Parking control methods, modules and vehicles

Abstract

This disclosure provides a parking control method, module, and vehicle, relating to the field of vehicle control technology. The method includes: acquiring multi-source perception data of a target vehicle, including geometric constraint information of the parking space boundary and the orientation information of neighboring vehicles; determining the target parking pose of the target vehicle based on the multi-source perception data and parking specifications matching the parking scenario in which the target vehicle is located; and performing coordinated parking control of the entire vehicle based on the target parking pose to enable the target vehicle to enter the target parking space. By fusing multi-source perception data and parking specifications to determine the parking angle, adaptive and precise control for various parking scenarios is achieved, improving the reliability and safety of the parking process.

Description

Technical Field

[0001] This disclosure relates to the field of vehicle control technology, specifically to parking control methods, modules, and vehicles. Background Technology

[0002] Automated parking systems assist users in automatically planning parking trajectories and controlling the vehicle to complete parking operations, thereby reducing parking difficulty and alleviating driving burden. In related technologies, automated parking systems typically operate as independent subsystems, relying on surround-view cameras and ultrasonic sensors for local environmental perception. They cannot collaborate with other intelligent ecosystems, resulting in parking planning depending solely on static geometric information such as parking space lines. They fail to perceive or utilize dynamic environmental information within the parking area, leading to poor compliance of the final parking posture. Summary of the Invention

[0003] In view of this, this disclosure provides a parking control method, module, and vehicle to address the problem of insufficient coordination between current automatic parking systems and other intelligent ecosystems during the parking process.

[0004] In a first aspect, one embodiment of this disclosure provides a parking control method, comprising: acquiring multi-source perception data of a target vehicle, the multi-source perception data including geometric constraint information of the parking space boundary and adjacent vehicle orientation information of surrounding vehicles; determining the target parking position of the target vehicle based on the multi-source perception data and parking specifications matching the parking scenario in which the target vehicle is located; and performing vehicle-wide cooperative parking control based on the target parking position to enable the target vehicle to enter the target parking space.

[0005] The above-mentioned parking control method improves the parking posture planning process by introducing multi-source perception data and combining it with scene-matching parking standards. While ensuring the parking success rate, it enhances the scene coordination between the final parking posture and the real environment, and ensures the compliance of parking behavior.

[0006] In conjunction with the first aspect, in some implementations of the first aspect, the target parking pose of the target vehicle is determined based on multi-source perception data and parking specifications that match the parking scenario in which the target vehicle is located. This includes: calculating the average heading angle of the surrounding vehicles based on the orientation information of the neighboring vehicles, which serves as a dynamic reference orientation, and the dynamic reference orientation represents the overall parking orientation of the surrounding vehicles; determining the direction of the static parking space line based on geometric constraint information; and determining the target parking pose based on the direction of the static parking space line, the dynamic reference orientation, and the parking specifications.

[0007] The above-mentioned parking control method integrates the dynamic orientation of surrounding vehicles with the static direction of parking lines, so that the calculated target parking position can reflect the parking habits of surrounding vehicles without deviating from the direction of the parking lines themselves. The information sources are more comprehensive, making the parking control method better adaptable to the environment.

[0008] In conjunction with the first aspect, in some implementations of the first aspect, the parking specifications include an angle specification threshold; based on the static parking space line direction, the dynamic reference orientation, and the parking specifications, the target parking pose is determined, including: fusing the dynamic reference orientation and the static parking space line direction to obtain the target parking angle; if the target parking angle exceeds the angle specification threshold, the target parking angle is corrected based on the static parking space line direction; and the target parking pose is determined based on the corrected target parking angle.

[0009] The above parking control method fuses the dynamic reference orientation and the static parking space line direction to obtain the target parking angle, and sets a standard threshold and correction mechanism for the angle. When the fused angle deviates from the standard, the static benchmark is mainly referenced to ensure the compliance of the parking result and avoid the overall posture from deviating from the standard due to excessive following of the orientation of neighboring vehicles.

[0010] In conjunction with the first aspect, in some implementations of the first aspect, the multi-source sensing data also includes the slope information of the target vehicle; the method also includes: determining the real-time slope angle of the target vehicle based on the slope information; determining the slope compensation angle based on the real-time slope angle when the real-time slope angle exceeds a preset slope threshold; and correcting the target parking pose based on the slope compensation angle.

[0011] The above-mentioned parking control method actively counteracts the gravitational force on the slope by acquiring slope information and calculating dynamic compensation angle, ensuring that the final posture of the vehicle on the slope is consistent with the preset target. This solves the problem of vehicle body deflection caused by gravity on non-level roads and improves parking accuracy and robustness in complex terrain.

