Vehicle-mounted function unit calculation method and device, computer device, and storage medium

Abstract

The application relates to a vehicle-mounted function unit computing method, a device, computer equipment and a storage medium. The method comprises the following steps: acquiring the unit type of a current vehicle-mounted function unit; in response to the unit type being a fast / slow road selectable function unit, determining a scene complexity score and a scene timeliness score according to each scene evaluation index, and determining the current fast road execution time and the current slow road execution time of the current frame of the current vehicle-mounted function unit according to the system load, the historical actual execution time of each historical frame and the scene complexity score, and acquiring a current residual time delay budget value; in response to the current slow road execution time being greater than the current residual time delay budget value, acquiring the historical computing output result of the previous vehicle-mounted function unit of the current vehicle-mounted function unit, and controlling the current vehicle-mounted function unit to perform computing processing according to the coarse-grained computing mode and the historical computing output result to obtain the current computing output result of the current vehicle-mounted function unit. The method can avoid timeout frame loss.

Description

Technical Field

[0001] This application relates to the field of data technology, and in particular to a method, apparatus, computer device, and storage medium for calculating vehicle-mounted functional units. Background Technology

[0002] With the rapid development of wireless communication technology and vehicle-to-everything (V2X) technology, intelligent transportation applications such as autonomous driving and in-vehicle video are constantly increasing. The computation of these in-vehicle functional modules and units needs to handle a large number of computationally complex tasks with strict latency sensitivity.

[0003] However, current methods for calculating vehicle-mounted functional units typically employ a single calculation path, which results in low computational efficiency or timeouts and frame drops. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, and storage medium for calculating vehicle-mounted functional units that can improve computing efficiency and avoid timeouts and frame drops, in order to address the aforementioned technical problems.

[0005] Firstly, a method for calculating an on-board functional unit is provided, the method being applied to a pre-configured current routing node corresponding to the current on-board functional unit; the method includes: Obtain the unit type of the current on-board functional unit; the unit type includes fast and slow road selectable functional units, skippable functional units, or necessary functional units; In response to the unit type being a selectable fast / slow road functional unit, the system load of the vehicle system, the evaluation indicators of each scenario of the current on-board functional unit, and the historical actual execution time of the preset number of historical frames of the current on-board functional unit are obtained. The scenario complexity score and scenario timeliness score are determined based on the evaluation indicators of each scenario. The current fast road execution time and current slow road execution time of the current frame of the current vehicle functional unit are determined based on the system load, the historical actual execution time of each historical frame and the scenario complexity score, and the current remaining latency budget value is obtained. In response to the current slow path execution time being greater than the current remaining delay budget value, the current calculation mode of the current vehicle functional unit is determined to be coarse-grained calculation mode. The historical calculation output result of the previous vehicle functional unit is obtained. The current vehicle functional unit is controlled to perform calculation processing based on the coarse-grained calculation mode and the historical calculation output result to obtain the current calculation output result of the current vehicle functional unit. The current calculation output result is then output to the routing node corresponding to the next vehicle functional unit.

[0006] In one embodiment, determining the scenario complexity score and scenario timeliness score based on each scenario evaluation index includes: normalizing each scenario evaluation index to obtain the corresponding normalized scenario evaluation index; evaluating the scenario complexity based on each normalized scenario evaluation index to obtain the scenario complexity score; and evaluating the scenario timeliness based on each normalized scenario evaluation index to obtain the scenario timeliness score.

[0007] In one embodiment, the current fast road execution time and current slow road execution time of the current frame of the current vehicle functional unit are determined based on system load, historical actual execution time of each historical frame and scene complexity score. This includes: online estimation based on a sliding window statistical algorithm or an exponential weighted statistical algorithm, according to system load, historical actual execution time of each historical frame and scene complexity score, to obtain the current fast road execution time and current slow road execution time.

[0008] In one embodiment, the method further includes: in response to a scenario complexity score greater than the complexity score threshold of the current vehicle-mounted functional unit and a scenario timeliness score greater than the timeliness score threshold of the current vehicle-mounted functional unit, determining that the current calculation mode is a coarse-grained calculation mode, obtaining historical calculation output results, controlling the current vehicle-mounted functional unit to perform calculation processing according to the coarse-grained calculation mode and the historical calculation output results to obtain the current calculation output result, and outputting the current calculation output result to the routing node corresponding to the next vehicle-mounted functional unit.

[0009] In one embodiment, the method further includes: obtaining the historical calculation mode of the previous historical frame; in response to the historical calculation mode being a fine-grained calculation mode and the current slow path execution time being less than the current remaining delay budget value, determining the current calculation mode as a fine-grained calculation mode, obtaining the historical calculation output result, controlling the current vehicle function unit to perform calculation processing based on the fine-grained calculation mode and the historical calculation output result to obtain the current calculation output result, and outputting the current calculation output result to the routing node corresponding to the next vehicle function unit.

[0010] In one embodiment, determining that the current computing mode of the current vehicle-mounted functional unit is a coarse-grained computing mode includes: counting the total number of computing mode switching times of the current vehicle-mounted functional unit within a preset time period; and switching the current computing mode to the coarse-grained computing mode in response to the fact that the total number of computing mode switching times is less than a switching time threshold.

[0011] In one embodiment, the method further includes: acquiring the nearest obstacle distance, longitudinal acceleration, and lateral acceleration of the vehicle system; in response to the nearest obstacle distance being less than a distance threshold, or the longitudinal acceleration being greater than a longitudinal acceleration threshold, or the lateral acceleration being greater than a lateral acceleration threshold, determining whether the node type of the current routing node is an emergency necessary slow road node; if the node type is an emergency necessary slow road node, setting the current calculation mode to a fine-grained calculation mode, acquiring historical calculation output results, controlling the current on-board functional unit to perform calculation processing according to the fine-grained calculation mode and the historical calculation output results to obtain the current calculation output result, and outputting the current calculation output result to the routing node corresponding to the next on-board functional unit; wherein, acquiring the unit type of the current on-board functional unit includes: if the node type is not an emergency necessary slow road node, acquiring the unit type.

