Automatic driving processing method, automatic driving processing device and electronic equipment

Abstract

The invention discloses an automatic driving processing method, an automatic driving processing device and electronic equipment.The method comprises the steps that under the condition that a vehicle is in an automatic driving state, main algorithms of automatic driving related functions are executed, and automatic driving of the vehicle is achieved through cooperative use of the main algorithms of different functions; each main algorithm is configured with at least one backup algorithm, and the backup algorithm is preset based on historical reliability data of the main algorithm; and determining the main algorithm with an error in the plurality of main algorithms as a target main algorithm, switching the target main algorithm into a target backup algorithm corresponding to the target main algorithm, and realizing the function of the target main algorithm through the target backup algorithm so as to maintain the operation of the automatic driving system.

Description

Technical Field

[0001] The present application relates to the technical field of processing for an autonomous driving system, and in particular to an autonomous driving processing method, an autonomous driving processing device, and an electronic device. Background Art

[0002] Autonomous driving systems are increasingly being used in vehicle driving. It is very important to ensure that the autonomous driving system can operate safely and reliably in the event of a failure or abnormality. In related technologies, the safety of autonomous driving systems is usually guaranteed by using a single redundant backup system, which backs up the autonomous driving system as a whole and starts the backup system when the autonomous driving system fails. However, this method does not back up different algorithms in the autonomous driving system, resulting in a waste of redundant backup resources and affecting the user experience. Summary of the invention

[0003] The embodiments of the present application provide a method for processing autonomous driving, a processing device for autonomous driving, and an electronic device.

[0004] On the one hand, an embodiment of the present application provides a method for processing autonomous driving, including:

[0005] When the vehicle is in an automatic driving state, a main algorithm for an automatic driving related function is executed, wherein the main algorithms for different functions are used in coordination to realize the automatic driving of the vehicle, and each main algorithm is configured with at least one backup algorithm, which is pre-set based on historical reliability data of the main algorithm;

[0006] A main algorithm in which an error occurs among multiple main algorithms is determined as a target main algorithm, and the target main algorithm is switched to a target backup algorithm corresponding to the target main algorithm. The function of the target main algorithm is implemented by the target backup algorithm to maintain the operation of the automatic driving system.

[0007] Optionally, the method further comprises:

[0008] Obtain the error probability of each main algorithm, and determine the historical reliability data of each main algorithm based on the error probability of each main algorithm; wherein the historical reliability data characterizes the operational reliability of the main algorithm, and the target number of backup algorithms required by the main algorithm is determined based on the operational reliability of the main algorithm.

[0009] Optionally, the method further comprises:

[0010] Each main algorithm is adjusted to a startup state, and based on a target number of backup algorithms required by the main algorithm, the startup number of the backup algorithms is determined.

[0011] Optionally, the switching of the target main algorithm to a target backup algorithm corresponding to the target main algorithm, and implementing the function of the target main algorithm through the target backup algorithm, includes:

[0012] The communication link of the main operating system corresponding to the target main algorithm is switched to the redundant system corresponding to the target backup algorithm, and data communication is performed between the redundant system corresponding to the target backup algorithm and the main operating system corresponding to the main algorithm adjacent to the main operating system corresponding to the target main algorithm.

[0013] Optionally, the method further comprises:

[0014] The running state of each main algorithm is detected to determine whether an error occurs in each main algorithm during vehicle operation, and the main algorithm with an error is determined as the target main algorithm.

[0015] Optionally, the method further comprises:

[0016] Detect the operating status of each backup algorithm to determine whether errors occur in each backup algorithm during vehicle operation;

[0017] When it is determined that an error occurs in the backup algorithm, the backup algorithm is switched to another backup algorithm having the same function to maintain the function of the backup algorithm.

[0018] Optionally, after switching the target main algorithm to a target backup algorithm corresponding to the target main algorithm, the method further includes:

[0019] The number of the redundant system corresponding to the target backup algorithm is replaced with the number of the main operating system corresponding to the target main algorithm, and the numbers of the redundant systems corresponding to other backup algorithms corresponding to the target main algorithm are adjusted in turn to change the usage priority of other backup algorithms corresponding to the target main algorithm.