[0012] In conjunction with the first aspect, in some implementations of the first aspect, vehicle-wide cooperative parking control is performed based on the target parking pose, including: generating a target parking trajectory based on the target parking pose, the target parking trajectory including a first path point and a second path point; performing parking control on the target vehicle to make the target vehicle travel along the target parking trajectory; when the target vehicle travels to the first path point, reducing the vehicle speed, and sequentially performing at least two levels of steering fine-tuning control, wherein the steering angle of the later level of fine-tuning is smaller than that of the earlier level, so as to make the target vehicle travel to the second path point.

[0013] The above-mentioned parking control method, by adopting a multi-level progressive angle steering control from coarse to fine, makes fine adjustments to the vehicle's posture at the end stage, reducing the problem of vehicle swaying or inaccurate stopping caused by a single large-angle steering, and solving the problem of insufficient accuracy in the end alignment stage of conventional parking control methods, thus achieving high-precision parking alignment.

[0014] In conjunction with the first aspect, some implementations of the first aspect further include: performing dynamic collision detection on the target parking trajectory based on the automatic braking system of the target vehicle; adjusting the target parking trajectory when the distance between any path point on the target parking trajectory and an obstacle is less than a safe distance; and / or performing emergency braking on the target vehicle.

[0015] The above parking control method incorporates an automatic braking system for real-time collision detection. Once a collision risk is detected, it responds immediately, reducing the potential collision risk caused by sensor blind spots or dynamic environmental changes, and improving the safety of the parking process.

[0016] In conjunction with the first aspect, some implementations of the first aspect further include: acquiring the overall vehicle status data of the target vehicle, the overall vehicle status data including the target vehicle's battery status information and / or driver attention information; adjusting the control precision of parking control based on the battery status information; and / or adjusting the safety distance based on the driver attention information.

[0017] The above-mentioned parking control method introduces vehicle status data as the basis for adjusting parking strategies, enabling the parking control process to be flexibly adjusted according to the vehicle's own status and the driver's status, thereby improving the safety and practicality of automatic parking under different working conditions.

[0018] In conjunction with the first aspect, some implementations of the first aspect also include: determining the real-time slope angle of the target vehicle based on multi-source sensing data; controlling the power system of the target vehicle to output reverse torque based on the real-time slope angle, and controlling the electronic parking brake system of the target vehicle to preload electronic parking force.

[0019] The above-mentioned parking control method addresses the safety hazard of vehicles easily rolling away on slopes. Depending on the slope, the power system outputs reverse torque to prevent rolling, while the electronic parking brake system preloads some braking force in advance, thus improving the safety and reliability of parking on slopes.

[0020] Secondly, one embodiment of this disclosure provides a parking control module deployed in a vehicle domain controller. The parking control module interacts with other functional modules deployed in the vehicle domain controller through a standardized service interface to implement the parking control method of the first aspect described above.

[0021] Thirdly, one embodiment of this disclosure provides a vehicle, including: a processor; and a memory for storing executable instructions of the processor; the processor is configured to execute the parking control method of the first aspect by executing the executable instructions. Attached Figure Description

[0022] Figure 1 The diagram shown is a schematic flowchart of a parking control method provided in an embodiment of this disclosure.

[0023] Figure 2 The diagram shown is a schematic flowchart of a parking control method provided in another embodiment of this disclosure.

[0024] Figure 3 The diagram shown is a schematic representation of a parking control device provided in an embodiment of this disclosure.

[0025] Figure 4 The diagram shown is a structural schematic of a vehicle provided in an embodiment of this disclosure. Detailed Implementation

[0026] The technical solutions of 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 of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0027] In related technologies, automatic parking control typically achieves automatic parking by using geometric parking space lines for path planning and attitude control. Specifically, this method uses surround-view cameras to identify ground markings and ultrasonic sensors to detect parking space boundaries and surrounding obstacles. Using the parking space lines as a reference, it calculates a collision-free parking path and controls the steering, power, and braking systems to follow this path until the vehicle stops inside the parking space, thus achieving automatic parking and reducing the driver's workload.

[0028] However, because this approach relies solely on static parking space line geometry, while it can achieve basic parking positioning, in complex scenarios with multiple parked vehicles, the vehicle's orientation after parking may differ from that of neighboring vehicles, leading to decreased user trust in the automatic parking function. This is because automatic parking systems typically operate as independent systems within the vehicle, lacking data collaboration with other intelligent ecosystem components (such as Advanced Driver Assistance Systems (ADAS), surround view systems, Vehicle-to-Everything (V2X) communication modules, high-precision maps, and energy management systems). This prevents parking decisions from utilizing deployed vehicle perception resources, and the parking strategy, based solely on geometric parking space lines, may result in a parking posture that does not conform to parking lot traffic regulations, affecting subsequent vehicle entry and exit and overall parking order.

