Diagonal driving control method and device, electronic equipment and storage medium

Abstract

This invention discloses a diagonal driving control method, device, electronic device, and storage medium. The method includes: acquiring the current motion state information and target motion trajectory of the vehicle; selecting one of multiple preset motion modes, including a diagonal motion mode, as the target motion mode based on the motion state information; calling the vehicle motion model corresponding to the target motion mode based on the target motion mode, and calculating the target motion trajectory into a target vehicle body motion vector based on the vehicle motion model; constructing an optimization allocation problem with the target vehicle body motion vector as a constraint and the target turning angle of each of the four wheels of the vehicle as a variable, and generating a set of turning angle commands containing four independent turning angle values ​​by solving the optimization allocation problem; and outputting the turning angle commands to the independent steering actuators of the four wheels of the vehicle to control the vehicle to achieve diagonal driving.

Description

Technical Field

[0001] This application belongs to the field of four-wheel steering control technology, specifically relating to a diagonal driving control method, a diagonal driving control device and electronic equipment, as well as a computer-readable storage medium. Background Technology

[0002] Currently, the mainstream mass-produced technologies in the field of intelligent connected vehicles with four-wheel steering (4WS) are based on Ackermann steering geometry constraints to achieve four-wheel coordinated control. They rely on a hierarchical control architecture to achieve functions such as trajectory tracking, stability control, and low-speed crab-like translation, and have been widely used in high-end intelligent vehicles. However, this type of solution is constrained by the inherent geometric constraints of Ackerman steering and has significant technical defects: First, the steering motion mode is closed, which can only realize fixed-axis rotation and approximate translational motion with the instantaneous steering center of the four wheels at the same point. It cannot realize new maneuvering modes such as non-point-of-concurrency and diagonal coordination, resulting in insufficient maneuverability in complex scenarios. Second, there is a serious waste of hardware degrees of freedom. The vehicle angle module has four degrees of freedom hardware capability for independent steering of the four wheels, but the existing solution forcibly associates the four wheel angles through Ackerman allocation logic and reduces the dimension to two degrees of freedom control of the front and rear axles. It cannot realize independent and differentiated angle control of each wheel, and the hardware performance cannot be fully utilized. Third, the accuracy and stability of special maneuvering functions are poor. The existing crab-walking and slanting functions rely on indirect conversion of trajectory planning rather than direct and precise control of the wheel ends. In complex working conditions, trajectory deviation and vehicle attitude instability are prone to occur, and the control accuracy and scenario adaptability are limited.

[0003] In summary, current intelligent connected vehicle four-wheel steering control technology suffers from inherent technical defects and shortcomings in industrial application, such as closed and single motion modes, wasted hardware control degrees of freedom, insufficient control precision for special maneuvering functions, and weak adaptability to complex scenarios. Therefore, breaking through the geometric constraints of traditional Ackerman steering, releasing the full-domain degrees of freedom of four-wheel independent steering hardware, and achieving multi-modal precise maneuvering control and stable adaptation to all scenarios have become the core challenges that urgently need to be addressed in optimizing four-wheel steering intelligent control technology. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a diagonal driving control method, device, electronic device, and storage medium. It can overcome the rigid constraints of Ackerman geometry in existing four-wheel steering control schemes, fully utilize the control capabilities of the corner module hardware, and directly and with high precision achieve various diagonal driving modes such as low-speed oblique parking and high-speed cooperative lane changing through precise and controllable control logic. This effectively improves the vehicle's spatial maneuverability, steering control accuracy, and overall driving stability in complex driving scenarios.

[0005] In a first aspect, embodiments of this application provide a diagonal driving control method, including: Obtain the vehicle's current motion status information and the target's motion trajectory; Based on motion status information, select one of the multiple preset motion modes, including diagonal motion mode, as the target motion mode. Based on the target motion pattern, the vehicle motion model corresponding to the target motion pattern is invoked, and the target motion trajectory is calculated into the target vehicle body motion vector based on the vehicle motion model; Using the target vehicle body motion vector as a constraint, an optimization allocation problem is constructed with the target steering angles of the four wheels of the vehicle as variables. By solving the optimization allocation problem, a set of steering commands containing four independent steering angle values ​​is generated. The steering commands are output to the independent steering actuators of the four wheels of the vehicle to control the vehicle to drive diagonally.

[0006] In some embodiments, selecting one of a plurality of preset motion modes, including a diagonal motion mode, as the target motion mode based on motion state information includes: When the vehicle speed is lower than the first threshold in the motion status information and the driver actively triggers the first instruction to perform angled parking, the low-speed angled parking mode is selected as the target motion mode. When the vehicle speed exceeds the second threshold and a second instruction from the intelligent driving system is received instructing the vehicle to perform an automatic lane change, the high-speed cooperative lane change mode is selected as the target motion mode.

[0007] In some embodiments, the target motion trajectory is solved into a target vehicle body motion vector based on the vehicle motion model, including: When the target motion mode is low-speed inclined parking mode, the low-speed inclined kinematic model is adopted to solve the target motion trajectory into the desired longitudinal velocity component and the desired lateral velocity component, and the desired yaw rate is zero. When the target motion mode is a high-speed cooperative lane change mode, a high-speed cooperative dynamics model is adopted to solve the target motion trajectory into the desired lateral velocity correction and / or the desired yaw rate correction.

[0008] In some embodiments, the optimization objective of the optimization allocation problem includes minimizing the change in the steering angle of each wheel of the vehicle relative to the previous moment and / or the estimated slip energy of each wheel of the vehicle.

[0009] In some embodiments, the diagonal driving control method further includes: Based on the safety boundary corresponding to the target motion mode, monitor the vehicle's safety status parameters in real time. In response to safety status parameters exceeding safety boundaries, intervention control is executed or the target motion mode is forcibly exited.

[0010] In some embodiments, implementing intervention control or forcibly exiting the target motion mode includes: In low-speed angled parking mode, the tire slip ratio and the safe distance between the vehicle and obstacles are monitored. When the tire slip ratio exceeds the preset slip threshold or the safe distance is less than the preset safe threshold, the wheel angle is reduced. In high-speed cooperative lane change mode, the system monitors the status of lateral acceleration and vehicle stability control system. In response to lateral acceleration exceeding the comfort threshold or the stability control system intervening, the system sets the desired lateral speed correction and / or the desired yaw rate correction to zero to exit high-speed cooperative lane change mode.