[0020] Optionally, the main algorithm includes at least one of the following: a first main algorithm, a second main algorithm, and a third main algorithm;

[0021] The first main algorithm is used to collect vehicle environment information, the second main algorithm is used to analyze the vehicle environment information collected by the first main algorithm and generate vehicle movement instructions; the third main algorithm is used to execute the instructions of the second main algorithm to control the movement of the vehicle;

[0022] The first main algorithm, the second main algorithm and the third main algorithm are all configured with different backup algorithms, and the backup algorithms are used to implement the functions of the target main algorithm.

[0023] The embodiment of the present application also provides an automatic driving processing device, including:

[0024] a determination module, configured to execute a main algorithm of a function related to the automatic driving when the vehicle is in the automatic driving state, wherein the coordinated use of the main algorithms of different functions realizes the automatic driving of the vehicle, and each main algorithm is configured with at least one backup algorithm, and the backup algorithm is pre-set based on the historical reliability data of the main algorithm;

[0025] A processing module is used to determine a main algorithm in which an error occurs among multiple main algorithms as a target main algorithm, switch the target main algorithm to a target backup algorithm corresponding to the target main algorithm, and implement the function of the target main algorithm through the target backup algorithm to maintain the operation of the automatic driving system.

[0026] On the other hand, an embodiment of the present application further provides an electronic device, including a memory and a processor, wherein an executable program is stored in the memory, and the processor executes the executable program to implement the steps of the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flowchart of a processing method for autonomous driving according to an embodiment of the present application;

[0028] Figure 2 A flowchart of an embodiment of a processing method for autonomous driving according to an embodiment of the present application;

[0029] Figure 3 A flowchart of another embodiment of the processing method for autonomous driving according to an embodiment of the present application;

[0030] Figure 4 A flowchart of another embodiment of the processing method for autonomous driving of the embodiment of the present application;

[0031] Figure 5 This is a structural block diagram of an autonomous driving processing device according to an embodiment of the present application.

[0032] Figure 6 This is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0033] Various aspects and features of the present application are described herein with reference to the accompanying drawings.

[0034] It should be understood that various modifications may be made to the embodiments of the present application. Therefore, the above description should not be considered as limiting, but only as an example of an embodiment. Other modifications within the scope and spirit of the present application will occur to those skilled in the art.

[0035] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present application and, together with the general description of the present application given above and the detailed description of the embodiments given below, serve to explain the principles of the present application.

[0036] These and other characteristics of the present application will become apparent from the following description of a preferred form of embodiment given as a non-limiting example with reference to the accompanying drawings.

[0037] It should also be understood that although the present application has been described with reference to some specific examples, those skilled in the art will be able to readily implement many other equivalent forms of the present application.

[0038] The above and other aspects, features and advantages of the present application will become more apparent in view of the following detailed description when taken in conjunction with the accompanying drawings.

[0039] Specific embodiments of the present application are described hereinafter with reference to the accompanying drawings; however, it should be understood that the embodiments applied for are merely examples of the present application, which may be implemented in a variety of ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that obscure the present application. Therefore, the specific structural and functional details applied for herein are not intended to be limiting, but merely serve as a basis and representative basis for the claims to teach those skilled in the art to use the present application in a variety of ways with substantially any suitable detailed structure.

[0040] This specification may use the phrases "in one embodiment," "in another embodiment," "in yet another embodiment," or "in other embodiments," all of which may refer to one or more of the same or different embodiments according to the present application.

[0041] Figure 1 A flowchart of an automatic driving processing method according to an embodiment of the present application is shown. An automatic driving processing method provided by an embodiment of the present application is as follows: Figure 1 As shown, including:

[0042] S100, when the vehicle is in an automatic driving state, executing a main algorithm of an automatic driving related function, wherein the coordinated use of main algorithms of different functions realizes the automatic driving of the vehicle, and each main algorithm is configured with at least one backup algorithm, and the backup algorithm is pre-set based on historical reliability data of the main algorithm;

[0043] The processing method of automatic driving of this embodiment can be applied to vehicles with self-driving functions. The vehicle with self-driving functions is provided with an automatic driving system. The main algorithms of the functions related to automatic driving constitute the automatic driving system. Each main algorithm has different functions. For example, the main algorithm can have the function of collecting and understanding the information of the vehicle's surrounding environment, the function of analyzing and making decisions on the vehicle's surrounding environment, and the function of executing decision instructions and controlling the movement of the vehicle. In this embodiment, when the vehicle is in the automatic driving state, the main algorithm of the automatic driving related function is executed, and the coordinated use of the main algorithms of different functions realizes the automatic driving of the vehicle. Before running the main algorithms of different functions, at least one backup algorithm is configured for each main algorithm. The backup algorithm is a redundant algorithm of the main algorithm. The specific execution mode of the backup algorithm is different from that of the main algorithm. The backup algorithm can realize the same function as the main algorithm.

