Automatic driving task processing system, method and device, electronic equipment and storage medium

Abstract

The invention provides an automatic driving task processing system, relates to the field of computers, in particular to the field of automatic driving task processing, and can be applied to an automatic driving scene. The system comprises a cloud configured to analyze a target task for a vehicle to obtain a plurality of first atomic instructions; a plurality of first atomic instructions are sent to the vehicle end, the plurality of first atomic instructions are obtained from a preset instruction set, the instruction set comprises a plurality of atomic instructions used for indicating the vehicle to execute a plurality of preset actions, and a plurality of combinations of the plurality of atomic instructions are used for executing a plurality of tasks of the vehicle; and the vehicle end is configured to receive and sequentially execute the plurality of first atomic instructions, so that a vehicle can complete a target task by implementing a plurality of target predetermined actions corresponding to the plurality of first atomic instructions. The invention further provides an automatic driving task processing method, a vehicle, a device, electronic equipment, a storage medium and a program product.

Description

Technical Field

[0001] The present disclosure relates to the field of computers, and more particularly to the field of autonomous driving task processing, and is applicable to autonomous driving scenarios. More specifically, the present disclosure provides an autonomous driving task processing system, method, apparatus, electronic device, storage medium, and program product. Background Art

[0002] Vehicles are typically task-driven. For example, the cloud sends different types of autonomous driving tasks to the vehicle. The vehicle then completes the autonomous driving task according to a mapping table of task types and execution logic. Each task type corresponds to a separate processing flow. Consequently, maintaining existing task types or developing new ones typically requires comprehensive adaptation of both the cloud and vehicle sides. Summary of the Invention

[0003] The present disclosure provides an autonomous driving task processing system, method, device, electronic device, storage medium, and program product.

[0004] According to one aspect of the present disclosure, an autonomous driving task processing system is provided, which includes a cloud side, configured to parse a target task for a vehicle to obtain a plurality of first atomic instructions; sending the plurality of first atomic instructions to a vehicle side, wherein the plurality of first atomic instructions are obtained from a preset instruction set, the instruction set including a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and a plurality of combinations of the plurality of atomic instructions are used to perform a plurality of tasks of the vehicle; and a vehicle side, configured to receive and sequentially execute the plurality of first atomic instructions so that the vehicle completes the target task by implementing a plurality of target predetermined actions corresponding to the plurality of first atomic instructions.

[0005] According to another aspect of the present disclosure, a method for processing an autonomous driving task is provided, the method comprising: parsing a target task for a vehicle to obtain a plurality of first atomic instructions, wherein the plurality of first atomic instructions are obtained from a preset instruction set, the instruction set including a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and a plurality of combinations of the plurality of atomic instructions are used to perform a plurality of tasks of the vehicle; sending the plurality of first atomic instructions to a vehicle end, so that the vehicle end receives and sequentially executes the plurality of first atomic instructions, so that the vehicle completes the target task by implementing a plurality of target predetermined actions corresponding to the plurality of first atomic instructions.

[0006] According to another aspect of the present disclosure, a method for processing an autonomous driving task is provided, the method comprising: receiving a plurality of first atomic instructions sent by a cloud, wherein the plurality of first atomic instructions are obtained by the cloud from a preset instruction set by parsing a target task for a vehicle, the instruction set comprising a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and a plurality of combinations of the plurality of atomic instructions are used to perform a plurality of tasks of the vehicle; executing the plurality of first atomic instructions of the target task in sequence, so that the vehicle completes the target task by implementing a plurality of target predetermined actions corresponding to the plurality of first atomic instructions.

[0007] According to another aspect of the present disclosure, an autonomous driving task processing device is provided, which includes: a task parsing module, configured to parse a target task for a vehicle to obtain multiple first atomic instructions, wherein the multiple first atomic instructions are obtained from a preset instruction set, and the instruction set includes multiple atomic instructions for instructing the vehicle to perform multiple predetermined actions, and multiple combinations of the multiple atomic instructions are used to perform multiple tasks of the vehicle; an information sending module, configured to send the multiple first atomic instructions to the vehicle end, so that the vehicle end receives and executes the multiple first atomic instructions in sequence, so that the vehicle completes the target task by implementing multiple target predetermined actions corresponding to the multiple first atomic instructions.

[0008] According to another aspect of the present disclosure, an autonomous driving task processing device is provided, which includes: an information receiving module configured to receive a plurality of first atomic instructions sent by the cloud, wherein the plurality of first atomic instructions are obtained by the cloud from a preset instruction set by parsing a target task for a vehicle, and the instruction set includes a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and a plurality of combinations of the plurality of atomic instructions are used to perform a plurality of tasks of the vehicle; an instruction execution module configured to execute the plurality of first atomic instructions of the target task in sequence, so that the vehicle completes the target task by implementing a plurality of target predetermined actions corresponding to the plurality of first atomic instructions.

[0009] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method provided according to the present disclosure.

[0010] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided. The computer instructions are used to cause a computer to execute the method provided according to the present disclosure.

[0011] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, which implements the method provided according to the present disclosure when executed by a processor.

[0012] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0014] Figure 1 This is a schematic diagram of an application scenario of an autonomous driving task processing method and apparatus according to an embodiment of the present disclosure;

[0015] Figure 2 is an architectural diagram of an autonomous driving task processing system according to one embodiment of the present disclosure;

[0016] Figure 3 is a schematic diagram of updating a first atomic instruction according to an embodiment of the present disclosure;

[0017] Figure 4 is a schematic diagram of updating a first atomic instruction according to another embodiment of the present disclosure;

[0018] Figure 5 is a schematic diagram of a calling instruction code according to an embodiment of the present disclosure;

[0019] Figure 6 is a flowchart of a method for processing an autonomous driving task in the cloud according to an embodiment of the present disclosure;

[0020] Figure 7 is a flowchart of a method for processing an autonomous driving task on a vehicle side according to an embodiment of the present disclosure;

[0021] Figure 8 is a block diagram of an autonomous driving task processing device applicable to the cloud according to one embodiment of the present disclosure;

[0022] Figure 9 is a block diagram of an autonomous driving task processing device applicable to a vehicle according to an embodiment of the present disclosure; and

[0023] Figure 10 A schematic block diagram of an example electronic device that can be used to implement embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0024] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0025] In the technical solution of this application, the user information involved (including but not limited to user personal information, user image information, user information such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, application and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0026] Take the example of the cloud sending tasks to the vehicle side to control the vehicle. Generally, the cloud side sends complete tasks to the vehicle side according to the business dimension, such as passenger tasks, scheduling tasks, charging tasks, battery replacement tasks, return tasks and OTA (Over The Air technology) tasks. The vehicle side executes the corresponding processing flow according to the task type-execution logic mapping table, where each task type corresponds to an independent processing flow. This task type-driven architecture makes the business logic and task logic of the cloud side strongly coupled with the execution logic of the vehicle side. When maintaining the original task type or developing a new task type, it is necessary to simultaneously modify the task generation logic and the vehicle side execution logic to adapt the code based on the overall task type. For example, each time a new type is added, a vehicle side adaptation module needs to be developed, a large number of code lines or code types are added, and the cloud side and the vehicle side need to be tested synchronously, resulting in a long test cycle. In addition, when the cloud side business logic and task logic are updated, the vehicle side needs to be upgraded synchronously through OTA. Too high a frequency will also cause failures during the OTA process.

