.

[0040] Optionally, the system parameters include vehicle-side remaining energy, roadside current computing power, vehicle-side average power, and roadside average power, and the preset index parameters are the computing service accuracy gains of the target vehicle within the monitoring blind zone.

[0041] Accordingly, the step of constructing an objective function for optimizing preset index parameters using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task load includes:

[0042] Under preset constraints, the objective function is constructed using the first travel time, the second travel time, the system parameters, and the total task volume.

[0043] The preset constraints include a first constraint and a second constraint.

[0044] The first constraint is that the sum of the time taken by the roadside equipment to process the roadside task and the time taken to deploy the roadside computing service and the vehicle-side computing service to the target vehicle is less than the first travel time.

[0045] The second constraint is that the total energy consumption of the target vehicle in running the roadside computing service and the vehicle-side computing service is less than the remaining energy on the vehicle side.

[0046] Optionally, solving the objective function to obtain the vehicle-side task quantity and the roadside task quantity includes:

[0047] The objective function is solved using the interior point method to obtain the task segmentation ratio.

[0048] The vehicle-side task volume and the roadside task volume are calculated based on the total task volume and the task segmentation ratio.

[0049] Optionally, solving the objective function to obtain the vehicle-side task quantity and the roadside task quantity includes:

[0050] The objective function is solved using the simplex method to obtain the task segmentation ratio;

[0051] The vehicle-side task volume and the roadside task volume are calculated based on the total task volume and the task segmentation ratio.

[0052] Optionally, the step of deploying the roadside processing results as roadside computing services and the vehicle-side tasks corresponding to the vehicle-side task volume as vehicle-side computing services to the target vehicle includes:

[0053] The roadside computing service is packaged to obtain a first container image;

[0054] The vehicle-side computing service is packaged to obtain a second container image;

[0055] Deploy the first container image and the second container image to the target vehicle.

[0056] Optionally, the task scheduling method further includes:

[0057] An online decision-making network model is constructed using historical data of the target vehicle at various historical moments. The historical data includes the system state, decision actions, system benefits at the corresponding historical moment, and the system state at the next historical moment. The system state includes the current vehicle-side computing power, roadside computing service volume, and vehicle-side computing service volume at the corresponding historical moment. The decision actions include the computing power allocated by the target vehicle to the roadside computing service and the computing power allocated to the vehicle-side computing service at the corresponding historical moment.

[0058] Obtain the system status of the target vehicle at the current moment;

[0059] The system state is processed using the online decision network model to obtain the decision action of the target vehicle at the current moment;

[0060] The decision-making action is performed using the target vehicle.

[0061] Optionally, the task scheduling method further includes:

[0062] The online decision network model is updated at preset time intervals to obtain the updated online decision network model.

[0063] Secondly, this application also discloses a task scheduling device for use in a vehicle-road cooperative system, comprising:

[0064] The monitoring module is used to monitor the target vehicle in real time using roadside equipment, and to obtain the first driving time of the target vehicle within the monitoring coverage area and the second driving time within the monitoring blind zone.

[0065] The construction module is used to construct an objective function about the optimal preset index parameters using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task volume;

[0066] The solution module is used to solve the objective function to obtain the vehicle-side task quantity and the roadside task quantity;

[0067] The processing module is used to process the roadside tasks corresponding to the roadside task volume using the roadside equipment, and obtain the roadside processing results.

[0068] The deployment module is used to deploy the roadside processing results as roadside computing services and the vehicle-side tasks corresponding to the vehicle-side task volume as vehicle-side computing services to the target vehicle, so as to provide computing services for the target vehicle when it is driving within the monitoring blind zone.

[0069] Thirdly, this application also discloses an electronic device, comprising:

[0070] Memory, used to store computer programs;

[0071] A processor, used to implement any of the task scheduling methods described above when executing the computer program.

[0072] Fourthly, this application also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of any of the task scheduling methods described above.

[0073] This application provides a task scheduling method, including using roadside equipment to monitor a target vehicle in real time, obtaining a first travel time of the target vehicle within the monitoring coverage area and a second travel time within the monitoring blind zone; constructing an objective function about optimal preset index parameters using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task volume; solving the objective function to obtain the vehicle-side task volume and the roadside task volume; processing the roadside tasks corresponding to the roadside task volume using the roadside equipment to obtain roadside processing results; deploying the roadside processing results as roadside computing services and the vehicle-side tasks corresponding to the vehicle-side task volume as vehicle-side computing services to the target vehicle to provide computing services when the target vehicle is traveling within the monitoring blind zone.

[0074] By applying the technical solution provided in this application, when a target vehicle is traveling within the monitoring coverage area of ​​a roadside device, the roadside device monitors it in real time to determine the time it spends within the monitoring coverage area and the time it spends within the monitoring blind zone. Then, combining system parameters and current total task volume, a target function is constructed to optimize a preset index parameter (such as maximum efficiency or maximum accuracy). The vehicle-side task volume and roadside task volume are calculated by solving this target function. The tasks corresponding to the vehicle-side task volume are executed by the target vehicle, while the tasks corresponding to the roadside task volume are executed in advance by the roadside device. Therefore, after the roadside device completes its roadside task processing, the roadside processing results are used as roadside computing services, and the vehicle-side tasks are used as vehicle-side computing services, both deployed to the target vehicle to provide computing services when the target vehicle is traveling within the monitoring blind zone. Thus, even when the vehicle is traveling within the monitoring blind zone, vehicle-road cooperative task processing can still be maintained, ensuring that the vehicle still has continuous and efficient computing power within the roadside device coverage blind zone, thereby guaranteeing the vehicle's continuous, safe, and stable operation.

[0075] The task scheduling device, electronic device, and computer-readable storage medium provided in this application also have the above-mentioned technical effects, and will not be described in detail here. Attached Figure Description

[0076] To more clearly illustrate the technical solutions in the prior art and the embodiments of this application, the accompanying drawings used in the description of the prior art and the embodiments of this application will be briefly introduced below. Of course, the accompanying drawings described below with respect to the embodiments of this application are only a part of the embodiments in this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort, and such other drawings also fall within the protection scope of this application.

[0077] Figure 1 A flowchart illustrating a task scheduling method provided in this application;

[0078] Figure 2 A schematic diagram of a task scheduling device provided in this application;

[0079] Figure 3 This is a schematic diagram of the structure of an electronic device provided in this application. Detailed Implementation

[0080] The core of this application is to provide a task scheduling method that can effectively ensure that vehicles have continuous and efficient computing power service capabilities in the blind spots of roadside equipment coverage, thereby ensuring the continuous, safe and stable operation of vehicles. Another core aspect of this application is to provide a task scheduling device, electronic device and computer-readable storage medium, all of which have the above-mentioned beneficial effects.