[0044] For example, for the function of collecting and understanding information about the vehicle's surroundings, the main algorithm can be a method based on deep learning, such as the pure visual perception algorithm Bev-former and the target detection algorithm YOLO; the backup algorithm can be an algorithm based on traditional computer vision, such as the image processing algorithm OpenCV and the Haar cascade classifier. For the function of analyzing and making decisions about the vehicle's surroundings, the main algorithm can be an algorithm based on graph search, such as the Dijkstra algorithm and the A* search algorithm; the backup algorithm can be an algorithm based on machine learning, such as the deep reinforcement learning algorithm and the neural network algorithm; the backup algorithm can also be an algorithm based on optimization, such as the dynamic programming algorithm and the gradient descent method. For the functions of executing decision instructions and controlling the movement of the vehicle, the main algorithm can be a model-based algorithm, such as the PID control algorithm (Proportional-Integral-Derivative Control) and the model predictive control algorithm MPC (Model Predictive Control); the backup algorithm can be a machine learning-based algorithm, such as a deep reinforcement learning algorithm and a neural network algorithm; the backup algorithm can also be an optimization-based algorithm, such as the optimal control algorithm LQR (Linear Quadratic Regulator) for linear systems and a nonlinear optimization algorithm (NonlinearOptimization).

[0045] After determining the main algorithms for different functions, one or more backup algorithms are pre-set for each main algorithm based on the historical reliability data of the main algorithm, and the historical reliability data of the main algorithm represents the probability of errors in the main algorithm during operation. If the historical reliability data of the main algorithm represents that the probability of errors in the main algorithm during operation is low, a backup algorithm can be configured for the main algorithm; if the historical reliability data of the main algorithm represents that the probability of errors in the main algorithm during operation is high, multiple backup algorithms can be configured for the main algorithm, for example, two backup algorithms can be configured for the main algorithm.

[0046] S200, determining a main algorithm in which an error occurs among multiple main algorithms as a target main algorithm, switching the target main algorithm to a target backup algorithm corresponding to the target main algorithm, and implementing the function of the target main algorithm through the target backup algorithm to maintain the operation of the autonomous driving system.

[0047] In this embodiment, during the execution of the automatic driving function, if the automatic driving function of the vehicle fails, the main algorithm in which the error occurs can be determined based on the specific failure of the automatic driving function. Since the automatic driving of the vehicle is realized based on the coordinated use of the main algorithms of different functions, if the automatic driving function fails, one or more main algorithms may have errors. The main algorithm in which the error occurs among the multiple running main algorithms is determined as the target main algorithm, and the target main algorithm is switched to the target backup algorithm corresponding to the target main algorithm, that is, the target backup algorithm is adjusted to the running state, and the operation of the target main algorithm is stopped. After the target main algorithm is switched to the target backup algorithm, the function of the target main algorithm can be realized by the target backup algorithm to maintain the operation of the automatic driving system.

[0048] For example, for the function of collecting and understanding the information of the vehicle's surrounding environment, the pure visual perception algorithm Bev-former is configured as the main algorithm, and the image processing algorithm OpenCV and the Haar cascade classifier are configured as backup algorithms; for the function of analyzing and making decisions about the vehicle's surrounding environment, the A* search algorithm is configured as the main algorithm, and the deep reinforcement learning algorithm is configured as the backup algorithm; for the function of executing decision instructions and controlling the movement of the vehicle, the PID control algorithm is configured as the main algorithm, and the nonlinear optimization algorithm is configured as the backup algorithm. After executing the main algorithm of the autonomous driving related functions, if the pure visual perception algorithm Bev-former is determined as the target main algorithm, the pure visual perception algorithm Bev-former can be switched to the image processing algorithm OpenCV, and the function of the pure visual perception algorithm Bev-former can be realized through the image processing algorithm OpenCV. If the pure visual perception algorithm Bev-former and the PID control algorithm are determined as the target main algorithms, the pure visual perception algorithm Bev-former can be switched to the image processing algorithm OpenCV, and the PID control algorithm can be switched to the nonlinear optimization algorithm, and the function of the pure visual perception algorithm Bev-former can be realized through the image processing algorithm OpenCV, and the function of the PID control algorithm can be realized through the nonlinear optimization algorithm.

