A parameter calibration method, apparatus, electronic device, and storage medium

Abstract

This disclosure provides a parameter calibration method, apparatus, electronic device, and storage medium, relating to computer technology, particularly to fields such as cloud computing, autonomous driving, autonomous parking, and intelligent transportation. The specific implementation scheme is as follows: Given that the parameters to be calibrated are steering wheel zero-offset and steering ratio, an input vector for a joint calibration model is constructed based on a first coefficient corresponding to the steering wheel zero-offset and a second coefficient corresponding to the steering ratio; based on iterative training of the joint calibration model using the input vector, the joint calibration result of the steering wheel zero-offset and steering ratio is output. Using this disclosure can improve the accuracy of parameter calibration.

Description

Technical Field

[0001] This disclosure relates to computer technology, particularly to fields such as cloud computing, autonomous driving, autonomous parking, and intelligent transportation. Background Technology

[0002] In autonomous driving scenarios, considering safe driving, it is necessary to calibrate the relevant parameters of autonomous driving in order to constrain autonomous driving behavior and thus control the vehicle to always be in a safe driving state. However, the current parameter calibration methods are not only costly but also inaccurate, thus bringing safety hazards. Summary of the Invention

[0003] This disclosure provides a parameter calibration method, apparatus, electronic device, and storage medium.

[0004] According to one aspect of this disclosure, a parameter calibration method is provided, comprising:

[0005] Given that the parameters to be calibrated are steering wheel zero bias and steering ratio, the input vector of the joint calibration model is constructed based on the first coefficient corresponding to steering wheel zero bias and the second coefficient corresponding to steering ratio.

[0006] Based on the iterative training of the joint calibration model using the input vector, the joint calibration results of the steering wheel zero bias and steering ratio are output.

[0007] According to another aspect of this disclosure, a parameter calibration apparatus is provided, comprising:

[0008] The construction module is used to construct the input vector of the joint calibration model based on the first coefficient corresponding to the steering wheel zero bias and the second coefficient corresponding to the steering ratio when the parameters to be calibrated are the steering wheel zero bias and the steering ratio.

[0009] The calibration module is used to output the joint calibration results of the steering wheel zero bias and the steering ratio based on the iterative training of the joint calibration model according to the input vector.

[0010] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0011] At least one processor; and

[0012] The memory is communicatively connected to the at least one processor; wherein,

[0013] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method provided in any embodiment of this disclosure.

[0014] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing the computer to perform the methods provided in any embodiment of this disclosure.

[0015] According to another aspect of this disclosure, a computer program product is provided, including computer instructions that, when executed by a processor, implement the methods provided in any embodiment of this disclosure.

[0016] By using this disclosure, when the parameters to be calibrated are determined to be steering wheel zero bias and steering ratio, the input vector of the joint calibration model can be constructed based on the first coefficient corresponding to the steering wheel zero bias and the second coefficient corresponding to the steering ratio. Thus, the joint calibration result of steering wheel zero bias and steering ratio can be output based on the iterative training of the joint calibration model using the input vector. Since it is a joint calibration of steering wheel zero bias and steering ratio, the accuracy of parameter calibration can be improved.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0018] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0019] Figure 1 This is a schematic diagram of a distributed cluster processing scenario according to an embodiment of the present disclosure;

[0020] Figure 2 This is a flowchart illustrating a parameter calibration method according to an embodiment of the present disclosure;

[0021] Figure 3 This is a flowchart illustrating another parameter calibration method according to an embodiment of the present disclosure;

[0022] Figure 4 This is a schematic diagram of the model input and output according to an embodiment of the present disclosure;

[0023] Figure 5 This is a flowchart illustrating another parameter calibration method according to an embodiment of the present disclosure.