[0081] To provide a clearer and more complete description of the technical solutions in the embodiments of this application, the technical solutions in the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0082] This application provides a task scheduling method.

[0083] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a task scheduling method provided in this application. The task scheduling method is applied to a vehicle-road cooperative system and may include the following steps S101 to S105.

[0084] S101: Real-time monitoring of target vehicles using roadside equipment to obtain the first travel time of the target vehicle within the monitoring coverage area and the second travel time within the monitoring blind zone.

[0085] This step aims to enable real-time monitoring of target vehicles by roadside equipment to determine the time spent traveling within the monitoring coverage area (i.e., the first travel time mentioned above) and the time spent traveling within the monitoring blind zone (i.e., the second travel time mentioned above). The target vehicle is the vehicle requiring vehicle-to-infrastructure (V2I) task scheduling. The monitoring coverage area refers to the area covered by the roadside equipment, while the monitoring blind zone refers to the area that the roadside equipment cannot monitor, generally the blind zone between two adjacent roadside equipment units. Based on this, adjacent roadside equipment units can communicate in real-time and simultaneously monitor the target vehicle. Then, based on the monitored information, the first travel time within the monitoring coverage area and the second travel time within the monitoring blind zone of the target vehicle are calculated.

[0086] S102: Construct an objective function for optimizing preset index parameters using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task.

[0087] This step aims to construct an objective function, which is used to optimize a certain preset index parameter. Its specific content can be set by technical personnel according to the current actual needs. For example, it can be an objective function to achieve the highest task scheduling efficiency, aiming to achieve the most efficient task scheduling; or it can be an objective function to achieve the highest precision computing service, aiming to achieve the most accurate task scheduling. This application does not limit this.

[0088] In the implementation process, after obtaining the first and second travel times, the system parameters of the vehicle-road cooperative system (such as the current computing power of the target vehicle and the current computing power of the roadside equipment) and the current total workload of the vehicle cooperative task can be further obtained. These data are then combined to construct the aforementioned objective function. It is understood that different preset index parameters may correspond to different types of system parameters; therefore, various types of data information can be obtained according to the actual situation.

[0089] S103: Solve the objective function to obtain the vehicle-side task quantity and the roadside task quantity.

[0090] This step aims to solve the objective function to obtain the vehicle-side task quantity and the roadside task quantity. The vehicle-side task quantity refers to the amount of computational tasks performed by the target vehicle (referred to as vehicle-side tasks) within the current total task quantity; the roadside task quantity refers to the amount of computational tasks pre-executed by roadside equipment within the current total task quantity. Clearly, vehicle task quantity + roadside task quantity = total task quantity.

[0091] S104: Use roadside equipment to process the roadside tasks corresponding to the roadside task volume and obtain the roadside processing results.

[0092] This step aims to enable roadside equipment to preprocess roadside tasks. Preprocessing here refers to handling roadside tasks before the target vehicle leaves the monitoring coverage area. Specifically, roadside equipment can be used directly to process the roadside tasks corresponding to the roadside task volume to obtain the corresponding processing results, which are the roadside processing results.

[0093] S105: The roadside processing results are used as roadside computing services, and the vehicle-side tasks corresponding to the vehicle-side task volume are used as vehicle-side computing services, which are deployed to the target vehicle to provide computing services when the target vehicle is driving within the monitoring blind zone.

[0094] This step aims to achieve the overall deployment of computing services. Building upon the partial computing tasks performed on the target vehicle and roadside equipment respectively, the roadside processing results can be deployed as roadside computing services, and the vehicle-side tasks as vehicle-side computing services, both deployed to the target vehicle. This allows computing services to continue being provided even when the target vehicle is traveling within the monitoring blind spot. Clearly, since the roadside computing services are already processed by the roadside equipment, when the vehicle is traveling within the monitoring blind spot, it only needs to run the roadside computing services and simultaneously execute its own vehicle-side computing services (i.e., execute its vehicle-side tasks) to maintain vehicle-road cooperative task processing.

[0095] As can be seen, the task scheduling method provided in this application, when the target vehicle is traveling within the monitoring coverage area of ​​the roadside equipment, utilizes the roadside equipment to monitor it in real time to determine the time it spends traveling within the monitoring coverage area and the time it spends traveling within the monitoring blind zone. Then, it combines system parameters, the current total task volume, and other data to construct an objective function for optimizing a certain preset index parameter (such as maximum efficiency, maximum accuracy, etc.). By solving this objective function, the vehicle-side task volume and the roadside task volume are calculated. The tasks corresponding to the vehicle-side task volume are executed by the target vehicle, while the tasks corresponding to the roadside task volume are executed in advance by the roadside equipment. Therefore, after the roadside equipment has processed the roadside tasks, the roadside processing results are used as roadside computing services, and the vehicle-side tasks are used as vehicle-side computing services, and both are deployed to the target vehicle to provide computing services when the target vehicle is traveling within the monitoring blind zone. Thus, when the vehicle is traveling within the monitoring blind zone, vehicle-road cooperative task processing can still be maintained, thereby ensuring that the vehicle still has continuous and efficient computing power service capabilities within the roadside equipment coverage blind zone, thus ensuring the continuous, safe, and stable operation of the vehicle.

[0096] Based on the above embodiments:

[0097] In one embodiment of this application, using roadside equipment to monitor a target vehicle in real time and obtain the first travel time of the target vehicle within the monitoring coverage area may include the following steps:

[0098] Real-time monitoring of target vehicles using roadside equipment can obtain their current location, speed, and planned trajectory, and determine the monitoring coverage area of ​​the roadside equipment.

[0099] The first travel time of the target vehicle within the monitoring coverage area is calculated using the current location, driving speed, and planned trajectory.

[0100] This application provides a method for determining the first travel time. First, a roadside device can be used to monitor the target vehicle in real time to obtain data such as its current location, speed, and planned trajectory (i.e., the planned travel trajectory of the target vehicle). At the same time, the monitoring coverage area of ​​the roadside device is obtained, and the first travel time of the target vehicle within the monitoring coverage area is calculated based on this data.

[0101] In one embodiment of this application, the above-mentioned use of roadside equipment to monitor the target vehicle in real time and obtain the target vehicle's current position, speed, and planned trajectory may include the following steps:

[0102] By using roadside equipment to communicate wirelessly with the target vehicle, the current location, speed, and planned trajectory of the target vehicle can be obtained.

