Reversing control method, device and equipment and storage medium

Abstract

The invention provides a reversing control method and device, electronic equipment and a storage medium, and the method comprises the steps: determining the transverse position deviation and course angle deviation of a vehicle according to the relation among the position, course angle and expected trajectory of the vehicle in the reversing driving process of the vehicle along the expected trajectory; adjusting parameters of a state error weight matrix and / or a control input weight matrix in a model predictive control algorithm for backing-up control at least according to the numerical values of the transverse position deviation and the course angle deviation; wherein the parameters of the state error weight matrix are used for adjusting the tracking precision of the transverse position and the course angle of the vehicle, and the parameters of the control input weight matrix are used for adjusting the control smoothness of the vehicle; and according to the adjusted matrix, controlling the vehicle to continuously reverse along the expected track. Therefore, the tracking precision and the control smoothness can be self-adaptively balanced according to the actual driving state of the vehicle, and the adaptability, the stability, the safety and the control precision in the vehicle reversing process are improved.

Description

TECHNICAL FIELD

[0001] One or more embodiments of the present disclosure relate to the field of automatic driving, and in particular to a reverse control method and device, an electronic device, and a storage medium. BACKGROUND

[0002] Due to the limited field of view of the driver and the high complexity of operation in the reverse driving scenario, the demand for automatic reverse driving function is increasing, and the lateral control of the vehicle is a key link to realize the tracking of the reverse driving path. The current mainstream lateral control method usually adopts MPC (Model Predictive Control), which minimizes the deviation between the actual trajectory and the expected trajectory by constructing a vehicle dynamics model under the premise of meeting the system constraints. The MPC controller balances the control accuracy and the action smoothness by adjusting the state error weight matrix Q and the control input weight matrix R, so as to realize stable and reliable steering control.

[0003] However, the existing MPC control method generally adopts fixedly calibrated weight matrix parameters, which are difficult to balance the control robustness and response speed under different reverse driving conditions. When the initial position or the heading angle of the vehicle deviates greatly, the fixed parameters may cause the deviation correction to be not timely and the adjustment time to be too long, and when the deviation is small, the high gain is easy to cause overshoot or oscillation. Therefore, in the face of diversified initial states of reverse driving and real-time changing dynamic characteristics of the vehicle, the existing MPC control method cannot meet the demand of all conditions, which leads to the decline of control performance and affects the safety and user experience of reverse driving. SUMMARY

[0004] Therefore, the present disclosure provides a reverse control method, which comprises: During the reverse driving of the vehicle along an expected trajectory, the lateral position deviation and the heading angle deviation of the vehicle are determined according to the relationship among the position, the heading angle of the vehicle and the expected trajectory; At least according to the numerical values of the lateral position deviation and the heading angle deviation, the parameters of the state error weight matrix and / or the control input weight matrix in the model predictive control algorithm for reverse control are adjusted; wherein the parameters of the state error weight matrix are used to adjust the tracking accuracy of the lateral position and the heading angle of the vehicle, and the parameters of the control input weight matrix are used to adjust the control smoothness of the vehicle; According to the adjusted state error weight matrix and / or the control input weight matrix, the vehicle continues to drive along the expected trajectory.

[0005] Optionally, before the at least according to the numerical values of the lateral position deviation and the heading angle deviation, the parameters of the state error weight matrix and / or the control input weight matrix in the model predictive control algorithm for reverse control are adjusted, the method further comprises: determining a reference deviation value range of each of the lateral position deviation and the heading angle deviation, respectively; initially calibrating parameters of a state error weight matrix and a control input weight matrix in a model predictive control algorithm for the reversing control according to the reference deviation value ranges of the lateral position deviation and the heading angle deviation.

[0006] Optionally, the method further comprises: if the lateral position deviation falls into the corresponding reference deviation value range or the heading angle deviation falls into the corresponding reference deviation value range, the parameters of the state error weight matrix and the control input weight matrix are not adjusted; controlling the vehicle to continue reversing along the desired trajectory according to the initially calibrated matrix.

[0007] Optionally, the adjusting the parameters of the state error weight matrix and / or the control input weight matrix in the model predictive control algorithm for the reversing control according to the values of the lateral position deviation and the heading angle deviation comprises: if the lateral position deviation is less than a lower limit value of the corresponding reference deviation value range and the heading angle deviation is less than a lower limit value of the corresponding reference deviation value range, increasing the parameter of the control input weight matrix; wherein the increased parameter of the control input weight matrix is used to enhance the control smoothness of the vehicle.

[0008] Optionally, the adjusting the parameters of the state error weight matrix and / or the control input weight matrix in the model predictive control algorithm for the reversing control according to the values of the lateral position deviation and the heading angle deviation comprises: if the lateral position deviation is greater than an upper limit value of the corresponding reference deviation value range and the heading angle deviation is less than a lower limit value of the corresponding reference deviation value range, increasing the parameter of the state error weight matrix for adjusting the lateral position deviation; wherein the increased parameter of the state error weight matrix for adjusting the lateral position deviation is used to enhance the tracking accuracy of the lateral position of the vehicle.

[0009] Optionally, the adjusting the parameters of the state error weight matrix and / or the control input weight matrix in the model predictive control algorithm for the reversing control according to the values of the lateral position deviation and the heading angle deviation comprises: if the lateral position deviation is less than a lower limit value of the corresponding reference deviation value range and the heading angle deviation is greater than an upper limit value of the corresponding reference deviation value range, increasing the parameter of the state error weight matrix for adjusting the heading angle deviation; Among them, the parameter for adjusting the heading angle deviation in the increased state error weight matrix is ​​used to enhance the tracking accuracy of the vehicle's heading angle.

[0010] Optionally, adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm used for reversing control, at least based on the values ​​of the lateral position deviation and the heading angle deviation, includes: If the lateral position deviation is greater than the upper limit of the corresponding reference deviation range, and the heading angle deviation is greater than the upper limit of the corresponding reference deviation range, then it is determined whether the lateral position deviation and the heading angle deviation are in the same direction. If the two directions are consistent, then the parameter for adjusting the lateral position deviation in the state error weight matrix is ​​decreased, and the parameter for adjusting the heading angle deviation in the state error weight matrix is ​​increased; wherein, the parameter for adjusting the lateral position deviation in the decreased state error weight matrix is ​​used to weaken the adjustment of the vehicle's lateral position tracking accuracy, and the parameter for adjusting the heading angle deviation in the increased state error weight matrix is ​​used to strengthen the adjustment of the vehicle's heading angle tracking accuracy. If the two directions are inconsistent, the parameters of the control input weight matrix are reduced; wherein, the reduced parameters of the control input weight matrix are used to weaken the control smoothness of the vehicle.