[0012] In one embodiment, the method further includes: in response to the unit type being a skippable execution functional unit, obtaining the historical obstacle detection result of the previous historical frame and the current obstacle detection result of the current frame; performing Hungarian matching processing on the historical obstacle detection result and the current obstacle detection result to obtain the matched historical obstacle detection result and the matched current obstacle detection result; performing a stability evaluation based on the matched historical obstacle detection result and the matched current obstacle detection result to obtain a stability evaluation score; in response to the stability evaluation score being greater than a stability evaluation score threshold, obtaining the historical calculation output result, and controlling the current vehicle functional unit to output the historical calculation output result to the routing node corresponding to the next vehicle functional unit.

[0013] In one embodiment, the method further includes: in response to a stability evaluation score being less than or equal to a stability evaluation score threshold, determining the current calculation mode as a fine-grained calculation mode, obtaining historical calculation output results, controlling the current vehicle functional unit to perform calculation processing based on the fine-grained calculation mode and historical calculation output results to obtain the current calculation output result, and outputting the current calculation output result to the routing node corresponding to the next vehicle functional unit.

[0014] In a second aspect, there is an in-vehicle functional unit computing device, which is applied to a pre-configured current routing node corresponding to the current in-vehicle functional unit; the device includes a unit type acquisition module, an evaluation index acquisition module, an execution time calculation module, and a coarse-grained calculation module.

[0015] The unit type acquisition module 810 is used to acquire the unit type of the current vehicle-mounted functional unit; the unit type includes a fast / slow road selectable functional unit, a skippable execution functional unit, or a necessary execution functional unit; the evaluation index acquisition module 820 is used to acquire the system load of the vehicle system, the evaluation indexes of each scenario of the current vehicle-mounted functional unit, and the historical actual execution time of the current vehicle-mounted functional unit in a preset number of historical frames in response to the unit type being a fast / slow road selectable functional unit; the execution time calculation module 830 is used to determine the scenario complexity score and scenario timeliness score based on each scenario evaluation index, and calculate the execution time based on the system load, the historical actual execution time of each historical frame, and the scenario complexity score. The system determines the current fast road execution time and the current slow road execution time of the current frame of the current vehicle functional unit, and obtains the current remaining delay budget value. The coarse-grained calculation module 840 is used to respond to the current slow road execution time being greater than the current remaining delay budget value, determine that the current calculation mode of the current vehicle functional unit is the coarse-grained calculation mode, obtain the historical calculation output result of the previous vehicle functional unit, control the current vehicle functional unit to perform calculation processing according to the coarse-grained calculation mode and the historical calculation output result to obtain the current calculation output result of the current vehicle functional unit, and output the current calculation output result to the routing node corresponding to the next vehicle functional unit.

[0016] Thirdly, a computer device is provided, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods described in the above method embodiments.

[0017] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of any of the methods described in the above method embodiments.

[0018] The aforementioned vehicle-mounted functional unit calculation method, apparatus, computer equipment, and storage medium obtain the unit type of the current vehicle-mounted functional unit. The unit type includes a fast / slow road selectable functional unit, a skippable execution functional unit, or a necessary execution functional unit. Then, in response to the unit type being a fast / slow road selectable functional unit, the system load of the vehicle system, the scenario evaluation indicators of the current vehicle-mounted functional unit, and the historical actual execution time of a preset number of historical frames for the current vehicle-mounted functional unit are obtained. Next, scenario complexity scores and scenario timeliness scores are determined based on the scenario evaluation indicators, and the current fast road execution time and current slow road execution time of the current frame of the current vehicle-mounted functional unit are determined based on the system load, the historical actual execution time of each historical frame, and the scenario complexity score. The current remaining latency budget value is determined. Finally, in response to the current slow path execution time exceeding the current remaining latency budget value, the current computation mode of the current vehicle functional unit is determined to be coarse-grained computation mode. The historical computation output results of the previous vehicle functional unit are obtained. The current vehicle functional unit is controlled to perform computation processing based on the coarse-grained computation mode and the historical computation output results to obtain the current computation output result of the current vehicle functional unit. The current computation output result is then output to the routing node corresponding to the next vehicle functional unit. By switching the current computation mode of the current vehicle functional unit to coarse-grained computation mode, timeout and frame drops are avoided. High-quality and high-accuracy current computation output results are also guaranteed under the premise of low latency, thus improving computation efficiency. Attached Figure Description

[0019] Figure 1 This is an application environment diagram of the vehicle-mounted functional unit calculation method in one embodiment; Figure 2 This is a schematic diagram of the first process of a vehicle-mounted functional unit calculation method in one embodiment; Figure 3 This is a schematic diagram of the process for determining scenario complexity scores and scenario timeliness scores based on various scenario evaluation indicators in one embodiment. Figure 4 This is a schematic diagram of the second process of the vehicle-mounted functional unit calculation method in one embodiment; Figure 5 This is a schematic diagram of the third process of the vehicle-mounted functional unit calculation method in one embodiment; Figure 6 This is a flowchart illustrating the process of determining the current calculation mode of the current vehicle-mounted functional unit as a coarse-grained calculation mode in one embodiment. Figure 7 This is a schematic diagram of the third process of the vehicle-mounted functional unit calculation method in one embodiment; Figure 8 This is a structural block diagram of an onboard functional unit computing device in one embodiment; Figure 9This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0021] To facilitate understanding of this application, a more complete description will be provided below with reference to the accompanying drawings, which illustrate embodiments of the present application. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of this application will be thorough and complete.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the specification of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application.

[0023] It is understood that the terms "first," "second," etc., used herein may be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of this application, a first resistor may be referred to as a second resistor, and similarly, a second resistor may be referred to as a first resistor. Both the first resistor and the second resistor are resistors, but they are not the same resistor.

[0024] It is understood that the term "connection" in the following embodiments should be understood as "electrical connection," "communication connection," etc., if the connected circuits, modules, units, etc., have electrical signal or data transmission with each other.

[0025] When used herein, the singular forms of “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising,” “including,” or “having,” etc., specify the presence of the stated feature, whole, step, operation, component, part, or combination thereof, but do not preclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, components, parts, or combinations thereof.

[0026] The vehicle-mounted functional unit calculation method provided in this application can be applied to, for example... Figure 1The vehicle system shown includes several in-vehicle functional modules 100 that are sequentially connected via communication. Each in-vehicle functional module 100 includes several in-vehicle functional units 110 that are sequentially connected via communication. Each in-vehicle functional unit 110 is pre-configured with a corresponding routing node 120 that is connected to the previous in-vehicle functional unit. It can be understood that the current in-vehicle functional unit is the in-vehicle functional unit 110 that needs to perform data calculation at the current moment; the current routing node is the routing node corresponding to the current in-vehicle functional unit.