[0103] This application provides a method for acquiring data related to target vehicles. Specifically, roadside equipment and each target vehicle can communicate in real time via wireless communication technology, thereby enabling the roadside equipment to monitor the target vehicles in real time and obtain the various types of data information fed back by the target vehicles.

[0104] In one embodiment of this application, the above-mentioned calculation of the first travel time of the target vehicle within the monitoring coverage area using the current location, driving speed, and planned trajectory may include the following steps:

[0105] Determine the intersection point between the planned trajectory and the monitoring coverage area to obtain the location of the first intersection point;

[0106] Calculate the first distance between the current position and the first intersection point;

[0107] The first travel time is calculated based on the first distance and the vehicle speed.

[0108] This application provides a method for calculating a first travel time based on various data information fed back by a target vehicle. First, after determining the planned trajectory of the target vehicle and the monitoring coverage area of ​​the roadside equipment, the intersection of the two can be further determined to obtain the first intersection point. Clearly, before the target vehicle reaches this first intersection point, it is within the monitoring coverage area; after the target vehicle leaves the first intersection point, it is within the monitoring blind zone. Next, the distance between the target vehicle's current location and the first intersection point is calculated, i.e., the aforementioned first distance, which is the travel distance of the target vehicle within the monitoring coverage area. Finally, the quotient between the first distance and the vehicle speed is calculated, and this quotient is the first travel time of the target vehicle within the monitoring coverage area.

[0109] In one embodiment of this application, using roadside equipment to monitor a target vehicle in real time and obtain the second driving time of the target vehicle within the monitoring blind zone may include the following steps:

[0110] Determine the next monitoring coverage area of ​​the next roadside device, and determine the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area;

[0111] The second travel time of the target vehicle within the monitoring blind zone is calculated using the current location, driving speed, and planned trajectory.

[0112] This application provides a method for determining a second travel time. Specifically, after calculating the first travel time, the current roadside device can further determine the next monitoring coverage area of ​​the next roadside device, which is the roadside device adjacent to the current roadside device that the target vehicle is about to pass through. Thus, the monitoring blind zone range that the target vehicle is about to enter can be determined based on the monitoring coverage areas of the current roadside device and the next roadside device. At the same time, the second travel time of the target vehicle within the monitoring blind zone range can be calculated by combining the various data information obtained above.

[0113] In one embodiment of this application, determining the next monitoring coverage area of ​​the next roadside device may include the following steps:

[0114] Based on the planned trajectory, predict the route and determine the next roadside equipment for the roadside equipment.

[0115] It communicates with the next-side device to obtain the next monitoring coverage area of ​​the next-side device.

[0116] This application provides a method for obtaining the next monitoring coverage area of ​​a roadside device. Specifically, the current roadside device can predict the route based on the planned trajectory of the target vehicle, thereby determining the next roadside device the target vehicle is about to approach; then, the various roadside devices can communicate wirelessly, thus enabling the current roadside device to communicate with the next roadside device to determine the next monitoring coverage area of ​​the next roadside device.

[0117] In one embodiment of this application, the above-mentioned calculation of the second driving time of the target vehicle within the monitoring blind zone using the current location, driving speed, and planned trajectory may include the following steps:

[0118] Determine the intersection point between the planned trajectory and the monitoring coverage area to obtain the location of the first intersection point;

[0119] Determine the intersection point between the planned trajectory and the next monitoring coverage area to obtain the location of the second intersection point;

[0120] Calculate the second distance between the first intersection point and the second intersection point;

[0121] The second travel time is calculated based on the second distance and the vehicle speed.

[0122] This application provides a method for calculating a second travel time based on various data. First, after determining the planned trajectory of the target vehicle and the monitoring coverage area of ​​the roadside equipment, the intersection of the two can be further determined to obtain the first intersection point. Clearly, before the target vehicle reaches this first intersection point, it is within the monitoring coverage area; after leaving the first intersection point, it is within the monitoring blind zone. Second, after determining the next monitoring coverage area of ​​the next roadside equipment, the intersection of the target vehicle's planned trajectory and the next monitoring coverage area can be further determined to obtain the second intersection point. Clearly, before the target vehicle reaches this second intersection point, it is within the monitoring blind zone; after reaching the second intersection point, it is within the monitoring coverage area. Therefore, the distance between the first and second intersection points (i.e., the aforementioned second distance) is the travel distance of the target vehicle within the monitoring blind zone. Thus, by calculating the quotient between the second distance and the vehicle speed, the second travel time of the target vehicle within the monitoring blind zone can be obtained.

[0123] In one embodiment of this application, before determining the next monitoring coverage area of ​​the next roadside device and determining the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area, the following steps may be included:

[0124] Determine whether the next-side equipment is in working condition;

[0125] If so, proceed with the steps of determining the next monitoring coverage area of ​​the next roadside device and determining the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area;

[0126] If not, determine the start time of the next roadside device, and calculate the second travel time based on the start time and the first travel time.

[0127] It is understandable that, for various reasons, the next roadside device may not be in operation, thus causing the original monitoring coverage area of ​​the next roadside device to become a blind spot for the target vehicle. Based on this, before determining the next monitoring coverage area of ​​the next roadside device, it is possible to first determine whether the next roadside device is in operation through data communication. If it is in operation, the original steps can be continued. If the next roadside device is not in operation, the start time of the next roadside device can be determined first, and then the second travel time can be calculated by combining it with the first travel time mentioned above.

[0128] In one embodiment of this application, the calculation of the second driving time based on the start time and the first driving time may include the following steps:

[0129] Determine the current time point;

[0130] The departure time of the target vehicle from the monitoring coverage area is calculated based on the current time and the first driving time.

[0131] The second travel time is calculated based on the start time and departure time.

[0132] This application provides an implementation method for calculating a second form based on a start time and a first travel time. It is understood that after the target vehicle leaves the monitoring coverage area of ​​the current roadside equipment but before the start time, it will continue to travel within the monitoring blind zone. At this time, the exit time of the target vehicle from the monitoring coverage area can be calculated based on the current time and the first travel time. The difference between the start time and the exit time is then used as the second travel time of the target vehicle within the monitoring blind zone.

[0133] In one embodiment of this application, the system parameters may include vehicle-side remaining energy, roadside current computing power, vehicle-side average power, and roadside average power. The preset index parameters may be the computing service accuracy gain of the target vehicle within the monitoring blind zone.