[0011] This disclosure also provides a reversing control device, the device comprising: The determining unit is used to determine the lateral position deviation and heading angle deviation of the vehicle based on the relationship between the vehicle position, heading angle and the desired trajectory during the process of the vehicle reversing along the desired trajectory. An adjustment unit is configured to adjust the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm used for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation; wherein the parameters of the state error weight matrix are used to adjust the tracking accuracy of the vehicle's lateral position and heading angle, and the parameters of the control input weight matrix are used to adjust the control smoothness of the vehicle. The control unit is used to control the vehicle to continue reversing along the desired trajectory based on the adjusted state error weight matrix and / or control input weight matrix.

[0012] This disclosure also provides an electronic device, including a communication interface, a processor, a memory, and a bus, wherein the communication interface, the processor, and the memory are interconnected via the bus; The memory stores machine-readable instructions, and the processor executes the above method by invoking the machine-readable instructions.

[0013] This disclosure also provides a machine-readable storage medium storing machine-readable instructions that, when called and executed by a processor, implement the above-described method.

[0014] Therefore, this disclosure obtains the lateral position deviation and heading angle deviation of the vehicle based on the relationship between the vehicle position, heading angle and desired trajectory during the reversing process, and adjusts the parameters of the state error weight matrix and control input weight matrix in the model predictive control algorithm at least according to the deviation values, and finally controls the vehicle to drive along the desired trajectory based on the adjusted matrix.

[0015] By means of the above methods, this disclosure enables the controller to adaptively optimize control behavior according to different deviation conditions by adjusting the parameters of the state error weight matrix and the control input weight matrix in real time. For example, it prioritizes rapid correction when there is a large deviation and focuses on control smoothness when there is a small deviation. In this way, it can adaptively balance tracking accuracy and control smoothness according to the actual driving state of the vehicle, thereby improving the adaptability, stability, safety and control accuracy of the vehicle during reversing. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram illustrating a vehicle body coordinate system as an exemplary embodiment; Figure 2 This is a flowchart illustrating a reversing control method as an exemplary embodiment; Figure 3 This is a schematic diagram illustrating the positional relationship between a vehicle and a desired trajectory, as shown in an exemplary embodiment. Figure 4 This is a schematic diagram illustrating the positional relationship between a vehicle and a desired trajectory, as shown in an exemplary embodiment. Figure 5 This is a schematic diagram illustrating the mapping relationship between lateral position deviation and adjustment parameters, as shown in an exemplary embodiment. Figure 6 This is an exemplary embodiment illustrating another mapping relationship between lateral position deviation and adjustment parameters; Figure 7 This is a schematic diagram illustrating the mapping relationship between heading angle deviation and adjustment parameters, as shown in an exemplary embodiment. Figure 8This is an exemplary embodiment illustrating the hardware structure of an electronic device; Figure 9 This is a block diagram illustrating a reversing control device as an exemplary embodiment. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions in this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this disclosure.

[0019] It should be noted that the steps of the corresponding methods are not necessarily performed in the order shown and described in this disclosure in other embodiments. In some other embodiments, the methods may include more or fewer steps than those described in this disclosure. Furthermore, a single step described in this disclosure may be broken down into multiple steps in other embodiments; and multiple steps described in this disclosure may be combined into a single step in other embodiments.

[0020] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating a vehicle body coordinate system as an exemplary embodiment. (As shown) Figure 1 As shown, the goal of lateral control during vehicle reversing is typically to reduce the lateral distance error between the vehicle and the desired trajectory by controlling the rotation of the steering wheel during reversing. and heading angle deviation Minimum. Here, the lateral distance error is the distance between the vehicle's center of gravity and the nearest trajectory point on the desired trajectory (or the pre-aiming point if pre-aiming control is considered) along the vehicle's Y-axis (e.g., ...). Figure 1 As shown, the deviation is in the direction perpendicular to the vehicle body. The heading angle deviation is the deviation between the vehicle's own heading direction and the direction of the nearest trajectory point on the desired trajectory (or the aiming point if aiming control is considered).

[0021] Constructing lateral distance error during vehicle reversing and heading angle deviation The state-space equations can be obtained as follows: in, This refers to the lateral distance error of the vehicle. This represents the rate of change of the vehicle's lateral distance error. This represents the vehicle's heading angle deviation. The rate of change of the vehicle's heading angle deviation. This refers to the lateral strength of the front axle tires of the vehicle. For the lateral strength of the vehicle's rear axle tires, This is the distance from the vehicle's center of gravity to the front axle. This is the distance from the vehicle's center of gravity to the rear axle. For the overall vehicle quality, For the vehicle along the X-axis (e.g.) Figure 1 The speed at which the vehicle is moving forward. For the vehicle around the Z-axis (e.g.) Figure 1 The moment of inertia (perpendicular to the ground). This refers to the front wheel steering angle. Let R be the rate of change of heading corresponding to the road radius R.

[0022] After discretizing the above state-space equations and undergoing a series of transformations, the cost function can be established as follows: in, For reference state value, Here is the state error weight matrix. To control the input weight matrix, For future control of the time domain.

[0023] The control objective of the MPC controller is to find the optimal control strategy within a finite future timeframe, based on the current state of the system, at the current moment. This strategy should minimize the defined cost function J while satisfying all system constraints, and then execute the first step of the strategy. Minimizing the cost function J is the most fundamental mathematical objective of the MPC controller. As an "optimizer," the controller's task is to find a set of decision variables that minimizes the value of the objective function J.

[0024] In the above cost function, the state error weight matrix The relative importance of state tracking accuracy is determined by the number of elements on the diagonal; the larger the element, the more closely the corresponding state variable needs to track the reference value. The state error weight matrix is ​​typically used to determine this. The matrix is ​​labeled as a diagonal matrix (or a positive semi-definite matrix) as follows: in, To control the weighting of the lateral distance error of the vehicle during reversing in the system, To control the weight of the rate of change of the lateral distance error of the vehicle during reversing in the system, To control the weight of the vehicle's reversing heading angle deviation in the system, This is the weight for the rate of change of the vehicle's reversing heading angle deviation in the control system.