[0027] In a specific example, the vehicle-mounted functional modules 100 may be, but are not limited to, a positioning module, a perception module, and a planning and control module, which are sequentially connected via communication. The perception module may include, but is not limited to, an image point cloud preprocessing unit, a target detection unit, and a target tracking unit, which are sequentially connected via communication. The above is merely a specific example; in actual applications, the configuration can be flexibly adjusted according to user needs and is not described here.

[0028] Firstly, such as Figure 2 As shown, a method for calculating an on-board functional unit is provided. The method is applied to the current routing node corresponding to the current on-board functional unit that is pre-configured. The method includes steps 201 to 204.

[0029] Step 201: Obtain the unit type of the current vehicle-mounted functional unit.

[0030] The unit types include selectable fast / slow path functional units, skippable execution functional units, and necessary execution functional units. It can be understood that selectable fast / slow path functional units are onboard functional units that can select the current calculation mode. Skipable execution functional units are onboard functional units that can skip calculation processing. Necessary execution functional units are onboard functional units that must undergo calculation processing. Specifically, the current routing node obtains the unit type of the current onboard functional unit.

[0031] Step 202: In response to the unit type being a selectable fast / slow road functional unit, obtain the system load of the vehicle system, the evaluation indicators of each scenario of the current on-board functional unit, and the historical actual execution time of the preset number of historical frames of the current on-board functional unit.

[0032] Specifically, when the current routing node identifies a unit type that can be selected as a fast or slow road functional unit, it obtains the system load of the vehicle system, the evaluation indicators of each scenario of the current vehicle functional unit, and the historical actual execution time of the preset number of historical frames of the current vehicle functional unit.

[0033] In a specific example, the preset number of historical frames is the preset number of frames preceding the current frame of the current onboard functional unit. This is merely a specific example; in actual applications, it should be flexibly set according to user needs, and will not be specified here.

[0034] In a specific example, the current vehicle-mounted functional unit target detection unit may be, but is not limited to, a target detection unit. The scene evaluation indicators of the target detection unit may include, but are not limited to, the nearest obstacle distance, relative speed, curvature, and obstacle density. The above is only a specific example and can be flexibly set according to user needs in actual applications. It will not be set here.

[0035] Step 203: Determine the scenario complexity score and scenario timeliness score based on the evaluation indicators of each scenario, and determine the current fast road execution time and current slow road execution time of the current frame of the current vehicle functional unit based on the system load, the historical actual execution time of each historical frame and the scenario complexity score, and obtain the current remaining latency budget value.

[0036] Specifically, the current routing node determines the scenario complexity score and scenario timeliness score based on the evaluation indicators of each scenario, and determines the current fast road execution time and current slow road execution time of the current frame of the current vehicle functional unit based on the system load, the historical actual execution time of each historical frame and the scenario complexity score, and obtains the current remaining latency budget value.

[0037] In a specific example, the current remaining latency budget value is obtained, including: Get the total latency budget and the current elapsed time; The remaining time budget is determined based on the difference between the total budget and the current elapsed time. The above is just a specific example. In actual applications, it can be flexibly set according to user needs, and will not be set here.

[0038] In one embodiment, such as Figure 3 As shown, the scenario complexity score and scenario timeliness score are determined based on the evaluation indicators of each scenario, including steps 301 to 303.

[0039] Step 301: Normalize the evaluation indicators for each scenario to obtain the corresponding normalized scenario evaluation indicators. Step 302: Evaluate the scene complexity based on the normalized scene evaluation metrics to obtain the scene complexity score. Step 303: Evaluate the timeliness of the scenario based on the normalized scenario evaluation indicators to obtain the scenario timeliness score.

[0040] Specifically, the current routing node normalizes the evaluation metrics for each scenario to obtain the corresponding normalized scenario evaluation metrics. Then, it evaluates the scenario complexity based on the normalized scenario evaluation metrics to obtain a scenario complexity score. Next, it evaluates the scenario timeliness based on the normalized scenario evaluation metrics to obtain a scenario timeliness score, thereby improving the efficiency and accuracy of scenario complexity and scenario timeliness analysis.

[0041] In this embodiment, the evaluation indicators for each scenario are normalized to obtain the corresponding normalized scenario evaluation indicators. Then, the scenario complexity is evaluated based on the normalized scenario evaluation indicators to obtain a scenario complexity score. Next, the scenario timeliness is evaluated based on the normalized scenario evaluation indicators to obtain a scenario timeliness score, thereby improving the efficiency and accuracy of scenario complexity and scenario timeliness analysis.

[0042] In one embodiment, the current fast road execution time and current slow road execution time of the current frame of the current onboard functional unit are determined based on system load, historical actual execution time of each historical frame, and scenario complexity score, including: Based on the sliding window statistical algorithm or the exponential weighted statistical algorithm, online estimation is performed according to the system load, the historical actual execution time of each historical frame, and the scene complexity score to obtain the current fast path execution time and the current slow path execution time.

[0043] Specifically, the current routing node uses a sliding window statistical algorithm or an exponential weighted statistical algorithm to perform online estimation based on system load, historical actual execution time of each historical frame, and scenario complexity score to obtain the current fast road execution time and the current slow road execution time, thereby improving the efficiency and accuracy of obtaining the current fast road execution time and the current slow road execution time of the current frame of the current on-board functional unit.

[0044] In this embodiment, based on the sliding window statistical algorithm or the exponential weighted statistical algorithm, online estimation is performed according to the system load, the historical actual execution time of each historical frame, and the scene complexity score to obtain the current fast road execution time and the current slow road execution time, which improves the efficiency and accuracy of obtaining the current fast road execution time and the current slow road execution time of the current frame of the current vehicle functional unit.

[0045] Step 204: In response to the current slow path execution time being greater than the current remaining delay budget value, determine that the current calculation mode of the current vehicle functional unit is coarse-grained calculation mode, obtain the historical calculation output result of the previous vehicle functional unit, control the current vehicle functional unit to perform calculation processing according to the coarse-grained calculation mode and the historical calculation output result to obtain the current calculation output result of the current vehicle functional unit, and output the current calculation output result to the routing node corresponding to the next vehicle functional unit.