[0027] Understandably, different mission types may contain the same atomic functions, such as traveling from location A to location B, opening and closing a car door, and so on. Because the mission-type-driven architecture uses mission-type granularity for overall maintenance or development, some atomic functions are duplicated across mission types. This leads to poor vehicle function reusability, high development costs, long iteration cycles, and significant risks to system stability. Furthermore, this mission-type-driven architecture requires the vehicle to maintain a complex state machine, exponentially increasing the complexity of exception handling.

[0028] In view of this, the embodiments of the present disclosure provide an autonomous driving task processing system, which can decouple the business logic and task logic on the cloud from the execution logic on the vehicle side through a preset instruction set, in which multiple combinations of multiple atomic instructions are used to execute multiple tasks of the vehicle. The vehicle side can execute multiple first atomic instructions of the target task in sequence, so that the vehicle can complete the target task by implementing the corresponding multiple target predetermined actions. Therefore, compared with the method of using task type as the granularity, the vehicle can be maintained, developed and controlled at a smaller granularity such as atomic instructions, and the same atomic instruction can be reused in different task types to avoid repeated development, thereby reducing development costs, shortening test cycles and iteration cycles, and reducing stability risks and the complexity of exception handling.

[0029] The technical solutions provided by the present disclosure will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0030] Figure 1 2 is a schematic diagram of an application scenario of an autonomous driving task processing method and device according to an embodiment of the present disclosure.

[0031] It should be noted that Figure 1 The examples shown are merely examples of system architectures to which the embodiments of the present disclosure may be applied, to help those skilled in the art understand the technical content of the present disclosure, but do not mean that the embodiments of the present disclosure may not be used in other systems, environments or scenarios.

[0032] like Figure 1 As shown, the scene 100 of this embodiment includes a server 110, a vehicle 120 traveling on a road, an obstacle 130, a communication base station 140 and a road traffic network, wherein the road traffic network may include roads (e.g., roads 151, 152, 153), and an intersection 154 formed by the intersection of roads. Figure 1 The arrow in the figure indicates the direction of travel of the road. The cloud side may include a server 110 and the vehicle side may include a vehicle 120.

[0033] For example, in this scenario 100, vehicle 120 traveling on the road may be a vehicle. Obstacle 130 may be a pedestrian or another vehicle around the vehicle. For example, vehicle 120 may travel straight along road 151 and pass through intersection 154. Obstacle 130 may need to travel from road 153 to intersection 154 and turn left onto road 152.

[0034] For example, the server 110 can generate a passenger task in response to a passenger order, and the vehicle 120 can receive the passenger task sent by the server 110 through the communication base station 140, and execute multiple atomic instructions of the passenger task in sequence so that the vehicle 120 can complete the passenger task by achieving multiple corresponding target predetermined actions, for example, to enable the vehicle 120 to achieve target predetermined actions such as picking up, navigating, and sending off to complete the passenger task.

[0035] The vehicle 120 may upload data to the server 110 via the communication base station 140. The server 110 may, for example, obtain the data uploaded by the vehicle 120, such as the command execution status of each command.

[0036] It should be understood that Figure 1 The numbers and types of servers, roads, vehicles, obstacles, and communication base stations in the figure are merely illustrative. Any number and type of servers, roads, vehicles, obstacles, and communication base stations may be provided as required.

[0037] Figure 2 1 is an architectural diagram of an autonomous driving task processing system according to an embodiment of the present disclosure.

[0038] like Figure 2 As shown, the autonomous driving task processing system 200 may include a cloud side and a vehicle side, the cloud side includes a business layer 210 and a platform layer 220 , and the vehicle side includes a vehicle 230 .

[0039] In some embodiments, the cloud side is configured to parse the target task for the vehicle 230 to obtain multiple first atomic instructions; send the multiple first atomic instructions to the vehicle side, wherein the multiple first atomic instructions are obtained from a preset instruction set, which includes multiple atomic instructions for instructing the vehicle 230 to perform multiple predetermined actions, and multiple combinations of the multiple atomic instructions are used to perform multiple tasks of the vehicle 230; the vehicle side is configured to receive and execute the multiple first atomic instructions in sequence, so that the vehicle 230 completes the target task by implementing multiple target predetermined actions corresponding to the multiple first atomic instructions.

[0040] For example, the target task may include an autonomous driving task to be performed by vehicle 230. Each autonomous driving task can be abstracted into multiple atomic instructions. For example, the instruction set may include a pick-up instruction, a passenger verification instruction, a navigation instruction, a delivery confirmation instruction, a battery status determination instruction, a charging instruction, and a delivery instruction. Atomic instructions include atomic instructions derived from the atomic functions of vehicle 230 and can be considered the minimum execution unit of the task. The corresponding predetermined action can be the minimum execution action required to complete the task. It is understood that the content of the minimum execution unit and the minimum execution action can be adjusted according to actual needs.

[0041] The business layer 210 can, for example, handle autonomous driving operation business, which may include autonomous driving tasks of one or more task types such as passenger tasks, scheduling tasks, charging tasks, battery replacement tasks and return tasks.

[0042] The platform layer 220 may include an instruction set, a task parsing engine 221, and an instruction combination engine 222. The instruction set includes multiple atomic instructions abstracted from one or more task types, such as passenger tasks, dispatch tasks, charging tasks, battery swap tasks, and return tasks. The task parsing engine 221 parses the target tasks issued by the business layer 210 and sends the parsed results to the instruction combination engine 222, which retrieves multiple first atomic instructions from the instruction set and sequentially combines them to form the execution logic of the target task.

[0043] The vehicle 230 can accept multiple first atomic instructions issued by the platform layer 220 (for example, the combination results output by the instruction combination engine 222), and use the instruction parsing module 231 to parse multiple first atomic instructions, for example, verifying the legitimacy of the instructions through digital signatures and CRC checks. The instruction executor 232 can then call the instruction code maintained locally according to each first atomic instruction to implement the corresponding target predetermined action. The execution monitoring module 233 can monitor the instruction execution status of each first atomic instruction and return it to the cloud, for example, by monitoring the instruction execution status in the form of main transactions and sub-transactions. The vehicle end 230 can also feedback the driving environment and vehicle status to the cloud.

[0044] In some examples, for example, the target task is a passenger task, such as transporting a passenger from a first destination to a second destination. The task parsing engine 221 can be used to parse the task to obtain the pick-up subtask, passenger verification subtask, navigation subtask, and delivery confirmation subtask. The instruction combination engine 222 then retrieves the first atomic instructions corresponding to each of these four subtasks, such as the pick-up instruction, passenger verification instruction, navigation instruction, and delivery confirmation instruction. The instruction parsing module 221 then parses the pick-up instruction, passenger verification instruction, navigation instruction, and delivery confirmation instruction to verify their legitimacy. The instruction executor 232 then calls the instruction codes for the pick-up instruction, passenger verification instruction, navigation instruction, and delivery confirmation instruction, so that the vehicle 230 performs the corresponding target predetermined actions to complete the passenger task.