[0049] The present application executes the main algorithm of the autonomous driving related functions through the above-mentioned method, realizes the autonomous driving of the vehicle through the coordinated use of the main algorithms of different functions, and pre-sets at least one backup algorithm for each main algorithm based on the historical reliability data of the main algorithm, and the backup algorithm has the same function as the corresponding main algorithm. In the process of executing the autonomous driving function, the main algorithm with an error is switched to the corresponding backup algorithm, and the function of the main algorithm with an error is realized through the backup algorithm to maintain the operation of the autonomous driving system, and only the algorithm with an error is switched to the redundant algorithm, which effectively utilizes the redundant backup resources and improves the user experience.

[0050] In one embodiment of the present application, Figure 2 As shown, the method also includes:

[0051] S300, obtaining the error probability of each main algorithm, and determining the historical reliability data of each main algorithm based on the error probability of each main algorithm; wherein the historical reliability data characterizes the operating reliability of the main algorithm, and the target number of backup algorithms required by the main algorithm is determined based on the operating reliability of the main algorithm.

[0052] In this embodiment, when the main algorithm is not configured with a backup algorithm, the main algorithm for the autonomous driving related functions is executed. During the operation of the autonomous driving, the operation status of each main algorithm is monitored to obtain the error probability of each main algorithm, and the historical reliability data of each main algorithm is determined based on the error probability of each main algorithm. The historical reliability data represents the operation reliability of the main algorithm. Based on the operation reliability represented by the historical reliability data of each main algorithm, the target number of backup algorithms required for each main algorithm can be determined, and a target number of backup algorithms can be set for each main algorithm. If the historical reliability data of the main algorithm represents a high operation reliability, a backup algorithm can be set for the main algorithm; if the historical reliability data of the main algorithm represents a low operation reliability, multiple backup algorithms can be set for the main algorithm to reduce the overall error probability of the autonomous driving system.

[0053] For example, the main algorithms for autonomous driving related functions are the main algorithms for collecting and understanding information about the vehicle's surroundings, the main algorithms for analyzing and making decisions about the vehicle's surroundings, and the main algorithms for executing decision instructions and controlling the vehicle's movement. Based on the historical data of autonomous driving, the error probabilities of the above three main algorithms are: P perception =0.01, P decision =0.005, P control =0.003, and then determine the reliability of the above three main algorithms are: R perception =1-0.01=0.99, R decision =1-0.005=0.995, R control =1-0.003=0.997, for this autonomous driving system, the reliability of the overall system R system for: The error probability of the overall system is P system error For: P system error =1-R system =1-0.982=0.018.

[0054] According to the reliability data of the above three main algorithms, after configuring a corresponding backup algorithm for each main algorithm, the reliability R of the above three main algorithms can be obtained based on the relevant data of autonomous driving. layer redundant They are: R perception redundant =1-(1-0.99) 2 =0.9999, R decision redundant =1-(1-0.995) 2 =0.999975, R control redundant =1-(1-0.997) 2 =0.999991, the reliability of the overall system Rredundant for: The error probability of the overall system is P redundant error For: P redundant error =1-R redundant =1-0.999866=0.000134.

[0055] According to the reliability data of the above three main algorithms with one backup algorithm, the reliability of the main algorithm corresponding to the function of collecting and understanding the information of the vehicle's surrounding environment is relatively low, and two backup algorithms can be set for the main algorithm. According to the relevant data of autonomous driving, the reliability of the above three main algorithms can be obtained. layer redundant They are: R perception redundant =1-(1-0.99) 3 =0.999999, R decision redundant =1-(1-0.995) 2 =0.999975, R control redundant =1-(1-0.997) 2 =0.999991, the reliability of the overall system R redundant for: The error probability of the overall system is P redundant error For: P redundant error =1-R redundant =1-0.999965=0.000035. By increasing the number of backup algorithms for the main algorithm with low operating reliability, the reliability of the main algorithm can be improved, thereby improving the overall reliability of the autonomous driving system.