[0029] To address the above issues, this disclosure provides a parking control method, module, and vehicle. The parking control method includes: acquiring multi-source perception data of the target vehicle, including geometric constraint information of the parking space boundary and the orientation information of neighboring vehicles; determining the target parking pose of the target vehicle based on the multi-source perception data and parking specifications matching the parking scenario of the target vehicle; and performing vehicle-wide cooperative parking control based on the target parking pose to enable the target vehicle to enter the target parking space. By introducing multi-source perception data and combining it with scenario-matched parking specifications, the planning stage of parking pose is improved. While ensuring parking success rate, the coordination between the final parking pose and the real-world environment is enhanced, and the compliance of parking behavior is ensured.

[0030] First, we introduce the functional architecture of a target vehicle provided in one embodiment of this disclosure. In this embodiment, the target vehicle includes a vehicle domain controller, multiple sensing components, and multiple actuators.

[0031] The target vehicle can be a vehicle with automatic parking function, such as a pure electric or plug-in hybrid new energy vehicle with automatic parking function.

[0032] The target vehicle's functional architecture is divided into a perception layer, a decision-making layer, and an execution layer. The perception layer is responsible for perceiving the surrounding environment and collecting and processing environmental and in-vehicle information. This includes perception components such as the surround-view system, the camera module of the Advanced Driver Assistance System (ADAS), the Vehicle to Everything (V2X) communication module, the Inertial Measurement Unit (IMU), and ultrasonic radar. Each perception component in the perception layer can be connected to the vehicle domain controller via a Controller Area Network with Flexible Data-Rate (CAN FD) or an Ethernet bus.

[0033] The decision-making layer is responsible for path planning and navigation, making driving decisions by executing corresponding control strategies. This includes the parking control module, the Automatic Emergency Braking (AEB) module, and the energy management system. The modules in the perception layer can be registered as Service-Oriented Architecture (SOA) services in the Zone Control Unit (ZCU) and communicate with other modules through service interfaces.

[0034] In related technologies, automatic parking functions typically operate on a separate controller, such as an Electronic Control Unit (ECU). This ECU is directly connected to sensors and actuators, forming a closed system that cannot effectively acquire information from other modules (such as battery level, driver status, and navigation maps). This results in parking decisions being limited to local environmental perception, making it difficult to achieve vehicle-wide intelligent coordination.

[0035] In this embodiment, the parking control module is deployed within the vehicle domain controller. The parking control module runs as a software process or a standard service module and interacts with other functional modules within the domain controller through standardized service interfaces.

[0036] Optionally, the parking control module is registered based on the Automotive Open System Architecture (AUTOSAR) SOA and is deeply integrated into the perception layer, decision-making layer, execution layer and human-machine interface (HMI) of the whole vehicle system.

[0037] The execution layer is responsible for the specific control operations of the target vehicle, such as acceleration, braking, and steering, including the braking system, power system, steering system, and electronic parking brake system. The modules of the execution layer can be connected to the vehicle domain controller via CAN FD.

[0038] The human-machine interaction system can obtain the status of the entire vehicle in real time and interact with the user.

[0039] Figure 1 The diagram shown is a schematic flowchart of a parking control method provided in an embodiment of this disclosure. Figure 1 As shown in the embodiments of this disclosure, the parking control method includes the following steps.

[0040] S110: Acquire multi-source perception data of the target vehicle.

[0041] The acquisition of multi-source sensing data can be achieved through the sensing components described above. Multi-source sensing data can be a collection of environmental information from multiple different sensors.

[0042] In this embodiment of the disclosure, the multi-source sensing data includes geometric constraint information of the parking space boundary and the orientation information of neighboring vehicles. For example, feature extraction, target recognition, and multi-sensor fusion can be performed on the raw data from the multi-source sensors to determine the orientation information of neighboring vehicles surrounding the target parking space, as well as the geometric constraint information of the parking space boundary.

[0043] The orientation information of neighboring vehicles can be used to determine the parking posture of the surrounding vehicles around the target parking space, providing a reference for planning the parking angle of the target vehicle. For example, the orientation information of neighboring vehicles can be the angle of the front of the surrounding vehicles in the vehicle coordinate system of the target vehicle.

[0044] Geometric constraint information can be information describing the geometric contour of the target parking space, providing a reference for determining the effective space and orientation of the target parking space. For example, geometric constraint information can be the position and direction of parking space lines and curbs. Optionally, geometric constraint information can be represented in the form of a set of coordinate points or line segments, for example, the coordinates of the four corner points of the parking space. In addition, geometric constraint information can also include the geometric contours of other parking spaces surrounding the target parking space.

[0045] In this embodiment of the disclosure, multi-source sensing data can be collected by at least two of the sensing components, such as a surround-view camera, an ADAS forward-view camera, or a V2X communication module.

[0046] The surround-view camera can capture surrounding images, identify parking lines or outlines of nearby parking spaces, and extract corresponding geometric constraint information. The ADAS forward-facing camera can capture parking lines or outlines of parking spaces at a greater distance and identify other vehicles around the target parking space, thereby obtaining neighboring vehicle orientation information and geometric constraint information. The V2X communication module can receive status information of surrounding vehicles to determine the neighboring vehicle orientation information.