[0024] Figure 6 This is another schematic diagram of the model input and output according to an embodiment of the present disclosure;

[0025] Figure 7 This is a schematic diagram of parameter compensation according to an embodiment of the present disclosure;

[0026] Figure 8 This is a schematic diagram of the composition of a parameter calibration device according to an embodiment of the present disclosure;

[0027] Figure 9 This is a block diagram of an electronic device used to implement the parameter calibration method of the embodiments of this disclosure. Detailed Implementation

[0028] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0029] In this document, the term "and / or" merely describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. The term "at least one" in this document indicates any combination of at least two of a plurality of elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C. The terms "first" and "second" in this document refer to and distinguish between multiple similar technical terms, not to restrict the order or to limit there to only two. For example, "first feature" and "second feature" refer to two categories / two features; the first feature can be one or more, and the second feature can also be one or more.

[0030] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0031] According to embodiments of this disclosure, Figure 1 This is a schematic diagram illustrating an application scenario of communication between an autonomous vehicle and the cloud according to an embodiment of this disclosure, such as... Figure 1 As shown, it includes: a backend server 100, multiple vehicles (e.g., vehicles 107-109), and a "cloud" 106 for communication between the backend server and the multiple vehicles. The backend server can utilize this distributed cluster system, such as... Figure 1As shown, this distributed cluster system includes multiple nodes (such as server cluster 101, server 102, server cluster 103, server 104, and server 105), and these nodes can jointly execute one or more parameter calibration tasks. Optionally, the multiple nodes in this distributed cluster system can execute the parameter calibration task in the same way, or they can execute the parameter calibration task in different ways. Optionally, after each round of data processing is completed, the multiple nodes can exchange data (such as data synchronization).

[0032] According to embodiments of this disclosure, a parameter calibration method is provided. Figure 2 This is a flowchart illustrating a parameter calibration method according to an embodiment of the present disclosure. This method can be applied to a parameter calibration device, for example, which can be deployed in a single-machine, multi-machine, or cluster system as a terminal, server, or other processing device to perform parameter calibration and other processing. The terminal can be a user equipment (UE), mobile device, personal digital assistant (PDA), handheld device, computing device, in-vehicle device, wearable device, etc. In some possible implementations, the method can also be implemented by a processor calling computer-readable instructions stored in memory. Figure 2 As shown, this method is applied to Figure 1 Any node in the cluster system shown includes:

[0033] S201. Given that the parameters to be calibrated are the steering wheel zero bias and the steering ratio, construct the input vector of the joint calibration model based on the first coefficient corresponding to the steering wheel zero bias and the second coefficient corresponding to the steering ratio.

[0034] S202. Based on the iterative training of the input vector in the joint calibration model, output the joint calibration results of the steering wheel zero bias and steering ratio.

[0035] In one example of S201-S202, the first coefficient can be the steering wheel zero-bias coefficient, and the second coefficient can be the steering ratio coefficient. The steering wheel zero-bias can be constrained by the steering wheel zero-bias coefficient during the iterative training of the joint calibration model, or the steering ratio can be constrained by the steering ratio coefficient. In addition to using the steering wheel zero-bias coefficient or the steering ratio coefficient for separate constraints, the steering wheel zero-bias coefficient and the steering ratio coefficient can also be used for joint constraints. Using the joint constraint method for joint calibration results in more accurate joint calibration results for steering wheel zero-bias and steering ratio compared to the separate constraint method, and the iteration speed is faster.

[0036] By using this disclosure, when the parameters to be calibrated are determined to be steering wheel zero bias and steering ratio, the input vector of the joint calibration model can be constructed based on the first coefficient corresponding to the steering wheel zero bias and the second coefficient corresponding to the steering ratio. Thus, the joint calibration result of steering wheel zero bias and steering ratio can be output based on the iterative training of the joint calibration model using the input vector. Since it is a joint calibration of steering wheel zero bias and steering ratio, the accuracy of parameter calibration can be improved.

[0037] According to embodiments of this disclosure, a parameter calibration method is provided. Figure 3 This is a flowchart illustrating a parameter calibration method according to an embodiment of the present disclosure. This method can be applied to a parameter calibration device, for example, the device can be deployed on a terminal, server, or other processing device in a single-machine, multi-machine, or cluster system, and can perform parameter calibration and other processing. Figure 3 As shown, this method is applied to Figure 1 Any node in the cluster system shown includes:

[0038] S301. Given that the parameters to be calibrated are the steering wheel zero bias and the steering ratio, construct the input vector of the joint calibration model based on the first coefficient corresponding to the steering wheel zero bias and the second coefficient corresponding to the steering ratio.

