Redundant sensing method, apparatus, electronic device, and storage medium

Abstract

To provide a redundant sensing method, an apparatus, an electronic device, and a storage medium.SOLUTION: The redundant sensing method includes a step of determining a first sensing result based on a first sensing application program in a first sensing function cluster, and a step of determining a second sensing result based on a second sensing application program in a second sensing function cluster. The first sensing application program in the first sensing function cluster is software-separated and / or hardware-separated from at least a target second sensing application program.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to driving assistance technology, and in particular to a redundant sensing method, device, electronic device, and storage medium. [Background technology]

[0002] In scenarios such as autonomous driving and driver assistance, sensing systems in vehicle driving control systems are required to provide the minimum sensing results necessary for a minimum risk policy to the vehicle driving control system if a malfunction or malfunction occurs and the driver cannot take over in a timely manner. For example, providing forward road environment sensing results for parking the vehicle in the current lane allows the vehicle to continue operating until parking is controlled, thereby avoiding accidents such as lane departure, collision, and rollover. Related technologies typically build redundant sensing systems using multiple independent electronic control units (ECUs). However, using multiple independent ECUs can significantly limit the utilization of the ECU's hardware resources. Summary of the Invention [Problem to be solved by the invention]

[0003] To address the above-mentioned issues, such as low utilization of hardware resources in electronic control units, embodiments of the present disclosure provide redundant sensing methods, devices, electronic devices, and storage media to improve utilization of hardware resources. [Means for solving the problem]

[0004] A redundant sensing method according to a first aspect of the present disclosure includes a step of determining a first sensing result based on a first sensing application program in a first sensing function cluster, wherein the first sensing result includes a first minimum sensing result, and the first minimum sensing result is a sensing result within a first preset minimum field of view angle range; and a step of determining a second sensing result based on a second sensing application program in a second sensing function cluster, wherein the second sensing result includes at least a second minimum sensing result, and the second minimum sensing result is a sensing result within a second preset minimum field of view angle range, wherein the first sensing application program in the first sensing function cluster is separated by software and / or hardware from at least a target second sensing application program, and the target second sensing application program represents a second sensing application program in the second sensing function cluster for obtaining the second minimum sensing result.

[0005] A redundant sensing device according to a second aspect of the present disclosure includes a first processing module for determining a first sensing result based on a first sensing application program in a first sensing function cluster, wherein the first sensing result includes a first minimum sensing result, the first minimum sensing result being a sensing result within a first preset minimum field of view angle range; and a second processing module for determining a second sensing result based on a second sensing application program in a second sensing function cluster, wherein the second sensing result includes at least a second minimum sensing result, the second minimum sensing result being a sensing result within a second preset minimum field of view angle range, wherein the first sensing application program in the first sensing function cluster is separated by software and / or hardware from at least a target second sensing application program, and the target second sensing application program represents a second sensing application program in the second sensing function cluster for obtaining the second minimum sensing result.

[0006] A computer-readable storage medium according to a third aspect of the present disclosure stores a computer program for executing the redundant sensing method according to any one of the above embodiments of the present disclosure.

[0007] An electronic device according to a fourth aspect of the present disclosure includes a processor and a memory for storing instructions executable by the processor, the processor being used to read and execute the executable instructions from the memory to realize a redundant sensing method described in any of the above embodiments.

[0008] A fifth aspect of the present disclosure provides a computer program product, wherein instructions in the computer program product, when executed by a processor, perform a redundant sensing method according to any of the above embodiments of the present disclosure. [Effects of the Invention]

[0009] Based on the redundant sensing method, device, electronic device, and storage medium according to the above embodiments of the present disclosure, a sensing application program can be divided into two sensing function clusters, and a first sensing result is determined based on a first sensing application program in the first sensing function cluster, and a second sensing result is determined based on a second sensing application program in the second sensing function cluster. Since the first sensing result includes the first minimum sensing result and the second sensing result includes at least the second minimum sensing result, both the first sensing application program and the target second sensing application program in the second sensing function cluster can provide minimum sensing results, redundant sensing is formed, and the first sensing application program and the target second sensing application program are separated by software and / or hardware, so that even if one of them (e.g., the first sensing application program) is unable to provide minimum sensing results due to a failure or other reason, the normal operation of the other (e.g., the target second sensing application program) can be continuously provided through the other to park the vehicle in the current lane, thereby allowing the vehicle to continue operating under control until parking, avoiding accidents such as lane departure, collision, and rollover, and thereby ensuring the safety of vehicle driving. Furthermore, since the first sensing application program and the target second sensing application program are separated by software and / or hardware, the first sensing function cluster and the second sensing function cluster can be realized using the hardware resources of the same electronic control unit or a controller of similar architecture that has a separation function, thereby improving the utilization rate of the hardware resources. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is an exemplary application scenario of the redundant sensing method according to the present disclosure. [Figure 2] 1 is a schematic flowchart of a redundant sensing method according to an exemplary embodiment of the present disclosure. [Figure 3] 10 is a schematic flowchart of a redundant sensing method according to another exemplary embodiment of the present disclosure. [Figure 4] FIG. 2 is a schematic diagram of a software partition according to an exemplary embodiment of the present disclosure. [Figure 5] FIG. 10 is a schematic diagram of a software partition according to another exemplary embodiment of the present disclosure. [Figure 6] FIG. 10 is a schematic diagram of a software partition according to yet another exemplary embodiment of the present disclosure. [Figure 7] 10 is a schematic flowchart of a redundant sensing method according to yet another exemplary embodiment of the present disclosure. [Figure 8] FIG. 1 is an exemplary schematic diagram of a redundant sensing system according to an exemplary embodiment of the present disclosure. [Figure 9] FIG. 10 is a schematic diagram of a software partition and sensing logic according to yet another exemplary embodiment of the present disclosure. [Figure 10] FIG. 2 is a schematic diagram of a sensing range of a first sensing function cluster according to an exemplary embodiment of the present disclosure. [Figure 11] FIG. 10 is a schematic diagram of a sensing range of a second sensing function cluster according to an exemplary embodiment of the present disclosure. [Figure 12] 10 is a schematic flowchart of a redundant sensing method according to yet another exemplary embodiment of the present disclosure. [Figure 13] 10 is a schematic flowchart of a redundant sensing method according to yet another exemplary embodiment of the present disclosure. [Figure 14] FIG. 1 is a schematic diagram of a principle for determining availability of sensing results according to an exemplary embodiment of the present disclosure. [Figure 15] FIG. 10 is a schematic diagram of a principle for determining availability of sensing results according to another exemplary embodiment of the present disclosure. [Figure 16] FIG. 2 is a structural schematic diagram of a redundant sensing device according to an exemplary embodiment of the present disclosure. [Figure 17] FIG. 10 is a structural schematic diagram of a redundant sensing device according to another exemplary embodiment of the present disclosure. [Figure 18] FIG. 10 is a structural schematic diagram of a redundant sensing device according to yet another exemplary embodiment of the present disclosure. [Figure 19] FIG. 10 is a structural schematic diagram of a redundant sensing device according to yet another exemplary embodiment of the present disclosure. [Figure 20] FIG. 10 is a structural schematic diagram of a redundant sensing device according to yet another exemplary embodiment of the present disclosure. [Figure 21] 1 is a structural diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] In order to explain the present disclosure, exemplary embodiments of the present disclosure will be described in detail below with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, not all of the embodiments, and the present disclosure is not limited to the exemplary embodiments. Unless otherwise specifically stated, the relative arrangements of components and steps, formulas and numerical values ​​described in these embodiments do not limit the scope of the present disclosure.

[0012] <Summary of this disclosure> In the process of realizing the present disclosure, the inventors discovered the following: In scenarios such as autonomous driving and driver assistance, if a sensing system in a vehicle driving control system fails or is unable to perform normal sensing functions, unless the driver can take over in a timely manner, the sensing system needs to additionally provide the vehicle driving control system with the minimum sensing results required for a minimum risk policy, such as providing forward road environment sensing results for parking the vehicle in the current lane, so that the vehicle can continue to operate under control until parking is completed, thereby avoiding accidents such as lane departure, collision, or rollover. Related technologies generally use multiple independent electronic control units (ECUs) to build redundant sensing systems. However, using multiple independent ECUs can significantly limit the utilization rate of the ECU's hardware resources and the flexibility of software deployment.