[0011] In some embodiments, the diagonal driving control method further includes: For each wheel of the vehicle, obtain the actual steering angle of the wheel; The deviation between the actual turning angle and the corresponding turning angle value in the turning command is used as feedback, and the optimization allocation problem at the next moment is corrected based on the feedback.

[0012] Secondly, embodiments of this application provide a diagonal driving control device, including: The acquisition module is configured to acquire the vehicle's current motion state information and the target's motion trajectory; The selected module is configured to select one of multiple preset motion modes, including diagonal motion mode, as the target motion mode based on motion state information. The solution module is configured to call the vehicle motion model corresponding to the target motion mode based on the target motion mode, and solve the target motion trajectory into the target vehicle motion vector based on the vehicle motion model; The construction module is configured to construct an optimization allocation problem with the target vehicle body motion vector as a constraint, and the target steering angle of each of the four wheels of the vehicle as a variable. By solving the optimization allocation problem, a set of steering commands containing four independent steering angle values ​​is generated. The execution module is configured to output steering commands to the independent steering actuators of the four wheels of the vehicle to control the vehicle to drive diagonally.

[0013] Thirdly, embodiments of this application provide an electronic device, including: a processor and a memory, wherein the memory stores a program or instructions that can run on the processor, and when the program or instructions are executed by the processor, they implement the steps of the diagonal driving control method, apparatus, electronic device and storage medium as described in the first aspect.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing a program or instructions, which, when executed by a processor, implement the steps of the diagonal driving control method, apparatus, electronic device, and storage medium as described in the first aspect.

[0015] The diagonal driving control method, device, electronic equipment, and storage medium provided in this application first acquire the vehicle's current motion state information and target trajectory. Then, based on the motion state information, one of several preset motion modes, including a diagonal motion mode, is selected as the target motion mode. Further, based on the target motion mode, the corresponding vehicle motion model is invoked, and the target trajectory is calculated into a target vehicle body motion vector. Using the target vehicle body motion vector as a constraint, an optimization allocation problem is constructed with the target turning angles of each of the vehicle's four wheels as variables. Solving the optimization allocation problem generates a set of turning angle commands containing four independent turning angle values. Finally, the turning angle commands are output to the independent steering actuators of the vehicle's four wheels to control the vehicle to achieve diagonal driving. This application selects the corresponding solution model according to different operating modes, which not only achieves a one-click diagonal entry operation effect, providing users with a stable and smooth user experience, but also makes the vehicle lane-changing process smoother, effectively optimizing the driving experience. It fully unleashes the performance potential of the corner module hardware, improves the control accuracy and response speed of the system, further expands the boundaries of the vehicle's motion function, and significantly improves the vehicle's maneuverability in various specific scenarios. The targeted safety strategy can also ensure that the corresponding functions have good reliability and availability, fully meet the requirements of actual engineering implementation, and continuously provide users with a stable and reliable user experience.

[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a schematic diagram of the overall principle of a diagonal driving control method according to an embodiment of this application; Figure 2 This is a flowchart illustrating a diagonal driving control method according to an embodiment of this application; Figure 3 This is a schematic diagram of a low-speed inclined parking mode in an embodiment of this application; Figure 4 This is a schematic diagram of a high-speed cooperative lane-changing mode in an embodiment of this application; Figure 5This is a schematic diagram of a process for optimizing wheel end angle allocation in an embodiment of this application; Figure 6 This is a control flowchart of a diagonal driving control method according to an embodiment of this application; Figure 7 This is a schematic diagram of the structure of a diagonal driving control device according to an embodiment of this application; Figure 8 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application.

[0018] Explanation of reference numerals in the attached drawings: diagonal driving control device 700, acquisition module 701, selection module 702, calculation module 703, construction module 704, execution module 705, processor 810, memory 820, input / output interface 830, communication interface 840, and bus 850. Detailed Implementation

[0019] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While some embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this application. It should be understood that the drawings and embodiments of this application are for illustrative purposes only and are not intended to limit the scope of protection of this application.

[0020] It should be understood that the steps described in the method embodiments of this application may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this application is not limited in this respect.

[0021] As described in the background section, current intelligent connected vehicle four-wheel steering control technology suffers from problems such as closed and single motion modes, wasted hardware control degrees of freedom, insufficient control precision for special maneuvering functions, and weak adaptability to complex scenarios. This application provides a diagonal driving control method, device, electronic device, and storage medium that can break through the rigid constraints of Ackerman geometry in existing four-wheel steering control schemes, fully utilize the control capabilities of the corner module hardware, and directly and with high precision realize various diagonal driving modes such as low-speed oblique parking and high-speed cooperative lane changing through precise and controllable control logic. This effectively improves the vehicle's spatial maneuverability, steering control precision, and overall driving stability in complex driving scenarios.

[0022] Currently, mainstream four-wheel steering control schemes adhere to Ackermann steering geometry rigidity constraints throughout the entire process. The system simultaneously receives a reference path generated by the path planning module, which includes the target position and heading angle, as well as real-time vehicle status data collected by the perception and positioning module, such as vehicle speed, yaw rate, center of gravity sideslip angle, and actual steering angles of the four wheels. Both types of information are simultaneously input into the path tracking controller, which employs a model predictive control (MPC) algorithm and integrates a simplified vehicle dynamics model to simulate the vehicle's basic driving dynamics. The path tracking controller aims to minimize the position and heading errors between the vehicle and the reference path. Combining actuator rate limits and driving stability constraints, it performs rolling optimization within a fixed prediction time domain, outputting the optimal front axle equivalent steering angle. This front axle steering angle, along with the vehicle's real-time yaw rate, is input into the wheel angle distribution controller. Through a fixed algorithm, the target steering angle of the rear axle is calculated, extending the single front axle control degree of freedom to a dual-degree-of-freedom system for both the front and rear axles. Subsequently, the target steering angles of the front and rear axles are fed into the Ackerman four-wheel steering angle distribution controller. This module rigidly constrains the steering vertical lines of the four wheels to intersect at the same instantaneous steering center. Through preset geometric trigonometric function formulas, the dual-axle steering angle is solved into independent target steering angles for the four wheels: left front wheel, right front wheel, left rear wheel, and right rear wheel. This forcibly eliminates the independent control degrees of freedom of the four wheels, completely binding the four steering angle variables as subordinate functions of the front and rear axle steering angles. Finally, the steering angle commands of the four wheels are sent to the four-wheel independent steering mechanism of the vehicle's corner module, driving the wheels to deflect. At the same time, the new state parameters generated by the vehicle's motion and the actual steering angle information are fed back to the upstream controller in real time, forming a complete closed-loop control loop. This loop is continuously iterated and optimized until the vehicle accurately tracks the reference path.