[0046] Among them, the historical calculation output result of the previous vehicle functional unit of the current vehicle functional unit refers to the calculation output result of the previous preceding vehicle functional unit of the current vehicle functional unit in the current frame; the current calculation output result of the current vehicle functional unit refers to the calculation output result of the current vehicle functional unit in the current frame.

[0047] Specifically, when the current routing node detects that the execution time of the current slow path is greater than the current remaining latency budget, it indicates that the current remaining latency budget is insufficient for the current vehicle functional unit to adopt fine-grained calculation mode. Therefore, it determines that the current calculation mode of the current vehicle functional unit is coarse-grained calculation mode, obtains the historical calculation output result of the previous vehicle functional unit, controls the current vehicle functional unit to perform calculation processing based on the coarse-grained calculation mode and the historical calculation output result to obtain the current calculation output result of the current vehicle functional unit, and outputs the current calculation output result to the routing node corresponding to the next vehicle functional unit. By switching the current calculation mode of the current vehicle functional unit to coarse-grained calculation mode, timeout and frame drop are avoided, and high-quality and high-accuracy current calculation output results are output under the premise of low latency, thus improving calculation efficiency.

[0048] Based on this, the above-mentioned vehicle-mounted functional unit calculation method obtains the unit type of the current vehicle-mounted functional unit; the unit type includes fast / slow road selectable functional units, skippable execution functional units, or necessary execution functional units; then, in response to the unit type being a fast / slow road selectable functional unit, it obtains the system load of the vehicle system, the scenario evaluation indicators of the current vehicle-mounted functional unit, and the historical actual execution time of the current vehicle-mounted functional unit for a preset number of historical frames; next, it determines the scenario complexity score and scenario timeliness score based on the scenario evaluation indicators, and determines the current fast road execution time and current slow road execution time of the current frame of the current vehicle-mounted functional unit based on the system load, the historical actual execution time of each historical frame, and the scenario complexity score, and obtains the current remaining time. Finally, in response to the current slow path execution time exceeding the current remaining delay budget value, the current computation mode of the current vehicle functional unit is determined to be coarse-grained computation mode. The historical computation output results of the previous vehicle functional unit are obtained, and the current vehicle functional unit is controlled to perform computation processing based on the coarse-grained computation mode and the historical computation output results to obtain the current computation output result of the current vehicle functional unit. The current computation output result is then output to the routing node corresponding to the next vehicle functional unit. By switching the current computation mode of the current vehicle functional unit to coarse-grained computation mode, timeout and frame drops are avoided, and high-quality and high-accuracy current computation output results are output under the premise of low latency, thus improving computation efficiency.

[0049] In one embodiment, such as Figure 4 As shown, the method further includes step 401.

[0050] Step 401: In response to the scenario complexity score being greater than the complexity score threshold of the current vehicle functional unit and the scenario timeliness score being greater than the timeliness score threshold of the current vehicle functional unit, determine that the current calculation mode is the coarse-grained calculation mode, obtain the historical calculation output results, control the current vehicle functional unit to perform calculation processing according to the coarse-grained calculation mode and the historical calculation output results to obtain the current calculation output result, and output the current calculation output result to the routing node corresponding to the next vehicle functional unit.

[0051] Specifically, when the current routing node identifies that the scene complexity score is greater than the complexity score threshold of the current vehicle functional unit and the scene timeliness score is greater than the timeliness score threshold of the current vehicle functional unit, it indicates that the scene complexity is high and the scene timeliness requirement is high. It needs to obtain the current calculation output result faster. The current calculation mode is determined to be coarse-grained calculation mode. The historical calculation output results are obtained, and the current vehicle functional unit is controlled to perform calculation processing according to the coarse-grained calculation mode and the historical calculation output results to obtain the current calculation output result. The current calculation output result is then output to the routing node corresponding to the next vehicle functional unit. This avoids timeout and frame drop in complex environments and also avoids the inability to guarantee latency when triggering global degradation, thus improving the stability of the calculation results.

[0052] In this embodiment, in response to the scenario complexity score being greater than the complexity score threshold of the current vehicle-mounted functional unit and the scenario timeliness score being greater than the timeliness score threshold of the current vehicle-mounted functional unit, the current calculation mode is determined to be a coarse-grained calculation mode. Historical calculation output results are obtained, and the current vehicle-mounted functional unit is controlled to perform calculation processing based on the coarse-grained calculation mode and historical calculation output results to obtain the current calculation output result. The current calculation output result is then output to the routing node corresponding to the next vehicle-mounted functional unit. This avoids timeouts and frame drops in complex environments and also avoids the inability to guarantee latency due to global degradation during the triggering process, thereby improving the stability of the calculation results.

[0053] In one embodiment, such as Figure 5 As shown, the method further includes steps 501 to 502.

[0054] Step 501: Obtain the historical calculation mode of the previous historical frame; Step 502: In response to the historical calculation mode being fine-grained calculation mode and the current slow road execution time being less than the current remaining delay budget value, determine that the current calculation mode is fine-grained calculation mode, obtain the historical calculation output results, control the current vehicle functional unit to perform calculation processing based on the fine-grained calculation mode and the historical calculation output results to obtain the current calculation output results, and output the current calculation output results to the routing node corresponding to the next vehicle functional unit.

[0055] Specifically, the current routing node obtains the historical calculation mode of the previous historical frame. Then, in response to the historical calculation mode being a fine-grained calculation mode and the current slow path execution time being less than the current remaining delay budget value, it indicates that the current remaining delay budget value is sufficient for the current vehicle functional unit to adopt a coarse-grained calculation mode without causing frequent changes in the calculation mode. The current calculation mode is then determined to be a fine-grained calculation mode. The historical calculation output results are obtained, and the current vehicle functional unit is controlled to perform calculation processing based on the fine-grained calculation mode and the historical calculation output results to obtain the current calculation output result. The current calculation output result is then output to the routing node corresponding to the next vehicle functional unit. This avoids frequent changes in the calculation mode, enriches the calculation modes of the current vehicle functional unit, and improves the stability and accuracy of the calculation results.