[0134] Accordingly, the above-mentioned construction of an objective function for optimizing preset index parameters using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task can include the following steps:

[0135] Under preset constraints, an objective function is constructed using the first driving time, the second driving time, system parameters, and the total task load.

[0136] The preset constraints include the first constraint and the second constraint.

[0137] The first constraint is that the sum of the time taken for the roadside equipment to process roadside tasks and the time taken to deploy roadside computing services and vehicle-side computing services to the target vehicle is less than the first travel time.

[0138] The second constraint is that the total energy consumption of the target vehicle's roadside computing service and vehicle-side computing service is less than the vehicle-side remaining energy.

[0139] This application provides a specific type of preset index parameter, namely, the computational service accuracy gain of the target vehicle within the monitoring blind zone. In other words, the objective function is a function for optimizing the computational service accuracy gain, aiming to achieve the most accurate task scheduling. Correspondingly, system parameters may include data such as vehicle-side remaining energy, roadside current computing power, vehicle-side average power, and roadside average power. Further, under the first and second constraints, the objective function can be constructed using the above-mentioned data information, wherein:

[0140] The first constraint is: the time taken for roadside equipment to process roadside tasks plus the time taken to deploy roadside computing services and vehicle-side computing services to the target vehicle < the first travel time.

[0141] The second constraint is: the total energy consumption of the target vehicle's roadside computing service and vehicle-side computing service is less than the vehicle-side remaining energy.

[0142] It is understandable that the preset constraints are intended to ensure that the target vehicle can drive safely and stably, and can be adaptively adjusted according to actual needs. This application does not limit this.

[0143] In one embodiment of this application, solving the objective function to obtain the vehicle-side task quantity and the roadside task quantity may include the following steps:

[0144] The objective function is solved using the interior point method to obtain the task segmentation ratio;

[0145] The vehicle-side task volume and roadside task volume are calculated based on the total task volume and the task segmentation ratio.

[0146] This application provides a method for solving the objective function, namely the interior-point method (a polynomial algorithm for solving optimization theory). The interior-point method is used to obtain the task segmentation ratio, that is, the task segmentation ratio with respect to the total task volume, which is the proportion of vehicle-side tasks in the total task and the proportion of roadside tasks in the total task. Then, the vehicle-side task volume and the roadside task volume are calculated based on the task segmentation ratio and the total task volume.

[0147] In one embodiment of this application, solving the objective function to obtain the vehicle-side task quantity and the roadside task quantity may include the following steps:

[0148] The objective function is solved using the simplex method to obtain the task segmentation ratio;

[0149] The vehicle-side task volume and roadside task volume are calculated based on the total task volume and the task segmentation ratio.

[0150] This application provides another method for solving the objective function, namely the simplex method (a linear programming algorithm). The task segmentation ratio is obtained by solving the simplex method, that is, the task segmentation ratio with respect to the total task volume, which is the proportion of vehicle-side tasks in the total task and the proportion of roadside tasks in the total task. The vehicle-side task volume and the roadside task volume are then calculated based on the task segmentation ratio and the total task volume.

[0151] In one embodiment of this application, the above-mentioned deployment of roadside processing results as roadside computing services and vehicle-side tasks corresponding to vehicle-side task quantities as vehicle-side computing services to the target vehicle may include the following steps:

[0152] Package the roadside computing service to obtain the first container image;

[0153] Package the vehicle-side computing service to obtain a second container image;

[0154] Deploy the first container image and the second container image to the target vehicle.

[0155] This application provides a method for deploying roadside computing services and vehicle-side computing services to a target vehicle, which can be implemented based on container technology. It is understood that container technology is a lightweight virtualization technology that can effectively manage and isolate computing resources, enabling the system to run multiple services simultaneously without interference. It can also easily migrate, scale up, and reduce service images across various computer hardware in response to changes in computing load, dynamically meeting task requirements and improving the utilization of computer hardware resources. Therefore, roadside computing services and vehicle-side computing services can be packaged into container images, namely the first container image and the second container image mentioned above, thereby deploying the roadside computing services and vehicle-side computing services to the target vehicle in the form of container images.

[0156] In one embodiment of this application, the task scheduling method may further include the following steps:

[0157] An online decision-making network model is constructed using historical data of the target vehicle at each historical moment. The historical data includes the system state, decision actions, system benefits at the corresponding historical moment, and the system state at the next historical moment. The system state includes the current computing power of the vehicle side, the amount of roadside computing services, and the amount of vehicle-side computing services at the corresponding historical moment. The decision actions include the computing power allocated to the roadside computing services and the computing power allocated to the vehicle-side computing services by the target vehicle at the corresponding historical moment.

[0158] Obtain the system status of the target vehicle at the current moment;

[0159] The system state is processed using an online decision network model to obtain the target vehicle's decision action at the current moment;

[0160] Utilize the target vehicle to execute decision-making actions.

[0161] After deploying vehicle-side computing services and roadside computing services to the target vehicle, the target vehicle can then implement vehicle-road cooperative computing services by running these services. To further achieve the rational allocation of computing resources within the vehicle and minimize task processing latency, an online decision network model corresponding to the target vehicle can be constructed. This online decision network model can then be used to rationally allocate computing resources between the vehicle-side computing services and roadside computing services within the target vehicle.

[0162] The online decision-making network model can be constructed based on historical data of the target vehicle at various historical moments, including the aforementioned data types. Furthermore, when using the online decision-making network model for computing resource allocation, the system state of the target vehicle at the current moment can be obtained first and input into the online decision-making network model for processing. The model's output is the target vehicle's decision action at the current moment, which the target vehicle then executes. Specifically, the decision action involves the target vehicle allocating the appropriate amount of computing resources to both vehicle-side computing services and roadside computing services.

[0163] In one embodiment of this application, the task scheduling method may further include the following steps:

[0164] The online decision network model is updated at preset time intervals to obtain the updated online decision network model.

[0165] To further ensure model accuracy and thus the accuracy of task scheduling results, the aforementioned online decision network model can be updated at preset time intervals. The data used for these updates includes various data generated by the target vehicle within the preset time period. The specific value of the preset time interval does not affect the implementation of this technical solution and can be set by technical personnel according to actual conditions; this application does not impose any limitations on it.

[0166] Based on the above embodiments:

[0167] This application provides another task scheduling method, the implementation process of which is as follows:

[0168] I. Early warning of blind spots in roadside equipment coverage:

[0169] (1) Current roadside equipment RSU's monitoring coverage area R curt The system tracks the location of vehicles within the monitoring area in real time, obtaining the current location and planned trajectory of the target vehicle to determine the estimated time T (first travel time) when the target vehicle is expected to leave the monitoring coverage area.