[0023] However, this solution can only achieve fixed-axis rotation and approximate translational motion with all four wheels pointing perpendicularly to the same point. It cannot support non-point-of-concurrency steering modes such as diagonal coordination, which significantly limits the vehicle's maneuverability in complex scenarios. Furthermore, this solution compresses the four-degree-of-freedom hardware capability of the corner module's independent four-wheel steering into a two-degree-of-freedom control of the front and rear axles, resulting in a serious waste of hardware potential. In addition, its mass-production application's crab-like mode relies on indirect calculations based on trajectory planning, resulting in a long control link, insufficient precision, and a tendency to malfunction under complex boundary conditions.

[0024] refer to Figure 1 This is a schematic diagram of the overall principle of a diagonal driving control method in an embodiment of this application.

[0025] like Figure 1As shown, the diagonal driving control method proposed in this application is deployed in the vehicle domain controller. Its core lies in constructing an intelligent control method capable of recognizing, planning, and precisely executing the specific motion mode of diagonal driving. This method is deeply integrated with the vehicle's intelligent driving system and chassis drive-by-wire system. Through an algorithmic process, it transforms high-level motion intentions into independent steering angle commands for the four corner modules at the lower level. The diagonal driving control method mainly consists of five functional modules: a multimodal decision-making and trajectory generation module, a kinematics / dynamics solver, a wheel-end angle optimization allocator, a command execution and closed-loop control module, and a safety monitoring and arbitration module. The multimodal decision-making and trajectory generation module receives raw commands from the driver or intelligent driving system and transmits the output mode signal and target trajectory to the kinematics / dynamics solver. The kinematics / dynamics solver calculates the target vehicle motion vector based on the mode signal and target trajectory and transmits the calculated motion vector to the wheel-end angle optimization allocator. The wheel-end angle optimization allocator generates steering angle commands by solving an optimization allocation problem and then transmits them to the command execution and closed-loop control module for execution. The instruction execution and closed-loop control module collects the actual vehicle status and the safety status monitored by the safety monitoring and arbitration module, and feeds them back to the multimodal decision and trajectory generation module, the kinematics / dynamics solver, and the wheel-end angle optimization distributor. These are used to adjust the decision-making, calculation, and optimization work in real time, and ultimately achieve a complete loop control in which perception, decision-making, execution, and feedback are coordinated.

[0026] refer to Figure 2 This is a flowchart illustrating a diagonal driving control method in an embodiment of this application.

[0027] like Figure 2 As shown, this application provides a diagonal driving control method, including: Step S201: Obtain the vehicle's current motion status information and the target motion trajectory.

[0028] In practical implementation, after the vehicle is powered on, the control system starts running, completes initialization, and performs status monitoring, continuously monitoring the vehicle's current speed, yaw rate, sideslip angle, actual wheel rotation angle, lateral acceleration, tire slip ratio, steering wheel angle, and gear position, among other motion status information. Once vehicle positioning is complete, path planning is initiated, or a driving command is received, the system acquires the position information, heading angle information, trajectory curvature, lateral displacement deviation, desired longitudinal velocity component, desired lateral velocity component, and desired yaw rate output by the intelligent driving system. The multimodal decision-making and trajectory generation module receives raw commands from the driver or intelligent driving system and transmits the motion status information and target trajectory to the kinematics / dynamics solver.

[0029] Step S202: Based on the motion state information, select one of the multiple preset motion modes, including the diagonal motion mode, as the target motion mode.

[0030] In practical implementation, the multimodal decision-making and trajectory generation module performs multimodal decision-making, receives internal and external instructions, determines the target motion mode to be entered, and generates the corresponding target motion trajectory. Based on vehicle speed and driver intent, this module constructs a unified diagonal kinematic model of the submodal motions, forming a dynamically switchable bimodal kinematic framework. The target motion mode includes at least a low-speed diagonal parking mode and a high-speed cooperative lane-changing mode.

[0031] Step S203: Based on the target motion mode, call the vehicle motion model corresponding to the target motion mode, and solve the target motion trajectory into the target vehicle body motion vector based on the vehicle motion model.

[0032] In practical implementation, the kinematics / dynamics solver selects the corresponding kinematic model based on the target motion mode, and calculates the target trajectory into the desired target vehicle body motion vector. The low-speed oblique parking mode uses the low-speed oblique kinematic model, suitable for manually triggered scenarios where vehicle speed is below the first threshold (e.g., 5 km / h). This model aims to minimize lateral displacement error, allowing for larger and independently optimizable unidirectional yaw angles (e.g., ±45°) for all four wheels, directly calculating wheel-end commands for precise oblique translation. The high-speed cooperative lane change mode uses the high-speed cooperative dynamics model, suitable for automatically triggered scenarios by intelligent driving systems where vehicle speed is above the second threshold (e.g., 60 km / h). This model aims to minimize the center of gravity sideslip angle and yaw overshoot, calculating a set of small four-wheel unidirectional compensation yaw angles (e.g., ±2°) to enhance the stability of traditional steering. Both models use independent four-wheel yaw angles as direct input and the synthesized vehicle body motion vector as output, supporting non-Ackermann steering geometry and fundamentally unlocking the vehicle's diagonal driving capability.

[0033] Step S204: Using the target vehicle body motion vector as a constraint, construct an optimization allocation problem with the target steering angles of the four wheels of the vehicle as variables, and generate a set of steering commands containing four independent steering angle values ​​by solving the optimization allocation problem.

[0034] In practical implementation, the wheel-end angle optimization allocator receives the target vehicle motion vector and calculates the optimal target steering angles for each of the four wheels by solving the optimization allocation problem, completely bypassing the traditional Ackermann allocation method. The core of the wheel-end angle optimization allocator is the real-time solution of the optimization allocation problem. The optimization objective is to minimize the deviation of the four wheel angles from the previous moment to ensure smooth driving, or to minimize tire slip ratio to ensure driving efficiency. Constraints must satisfy the desired target vehicle motion vector defined by the kinematic model in the current activation mode, with each wheel angle not exceeding physical limits and each wheel slip ratio below a safety threshold. By solving this optimization problem online, the system can dynamically allocate a set of independent and optimal target steering angles to the four wheels, fully utilizing the control degrees of freedom of the four steering actuators and achieving a paradigm shift from axle control to wheel control. This allocation method can fully unleash the performance potential of the angle module hardware, improving the system's control accuracy and response speed.