[0056] In one embodiment of the present application, Figure 3 As shown, the method also includes:

[0057] S30, adjusting each main algorithm to a startup state, and determining the startup quantity of the backup algorithms based on the target quantity of the backup algorithms required by the main algorithm.

[0058] In this embodiment, after determining the target number of backup algorithms required for the main algorithm based on the operational reliability represented by the historical reliability data of the main algorithm, one or more backup algorithms may be pre-set for each main algorithm according to the target number of backup algorithms required for different main algorithms. Alternatively, multiple backup algorithms may be pre-set for each main algorithm, and the number of all backup algorithms pre-set for each main algorithm that need to be started is determined according to the actual error probability of each main algorithm, that is, the target number of backup algorithms required for the main algorithm. For example, for a main algorithm that has the function of collecting and understanding information about the vehicle's surroundings, historical reliability data represents that the error probability of the main algorithm is high, and the target number of backup algorithms required for the main algorithm may be set to multiple. For a main algorithm that has the function of analyzing and making decisions about the vehicle's surroundings, historical reliability data represents that the error probability of the main algorithm is low, and the target number of backup algorithms required for the main algorithm may be set to one. For a main algorithm that has the function of executing decision instructions and controlling the movement of the vehicle, historical reliability data represents that the error probability of the main algorithm is low, and the target number of backup algorithms required for the main algorithm may be set to one. When the vehicle is in the automatic driving state, each main algorithm is adjusted to the startup state, so that the automatic driving of the vehicle is realized by the coordinated use of the main algorithms with different functions, and the startup number of the backup algorithms is determined based on the target number of backup algorithms required by the main algorithm, and the backup algorithms have the same functions as the corresponding main algorithms. If an error occurs in the main algorithm, the main algorithm with the error can be switched to the corresponding backup algorithm, and the function of the main algorithm can be realized by the backup algorithm.

[0059] In one embodiment of the present application, the switching of the target main algorithm to a target backup algorithm corresponding to the target main algorithm, and implementing the function of the target main algorithm through the target backup algorithm, includes:

[0060] The communication link of the main operating system corresponding to the target main algorithm is switched to the redundant system corresponding to the target backup algorithm, and data communication is performed between the redundant system corresponding to the target backup algorithm and the main operating system corresponding to the main algorithm adjacent to the main operating system corresponding to the target main algorithm.

[0061] In this embodiment, the main operating system corresponding to each main algorithm has a communication link, and the main operating systems corresponding to each main algorithm will perform data communication transmission through their respective communication links, so that each main algorithm can be used in conjunction to realize the automatic driving of the vehicle. The redundant system corresponding to each backup algorithm has a communication link. In the event of an error in the target main algorithm, the communication link of the main operating system corresponding to the target main algorithm can be switched to the redundant system corresponding to the target backup algorithm, and data communication is performed between the redundant system corresponding to the target backup algorithm and the main operating system corresponding to the main algorithm adjacent to the main operating system corresponding to the target main algorithm, thereby realizing the function of the target main algorithm through the target backup algorithm.

[0062] In one embodiment of the present application, the method further includes:

[0063] The running state of each main algorithm is detected to determine whether an error occurs in each main algorithm during vehicle operation, and the main algorithm with an error is determined as the target main algorithm.

[0064] In this embodiment, when the vehicle is in an automatic driving state, the operating state of each main algorithm is monitored in real time to determine whether an error occurs in each main algorithm during vehicle operation. Specifically, the corresponding functions of each main algorithm during operation can be monitored. If the function corresponding to the main algorithm is not implemented or a problem occurs, it indicates that an error occurs in the main algorithm, resulting in a failure in the automatic driving system. One or more main algorithms with errors can be determined as target main algorithms. For example, when the vehicle is in an automatic driving state, the functions of collecting and understanding information about the vehicle's surroundings, analyzing and making decisions about the vehicle's surroundings, and executing decision instructions and controlling the movement of the vehicle are monitored in real time. If one or more of the above functions are not implemented or a problem occurs, it indicates that an error occurs in the corresponding main algorithm. One or more main algorithms with errors can be determined as target main algorithms.

[0065] In one embodiment of the present application, Figure 4 As shown, the method also includes:

[0066] S400, detecting the operating status of each backup algorithm to determine whether an error occurs in each backup algorithm during vehicle operation;

[0067] S500: When it is determined that an error occurs in the backup algorithm, the backup algorithm is switched to another backup algorithm with the same function to maintain the function of the backup algorithm.