[0056] In a specific example, determining that the current computation mode is a fine-grained computation mode includes: The total number of times the calculation mode of the current in-vehicle functional unit switches within a preset time period is counted. In response to the total number of computing mode switching times being less than the switching time threshold, the current computing mode is determined to be a fine-grained computing mode; In response to the total number of calculation mode switching times being greater than or equal to the switching time threshold, the current calculation mode is determined as the historical calculation mode of the previous historical frame. The above is only a specific example. In actual applications, it can be flexibly set according to user needs, and no restrictions are imposed here.

[0057] In this embodiment, the historical calculation mode of the previous historical frame is obtained; then, in response to the historical calculation mode being a fine-grained calculation mode and the current slow path execution time being less than the current remaining delay budget value, the current calculation mode is determined to be a fine-grained calculation mode, the historical calculation output result is obtained, and the current vehicle functional unit is controlled to perform calculation processing based on the fine-grained calculation mode and the historical calculation output result to obtain the current calculation output result, and the current calculation output result is output to the routing node corresponding to the next vehicle functional unit. This avoids frequent changes in the calculation mode, enriches the calculation modes of the current vehicle functional unit, and improves the stability and accuracy of the calculation results.

[0058] In one embodiment, such as Figure 6 As shown, the current calculation mode of the current vehicle-mounted functional unit is determined to be coarse-grained calculation mode, including steps 601 and 602.

[0059] Step 601: Calculate the total number of times the calculation mode of the current vehicle-mounted functional unit switches within a preset time period; Step 602: In response to the total number of calculation mode switching times being less than the switching time threshold, the current calculation mode is switched to the coarse-grained calculation mode.

[0060] Specifically, the current routing node counts the total number of times the current vehicle-mounted functional unit switches its calculation mode within a preset time period. Then, in response to the total number of calculation mode switches being less than the switching threshold, the current calculation mode is switched to a coarse-grained calculation mode. This avoids frequent switching of the current calculation mode of the current vehicle-mounted functional unit due to jitter, and improves the stability and accuracy of the calculation results.

[0061] In a specific example, determining that the current calculation mode of the current on-board functional unit is a coarse-grained calculation mode also includes: In response to the total number of calculation mode switching times being greater than or equal to the switching time threshold, the current calculation mode is determined as the historical calculation mode of the previous historical frame. The above is only a specific example. In actual applications, it can be flexibly set according to user needs, and no restrictions are imposed here.

[0062] In this embodiment, the total number of calculation mode switching times of the current vehicle-mounted functional unit within a preset time period is counted; then, in response to the total number of calculation mode switching times being less than the switching time threshold, the current calculation mode is switched to a coarse-grained calculation mode, which avoids frequent switching of the current calculation mode of the current vehicle-mounted functional unit and improves the stability and accuracy of the calculation results.

[0063] In one embodiment, the method further includes: Obtain the nearest obstacle distance, longitudinal acceleration, and lateral acceleration of the vehicle system; In response to the nearest obstacle being less than a distance threshold, or the longitudinal acceleration being greater than a longitudinal acceleration threshold, or the lateral acceleration being greater than a lateral acceleration threshold, determine whether the current routing node is an emergency necessary slow path node. If the node type is an emergency necessary slow road node, the current calculation mode is set to fine-grained calculation mode, the historical calculation output results are obtained, the current vehicle functional unit is controlled to perform calculation processing according to the fine-grained calculation mode and the historical calculation output results to obtain the current calculation output result, and the current calculation output result is output to the routing node corresponding to the next vehicle functional unit. Among them, obtaining the unit type of the current vehicle-mounted functional unit includes: if the node type is not an emergency necessary slow road node, obtaining the unit type.

[0064] Specifically, the current routing node acquires the nearest obstacle distance, longitudinal acceleration, and lateral acceleration of the vehicle system. Then, in response to the nearest obstacle distance being less than a distance threshold, or the longitudinal acceleration being greater than a longitudinal acceleration threshold, or the lateral acceleration being greater than a lateral acceleration threshold, it determines whether the current routing node is an emergency necessary slow road node. Next, if the node type is an emergency necessary slow road node, the current calculation mode is set to fine-grained calculation mode, historical calculation output results are acquired, and the current on-board functional unit is controlled to perform calculation processing based on the fine-grained calculation mode and historical calculation output results to obtain the current calculation output result, which is then output to the routing node corresponding to the next on-board functional unit. Simultaneously, if the node type is not an emergency necessary slow road node, the unit type is acquired, and when the nearest obstacle distance is less than a distance threshold, or the longitudinal acceleration is greater than a longitudinal acceleration threshold, or the lateral acceleration is greater than a lateral acceleration threshold, the on-board functional unit corresponding to the emergency necessary slow road node is forced to adopt fine-grained calculation mode to ensure the minimum safety of the on-board functional unit's calculation.

[0065] In this embodiment, the nearest obstacle distance, longitudinal acceleration, and lateral acceleration of the vehicle system are obtained. Then, in response to the nearest obstacle distance being less than a distance threshold, or the longitudinal acceleration being greater than a longitudinal acceleration threshold, or the lateral acceleration being greater than a lateral acceleration threshold, it is determined whether the node type of the current routing node is an emergency necessary slow road node. Next, if the node type is an emergency necessary slow road node, the current calculation mode is set to fine-grained calculation mode, historical calculation output results are obtained, and the current on-board functional unit is controlled to perform calculation processing based on the fine-grained calculation mode and historical calculation output results to obtain the current calculation output result, and the current calculation output result is output to the routing node corresponding to the next on-board functional unit. At the same time, if the node type is not an emergency necessary slow road node, the unit type is obtained, and when the nearest obstacle distance is less than a distance threshold, or the longitudinal acceleration is greater than a longitudinal acceleration threshold, or the lateral acceleration is greater than a lateral acceleration threshold, the on-board functional unit corresponding to the emergency necessary slow road node is forced to adopt fine-grained calculation mode to ensure the minimum safety of the on-board functional unit calculation.

[0066] In one embodiment, such as Figure 7 As shown, the method further includes steps 701 to 704.