[0025] In the above cost function, the control input weight matrix This determines the relative importance of smooth control actions and energy efficiency, which is crucial for preventing system oscillations and saving energy. The control weight matrix is ​​typically used... Labeled as a diagonal matrix: in, The weights are used to control the vehicle's steering.

[0026] For a reversing control system, the state error weight matrix This is used to adjust the priority of each state variable (such as lateral position deviation and its rate of change, heading angle deviation and its rate of change) in the optimization objective; a larger value indicates a higher requirement for the tracking accuracy of that state; control input weight matrix. This is used to penalize changes in control inputs (such as front wheel steering angle); a larger value results in a smoother system response but may sacrifice response speed. Both factors together determine the dynamic performance and stability of the MPC controller. In practical applications, for... and By performing actual calibration, it is possible to achieve tracking and control of the vehicle's desired trajectory.

[0027] However, existing MPC control methods generally use fixed-calibration weight matrix parameters, which makes it difficult to balance control robustness and response speed under different reversing conditions.

[0028] In view of this, the present disclosure aims to propose a technical solution for dynamically adjusting the parameters of the MPC weight matrix based on the vehicle's lateral position deviation and heading angle deviation.

[0029] The present disclosure is described below through specific embodiments and in conjunction with specific application scenarios.

[0030] Please see Figure 2 , Figure 2 This is a flowchart illustrating a reversing control method as an exemplary embodiment. The method may perform the following steps: Step 202: During the reverse driving of the vehicle along the desired trajectory, determine the lateral position deviation and heading angle deviation of the vehicle based on the relationship between the vehicle position, heading angle and the desired trajectory.

[0031] For example, see Figure 3 , Figure 3 This is a schematic diagram illustrating the positional relationship between a vehicle and a desired trajectory, as shown in an exemplary embodiment. Figure 3As shown, during reversing, the vehicle exhibits several typical postures relative to the desired trajectory (black dashed line): the lateral position deviation can be positive (vehicle is on the left side of the desired trajectory), zero (vehicle is centered on the desired trajectory), or negative (vehicle is on the right side of the desired trajectory); the heading angle deviation can also be positive (vehicle head is offset counterclockwise towards the tangent of the desired trajectory), zero (vehicle head is aligned with the desired trajectory), or negative (vehicle head is offset clockwise towards the tangent of the desired trajectory). These two combinations form nine typical operating conditions. The current vehicle centroid position and heading angle can be obtained through the vehicle positioning system. Combined with the desired trajectory in the high-precision map, the distance from the vehicle centroid to the nearest trajectory point (or the pre-aiming point if pre-aiming control is considered) in the direction perpendicular to the trajectory tangent is calculated as the lateral position deviation, and the angle between the vehicle's current heading angle and the trajectory tangent direction is calculated as the heading angle deviation. This provides a basis for subsequent dynamic adjustment of the MPC weight matrix.

[0032] The desired trajectory, generated by the upper-level path planning module, is typically a smooth curve representing the ideal reversing path that the vehicle's center of gravity should follow under ideal conditions. It may originate from a parking path automatically generated by the automatic parking system or a reference route manually specified by the driver. This trajectory is usually stored in the controller as a discrete point sequence or a parameterized function. Lateral position deviation reflects whether the vehicle deviates from the predetermined path. Heading angle deviation reflects the degree of consistency between the vehicle's direction of travel and the desired trajectory.

[0033] Step 204: Adjust the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm used for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation; wherein the parameters of the state error weight matrix are used to adjust the tracking accuracy of the vehicle's lateral position and heading angle, and the parameters of the control input weight matrix are used to adjust the control smoothness of the vehicle.

[0034] For example, throughout the reversing process, the system continuously evaluates the degree of matching between the vehicle's current state and the desired trajectory, and adjusts the state error weight matrix based on the values ​​of the vehicle's lateral position deviation and heading angle deviation. and control input weight matrix The value of at least one term in the weight matrix. State error weight matrix. For a diagonal matrix, increase The value of a particular term indicates that the controller prioritizes suppressing the error associated with that term, thereby accelerating response speed and improving tracking accuracy. When the lateral position deviation is large, increasing... Weight of lateral distance error when reversing a vehicle This allows the system to quickly correct its lateral position. When the vehicle's reversing heading angle deviation is large, it improves... Weight of vehicle reversing heading angle deviation This enables the system to quickly correct its heading angle. Control input weight matrix Typically simplified to a scalar or low-dimensional matrix, representing the penalty applied to control actions (such as the front wheel steering angle and its rate of change). When lateral position deviation or reversing heading angle deviation is small, the control input weight matrix is ​​increased. Adjusting the parameters will make the control behavior smoother, reduce sharp steering, and benefit comfort and actuator lifespan protection; when the lateral position deviation or reversing heading angle deviation is large, increasing the control input weight matrix .... The parameters allow for greater control freedom, making it suitable for emergency correction scenarios.

[0035] Model predictive control (MMC) is an optimal control strategy based on a dynamic model. It solves a quadratic programming problem in a finite-time domain online, outputs the optimal control sequence for the next few steps, and executes only the first control command. Its core cost function consists of two terms: a state error term and a... With control input items The weights are respectively composed of the state error weight matrix. and control input weight matrix Weighting is applied.

[0036] In addition to adjusting matrix parameters based on the values ​​of lateral position deviation and heading angle deviation, other auxiliary variables (such as vehicle speed, road surface adhesion coefficient, historical error trends, or directional consistency of lateral position deviation and heading angle deviation) can be introduced for comprehensive judgment. Methods for adjusting parameters include multiplying the elements of the base matrix by a dynamically adjusted scaling factor, table lookup, piecewise function mapping, fuzzy logic reasoning, or neural network prediction.

[0037] Step 206: Based on the adjusted state error weight matrix and / or control input weight matrix, control the vehicle to continue reversing along the desired trajectory.

[0038] For example, during vehicle reversing, due to a large lateral position deviation, the controller will adjust the state error weight matrix. Weight of lateral distance error when reversing a vehicle Increased from the default value of 1.0 to 1.8. Adjusted state error weight matrix. The input to the MPC solver generates more aggressive steering commands, causing the steering wheel to turn quickly to the right to correct the position. Subsequently, under the action of the updated control law, the vehicle gradually converges to the desired trajectory and resumes smooth control when it is close to alignment, avoiding overshoot or oscillation. Through the aforementioned adjustments, the vehicle ultimately achieves stable and smooth tracking of the desired trajectory.