[0068] In this embodiment, when the vehicle is in the automatic driving state, after the main algorithms of different functions are started, the target number of backup algorithms required by each main algorithm will also be started. The backup algorithm can replace the function of the main algorithm when an error occurs in the main algorithm. After the backup algorithm corresponding to each main algorithm is started, the backup algorithm in the started state is monitored in real time to determine whether an error occurs in each backup algorithm during vehicle operation.

[0069] When it is determined that an error occurs in the backup algorithm, the backup algorithm is switched to another backup algorithm with the same function to maintain the function of the backup algorithm. Since the backup algorithm has the same function as the corresponding main algorithm, if an error occurs in the backup algorithm, the main algorithm cannot continue to maintain the relevant function in the event of an error. The number of backup algorithms that need to be started can be set to multiple, and after each backup algorithm is started, the operating status of each backup algorithm is detected. For example, for the function of collecting and understanding information about the vehicle's surroundings, the pure visual perception algorithm Bev-former is configured as the main algorithm and started, and the image processing algorithm OpenCV and the Haar cascade classifier are configured as backup algorithms and started. After detecting the operating status of each backup algorithm, if an error occurs in the image processing algorithm OpenCV, the image processing algorithm OpenCV can be switched to the Haar cascade classifier, and the function of the backup algorithm can be maintained through the Haar cascade classifier.

[0070] In one embodiment of the present application, after the target main algorithm is switched to a target backup algorithm corresponding to the target main algorithm, the method further includes:

[0071] The number of the redundant system corresponding to the target backup algorithm is replaced with the number of the main operating system corresponding to the target main algorithm, and the numbers of the redundant systems corresponding to other backup algorithms corresponding to the target main algorithm are adjusted in turn to change the usage priority of other backup algorithms corresponding to the target main algorithm.

[0072] In this embodiment, after determining the main algorithms of different functions and the target number of backup algorithms, a number is set for the main operating system corresponding to each main algorithm, and a number is set for the redundant system corresponding to each backup algorithm, and the number can be represented by a number. If the target number of backup algorithms corresponding to the main algorithm is multiple, the number of the redundant system corresponding to each backup algorithm in the multiple backup algorithms can be arranged and set in sequence, so as to determine the use priority of the multiple backup algorithms. After determining the target main algorithm in which an error occurs, if the target number of backup algorithms corresponding to the main algorithm is multiple, the target backup algorithm can be determined from the multiple backup algorithms according to the use priority of the multiple backup algorithms. The target main algorithm is switched to the target backup algorithm corresponding to the target main algorithm, and the number of the redundant system corresponding to the target backup algorithm is replaced with the number of the main operating system corresponding to the target main algorithm, that is, the main operating system corresponding to the target main algorithm can be replaced with the redundant system corresponding to the target backup algorithm. And the numbers of the redundant systems corresponding to other backup algorithms corresponding to the target main algorithm are adjusted in sequence to change the use priority of other backup algorithms corresponding to the target main algorithm.

[0073] For example, for the function of collecting and understanding the information of the vehicle's surrounding environment, the pure visual perception algorithm Bev-former is configured as the main algorithm, and the number of the main operating system corresponding to the main algorithm is set to 1; the image processing algorithm OpenCV, the Haar cascade classifier, and the target detection algorithm YOLO are configured as backup algorithms, and the numbers of the redundant systems corresponding to the three backup algorithms are set to 2, 3, and 4 respectively. If an error occurs in the pure visual perception algorithm Bev-former, the pure visual perception algorithm Bev-former can be switched to the image processing algorithm OpenCV, and the number 2 of the redundant system corresponding to the image processing algorithm OpenCV is replaced with the number 1 of the main operating system corresponding to the pure visual perception algorithm Bev-former. Then, the number of the redundant system corresponding to the Haar cascade classifier is changed from 3 to 2, and the number of the redundant system corresponding to the target detection algorithm YOLO is changed from 4 to 3, so as to change the priority of the use of the above backup algorithms.

[0074] In one embodiment of the present application, the main algorithm includes at least one of the following: a first main algorithm, a second main algorithm and a third main algorithm; the first main algorithm, the second main algorithm and the third main algorithm are all configured with different backup algorithms, and the backup algorithms are used to implement the function of the target main algorithm.