[0035] In step S205, the turning command is output to the independent steering actuators of the four wheels of the vehicle to control the vehicle to drive diagonally.

[0036] In practice, the command execution and closed-loop control module sends the optimized wheel-end steering angle command to the independent steering actuators of the four wheels via the vehicle network. Simultaneously, it collects actual steering angle and vehicle status information and provides feedback correction. Furthermore, this module continuously provides feedback on the actual motion state, fine-tuning the wheel-end steering angle to ensure the vehicle accurately follows the trajectory provided by the multimodal decision-making and trajectory generation module. The independent steering capability of the angle module reduces the deviation between the vehicle's actual driving path and the preset trajectory, resulting in a more stable vehicle driving state.

[0037] In some embodiments, selecting one of a plurality of preset motion modes, including a diagonal motion mode, as a target motion mode based on motion state information includes: selecting a low-speed diagonal parking mode as the target motion mode when the vehicle speed is lower than a first threshold in the motion state information and a first instruction initiated by the driver to instruct the vehicle to perform diagonal parking is received; and selecting a high-speed cooperative lane changing mode as the target motion mode when the vehicle speed is higher than a second threshold and a second instruction initiated by the intelligent driving system to instruct the vehicle to perform automatic lane changing is received.

[0038] In practical implementation, the multimodal decision-making and trajectory generation module combines information such as vehicle speed, yaw rate, center of gravity sideslip angle, actual wheel turning angle, lateral acceleration, tire slip ratio, steering wheel angle, and gear position to select a suitable target motion mode. The diagonal motion modes include at least a low-speed diagonal parking mode and a high-speed cooperative lane-changing mode. In low-speed scenarios, the multimodal decision-making and trajectory generation module obtains the diagonal parking lines and trajectory through the automatic parking system. In high-speed scenarios, it directly reuses the standard lane-changing trajectory generated by the intelligent driving system. Furthermore, it can achieve adaptive switching between the two modes based on multimodal decision-making logic. This logic references predetermined vehicle speed thresholds and various trigger signals, enabling the multimodal decision-making and trajectory generation module to call the corresponding underlying models and control parameters in different scenarios, achieving one-click diagonal entry and a seamless and stable user experience.

[0039] refer to Figure 3 This is a schematic diagram of a low-speed inclined parking mode in an embodiment of this application.

[0040] like Figure 3 As shown, the multimodal decision-making and trajectory generation module detects that the vehicle speed is below a first threshold (e.g., below 5 km / h) and receives a driver-initiated diagonal parking command via the in-vehicle screen or button. The system then enters a low-speed diagonal parking mode. This module calls a parking planning algorithm, combines it with parking space information perceived by the surround-view camera, and generates an ideal diagonal straight trajectory (e.g., a straight line that forms a 30° angle with the roadway) from the vehicle's current position to the target parking space, while simultaneously setting the corresponding mode identifier.

[0041] refer to Figure 4 This is a schematic diagram of a high-speed cooperative lane-changing mode in an embodiment of this application.

[0042] like Figure 4 As shown, the multimodal decision-making and trajectory generation module detects that the vehicle speed is higher than the second threshold (e.g., the vehicle speed exceeds 60km / h). At the same time, it receives the automatic lane change or lever lane change confirmation request issued by the intelligent driving system via the CAN bus. The system then enters the high-speed cooperative lane change mode. This module takes the standard lane change trajectory planned by the intelligent driving system as input and sets the corresponding mode identifier.

[0043] If any of the aforementioned operating conditions are not met, the system maintains the conventional steering mode and executes the traditional steering control strategy. The multi-modal diagonal kinematic model equipped in this method allows the control system to recognize and generate novel motion commands such as diagonal driving. This enables the vehicle to smoothly complete a single, one-time maneuver along the optimal diagonal line when parallel parking, reducing the number of adjustments needed. When driving on highways, lane changes and overtaking are also smoother and faster. This solves the problem of existing technologies being limited to a single motion mode, resulting in insufficient scenario adaptability and low operational efficiency. The vehicle's motion function boundaries are further expanded, and the vehicle's maneuverability in various specific scenarios is significantly improved.

[0044] In some embodiments, the target motion trajectory is calculated into a target vehicle motion vector based on a vehicle motion model, including: when the target motion mode is a low-speed oblique parking mode, a low-speed oblique kinematic model is used to calculate the target motion trajectory into a desired longitudinal velocity component and a desired lateral velocity component, and the desired yaw rate is zero; when the target motion mode is a high-speed cooperative lane change mode, a high-speed cooperative dynamics model is used to calculate the target motion trajectory into a desired lateral velocity correction and / or a desired yaw rate correction.

[0045] In practice, the kinematics / dynamics solver selects the appropriate solution model based on different operating modes. When the target motion mode is low-speed oblique parking, the solver calls the low-speed oblique kinematics model. This model treats the vehicle as a rigid body for calculation, taking into account the curvature and lateral displacement deviation of the target oblique trajectory. Under low-speed conditions, the trajectory curvature is approximately zero. Finally, it outputs the longitudinal and lateral velocity components corresponding to the vehicle's composite velocity vector in the vehicle coordinate system, and sets the desired yaw rate to zero, thereby achieving pure oblique movement of the vehicle and ensuring that the vehicle's center of gravity travels along the oblique line. When the target motion mode is high-speed cooperative lane change, the solver calls the high-speed cooperative dynamics model. This model is extended from the vehicle's two-degree-of-freedom dynamics model, using the standard front wheel steering angle and the vehicle's real-time state as the calculation basis, and outputs the desired motion vector correction. This correction can suppress vehicle yaw and lateral acceleration overshoot, specifically including the desired lateral velocity correction and the desired yaw rate correction. Selecting the appropriate solution model based on different operating modes can make the lane-changing process smoother and effectively improve the driving experience.

[0046] In some embodiments, the optimization objective of the optimization allocation problem includes minimizing the change in the steering angle of each wheel of the vehicle relative to the previous moment and / or the estimated slip energy of each wheel of the vehicle.

[0047] refer to Figure 5 This is a schematic diagram of a process for optimizing wheel end angle allocation in an embodiment of this application.

[0048] like Figure 5 As shown, the wheel-end steering angle optimization allocator is based on the target vehicle motion vector and assigns independent optimal target steering angles to the four wheels through constrained optimization solutions, supporting the wheel control paradigm of non-Ackerman steering.