[0013] <Illustrative Overview> FIG. 1 illustrates an exemplary application scenario of the redundant sensing method according to the present disclosure. As illustrated in FIG. 1, two sensing function clusters may be installed, which may include a first sensing function cluster and a second sensing function cluster. A first sensing application program in the first sensing function cluster may be used to perform environmental sensing based on images collected by a first preset minimum viewing angle camera 1. A second sensing application program in the second sensing function cluster may be used to perform environmental sensing based on images collected by a second preset minimum viewing angle camera 2 and images collected by a camera with another viewpoint. For example, the first preset minimum viewing angle camera 1 may be a forward-view camera 1, and the second preset minimum viewing angle camera 2 may be a forward-view camera 2. According to the redundant sensing method of the present disclosure, a first sensing result may be determined based on the first sensing application program in the first sensing function cluster. The first sensing result includes a first minimum sensing result, which is a sensing result within a first preset minimum field of view angle range, such as a sensing result within the field of view angle range of the forward-view camera 1. The second sensing result can be determined based on a second sensing application program in a second sensing function cluster. The second sensing result includes at least a second minimum sensing result, which is a sensing result within a second preset minimum field of view angle range, such as a sensing result within the field of view angle range of the forward-view camera 2. For example, the second sensing application program includes a target second sensing application program, which represents a second sensing application program for obtaining the second minimum sensing result in the second sensing function cluster. In actual use, the second sensing result may also include sensing results within field of view angle ranges of other viewpoints, such as a sensing result within a peripheral viewpoint field of view angle range or a rear viewpoint field of view angle range.Decision-making and control can be performed based on the first sensing result and the second sensing result, thereby controlling the vehicle's travel. The first sensing application program in the first sensing function cluster is separated by software and / or hardware from at least the target second sensing application program, ensuring that the first sensing application program and the target second sensing application program do not affect each other. For example, if the first sensing application program is unable to obtain the first minimum sensing result due to a malfunction of the first preset minimum field-of-view camera 1 and an execution error in the first sensing application program, the acquisition of the second minimum sensing result will not be affected. As a result, the first minimum sensing result and the second minimum sensing result are redundant sensing results for each other, the first preset minimum viewing angle camera 1 and the second preset minimum viewing angle camera 2 are redundant cameras for each other, and the first sensing application program and the target second sensing application program are redundant sensing application programs for each other, thereby forming a redundant sensing system. If one of the two (e.g., the first sensing application program) is unable to provide a minimum sensing result due to a failure in the corresponding software or hardware, the normal operation of the other (e.g., the target second sensing application program) will not be affected, and the other can continue to provide a minimum sensing result (e.g., a forward road environment sensing result) for the vehicle to park in the current lane, so that the vehicle can continue to operate under control until parking, avoiding accidents such as lane departure, collision, and rollover, thereby ensuring the safety of vehicle driving.Furthermore, since the first sensing application program and the target second sensing application program are separated by software and / or hardware, the first sensing function cluster and the second sensing function cluster can be realized using the hardware resources of the same system on chip (abbreviated as SoC), electronic control unit, or controller of similar architecture that has separation functions, thereby improving the utilization rate of hardware resources and improving the flexibility of software deployment.

[0014] Exemplary Methods 2 is a schematic flowchart of a redundant sensing method according to an exemplary embodiment of the present disclosure. This embodiment is particularly applicable to electronic devices such as in-vehicle computing platforms. As shown in FIG. 2, the method according to the embodiment of the present disclosure may include the following steps 201 to 202:

[0015] In step 201, a first sensing result is determined based on a first sensing application program in a first sensing function cluster.

[0016] Here, the first sensing result may include a first minimum sensing result, and the first minimum sensing result may be a sensing result within a first preset minimum viewing angle range.

[0017] In some alternative embodiments, the first preset minimum viewing angle range may be set according to the needs of the vehicle to determine the minimum risk policy, for example, the first preset minimum viewing angle range may be a preset range of forward-view viewing angles.

[0018] In some alternative embodiments, an image of a first preset minimum viewing angle range can be collected by a first preset viewing angle camera, and a sensing process can be performed on the image of the first preset minimum viewing angle range based on a first sensing application program in a first sensing function cluster to obtain a first sensing result.

[0019] In some alternative embodiments, the first sensing application program may include one or more sensing application programs. For example, the first sensing application program may include a first forward-view sensing application program. Alternatively, the first sensing application program may include a first forward-view sensing application program and a first post-processing application program. The first forward-view sensing application program may be used to perform sensing processing on sensor data (e.g., image and / or radar point cloud data) within a first predetermined minimum field of view angle range to obtain a first forward-view sensing result, and the first post-processing application program may be used to perform post-processing on the first forward-view sensing result to obtain a first sensing result. In this way, the first minimum sensing result may be a sensing result within the first forward-view field of view angle range.

[0020] In some alternative embodiments, the first sensing result may include status information of static objects such as sensed lane markings and road shoulders, as well as status information of dynamic objects such as other surrounding vehicles, pedestrians, and riders. The status information of static objects may include, for example, the type of lane marking, the start point of the lane marking, the end point of the lane marking, the curve equation of the lane marking, the type of shoulder, the start point of the shoulder, the end point of the shoulder, and the curve equation of the shoulder. The status information of dynamic objects may include, for example, the classification, position, speed, acceleration, and the like of the dynamic object. In actual use, the status information of dynamic objects may include the availability status of the information, such as a flag bit indicating failure or unavailable information.

[0021] In some embodiments, the coverage range of the first minimum sensing result may include a lateral range of the lane in which the host vehicle is located, adjacent lanes (e.g., including the lanes to the left and right of the lane in which the host vehicle is located), and adjacent lanes (i.e., adjacent lanes of adjacent lanes; for example, adjacent lanes may include the left lane of the lane in which the host vehicle is located and the right lane of the lane in which the host vehicle is located), and a longitudinal range of a preset distance. The specific range may be set according to the actual usage situation of the vehicle; for example, the sensing range of the first minimum sensing result may be determined according to the maximum driving speed of the vehicle, vehicle control performance, etc.

[0022] In step 202, a second sensing result is determined based on a second sensing application program in a second sensing function cluster.

[0023] Here, the second sensing result may include at least a second minimum sensing result, and the second minimum sensing result may be a sensing result within a second preset minimum viewing angle range.

[0024] Here, the first sensing application program in the first sensing function cluster is separated by software and / or hardware from at least the target second sensing application program, which may represent a second sensing application program in the second sensing function cluster for obtaining a second minimum sensing result.

[0025] In some alternative embodiments, the first sensing function cluster and the second sensing function cluster are two groups obtained by logically dividing the sensing application program, and the first sensing function cluster is a minimum redundant sensing function cluster of the second sensing function cluster, i.e., the first sensing function cluster can provide a minimum sensing result (i.e., the first minimum sensing result), and the second sensing function cluster can also provide a minimum sensing result (i.e., the second minimum sensing result), thereby realizing minimum redundant sensing and ensuring the provision of a minimum sensing result required for a minimum risk policy of the vehicle at a minimum cost.

[0026] In some alternative embodiments, the number of second sensing application programs in the second sensing function cluster may be one or more. For example, the second sensing function cluster may include at least a second target sensing application program, which may be used to obtain a second minimum sensing result. Specifically, the second target sensing application program may determine the sensing result within the second preset minimum field of view angle range as the second minimum sensing result based on sensor data within the second preset minimum field of view angle range. In actual use, the second sensing function cluster may include, in addition to the second target sensing application program, other second sensing application programs, such as at least one sensing application program with another perspective. The other perspectives may be set according to actual needs. For example, the other perspectives may include a peripheral perspective, a rear perspective, an omnidirectional perspective, etc., so that the second sensing function cluster can sense a wide range around the host vehicle and obtain the second sensing result.

[0027] In some alternative embodiments, the target second sensing application program may include a second forward-looking sensing application program, or the target second sensing application program may include a second forward-looking sensing application program and a second post-processing application program.

[0028] In some alternative embodiments, first sensing application programs in a first sensing function cluster may have a certain data dependency relationship. Second sensing application programs in a second sensing function cluster may have a certain data dependency relationship. The first sensing application program and the second sensing application program may have a certain data dependency relationship. Specifically, the data dependency relationship between each sensing application program (which may be abbreviated as application program) can be set according to the function of each sensing application program.