[0170]

[0171] in, Indicates the target vehicle's current location to the planned trajectory and the monitoring coverage area R. curt Intersection point O curt The length of the trajectory between the two points, where v represents the speed of the target vehicle.

[0172] (2) The current RSU is based on the target vehicle in O curt The planned trajectory after the point is determined, the next RSU is identified and communicated with to obtain its working status;

[0173] If the next RSU can achieve seamless coverage (i.e., there is no monitoring blind zone between two adjacent RSUs), then no further predictions or actions will be taken.

[0174] If there is a blind spot between the next RSU and the current RSU, calculate the time H (second travel time) for the target vehicle to pass through the blind spot:

[0175]

[0176] in, This indicates the planned trajectory of the target vehicle and the monitoring coverage area R of the next RSU. next The intersection point to O curt The length of the trajectory between points, where v represents the speed of the target vehicle.

[0177] In addition, if the next RSU is not in a working state, its startup time T is obtained. rst This allows us to calculate the time the target vehicle spends within the monitoring blind spot.

[0178]

[0179] in, This indicates that the target vehicle has arrived at O. curt The estimated time for the point.

[0180] II. Vehicle-Road Cooperative Computing Task Scheduling:

[0181] (1) System parameter settings:

[0182] In the vehicle-road cooperative system, the main system parameters are: the target vehicle's travel time T within the current RSU's monitoring coverage area, the estimated travel time H within the monitoring blind zone, the target vehicle's current remaining energy E (i.e., the vehicle-side remaining energy mentioned above), the total workload W of vehicle-road cooperation, and the current RSU's computing power C (i.e., the current roadside computing power mentioned above).

[0183] If the current RSU decides to preprocess some computational tasks for the target vehicle to meet the goals of optimal accuracy of vehicle driving services and optimal vehicle driving energy efficiency, it is necessary to calculate the division ratio x of the total computational tasks. x represents the proportion of tasks in W that need to be preprocessed by the current RSU and a container image is built and sent to the target vehicle to provide continuous services. This is called offline tasks (i.e., the roadside computing services mentioned above). The remaining tasks in W are sent to the target vehicle for real-time computation. This is called online tasks (i.e., the vehicle-side computing services mentioned above).

[0184] For example, the driving decisions of autonomous vehicles need to be generated based on traffic congestion and other conditions along the current route. When the vehicle can obtain road condition information ahead, it can generate more accurate driving behavior decisions, such as reducing cruising speed before entering congested areas. When the vehicle cannot obtain real-time road condition information within a certain period, it can calculate in advance, based on the current road conditions, to predict the road conditions a certain period later, and plan driving decisions based on the predicted information, providing driving decision information as close as possible to the actual situation.

[0185] Considering the positive and negative impacts of real-time and offline execution of computational tasks on vehicle service accuracy and safety, we assume that the coefficient of influence of a unit of real-time computational task on vehicle service accuracy is k1, the coefficient of influence of a unit of offline execution task on vehicle service accuracy is k2, the average power of the target vehicle processing a unit of real-time task is P1 (i.e., the average power on the vehicle side mentioned above), and the average power of processing a unit of offline task is P2 (i.e., the average power on the roadside mentioned above). Since real-time computation requires acquiring and rapidly processing real-time perception information, its power consumption is higher than that of offline task processing, but the computational benefits are relatively higher. Therefore, the more computational tasks are processed in real-time, the higher the vehicle service accuracy and safety, but the energy consumption per vehicle will increase significantly.

[0186] (2) Establish the objective function:

[0187] Based on the above parameters, an objective function is established to maximize the computational service accuracy benefit of the target vehicle within the monitoring blind zone, while satisfying the corresponding constraints:

[0188] max k1(1-x)W+k2xW;

[0189]

[0190] H[P1(1-x)W+P2W D ] <E:

[0191] The objective function is to maximize the computational gain of the target vehicle within the monitoring blind zone.

[0192] Constraint 1 (first constraint) is that the sum of the time taken for the current RSU to compute the offline task and the time taken to package the task results and the online task into container images and send them to the target vehicle is less than the time it takes for the target vehicle to leave its monitoring coverage area (i.e., the first driving time), to ensure that the computing task can be configured to the vehicle in a timely manner. Wherein, W D T represents the amount of data required to run an offline task container image. RV This indicates the transmission bandwidth between the target vehicle and the roadside equipment.

[0193] Constraint 2 (second constraint) requires that the total energy consumption of the target vehicle running real-time and offline tasks within the monitoring coverage blind zone be less than the remaining energy consumption of the target vehicle (remaining energy on the vehicle side) to ensure the continuity of single-vehicle service.

[0194] (3) Solve for the objective function:

[0195] The objective function is solved using the simplex method or the interior point method to obtain the optimal solution for x.

[0196] III. Dynamic allocation of container resources:

[0197] (1) Based on the optimal x value, the current RSU determines the amount of offline tasks (i.e., roadside tasks) that need to be pre-calculated for the target vehicle. After the RSU completes this part of the calculation task, it packages it into a container image (first container image) and pushes it to the target vehicle. The remaining calculation tasks are real-time calculation tasks, which are also packaged into a container image (second container image) and pushed to the target vehicle. The two container images run simultaneously to provide continuous calculation services for the target vehicle while it is driving within the monitoring coverage area.

[0198] (2) When two container images are deployed on the target vehicle, the corresponding computing resources can be allocated to each container according to the vehicle’s current available computing power.

[0199] Let W represent the maximum computing power of the target vehicle. Let W be the real-time and offline task loads at time h (here, a historical time). os and W rt The computing power allocated to the offline service container is C. os The computing power allocated to the real-time service container is C. rt Therefore, the average latency of the task needs to be minimized:

[0200]

[0201] Since the amount of input tasks at each moment is random, it is necessary to dynamically adjust the computing resources allocated to different containers. To address this, an online resource allocation method can be used to allocate computing resources to containers at different times.

[0202] (3) Determine the decision variable C at each time point h. os and And satisfy Let the observed system state at time h be... Then decision action The system benefit obtained from executing this decision action is:

[0203]

[0204] (4) Construct an online decision network model Q(ω) based on deep reinforcement learning, where ω is the parameter of the Q network. The network input is s, and the output is the decision action a, which is executed to obtain the system reward r at the current stage. Then the system enters the next observation state s′. Further, {s, a, r, s′} are stored in the cache pool D. When the information stored in D is greater than M, (5) is executed.