[0047] Optionally, acquiring multi-source perception data of the target vehicle also includes: echo signals from ultrasonic radar, used to simultaneously detect whether there are obstacles around the target vehicle in order to determine the safety boundary.

[0048] S120 determines the target parking position of the target vehicle based on multi-source perception data and parking specifications that match the parking scenario of the target vehicle.

[0049] Parking regulations can be rules or standards that match specific parking scenarios. For example, in the underground parking lot of a shopping mall, the parking regulation is that the angle between the front of the car and the parking space line should be within 3°; in the ground parking lot of a residential community, the parking regulation is that the angle between the front of the car and the parking space line should be within 5°; and in a multi-level mechanical parking garage, the parking regulation is that the angle between the front of the car and the parking space line should be within 2°.

[0050] Optionally, parking rules can be pre-stored in the target vehicle's memory or dynamically obtained via V2X or other methods.

[0051] For example, the current parking lot type can be determined using the vehicle location information of the target vehicle; subsequently, parking specifications matching that parking lot type are read from a parking specification library stored in memory. This parking specification library can be stored in JSON format and supports updates via Over-The-Air (OTA) technology, thereby ensuring the reliability of the parking specifications.

[0052] The target parking pose refers to the final spatial state parameters that the target vehicle needs to achieve when parking in the target parking space; that is, the target parking pose represents the final target posture of the parking path planning. For example, the target parking pose includes the target parking angle, target position coordinates, etc. The target parking angle represents the desired heading angle of the target vehicle after parking in the target parking space; the target position coordinates represent the desired position of the reference point of the target vehicle after parking in the target parking space.

[0053] The target parking position is determined after comprehensively considering factors such as the posture of adjacent vehicles, the direction of the parking space, the size of the parking space, and the safety clearance, while ensuring that parking regulations are met.

[0054] Optionally, based on multi-source perception data and parking rules matching the parking scenario of the target vehicle, the target parking pose is determined, including: identifying the type of the current parking scenario and determining the corresponding parking rules based on the parking scenario; based on the orientation information and geometric constraint information of neighboring vehicles in the multi-source perception data, analyzing the actual parking posture of surrounding vehicles and the geometric contour of the parking space, and determining the target parking pose that can both meet the parking rules and avoid collision risks. This target parking pose is the benchmark for subsequent vehicle cooperative parking control.

[0055] In some embodiments, statistical indices of the adjacent vehicle orientation information of surrounding vehicles can be calculated first, and then the target parking position can be determined based on the statistical indices. For example, the statistical indices can be one of the maximum, minimum, median, or mean of the adjacent vehicle orientation information of surrounding vehicles.

[0056] To improve the accuracy of the target parking position, fixed static geometric constraints can be extracted from the parking space itself, and dynamic reference orientations can be extracted from the parking postures of surrounding vehicles. Then, combined with parking regulations, a target parking position that meets the site geometry requirements and can safely avoid neighboring vehicles can be determined.

[0057] In this embodiment of the disclosure, the target parking pose of the target vehicle is determined based on multi-source perception data and parking specifications matching the parking scenario of the target vehicle. This includes: calculating the average heading angle of the surrounding vehicles based on their heading angle information, which serves as a dynamic reference orientation, whereby the dynamic reference orientation represents the overall parking orientation of the surrounding vehicles; determining the static parking line direction of the target parking space based on geometric constraint information; and determining the target parking pose based on the static parking line direction, the dynamic reference orientation, and the parking specifications.

[0058] Dynamic reference orientation reflects the parking posture of surrounding vehicles and serves as an alignment reference for the target vehicle when parking. For example, dynamic reference orientation can be a reference angle derived from statistical indices of the actual heading angles of vehicles surrounding the target parking space.

[0059] Optionally, the dynamic reference orientation can be a reference angle obtained based on the average of the actual heading angles of vehicles surrounding the target parking space, used to characterize the overall parking trend of the neighboring vehicle group.

[0060] Optionally, the dynamic reference orientation can be the angle obtained by taking the arithmetic mean or weighted average of the heading angles of all vehicles around the target parking space.

[0061] The direction of static parking space lines can reflect the geometric characteristics of the parking space itself and can serve as a reference direction for the target vehicle when parking. For example, the direction of static parking space lines can be a reference angle extracted based on the geometric constraint information of the target parking space.

[0062] Optionally, the direction of the static parking space line can be a reference angle obtained based on the direction of the parking space line, the direction of the curb, or the direction of the parking lot markings, representing the orientation of the parking space. For example, the direction of the static parking space line can be the direction of the long side line of the target parking space.

[0063] Optionally, the direction of the static parking space line can be the orientation of the parking space obtained by fitting a straight line based on the set of parking space boundary points or by identifying the direction of the parking space line.