[0075] In this embodiment, the first main algorithm is used to collect vehicle environment information. The first main algorithm may have the following functions: identify and locate moving objects such as vehicles, pedestrians, and animals; distinguish static objects such as roads, lane lines, and obstacles through semantic segmentation; combine object detection and semantic segmentation to provide detailed information of all objects in the scene through panoramic segmentation; combine data from multiple sensors such as lidar, camera, millimeter wave radar, etc. to improve the accuracy and robustness of perception. The second main algorithm is used to analyze the vehicle environment information collected by the first main algorithm and generate vehicle movement instructions. The second main algorithm may have the following functions: generate a path planning function for a global path from the starting point to the end point; generate a local path adapted to the current environment based on the global path; make behavioral decisions based on the real-time environment and traffic rules to determine the specific actions of the vehicle, such as changing lanes, overtaking, and parking; generate a motion trajectory that meets physical constraints and safety requirements. The third main algorithm is used to execute the instructions of the second main algorithm to control the movement of the vehicle; the third main algorithm may have the following functions: by controlling the throttle and brake, longitudinal control of acceleration and deceleration is achieved; by controlling the steering wheel, lateral control of vehicle steering is achieved; ensure that the vehicle travels along the planned trajectory, and adjusts the speed and direction to maintain the accuracy of trajectory tracking. The first main algorithm, the second main algorithm, and the third main algorithm are all configured with backup algorithms with the same functions. The automatic driving function is achieved through the coordinated use of the above main algorithms. The main algorithm in which an error occurs in the above main algorithms is determined as the target main algorithm, and the target main algorithm is switched to the corresponding target backup algorithm. The target backup algorithm is used to implement the function of the target main algorithm.

[0076] Based on the same inventive concept, the second aspect of the present application also provides an autonomous driving processing device corresponding to an autonomous driving processing method. Since the principle of solving the problem by the autonomous driving processing device in the present application is similar to the above-mentioned autonomous driving processing method of the present application, the implementation of the autonomous driving processing device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0077] Figure 5 The schematic diagram of the structure of the processing device for autonomous driving provided in the embodiment of the present application is shown, which specifically includes:

[0078] a determination module, configured to execute a main algorithm of a function related to the automatic driving when the vehicle is in the automatic driving state, wherein the coordinated use of the main algorithms of different functions realizes the automatic driving of the vehicle, and each main algorithm is configured with at least one backup algorithm, and the backup algorithm is pre-set based on the historical reliability data of the main algorithm;

[0079] A processing module is used to determine a main algorithm in which an error occurs among multiple main algorithms as a target main algorithm, switch the target main algorithm to a target backup algorithm corresponding to the target main algorithm, and implement the function of the target main algorithm through the target backup algorithm to maintain the operation of the automatic driving system.

[0080] In one embodiment of the present application, the determination module is further configured to:

[0081] Obtaining an error probability of each main algorithm, and determining historical reliability data of each main algorithm based on the error probability of each main algorithm;

[0082] The historical reliability data represents the operational reliability of the main algorithm, and the target number of backup algorithms required by the main algorithm is determined based on the operational reliability of the main algorithm.

[0083] In one embodiment of the present application, the determination module is further configured to:

[0084] Each main algorithm is adjusted to a startup state, and based on a target number of backup algorithms required by the main algorithm, the startup number of the backup algorithms is determined.

[0085] In one embodiment of the present application, the processing module is further configured to:

[0086] The communication link of the main operating system corresponding to the target main algorithm is switched to the redundant system corresponding to the target backup algorithm, and data communication is performed between the redundant system corresponding to the target backup algorithm and the main operating system corresponding to the main algorithm adjacent to the main operating system corresponding to the target main algorithm.

[0087] In one embodiment of the present application, the determination module is further configured to:

[0088] The running state of each main algorithm is detected to determine whether an error occurs in each main algorithm during vehicle operation, and the main algorithm with an error is determined as the target main algorithm.

[0089] In one embodiment of the present application, the processing module is further configured to:

[0090] Detect the operating status of each backup algorithm to determine whether errors occur in each backup algorithm during vehicle operation;

[0091] When it is determined that an error occurs in the backup algorithm, the backup algorithm is switched to another backup algorithm having the same function to maintain the function of the backup algorithm.