[0045] In other examples, for example, if the target task is a charging task, such as charging at a first charging station, the task parsing engine 221 can be used to parse the task to obtain the battery status determination subtask, the pause pick-up subtask, the navigation subtask, and the charging subtask. The instruction combination engine 222 then obtains the battery status determination instruction, the pause pick-up instruction, the navigation instruction, and the charging instruction, respectively. The instruction parsing module 221 then parses the obtain battery status determination instruction, the pause pick-up instruction, the navigation instruction, and the charging instruction to verify their legitimacy. The instruction executor 232 then calls the instruction codes for the obtain battery status determination instruction, the pause pick-up instruction, the navigation instruction, and the charging instruction, so that the vehicle 230 performs the corresponding target predetermined actions to complete the charging task.

[0046] In other examples, a combination of atomic instructions such as the delivery instruction, navigation instruction, and delivery confirmation instruction can control the vehicle to perform the delivery task, and the battery status judgment instruction, pause delivery instruction, navigation instruction, and parking instruction can control the vehicle to perform the fixed-point parking task.

[0047] It can be found that different target tasks may require the execution of navigation instructions, and the navigation instructions are obtained from a preset instruction set. Therefore, the navigation instructions can be reused in different types of tasks to avoid repeated development, thereby reducing development costs and shortening iteration cycles.

[0048] It should be noted that the use of the task parsing engine 221 to parse the target task in the cloud-based platform layer 220 and the use of the instruction combination engine 222 to derive multiple first atomic instructions from the instruction set is one embodiment of the autonomous driving task processing system 200, and the present disclosure is not limited thereto. For example, the target task may also be parsed on the vehicle side using a task-instruction mapping table, or the target task may be parsed on the vehicle side or in the cloud based on a pre-configured task template, and then multiple first atomic instructions may be derived from the instruction set.

[0049] According to the embodiments of the present disclosure, the autonomous driving task processing system can decouple the business logic and task logic on the cloud from the execution logic on the vehicle side. Compared with the task-based granularity approach, the vehicle can be maintained, developed and controlled at a smaller granularity such as atomic instructions. The same atomic instructions can be reused in different task types to avoid repeated development, thereby reducing development costs, shortening testing cycles and iteration cycles, and reducing stability risks and the complexity of exception handling.

[0050] Figure 3 is a schematic diagram of updating a first atomic instruction according to an embodiment of the present disclosure.

[0051] In some embodiments, as Figure 3As shown, vehicle-side 310 is configured to send execution information to cloud 320 during the execution of multiple first atomic instructions. The execution information is determined based on at least one of the vehicle's driving environment, the vehicle state, and the instruction execution state. Cloud 320 is configured to receive the execution information. In response to the execution information not meeting a preset condition, at least one of the multiple first atomic instructions is updated to ensure that the execution information meets the preset condition. The preset condition is pre-set based on the target task completion condition. In response to the execution information not meeting the preset condition, vehicle-side 310 continues execution.

[0052] Exemplarily, the driving environment includes external environmental information during vehicle driving, for example, it may include the route (including the starting position and the target position), road conditions, construction conditions, congestion conditions, weather conditions, etc.; the vehicle status is obtained based on the vehicle's own information during vehicle driving, and may include the vehicle's battery or fuel consumption, vehicle tire status or vehicle system status, etc.; the instruction execution status may include execution success or execution failure.

[0053] For example, the execution information obtained based on at least one of the driving environment, vehicle status, and instruction execution status includes at least one of road congestion, inability to drive due to road construction, insufficient vehicle battery, unstable tire pressure, and instruction execution failure. For example, the vehicle's target task is to send the driver, and the task of sending the passenger to the destination is completed. The preset condition is that each target predetermined action can be successfully executed to send the passenger to the destination. And the specific rules for successfully executing each target predetermined action. If the execution information obtained during the vehicle's execution of the target task is that the vehicle battery is insufficient to send the passenger to the destination, the preset condition is not met. Therefore, at least one atomic instruction can be updated to enable the vehicle to perform subsequent operations, such as charging, ending the send-driver task, etc.

[0054] For example, the target task is to transport a passenger from a first destination to a second destination. The first atomic instructions executed in sequence are the pick-up instruction, passenger verification instruction, navigation instruction, and delivery confirmation instruction. During the execution of the pick-up instruction and passenger verification instruction, if the execution information of the vehicle-side 310 meets the preset conditions, the vehicle-side 310 will continue to execute the subsequent instructions. During the execution of the navigation instruction, the execution information of the vehicle-side 310 indicates that the vehicle is out of power. At this time, the execution information does not meet the preset conditions, and the cloud 320 needs to update the navigation instruction, update the destination in the navigation instruction to a nearby charging station, and add a charging instruction, which is sent to the vehicle-side 310 so that the vehicle can charge at a nearby charging station according to the updated navigation information. If, during the execution of the navigation instruction, the execution information of the vehicle-side 310 indicates that the road ahead is congested, the cloud 320 will update the navigation route given by the current navigation instruction to obtain a new navigation instruction. The vehicle-side 310 executes the new navigation instruction to bypass the congested section. After executing the new navigation instruction, if the execution information meets the preset conditions, the delivery confirmation instruction will continue to be executed, thereby improving the passenger experience.

[0055] For example, based on the execution information, a decision tree is used to determine the update strategy for the target task. A decision tree can be pre-constructed for each target task, and the first atomic instruction corresponds to a branch node of the decision tree. The branch node is configured with an evaluation rule for evaluating whether the predetermined action corresponding to the current atomic instruction is successfully implemented. Multiple first atomic instructions correspond to a target path of the decision tree that can achieve the target task. During the calculation process of the decision tree, the feasibility of the target path is evaluated based on the execution information. For example, in response to the evaluation rule of the node corresponding to any first atomic instruction of the decision tree being failed (such as the vehicle is insufficiently charged and cannot reach the destination), the target path is updated, specifically by replacing the branch node (such as replacing the node corresponding to the charging instruction), thereby determining a new atomic instruction.

[0056] During a vehicle's mission, the entire mission is typically updated. For example, if the execution information doesn't meet the pre-set conditions, the cloud 320 reissues a new autonomous driving mission to the vehicle 310 to control the vehicle's continued completion. However, the mission execution logic is a fixed process pre-established by the vehicle 310, which lacks flexibility and often mismatches the execution information. New missions may require the vehicle to perform redundant actions, or the execution information for new missions may still not meet the pre-set conditions.

[0057] According to an embodiment of the present disclosure, if the execution information does not meet the preset conditions, one or more atomic instructions can be updated through the cloud 320. Compared with the method of updating the task as a whole, the update granularity is finer and the flexibility is higher, so that the vehicle can be quickly optimized in the process of completing the target task, thereby improving the success rate of completing the target task and enhancing the passenger experience.

[0058] In some embodiments, the vehicle side 310 is configured to send the instruction execution status corresponding to the currently executed first atomic instruction to the cloud side 320 based on the execution result of the vehicle's predetermined action for the current target; the cloud side 320 is configured to receive the instruction execution status; in response to the instruction execution status being a failure state, update at least one of the multiple first atomic instructions.