[0067] Step 701: In response to the unit type being a skippable execution function unit, obtain the historical obstacle detection results of the previous historical frame and the current obstacle detection results of the current frame; Step 702: After performing Hungarian matching on the historical obstacle detection results and the current obstacle detection results, the matched historical obstacle detection results and the matched current obstacle detection results are obtained. Step 703: Perform a stability assessment based on the historical obstacle detection results after matching and the current obstacle detection results after matching to obtain a stability assessment score; Step 704: In response to the stability assessment score being greater than the stability assessment score threshold, obtain the historical calculation output results and control the current vehicle functional unit to output the historical calculation output results to the routing node corresponding to the next vehicle functional unit.

[0068] Specifically, in response to the unit type being a skippable execution function unit, the current routing node obtains the historical obstacle detection results of the previous historical frame and the current obstacle detection results of the current frame. Then, it performs Hungarian matching on the historical and current obstacle detection results to obtain matched historical and current obstacle detection results. Next, it performs a stability evaluation based on the matched historical and current obstacle detection results to obtain a stability evaluation score. Finally, in response to the stability evaluation score being greater than a stability evaluation score threshold, it obtains the historical calculation output results and controls the current vehicle-mounted function unit to output the historical calculation output results to the routing node corresponding to the next vehicle-mounted function unit. This enriches the calculation mode of the current vehicle-mounted function unit, improves the stability and accuracy of the calculation results, and improves computational efficiency while ensuring the accuracy and stability of the vehicle-mounted function unit's calculations by directly skipping the calculation process.

[0069] In this embodiment, in response to the unit type being a skippable execution function unit, the historical obstacle detection results of the previous historical frame and the current obstacle detection results of the current frame are obtained. Then, Hungarian matching processing is performed on the historical obstacle detection results and the current obstacle detection results to obtain the matched historical obstacle detection results and the matched current obstacle detection results. Next, a stability evaluation is performed based on the matched historical obstacle detection results and the matched current obstacle detection results to obtain a stability evaluation score. Finally, in response to the stability evaluation score being greater than the stability evaluation score threshold, the historical calculation output results are obtained, and the current vehicle-mounted function unit is controlled to output the historical calculation output results to the routing node corresponding to the next vehicle-mounted function unit. This enriches the calculation mode of the current vehicle-mounted function unit, reduces redundant calculations, and improves the stability and accuracy of the calculation results. By allowing the current vehicle-mounted function unit to directly skip the calculation processing, the calculation efficiency is improved while ensuring the accuracy and stability of the vehicle-mounted function unit's calculations.

[0070] In one embodiment, such as Figure 7 As shown, the method further includes step 705.

[0071] Step 705: In response to the stability assessment score being less than or equal to the stability assessment score threshold, determine that the current calculation mode is fine-grained calculation mode, obtain the historical calculation output results, control the current vehicle functional unit to perform calculation processing according to the fine-grained calculation mode and the historical calculation output results to obtain the current calculation output result, and output the current calculation output result to the routing node corresponding to the next vehicle functional unit.

[0072] Specifically, when the stability evaluation score is less than or equal to the stability evaluation score threshold, the current routing node determines the current calculation mode to be fine-grained calculation mode, obtains the historical calculation output results, controls the current vehicle functional unit to perform calculation processing based on the fine-grained calculation mode and the historical calculation output results to obtain the current calculation output result, and outputs the current calculation output result to the routing node corresponding to the next vehicle functional unit. This enriches the calculation mode of the current vehicle functional unit and improves the stability and accuracy of the calculation results.

[0073] In this embodiment, in response to a stability evaluation score being less than or equal to a stability evaluation score threshold, the current calculation mode is determined to be a fine-grained calculation mode. Historical calculation output results are obtained, and the current vehicle functional unit is controlled to perform calculation processing based on the fine-grained calculation mode and historical calculation output results to obtain the current calculation output result. The current calculation output result is then output to the routing node corresponding to the next vehicle functional unit, which enriches the calculation mode of the current vehicle functional unit and improves the stability and accuracy of the calculation results.

[0074] In a specific example, the method further includes: responding to the unit type being a necessary execution functional unit, determining the current calculation mode as a fine-grained calculation mode, obtaining historical calculation output results, controlling the current vehicle-mounted functional unit to perform calculation processing based on the fine-grained calculation mode and the historical calculation output results to obtain the current calculation output result, and outputting the current calculation output result to the routing node corresponding to the next vehicle-mounted functional unit. The above is only a specific example; in actual applications, it can be flexibly configured according to user needs, and is not described here.

[0075] In a specific example, the method further includes: storing the unit type of the current vehicle functional unit, the system load of the vehicle system, the evaluation indicators of each scenario of the current vehicle functional unit, the current calculation mode, and the current calculation output results into the working log of the current vehicle functional unit, so as to facilitate the reproduction of the current working state in subsequent offline testing. The above is only a specific example, and it can be flexibly set according to user needs in actual applications. It is not set here.

[0076] In a specific example, the following table compares the technical effects of traditional vehicle-mounted functional unit calculation methods (only fine-grained calculation mode) with those of this application: The above are just specific examples. In actual applications, the settings should be flexibly adjusted according to user needs, and no settings will be made here.

[0077] It should be understood that, although Figures 2-7 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figures 2-7 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0078] Secondly, such as Figure 8 As shown, an in-vehicle functional unit computing device is provided. The device is applied to the current routing node corresponding to the current in-vehicle functional unit in a pre-configured manner. The device includes a unit type acquisition module 810, an evaluation index acquisition module 820, an execution time calculation module 830, and a coarse-grained calculation module 840.

[0079] The system includes several modules: a unit type acquisition module for obtaining the unit type of the current vehicle-mounted functional unit; unit types include selectable fast / slow road functional units, skippable execution functional units, and necessary execution functional units; an evaluation index acquisition module for obtaining the system load of the vehicle system, the evaluation indexes for each scenario of the current vehicle-mounted functional unit, and the historical actual execution time of a preset number of historical frames for the current vehicle-mounted functional unit, in response to the unit type being a selectable fast / slow road functional unit; and an execution time calculation module for determining the scenario complexity score and scenario timeliness score based on the evaluation indexes for each scenario, and determining the execution time calculation based on the system load, the historical actual execution time of each historical frame, and the scenario complexity score. The current fast road execution time and slow road execution time of the current frame of the current vehicle functional unit are determined, and the current remaining delay budget value is obtained. The coarse-grained calculation module is used to determine the current calculation mode of the current vehicle functional unit as coarse-grained calculation mode in response to the current slow road execution time being greater than the current remaining delay budget value. It obtains the historical calculation output result of the previous vehicle functional unit, controls the current vehicle functional unit to perform calculation processing according to the coarse-grained calculation mode and the historical calculation output result to obtain the current calculation output result of the current vehicle functional unit, and outputs the current calculation output result to the routing node corresponding to the next vehicle functional unit.