[0029] In some alternative embodiments, the separation between the first sensing application program and the target second sensing application program may be software separation, hardware separation, or a combination of software and hardware separation, thereby ensuring that the first sensing application program and the target second sensing application program do not affect each other and enabling at least one of the first minimum sensing result and the second minimum sensing result to be obtained. For example, if the first sensing application program is unable to obtain an available first minimum sensing result (e.g., the first sensing application program is unable to obtain the first minimum sensing result due to a malfunction or the like, or obtains an incorrect first minimum sensing result), the second minimum sensing result obtained by the target second sensing application program can provide a minimum-risk policy for vehicle driving, ensuring that the vehicle drives to parking in a controlled manner and avoiding accidents such as lane departure, collision, or rollover. Similarly, if the target second sensing application program is unable to provide a usable second minimum sensing result (for example, if the target second sensing application program is unable to obtain the second minimum sensing result due to a malfunction or the like, or obtains an incorrect second minimum sensing result), a minimum risk policy can be provided for vehicle operation using the first minimum sensing result obtained by the first sensing application program. Here, isolation by software may mean that an operating system (abbreviated as OS) or the like allocates shared resources such as a processor, memory, or bus to each application program that requires isolation, thereby preventing any application program from being affected by or affecting other application programs when it is running.Hardware isolation may refer to binding hardware resources to application programs that require isolation, so that when an application program is running, it is not affected by or affects other application programs. For example, hardware isolation can be achieved by performing memory management through a memory management unit (MMU) and binding different memory resources to application programs that require isolation.

[0030] In some alternative embodiments, software isolation can be achieved by software partitions or any other possible implementation, which may mean that application programs requiring isolation through an operating system can be assigned different partitions of memory for storing the application programs, and the application programs stored in different partitions can be assigned different processor resources for executing the application programs, so that the application programs requiring isolation do not affect each other during execution.

[0031] In some alternative embodiments, hardware isolation can be achieved by binding hardware resources or other forms, which may mean binding hardware resources corresponding to application programs that require isolation so that different application programs are executed by different bound hardware resources, so as to achieve the goal that the application programs are independent of each other and do not affect each other.

[0032] In some alternative embodiments, the sensing functions of the first sensing function cluster and the second sensing function cluster are heterogeneous, i.e., the first sensing function cluster and the second sensing function cluster satisfy at least one of the following: 1) sensing algorithm parameters are inconsistent; 2) sensing algorithm input sources are inconsistent; and 3) sensing algorithm architectures are inconsistent. This heterogeneity configuration can prevent the two sensing function clusters from simultaneously failing or failing to operate normally. Inconsistent sensing algorithm parameters may refer to different specific parameters when the sensing algorithm architectures are the same, e.g., sensing models with the same network architecture have different network parameters. Inconsistent sensing algorithm input sources refer to sensor data input to the first sensing function cluster and the second sensing function cluster being data collected by different sensors. For example, the input source of the first sensing function cluster is images collected by a forward-view camera 1. The input sources of the second sensing function cluster include images collected by the forward viewpoint camera 2 and may also include images collected by cameras with other viewpoints. A mismatch in the sensing algorithm architecture may mean that the first sensing function cluster and the second sensing function cluster employ different sensing algorithms, e.g., different sensing models.

[0033] In embodiments of the present disclosure, sensors for acquiring input sources are not limited to cameras, but may include other types of sensors, such as, but not limited to, lidar, millimeter wave radar, ultrasonic radar, and the like.

[0034] In some alternative embodiments, the second preset minimum viewing angle range and the first preset minimum viewing angle range may be the viewing angle ranges of two sensors, for example, the viewing angle ranges of two front-view cameras, Front View Camera 1 and Front View Camera 2. The two sensors may have the same or different parameters, for example, the two front-view cameras may have the same or different viewing angles (e.g., 90 degrees and 120 degrees) and other parameters, and therefore the second preset minimum viewing angle range and the first preset minimum viewing angle range may be the same or may have a certain difference.

[0035] The order of steps 201 and 202 does not matter.

[0036] The redundant sensing method of this embodiment can divide a sensing application program into two sensing function clusters, and determine a first sensing result based on a first sensing application program in the first sensing function cluster, and determine a second sensing result based on a second sensing application program in the second sensing function cluster. Since the first sensing result includes the first minimum sensing result and the second sensing result includes at least the second minimum sensing result, both the first sensing application program and the target second sensing application program in the second sensing function cluster can provide minimum sensing results, redundant sensing is formed, and the first sensing application program and the target second sensing application program are separated by software and / or hardware, so that even if one of them (e.g., the first sensing application program) is unable to provide minimum sensing results due to a failure or other reason, the normal operation of the other (e.g., the target second sensing application program) can be continuously provided through the other to park the vehicle in the current lane, thereby allowing the vehicle to continue operating under control until parking, avoiding accidents such as lane departure, collision, and rollover, and thereby ensuring the safety of vehicle driving. Furthermore, since the first sensing application program and the target second sensing application program are separated by software and / or hardware, the first sensing function cluster and the second sensing function cluster can be realized using the hardware resources of the same system-on-chip, electronic control unit, or controller of similar architecture that has separation functions, thereby improving the utilization rate of hardware resources and improving the flexibility of software deployment.

[0037] FIG. 3 is a schematic flowchart of a redundant sensing method according to another exemplary embodiment of the present disclosure.

[0038] In some alternative embodiments, as shown in FIG. 3, step 201 of determining a first sensing result based on a first sensing application program in a first sensing function cluster may specifically include the following steps 2011 to 2013.

[0039] In step 2011, a first sensing application program in a first sensing function cluster is obtained from a first software partition.

[0040] In some alternative embodiments, the method of the presently disclosed embodiment can be implemented on a platform capable of implementing software partitions, thereby enabling separation of the first sensing application program and the target second sensing application program through the software partitions. The partition in which the first sensing application program resides is referred to as the first software partition. During the sensing process, the first sensing application program in the first sensing function cluster can be obtained from the first software partition. The target second sensing application program is deployed to a software partition different from the first software partition.

[0041] In step 2012, a first processor resource for executing the first sensing application program is determined from the shared resource.

[0042] Here, the shared resources may include processor resources necessary for implementing the sensing application program, such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The shared resources can be shared and used by the first sensing function cluster and the second sensing function cluster. Note that the same processor resource is used by only one software partition's sensing application program at a time. That is, at any given time, the first sensing application program and the target second sensing application program run on different processor resources to ensure their isolation.

[0043] In some alternative embodiments, the first processor resource for executing the first sensing application program can be determined according to the availability of each processor resource in the shared resource. For example, the shared resource may include multiple systems-on-chips SoC1, SoC2, ..., SoCn, and any available chip among them may be determined as the first processor resource for executing the first sensing application program. In actual use, the shared resource may include one or more different processing modules (e.g., different processor cores) in the chip. That is, the shared resource may be multiple systems-on-chips with relatively coarse granularity or multiple processing modules with relatively fine granularity. The specific form and granularity of the shared resource are not limited.

[0044] In step 2013, a first sensing application program is executed based on the first processor resource, and a first sensing result is obtained.

[0045] Here, after determining a first processor resource for executing the first sensing application program, the first sensing application program is executed based on the first processor resource, and sensing processing is performed on sensor data of a first preset minimum field of view range collected by a first preset field of view sensor (e.g., a camera and / or radar), thereby obtaining a first sensing result.

[0046] This embodiment facilitates separation of a first sensing application program from a target second sensing application program by a software partition by deploying the first sensing application program in a first software partition; for example, the target second sensing application program can be deployed in a software partition different from the first software partition, and the first sensing application program and the second sensing application program can be made to share processor resources, thereby improving the utilization rate of hardware resources.

[0047] In some alternative embodiments, as shown in FIG. 3, step 202 of determining a second sensing result based on a second sensing application program in a second sensing function cluster may include steps 2021 to 2024.

[0048] In step 2021, obtain a target second sensing application program from a second software partition.

[0049] Here, the second software partition and the first software partition are different software partitions, and the target second sensing application program can be deployed to the second software partition so that the target second sensing application program can be obtained from the second software partition during the sensing process.

[0050] In step 2022, a second processor resource for executing the target second sensing application program is determined from the shared resource.