[0059] During the process of executing multiple first atomic instructions on the vehicle side 310, the execution status of each first atomic instruction is obtained. For example, if the first first atomic instruction is a door opening instruction, the target predetermined action corresponding to the instruction is door opening, and the corresponding execution result is door opening success or failure. If the execution result is door opening failure, the instruction execution status is execution failure. The cloud side 320 can further update at least one first atomic instruction in response to the door opening failure.

[0060] In the case of task-based granularity, the cloud 320 or vehicle 310 usually waits until the task completely fails or is successfully executed before obtaining the task execution status. During the task execution process, if the vehicle encounters navigation errors, remains stationary, or the battery is exhausted, it is difficult to handle it in a timely manner.

[0061] According to an embodiment of the present disclosure, the vehicle side 310 can individually feedback the instruction execution status for each atomic instruction, so that the cloud side 320 can be updated in time according to the latest instruction execution status, effectively improving the success rate of target task execution.

[0062] In some embodiments, the vehicle end 310 is configured to obtain the execution status of multiple sub-transactions corresponding to the multiple first atomic instructions during the execution of multiple first atomic instructions; determine the execution status of the main transaction corresponding to the target task based on the execution status of the multiple sub-transactions, and the execution status of the main transaction represents the completion progress of the target task.

[0063] For example, the vehicle-side 310 can generate multiple subtransactions for multiple first atomic instructions, record the identifiers of each subtransaction, the subtransaction execution order, and the corresponding open status in the main transaction table, and accordingly generate a main transaction for the target task and assign a main transaction identifier. The vehicle-side 310 can then execute specific atomic instructions based on the subtransaction execution order to cause the vehicle to perform the corresponding action. During this process, the execution status of each subtransaction can be obtained in a timely manner.

[0064] For example, in a vehicle fixed-point parking task, the first atomic instructions executed sequentially are the designated parking location navigation instruction and the parking instruction. The target task's main transaction corresponds to the fixed-point parking main transaction, the subtransaction corresponding to the designated parking location navigation instruction is the designated parking location navigation subtransaction, and the subtransaction corresponding to the parking instruction is the parking subtransaction. If the designated parking location navigation subtransaction executes successfully, but the parking subtransaction executes unsuccessfully, the fixed-point parking instruction will also execute unsuccessfully. Only when the atomic instructions corresponding to each subtransaction in the main transaction execute successfully will the fixed-point parking main transaction display a successful execution status, indicating that the fixed-point parking task is complete.

[0065] In some embodiments, updating at least one of the multiple first atomic instructions may include, for example, adding corresponding subtransactions for each new atomic instruction, and recording each subtransaction identifier, subtransaction execution order, and corresponding open status in a main transaction table to achieve binding with the main transaction to continue managing the execution progress of the target task. Corresponding subtransactions may also be canceled for each canceled atomic instruction.

[0066] For example, the cloud 320 can synchronize subtransactions and main transactions with the vehicle 310. The cloud 320 can generate multiple subtransactions for multiple first atomic instructions, record the identifiers of each subtransaction, the order in which it is executed, and the corresponding open status in the main transaction table, and then generate a main transaction for the target task and assign a main transaction identifier. The cloud 320 can then promptly obtain the execution status of each subtransaction from the vehicle 310 and synchronize the updates. This ensures consistency between the cloud 320 and the vehicle 310.

[0067] According to the embodiments of the present disclosure, through the combined use of main transactions and sub-transactions, the reliability of vehicle task execution can be improved through hierarchical management, the consistency of multiple first atomic instructions of the target task can be maintained during the execution process, errors in the execution logic of the vehicle end 310 can be avoided, and the success rate of the target task can be improved.

[0068] Figure 4 is a schematic diagram of updating a first atomic instruction according to another embodiment of the present disclosure.

[0069] In some embodiments, the cloud 420 is configured to send processing instructions for multiple first atomic instructions to the vehicle side 410 in response to the execution information not satisfying a preset condition based on the execution progress of the multiple first atomic instructions executed by the vehicle side 410; the vehicle side 410 is configured to receive the processing instructions; and perform at least one of the following operations based on the processing instructions: cancel at least one unexecuted first atomic instruction, roll back at least one target predetermined action that has been executed by the vehicle (for example, determined by the vehicle side 410 based on the executed atomic instructions).

[0070] In some embodiments, the cloud 420 is configured to obtain at least one target predetermined action to be implemented by the vehicle among multiple target predetermined actions based on the execution progress of multiple first atomic instructions executed by the vehicle side 410; determine at least one second atomic instruction from the instruction set based on at least one target predetermined action to be implemented by the vehicle; and send at least one second atomic instruction to the vehicle side 410 to update at least one of the multiple first atomic instructions so that the execution information meets the preset conditions.

[0071] For example, in response to execution information not meeting preset conditions, cloud-side 420 can obtain the latest execution status of each first atomic instruction to determine the execution progress. Vehicle-side 410 can cancel or roll back multiple first atomic instructions based on the processing instructions (e.g., the vehicle rolls back the corresponding actions) and obtain a processing result. If the processing result is a target result, at least one second atomic instruction is added so that the vehicle can implement at least one target predetermined action corresponding to the at least one second atomic instruction. Target results include successful cancellation and successful rollback results.

[0072] For example, if the target task is to park at a designated location, the first atomic commands to be executed sequentially are: a passenger disembarkation confirmation command, a door closing command, a parking guidance command, and a parking command. After the passenger disembarkation confirmation command is executed, the vehicle-side 410 executes the door closing command to close the vehicle's doors. If the door closing fails, the unexecuted parking guidance command and the parking command are canceled, and the door closing command is rolled back (for example, requiring the vehicle to execute the door closing action again). The rollback operation restores the vehicle to a predetermined state, such as a door-closed state, allowing the vehicle to continue executing the updated command.

[0073] Then, when the cancellation operation and / or rollback operation is successful, at least one second atomic instruction is added for execution. For example, the unexecuted parking navigation instruction and parking instruction are canceled. After the cancellation is successful, the second atomic instruction capable of driving to the designated location and parking and other target predetermined actions is added for execution, so that the vehicle can complete the task of parking at the designated location.

[0074] For example, if cloud 420 receives feedback from vehicle 410 indicating that the door closing command failed, the parking guidance command, and the parking command were not executed, and vehicle 410 sends a notification, the vehicle continues to execute the parking guidance command and the parking command without closing the door. In some examples, for example, after vehicle 410 successfully executes the parking guidance command and the parking command, it sends a notification to cloud 420. In this case, if vehicle 410 fails to cancel the parking guidance command and the parking command, which is not the target result, cloud 420 will not determine the second atomic instruction for vehicle 410 to execute, and may only instruct vehicle 410 to roll back the door closing command. In some examples, if vehicle 410 sends a notification to the cloud after failing to execute the parking guidance command, it may be because the vehicle has traveled to a new location with new execution information. Cloud 420 then re-determines at least one target predetermined action to be executed, determines at least one second atomic instruction, and sends a new processing instruction based on the latest execution progress. These cancellation, rollback, and addition operations can be implemented at the sub-transaction level.