[0080] In one embodiment, the execution time calculation module 830 includes a scene evaluation unit.

[0081] The scenario evaluation unit is used to normalize each scenario evaluation index to obtain the corresponding normalized scenario evaluation index; the scenario evaluation unit is used to evaluate the scenario complexity based on each normalized scenario evaluation index to obtain the scenario complexity score; the scenario evaluation unit is used to evaluate the scenario timeliness based on each normalized scenario evaluation index to obtain the scenario timeliness score.

[0082] In one embodiment, the execution time calculation module 830 further includes an online estimation unit.

[0083] The online estimation unit is used to estimate the current fast path execution time and the current slow path execution time based on the sliding window statistical algorithm or the exponential weighted statistical algorithm, according to the system load, the historical actual execution time of each historical frame and the scene complexity score.

[0084] In one embodiment, the coarse-grained calculation module 840 is further configured to, in response to a scenario complexity score greater than the complexity score threshold of the current vehicle-mounted functional unit and a scenario timeliness score greater than the timeliness score threshold of the current vehicle-mounted functional unit, determine that the current calculation mode is a coarse-grained calculation mode, obtain historical calculation output results, control the current vehicle-mounted functional unit to perform calculation processing based on the coarse-grained calculation mode and historical calculation output results to obtain the current calculation output result, and output the current calculation output result to the routing node corresponding to the next vehicle-mounted functional unit.

[0085] In one embodiment, the device further includes a fine-grained computing module.

[0086] The fine-grained calculation module is used to obtain the historical calculation mode of the previous historical frame. In response to the historical calculation mode being fine-grained and the current slow path execution time being less than the current remaining delay budget value, the fine-grained calculation module determines the current calculation mode as fine-grained calculation mode, obtains the historical calculation output result, controls the current vehicle function unit to perform calculation processing based on the fine-grained calculation mode and the historical calculation output result to obtain the current calculation output result, and outputs the current calculation output result to the routing node corresponding to the next vehicle function unit.

[0087] In one embodiment, the coarse-grained calculation module 840 includes a coarse-grained calculation mode determination unit.

[0088] The coarse-grained calculation mode determination unit is used to count the total number of calculation mode switching times of the current vehicle-mounted functional unit within a preset time period; the coarse-grained calculation mode determination unit is used to switch the current calculation mode to the coarse-grained calculation mode in response to the total number of calculation mode switching times being less than the switching time threshold.

[0089] In one embodiment, the device further includes a minimum security assurance calculation module.

[0090] The minimum safety assurance calculation module is used to obtain the nearest obstacle distance, longitudinal acceleration, and lateral acceleration of the vehicle system. In response to the nearest obstacle distance being less than a distance threshold, or the longitudinal acceleration being greater than a longitudinal acceleration threshold, or the lateral acceleration being greater than a lateral acceleration threshold, the minimum safety assurance calculation module determines whether the node type of the current routing node is an emergency necessary slow road node. If the node type is an emergency necessary slow road node, the minimum safety assurance calculation module sets the current calculation mode to fine-grained calculation mode, obtains historical calculation output results, controls the current on-board functional unit to perform calculation processing based on the fine-grained calculation mode and historical calculation output results to obtain the current calculation output result, and outputs the current calculation output result to the routing node corresponding to the next on-board functional unit. Obtaining the unit type of the current on-board functional unit includes: if the node type is not an emergency necessary slow road node, obtaining the unit type.

[0091] In one embodiment, the apparatus further includes a skippable execution module.

[0092] The skippable execution module is used to obtain the historical obstacle detection results of the previous historical frame and the current obstacle detection results of the current frame in response to the unit type being a skippable execution functional unit; the skippable execution module is used to perform Hungarian matching processing on the historical obstacle detection results and the current obstacle detection results to obtain the matched historical obstacle detection results and the matched current obstacle detection results; the skippable execution module is used to perform stability evaluation based on the matched historical obstacle detection results and the matched current obstacle detection results to obtain a stability evaluation score; the skippable execution module is used to obtain the historical calculation output results in response to the stability evaluation score being greater than the stability evaluation score threshold, and control the current vehicle functional unit to output the historical calculation output results to the routing node corresponding to the next vehicle functional unit.

[0093] In one embodiment, the skippable execution module is further configured to, in response to a stability evaluation score less than or equal to a stability evaluation score threshold, determine that the current calculation mode is a fine-grained calculation mode, obtain historical calculation output results, control the current vehicle functional unit to perform calculation processing based on the fine-grained calculation mode and historical calculation output results to obtain the current calculation output result, and output the current calculation output result to the routing node corresponding to the next vehicle functional unit.

[0094] Specific limitations regarding the in-vehicle functional unit computing device can be found in the limitations of the in-vehicle functional unit computing method described above, and will not be repeated here. Each module in the aforementioned in-vehicle functional unit computing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in the computer device, or stored in software in the memory of the computer device, so that the processor can call and execute the operations corresponding to each module.

[0095] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a vehicle-mounted functional unit computing method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0096] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0097] Thirdly, a computer device is provided, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods described in the above method embodiments.

[0098] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of any of the methods described in the above method embodiments.

[0099] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0100] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for calculating an on-board functional unit, the method being applied to a pre-configured current routing node corresponding to the current on-board functional unit; the method comprising: Obtain the unit type of the current vehicle-mounted functional unit; The unit types include fast / slow path selectable functional units, skippable execution functional units, or necessary execution functional units; In response to the unit type being the fast / slow road selectable functional unit, the system load of the vehicle system, the evaluation indicators of each scenario of the current vehicle functional unit, and the historical actual execution time of the current vehicle functional unit in a preset number of historical frames are obtained. The scenario complexity score and scenario timeliness score are determined based on the evaluation indicators of each scenario. The current fast road execution time and current slow road execution time of the current frame of the current vehicle functional unit are determined based on the system load, the historical actual execution time of each historical frame and the scenario complexity score. The current remaining latency budget value is obtained. In response to the current slow path execution time being greater than the current remaining delay budget value, the current calculation mode of the current vehicle functional unit is determined to be coarse-grained calculation mode. The historical calculation output result of the previous vehicle functional unit is obtained. The current vehicle functional unit is controlled to perform calculation processing according to the coarse-grained calculation mode and the historical calculation output result to obtain the current calculation output result of the current vehicle functional unit. The current calculation output result is then output to the routing node corresponding to the next vehicle functional unit.