[0039] S302. Iteratively train the joint calibration model based on the input vector, take the first and second coefficients in the input vector as constraints, and perform joint online calibration of the steering wheel zero bias and steering ratio under constraints based on the constraints. The output calibration result is determined as the joint calibration result of the steering wheel zero bias and steering ratio.

[0040] In some examples, such as Figure 4 As shown, the input vector includes a first coefficient and a second coefficient, which are input into the joint calibration model 401. The first coefficient can be the steering wheel zero-bias coefficient, and the second coefficient can be the steering ratio coefficient. In the iterative training of the joint calibration model, one approach is to use separate constraints, i.e., constraining the steering wheel zero-bias using the steering wheel zero-bias coefficient, or constraining the steering ratio using the steering ratio coefficient; another approach is to use joint constraints, i.e., using both the steering wheel zero-bias coefficient and the steering ratio coefficient for joint constraints. Both separate and joint constraint methods can achieve the purpose of joint calibration, and the output vector of the joint calibration model is the joint calibration result (i.e., the joint calibration result of the steering wheel zero-bias and the steering ratio).

[0041] It's important to note that steering wheel zero-bias refers to the deviation between the actual zero position of the steering wheel and the 0° steering angle, while the steering ratio refers to the ratio of the steering wheel angle to the front wheel deflection angle. In autonomous driving, zero-bias in the steering wheel often leads to lateral deviation of the vehicle, posing a collision risk and compromising safe driving. Furthermore, in autonomous vehicle control, if front wheel zero-bias is used as a direct control variable and the corresponding steering wheel zero-bias is calculated using the steering ratio, neglecting the steering ratio or having an inaccurate steering ratio calibration can result in excessive lateral control errors, causing the vehicle to deviate from the planned path and posing a safety risk.

[0042] In other words, if only front wheel zero-bias is considered, there will be errors, which will be detrimental to safe driving in autonomous driving scenarios. One evaluation indicator used in the industry to determine whether a safe driving state is in is steering wheel zero-bias. Obviously, steering wheel zero-bias and front wheel zero-bias are different concepts. Steering wheel zero-bias can be obtained from front wheel zero-bias and steering ratio. If the steering ratio is not calibrated to identify whether it also has deviations (i.e., whether there is zero-bias in the steering ratio), and only the front wheel zero-bias is calibrated to identify only the deviations present in that front wheel zero-bias, for example, by taking a fixed value based on experience as the steering ratio, then even for the same vehicle model, the zero-bias of the steering ratio will differ between different vehicles.

[0043] In summary, calibrating only the front wheel zero-offset without considering the steering ratio results in inaccurate calculations of the steering wheel zero-offset. Treating the steering ratio of the same vehicle model as a fixed value ignores the differences in steering ratios between different vehicles of the same model. In reality, the steering ratio is closely related to the steering steer-by-wire system, the bogie, and even tire characteristics. Different vehicles of the same model exhibit consistent differences, and consequently, their steering ratios will also vary.

[0044] Steering wheel deflection and steering ratio can be obtained offline through methods such as four-wheel alignment. However, offline calibration is not only costly but also inefficient, making it unsuitable for autonomous driving in mass-produced vehicles. In particular, due to factors such as mechanical wear, compression, and tire pressure changes, steering wheel deflection and steering ratio are actually dynamic, rendering offline calibration results inapplicable.

[0045] By employing this implementation method, the joint calibration of steering wheel zero bias and steering ratio not only overcomes the calibration accuracy problem caused by the difference in steering ratio between different vehicles of the same model, but also, the online calibration method, which differs from the offline calibration method, is matched with dynamic changes, thereby accurately calibrating steering wheel zero bias and steering ratio. This has a significant effect on improving the lateral control accuracy and stability of vehicles in autonomous driving scenarios.