[0075] For example, the target task is to deliver a passenger to the first destination. Specifically, the first atomic instructions executed in sequence are the pick-up instruction, the passenger code scanning verification instruction, the first navigation instruction, and the delivery confirmation instruction. During the vehicle's execution of the pick-up instruction, the execution information meets the preset conditions. However, during the vehicle's execution of subsequent instructions, the execution information does not meet the preset conditions. In other words, the execution progress is that only the first first atomic instruction is completed, and the subsequent first atomic instructions are not executed. The second atomic instruction is obtained from the instruction set based on the target predetermined action corresponding to the unexecuted first atomic instruction, so that the vehicle executes the second atomic instruction to complete the task.

[0076] For example, during the execution of the passenger code scanning verification instruction, if the passenger code scanning verification fails, the execution status of the instruction is failure. For example, the second atomic instruction selected by the cloud 420 is the passenger mobile phone number verification instruction, and the passenger mobile phone number verification instruction is updated to the vehicle end 410 for additional execution. When the passenger mobile phone number verification is successful, the vehicle end 410 executes the next atomic instruction (i.e., the first navigation instruction).

[0077] Then, during the execution of the first navigation instruction, the vehicle cannot continue driving due to road construction. The instruction fails, and the execution information includes the road not meeting the driving conditions and the failure of the first navigation instruction. The target action corresponding to the first navigation instruction is driving to the first destination. Cloud 420 selects a second atomic instruction from the instruction set that can complete the target action of driving to the first destination. For example, the selected second atomic instruction is a second navigation instruction (i.e., an instruction of the same type is selected, but the route and destination have been updated). This instruction has a different driving route than the first navigation instruction, bypassing the construction section. The second navigation instruction is then updated to vehicle-side 410 for execution. At this point, the execution information meets the preset conditions. Vehicle-side 410 then executes the second navigation instruction to reach the first destination, and then executes the delivery confirmation instruction. The vehicle door opens, the delivery confirmation instruction is executed, and the vehicle door closes, completing the task of delivering the passenger to the first destination.

[0078] According to the embodiments of the present disclosure, the execution progress and consistency before and after the update can be considered, and the vehicle can be provided with the conditions for executing the updated instructions through cancellation or rollback operations. Furthermore, based on the flexibility of the instructions, combined with the execution progress and execution information, the success rate and efficiency of the vehicle completing the target task can be improved through instruction updates (such as recombining instructions).

[0079] In some embodiments, the cloud side 420 is configured to obtain multiple priorities of multiple target tasks when the vehicle performs multiple target tasks. The priorities are obtained based on the preset weights of the multiple first atomic instructions corresponding to the target tasks. The preset weights represent the importance of the corresponding first atomic instructions; the vehicle side 410 is configured to determine the completion order of the multiple target tasks based on the multiple priorities.

[0080] For example, the vehicle performs target tasks 1 and 2. Target task 1 is completed by executing a first set of multiple first atomic instructions on the vehicle side 410, and target task 2 is completed by executing a second set of multiple first atomic instructions on the vehicle side 410. Each first atomic instruction in the first set of multiple first atomic instructions has a corresponding preset weight, and the priority of the first set of multiple first atomic instructions is calculated based on the preset weight of the first atomic instruction in the first set of multiple first atomic instructions. Each first atomic instruction in the second set of multiple first atomic instructions also has a corresponding preset weight, and the priority of the second set of multiple first atomic instructions is calculated based on the preset weight of the first atomic instruction in the second set of multiple first atomic instructions. The priorities of the first set of multiple first atomic instructions are compared with the priorities of the second set of multiple first atomic instructions, and the target task with the higher priority is placed in front, and the target task with the higher priority is completed first; for example, the vehicle performs the two tasks of charging and sending a driver. The priority of the charging task is higher than the priority of the sending a driver task, so the vehicle should perform the charging task first and then the sending a driver task.

[0081] Furthermore, for situations where urgent tasks need to be handled with priority, task queue jumping can be achieved by adjusting the weights of atomic instructions; for example, the vehicle's target task is to send passenger 1 to the first destination and passenger 2 to the second destination respectively, but during the execution of the task, the vehicle needs to be charged. At this time, the charging task is an urgent task. The cloud 420 can adjust the weights of each atomic instruction corresponding to the charging task to make the charging task have the highest priority, thereby giving priority to the charging task.

[0082] According to the embodiments of the present disclosure, task priorities can be calculated by calculating instruction weights, effectively resolving multi-task conflicts and increasing the flexibility of multi-autonomous driving task processing.

[0083] In some embodiments, priorities can also be assigned based on task types. When processing multiple autonomous driving tasks, priorities can be determined by type to resolve multi-task conflicts.

[0084] In some embodiments, when a vehicle executes multiple target tasks, at least one first atomic instruction of the multiple target tasks may be updated in a linked manner. For example, if at least two target tasks contain the same first atomic instruction, and when executing one of the target tasks, the cloud 420 determines that the current execution information does not meet the execution conditions, such as if the same first atomic instruction fails to execute, the cloud 420 may determine a second atomic instruction and synchronously update the at least two target tasks. Furthermore, the cloud 420 may determine whether the first atomic instruction to be executed subsequently for each target task is executable, and if it fails, the cloud 420 may continue to update the second atomic instruction.

[0085] Figure 5 2 is a schematic diagram of a calling instruction code according to an embodiment of the present disclosure.

[0086] In some embodiments, the vehicle end 510 is configured to call multiple instruction codes of multiple first atomic instructions in sequence. The instruction codes are obtained by the vehicle end 510 from the cloud 520 through over-the-air download technology. The instruction codes are used to instruct the vehicle to achieve corresponding target predetermined actions.

[0087] For example, the cloud 520 may pre-store instruction codes for each atomic instruction in the instruction set. The instruction codes may be pre-written and stored by the user. The vehicle 510 may download the instruction code files for each atomic instruction in the instruction set via OTA and store them locally.

[0088] According to an embodiment of the present disclosure, the vehicle-side 510 can call multiple instruction codes of multiple first atomic instructions based on the target task information issued by the cloud 520, and can perform maintenance, development, and control of the vehicle at a smaller granularity of atomic instructions, thereby decoupling the business logic and task logic of the cloud 520 from the execution logic of the vehicle-side 510. In addition, when updating atomic instructions, the instruction code of the atomic instructions can be updated. When adding a new task type, the original atomic instructions and instruction code can be reused, and the instruction code can be developed for the newly added atomic instructions. The vehicle-side 510 downloads the instruction code of one or more atomic instructions via OTA, not the entire task.

[0089] In some embodiments, the cloud 520 is configured to parse the target task in response to the existence of multiple versions of instruction sets to obtain multiple first atomic instructions from the target instruction set, and the version of the target instruction set is adapted to the instruction code version in the vehicle side 510; and send the multiple first atomic instructions to the vehicle side 510.

[0090] Reference Figure 5 With continuous iterations, the cloud 520 may store multiple versions of instruction sets, such as instruction set v1, instruction set v2, etc., where different versions may include at least one different atomic instruction. For example, instruction set v1 and instruction set v2 have different instruction codes for atomic instructions corresponding to the same predetermined action. For example, instruction set v2 adds new atomic instructions compared to instruction set v1.

[0091] In some embodiments, at least one of the following operations may be performed: A / B (control) testing, simulation testing.