2. The method according to claim 1, characterized in that, The determination of scenario complexity score and scenario timeliness score based on each of the scenario evaluation indicators includes: The scenario evaluation indicators are normalized to obtain the corresponding normalized scenario evaluation indicators. The scene complexity is evaluated based on the scene evaluation indicators after normalization, and the scene complexity score is obtained. The timeliness of the scenario is evaluated based on the normalized scenario evaluation indicators to obtain the scenario timeliness score.

3. The method according to claim 1, characterized in that, The step of determining the current fast road execution time and current slow road execution time of the current frame of the current on-board functional unit based on the system load, the historical actual execution time of each historical frame, and the scene complexity score includes: Based on a sliding window statistical algorithm or an exponential weighted statistical algorithm, the current fast path execution time and the current slow path execution time are estimated online according to the system load, the historical actual execution time of each historical frame, and the scene complexity score.

4. The method according to claim 1, characterized in that, The method further includes: In response to the scenario complexity score being greater than the complexity score threshold of the current vehicle-mounted functional unit and the scenario timeliness score being greater than the timeliness score threshold of the current vehicle-mounted functional unit, the current calculation mode is determined to be the coarse-grained calculation mode. The historical calculation output results are obtained, and the current vehicle-mounted functional unit is controlled to perform calculation processing based on the coarse-grained calculation mode and the historical calculation output results to obtain the current calculation output result. The current calculation output result is then output to the routing node corresponding to the next vehicle-mounted functional unit.

5. The method according to claim 1, characterized in that, The method further includes: Obtain the historical calculation pattern of the previous historical frame; In response to the historical calculation mode being fine-grained calculation mode and the current slow path execution time being less than the current remaining delay budget value, the current calculation mode is determined to be fine-grained calculation mode. The historical calculation output result is obtained, and the current vehicle function unit is controlled to perform calculation processing based on the fine-grained calculation mode and the historical calculation output result to obtain the current calculation output result. The current calculation output result is then output to the routing node corresponding to the next vehicle function unit.

6. The method according to claim 1, characterized in that, The step of determining that the current calculation mode of the current vehicle-mounted functional unit is a coarse-grained calculation mode includes: The total number of times the calculation mode of the current vehicle-mounted functional unit switches within a preset time period is counted. In response to the total number of computing mode switching times being less than the switching time threshold, the current computing mode is switched to the coarse-grained computing mode.

7. The method according to claim 1, characterized in that, The method further includes: The nearest obstacle distance, longitudinal acceleration, and lateral acceleration of the vehicle system are obtained. In response to the nearest obstacle distance being less than a distance threshold, or the longitudinal acceleration being greater than a longitudinal acceleration threshold, or the lateral acceleration being greater than a lateral acceleration threshold, it is determined whether the node type of the current routing node is an emergency necessary slow road node; If the node type is the emergency necessary slow road node, the current calculation mode is set to the fine-grained calculation mode, the historical calculation output result is obtained, the current vehicle function unit is controlled to perform calculation processing according to the fine-grained calculation mode and the historical calculation output result to obtain the current calculation output result, and the current calculation output result is output to the routing node corresponding to the next vehicle function unit. The step of obtaining the unit type of the current vehicle-mounted functional unit includes: if the node type is not the emergency necessary slow road node, obtaining the unit type.

8. The method according to claim 1, characterized in that, The method further includes: In response to the unit type being the skippable execution function unit, the historical obstacle detection result of the previous historical frame and the current obstacle detection result of the current frame are obtained; After performing Hungarian matching on the historical obstacle detection results and the current obstacle detection results, the matched historical obstacle detection results and the matched current obstacle detection results are obtained. A stability assessment is performed based on the historical obstacle detection results after matching and the current obstacle detection results after matching to obtain a stability assessment score; In response to the stability assessment score being greater than the stability assessment score threshold, the historical calculation output result is obtained, and the current vehicle functional unit is controlled to output the historical calculation output result to the routing node corresponding to the next vehicle functional unit.

9. The method according to claim 8, characterized in that, The method further includes: In response to the stability evaluation score being less than or equal to the stability evaluation score threshold, the current calculation mode is determined to be a fine-grained calculation mode. The historical calculation output results are obtained, and the current vehicle functional unit is controlled to perform calculation processing based on the fine-grained calculation mode and the historical calculation output results to obtain the current calculation output result. The current calculation output result is then output to the routing node corresponding to the next vehicle functional unit.

10. A vehicle-mounted functional unit computing device, the device being applied to a pre-configured current routing node corresponding to the current vehicle-mounted functional unit; the device: The unit type acquisition module is used to acquire the unit type of the current vehicle-mounted functional unit; the unit type includes a fast / slow road selectable functional unit, a skippable execution functional unit, or a necessary execution functional unit; The evaluation index acquisition module is used to acquire the system load of the vehicle system, the evaluation index of each scenario of the current vehicle-mounted functional unit, and the historical actual execution time of the current vehicle-mounted functional unit in response to the unit type being the fast and slow road selectable functional unit. The execution time calculation module is used to determine the scenario complexity score and scenario timeliness score according to the evaluation indicators of each scenario, and to determine the current fast road execution time and current slow road execution time of the current frame of the current vehicle functional unit according to the system load, the historical actual execution time of each historical frame and the scenario complexity score, and to obtain the current remaining latency budget value. The coarse-grained calculation module is used to respond to the current slow path execution time being greater than the current remaining delay budget value, determine that the current calculation mode of the current vehicle function unit is the coarse-grained calculation mode, obtain the historical calculation output result of the previous vehicle function unit, control the current vehicle function unit to perform calculation processing according to the coarse-grained calculation mode and the historical calculation output result to obtain the current calculation output result of the current vehicle function unit, and output the current calculation output result to the routing node corresponding to the next vehicle function unit.

11. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 9.

Images