[0205] (5) Randomly sample M samples from D, and calculate the target Q value according to the following formula:

[0206] Q target =r+γ*min a′ (Q(s′,a′,ω));

[0207] Where γ is a real coefficient between 0 and 1, the loss function L is obtained:

[0208]

[0209] The network parameters ω are updated using gradient backpropagation until the network converges, resulting in an online decision network model.

[0210] (6) The online decision network model processes the system state at the current moment, outputs the decision action at the current moment, and executes it so as to allocate the computing resources of the target vehicle to different containers.

[0211] As can be seen, the task scheduling method provided in this application, when the target vehicle is traveling within the monitoring coverage area of ​​the roadside equipment, utilizes the roadside equipment to monitor it in real time to determine the time it spends traveling within the monitoring coverage area and the time it spends traveling within the monitoring blind zone. Then, it combines system parameters, the current total task volume, and other data to construct an objective function for optimizing a certain preset index parameter (such as maximum efficiency, maximum accuracy, etc.). By solving this objective function, the vehicle-side task volume and the roadside task volume are calculated. The tasks corresponding to the vehicle-side task volume are executed by the target vehicle, while the tasks corresponding to the roadside task volume are executed in advance by the roadside equipment. Therefore, after the roadside equipment has processed the roadside tasks, the roadside processing results are used as roadside computing services, and the vehicle-side tasks are used as vehicle-side computing services, and both are deployed to the target vehicle to provide computing services when the target vehicle is traveling within the monitoring blind zone. Thus, when the vehicle is traveling within the monitoring blind zone, vehicle-road cooperative task processing can still be maintained, thereby ensuring that the vehicle still has continuous and efficient computing power service capabilities within the roadside equipment coverage blind zone, thus ensuring the continuous, safe, and stable operation of the vehicle.

[0212] This application provides a task scheduling device.

[0213] Please refer to Figure 2 , Figure 2 This application provides a schematic diagram of a task scheduling device, which is applied to a vehicle-road cooperative system and may include:

[0214] Monitoring module 1 is used to monitor the target vehicle in real time using roadside equipment, and obtain the first travel time of the target vehicle within the monitoring coverage area and the second travel time within the monitoring blind zone.

[0215] Module 2 is used to construct an objective function about the optimal preset index parameters using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task volume;

[0216] Solver module 3 is used to solve the objective function to obtain the vehicle-side task quantity and the roadside task quantity;

[0217] Processing module 4 is used to process roadside tasks corresponding to roadside task volume using roadside equipment and obtain roadside processing results;

[0218] Deployment module 5 is used to deploy the roadside processing results as roadside computing services and the vehicle-side tasks corresponding to the vehicle-side task volume as vehicle-side computing services to the target vehicle, so as to provide computing services for the target vehicle when it is driving within the monitoring blind zone.

[0219] As can be seen, the task scheduling device provided in this application, when the target vehicle is traveling within the monitoring coverage area of ​​the roadside equipment, utilizes the roadside equipment to monitor it in real time to determine the time it spends traveling within the monitoring coverage area and the time it spends traveling within the monitoring blind zone. Then, it combines system parameters, the current total task volume, and other data to construct an objective function for optimizing a certain preset index parameter (such as maximum efficiency, maximum accuracy, etc.). By solving this objective function, the vehicle-side task volume and the roadside task volume are calculated. The tasks corresponding to the vehicle-side task volume are executed by the target vehicle, while the tasks corresponding to the roadside task volume are executed in advance by the roadside equipment. Therefore, after the roadside equipment completes the processing of the roadside tasks, the roadside processing results are used as roadside computing services, and the vehicle-side tasks are used as vehicle-side computing services, and both are deployed to the target vehicle to provide computing services when the target vehicle is traveling within the monitoring blind zone. Thus, when the vehicle is traveling within the monitoring blind zone, it can still maintain vehicle-road cooperative task processing, thereby ensuring that the vehicle still has continuous and efficient computing power service capabilities within the roadside equipment coverage blind zone, thus ensuring the continuous, safe, and stable operation of the vehicle.

[0220] In one embodiment of this application, the monitoring module 1 may include:

[0221] The monitoring unit is used to monitor target vehicles in real time using roadside equipment, obtain the target vehicle's current location, driving speed, planned trajectory, and determine the monitoring coverage area of ​​the roadside equipment;

[0222] The first calculation unit is used to calculate the first travel time of the target vehicle within the monitoring coverage area using the current location, driving speed, and planned trajectory.

[0223] In one embodiment of this application, the monitoring unit described above can be specifically used to wirelessly communicate with the target vehicle using roadside equipment to obtain the target vehicle's current location, driving speed, and planned trajectory.

[0224] In one embodiment of this application, the first calculation unit can be specifically used to determine the intersection of the planned trajectory and the monitoring coverage area to obtain the first intersection location; calculate the first distance between the current location and the first intersection location; and calculate the first travel time based on the first distance and the driving speed.

[0225] In one embodiment of this application, the monitoring module 1 may include:

[0226] The determination unit is used to determine the next monitoring coverage area of ​​the next roadside device of the roadside device, and to determine the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area;

[0227] The second calculation unit is used to calculate the second driving time of the target vehicle within the monitoring blind zone using the current location, driving speed, and planned trajectory.

[0228] In one embodiment of this application, the aforementioned determining unit may be specifically used to predict the route based on the planned trajectory, determine the next roadside device of the roadside equipment, and communicate with the next roadside device to obtain the next monitoring coverage area of ​​the next roadside device.

[0229] In one embodiment of this application, the second calculation unit can be specifically used to determine the intersection of the planned trajectory and the monitoring coverage area to obtain the first intersection point position; determine the intersection of the planned trajectory and the next monitoring coverage area to obtain the second intersection point position; calculate the second distance between the first intersection point position and the second intersection point position; and calculate the second travel time based on the second distance and the driving speed.

[0230] In one embodiment of this application, the monitoring module 1 may further include a judgment unit, used to determine whether the next roadside device is in operation before determining the next monitoring coverage area of ​​the next roadside device and determining the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area; if yes, then the steps of determining the next monitoring coverage area of ​​the next roadside device and determining the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area are executed; if no, then the start time of the next roadside device is determined, and the second travel time is calculated based on the start time and the first travel time.

[0231] In one embodiment of this application, the aforementioned determination unit may be specifically used to determine the current time point; calculate the exit time point of the target vehicle from the monitoring coverage area based on the current time point and the first driving time; and calculate the second driving time based on the start time point and the exit time point.