[0092] For example, for A / B testing, different business strategies can be compared and different execution logic can be configured for the same type of tasks based on the same version of the instruction set and code set, such as different first atomic instruction combinations, which can be sent from the cloud 520 to the vehicle side 510 for execution and comparison of the execution results. It is also possible to configure the same execution logic for the same type of tasks based on different versions of the instruction set and code set, but with multiple first atomic instructions of different versions (at least one of which is different) which can be sent from the cloud 520 to the vehicle side 510 for execution and comparison of the execution results. It is also possible to send from the cloud 520 to each vehicle side 510 for execution and comparison of the execution results for different vehicles.

[0093] For example, for simulation testing, real road conditions, including road signs, obstacles, etc., are simulated in the cloud 520, and a digital twin model of vehicle capabilities can be established based on the vehicle side 510 to simulate and verify different vehicles and different versions of instruction sets and code sets.

[0094] Compared with the task type driven architecture that strongly couples the business logic and task logic of the cloud 520 with the execution logic of the vehicle 510, in some embodiments, a new version of the instruction set and the corresponding code set can be pushed to different vehicles, such as Figure 5 If the vehicle-side 510 has not yet OTA downloaded the instruction set v2, it can push information about the adapted target task based on the version downloaded locally by each vehicle via OTA, such as multiple first atomic instructions of the adapted version; and when the business logic or task logic is changed on the cloud side 520, for example, the combination of the first atomic instructions is changed for the same type of task, the vehicle-side 510 can directly call the corresponding instruction code in the local code set, which can also enable the vehicle to complete the target task, thus avoiding the development of the vehicle-side 510 adaptation module, adding a large number of code lines or code types, and synchronously testing the cloud side 520 and the vehicle-side 510, thus avoiding duplicate development work to a certain extent.

[0095] It should be noted that "the target instruction set version is compatible with the instruction code version in the vehicle 510" can include version consistency or inconsistency. For example, in the case of consistency, instruction set v1 is compatible with the code of code set v1, and instruction set v2 is compatible with the code of code set v2. In the case of inconsistency, the lower version of the code set can be compatible with the instruction set of different versions, and code set v1 is compatible with instruction set v1 and instruction set v2. That is, when the cloud 520 updates the instruction set version, the vehicle 510 does not need to download the new version of the code set via OTA.

[0096] In some embodiments, communication between the cloud 520 and the vehicle 510 can be performed using a resumable transmission protocol. Network signal fluctuations may occur due to vehicle movement. When the network is poor (e.g., weak signal, slow speed, or disconnection), data (e.g., execution information, execution progress, processing instructions, or at least one second atomic instruction to be updated) can be packetized and the context of the currently executing target task can be saved for a certain period of time. Communication can then resume using a resumable transmission protocol after the network is restored. This prevents data loss and reduces duplicate data transmission. Saving contextual information (e.g., progress and status) during task execution ensures the continuity of task logic.

[0097] In some embodiments, the instruction set and code set may include instructions and codes officially provided by the vehicle manufacturer, and may also be open to third-party developers in compliance with relevant regulations (such as authority verification), allowing third-party developers to provide atomic instructions, instruction codes, or combinations of atomic instructions, thereby enabling flexible configuration of various new services and supporting complex needs, complex scenarios, and multi-scenario integration.

[0098] Figure 6 This is a flowchart of a method for processing autonomous driving tasks in the cloud according to an embodiment of the present disclosure.

[0099] Reference Figure 6 The autonomous driving task processing method 600 may include operations S610 to S620.

[0100] In operation S610, a plurality of first atomic instructions are determined by parsing the target task for the vehicle, wherein the plurality of first atomic instructions are obtained from a preset instruction set, the instruction set including a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and a plurality of combinations of the plurality of atomic instructions are used to perform a plurality of tasks of the vehicle.

[0101] In operation S620, multiple first atomic instructions are sent to the vehicle side, so that the vehicle side receives and executes the multiple first atomic instructions in sequence, so that the vehicle completes the target task by implementing multiple target predetermined actions corresponding to the multiple first atomic instructions.

[0102] In some embodiments, the cloud can receive execution information sent by the vehicle, based on at least one of the vehicle's driving environment, vehicle status, and instruction execution status. In response to the execution information not meeting a preset condition, the cloud updates at least one of the multiple first atomic instructions to ensure that the execution information meets the preset condition, which is pre-set based on the target task completion conditions. Compared to updating tasks as a whole, this approach provides finer granularity and greater flexibility, enabling rapid optimization of the vehicle during task completion, improving the success rate of completing the target task and enhancing the passenger experience.

[0103] It should be noted that the part of the autonomous driving task processing method applicable to the cloud can also include the various steps executed on the cloud in any one of the autonomous driving task processing system embodiments described above, and the implementation methods, technical problems solved, functions implemented, and technical effects achieved of each step are the same or similar to the corresponding implementation methods, technical problems solved, functions implemented, and technical effects achieved in the cloud in the autonomous driving task processing system embodiment, and will not be repeated here.

[0104] Figure 7 This is a flowchart of a method for processing autonomous driving tasks on a vehicle side according to an embodiment of the present disclosure.

[0105] Reference Figure 7 The autonomous driving task processing method 700 may include operations S710 to S720.

[0106] In operation S710, multiple first atomic instructions sent by the cloud are received, wherein the multiple first atomic instructions are obtained by the cloud from a preset instruction set by parsing the target tasks for the vehicle, and the instruction set includes multiple atomic instructions for instructing the vehicle to perform multiple predetermined actions, and multiple combinations of the multiple atomic instructions are used to perform multiple tasks of the vehicle.

[0107] In operation S720 , a plurality of first atomic instructions of the target task are sequentially executed so that the vehicle completes the target task by implementing a plurality of target predetermined actions corresponding to the plurality of first atomic instructions.

[0108] In some embodiments, the vehicle side can send execution information to the cloud during the execution of multiple first atomic instructions, and the execution information is obtained based on at least one of the vehicle's driving environment, vehicle status, and instruction execution status; in response to receiving the processing instruction sent by the cloud, perform at least one of the following operations: cancel at least one unexecuted first atomic instruction, control the vehicle to roll back at least one executed first atomic instruction; in response to receiving at least one second atomic instruction sent by the cloud, update at least one of the multiple first atomic instructions so that the execution information meets the preset conditions, and the preset conditions are pre-set according to the target task completion conditions. It is possible to consider the execution progress and the consistency before and after the update, and provide the conditions for the vehicle to execute the second atomic instruction through cancellation or rollback operations. It is also possible to improve the success rate and efficiency of the vehicle in completing the target task by updating the instructions (such as the re-combination of instructions) based on the flexibility of the instructions, combined with the execution progress and execution information.

[0109] In some embodiments, the vehicle-side can obtain the execution status of multiple sub-transactions corresponding to the multiple first atomic instructions during the execution of the multiple first atomic instructions; based on the execution status of the multiple sub-transactions, the execution status of the main transaction corresponding to the target task is determined, and the execution status of the main transaction represents the progress of the target task. By combining the main transaction and the sub-transactions, the vehicle-side can separately feedback the instruction execution status for each atomic instruction, allowing the cloud to be updated in a timely manner based on the latest instruction execution status, effectively improving the success rate of the target task execution.