[0046] According to embodiments of this disclosure, a parameter calibration method is provided. Figure 5This is a flowchart illustrating a parameter calibration method according to an embodiment of the present disclosure. This method can be applied to a parameter calibration device, for example, the device can be deployed on a terminal, server, or other processing device in a single-machine, multi-machine, or cluster system, and can perform parameter calibration and other processing. Figure 5 As shown, this method is applied to Figure 1 Any node in the cluster system shown includes:

[0047] S501. Given that the parameters to be calibrated are steering wheel zero bias and steering ratio, construct the input vector of the joint calibration model based on the first coefficient corresponding to steering wheel zero bias and the second coefficient corresponding to steering ratio.

[0048] S502. Iteratively train the joint calibration model based on the input vector, and use the first and second coefficients in the input vector as constraints. Perform joint online calibration of the steering wheel zero bias and steering ratio under constraints based on these constraints.

[0049] S503. When performing multiple iterations of training, obtain the iteration processing results corresponding to each iteration of training.

[0050] S504. After preprocessing the iterative processing result, the iterative processing result to be compared is obtained.

[0051] S505. Compare the iterative processing result to be compared with the threshold. If the iterative processing result to be compared is less than the threshold, stop the iterative training and determine the output result obtained after stopping the iterative training as the output calibration result.

[0052] S506. Determine the calibration result of this output as the joint calibration result of steering wheel zero bias and steering ratio.

[0053] In some examples, such as Figure 6 As shown, the input vector includes a first coefficient and a second coefficient, which are input into the joint calibration model 601. The first coefficient can be the steering wheel zero-bias coefficient, and the second coefficient can be the steering ratio coefficient. In the iterative training of the joint calibration model, multiple iterations are performed until the steering wheel zero-bias and steering ratio are in a convergent state. At this point, after multiple iterations of training until convergence, the joint calibration model 601 yields the target joint calibration model 602. The output vector of the target joint calibration model 602 is the joint calibration result (i.e., the joint calibration result of the steering wheel zero-bias and steering ratio).

[0054] In this implementation method, the iterative processing results obtained after multiple iterations of training are compared with the threshold to serve as the calibration result of the output. In other words, the converged result obtained after the joint calibration result converges and the iterative training stops is the final steering wheel zero bias and steering ratio, instead of directly using the result obtained after each iteration of training as the calibration result of the output, thus improving the accuracy of online joint calibration.

[0055] In one embodiment, when performing multiple iterations of training, the iteration processing result corresponding to each iteration can be preprocessed as follows to obtain the iteration processing result to be compared, and then the iteration processing result to be compared is compared with a threshold. If the iteration processing result to be compared is less than the threshold, the iteration training is stopped, and the output result obtained after stopping the iteration training is determined as the output calibration result.

[0056] Specifically, preprocessing includes any of the following methods:

[0057] 1) Calculate the ratio of variance to mean based on the iterative processing results, and use the ratio result as the iterative processing result of the comparison.

[0058] 2) Perform weighted processing based on the iterative processing results, and use the weighted processing results as the iterative processing results of the comparison.

[0059] 3) Take the average value of the results of the iteration process and use the average value as the result of the iteration process to be compared.

[0060] By adopting this implementation method, the above preprocessing can obtain the iterative processing result to be compared. Since it is not a simple superposition of multiple iterative training cases, a more accurate iterative processing result to be compared can be obtained. Thus, the goal of joint calibration result convergence can be achieved better and faster. The converged result obtained after stopping iterative training can be used as the final steering wheel zero bias and steering ratio, thereby improving the accuracy of online joint calibration.

[0061] In one embodiment, iterative training of the joint calibration model based on the input vector includes: performing one or more iterative training iterations in the joint calibration model based on the input vector and the acquired control parameters to obtain the output joint calibration result. The control parameters include at least one of the following: vehicle yaw rate, vehicle speed, steering wheel angle, and vehicle wheelbase.