[0051] According to the occupancy status of the shared resource, a second processor resource for executing the target second sensing application program can be determined from the free resources of the shared resource. At this time, since the first processor resource has already been allocated to the first sensing application program, the second processor resource and the first processor resource are different processor resources in the shared resource. However, the first processor resource and the second processor resource can be shared among all software partitions, and in the previous or subsequent execution process, either processor resource can be used to execute the sensing application program of any software partition. That is, each time a sensing application program is executed, resource allocation can be performed according to the real-time occupancy status of the shared resource.

[0052] In step 2023, for other second sensing application programs other than the target second sensing application program in the second sensing function cluster, the other second sensing application programs are obtained from software partitions corresponding to the other second sensing application programs, and processor resources corresponding to the other second sensing application programs are determined based on the software partitions corresponding to the other second sensing application programs.

[0053] Here, the processor resource corresponding to the other second sensing application program is any one of the first processor resource, the second processor resource, and the third processor resource.

[0054] In some alternative embodiments, the other second sensing application programs may include, for example, a rear-view sensing application program, a peripheral-view sensing application program, an omnidirectional-view sensing application program, etc., and the target second sensing application program and the other second sensing application programs can cooperate to achieve wide-area sensing around the vehicle.

[0055] In some alternative embodiments, the other second sensing application program may be deployed in any software partition, for example, in the first software partition or the second software partition, or in a third software partition different from the first software partition and the second software partition. If there are multiple other second sensing application programs, each other second sensing application program may be deployed in multiple software partitions so as to be distributed in an arbitrary manner. For example, a portion of the other second sensing application program may be deployed in the first software partition, and another portion of the other second sensing application program may be deployed in the second software partition. The partition deployment form of the other second sensing application program is not limited.

[0056] In some alternative embodiments, different processor resources may be assigned to different software partitions during each sensing process. For example, if another second sensing application program is deployed in a first software partition, the processor resource corresponding to the other second sensing application program may be determined to be the first processor resource. If another second sensing application program is deployed in a second software partition, the processor resource corresponding to the other second sensing application program may be determined to be the second processor resource. If another second sensing application program is deployed in a third software partition, the processor resource corresponding to the other second sensing application program may be determined to be the third processor resource in the shared resources.

[0057] In step 2024, a target second sensing application program is executed based on the second processor resource, and another second sensing application program is executed based on the processor resource corresponding to the other second sensing application program, and a second sensing result is obtained.

[0058] Here, the second processor resource and the first processor resource are different resources in the shared resource. After determining the processor resources corresponding to the target second sensing application program and the other second sensing application program, respectively, the target second sensing application program is executed based on the second processor resource, and the other second sensing application program is executed based on the processor resources corresponding to the other second sensing application program, thereby performing sensing processing on the sensor data input to the second sensing function cluster and obtaining a second sensing result. A data dependency may exist between the target second sensing application program and the other second sensing application program in the second sensing function cluster, and thus the target second sensing application program and the other second sensing application program may be executed based on the corresponding data dependency to achieve accurate streaming of the data stream.

[0059] In some alternative embodiments, Figure 4 is a schematic diagram of a software partition according to an exemplary embodiment of the present disclosure. As shown in Figure 4, another second sensing application program in the second sensing function cluster may be deployed in a second software partition.

[0060] In some alternative embodiments, Figure 5 is a schematic diagram of a software partition according to another exemplary embodiment of the present disclosure. As shown in Figure 5, another second sensing application program in a second sensing function cluster may be deployed in the first software partition.

[0061] 6 is a schematic diagram of software partitions according to another exemplary embodiment of the present disclosure. As shown in FIG. 6, for another second sensing application program in a second sensing function cluster, a part of the other second sensing application program may be deployed in a first software partition, and another part of the other second sensing application program may be deployed in a second software partition.

[0062] In this embodiment, the target second sensing application program is deployed in a second software partition, thereby separating it from the first software partition in which the first sensing application program resides. During the sensing process, different processor resources are allocated to the sensing application programs in different software partitions, so that the acquisition of the first minimum sensing results and the second minimum sensing results is separated from each other and does not affect each other, effectively realizing minimal redundant sensing and allowing resources to be shared with the first software partition, thereby improving resource utilization.

[0063] FIG. 7 is a schematic flowchart of a redundant sensing method according to yet another exemplary embodiment of the present disclosure.

[0064] In some alternative embodiments, the first sensing application program may include a first forward-looking sensing application program and a first post-processing application program. The second sensing application program may include a second forward-looking sensing application program, at least one other-viewpoint sensing application program, and a second post-processing application program. The first forward-looking sensing application program may correspond to a first software partition, and the second forward-looking sensing application program and the second post-processing application program may correspond to a second software partition. The first post-processing application program may correspond to a third software partition. Any other-viewpoint sensing application program may correspond to any of the first, second, third, or fourth software partitions.

[0065] Here, the first forward-view sensing application program may be used to perform sensing processing on sensor data (e.g., images and / or radar point cloud data) within a first preset minimum field of view angle range to obtain first forward-view sensing results. The first forward-view sensing results may include sensed object information in the sensor data within the forward-view field of view angle range, such as a detection frame for each object in the image, an object type corresponding to the detection frame, and semantic segmentation results. The first post-processing application program may be used to perform post-processing on the first forward-view sensing results to obtain first sensing results. The post-processing may include dynamic sensing post-processing, static fusion post-processing, and the like. For example, post-processing may be performed based on object information for each object included in the first forward-view sensing results to obtain state information for dynamic objects, state information for static objects, and the like. The second forward-view sensing application program may be used to perform sensing processing on sensor data within a second preset minimum field of view angle range to obtain second forward-view sensing results. The sensing application program for another viewpoint can be used to perform sensing processing on sensor data within a range corresponding to the other viewpoint to obtain sensing results for the other viewpoint. The second post-processing application program can be used to fuse and post-process the sensing results for each viewpoint to obtain wide-range sensing results as a second sensing result. Alternatively, state information of dynamic objects and state information of static objects can be obtained by performing dynamic sensing post-processing and static fusion post-processing on objects within a wide range.

[0066] In some alternative embodiments, the second forward-looking application program and the second post-processing application program may be determined as target second sensing application programs and may be separate from the first sensing application program. The first forward-looking sensing application program may be deployed in a first software partition, the second forward-looking sensing application program and the second post-processing application program may be deployed in a second software partition, and the first post-processing application program may be deployed in a third software partition. Any other perspective sensing application program may be deployed in any software partition. For example, it may be deployed in the first software partition, the second software partition, the third software partition, or even a fourth software partition.

[0067] In some alternative embodiments, as shown in FIG. 7, step 201 of determining a first sensing result based on a first sensing application program in a first sensing function cluster may include steps 201a to 201d.

[0068] In step 201a, a first processor resource for executing a first forward viewpoint sensing application program from the shared resource is determined based on the first software partition.

[0069] In step 201b, determine a third processor resource for executing the first post-processing application program from the shared resource based on the third software partition.

[0070] The order of steps 201a and 201b does not matter.

[0071] In step 201c, a first forward viewpoint sensing application program is executed based on the first processor resource, and a first forward viewpoint sensing result is obtained.

[0072] In step 201d, a first post-processing application program is executed based on the third processor resource to perform post-processing on the first forward viewpoint sensing result to obtain the first sensing result.

[0073] The resource allocation and execution in steps 201a to 201d are the same as in the above-described embodiment, so a duplicated explanation will be omitted here.

[0074] In some alternative embodiments, determining 202 a second sensing result based on a second sensing application program in a second sensing function cluster may include steps 202a-202e.

[0075] In step 202a, determine second processor resources for executing a second forward viewpoint sensing application program and a second post-processing application program from the shared resources based on the second software partition.

[0076] In step 202b, determine corresponding processor resources for executing each other-perspective sensing application program based on the software partitions corresponding to each other-perspective sensing application program.

[0077] The order of steps 202a and 202b does not matter.

[0078] In step 202c, a second forward viewpoint sensing application program is executed based on the second processor resource, and a second forward viewpoint sensing result is obtained.

[0079] In step 202d, the sensing application program of each other viewpoint is executed based on the processor resources corresponding to the sensing application program of each other viewpoint, and other viewpoint sensing results corresponding to each other viewpoint are obtained.

[0080] The order of steps 202c and 202d does not matter.

[0081] In step 202e, based on the second processor resource, a second post-processing application program is executed to perform post-processing on the second forward viewpoint sensing result and the other viewpoint sensing results corresponding to each other viewpoint, and obtain the second sensing result.