[0110] It should be noted that the part of the autonomous driving task processing method applicable to the vehicle side may also include the various steps executed by the vehicle side in any of the autonomous driving task processing system embodiments described above, and the implementation methods, technical problems solved, functions realized, and technical effects achieved of each step are respectively the same or similar to the corresponding implementation methods, technical problems solved, functions realized, and technical effects achieved by the vehicle side in the autonomous driving task processing system embodiment, and will not be repeated here.

[0111] Figure 8 1 is a block diagram of an autonomous driving task processing device suitable for the cloud according to an embodiment of the present disclosure.

[0112] like Figure 8 As shown, the autonomous driving task processing device 800 may include a task parsing module 810 and an information sending module 820.

[0113] The task parsing module 810 can execute operation S610, which is configured to parse the target task for the vehicle to obtain multiple first atomic instructions, wherein the multiple first atomic instructions are obtained from a preset instruction set, and the instruction set includes multiple atomic instructions for instructing the vehicle to perform multiple predetermined actions, and multiple combinations of the multiple atomic instructions are used to execute multiple tasks of the vehicle.

[0114] The information sending module 820 can perform operation S620 and is configured to send multiple first atomic instructions to the vehicle end, so that the vehicle end receives and executes the multiple first atomic instructions in sequence, so that the vehicle completes the target task by implementing multiple target predetermined actions corresponding to the multiple first atomic instructions.

[0115] In some embodiments, the autonomous driving task processing device 800 may further include: an execution information receiving module, configured to receive execution information sent by the vehicle side, the execution information being obtained based on at least one of the vehicle's driving environment, vehicle status, and instruction execution status; a cloud update module, configured to update at least one of the multiple first atomic instructions in response to the execution information not meeting the preset conditions, so that the execution information meets the preset conditions, and the preset conditions are pre-set according to the target task completion conditions.

[0116] Any portions not mentioned in the autonomous driving task processing apparatus 800 can be understood by referring to the various embodiments of the aforementioned cloud-based autonomous driving task processing method. Specifically, the apparatus portion includes modules for executing each step of any of the aforementioned cloud-based autonomous driving task processing method embodiments. Furthermore, the implementation methods, technical problems solved, functions achieved, and technical effects achieved of each module / unit / subunit in the apparatus portion embodiments are identical or similar to the implementation methods, technical problems solved, functions achieved, and technical effects achieved of the corresponding steps in the cloud-based autonomous driving task processing method embodiments, and are not further described here.

[0117] Figure 9 This is a block diagram of an autonomous driving task processing device applicable to a vehicle according to an embodiment of the present disclosure.

[0118] like Figure 9 As shown, the autonomous driving task processing device 900 may include an information receiving module 910 and an instruction execution module 920.

[0119] The information receiving module 910 can execute operation S710 and is configured to receive multiple first atomic instructions sent by the cloud, wherein the multiple first atomic instructions are obtained by the cloud from a preset instruction set by parsing the target tasks for the vehicle, and the instruction set includes multiple atomic instructions for instructing the vehicle to perform multiple predetermined actions, and multiple combinations of the multiple atomic instructions are used to execute multiple tasks of the vehicle.

[0120] The instruction execution module 920 may perform operation S720 and be configured to sequentially execute multiple first atomic instructions of the target task so that the vehicle completes the target task by implementing multiple target predetermined actions corresponding to the multiple first atomic instructions.

[0121] In some embodiments, the autonomous driving task processing device 900 may include: an execution information sending module, configured to send execution information to the cloud during the execution of multiple first atomic instructions, the execution information being obtained based on at least one of the vehicle's driving environment, vehicle status, and instruction execution status; a processing module, configured to perform at least one of the following operations in response to receiving a processing instruction sent from the cloud: canceling at least one unexecuted first atomic instruction, controlling the vehicle to roll back at least one executed first atomic instruction; a vehicle-side update module, configured to update at least one of the multiple first atomic instructions in response to receiving at least one second atomic instruction sent from the cloud, so that the execution information meets preset conditions, and the preset conditions are pre-set according to the target task completion conditions.

[0122] In some embodiments, the autonomous driving task processing device 900 may include: a sub-transaction module, configured to obtain the execution status of multiple sub-transactions corresponding to multiple first atomic instructions during the execution of multiple first atomic instructions; a main transaction module, configured to determine the execution status of the main transaction corresponding to the target task based on the execution status of the multiple sub-transactions, and the execution status of the main transaction represents the completion progress of the target task.

[0123] For portions not mentioned in the autonomous driving task processing apparatus 900, reference can be made to the various embodiments of the vehicle-side autonomous driving task processing method described above. That is, the apparatus portion includes modules for executing each step of any of the vehicle-side autonomous driving task processing method embodiments described above. Furthermore, the implementation methods, technical problems solved, functions achieved, and technical effects achieved of each module / unit / subunit, etc. in the apparatus portion embodiments are the same or similar to the implementation methods, technical problems solved, functions achieved, and technical effects achieved of each corresponding step in the vehicle-side autonomous driving task processing method embodiment, and are not further described here.

[0124] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0125] According to an embodiment of the present disclosure, an electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above method.

[0126] According to an embodiment of the present disclosure, a non-transitory computer-readable storage medium stores computer instructions, wherein the computer instructions are used to enable a computer to execute the above method.

[0127] According to an embodiment of the present disclosure, a computer program product includes a computer program, and the computer program implements the above method when executed by a processor.

[0128] According to an embodiment of the present disclosure, an autonomous driving vehicle may also be provided, which may utilize the autonomous driving task processing system described above to complete a target task; and / or, may execute an autonomous driving task processing method applicable to the vehicle side; and / or, may include an autonomous driving task processing device applicable to the vehicle side.

[0129] Figure 10A schematic block diagram of an example electronic device 1000 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearables, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0130] like Figure 10 As shown, electronic device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. Various programs and data required for the operation of 1000 may also be stored in RAM 1003. Computing unit 1001, ROM 1002, and RAM 1003 are interconnected via a bus 1004. An input / output (I / O) interface 1005 is also connected to bus 1004.

[0131] Multiple components in the electronic device 1000 are connected to the I / O interface 1005, including an input unit 1006, such as a keyboard, a mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a magnetic disk, an optical disk, etc.; and a communication unit 1009, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1009 allows the electronic device 1000 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0132] Computing unit 1001 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Computing unit 1001 performs the various methods and processes described above, such as the autonomous driving task processing method. For example, in some embodiments, the autonomous driving task processing method may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed onto 1000 via ROM 1002 and / or communication unit 1009. When the computer program is loaded into RAM 1003 and executed by computing unit 1001, one or more steps of the autonomous driving task processing method described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to execute the autonomous driving task processing method in any other appropriate manner (for example, by means of firmware).

[0133] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-a-chip systems (SOCs), complex programmable logic (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0134] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable autonomous driving task processing device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0135] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or the like. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or the like, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

[0136] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0137] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0138] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0139] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0140] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. An autonomous driving task processing system, comprising: The cloud is configured to parse a target task for the vehicle to obtain a plurality of first atomic instructions; Sending the plurality of first atomic instructions to the vehicle end, wherein the plurality of first atomic instructions are obtained from a preset instruction set, the instruction set including a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and multiple combinations of the plurality of atomic instructions are used to perform a plurality of tasks of the vehicle; The vehicle side is configured to receive and sequentially execute the multiple first atomic instructions so that the vehicle completes the target task by implementing multiple target predetermined actions corresponding to the multiple first atomic instructions.