[0064] In some embodiments, the target parking pose includes a target parking angle. The target parking angle is adaptively adjusted based on a dynamic reference direction, provided that parking regulations are met. Specifically, the target parking angle is a value that conforms to parking regulations and is close to the dynamic reference orientation, so that the target vehicle maintains the same posture as adjacent vehicles. Correspondingly, determining the target parking pose based on the static parking line direction, the dynamic reference orientation, and the parking regulations includes: first, determining a parking angle range according to the parking regulations; then, determining the deviation between the dynamic reference orientation and the static parking line direction. If the deviation is small and conforms to the parking angle range required by the parking regulations, the value corresponding to the dynamic reference orientation can be set as the target parking angle.

[0065] Optionally, the multi-source sensing data, target parking angle, etc., can be in the world coordinate system or in the vehicle coordinate system of the target vehicle. This disclosure does not make specific limitations in this regard.

[0066] For example, if the parking rule stipulates that the angle between the front of the car and the parking space line is within 3°, and the static parking space line direction is 90°, and the heading angles of two cars surrounding the target parking space are 92° and 88° respectively (i.e., their angles with the parking space line are +2° and -2° respectively), the average heading angle is 90°, meaning the dynamic reference orientation is 90°. In this case, the dynamic reference orientation is consistent with the static parking space line direction, and combined with the parking rule, the target parking angle is set to 90°.

[0067] The parking control method in this embodiment introduces a dynamic reference orientation, which enables the target parking posture to reflect the parking posture of surrounding vehicles without deviating from the direction of the parking space line itself. The information source is more comprehensive, improving the scene coordination between the final parking posture and the real environment.

[0068] In some embodiments, the parking specifications include an angle specification threshold; determining the target parking pose based on the static parking line direction, the dynamic reference orientation, and the parking specifications includes: fusing the dynamic reference orientation and the static parking line direction to obtain the target parking angle; if the target parking angle exceeds the angle specification threshold, correcting the target parking angle based on the static parking line direction; and determining the target parking pose based on the corrected target parking angle.

[0069] The included angle standard threshold can be used to limit the reasonable range of the target parking angle. It can be the maximum allowable deviation range between the direction of the target vehicle's front and the direction of the static parking space line specified in the parking regulations. For example, the included angle standard threshold can be 2°, and correspondingly, the reasonable range of the target parking angle can be 90°±2°.

[0070] The target parking angle is obtained by fusing the dynamic reference orientation and the static parking line direction. Optionally, a fusion rule can be preset, and the dynamic reference orientation and the static parking line direction can be processed based on the fusion rule to obtain the target parking angle.

[0071] In some embodiments, the fusion rule can be a weighted average, with weights set according to confidence levels. For example, the target parking angle θ_target can be calculated using the following formula: θ_target = a × θ_avg + b × θ_line, where a and b are the weights of the dynamic reference orientation (θ_avg) and the static parking line direction (θ_line), respectively. As a specific example, if the weight of the static parking line direction is 0.4, the weight of the dynamic reference direction is 0.6, the static parking line direction is 90°, and the dynamic reference orientation is 93°, then the target parking angle is 0.6 × 93° + 0.4 × 90° = 91.8°. Further, parking regulations specify an included angle threshold of 2°, meaning the parking angle range is 88° to 92°. The target parking angle falls within this range and therefore requires no correction. If it exceeds this range, the target parking angle can be corrected based on the static parking line direction.

[0072] As an optional implementation, if the deviation between the target parking angle (θ_target) and the direction of the static parking line (θ_line) exceeds the angle specification threshold (θ_th), i.e. |θ_target-θ_line|>θ_th, then the direction of the static parking line is taken as the new target parking angle.

[0073] As another optional implementation method, the target parking angle is corrected based on the direction of the static parking line, including: determining the boundary value of the target parking angle based on the direction of the static parking line and the included angle standard threshold, and selecting the boundary value of the adjacent target parking angle as the new target parking angle.

[0074] In some embodiments, the included angle specification threshold defines the upper and lower boundary values ​​of the parking angle. When the target parking angle exceeds the parking angle range, the target parking angle is set to the value of the static parking line direction plus or minus the threshold boundary. Specifically, the upper limit and lower limit of the angle are determined based on the static parking line direction and the included angle specification threshold. When the target parking angle exceeds the upper limit, the upper limit is used as the new target parking angle; when the target parking angle is lower than the lower limit, the lower limit is used as the new target parking angle.

[0075] For example, if the upper limit of the angle is 92°, the lower limit of the angle is 88°, and the target parking angle is 94°, then when the upper limit of the angle is exceeded, the target parking angle is determined to be 92°; when the target parking angle is lower than the lower limit of the angle, 88°, the target parking angle is determined to be 88°.