[0232] In one embodiment of this application, the system parameters may include vehicle-side remaining energy, roadside current computing power, vehicle-side average power, and roadside average power. The preset index parameters may be the computing service accuracy gain of the target vehicle within the monitoring blind zone.

[0233] Accordingly, the aforementioned construction module 2 can be specifically used to construct an objective function using a first driving time, a second driving time, system parameters, and total task volume under preset constraints; wherein, the preset constraints include a first constraint and a second constraint.

[0234] The first constraint is that the sum of the time taken for the roadside equipment to process roadside tasks and the time taken to deploy roadside computing services and vehicle-side computing services to the target vehicle is less than the first travel time.

[0235] The second constraint is that the total energy consumption of the target vehicle's roadside computing service and vehicle-side computing service is less than the vehicle-side remaining energy.

[0236] In one embodiment of this application, the solution module 3 described above can be specifically used to solve the objective function using the interior point method to obtain the task segmentation ratio; and to calculate the vehicle-side task volume and roadside task volume based on the total task volume and the task segmentation ratio.

[0237] In one embodiment of this application, the solution module 3 can be specifically used to solve the objective function using the simplex method to obtain the task segmentation ratio; and to calculate the vehicle-side task volume and roadside task volume based on the total task volume and the task segmentation ratio.

[0238] In one embodiment of this application, the deployment module 5 can be specifically used to package the roadside computing service to obtain a first container image; package the vehicle-side computing service to obtain a second container image; and deploy the first container image and the second container image to the target vehicle.

[0239] In one embodiment of this application, the task scheduling device may further include a computing power allocation module, used to construct an online decision-making network model using historical data of the target vehicle at various historical moments; the historical data includes the system state, decision actions, system benefits at the corresponding historical moment, and the system state at the next historical moment; the system state includes the current vehicle-side computing power, roadside computing service volume, and vehicle-side computing service volume at the corresponding historical moment; the decision actions include the computing power allocated to the roadside computing service and the computing power allocated to the vehicle-side computing service at the corresponding historical moment; obtaining the system state of the target vehicle at the current moment; processing the system state using the online decision-making network model to obtain the decision actions of the target vehicle at the current moment; and executing the decision actions using the target vehicle. A model update computing power allocation module is used to update the online decision-making network model at preset time intervals to obtain an updated online decision-making network model.

[0240] For a description of the apparatus provided in the embodiments of this application, please refer to the above method embodiments; further details will not be repeated here.

[0241] This application provides an electronic device.

[0242] Please refer to Figure 3 , Figure 3 This application provides a schematic diagram of the structure of an electronic device, which may include:

[0243] Memory, used to store computer programs;

[0244] A processor, used to execute computer programs, can implement the steps of any of the task scheduling methods described above.

[0245] like Figure 3 The diagram shows the structural composition of an electronic device, which may include a processor 10, a memory 11, a communication interface 12, and a communication bus 13. The processor 10, memory 11, and communication interface 12 all communicate with each other through the communication bus 13.

[0246] In this embodiment, the processor 10 may be a central processing unit (CPU), an application-specific integrated circuit, a digital signal processor, a field-programmable gate array, or other programmable logic devices.

[0247] The processor 10 can call programs stored in the memory 11. Specifically, the processor 10 can execute operations in the embodiments of the task scheduling method.

[0248] The memory 11 is used to store one or more programs. The programs may include program code, which includes computer operation instructions. In this embodiment, the memory 11 stores at least a program for implementing the following functions:

[0249] The target vehicle is monitored in real time using roadside equipment to obtain the first travel time of the target vehicle within the monitoring coverage area and the second travel time within the monitoring blind zone.

[0250] Construct an objective function for optimizing preset index parameters using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task.

[0251] Solve the objective function to obtain the vehicle-side task quantity and the roadside task quantity;

[0252] The roadside equipment is used to process the roadside tasks corresponding to the roadside workload and obtain the roadside processing results.

[0253] The roadside processing results are used as roadside computing services, and the vehicle-side tasks corresponding to the vehicle-side task volume are used as vehicle-side computing services. These are then deployed to the target vehicle to provide computing services when the target vehicle is driving within the monitoring blind zone.

[0254] In one possible implementation, the memory 11 may include a program storage area and a data storage area, wherein the program storage area may store the operating system and applications required for at least one function; and the data storage area may store data created during use.

[0255] In addition, memory 11 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device or other volatile solid-state storage device.

[0256] Communication interface 12 can be an interface for the communication module, used to connect with other devices or systems.

[0257] Of course, it should be noted that, Figure 3 The structure shown does not constitute a limitation on the electronic device in the embodiments of this application. In practical applications, the electronic device may include more than Figure 3 More or fewer components as shown, or combinations of certain components.

[0258] This application provides a computer-readable storage medium.

[0259] The computer-readable storage medium provided in this application embodiment stores a computer program, which, when executed by a processor, can implement the steps of any of the task scheduling methods described above.

[0260] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0261] For a description of the computer-readable storage medium provided in the embodiments of this application, please refer to the above method embodiments; further details will not be repeated here.

[0262] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0263] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0264] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0265] The task scheduling method, apparatus, electronic device, and computer-readable storage medium provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A task scheduling method, characterized in that, Applications in vehicle-road cooperative systems include: The target vehicle is monitored in real time using roadside equipment to obtain a first travel time within the monitoring coverage area and a second travel time within the monitoring blind zone. Within the monitoring coverage area, the first travel time is calculated using the target vehicle's current position, speed, and planned trajectory. Within the monitoring blind zone, the second travel time is calculated using the current position, speed, and planned trajectory. A target function for optimizing preset index parameters is constructed using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task load. The objective function is solved to obtain the vehicle-side task quantity and the roadside task quantity; wherein, the objective function is solved by the simplex method or the interior point method. The roadside equipment is used to process the roadside tasks corresponding to the roadside task volume to obtain the roadside processing results; The roadside processing results are used as roadside computing services, and the vehicle-side tasks corresponding to the vehicle-side task volume are used as vehicle-side computing services, which are deployed to the target vehicle to provide computing services when the target vehicle is driving within the monitoring blind zone. Under preset constraints, the objective function is constructed using the first travel time, the second travel time, the system parameters, and the total task volume. The preset constraints include a first constraint and a second constraint. The first constraint is that the sum of the time taken by the roadside equipment to process the roadside task and the time taken to deploy the roadside computing service and the vehicle-side computing service to the target vehicle is less than the first travel time. The second constraint is that the total energy consumption of the target vehicle running the roadside computing service and the vehicle-side computing service is less than the vehicle-side remaining energy included in the system parameters.