[0049] First, the optimization problem is constructed to establish a mathematical framework for the optimization solution. The input information of the wheel-end angle optimization allocator consists of two parts: one is the target vehicle motion vector output by the kinematics / dynamics solver, namely the desired longitudinal velocity component, the desired lateral velocity component, and the desired yaw rate, representing the composite motion state that the vehicle needs to achieve; the other is the wheel-end angle at the previous moment, used to constrain the smoothness of the angle change in the target, avoiding abrupt changes in control commands. The optimization variables are the steering angles of the four wheels, and the formula is:

[0050] in, For the corner of the front left wheel, The turning angle of the front right wheel. For the left rear wheel's turning angle, The turning angle for the rear right wheel.

[0051] Then, based on the input, the objective function of the optimization problem needs to be defined, which needs to take into account two core requirements: first, to minimize the deviation between the current turning angle and the previous turning angle to ensure smooth driving; and second, to minimize the estimated tire slip energy to improve driving efficiency. These two requirements are balanced by weighting coefficients to clarify the optimization direction for subsequent solutions. The formula is as follows:

[0052] in, The target steering angle vectors for the four wheels. This is the wheel end angle vector output from the previous control cycle. These are the weighting coefficients. To predict tire slip ratio, It can be any integer between 1 and 4, corresponding to the four wheels respectively.

[0053] To ensure the feasibility, safety, and effectiveness of the optimization solution, three types of constraints are set to limit the range of variable values ​​and mapping relationships. The first type is kinematic constraints, which stipulate that the steering angles of the four wheels, after being mapped through the vehicle kinematic model, must be equal to the input target vehicle body motion vector, ensuring that the calculated steering angles enable the vehicle to achieve the desired motion state. The formula is as follows:

[0054] in, This refers to the actual longitudinal speed of the vehicle body; This refers to the actual lateral speed of the vehicle body; This refers to the vehicle's actual yaw rate. The desired longitudinal velocity component; The desired lateral velocity component; The desired yaw rate; To create an accurate vehicle motion model selected based on the current mode, it establishes a mathematical relationship between the four-wheel steering angle and the vehicle motion vector. This model does not presuppose Ackerman geometry.

[0055] The second type is actuator constraints, which limit the steering angle of each wheel to the physical limits of the steering actuator, such as the upper and lower limits of the maximum achievable steering angle. This avoids generating invalid instructions that exceed the hardware's capabilities. For example, in low-speed angled parking mode, the turning angle of each wheel cannot exceed 45°; in high-speed cooperative lane change mode, the turning angle of each wheel is usually limited to ±2° to ±5°.

[0056] The third type is safety constraints, which estimate the slip ratio of each tire at the current steering angle and limit the estimated slip ratio of each wheel to not exceed a preset safety threshold. This prevents excessive tire slippage and ensures stability and safety during driving.

[0057] The wheel-end angle optimization allocator employs algorithms suitable for constrained nonlinear optimization (such as real-time optimization algorithms like sequential quadratic programming) to solve the aforementioned constrained nonlinear optimization problem online, obtaining the optimal four-wheel steering angle solution at the current moment. This solution allows the four wheel angles to be in the same direction but with different values. For example, when driving at low speeds at an angle, the four wheel angles can exhibit a state of slight differences in the same direction (the steering angles of the four wheel ends are 30°, 28°, 32°, and 30°, respectively), thereby actively optimizing tire forces and reducing slippage while achieving oblique movement. After the solution is completed, the wheel-end angle optimization allocator outputs the optimal target steering angles for the four wheels, corresponding to the steering angle commands for the four wheel ends. Simultaneously, the optimal steering angle obtained at the current solution is fed back to the step of constructing the optimization problem, serving as the steering angle input for the previous moment in the next optimization problem, forming a closed-loop iteration to ensure the continuity and smoothness of the wheel-end angle commands and avoid sudden shocks during the control process.

[0058] The wheel-end angle optimization allocator transforms angle calculation into an optimal solution problem. While satisfying the desired vehicle movement, it can freely adjust the four angles to optimize smoothness or efficiency, naturally generating optimal solutions in the same direction but with different angles. This achieves direct, independent, and optimal control of the four steering actuators, completely bypassing the fixed proportional allocation of traditional Ackerman steering and maximizing the control freedom of the angle module. It can more accurately track complex desired motion trajectories and reduce trajectory errors caused by the Ackerman approximation. Simultaneously, the optimization algorithm can dynamically adjust the four angles according to the vehicle's real-time status, achieving better adaptability and robustness. This translates the hardware advantages of the angle module system into tangible improvements in handling performance, providing precise wheel-end angle commands for subsequent instruction execution and closed-loop control.

[0059] In some embodiments, the diagonal driving control method further includes: monitoring the vehicle's safety status parameters in real time according to the safety boundary corresponding to the target motion mode; and performing intervention control or forcibly exiting the target motion mode in response to the safety status parameters exceeding the safety boundary.

[0060] During implementation, the safety monitoring and arbitration module continuously acquires and monitors real-time data from the command execution and closed-loop control modules, including the estimated slip ratio at all four wheels, vehicle yaw rate, actual steering angle from the steering actuators at all four wheels, and lateral acceleration. This ensures the vehicle remains within safe boundaries during the target motion mode and triggers intervention control or forces exit the target motion mode when necessary. Safety arbitration has a higher priority than all motion control commands to ensure the vehicle does not become unstable under any circumstances. When low-speed parking is completed, high-speed lane changing ends, the driver manually cancels the function, or the safety system forcibly takes over control, the multimodal decision-making and trajectory generation module clears the current mode identifier, smoothly transferring control to the conventional steering strategy, and the entire control process ends.

[0061] This modal adaptive safety arbitration mechanism sets targeted safety constraints for different diagonal driving scenarios. The angled parking function can operate stably while ensuring that the tires do not experience abnormal wear, and the high-speed lane change assist function can improve driving stability without causing vehicle body sway. This mechanism achieves a precise match between function and safety, effectively ensuring the reliability and usability of new functions. This targeted safety strategy can meet the requirements of actual engineering implementation and provide users with a reliable user experience.

[0062] In some embodiments, performing intervention control or forcibly exiting the target motion mode includes: in a low-speed inclined parking mode, monitoring the tire slip ratio and the safe distance between the vehicle and an obstacle, and reducing the wheel angle in response to the tire slip ratio exceeding a preset slip threshold or the safe distance being less than a preset safe threshold; in a high-speed cooperative lane change mode, monitoring the state of the lateral acceleration and the vehicle stability control system, and setting the desired lateral speed correction and / or the desired yaw rate correction to zero in response to the lateral acceleration exceeding a comfort threshold or the stability control system intervening, so as to exit the high-speed cooperative lane change mode.