[0076] The parking control method described above achieves a balance between environmental adaptability and regulatory compliance by fusing the dynamic reference orientation with the static parking line direction and correcting it based on the static parking line direction when the fusion result exceeds the specified angle threshold. This ensures safe and compliant parking and improves the robustness and generalization ability of automatic parking control in complex scenarios.

[0077] In some embodiments, to improve the safety and accuracy of automatic parking control, the target parking angle can be compensated based on slope information to adapt the parking angle to the sloping road conditions, thereby improving the safety of parking on slopes. This disclosure provides an optional embodiment, the specific implementation of which is described below.

[0078] In this embodiment of the disclosure, the multi-source sensing data also includes the slope information of the target vehicle; the automatic parking control method further includes: determining the real-time slope angle of the target vehicle based on the slope information; determining the slope compensation angle based on the real-time slope angle when the real-time slope angle exceeds a preset slope threshold; and correcting the target parking posture based on the slope compensation angle.

[0079] Slope information can be information describing the degree of inclination of the road surface where the target vehicle is located, such as the inclination direction and angle of the slope where the vehicle is located.

[0080] Alternatively, slope information can be obtained through an IMU, vehicle height sensor, or wheel speed sensor.

[0081] The real-time slope angle represents the actual inclination angle of the current road surface. Optionally, the real-time slope angle can be the longitudinal slope and / or the lateral slope. The longitudinal slope can be the angle between the vehicle's longitudinal axis and the horizontal plane; the lateral slope can be the angle between the transverse axis and the horizontal plane.

[0082] The preset slope threshold can be a set angle threshold used to determine whether slope compensation needs to be activated. For example, the preset slope threshold can be set to 2°. When the real-time slope angle is less than or equal to the preset slope threshold, it is determined that the current slope has little impact on the parking posture and no correction is required; when the real-time slope angle exceeds the preset slope threshold, it is determined that the current slope has a significant impact on the parking posture and correction is required.

[0083] The slope compensation angle can be an additional angle correction calculated for slope conditions. It is used to adjust the target parking posture (more specifically, to adjust the target parking angle in the target parking posture) so that the vehicle body posture maintains a reasonable alignment with the horizontal plane after parking on the slope, or to avoid the risk of slippage caused by the slope.

[0084] As an optional implementation method, the slope compensation angle can be calculated by looking up a table or by an algorithm. Alternatively, a correspondence can be established between multiple slope angles and multiple slope compensation angles; based on this correspondence, the slope compensation angle corresponding to the real-time slope angle can be determined, thus obtaining the slope compensation angle.

[0085] In addition, a correspondence can be established between multiple slope angles and multiple longitudinal slope compensation angles and multiple transverse slope compensation angles. Based on this correspondence, the longitudinal slope compensation angle and transverse slope compensation angle corresponding to the real-time slope angle can be calculated to obtain the longitudinal slope compensation angle and transverse slope compensation angle. Based on the longitudinal slope compensation angle and transverse slope compensation angle, the slope compensation angle can be generated.

[0086] In some embodiments, correcting the target parking angle based on the ramp compensation angle includes: superimposing the ramp compensation angle onto the target parking angle, incrementally adjusting the target parking angle to obtain the corrected target parking angle, and performing parking control based on the corrected target parking angle.

[0087] The parking control method described above introduces a slope compensation angle when the slope exceeds a preset slope threshold, enabling the parking control module to adapt to the slope conditions and reducing the risk of vehicle body posture imbalance or slippage on the slope.

[0088] S130 performs coordinated parking control of the entire vehicle based on the target parking position and posture, so that the target vehicle can drive into the target parking space.

[0089] Vehicle-wide cooperative parking control allows the vehicle domain controller to coordinate the steering, power, braking, and parking systems during the parking process, responding synchronously to control commands to achieve precise trajectory tracking and attitude adjustment.

[0090] In some embodiments, this step can be implemented by controlling the modules in the execution layer. For example, the steering system controls the direction of the target vehicle, the power system outputs corresponding torque to maintain the target speed, the braking system applies braking force when needed, and the parking system preloads a certain parking force to better adapt to slope parking scenarios.

[0091] Specifically, the parking control module plans a parking trajectory based on the target parking position. It then sends commands to each actuator, which coordinates their actions according to a unified timing sequence to drive into the target parking space. During the journey, the sensing components continuously monitor the surroundings, achieving closed-loop control.

[0092] The parking control method described above improves the parking posture planning process by introducing multi-source perception data and combining it with scene-matching parking rules. While ensuring parking success rate, it enhances the scene coordination between the final parking posture and the real environment, making the parking behavior closer to the actual scene and breaking the limitation of traditional automatic parking that relies solely on fixed rules.

[0093] To make the execution of parking control smoother and more precise, this disclosure provides an optional embodiment, the specific implementation of which is described below.