[0082] The resource allocation and program execution in steps 202a to 202e are the same as in the above-described embodiment, so a duplicated explanation will be omitted here.

[0083] In this embodiment, the first forward-view sensing application program, the first post-processing application program, and the target second sensing application program (which may include a second forward-view sensing application program and a second post-processing application program) are deployed in different software partitions, respectively, to facilitate the establishment of data dependency between the first sensing function cluster and the second sensing function cluster. For example, the output result of the second forward-view sensing application program is transmitted to the first post-processing application program, and the output result of the first forward-view sensing application program is transmitted to the second post-processing application program. For example, if the second post-processing application program and the first forward viewpoint sensing application program fail, the first post-processing application program can perform post-processing on the second forward viewpoint sensing result to obtain a minimum sensing result. Similarly, if the second forward viewpoint sensing application program and the first post-processing application program fail, the second post-processing application program can perform post-processing on the first forward viewpoint sensing result to obtain a minimum sensing result. This further improves the effectiveness and reliability of redundant sensing.

[0084] In some alternative embodiments, the second forward-looking sensing application program and the second post-processing application program may be deployed in different software partitions, for example, the second forward-looking sensing application program may be deployed in a second software partition and the second post-processing application program may be deployed in a fourth software partition. In this way, the first forward-looking sensing application program, the first post-processing application program, the second forward-looking sensing application program, and the second post-processing application program may be separated from each other, further improving the separation granularity and thereby further improving the effectiveness and reliability of minimally redundant sensing.

[0085] In some alternative embodiments, step 201d of executing a first post-processing application program based on a third processor resource to perform post-processing on at least the first forward viewpoint sensing result to obtain the first sensing result includes: The method may include a step of executing a first post-processing application program based on a third processor resource, performing fusion processing on the first forward viewpoint sensing result and the second forward viewpoint sensing result, and obtaining the first sensing result.

[0086] In this case, during the sensing process, both the first forward viewpoint sensing result and the second forward viewpoint sensing result can be transmitted to a third processor resource, which then executes a first post-processing application program to perform fusion processing on the first forward viewpoint sensing result and the second forward viewpoint sensing result to obtain a first sensing result. The specific fusion method is not limited. For example, if the second forward viewpoint sensing result covers a relatively large range of forward viewpoint field of view angles and there is an object not sensed in the first forward viewpoint sensing result, the first post-processing application program can fuse the object in the second forward viewpoint sensing result with the first sensing result to obtain a more effective first sensing result.

[0087] Step 202e of executing a second post-processing application program based on a second processor resource, and performing post-processing on at least the second forward viewpoint sensing result and the other viewpoint sensing results corresponding to each of the other viewpoints to obtain a second sensing result includes: The method may include a step of executing a second post-processing application program based on a second processor resource, performing post-processing on the first forward viewpoint sensing result, the second forward viewpoint sensing result, and other viewpoint sensing results corresponding to each of the other viewpoints, and obtaining a second sensing result.

[0088] Here, the first forward viewpoint sensing result can be transmitted to a second processor resource to participate in obtaining the second sensing result, thereby further improving the validity of the second sensing result.

[0089] This embodiment allows the first forward viewpoint sensing result to participate in determining the second sensing result, and the second forward viewpoint sensing result to participate in determining the first sensing result, thereby improving the validity and reliability of the first sensing result and the second sensing result, while providing more redundant sensing paths for minimum redundant sensing, thereby improving the validity and reliability of minimum redundant sensing.

[0090] In some alternative embodiments, step 202e of executing a second post-processing application program based on a second processor resource, and performing post-processing on at least the second forward viewpoint sensing result and the other viewpoint sensing results corresponding to each of the other viewpoints to obtain the second sensing result includes: The method may include a step of executing a second post-processing application program based on a second processor resource, and performing post-processing on the first forward viewpoint sensing result, the second forward viewpoint sensing result, other viewpoint sensing results corresponding to each of the other viewpoints, and the first sensing result to obtain a second sensing result.

[0091] Here, by further including the first sensing result in determining the second sensing result, the reliability of the second sensing result can be further improved.

[0092] 8 is an exemplary schematic diagram of a redundant sensing system according to one exemplary embodiment of the present disclosure. As shown in FIG. 8, the first sensing function cluster may include a sensing application program for a single camera with a forward viewpoint 1 (i.e., a first forward viewpoint sensing application program), and post-processing fusion application programs for the forward viewpoint 1 and the forward viewpoint 2 (i.e., a first post-processing application program). The second sensing function cluster may include a sensing application program for a single camera with a forward viewpoint 2 (i.e., a second forward viewpoint sensing application program), a sensing application program for a peripheral viewpoint single camera *N, a sensing application program for a rearward viewpoint single camera, a sensing application program for an omnidirectional fisheye single camera *M, a wide-range sensing application program, a dynamic sensing post-processing application program, and a static fusion post-processing application program. Here, the forward viewpoint 1 single camera and the forward viewpoint 2 single camera refer to a single forward viewpoint camera, the rearward viewpoint single camera refers to a single rearward viewpoint camera, the peripheral viewpoint single camera *N refers to a single camera with N different viewpoints used to realize peripheral viewpoint sensing, and the omnidirectional fisheye single camera *M refers to a fisheye camera with M different viewpoints used to realize omnidirectional viewpoint sensing. The sensing application program for each viewpoint can obtain sensing results corresponding to that viewpoint. For example, the peripheral viewpoint single camera *N sensing application program can be used to perform peripheral viewpoint sensing based on images collected by the N peripheral viewpoint cameras and obtain peripheral viewpoint sensing results. The wide-range sensing application program can combine the sensing results of the sensing application programs for each viewpoint to obtain wide-range sensing results around the host vehicle. The dynamic sensing post-processing application program can be used to perform post-processing on dynamic objects in the wide-range sensing results and obtain status information of the dynamic objects.The static fusion post-processing application program can be used to perform fusion post-processing on static objects in the wide-range sensing results and obtain status information of the static objects. The status information of the wide-range dynamic objects and the status information of the static objects are used as the second sensing results. The forward view 1 and forward view 2 post-processing fusion application programs can be used to perform fusion, dynamic sensing post-processing, static fusion post-processing, etc. on the sensing results of forward view 1 and forward view 2 to obtain the first sensing results.

[0093] Furthermore, when each application program is executed by the corresponding processor resource, the execution of each application program can be controlled according to the data dependency between the application programs, thereby effectively obtaining the first sensing result and the second sensing result.

[0094] In some alternative embodiments, FIG. 9 is a schematic diagram of software partitions and sensing logic according to another exemplary embodiment of the present disclosure. As shown in FIG. 9, a sensing application program from another perspective corresponds to a fourth software partition. The first processor resource may be, for example, SoC1 in the shared resources, the third processor resource may be, for example, SoC2 in the shared resources, the second processor resource may be, for example, SoC3, and the processor resource corresponding to the sensing application program from another perspective may be, for example, SoC4. Arrows in the figure indicate the direction of data flow during the sensing process. Dashed arrows indicate selectable settings, which can be configured according to actual needs. For example, a first forward-view sensing result obtained by running a first forward-view sensing application program may be transmitted to a second post-processing application program, which then participates in post-processing of wide-area sensing to obtain a second sensing result. The second forward viewpoint sensing result obtained by running the second forward viewpoint sensing application program can be transmitted to the first post-processing application program, whereby the second forward viewpoint sensing result is fused with the first forward viewpoint sensing result and post-processed to obtain the first sensing result. The first sensing result obtained by running the first post-processing application program can also be transmitted to the second post-processing application program to participate in post-processing of wide-range sensing.

[0095] In some alternative embodiments, Fig. 10 is a schematic diagram of a sensing range of a first sensing function cluster according to an exemplary embodiment of the present disclosure. Fig. 11 is a schematic diagram of a sensing range of a second sensing function cluster according to an exemplary embodiment of the present disclosure. As shown in Fig. 10, the sensing range of the first sensing function cluster is an area within a certain range of the forward viewpoint field of view angle of the host vehicle. As shown in Fig. 11, the sensing range of the second sensing function cluster is an area within a wide range around the host vehicle.

[0096] FIG. 12 is a schematic flowchart of a redundant sensing method according to yet another exemplary embodiment of the present disclosure.