[0063] In practice, the adaptive safety arbitration mechanism is linked to the vehicle's target motion mode. When the vehicle switches to low-speed angled parking mode, if any wheel experiences continuous ground slippage and fails to roll normally, causing the wheel slip rate to exceed a preset slip threshold, or if the distance between the vehicle and surrounding obstacles is less than a preset safety threshold, the safety monitoring and arbitration module will output an alarm signal to the multimodal decision-making and trajectory generation module. This module simultaneously coordinates with the command execution and closed-loop control module to narrow the turning angle of the corresponding wheel and continuously adjusts it until the system exits the low-speed angled parking mode or the driver takes over the vehicle. When the vehicle is operating in high-speed cooperative lane-changing mode, if the safety monitoring and arbitration module detects that the lateral acceleration exceeds the comfort threshold, or if the vehicle stability control system intervenes, it will immediately clear the expected lateral speed correction and yaw rate correction, and then seamlessly switch back to the normal steering mode to ensure the safety of the vehicle during high-speed driving.

[0064] The adaptive safety arbitration mechanism achieves deep coupling between safety logic and motion modes, and matches exclusive safety boundaries for various target motion modes. The vehicle's diagonal driving function can operate stably and reliably within the clearly defined safety boundaries, and can also avoid conflicts between the function and traditional safety objectives, preventing related functions from being restricted or experiencing operational interruptions.

[0065] In some embodiments, the diagonal driving control method further includes: for each wheel of the vehicle, obtaining the actual turning angle of the wheel; using the deviation between the actual turning angle and the corresponding turning angle value in the turning angle command as a feedback quantity, and correcting the optimization allocation problem at the next moment based on the feedback quantity.

[0066] In the specific implementation process, the command execution and closed-loop control module sends the optimized target steering angle of each wheel to the steer-by-wire actuators of the four wheel-end angle modules via the CAN bus. The actuators drive the wheels to rotate to the target angle. This module simultaneously uses sensors to obtain the actual wheel angle and the actual vehicle motion state, compares the actual angle with the corresponding value in the angle command to form a closed-loop feedback, and then corrects the optimization allocation for the next moment.

[0067] refer to Figure 6 This is a control flowchart of a diagonal driving control method in an embodiment of this application.

[0068] like Figure 6 The diagram illustrates the fully closed-loop control logic of the vehicle from system power-on to the completion and exit of the diagonal driving function. This logic combines vehicle speed and trigger signals to automatically switch between low-speed diagonal parking mode and high-speed cooperative lane-changing mode, and achieves precise diagonal motion control through optimized distribution of independent wheel-end steering angles. Simultaneously, it performs safety monitoring throughout the entire process to ensure vehicle driving safety.

[0069] After the vehicle is powered on, the system first completes hardware self-checks and parameter initialization of the control system. It then collects real-time data on vehicle speed, steering wheel angle, gear position, braking status, driver control commands, intelligent driving requests, as well as parking space information from surround-view cameras and distances to surrounding obstacles detected by millimeter-wave radar and lidar. Multimodal decision-making is then performed on all collected status and command signals. When the vehicle's real-time speed is below a first threshold and the system receives the first command from the driver, the vehicle enters a low-speed angled parking mode. This activates the multimodal decision-making and trajectory generation module, generating an ideal angled straight trajectory from the vehicle's current position to the target parking space based on parking space information perceived by the surround-view cameras. This mode is suitable for low-speed scenarios such as parallel parking in tight spaces, parking in non-standard spaces, and angled entry into narrow passages. When the vehicle's real-time speed exceeds a second threshold and the system receives a second command from the intelligent driving system, the vehicle enters a high-speed cooperative lane-change mode. This mode directly uses the standard lane-change trajectory planned by the intelligent driving system as the input. This mode is suitable for smooth lane-change and overtaking scenarios on highways and urban expressways. When the operating conditions cannot match the above two modes, the vehicle maintains its normal operating state or exits the target mode, no longer intervenes in steering control, and completely transfers vehicle control to the traditional Ackermann steering control system, which then executes the normal front and rear axle steering logic.

[0070] After the vehicle enters the target motion mode, the kinematics / dynamics solver calls the corresponding kinematic model according to the currently active motion mode, transforming the target trajectory generated by the upper layer into the desired motion vector of the vehicle's center of mass, providing core constraints for subsequent wheel-end angle optimization. After the vehicle enters the low-speed oblique parking mode, the kinematics / dynamics solver treats the vehicle as a rigid body, inputting the curvature of the target oblique trajectory and the vehicle's current lateral displacement deviation into the low-speed oblique kinematic model, outputting the vehicle's desired longitudinal velocity component, lateral velocity component, and yaw rate. After the vehicle enters the high-speed cooperative lane change mode, the kinematics / dynamics solver inputs the standard front wheel steering angle and the vehicle's real-time operating state into the high-speed cooperative dynamics model, outputting the vehicle's desired lateral velocity correction and yaw rate correction.

[0071] After completing the motion vector calculation, the wheel-end angle optimization allocator transforms the angle calculation into a constrained optimization problem. Using real-time optimization algorithms such as sequential quadratic programming, it completes the online solution within a single control cycle, obtaining the optimal four-wheel angle combination for the current moment. It outputs independent target steering angle commands for each of the four wheels. The command execution and closed-loop control module sends the optimized steering commands to the steer-by-wire actuators of each wheel-end angle module. Each actuator drives the wheel to deflect to the target angle via a drive motor. The system collects actual wheel angle data through wheel-end angle sensors and actual vehicle motion state data through an inertial measurement unit. It compares various actual operating parameters with preset expected values, feeding back any errors to the expected motion vector calculation and wheel-end angle optimization allocator stages. This real-time correction of the calculation results and initial optimization values ​​constructs a complete vehicle steering control closed loop, achieving closed-loop tracking.

[0072] At the same time, the safety monitoring and arbitration module will collect all key safety parameters in real time, such as wheel slip rate, vehicle yaw rate, lateral acceleration, and distance to surrounding obstacles. Combined with the safety boundary corresponding to the target motion mode, it will generate a safety status signal and feed the safety status signal back to each link of the desired motion vector calculation, wheel end angle optimization allocation, command issuance and closed-loop tracking in real time, so as to realize dynamic safety intervention in the entire process of vehicle control.

[0073] The system will exit the current diagonal driving control process when any of the following conditions are met: completion of a low-speed diagonal parking task or the vehicle reaches the target parking space; completion of a high-speed coordinated lane change task or the vehicle enters the target lane; the driver manually cancels the diagonal driving function; or the safety monitoring and arbitration module detects a serious safety risk and executes a forced takeover operation. The system will then clear the current mode identifier, smoothly transfer vehicle steering control to the regular steering control system, and the process will end.