[0094] In this embodiment of the disclosure, vehicle-wide cooperative parking control is performed based on the target parking pose, including: generating a target parking trajectory based on the target parking pose, the target parking trajectory including a first path point and a second path point; performing parking control on the target vehicle to make the target vehicle travel along the target parking trajectory; when the target vehicle travels to the first path point, reducing the vehicle speed, and sequentially performing at least two levels of steering fine-tuning control, wherein the steering angle of the later level of fine-tuning is smaller than that of the earlier level, so as to make the target vehicle travel to the second path point.

[0095] The target parking trajectory can be a vehicle driving path planned based on the target parking position, including the starting position of the target vehicle to the final position where it is parked in the target parking space.

[0096] The target parking trajectory may include a series of waypoints, which include a first waypoint and a second waypoint.

[0097] The first path point is a node on the target parking trajectory. During the normal parking phase (i.e., the process of the target vehicle traveling from the starting position to the first path point), the target vehicle travels at a set speed (e.g., 1.5 km / h), and the steering system takes over the steering wheel with a control accuracy of ±0.5°.

[0098] After the first waypoint, the target vehicle enters the fine-tuning phase. For example, the first waypoint could be located in front of a parking space, such as 0.5 meters from the parking space entrance. At the first waypoint, the target vehicle has completed approximate position or orientation adjustments, laying the foundation for subsequent precise fine-tuning.

[0099] During the fine-tuning phase (i.e., the process of the target vehicle traveling from the first path point to the second path point), the powertrain reduces the target vehicle's speed, for example (0.8 km / h), while simultaneously coordinating with the steering system to make minor directional adjustments to achieve precise alignment.

[0100] The second waypoint can be another node on the target parking trajectory, at which the target vehicle has completed the fine-tuning phase. Optionally, the second waypoint can be the end point of the target parking trajectory. Alternatively, the second waypoint can be located within the target parking space, after which the target vehicle can drive in a straight line to the end point of the target parking trajectory.

[0101] In some embodiments, parking control is performed on the target vehicle to make the target vehicle travel along a target parking trajectory, including: the parking control module coordinates the steering system, power system, braking system and electronic parking brake system to control the vehicle to park in the target parking space according to the target parking trajectory until the first path point is reached.

[0102] In this embodiment of the disclosure, when the vehicle reaches the first waypoint, it enters steering fine-tuning control, which includes at least two levels, and the steering angle of the at least two levels gradually decreases.

[0103] Among them, steering fine-tuning control refers to a control method that uses a gradually decreasing steering angle adjustment to make closed-loop corrections to the vehicle's heading angle during the parking alignment phase.

[0104] Optionally, the execution cycle of each fine-tuning step can be determined based on the sensor feedback frequency and control cycle. For example, each fine-tuning step can last for 0.2 seconds or one control cycle. This progressively decreasing design can effectively avoid overshooting or oscillation of the vehicle due to excessive single correction.

[0105] For example, the steering fine-tuning control includes two sequentially executed fine-tuning stages. The first stage involves adjusting the steering angle by a larger step size for a first duration, rapidly converging the vehicle's heading angle deviation to a smaller range. The second stage involves adjusting the steering angle by a smaller step size for a second duration, achieving precise alignment of the vehicle's heading angle. Optionally, the vehicle speed in the first stage is greater than the speed in the second stage; using a higher speed in the first stage improves parking efficiency, while using a lower speed in the second stage ensures alignment accuracy, thus achieving a balance between efficiency and precision.

[0106] For example, fine-tuning control can also employ a three-level closed-loop fine-tuning control, with each level progressively improving steering accuracy: the steering system first performs an adjustment of ±0.8° for 0.3 seconds, obtaining the current deviation based on sensor feedback; then, a closed-loop correction of ±0.4° for 0.2 seconds is performed to further reduce the error; finally, a fine-tuning of ±0.2° for 0.1 seconds is performed to achieve precise alignment. The entire process embodies a hierarchical control strategy from coarse to fine adjustment.

[0107] Optionally, during automatic parking, a maximum speed (e.g., 5 km / h) can be set to allow the target vehicle to park at a lower speed to ensure safety and accuracy.

[0108] The parking control method described above introduces at least two levels of steering fine-tuning control, with the steering angle of the later level being smaller than that of the earlier level. This achieves precise and smooth correction of the parking posture in the later stages. This progressive control can reduce over-adjustment and vehicle body vibration caused by a single large angle adjustment, and improve the tracking accuracy and stability of the final stage of the parking trajectory.

[0109] In some embodiments, to improve safety, collision detection can also be performed on the target parking trajectory, as described below.

[0110] In this embodiment of the disclosure, the parking control method further includes: performing dynamic collision detection on the target parking trajectory based on the automatic braking system of the target vehicle; adjusting the target parking trajectory when the distance between any path point on the target parking trajectory and an obstacle is less than a safe distance; and / or performing emergency braking on the target vehicle.