[0039] Among them, "according to the adjusted state error weight matrix and / or control input weight matrix" refers to the dynamically corrected state error weight matrix. and control input weight matrix Substituting the values ​​into the cost function of the MPC algorithm, the optimization problem is solved again to generate a new optimal control sequence. "Controlling the vehicle" refers to sending the calculated initial control quantity (usually the front wheel steering angle command) to the EPS (Electric Power Steering) or other actuators to drive the vehicle to perform the corresponding action. As the vehicle continues to reverse, the entire control process iterates continuously, with each step repeatedly executing deviation detection, parameter adjustment, and control output, forming a closed-loop feedback.

[0040] In one embodiment shown, before adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control based at least on the values ​​of the lateral position deviation and the heading angle deviation, the method further includes: determining reference deviation value ranges for the lateral position deviation and the heading angle deviation, respectively; and performing initial calibration on the parameters of the state error weight matrix and the control input weight matrix in the model predictive control algorithm for reversing control based on the reference deviation value ranges of the lateral position deviation and the heading angle deviation.

[0041] For example, the reference deviation range is a pre-defined threshold interval used to define the lateral position deviation range and heading angle deviation range under "typical reversing conditions". For instance, during the development phase of the MPC controller, the reference deviation range for lateral position deviation is determined to be [-0.15m, -0.05m] and [0.05m, 0.15m], and the reference deviation range for heading angle deviation is determined to be [-10°, -5°] and [5°, 10°]. These two ranges cover most smooth reversing scenarios. Subsequently, through real-vehicle testing or high-fidelity simulation, different... and The tracking performance and control smoothness under parameter combinations were analyzed, and finally, an optimal set of initial calibration state error weight matrices was obtained. and control input weight matrix : Includes 4 coefficients , , , The values ​​were 1.2, 0.4, 1.0, and 0.2, respectively. Includes 1 coefficient The value is 0.7. This set of parameters represents a baseline configuration that balances accuracy and comfort within the reference deviation range, and serves as the starting point for subsequent dynamic adjustment of parameters based on real-time deviations.

[0042] The reference deviation ranges for lateral position deviation and heading angle deviation can be determined through real-vehicle calibration experiments based on different vehicle models, parking scenarios (such as parallel parking and perpendicular parking), and road conditions, and stored in the controller's non-volatile memory. Initial calibration refers to determining a set of basic weight parameters suitable for "typical reversing conditions" under typical operating conditions, based on a large amount of real-vehicle test data or simulation platform optimization. These parameters constitute the initial calibration state error weight matrix. and control input weight matrix This process can be completed using an automatic optimization algorithm. After calibration, and This serves as a benchmark for subsequent online dynamic adjustments, ensuring relatively stable parameter adjustments.

[0043] In one embodiment shown, the method further includes: if the lateral position deviation falls within the corresponding reference deviation value range, or the heading angle deviation falls within the corresponding reference deviation value range, then the parameters of the state error weight matrix and the control input weight matrix are not adjusted; and the vehicle is controlled to continue reversing along the desired trajectory according to the initially calibrated matrix.

[0044] For example, if the vehicle's lateral position deviation during reversing is +0.18 m and its heading angle deviation is +3°, both of which fall outside the reference deviation range (the reference deviation range for lateral position deviation is [0.05 m, 0.15 m] and [-0.15 m, -0.05 m], and the reference deviation range for heading angle deviation is [5°, 10°] and [-10°, -5°]), the system determines that the current operating condition is an "atypical reversing condition" and will trigger the parameter adjustment mechanism of the weight matrix. Conversely, if the lateral position deviation is +0.06 m or the heading angle deviation is +7°, at least one of which falls within the aforementioned preset reference deviation range, the controller determines that the vehicle is in a typical smooth reversing condition and maintains the initially calibrated state error weight matrix. and control input weight matrix If the parameters remain unchanged, use this set of parameters directly to solve the MPC problem and output steering commands to control the vehicle to continue reversing along the desired trajectory.

[0045] Please see Figure 4 , Figure 4 This is a schematic diagram illustrating the positional relationship between a vehicle and a desired trajectory, as shown in an exemplary embodiment. Figure 4 As shown, there are three situations that belong to the "typical reversing condition" and the weight matrix parameters are not adjusted: Scenario 1: If the lateral position deviation during vehicle reversing is less than the lower limit of the reference deviation range corresponding to the lateral position deviation, and the heading angle deviation is within the corresponding reference deviation range (i.e., "small lateral deviation, moderate heading deviation"), the weight matrix parameters are not adjusted. This is because the lateral deviation is already very close to the desired trajectory, and the "moderate deviation" in heading is within the "natural fluctuation range" of the reversing process. If the heading weight is adjusted at this time, it is easy for the vehicle to make unnecessary steering movements in order to "pursue the ultimate heading," which would disrupt the smoothness of reversing—maintaining the status quo is the optimal solution to "maintain stability at the lowest cost."

[0046] Scenario 2: If the lateral position deviation of the vehicle during reversing is within the corresponding reference deviation range (i.e., "moderate lateral deviation"), while the heading angle deviation is arbitrary, it indicates that the current control strategy for lateral deviation is effective (lateral deviation is not out of control). Blindly adjusting the weight matrix parameters may break this balance, so "no adjustment" is chosen to maintain the continuity of control.

[0047] Scenario 3: If the lateral position deviation during vehicle reversing exceeds the upper limit of the corresponding reference deviation range (i.e., "large lateral deviation"), while the heading angle deviation is within the corresponding reference deviation range (i.e., "large lateral deviation, moderate heading deviation"), the weight matrix parameters are not adjusted. This is because although the lateral deviation is large, the heading deviation is relatively stable. In this case, the "moderate heading" provides a "reference direction" for lateral correction—adjusting the weight matrix parameters might cause a conflict between lateral correction and heading stability. Maintaining the existing weight matrix parameters allows the controller to use a stable heading as a reference when "correcting lateral deviation," avoiding a trade-off.

[0048] The reference deviation range is a pre-defined set of threshold intervals used to define the "acceptable" or "normally controllable" error level and to define a "typical reversing condition." The reference deviation range typically includes a lower limit and an upper limit, forming a band-shaped area that allows for fluctuations. When the lateral position deviation or heading angle deviation falls within this range, it indicates that the vehicle's current position is relatively ideal and no drastic adjustments are needed.