[0097] In some alternative embodiments, as shown in FIG. 12, step 201 of determining a first sensing result based on a first sensing application program in a first sensing function cluster may include steps 201A-201B.

[0098] In step 201A, a first processor resource bound to a first sensing application program is determined.

[0099] Here, a processor resource may be bound to each sensing application program in advance, i.e., a binding relationship (also called a correspondence relationship) between the sensing application program and the processor resource may be established, and each time the same sensing application program is executed, it is executed by the processor resource corresponding to that sensing application program, thereby achieving separation between the first sensing application program and the target second sensing application program through hardware separation. Therefore, based on the binding relationship between the sensing application program and the processor resource, the first processor resource bound to correspond to the first sensing application program can be determined.

[0100] In step 201B, a first sensing application program is executed based on a first processor resource, and a first sensing result is obtained.

[0101] By binding the sensing application program and processor resources, this embodiment facilitates the separation of the first sensing application program and the target second sensing application program through hardware separation, ensuring minimal redundant sensing and improving resource utilization.

[0102] In some alternative embodiments, as shown in FIG. 12, step 202 of determining a second sensing result based on a second sensing application program in a second sensing function cluster may include steps 202A-202B.

[0103] In step 202A, a second processor resource bound corresponding to the target second sensing application program and a processor resource bound corresponding to another second sensing application program other than the target second sensing application program in the second sensing function cluster are determined.

[0104] In step 202B, a target second sensing application program is executed based on the second processor resource, and another second sensing application program is executed based on the processor resource bound to the other second sensing application program to obtain a second sensing result.

[0105] Here, the second processor resource and the first processor resource are different resources. For example, the second processor resource and the first processor resource may be different ECUs or different processor resources on the same SoC.

[0106] The determination of the bound resources and the execution of the application program in steps 202A and 202B are the same as in the above-described embodiment, and therefore a duplicated description will be omitted here.

[0107] In this embodiment, the target second sensing application program is bound to the second processor resource, and the second processor resource and the first processor resource are isolated by hardware, thereby realizing hardware isolation between the target second sensing application program and the first sensing application program. As a result, in the sensing process, the processes of obtaining the first minimum sensing result and the second minimum sensing result are isolated from each other and do not affect each other. This ensures the effectiveness and reliability of minimum redundant sensing, and minimum redundant sensing can be realized by the same ECU or a controller with a similar architecture, thereby improving resource utilization.

[0108] In some alternative embodiments, the isolation between the first sensing application program and the target second sensing application program can be further achieved by combining software partitions and hardware isolation, for example, by deploying the first sensing application program and the target second sensing application program in different software partitions and binding different processor resources to the different software partitions, thereby executing the application program of each software partition using the bound processor resource to obtain the first sensing result and the second sensing result.

[0109] FIG. 13 is a schematic flowchart of a redundant sensing method according to yet another exemplary embodiment of the present disclosure.

[0110] In some alternative embodiments, as shown in FIG. 13, the method of this embodiment may include steps 301 to 304.

[0111] In step 301, a first availability status of a first sensing result is determined.

[0112] Here, an availability status detection subroutine may be set in the first sensing application program to determine the first availability status of the first sensing result. The first availability status of the first sensing result may be determined in other ways, and is not limited to these in the embodiments of the present disclosure.

[0113] In step 302, output availability status indication information of the first sensing result based on the first availability status.

[0114] Here, the availability status indication information can be indicated in any possible manner, for example, the availability status indication information may be indicated by a status flag bit, for example, 1 indicates availability and 0 indicates non-availability, or 0 indicates availability and 1 indicates non-availability.

[0115] In step 303, a second availability status of the second sensing result is determined.

[0116] In step 304, output an availability status indication of the second sensing result based on the second availability status.

[0117] Here, the second availability state is similar to the first availability state, so a duplicated description will be omitted here.

[0118] In some alternative embodiments, the order of steps 301 to 302 and steps 303 to 304 may be interchangeable. For example, if there is no dependency between the first sensing result and the second sensing result, the order may be interchangeable.

[0119] In this embodiment, by installing an availability status detection subroutine in the sensing application program, the availability status of the sensing results can be determined, which makes it easier to provide availability status indication information that is effective for planning and controlling vehicle driving, and avoids the adverse impact of unavailable sensing results on driving plans, thereby further improving the safety of vehicle driving.

[0120] In some alternative embodiments, FIG. 14 is a schematic diagram of a principle for determining the availability of sensing results according to an exemplary embodiment of the present disclosure. As shown in FIG. 14 , an availability detection subroutine may be provided in a sensing application program to detect the availability of sensing results. For example, the availability detection subroutine in sensing application program 1 is used to detect the availability of upstream data for sensing application program 1, and the availability detection subroutine in sensing application program 2 is used to detect the availability of output results from sensing application program 1. The availability detection subroutine may output availability indication information. The operation of sensing application program 2 depends on the output results of sensing application program 1. For example, sensing application program 1 may be a first forward-view sensing application program, and sensing application program 2 may be a first post-processing application program. For example, sensing application program 1 may be either a second forward-view sensing application program or a sensing application program for another viewpoint, and sensing application program 2 may be a second post-processing application program.

[0121] In some alternative embodiments, Figure 15 is a schematic diagram of a principle for determining the availability status of a sensing result according to another exemplary embodiment of the present disclosure. As shown in Figure 15, the availability status detection subroutine of each sensing application program can be used to individually output availability status indication information. A status collection module collects the availability status indication information output from each sensing application program, and outputs availability status indication information, such as a first sensing result or a second sensing result, based on the availability status indication information of each sensing application program.

[0122] In a method according to an embodiment of the present invention, two sensing function clusters are installed, one of which (i.e., the first sensing function cluster) can provide only minimal environmental sensing results, while the other (i.e., the second sensing function cluster) can provide main environmental sensing results (wide-range sensing results) including minimal environmental sensing. The real-time sensing information and sensing ranges provided by the different function clusters can be determined according to the maximum vehicle speed and vehicle control performance in actual use. During the sensing process, the two sensing function clusters operate simultaneously to provide real-time results including minimal environmental sensing. The separation is achieved by software partitioning, so that if the sensing application program in one sensing function cluster becomes invalid or cannot continue to operate normally, the sensing system can still continue to provide minimal environmental sensing through the other sensing function cluster. For example, if the sensing of the single camera at the forward viewpoint 1 freezes, the sensing results provided by the second sensing function cluster can cover minimal environmental sensing; if a peripheral viewpoint single camera in the second sensing function cluster freezes, both the first sensing function cluster and the second sensing function cluster can provide minimal environmental sensing functions. Software partitioning prevents the sensing application program from directly recognizing hardware resources, and the corresponding resource allocation and deployment are separated in the form of software partitioning, thereby effectively meeting the redundancy requirement for minimal environmental sensing. By binding and running the first sensing application program and the target second sensing application program on different processors, the redundancy requirement for minimal environmental sensing can be achieved through hardware separation.Therefore, the method of the present disclosure can achieve redundant minimum environmental sensing functions at a relatively low cost while realizing the main sensing function by combining environmental sensing channels, thereby improving driving safety and highly scalable software deployment. Furthermore, the present disclosure can combine multiple sensor data streams depending on the fault conditions of software and hardware, such as sensors, to ensure minimum sensing functions, improving performance and expanding the system security boundary when implementing minimum risk policies, thereby further improving vehicle driving safety.

[0123] The above-described embodiments of the present disclosure may be implemented alone or in any combination if not inconsistent, and specifically may be set according to actual needs, and are not limited by the present disclosure.

[0124] The redundant sensing method according to the embodiment of the present disclosure may be executed by any suitable device having data processing capabilities, including, but not limited to, a terminal device, a server, etc. Alternatively, the redundant sensing method according to the embodiment of the present disclosure may be executed by a processor, for example, the processor executes any of the redundant sensing methods mentioned in the embodiment of the present disclosure by calling corresponding instructions stored in a memory. Hereinafter, redundant descriptions will be omitted.

[0125] Exemplary Devices 16 is a structural schematic diagram of a redundant sensing device according to an exemplary embodiment of the present disclosure, which is used to realize a corresponding redundant sensing method embodiment of the present disclosure, and the device shown in FIG. 16 includes a first processing module 51 and a second processing module 52.

[0126] The first processing module 51 is used to determine a first sensing result based on a first sensing application program in a first sensing function cluster.