[0062] like Figure 7The diagram illustrates the zero-bias calibration required for vehicle 701 in an autonomous driving scenario. Vehicle 701 is controlled by steering wheel 704 within a pair of lane lines (lane line 702-lane line 703). During vehicle movement, front wheel zero-bias 705 is obtained, which is the difference between the actual front wheel steering angle (as shown by the solid arrow) and the measured front wheel steering angle (as shown by the dashed arrow). The actual front wheel steering angle is the perceived steering angle of the front wheels, and the measured steering angle is the value collected by sensors (such as gyroscopes) on the vehicle. Steering wheel zero-bias 706 is also obtained during vehicle movement, obtained by the difference between the actual steering wheel steering angle (as shown by the solid arrow) and the measured steering wheel steering angle (as shown by the dashed arrow). Again, the actual steering wheel steering angle is the perceived steering angle of the vehicle, and the measured steering wheel steering angle is the value collected by sensors (such as gyroscopes) on the vehicle. Furthermore, considering the measurement errors of the sensors, compensation can be made separately for the front wheel zero bias and the steering wheel zero bias. For example, the joint calibration result obtained in the above embodiment can be used as the compensation coefficient. Specifically, the steering wheel zero bias from the joint calibration result can be used directly for compensation, or the steering ratio from the joint calibration result can be used for compensation, and the steering ratio and front wheel zero bias can be combined for calculation to obtain the steering wheel zero bias. Alternatively, the steering wheel zero bias from the joint calibration result can be used directly for compensation. Since the steering wheel zero bias can be obtained from the steering ratio and the front wheel zero bias, the front wheel zero bias can be derived from the steering wheel zero bias and steering ratio in the joint calibration result, thereby compensating for the front wheel zero bias. In other words, depending on actual needs, the joint calibration result can be used in any combination, or any parameter from the joint calibration result can be selected, to compensate for relevant parameters in the autonomous driving scenario, such as front wheel zero bias, steering wheel zero bias, and steering ratio zero bias, or to compensate for parameters related to front wheel zero bias, steering wheel zero bias, and steering ratio zero bias, thereby improving the safety of autonomous driving.

[0063] In one application example, the joint calibration model in the above example performs joint online calibration of the two parameters, steering wheel zero bias and steering ratio, and models them into the joint calibration model. The input of the joint calibration model is the steering wheel zero bias coefficient and steering ratio coefficient mentioned above, and the output is the joint calibration result of steering wheel zero bias and steering ratio. This joint calibration model can also be called the joint estimation model of steering wheel zero bias and steering ratio (hereinafter referred to as the joint estimation model).

[0064] The mathematical modeling of the joint estimation model, and the online joint calibration of steering wheel zero bias and steering ratio based on the mathematical modeling of the joint estimation model, to calibrate steering wheel zero bias and steering ratio in real time, include the following:

[0065] (1) Joint estimation model

[0066] Through mathematical modeling using formula (1), the joint estimation model can be obtained as follows:

[0067]

[0068] In formula (1), Here is the steering wheel angle feedback value, and ρ is the steering ratio. To achieve zero steering wheel deviation, Let ρ represent the yaw rate, v be the vehicle speed, L be the vehicle wheelbase, and arctan() represent the arctangent function. In this joint estimation model, ρ and The value to be calibrated. v can be obtained through vehicle-mounted integrated inertial navigation. This can be obtained through chassis feedback, where L is an inherent vehicle parameter. When applying this joint estimation model, to prevent division by zero errors, the vehicle speed v > v0 is required. min Among them, v min The value is greater than 0, with an empirical value of 0.1 m / s. The application of this joint estimation model does not limit the straight-ahead state, as long as the vehicle speed v > v min It is applicable to all application scenarios.

[0069] The specific derivation process of the above formula (1) is as follows:

[0070]

[0071] in, The value represents the yaw rate, v is the vehicle speed, δ is the front wheel deflection angle, and L is the vehicle wheelbase.

[0072]

[0073] Let the steering wheel angle feedback value be... Steering wheel zero offset With a steering ratio of ρ, we obtain the following formula (4):

[0074]

[0075] Combining formulas (3) and (4), we can obtain the following formula (5):

[0076]

[0077] Based on the derivation of the above vehicle kinematics model, i.e., the above formulas (2)-(5), the above joint estimation model obtained by mathematical modeling through formula (1) can be obtained.