[0049] In this embodiment, by completing the initial calibration of the matrix based on the reference deviation range during the development phase, and only initiating the adjustment of the weight matrix parameters during runtime under "atypical reversing conditions", the control stability under "typical reversing conditions" is ensured, while parameter drift and system oscillation caused by frequent adjustments are avoided. This effectively suppresses high-frequency oscillation and over-adjustment, improves the smoothness of the reversing process and ride comfort, and reduces the wear and energy consumption of the actuator.

[0050] During the reversing motion of the vehicle along the desired trajectory, if the vehicle is determined to be in a "typical reversing condition" based on its lateral position deviation and heading angle deviation, then this disclosure requires adjusting the initially calibrated state error weight matrix. and control input weight matrix The parameters in the settings can be adjusted, and the adjustment method can be weighted by the scaling factor. Details are as follows: For the initially calibrated state error weight matrix The design includes , , and The scaling factor of the four coefficients The four coefficients have a default value of 1.0, and are adjusted accordingly. The four weights are adjusted as follows: For the initially calibrated control input weight matrix The design includes A coefficient's scaling factor ,coefficient The default value is 1.0, which corresponds to adjusting the control input weight matrix. The steering control input weights are adjusted as follows: It can be seen that, under the "atypical reversing condition", the state error weight matrix actually used by the MPC algorithm is... and control input weight matrix They are respectively: In the formula, the initial calibration state error weight matrix and the initially calibrated control input weight matrix The calibration was obtained under "typical reversing conditions". , The adjustment rules can be set based on experience or data-driven methods.

[0051] The following describes the procedures for "atypical reversing conditions". , The adjustment rules.

[0052] In one embodiment shown, adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation, includes: increasing the parameters of the control input weight matrix if the lateral position deviation is less than the lower limit of the corresponding reference deviation range and the heading angle deviation is less than the lower limit of the corresponding reference deviation range; wherein the increased parameters of the control input weight matrix are used to enhance the control smoothness of the vehicle.

[0053] For example, such as Figure 4 As shown, if the lateral position deviation of the vehicle during reversing... Less than the lower limit of the reference deviation range corresponding to the lateral position deviation. And the heading angle deviation Less than the lower limit of the corresponding reference deviation range (i.e., "small lateral error, small heading error"), then increase the control input weight matrix. parameters This enhances the smoothness of vehicle control.

[0054] Among them, the control input weight matrix In the MPC algorithm, this matrix is ​​used to penalize changes in control inputs. Its function is to constrain the magnitude of changes in steering commands or acceleration, preventing drastic fluctuations in control output. Increasing this matrix... The parameters will enhance the suppression of changes in steering wheel angle, making steering actions slower and smoother.

[0055] In this embodiment, when the controller detects that both the lateral position deviation and the heading angle deviation are less than the lower limit of their respective reference ranges, it indicates that the vehicle has basically aligned with the target trajectory. Continuing to correct the deviation with high response gain may lead to frequent fine adjustments in the steering system, causing high-frequency oscillations or steering wheel "shaking," affecting ride comfort and actuator lifespan. At this time, the control system should prioritize smooth operation rather than further improving tracking accuracy.

[0056] This embodiment increases the control input weight matrix. parameters This method enhances the smoothness of steering input adjustments. By reducing unnecessary high-frequency control outputs, it effectively suppresses oscillations and over-adjustment within small error ranges, improving driving comfort and reducing the risk of mechanical wear. Furthermore, this method requires no redesign of the controller structure, is compatible with existing MPC frameworks, and possesses good engineering practicality.

[0057] In one embodiment shown, adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation, includes: if the lateral position deviation is greater than the upper limit of the corresponding reference deviation range and the heading angle deviation is less than the lower limit of the corresponding reference deviation range, then increasing the parameter for adjusting the lateral position deviation in the state error weight matrix; wherein, the increased parameter for adjusting the lateral position deviation in the state error weight matrix is ​​used to enhance the tracking accuracy of adjusting the lateral position of the vehicle.

[0058] For example, see Figure 4 , Figure 4 This is a schematic diagram illustrating the positional relationship between a vehicle and a desired trajectory, as shown in an exemplary embodiment. Figure 4 As shown, if the lateral position deviation during the vehicle reversing process is different... The value is greater than the upper limit of the reference deviation range corresponding to the lateral position deviation. And the heading angle deviation Less than the lower limit of the corresponding reference deviation range (i.e., "large lateral deviation, small heading deviation"), then increase the state error weight matrix. Parameters for adjusting lateral position deviation This enhances the tracking accuracy of adjusting the vehicle's lateral position. State error weight matrix. Parameters for adjusting lateral position deviation Numerical settings and lateral position deviation related.

[0059] Please see Figure 5 , Figure 5 This is a schematic diagram illustrating the mapping relationship between lateral position deviation and adjustment parameters, as shown in an exemplary embodiment. Figure 5 As shown in the figure, this illustrates a typical nonlinear gain mapping relationship, with the horizontal axis representing the input variable (such as lateral positional deviation). absolute value The vertical axis represents output variables (such as parameters that adjust the lateral position deviation). The curve exhibits an S-shaped growth characteristic: (Regarding input variables...) Less than the upper limit of the reference deviation range corresponding to the lateral position deviation When adjusting the parameters of lateral position deviation The default value is 1.0; in the input variable Exceeding the upper limit of the reference deviation range corresponding to the lateral position deviation. When the input increases, the output rises rapidly, indicating that the controller gradually increases its sensitivity to the error; when the input reaches a certain threshold... Afterwards, the output tends to saturate and remains constant. The value remains unchanged to prevent excessive control from causing system oscillations.

[0060] Specifically, when the lateral position deviation is large while the heading angle deviation is small, the state error weight matrix is ​​increased. Parameters for adjusting lateral position deviation This allows the controller to increase its control over lateral position tracking. By enhancing sensitivity to lateral deviations and errors, the system can drive the vehicle back to the desired trajectory more quickly, shortening adjustment time and avoiding reversing failures or safety hazards caused by slow response, thus improving the robustness and practicality of reversing control.

[0061] In this embodiment, when the controller detects that the vehicle has deviated significantly from the desired trajectory but its heading is correct, it prioritizes increasing the lateral position control weight. This allows the MPC controller to concentrate resources to quickly reduce the lateral deviation while maintaining the original heading stability. This adaptive weight adjustment mechanism based on deviation status achieves a reasonable allocation of control degrees of freedom.