[0074] In summary, the diagonal driving control method proposed in this application first acquires the vehicle's current motion state information and target trajectory. Then, based on the motion state information, it selects one of several preset motion modes, including a diagonal motion mode, as the target motion mode. Further, based on the target motion mode, it calls the vehicle motion model corresponding to that mode and calculates the target trajectory as a target vehicle body motion vector. Using the target vehicle body motion vector as a constraint, it constructs an optimization allocation problem with the target turning angles of each of the vehicle's four wheels as variables. Solving this optimization allocation problem generates a set of turning angle commands containing four independent turning angle values. Finally, the turning angle commands are output to the independent steering actuators of the vehicle's four wheels to control the vehicle to achieve diagonal driving. This application selects the corresponding solution model according to different operating modes, which not only achieves a one-click diagonal entry operation effect, providing users with a stable and smooth user experience, but also makes the vehicle lane-changing process smoother, effectively optimizing the driving experience. It fully unleashes the performance potential of the corner module hardware, improves the control accuracy and response speed of the system, further expands the boundaries of the vehicle's motion function, and significantly improves the vehicle's maneuverability in various specific scenarios. The targeted safety strategy can also ensure that the corresponding functions have good reliability and availability, fully meet the requirements of actual engineering implementation, and continuously provide users with a stable and reliable user experience.

[0075] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described above.

[0076] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0077] Corresponding to the above embodiments, this application also proposes a diagonal driving control device.

[0078] refer to Figure 7 This is a schematic diagram of the structure of a diagonal driving control device in an embodiment of this application.

[0079] like Figure 7As shown, this application embodiment provides a diagonal driving control device 700, including: The acquisition module 701 is configured to acquire the vehicle's current motion state information and the target motion trajectory; The selected module 702 is configured to select one of multiple preset motion modes, including diagonal motion mode, as the target motion mode based on motion state information. The calculation module 703 is configured to call the vehicle motion model corresponding to the target motion mode based on the target motion mode, and calculate the target motion trajectory into the target vehicle motion vector based on the vehicle motion model. Module 704 is configured to construct an optimization allocation problem with the target vehicle body motion vector as a constraint and the target steering angle of each of the four wheels of the vehicle as a variable, and generate a set of steering commands containing four independent steering angle values ​​by solving the optimization allocation problem. The execution module 705 is configured to output steering commands to the independent steering actuators of the four wheels of the vehicle to control the vehicle to drive diagonally.

[0080] Optionally, the selected module 702 is also configured as follows: Based on motion status information, one of several preset motion modes, including diagonal motion mode, is selected as the target motion mode, including: When the vehicle speed is lower than the first threshold in the motion status information and the driver actively triggers the first instruction to perform angled parking, the low-speed angled parking mode is selected as the target motion mode. When the vehicle speed exceeds the second threshold and a second instruction from the intelligent driving system is received instructing the vehicle to perform an automatic lane change, the high-speed cooperative lane change mode is selected as the target motion mode.

[0081] Optionally, the solver module 703 is also configured as follows: Based on the vehicle motion model, the target trajectory is solved into the target vehicle body motion vector, including: When the target motion mode is low-speed inclined parking mode, the low-speed inclined kinematic model is adopted to solve the target motion trajectory into the desired longitudinal velocity component and the desired lateral velocity component, and the desired yaw rate is zero. When the target motion mode is a high-speed cooperative lane change mode, a high-speed cooperative dynamics model is adopted to solve the target motion trajectory into the desired lateral velocity correction and / or the desired yaw rate correction.

[0082] Optionally, build module 704 is also configured as follows: The optimization objectives of the optimization allocation problem include minimizing the change in the steering angle of each wheel of the vehicle relative to the previous moment and / or the estimated slip energy of each wheel of the vehicle.

[0083] Optionally, execution module 705 is also configured as follows: Based on the safety boundary corresponding to the target motion mode, monitor the vehicle's safety status parameters in real time. In response to safety status parameters exceeding safety boundaries, intervention control is executed or the target motion mode is forcibly exited.

[0084] Optionally, execution module 705 is also configured as follows: Implementing intervention controls or forcibly exiting the target movement mode, including: In low-speed angled parking mode, the tire slip ratio and the safe distance between the vehicle and obstacles are monitored. When the tire slip ratio exceeds the preset slip threshold or the safe distance is less than the preset safe threshold, the wheel angle is reduced. In high-speed cooperative lane change mode, the system monitors the status of lateral acceleration and vehicle stability control system. In response to lateral acceleration exceeding the comfort threshold or the stability control system intervening, the system sets the desired lateral speed correction and / or the desired yaw rate correction to zero to exit high-speed cooperative lane change mode.

[0085] Optional, also includes: For each wheel of the vehicle, obtain the actual steering angle of the wheel; The deviation between the actual turning angle and the corresponding turning angle value in the turning command is used as feedback, and the optimization allocation problem at the next moment is corrected based on the feedback.

[0086] In summary, the diagonal driving control device proposed in this application first acquires the vehicle's current motion state information and target trajectory. Then, based on the motion state information, it selects one of several preset motion modes, including a diagonal motion mode, as the target motion mode. Further, based on the target motion mode, it calls the vehicle motion model corresponding to that mode and calculates the target trajectory as a target vehicle body motion vector. Using the target vehicle body motion vector as a constraint, it constructs an optimization allocation problem with the target turning angles of each of the vehicle's four wheels as variables. Solving this optimization allocation problem generates a set of turning angle commands containing four independent turning angle values. Finally, the turning angle commands are output to the independent steering actuators of the vehicle's four wheels to control the vehicle to achieve diagonal driving. This application selects the corresponding solution model according to different operating modes, which not only achieves a one-click diagonal entry operation effect, providing users with a stable and smooth user experience, but also makes the vehicle lane-changing process smoother, effectively optimizing the driving experience. It fully unleashes the performance potential of the corner module hardware, improves the control accuracy and response speed of the system, further expands the boundaries of the vehicle's motion function, and significantly improves the vehicle's maneuverability in various specific scenarios. The targeted safety strategy can also ensure that the corresponding functions have good reliability and availability, fully meet the requirements of actual engineering implementation, and continuously provide users with a stable and reliable user experience.

[0087] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, in implementing this application, the functions of each module can be implemented in one or more software and / or hardware.