[0092] In one embodiment of the present application, the processing module is further configured to:

[0093] The number of the redundant system corresponding to the target backup algorithm is replaced with the number of the main operating system corresponding to the target main algorithm, and the numbers of the redundant systems corresponding to other backup algorithms corresponding to the target main algorithm are adjusted in turn to change the usage priority of other backup algorithms corresponding to the target main algorithm.

[0094] Based on the same inventive concept, Figure 6 As shown, this embodiment also includes an electronic device, including:

[0095] A memory for storing executable programs;

[0096] The processor is used to execute the executable program to implement the above method.

[0097] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A method for processing an automatic driving, comprising: When the vehicle is in an automatic driving state, a main algorithm for an automatic driving related function is executed, wherein the main algorithms for different functions are used in coordination to realize the automatic driving of the vehicle, and each main algorithm is configured with at least one backup algorithm, which is pre-set based on historical reliability data of the main algorithm; A main algorithm in which an error occurs among multiple main algorithms is determined as a target main algorithm, and the target main algorithm is switched to a target backup algorithm corresponding to the target main algorithm. The function of the target main algorithm is implemented by the target backup algorithm to maintain the operation of the automatic driving system.

2. The method according to claim 1, further comprising: Obtain the error probability of each main algorithm, and determine the historical reliability data of each main algorithm based on the error probability of each main algorithm; wherein the historical reliability data characterizes the operational reliability of the main algorithm, and the target number of backup algorithms required by the main algorithm is determined based on the operational reliability of the main algorithm.

3. The method according to claim 2, further comprising: Each main algorithm is adjusted to a startup state, and based on a target number of backup algorithms required by the main algorithm, the startup number of the backup algorithms is determined.

4. The method according to claim 1, wherein the target main algorithm is switched to a target backup algorithm corresponding to the target main algorithm, and the function of the target main algorithm is realized by the target backup algorithm, comprising: The communication link of the main operating system corresponding to the target main algorithm is switched to the redundant system corresponding to the target backup algorithm, and data communication is performed between the redundant system corresponding to the target backup algorithm and the main operating system corresponding to the main algorithm adjacent to the main operating system corresponding to the target main algorithm.

5. The method according to claim 1, further comprising: The running state of each main algorithm is detected to determine whether an error occurs in each main algorithm during vehicle operation, and the main algorithm with an error is determined as the target main algorithm.

6. The method according to claim 1, further comprising: Detect the operating status of each backup algorithm to determine whether errors occur in each backup algorithm during vehicle operation; When it is determined that an error occurs in the backup algorithm, the backup algorithm is switched to another backup algorithm having the same function to maintain the function of the backup algorithm.

7. The method according to claim 1, after switching the target main algorithm to a target backup algorithm corresponding to the target main algorithm, the method further comprises: The number of the redundant system corresponding to the target backup algorithm is replaced with the number of the main operating system corresponding to the target main algorithm, and the numbers of the redundant systems corresponding to other backup algorithms corresponding to the target main algorithm are adjusted in turn to change the usage priority of other backup algorithms corresponding to the target main algorithm.

8. The method according to claim 1, wherein the main algorithm comprises at least one of the following: a first main algorithm, a second main algorithm, and a third main algorithm; The first main algorithm is used to collect vehicle environment information, and the second main algorithm is used to analyze the vehicle environment information collected by the first main algorithm and generate vehicle movement instructions; The third main algorithm is used to execute the instructions of the second main algorithm to control the movement of the vehicle; The first main algorithm, the second main algorithm and the third main algorithm are all configured with different backup algorithms, and the backup algorithms are used to implement the functions of the target main algorithm.

9. A processing device for automatic driving, comprising: a determination module, configured to execute a main algorithm of a function related to the automatic driving when the vehicle is in the automatic driving state, wherein the coordinated use of the main algorithms of different functions realizes the automatic driving of the vehicle, and each main algorithm is configured with at least one backup algorithm, and the backup algorithm is pre-set based on the historical reliability data of the main algorithm; A processing module is used to determine a main algorithm in which an error occurs among multiple main algorithms as a target main algorithm, switch the target main algorithm to a target backup algorithm corresponding to the target main algorithm, and implement the function of the target main algorithm through the target backup algorithm to maintain the operation of the automatic driving system.

10. An electronic device comprising a processor and a memory, wherein the memory stores an executable program, and the processor executes the executable program to perform the autonomous driving processing method as described in any one of claims 1 to 8.

Images