[0062] In one embodiment shown, adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation, includes: if the lateral position deviation is less than the lower limit of the corresponding reference deviation range and the heading angle deviation is greater than the upper limit of the corresponding reference deviation range, then increasing the parameter for adjusting the heading angle deviation in the state error weight matrix; wherein, the increased parameter for adjusting the heading angle deviation in the state error weight matrix is ​​used to enhance the tracking accuracy of adjusting the heading angle of the vehicle.

[0063] For example, see Figure 4 , Figure 4 This is a schematic diagram illustrating the positional relationship between a vehicle and a desired trajectory, as shown in an exemplary embodiment. Figure 4 As shown, if the lateral position deviation of the vehicle during reversing... Less than the lower limit of the reference deviation range corresponding to the lateral position deviation. And the heading angle deviation Greater than the upper limit of the corresponding reference deviation range (i.e., "small lateral deviation, large heading deviation"), then increase the state error weight matrix. Parameters for adjusting heading angle deviation This improves the tracking accuracy of adjusting the vehicle's heading angle.

[0064] Specifically, when the system detects a small lateral position deviation, it indicates that the vehicle is basically near the desired trajectory. However, if the heading angle deviation is large, it indicates that the driving direction has significantly deviated from the trajectory. If the heading is not corrected in time, even if the current position is good, a new lateral deviation will quickly occur during the reversing process due to the incorrect direction, leading to trajectory divergence or even a collision risk. Therefore, the controller needs to identify such conditions and respond dynamically.

[0065] A large deviation in heading angle means that the vehicle's attitude deviates significantly from the expected direction, which may lead to oversteering or path oscillation during subsequent reversing. Increasing the state error weight matrix... Parameters for adjusting heading angle deviation This error can be penalized more severely in MPC optimization, prompting the controller to adjust the front wheel steering angle, allowing the vehicle to align with the target trajectory more quickly. This adjustment can be completed online without interrupting the control flow. After adjustment, MPC will focus more on eliminating heading angle deviation, applying a larger steering input in the short term. Since the lateral position deviation is small at this point, appropriately reducing the position control weight will not cause significant position oscillations; instead, it helps to fundamentally avoid future deviation accumulation.

[0066] In this embodiment, when the vehicle approaches the desired trajectory but the heading deviates significantly, the weight of the heading angle is increased, and the controller can focus more on directional alignment, effectively suppressing trajectory divergence and improving the long-term stability and safety of the reversing process.

[0067] In one embodiment, adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation, includes: if the lateral position deviation is greater than the upper limit of the corresponding reference deviation range and the heading angle deviation is greater than the upper limit of the corresponding reference deviation range, then determining whether the lateral position deviation and the heading angle deviation are in the same direction; if they are in the same direction, then decreasing the parameter for adjusting the lateral position deviation in the state error weight matrix and increasing the parameter for adjusting the heading angle deviation in the state error weight matrix; wherein, the parameter for adjusting the lateral position deviation in the decreased state error weight matrix is ​​used to weaken the lateral position tracking accuracy of the vehicle, and the parameter for adjusting the heading angle deviation in the increased state error weight matrix is ​​used to strengthen the heading angle tracking accuracy of the vehicle; if they are not in the same direction, then decreasing the parameter of the control input weight matrix; wherein, the parameter of the decreased control input weight matrix is ​​used to weaken the control smoothness of the vehicle.

[0068] For example, see Figure 4 , Figure 4This is a schematic diagram illustrating the positional relationship between a vehicle and a desired trajectory, as shown in an exemplary embodiment. Figure 4 As shown, if the lateral position deviation of the vehicle during reversing... The value is greater than the upper limit of the reference deviation range corresponding to the lateral position deviation. And the heading angle deviation Greater than the upper limit of the corresponding reference deviation range (That is, "large lateral deviation, large heading deviation"), then determine the lateral position deviation. and heading angle deviation Are their directions consistent? If they are consistent, then decrease the state error weight matrix. Parameters for adjusting lateral position deviation This reduces the lateral position tracking accuracy of the vehicle and increases the state error weight matrix. Parameters for adjusting heading angle deviation This is to improve the accuracy of the vehicle's heading angle tracking. If the two directions are inconsistent, the control input weight matrix is ​​reduced. parameters This reduces the smoothness of vehicle control.

[0069] Please see Figure 6 , Figure 6 This is an exemplary embodiment illustrating another mapping relationship between lateral position deviation and adjustment parameters. For example... Figure 6 As shown in the figure, this plot illustrates a typical nonlinear decay curve, with the horizontal axis representing the input variable (such as lateral positional deviation). absolute value The vertical axis represents output variables (such as parameters that adjust the lateral position deviation). The curve shows a trend of first flattening and then rapidly decreasing: (Regarding input variables) Less than the upper limit of the reference deviation range corresponding to the lateral position deviation When adjusting the parameters of lateral position deviation The default value is 1.0, and the system maintains the default sensitivity to lateral position errors; in the input variables Exceeding the upper limit of the reference deviation range corresponding to the lateral position deviation. As the input increases, the output begins to decay exponentially, indicating that the controller gradually reduces its sensitivity to the error; when the input reaches a certain threshold... Afterwards, the output tends to a lower stable value. And keep it unchanged, thereby prioritizing rapid correction of the heading angle and improving control stability and safety.

[0070] Please see Figure 7 , Figure 7This is a schematic diagram illustrating the mapping relationship between heading angle deviation and adjustment parameters, as shown in an exemplary embodiment. Figure 7 As shown in the figure, this plot illustrates a typical non-linear growth curve, with the horizontal axis representing the input variable (such as heading angle deviation). absolute value The vertical axis represents output variables (such as parameters for adjusting heading angle deviation). The curve exhibits an S-shaped characteristic: "first flattening out, then accelerating upwards, and finally saturating." This is true for the input variables. Less than the upper limit of the corresponding reference deviation range At that time, adjust the parameters of the heading angle deviation. The default value is 1.0, and the system maintains the default sensitivity to heading angle deviation; in the input variables Exceeding the upper limit of the corresponding reference deviation range When the input increases, the output rises rapidly, indicating that the controller gradually increases its sensitivity to the error, achieving adaptive control; when the input... Reaching a certain threshold Afterwards, the output tends to a stable value. It remains unchanged, thus preventing excessive control from causing oscillations and balancing stability and response efficiency.