[0078] (2) Based on the mathematical modeling of the joint estimation model, the joint online calibration of steering wheel zero bias and steering ratio is realized to calibrate steering wheel zero bias and steering ratio in real time. For the joint estimation model shown in formula (1), the recursive least squares method can be used to calibrate the steering ratio ρ and steering wheel zero bias online. The steps are as follows:

[0079] The first step is to obtain the vehicle's yaw rate using an onboard integrated inertial navigation system. The vehicle speed v is used to obtain the actual steering wheel angle through chassis feedback.

[0080] The second step is to check if the system is booting up and initializing. If so, perform the initialization operation; otherwise, proceed directly to the third step.

[0081] The initialization operation is to initialize P0 = eps, where eps is a value greater than 0 and less than 1, with an empirical value of 0.0001;

[0082] The third step is to apply the recursive least squares method to calibrate the steering ratio ρ and the zero steering wheel deflection online.

[0083] In the iterative training, the (k+1)th online calibration iteration (k>=0) includes the following:

[0084] 1) Calculation The input vector is X k+1 =[x k+1 [, -1]; where x k+1 is the steering ratio coefficient, and -1 is the coefficient of steering wheel zero bias. That is, the input vector includes the steering ratio coefficient and the steering wheel zero bias coefficient. In other words, the input vector can be constructed using the steering ratio coefficient and the steering wheel zero bias coefficient.

[0085] 2) Perform recursive calculations using the recursive least squares method. First, calculate the intermediate variable P. k+1 , Next, calculate the output vector Y. k+1 ,

[0086] in, The actual steering wheel angle obtained during the k-th online calibration iteration. ρ k+1 The steering ratio during the (k+1)th online calibration. The zero bias of the steering wheel is set at the (k+1)th online calibration, and T is the transpose operator.

[0087] 3) Convergence criterion: Record the online calibration results of n iterations (n ​​is an integer greater than or equal to k+1), and calculate the ratio σ of the variance to the mean of the n results. If the steering ratio ρ and the steering wheel zero offset are... If the ratio σ (a specific form of the iterative processing result to be compared above) is less than a certain value ε (a specific form of the threshold), then the online calibration result is considered to have converged, and iterative training is stopped. The output result after convergence is the joint calibration result (i.e., the online calibrated steering ratio and steering wheel zero bias). Otherwise, return to step 1) to continue iterative training. Here, the empirical value of n is 10, and the empirical value of ε is 0.1.

[0088] Using this application example, online calibration and joint estimation of steering ratio and steering wheel zero-off are performed. This not only considers front wheel zero-off but also the influence of steering ratio on steering wheel zero-off, resulting in higher identification accuracy. In practical applications, a perfectly straight-ahead state is difficult to obtain. If the constraints on the straight-ahead state are too tight, the application scenarios are limited; if they are too loose, the calibration accuracy will be affected. A joint estimation model derived from the vehicle kinematics model is used, with clear physical meaning, as long as the vehicle speed is higher than v. min The application scenarios with an empirical value of 0.1 m / s are all applicable. This application example is not limited to straight-line driving, and its applicability is wider. Online calibration and joint estimation of steering ratio and steering wheel deflection can effectively solve the steering consistency problem of mass-produced vehicles, especially the consistency problem of different vehicles of the same model.

[0089] According to embodiments of this disclosure, a parameter calibration device is provided. Figure 8 This is a schematic diagram of the composition of a parameter calibration device according to an embodiment of the present disclosure, as shown below. Figure 8 As shown, the parameter calibration device includes: a construction module 801, used to construct an input vector of a joint calibration model based on a first coefficient corresponding to the steering wheel zero bias and a second coefficient corresponding to the steering ratio when the parameters to be calibrated are steering wheel zero bias and steering ratio; and a calibration module 802, used to output the joint calibration result of the steering wheel zero bias and the steering ratio based on the iterative training of the joint calibration model using the input vector.

[0090] In one embodiment, the calibration module 802 is used to: use the first coefficient and the second coefficient as constraints; perform joint online calibration of the steering wheel zero bias and the steering ratio under constraints according to the constraints, and determine the output calibration result as the joint calibration result of the steering wheel zero bias and the steering ratio.