2. The task scheduling method according to claim 1, characterized in that, The roadside equipment is used to monitor the target vehicle in real time to obtain the target vehicle's current position, speed, and planned trajectory, including: The roadside equipment is used to communicate wirelessly with the target vehicle to obtain the target vehicle's current location, driving speed, and planned trajectory.

3. The task scheduling method according to claim 1, characterized in that, Within the monitoring coverage area, the first travel time is calculated using the target vehicle's current location, speed, and planned trajectory, including: Determine the intersection point between the planned trajectory and the monitoring coverage area to obtain the location of the first intersection point; Calculate the first distance between the current position and the first intersection point position; The first travel time is calculated based on the first distance and the driving speed.

4. The task scheduling method according to claim 1, characterized in that, Determine the next monitoring coverage area of ​​the next roadside device of the roadside device, and determine the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area.

5. The task scheduling method according to claim 4, characterized in that, Determining the next monitoring coverage area of ​​the next roadside device includes: Based on the planned trajectory, route prediction is performed to determine the next roadside device for the roadside equipment. Data communication is performed with the next-side device to obtain the next monitoring coverage area of ​​the next-side device.

6. The task scheduling method according to claim 4, characterized in that, Within the monitoring blind zone, the second travel time is calculated using the current location, the vehicle speed, and the planned trajectory, including: Determine the intersection point between the planned trajectory and the monitoring coverage area to obtain the location of the first intersection point; Determine the intersection point between the planned trajectory and the next monitoring coverage area to obtain the location of the second intersection point; Calculate the second distance between the first intersection point and the second intersection point; The second travel time is calculated based on the second distance and the vehicle speed.

7. The task scheduling method according to claim 4, characterized in that, Before determining the next monitoring coverage area of ​​the next roadside device of the roadside device, and determining the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area, the method further includes: Determine whether the next-side device is in working condition; If so, then the steps of determining the next monitoring coverage area of ​​the next roadside device of the roadside device and determining the monitoring blind zone range based on the monitoring coverage area and the next monitoring coverage area are executed; If not, then determine the start time of the next roadside device, and calculate the second travel time based on the start time and the first travel time.

8. The task scheduling method according to claim 7, characterized in that, The step of calculating the second driving time based on the start time and the first driving time includes: Determine the current time point; The departure time of the target vehicle from the monitoring coverage area is calculated based on the current time and the first driving time. The second travel time is calculated based on the start time and the departure time.

9. The task scheduling method according to claim 1, characterized in that, The system parameters include the current roadside computing power, the vehicle-side average power, and the roadside average power. The preset index parameter is the computing service accuracy gain of the target vehicle within the monitoring blind zone.

10. The task scheduling method according to claim 9, characterized in that, Solving the objective function to obtain the vehicle-side task quantity and the roadside task quantity includes: The objective function is solved using the interior point method to obtain the task segmentation ratio. The vehicle-side task volume and the roadside task volume are calculated based on the total task volume and the task segmentation ratio.

11. The task scheduling method according to claim 9, characterized in that, Solving the objective function to obtain the vehicle-side task quantity and the roadside task quantity includes: The objective function is solved using the simplex method to obtain the task segmentation ratio; The vehicle-side task volume and the roadside task volume are calculated based on the total task volume and the task segmentation ratio.

12. The task scheduling method according to claim 1, characterized in that, The step of deploying the roadside processing results as roadside computing services and the vehicle-side tasks corresponding to the vehicle-side task volume as vehicle-side computing services to the target vehicle includes: The roadside computing service is packaged to obtain a first container image; The vehicle-side computing service is packaged to obtain a second container image; Deploy the first container image and the second container image to the target vehicle.

13. The task scheduling method according to any one of claims 1 to 12, characterized in that, Also includes: An online decision-making network model is constructed using historical data of the target vehicle at various historical moments. The historical data includes the system state, decision actions, system benefits at the corresponding historical moment, and the system state at the next historical moment. The system state includes the current vehicle-side computing power, roadside computing service volume, and vehicle-side computing service volume at the corresponding historical moment. The decision actions include the computing power allocated by the target vehicle to the roadside computing service and the computing power allocated to the vehicle-side computing service at the corresponding historical moment. Obtain the system status of the target vehicle at the current moment; The system state is processed using the online decision network model to obtain the decision action of the target vehicle at the current moment; The decision-making action is performed using the target vehicle.

14. The task scheduling method according to claim 13, characterized in that, Also includes: The online decision network model is updated at preset time intervals to obtain the updated online decision network model.

15. A task scheduling device, characterized in that, Applications in vehicle-road cooperative systems include: The monitoring module is used to monitor a target vehicle in real time using roadside equipment, and to obtain the first travel time of the target vehicle within the monitoring coverage area and the second travel time within the monitoring blind zone. Within the monitoring coverage area, the first travel time is calculated using the target vehicle's current position, speed, and planned trajectory; within the monitoring blind zone, the second travel time is calculated using the current position, speed, and planned trajectory. The construction module is used to construct an objective function about the optimal preset index parameters using the first travel time, the second travel time, the system parameters of the vehicle-road cooperative system, and the total task volume; The solution module is used to solve the objective function to obtain the vehicle-side task quantity and the roadside task quantity; wherein, the objective function is solved by the simplex method or the interior point method; The processing module is used to process the roadside tasks corresponding to the roadside task volume using the roadside equipment, and obtain the roadside processing results. The deployment module is used to deploy the roadside processing results as roadside computing services and the vehicle-side tasks corresponding to the vehicle-side task volume as vehicle-side computing services to the target vehicle, so as to provide computing services for the target vehicle when it is driving within the monitoring blind zone. Under preset constraints, the objective function is constructed using the first travel time, the second travel time, the system parameters, and the total task volume. The preset constraints include a first constraint and a second constraint. The first constraint is that the sum of the time taken by the roadside equipment to process the roadside task and the time taken to deploy the roadside computing service and the vehicle-side computing service to the target vehicle is less than the first travel time. The second constraint is that the total energy consumption of the target vehicle running the roadside computing service and the vehicle-side computing service is less than the vehicle-side remaining energy included in the system parameters.

16. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the task scheduling method as described in any one of claims 1 to 14 when executing the computer program.

17. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the task scheduling method as described in any one of claims 1 to 14.

Images