2. The system according to claim 1, wherein: The vehicle side is configured to send execution information to the cloud during the execution of the plurality of first atomic instructions, wherein the execution information is obtained based on at least one of the driving environment of the vehicle, the vehicle state, and the instruction execution state; The cloud is configured to receive the execution information; in response to the execution information not meeting a preset condition, update at least one of the multiple first atomic instructions so that the execution information meets the preset condition, and the preset condition is pre-set according to the target task completion condition.

3. The system according to claim 2, wherein: The cloud is configured to send, in response to the execution information not satisfying a preset condition, processing instructions for the plurality of first atomic instructions to the vehicle side based on the execution progress of the plurality of first atomic instructions by the vehicle side; The vehicle end is configured to receive the processing instruction; At least one of the following operations is performed based on the processing instruction: canceling at least one unexecuted first atomic instruction, and controlling the vehicle to roll back at least one executed target predetermined action.

4. The system according to claim 2 or 3, wherein: The cloud configuration is: Obtaining, based on the execution progress of the plurality of first atomic instructions by the vehicle side, at least one target predetermined action to be implemented by the vehicle among the plurality of target predetermined actions; determining at least one second atomic instruction from the instruction set based on at least one target predetermined action to be achieved by the vehicle; The at least one second atomic instruction is sent to the vehicle end to update at least one of the multiple first atomic instructions so that the execution information meets the preset condition.

5. The system according to claim 1 or 2, wherein: The vehicle side is configured to send an instruction execution status corresponding to the currently executed first atomic instruction to the cloud based on the execution result of the vehicle's predetermined action for the current target; The cloud is configured to receive the instruction execution status; in response to the instruction execution status being a failure state, update at least one of the multiple first atomic instructions.

6. The system according to claim 1, wherein: The cloud is configured to obtain, when the vehicle executes the plurality of target tasks, a plurality of priorities of the plurality of target tasks, the priorities being obtained according to respective preset weights of the plurality of first atomic instructions corresponding to the target tasks, the preset weights representing the importance of the corresponding first atomic instructions; The vehicle side is configured to determine the completion order of the multiple target tasks based on the multiple priorities.

7. The system according to claim 1, wherein: The vehicle-side configuration is: During the execution of the plurality of first atomic instructions, obtaining execution statuses of a plurality of sub-transactions corresponding to the plurality of first atomic instructions; An execution status of a main transaction corresponding to the target task is determined based on the execution statuses of the multiple sub-transactions, where the execution status of the main transaction represents the completion progress of the target task.

8. The autonomous driving task processing system according to claim 1, wherein: The vehicle-side is configured to sequentially execute a plurality of first atomic instructions of the target task, including: The vehicle side is configured to call multiple code instructions of the multiple first atomic instructions in sequence. The code instructions are obtained by the vehicle side from the cloud through over-the-air download technology. The code instructions are used to instruct the vehicle to achieve corresponding target predetermined actions.

9. The system according to claim 8, wherein: The cloud-side configuration for parsing the target task for the vehicle to obtain a plurality of first atomic instructions includes: The cloud side is configured to parse the target task to obtain the multiple first atomic instructions from the target instruction set in response to the existence of multiple versions of the instruction set, and the version of the target instruction set is adapted to the code instruction version in the vehicle side.

10. A method for processing an autonomous driving task, comprising: Parsing a target task for the vehicle to obtain a plurality of first atomic instructions, wherein the plurality of first atomic instructions are obtained from a preset instruction set, the instruction set including a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and a plurality of combinations of the plurality of atomic instructions for performing a plurality of tasks for the vehicle; The multiple first atomic instructions are sent to the vehicle end, so that the vehicle end receives and executes the multiple first atomic instructions in sequence, so that the vehicle completes the target task by implementing multiple target predetermined actions corresponding to the multiple first atomic instructions.

11. The method according to claim 10, wherein: Also includes: receiving execution information sent by the vehicle end, wherein the execution information is obtained based on at least one of a driving environment of the vehicle, a vehicle state, and a command execution state; In response to the execution information not satisfying a preset condition, at least one of the plurality of first atomic instructions is updated so that the execution information satisfies the preset condition, where the preset condition is pre-set according to a target task completion condition.

12. A method for processing an autonomous driving task, comprising: Receiving a plurality of first atomic instructions sent by a cloud, wherein the plurality of first atomic instructions are obtained by the cloud from a preset instruction set by parsing target tasks for the vehicle, the instruction set including a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and multiple combinations of the plurality of atomic instructions are used to perform a plurality of tasks for the vehicle; The plurality of first atomic instructions of the target task are executed in sequence, so that the vehicle completes the target task by implementing a plurality of target predetermined actions corresponding to the plurality of first atomic instructions.

13. The method according to claim 12, wherein: Also includes: During the execution of the plurality of first atomic instructions, execution information is sent to the cloud, where the execution information is obtained based on at least one of a driving environment of the vehicle, a vehicle state, and an instruction execution state; In response to receiving the processing instruction sent by the cloud, performing at least one of the following operations: canceling at least one unexecuted first atomic instruction, controlling the vehicle to roll back at least one executed target predetermined action; In response to receiving at least one second atomic instruction sent by the cloud, at least one of the multiple first atomic instructions is updated so that the execution information meets a preset condition, and the preset condition is pre-set according to the target task completion condition.

14. The method according to claim 12, wherein: Also includes: During the execution of the plurality of first atomic instructions, obtaining execution statuses of a plurality of sub-transactions corresponding to the plurality of first atomic instructions; An execution status of a main transaction corresponding to the target task is determined based on the execution statuses of the multiple sub-transactions, where the execution status of the main transaction represents the completion progress of the target task.

15. An autonomous driving task processing device, comprising: a task parsing module configured to parse a target task for the vehicle to obtain a plurality of first atomic instructions, wherein the plurality of first atomic instructions are obtained from a preset instruction set, the instruction set including a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and a plurality of combinations of the plurality of atomic instructions for performing a plurality of tasks for the vehicle; The information sending module is configured to send the multiple first atomic instructions to the vehicle end, so that the vehicle end receives and executes the multiple first atomic instructions in sequence, so that the vehicle completes the target task by implementing multiple target predetermined actions corresponding to the multiple first atomic instructions.

16. An autonomous driving task processing device, comprising: an information receiving module configured to receive a plurality of first atomic instructions sent by a cloud, wherein the plurality of first atomic instructions are obtained by the cloud from a preset instruction set by parsing target tasks for the vehicle, the instruction set including a plurality of atomic instructions for instructing the vehicle to perform a plurality of predetermined actions, and multiple combinations of the plurality of atomic instructions are used to perform a plurality of tasks for the vehicle; The instruction execution module is configured to execute multiple first atomic instructions of the target task in sequence, so that the vehicle completes the target task by implementing multiple target predetermined actions corresponding to the multiple first atomic instructions.

17. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 10 to 11 or claims 12 to 14.

18. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 10 to 11 or claims 12 to 14.

19. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method according to any one of claims 10 to 11 or claims 12 to 14.

Images