[0091] In one embodiment, the system further includes an iterative processing module, configured to: obtain the iterative processing result corresponding to each iterative training when performing multiple iterative training sessions; preprocess the iterative processing result to obtain the iterative processing result to be compared; compare the iterative processing result to be compared with a threshold; if the iterative processing result to be compared is less than the threshold, stop the iterative training; and determine the output result obtained after stopping the iterative training as the calibration result of the output.

[0092] In one embodiment, the calibration module 802 is used to: perform a ratio calculation of variance and mean based on the iterative processing result, and use the ratio calculation result as the iterative processing result to be compared.

[0093] In one embodiment, the calibration module 802 is used to: perform weighted processing based on the iterative processing result, and use the weighted processing result as the iterative processing result to be compared.

[0094] In one embodiment, the calibration module 802 is used to: perform average averaging based on the iterative processing result, and use the average averaging result as the iterative processing result to be compared.

[0095] In one embodiment, the calibration module 802 is used to: perform the iterative training in the joint calibration model according to the input vector and the acquired control parameters; wherein the control parameters include at least one of the following: vehicle yaw rate, vehicle speed, steering wheel angle, and vehicle wheelbase.

[0096] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0097] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0098] Figure 9 A schematic block diagram of an example electronic device 900 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0099] like Figure 9As shown, the electronic device 900 includes a computing unit 901, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 902 or a computer program loaded into a random access memory (RAM) 903 from a storage unit 908. The RAM 903 may also store various programs and data required for the operation of the electronic device 900. The computing unit 901, ROM 902, and RAM 903 are interconnected via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0100] Multiple components in electronic device 900 are connected to I / O interface 905, including: input unit 906, such as keyboard, mouse, etc.; output unit 907, such as various types of displays, speakers, etc.; storage unit 908, such as disk, optical disk, etc.; and communication unit 909, such as network card, modem, wireless transceiver, etc. Communication unit 909 allows electronic device 900 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0101] The computing unit 901 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs the various methods and processes described above, such as parameter calibration methods. For example, in some embodiments, the parameter calibration method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 908. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 900 via ROM 902 and / or communication unit 909. When the computer program is loaded into RAM 903 and executed by the computing unit 901, one or more steps of the parameter calibration method described above may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform parameter calibration methods by any other suitable means (e.g., by means of firmware).

[0102] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0103] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0104] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0105] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0106] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0107] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0108] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this disclosure can be achieved, and this is not limited herein.

[0109] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A parameter calibration method, comprising: in a case where a parameter to be calibrated is a steering wheel zero offset and a steering ratio, constructing an input vector of a joint calibration model according to a first coefficient corresponding to the steering wheel zero offset and a second coefficient corresponding to the steering ratio; the steering ratio is a ratio of a steering wheel angle to a front wheel deflection angle; wherein the first coefficient is a steering wheel zero offset coefficient, and the second coefficient is a steering ratio coefficient; outputting a joint calibration result of the steering wheel zero offset and the steering ratio in iterative training of the joint calibration model according to the input vector, comprising: taking the steering wheel zero offset coefficient and the steering ratio coefficient as constraint conditions; performing joint online calibration under the constraint conditions of the steering wheel zero offset and the steering ratio and modeling into the joint calibration model, and determining the output calibration result as the joint calibration result of the steering wheel zero offset and the steering ratio; wherein performing joint online calibration under the constraint conditions of the steering wheel zero offset and the steering ratio and modeling into the joint calibration model comprises: mathematical modeling by the following formula to obtain the joint calibration model; ； is a steering wheel angle feedback value, is a steering ratio, is a steering wheel zero offset, represents a yaw rate, is a vehicle speed, is a vehicle wheelbase, in the joint calibration model, and is a value to be calibrated, , can be obtained through an on-board integrated inertial navigation system, can be obtained through chassis feedback, is a vehicle intrinsic parameter; is a second coefficient, and -1 is a first coefficient. 2.The method of claim 1, further comprising: in a case where the iterative training is performed multiple times, obtaining an iterative processing result corresponding to each of the iterative training; obtaining a to-be-compared iterative processing result after pre-processing the iterative processing result; comparing the to-be-compared iterative processing result with a threshold value, and in a case where the to-be-compared iterative processing result is less than the threshold value, stopping the iterative training, and determining an output result obtained after the iterative training is stopped as the output calibration result.