[0127] Here, the first sensing result includes a first minimum sensing result, and the first minimum sensing result is a sensing result within a first preset minimum viewing angle range.

[0128] The second processing module 52 is used to determine a second sensing result based on a second sensing application program in the second sensing function cluster.

[0129] Here, the second sensing result includes at least a second minimum sensing result, and the second minimum sensing result is a sensing result within a second preset minimum viewing angle range.

[0130] The first sensing application program in the first sensing function cluster is separated by software and / or hardware from at least the target second sensing application program, and the target second sensing application program represents a second sensing application program in the second sensing function cluster for obtaining a second minimum sensing result.

[0131] FIG. 17 is a structural schematic diagram of a redundant sensing device according to another exemplary embodiment of the present disclosure.

[0132] In some alternative embodiments, as shown in FIG. 17, the first processing module 51 may include a first obtaining unit 511, a first determining unit 512, and a first processing unit 513.

[0133] The first acquiring unit 511 can be used to acquire a first sensing application program in a first sensing function cluster from a first software partition.

[0134] The first determining unit 512 can be used to determine a first processor resource for executing the first sensing application program from the shared resource.

[0135] The first processing unit 513 can be used to execute a first sensing application program and obtain a first sensing result based on a first processor resource.

[0136] In some alternative embodiments, as shown in FIG. 17, the second processing module 52 may include a second acquisition unit 521, a second determination unit 522, a third acquisition unit 523, a third determination unit 524, and a second processing unit 525.

[0137] The second acquiring unit 521 can be used to acquire a target second sensing application program from a second software partition.

[0138] The second determining unit 522 can be used to determine a second processor resource for executing a target second sensing application program from the shared resource.

[0139] The third acquisition unit 523 can be used for acquiring other second sensing application programs other than the target second sensing application program in the second sensing function cluster from software partitions corresponding to the other second sensing application programs.

[0140] The third determining unit 524 can be used to determine processor resources corresponding to the other second sensing application programs based on the software partitions corresponding to the other second sensing application programs.

[0141] Here, the processor resource corresponding to the other second sensing application program is any one of the first processor resource, the second processor resource, and the third processor resource.

[0142] The second processing unit 525 can be used to execute a target second sensing application program based on second processor resources, and to execute other second sensing application programs based on processor resources corresponding to the other second sensing application programs to obtain second sensing results.

[0143] Here, the second processor resource and the first processor resource are different resources among the shared resources.

[0144] FIG. 18 is a structural schematic diagram of a redundant sensing device according to yet another exemplary embodiment of the present disclosure.

[0145] In some alternative embodiments, the first sensing application program may include a first forward-looking sensing application program and a first post-processing application program. The system may include a second forward-looking sensing application program, at least one other-viewpoint sensing application program, and a second post-processing application program. The first forward-looking sensing application program may correspond to a first software partition, and the second forward-looking sensing application program and the second post-processing application program may correspond to a second software partition. The first post-processing application program may correspond to a third software partition. Any other-viewpoint sensing application program may correspond to any of the first, second, third, and fourth software partitions.

[0146] In some alternative embodiments, as shown in FIG. 18, the first processing module 51 may include a first determining unit 512, a first processing unit 513, and a third processing unit 514.

[0147] The first determination unit 512 can be used to determine, based on the first software partition, a first processor resource for executing a first forward viewpoint sensing application program from the shared resource, and to determine, based on the third software partition, a third processor resource for executing a first post-processing application program from the shared resource.

[0148] The first processing unit 513 may be used to execute a first forward viewpoint sensing application program and obtain a first forward viewpoint sensing result based on a first processor resource.

[0149] The third processing unit 514 can be used to execute a first post-processing application program based on the third processor resource, perform post-processing on the first forward viewpoint sensing result, and obtain the first sensing result.

[0150] In some alternative embodiments, the second processing module 52 may include a second determining unit 522 , a third determining unit 524 , a second processing unit 525 and a fourth processing unit 526 .

[0151] The second determination unit 522 can be used to determine, based on the second software partition, second processor resources for executing a second forward viewpoint sensing application program and a second post-processing application program from the shared resources.

[0152] The third determination unit 524 can be used to determine corresponding processor resources for executing the sensing application program of each other perspective based on the software partition corresponding to the sensing application program of each other perspective.

[0153] The second processing unit 525 may be used to execute a second forward-view sensing application program and obtain a second forward-view sensing result based on a second processor resource.

[0154] The fourth processing unit 526 can be used to execute the sensing application program of each other viewpoint based on the processor resources corresponding to the sensing application program of each other viewpoint, and obtain other viewpoint sensing results corresponding to each other viewpoint.

[0155] The second processing unit 525 can further be used to execute a second post-processing application program based on the second processor resource, perform post-processing on the second forward viewpoint sensing result, other viewpoint sensing results corresponding to each other viewpoint, and obtain the second sensing result.

[0156] In some alternative embodiments, the third processing unit 514 can be specifically used to execute a first post-processing application program based on the third processor resource, perform fusion processing on the first forward viewpoint sensing result and the second forward viewpoint sensing result, and obtain the first sensing result.

[0157] The second processing unit 525 can specifically be used to execute a second post-processing application program based on the second processor resource, perform post-processing on the first forward viewpoint sensing result, the second forward viewpoint sensing result, and other viewpoint sensing results corresponding to each other viewpoint, respectively, to obtain a second sensing result.

[0158] In some alternative embodiments, the second processing unit 525 can be specifically used to execute a second post-processing application program based on a second processor resource, and perform post-processing on the first forward viewpoint sensing result, the second forward viewpoint sensing result, other viewpoint sensing results corresponding to each other viewpoint, and the first sensing result to obtain a second sensing result.

[0159] FIG. 19 is a structural schematic diagram of a redundant sensing device according to yet another exemplary embodiment of the present disclosure.

[0160] In some alternative embodiments, as shown in FIG. 19, the first processing module 51 may include a first determining unit 512 and a first processing unit 513.

[0161] The first determining unit 512 may be used to determine a first processor resource bound corresponding to the first sensing application program.

[0162] The first processing unit 513 can be used to execute a first sensing application program and obtain a first sensing result based on a first processor resource.

[0163] In some alternative embodiments, as shown in FIG. 19, the second processing module 52 may include a second determining unit 522 and a second processing unit 525.

[0164] The second determination unit 522 can be used to determine second processor resources bound corresponding to the target second sensing application program and processor resources bound corresponding to other second sensing application programs other than the target second sensing application program in the second sensing function cluster.

[0165] The second processing unit 525 can be used to execute a target second sensing application program based on the second processor resource, and to execute other second sensing application programs based on the processor resource bound corresponding to the other second sensing application programs, to obtain second sensing results.

[0166] FIG. 20 is a structural schematic diagram of a redundant sensing device according to yet another exemplary embodiment of the present disclosure.

[0167] In some alternative embodiments, as shown in FIG. 20 , the first processing module 51 may include a fourth determination unit 515 that can be used to determine a first availability status of the first sensing result and output availability status indication information of the first sensing result based on the first availability status.

[0168] The second processing module 52 may include a fifth determining unit 527 that can be used to determine a second availability status of the second sensing result and output availability status indication information of the second sensing result based on the second availability status.

[0169] The beneficial technical effects corresponding to the exemplary embodiments of the present device can be referred to the beneficial technical effects corresponding to the exemplary method part described above, and therefore, redundant description will be omitted here.

[0170] <Example Electronic Devices> FIG. 21 is a structural diagram of an electronic device according to an embodiment of the present disclosure, which includes at least one processor 11 and a memory 12.

[0171] The processor 11 may be a central processing unit (CPU) or other form of processing unit having data processing and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0172] The memory 12 may include one or more computer program products, which may include various forms of computer-readable storage media, such as, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, and flash memory. The computer-readable storage media may store one or more computer program instructions, and the processor 11 may execute the one or more computer program instructions to implement the methods and / or other desired functions described above in accordance with the embodiments of the present disclosure.

[0173] In one example, electronic device 10 may further include input devices 13 and output devices 14, with these components interconnected via a bus system and / or other form of connection (not shown).

[0174] The input device 13 may further include, for example, a keyboard and a mouse.

[0175] The output device 14 can output various types of information to the outside, and may include, for example, a display, a speaker, a printer, a communication network, and remote output devices connected thereto.