[0071] When both lateral position deviation and heading angle deviation exceed the upper limit of their respective reference ranges, it indicates that the vehicle has significantly deviated from the expected trajectory and entered a large error condition. At this point, it is necessary to determine the consistency of their directions: if the signs are the same (e.g., lateral deviation to the left and heading to the left), they are considered to be in the same direction; if the signs are opposite (e.g., lateral deviation to the left and heading to the right), they are considered to be incompatible. This determination can be achieved by directly comparing the signs of the deviations, or by using a normalized vector dot product for quantitative evaluation.

[0072] When the directions are consistent, it usually reflects the orderly deviation of the vehicle as a whole. If the lateral position is forcibly corrected first, the deviation may be aggravated by the superposition of steering. Therefore, this embodiment chooses to reduce the weight of lateral position deviation, appropriately relax its tracking accuracy requirements, and increase the weight of heading angle deviation, so that the control is dominated by directional correction - similar to the driving experience of "first straighten the direction, then correct the position", which helps to improve convergence stability.

[0073] When directions are inconsistent, it indicates a contradiction in attitude and a risk of adjustment conflict, making the system prone to oscillation or correction failure. In such cases, traditional fixed-weight MPC often responds slowly. This embodiment reduces the parameters of the control input weight matrix, actively relaxing smoothness constraints and granting the controller greater intervention freedom to quickly break the deadlock. For example, the value can be temporarily reduced to 60%~80% of the original value, thereby outputting a stronger steering command. This adjustment can be conditionally triggered or combined with fuzzy logic to achieve gradual adjustment.

[0074] Through the above steps, this embodiment achieves intelligent adaptive control under dual large deviation conditions: when the lateral and heading deviations are in the same direction, heading correction takes the lead to avoid instability caused by sudden position correction; when they are in opposite directions, the control input constraints are reduced to release stronger steering capability to break the adjustment deadlock. This method does not require modification of the basic MPC structure, and can effectively cope with complex reversing postures by dynamically adjusting the weight parameters. On the one hand, it improves the robustness and response efficiency of trajectory tracking; on the other hand, it has low engineering implementation cost and good practicality and expansion potential.

[0075] Corresponding to the embodiments of the above-described reversing control method, this disclosure also provides an embodiment of a reversing control device.

[0076] Please see Figure 8 , Figure 8 This is an exemplary embodiment illustrating the hardware structure of an electronic device. At the hardware level, the device includes a processor 802, an internal bus 804, a network interface 806, memory 808, and non-volatile memory 810, and may also include other necessary hardware. One or more embodiments of this disclosure can be implemented in software, for example, the processor 802 reads the corresponding computer program from the non-volatile memory 810 into memory 808 and then runs it. Of course, besides software implementation, one or more embodiments of this disclosure do not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution entity of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0077] Please see Figure 9 , Figure 9 This is a block diagram illustrating an exemplary embodiment of a reversing control device 900. This reversing control device 900 can be applied to, for example... Figure 8 The illustrated electronic device is used to implement the technical solution of this disclosure. The device includes: The first determining unit 902 is used to determine the lateral position deviation and heading angle deviation of the vehicle based on the relationship between the vehicle position, heading angle and the desired trajectory during the process of the vehicle reversing along the desired trajectory. The first adjustment unit 904 is used to adjust the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation; wherein the parameters of the state error weight matrix are used to adjust the tracking accuracy of the vehicle's lateral position and heading angle, and the parameters of the control input weight matrix are used to adjust the control smoothness of the vehicle. The first control unit 906 is used to control the vehicle to continue reversing along the desired trajectory based on the adjusted state error weight matrix and / or control input weight matrix.

[0078] In some embodiments, before adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control based at least on the values ​​of the lateral position deviation and the heading angle deviation, the apparatus further includes: The second determining unit 908 is used to determine the reference deviation value range of the lateral position deviation and the heading angle deviation, respectively. The calibration unit 910 is used to initially calibrate the parameters of the state error weight matrix and the control input weight matrix in the model predictive control algorithm used for reversing control, based on the reference deviation range of the lateral position deviation and the heading angle deviation.

[0079] In some embodiments, the apparatus further includes: The second adjustment unit 912 is used to not adjust the parameters of the state error weight matrix and the control input weight matrix if the lateral position deviation falls within the corresponding reference deviation value range or the heading angle deviation falls within the corresponding reference deviation value range. The second control unit 914 is used to control the vehicle to continue reversing along the desired trajectory according to the initially calibrated matrix.

[0080] In some embodiments, adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control, at least based on the values ​​of the lateral position deviation and the heading angle deviation, includes: If the lateral position deviation is less than the lower limit of the corresponding reference deviation range and the heading angle deviation is less than the lower limit of the corresponding reference deviation range, then the parameters of the control input weight matrix are increased. The parameters of the increased control input weight matrix are used to enhance the control smoothness of the vehicle.

[0081] In some embodiments, adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control, at least based on the values ​​of the lateral position deviation and the heading angle deviation, includes: If the lateral position deviation is greater than the upper limit of the corresponding reference deviation range and the heading angle deviation is less than the lower limit of the corresponding reference deviation range, then the parameter for adjusting the lateral position deviation in the state error weight matrix is ​​increased. Among them, the parameter for adjusting the lateral position deviation in the increased state error weight matrix is ​​used to enhance the tracking accuracy of adjusting the lateral position of the vehicle.

[0082] In some embodiments, adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control, at least based on the values ​​of the lateral position deviation and the heading angle deviation, includes: If the lateral position deviation is less than the lower limit of the corresponding reference deviation range and the heading angle deviation is greater than the upper limit of the corresponding reference deviation range, then the parameter for adjusting the heading angle deviation in the state error weight matrix is ​​increased. Among them, the parameter for adjusting the heading angle deviation in the increased state error weight matrix is ​​used to enhance the tracking accuracy of the vehicle's heading angle.