3. The method of claim 2, wherein, the pre-processing of the iterative processing result to obtain the to-be-compared iterative processing result comprises: performing variance-to-average value ratio operation on the iterative processing result, and taking a ratio operation result as the to-be-compared iterative processing result.

4. The method of claim 2, wherein, the pre-processing of the iterative processing result to obtain the to-be-compared iterative processing result comprises: performing weighted processing on the iterative processing result, and taking a weighted processing result as the to-be-compared iterative processing result.

5. The method of claim 2, wherein, the pre-processing of the iterative processing result to obtain the to-be-compared iterative processing result comprises: performing average value processing on the iterative processing result, and taking an average value processing result as the to-be-compared iterative processing result.

6. The method of any one of claims 1-5, wherein, the iterative training of the joint calibration model according to the input vector comprises: performing the iterative training in the joint calibration model according to the input vector and an obtained control parameter; wherein the control parameter comprises at least one of a vehicle yaw rate, a vehicle speed, a steering wheel angle, and a vehicle wheelbase. 7.A parameter calibration device, comprising: The construction module is configured to determine that the to-be-calibrated parameter is a steering wheel zero offset and a steering ratio, and construct an input vector of a joint calibration model according to a first coefficient corresponding to the steering wheel zero offset and a second coefficient corresponding to the steering ratio; the steering ratio is a ratio of a steering wheel angle to a front wheel deflection angle; the first coefficient is a steering wheel zero offset coefficient, and the second coefficient is a steering ratio coefficient. The calibration module is configured to: During iterative training of the joint calibration model according to the input vector, the steering wheel zero offset coefficient and the steering ratio coefficient are taken as constraint conditions in outputting a joint calibration result of the steering wheel zero offset and the steering ratio; the joint online calibration of the steering wheel zero offset and the steering ratio under the constraint conditions is modeled into the joint calibration model, and the output calibration result is determined as the joint calibration result of the steering wheel zero offset and the steering ratio, wherein the joint online calibration of the steering wheel zero offset and the steering ratio under the constraint conditions and modeling into the joint calibration model include: The mathematical modeling is performed by the following formula to obtain the joint calibration model; ； is a steering wheel angle feedback value, is a steering ratio, is a steering wheel zero offset, represents a yaw rate, is a vehicle speed, is a vehicle wheelbase, in the joint calibration model, and is a value to be calibrated, , can be obtained through an on-board integrated inertial navigation system, can be obtained through chassis feedback, is a vehicle intrinsic parameter; is a second coefficient, and -1 is a first coefficient.

8. The apparatus of claim 7, further comprising: The iteration processing module is configured to: In a case where the iterative training is performed multiple times, an iteration processing result corresponding to each time of the iterative training is obtained; After pre-processing the iteration processing result, a to-be-compared iteration processing result is obtained; The to-be-compared iteration processing result is compared with a threshold value, and in a case where the to-be-compared iteration processing result is less than the threshold value, the iterative training is stopped, and an output result obtained by stopping the iterative training is determined as the output calibration result.

9. The apparatus of claim 8, wherein, The calibration module is configured to: The ratio of the variance to the average value is calculated according to the iteration processing result, and the ratio calculation result is taken as the to-be-compared iteration processing result.

10. The apparatus of claim 8, wherein, The calibration module is configured to: The weighting processing result is taken as the to-be-compared iteration processing result.

11. The apparatus of claim 8, wherein, The calibration module is configured to: The average value processing result is taken as the to-be-compared iteration processing result.

12. The apparatus of any one of claims 7-11, wherein, The calibration module is configured to: The iterative training is performed in the joint calibration model according to the input vector and an obtained control parameter; The control parameter includes at least one of a vehicle yaw rate, a vehicle speed, a steering wheel angle, and a vehicle wheelbase.

13. An electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

14. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method of any one of claims 1-6.

15. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-6.

Images