[0176] 21 shows only some of the components of the electronic device 10 that are relevant to the present disclosure, and omits components such as buses, input / output interfaces, etc. In addition, the electronic device 10 may further include any other appropriate components depending on the specific application.

[0177] Exemplary Computer Program Products and Computer-Readable Storage Media In addition to the above methods and apparatus, embodiments of the present disclosure may further provide a computer program product including computer program instructions that, when executed by a processor, cause the processor to perform the steps of the methods of various embodiments of the present disclosure described in the "Exemplary Methods" section above.

[0178] The computer program product may be written with program code for carrying out operations of embodiments of the present disclosure in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and traditional procedural programming languages ​​such as "C" or similar programming languages. The program code may execute entirely on a user's computing device, partially on a user's device, as a separate software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or a server.

[0179] Furthermore, an embodiment of the present disclosure may be a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform the steps of the methods of the various embodiments of the present disclosure described in the "Exemplary Method" section above.

[0180] The computer-readable storage medium can be any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable medium may include, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include an electrical connection having one or more leads, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), fiber optics, a compact disc read-only memory (CD-ROM), an optical storage element, a magnetic storage element, or any suitable combination of the above.

[0181] Although the basic principles of the present disclosure have been described above with reference to specific embodiments, the benefits, advantages, effects, etc. mentioned in the present disclosure are not limited but merely illustrative, and these benefits, advantages, effects, etc. do not necessarily belong to each embodiment of the present disclosure. Furthermore, the specific details disclosed above are not limited but merely serve to serve as examples and to facilitate understanding, and the above details do not necessarily limit the present disclosure to be realized by the above specific details.

[0182] Those skilled in the art can make various modifications and variations to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure also intends to include these modifications and variations.

Claims

1. determining a first sensing result based on a first sensing application program in a first sensing function cluster, the first sensing result including a first minimum sensing result, the first minimum sensing result being a sensing result within a first preset minimum viewing angle range; determining a second sensing result based on a second sensing application program in a second sensing function cluster, the second sensing result including at least a second minimum sensing result, the second minimum sensing result being a sensing result within a second preset minimum viewing angle range; A redundant sensing method executed by a redundant sensing device, in which a first sensing application program in the first sensing function cluster is separated by software and / or hardware from at least a target second sensing application program, and the target second sensing application program represents a second sensing application program in the second sensing function cluster for obtaining the second minimum sensing result.

2. The step of determining a first sensing result based on a first sensing application program in a first sensing function cluster includes: obtaining the first sensing application program in the first sensing function cluster from a first software partition; determining a first processor resource for executing the first sensing application program from a shared resource; 2. The redundant sensing method according to claim 1, further comprising: executing the first sensing application program based on the first processor resource and obtaining the first sensing result.

3. The step of determining a second sensing result based on a second sensing application program in a second sensing function cluster includes: obtaining the target second sensing application program from a second software partition; determining a second processor resource for executing the target second sensing application program from a shared resource; For another second sensing application program other than the target second sensing application program in the second sensing function cluster, acquiring the other second sensing application program from a software partition corresponding to the other second sensing application program, and determining a processor resource corresponding to the other second sensing application program based on the software partition corresponding to the other second sensing application program, wherein the processor resource corresponding to the other second sensing application program is any one of the first processor resource, the second processor resource, and a third processor resource; executing the target second sensing application program based on the second processor resource, and executing the other second sensing application program based on processor resources corresponding to the other second sensing application program to obtain the second sensing result; The redundant sensing method according to claim 2 , wherein the second processor resource and the first processor resource are different resources in the shared resource.

4. the first sensing application program includes a first forward-viewpoint sensing application program and a first post-processing application program; the second sensing application program includes a second forward-viewpoint sensing application program, at least one other-viewpoint sensing application program, and a second post-processing application program; the first forward-viewpoint sensing application program corresponds to a first software partition; the second forward-viewpoint sensing application program and the second post-processing application program correspond to a second software partition; the first post-processing application program corresponds to a third software partition; and any of the other-viewpoint sensing application programs corresponds to any of the first software partition, the second software partition, the third software partition, and a fourth software partition; The step of determining a first sensing result based on a first sensing application program in a first sensing function cluster includes: determining a first processor resource for executing the first forward viewpoint sensing application program from a shared resource based on the first software partition; determining, based on the third software partition, a third processor resource for executing the first post-processing application program from shared resources; Executing the first forward viewpoint sensing application program based on the first processor resource and obtaining a first forward viewpoint sensing result; executing the first post-processing application program based on the third processor resource, and performing post-processing on at least the first forward viewpoint sensing result to obtain the first sensing result; The step of determining a second sensing result based on a second sensing application program in a second sensing function cluster includes: determining, based on the second software partition, second processor resources for executing the second forward viewpoint sensing application program and the second post-processing application program from shared resources; determining corresponding processor resources for executing each of the sensing application programs of the other viewpoints based on software partitions corresponding to the sensing application programs of the other viewpoints; Executing the second forward viewpoint sensing application program based on the second processor resource and obtaining a second forward viewpoint sensing result; Executing the sensing application program for each of the other viewpoints based on a processor resource corresponding to the sensing application program for each of the other viewpoints, and obtaining other viewpoint sensing results corresponding to each of the other viewpoints; 2. The redundant sensing method of claim 1, further comprising: a step of executing the second post-processing application program based on the second processor resource, performing post-processing on at least the second forward viewpoint sensing result and the other viewpoint sensing results corresponding to each of the other viewpoints, and obtaining the second sensing result.

5. The step of executing the first post-processing application program based on the third processor resource, performing post-processing on at least the first forward viewpoint sensing result, and acquiring the first sensing result includes: executing the first post-processing application program based on the third processor resource, and performing fusion processing on the first forward viewpoint sensing result and the second forward viewpoint sensing result to obtain the first sensing result; the step of executing the second post-processing application program based on the second processor resource, performing post-processing on at least the second forward viewpoint sensing result and the other viewpoint sensing results corresponding to each of the other viewpoints, and acquiring the second sensing result, executing the second post-processing application program based on the second processor resource, and performing post-processing on the first forward viewpoint sensing result, the second forward viewpoint sensing result, and the other viewpoint sensing results corresponding to each of the other viewpoints, thereby obtaining the second sensing result; or 5. The redundant sensing method of claim 4, further comprising: executing the second post-processing application program based on the second processor resource, and performing post-processing on the first forward viewpoint sensing result, the second forward viewpoint sensing result, the other viewpoint sensing results corresponding to each of the other viewpoints, and the first sensing result, to obtain the second sensing result.

6. The step of determining a first sensing result based on a first sensing application program in a first sensing function cluster includes: determining a first processor resource bound corresponding to the first sensing application program; 2. The redundant sensing method according to claim 1, further comprising: executing the first sensing application program based on the first processor resource and obtaining the first sensing result.

7. The step of determining a second sensing result based on a second sensing application program in a second sensing function cluster includes: determining second processor resources bound to correspond to the target second sensing application program and processor resources bound to correspond to other second sensing application programs other than the target second sensing application program in the second sensing function cluster; 7. The redundant sensing method according to claim 6, further comprising: executing the target second sensing application program based on the second processor resource, and executing the other second sensing application program based on processor resources bound to the other second sensing application program, thereby obtaining the second sensing result.

8. determining a first availability state of the first sensing result; outputting an availability status indication of the first sensing result based on the first availability status; determining a second availability status of the second sensing result; and outputting availability status indication information of the second sensing result based on the second availability status.

9. a first processing module for determining a first sensing result based on a first sensing application program in a first sensing function cluster, the first sensing result including a first minimum sensing result, the first minimum sensing result being a sensing result within a first preset minimum field of view angle range; a second processing module for determining a second sensing result based on a second sensing application program in a second sensing function cluster, the second sensing result including at least a second minimum sensing result, the second minimum sensing result being a sensing result within a second preset minimum field of view angle range; A redundant sensing device in which a first sensing application program in the first sensing function cluster is separated by software and / or hardware from at least a target second sensing application program, and the target second sensing application program represents a second sensing application program in the second sensing function cluster for obtaining the second minimum sensing result.

10. A computer-readable storage medium storing a computer program for carrying out the method according to any one of claims 1 to 7.

11. a processor; a memory for storing instructions executable by said processor; An electronic device, wherein the processor is adapted to read and execute the executable instructions from the memory to implement the method of any one of claims 1 to 7.

Images