[0083] In some embodiments, adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm for reversing control, at least based on the values ​​of the lateral position deviation and the heading angle deviation, includes: If the lateral position deviation is greater than the upper limit of the corresponding reference deviation range, and the heading angle deviation is greater than the upper limit of the corresponding reference deviation range, then it is determined whether the lateral position deviation and the heading angle deviation are in the same direction. If the two directions are consistent, then the parameter for adjusting the lateral position deviation in the state error weight matrix is ​​decreased, and the parameter for adjusting the heading angle deviation in the state error weight matrix is ​​increased; wherein, the parameter for adjusting the lateral position deviation in the decreased state error weight matrix is ​​used to weaken the adjustment of the vehicle's lateral position tracking accuracy, and the parameter for adjusting the heading angle deviation in the increased state error weight matrix is ​​used to strengthen the adjustment of the vehicle's heading angle tracking accuracy. If the two directions are inconsistent, the parameters of the control input weight matrix are reduced; wherein, the reduced parameters of the control input weight matrix are used to weaken the control smoothness of the vehicle.

[0084] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0085] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0086] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which can take the form of a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email sending and receiving device, game console, tablet computer, wearable device, or any combination of these devices.

[0087] In a typical configuration, a computer includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0088] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0089] Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0090] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points shall be provided for users to choose to authorize or refuse.

[0091] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0092] The foregoing has described specific embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] The terminology used in one or more embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this disclosure. The singular forms “a,” “the,” and “the” as used in one or more embodiments of this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0094] It should be understood that although the terms first, second, third, etc., may be used to describe various information in one or more embodiments of this disclosure, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of one or more embodiments of this disclosure, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0095] The above description is merely a preferred embodiment of one or more embodiments of this disclosure and is not intended to limit the scope of one or more embodiments of this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this disclosure should be included within the scope of protection of one or more embodiments of this disclosure.

Claims

1. A reversing control method, characterized in that, The method includes: During the reverse driving of the vehicle along the desired trajectory, the lateral position deviation and heading angle deviation of the vehicle are determined based on the relationship between the vehicle position, heading angle and the desired trajectory. At least based on the values ​​of the lateral position deviation and the heading angle deviation, the parameters of the state error weight matrix and / or the control input weight matrix in the model predictive control algorithm used for reversing control are adjusted; wherein, the parameters of the state error weight matrix are used to adjust the tracking accuracy of the vehicle's lateral position and heading angle, and the parameters of the control input weight matrix are used to adjust the control smoothness of the vehicle. Based on the adjusted state error weight matrix and / or control input weight matrix, the vehicle is controlled to continue reversing along the desired trajectory.

2. The method according to claim 1, characterized in that, Before adjusting the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm used for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation, the method further includes: Determine the reference deviation ranges for the lateral position deviation and the heading angle deviation, respectively; Based on the reference deviation range of the lateral position deviation and the heading angle deviation, the parameters of the state error weight matrix and the control input weight matrix in the model predictive control algorithm used for reversing control are initially calibrated.

3. The method according to claim 2, characterized in that, The method further includes: If the lateral position deviation falls within the corresponding reference deviation range, or the heading angle deviation falls within the corresponding reference deviation range, then the parameters of the state error weight matrix and the control input weight matrix will not be adjusted. Based on the initially calibrated matrix, the vehicle is controlled to continue reversing along the desired trajectory.

4. The method according to claim 2, characterized in that, The adjustment of the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm used for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation, includes: If the lateral position deviation is less than the lower limit of the corresponding reference deviation range and the heading angle deviation is less than the lower limit of the corresponding reference deviation range, then the parameters of the control input weight matrix are increased. The parameters of the increased control input weight matrix are used to enhance the control smoothness of the vehicle.

5. The method according to claim 2, characterized in that, The adjustment of the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm used for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation, includes: If the lateral position deviation is greater than the upper limit of the corresponding reference deviation range and the heading angle deviation is less than the lower limit of the corresponding reference deviation range, then the parameter for adjusting the lateral position deviation in the state error weight matrix is ​​increased. Among them, the parameter for adjusting the lateral position deviation in the increased state error weight matrix is ​​used to enhance the tracking accuracy of adjusting the lateral position of the vehicle.

6. The method according to claim 2, characterized in that, The adjustment of the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm used for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation, includes: If the lateral position deviation is less than the lower limit of the corresponding reference deviation range and the heading angle deviation is greater than the upper limit of the corresponding reference deviation range, then the parameter for adjusting the heading angle deviation in the state error weight matrix is ​​increased. Among them, the parameter for adjusting the heading angle deviation in the increased state error weight matrix is ​​used to enhance the tracking accuracy of the vehicle's heading angle.

7. The method according to claim 2, characterized in that, The adjustment of the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm used for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation, includes: If the lateral position deviation is greater than the upper limit of the corresponding reference deviation range, and the heading angle deviation is greater than the upper limit of the corresponding reference deviation range, then it is determined whether the lateral position deviation and the heading angle deviation are in the same direction. If the two directions are consistent, then the parameter for adjusting the lateral position deviation in the state error weight matrix is ​​decreased, and the parameter for adjusting the heading angle deviation in the state error weight matrix is ​​increased; wherein, the parameter for adjusting the lateral position deviation in the decreased state error weight matrix is ​​used to weaken the adjustment of the vehicle's lateral position tracking accuracy, and the parameter for adjusting the heading angle deviation in the increased state error weight matrix is ​​used to strengthen the adjustment of the vehicle's heading angle tracking accuracy. If the two directions are inconsistent, the parameters of the control input weight matrix are reduced; wherein, the reduced parameters of the control input weight matrix are used to weaken the control smoothness of the vehicle.

8. A reversing control device, characterized in that, The device includes: The determining unit is used to determine the lateral position deviation and heading angle deviation of the vehicle based on the relationship between the vehicle position, heading angle and the desired trajectory during the process of the vehicle reversing along the desired trajectory. An adjustment unit is configured to adjust the parameters of the state error weight matrix and / or control input weight matrix in the model predictive control algorithm used for reversing control, based at least on the values ​​of the lateral position deviation and the heading angle deviation; wherein the parameters of the state error weight matrix are used to adjust the tracking accuracy of the vehicle's lateral position and heading angle, and the parameters of the control input weight matrix are used to adjust the control smoothness of the vehicle. The control unit is used to control the vehicle to continue reversing along the desired trajectory based on the adjusted state error weight matrix and / or control input weight matrix.

9. An electronic device, characterized in that, It includes a communication interface, a processor, a memory, and a bus, wherein the communication interface, the processor, and the memory are interconnected via the bus; The memory stores machine-readable instructions, and the processor executes the method according to any one of claims 1 to 7 by invoking the machine-readable instructions.

10. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores machine-readable instructions, which, when invoked and executed by a processor, implement the method described in any one of claims 1 to 7.

Images