[0088] The apparatus of the above embodiments is used to implement the corresponding method in any of the foregoing embodiments and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0089] Corresponding to the above embodiments, this application also proposes an electronic device. (See reference...) Figure 8 This is a block diagram of an electronic device according to some embodiments of the present application, illustrating a more specific hardware structure of an electronic device provided by an embodiment of the present application. The device may include: a processor 810, a memory 820, an input / output interface 830, a communication interface 840, and a bus 850. The processor 810, memory 820, input / output interface 830, and communication interface 840 are internally connected to each other via the bus 850.

[0090] The processor 810 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0091] The memory 820 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 820 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 820 and is called and executed by the processor 810.

[0092] The input / output interface 830 is used to connect input / output modules to enable information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0093] The communication interface 840 is used to connect the communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0094] Bus 850 includes a pathway for transmitting information between various components of the device, such as processor 810, memory 820, input / output interface 830, and communication interface 840.

[0095] It should be noted that although the above-described device only shows the processor 810, memory 820, input / output interface 830, communication interface 840, and bus 850, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0096] The electronic devices described above are used to implement the corresponding diagonal driving control methods in any of the foregoing embodiments and have corresponding beneficial effects, which will not be elaborated further here.

[0097] Based on the same concept, corresponding to the diagonal driving control method provided in any of the above embodiments, this application also provides a computer-readable storage medium storing a program or instructions, which, when executed by a processor, implements the diagonal driving control method as described above.

[0098] The aforementioned computer-readable storage medium can be any available medium or data storage device that a computer can access, including but not limited to magnetic storage (e.g., floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical storage (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (e.g., ROMs, EPROMs, EEPROMs, non-volatile memory (NAND flash), solid-state drives (SSDs)).

[0099] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the corresponding diagonal driving control method in any of the foregoing embodiments, and have corresponding beneficial effects, which will not be described in detail here.

[0100] It should be noted that, in this document, 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 limitations, 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 that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0101] From the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of computer software products plus necessary general-purpose hardware platforms, and of course, they can also be implemented by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes several instructions to cause the terminal or network-side device to execute the methods described in the various embodiments of this application.

[0102] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other implementations under the guidance of this application without departing from the spirit and scope of the claims. All of these implementations are within the protection scope of this application.

Claims

1. A diagonal driving control method, characterized in that, include: Obtain the vehicle's current motion status information and the target's motion trajectory; Based on the motion state information, select one of the multiple preset motion modes, including the diagonal motion mode, as the target motion mode. Based on the target motion mode, the vehicle motion model corresponding to the target motion mode is invoked, and the target motion trajectory is calculated into a target vehicle body motion vector based on the vehicle motion model; Using the target vehicle body motion vector as a constraint, an optimization allocation problem is constructed with the target steering angles of the four wheels of the vehicle as variables. A set of steering commands containing four independent steering angle values ​​is generated by solving the optimization allocation problem. The turning command is output to the independent steering actuators of the four wheels of the vehicle to control the vehicle to drive diagonally.

2. The diagonal driving control method according to claim 1, characterized in that, The step of selecting one of multiple preset motion modes, including a diagonal motion mode, as the target motion mode based on the motion state information includes: When the vehicle speed is lower than a first threshold in the motion state information and a first instruction triggered by the driver to perform angled parking is received, a low-speed angled parking mode is selected as the target motion mode. When the vehicle speed exceeds the second threshold and a second instruction is received from the intelligent driving system instructing the vehicle to perform an automatic lane change, the high-speed cooperative lane change mode is selected as the target motion mode.

3. The diagonal driving control method according to claim 2, characterized in that, The step of solving the target motion trajectory into a target vehicle body motion vector based on the vehicle motion model includes: When the target motion mode is the low-speed inclined parking mode, the low-speed inclined kinematic model is used to solve the target motion trajectory into the desired longitudinal velocity component and the desired lateral velocity component, and the desired yaw rate is zero. When the target motion mode is the high-speed cooperative lane change mode, a high-speed cooperative dynamics model is used to solve the target motion trajectory into the desired lateral velocity correction and / or the desired yaw rate correction.

4. The diagonal driving control method according to claim 1, characterized in that, The optimization objective of the optimization allocation problem includes minimizing the change in the steering angle of each wheel of the vehicle relative to the previous moment and / or the estimated slip energy of each wheel of the vehicle.

5. The diagonal driving control method according to claim 3, characterized in that, The method further includes: Based on the safety boundary corresponding to the target motion mode, monitor the vehicle's safety status parameters in real time. In response to the safety status parameter exceeding the safety boundary, intervention control is performed or the target motion mode is forcibly exited.

6. The diagonal driving control method according to claim 5, characterized in that, The execution of intervention control or forced exit from the target motion mode includes: In the low-speed inclined parking mode, the tire slip ratio and the safe distance between the vehicle and the obstacle are monitored. In response to the tire slip ratio exceeding a preset slip threshold or the safe distance being less than a preset safe threshold, the wheel angle is reduced. In the high-speed cooperative lane change mode, the status of the lateral acceleration and vehicle stability control system is monitored. In response to the lateral acceleration exceeding the comfort threshold or the stability control system intervening, the desired lateral speed correction and / or desired yaw rate correction are set to zero to exit the high-speed cooperative lane change mode.

7. The diagonal driving control method according to claim 1, characterized in that, The method further includes: For each wheel of the vehicle, obtain the actual rotation angle of that wheel; The deviation between the actual turning angle and the corresponding turning angle value in the turning angle command is used as a feedback quantity, and the optimization allocation problem at the next moment is corrected based on the feedback quantity.

8. A diagonal driving control device, characterized in that, include: The acquisition module is configured to acquire the vehicle's current motion state information and the target's motion trajectory; The selected module is configured to select one of multiple preset motion modes, including diagonal motion mode, as the target motion mode based on the motion state information. The calculation module is configured to, based on the target motion mode, call the vehicle motion model corresponding to the target motion mode, and calculate the target motion trajectory into a target vehicle motion vector based on the vehicle motion model; The construction module is configured to construct an optimization allocation problem with the target vehicle body motion vector as a constraint and the target steering angle of each of the four wheels of the vehicle as a variable, and generate a set of steering commands containing four independent steering angle values ​​by solving the optimization allocation problem; The execution module is configured to output the turning command to the independent steering actuators of the four wheels of the vehicle to control the vehicle to drive diagonally.

9. An electronic device, characterized in that, include: A processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the diagonal driving control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the diagonal driving control method as described in any one of claims 1 to 7.

Images