[0111] Understandably, if the distance between any waypoint on the target parking trajectory and an obstacle is less than a safe distance, the trajectory adjustment should be attempted first to maintain parking continuity. If adjustment is not feasible, emergency braking should be triggered directly to ensure safety.

[0112] For example, the automatic braking system may be an AEB module.

[0113] In some embodiments, dynamic collision detection is performed based on real-time perception data and the target parking trajectory. Correspondingly, dynamic collision detection is performed on the target parking trajectory based on the automatic braking system of the target vehicle. This includes: during the parking process, continuously evaluating the distance relationship between each path point on the target parking trajectory and surrounding obstacles based on the real-time detected perception data and the target parking trajectory, so as to determine whether there is a collision risk based on the distance relationship.

[0114] The safe distance is the minimum permissible distance threshold used to determine whether to trigger avoidance or braking. Understandably, this safe distance can be dynamically adjusted based on factors such as vehicle speed, obstacle type (e.g., static pillars, moving pedestrians), and road surface adhesion. For example, in low-speed parking scenarios, the safe distance for static obstacles can be set to 0.3 meters, and the safe distance for dynamic obstacles can be set to 0.8 meters.

[0115] Optionally, if the distance between any path point on the target parking trajectory and an obstacle is less than a safe distance, the target parking trajectory may be adjusted. This includes: if a collision risk is determined when the distance between any path point on the target parking trajectory and an obstacle is less than a safe distance, the target parking trajectory may be locally or globally corrected. For example, the positions of some path points may be adjusted, the curvature may be changed, or detour points may be inserted, so that the adjusted target parking trajectory meets the safe distance requirements. The new target parking trajectory can still be accurately tracked, and parking control can continue.

[0116] In some embodiments, when the distance between any waypoint on the target parking trajectory and an obstacle is less than a safe distance, emergency braking is performed on the target vehicle, including: if the obstacle suddenly approaches, or if the distance between the waypoint and the obstacle is a critical safe value, the automatic braking system is immediately triggered to intervene and apply braking force to bring the vehicle to a smooth stop.

[0117] The parking control method described above provides a dual-mode response mechanism by coupling dynamic collision detection with the automatic braking system. It maintains the availability of the automatic parking function through trajectory adjustment, and at the same time avoids collision risks through emergency braking, thereby improving the reliability and user trust of the parking control method in open and complex environments.

[0118] In some embodiments, the parking control module may also incorporate vehicle status data as a basis for adjusting the safe distance. Correspondingly, the parking control method further includes: acquiring vehicle status data of the target vehicle, the vehicle status data including battery status information and / or driver attention information of the target vehicle; adjusting the control precision of the parking control based on the battery status information; and / or adjusting the safe distance based on the driver attention information.

[0119] Vehicle status data can be a set of data describing the current overall operating status of the target vehicle. In this embodiment of the disclosure, vehicle status data includes battery status information and / or driver attention information.

[0120] Optionally, battery status information can be collected by the Battery Management System (BMS) module in the vehicle domain controller. Battery status information may include battery state of charge (SOC), battery temperature, or available power, etc.

[0121] If the battery status information indicates that the target vehicle's battery charge is low (e.g., below the first charge threshold, such as 20%), the control precision of parking control can be appropriately reduced, and ramp parking can be prohibited.

[0122] Control accuracy can be used to describe the accuracy with which parking control tracks the target trajectory. In some embodiments, reducing control accuracy can be achieved by at least one of the following measures: relaxing alignment accuracy, increasing the minimum step size for steering fine-tuning, and reducing path point density. These adjustments can effectively reduce the operating load on the domain controller and motors, thereby reducing power consumption and computing resource usage.

[0123] If the battery status information indicates that the target vehicle's battery has sufficient charge (e.g., above the second charge threshold, such as 80%), the control precision can be appropriately increased to ensure smooth and accurate parking.

[0124] Driver attention information can be collected by a Driver Monitoring System (DMS). Driver attention information describes whether the driver's attention is currently focused on the driving task, and is typically generated by a camera in conjunction with facial recognition algorithms (such as eye tracking and head posture analysis). For example, driver attention information can be represented as an attention level, specifically categorized as high, medium, or low.

[0125] In some embodiments, the vehicle status data includes driver attention information of the target vehicle. Correspondingly, adjusting the safe distance based on the vehicle status data includes: if the driver attention information indicates low driver attention (e.g., below a preset attention threshold), the safe distance can be increased to prioritize safety redundancy. If the driver attention information indicates high driver attention (e.g., above a preset attention threshold), the default safe distance can be maintained.

[0126] In some embodiments, the above two adjustments can be performed independently or simultaneously. Furthermore, the adjusted control accuracy parameters and safety distance thresholds are applied to the dynamic collision detection and trajectory adjustment in the embodiments of this disclosure to achieve adaptive parking safety control of the vehicle state.

[0127] The parking control method described above, by introducing vehicle status data, can effectively extend system operating time or increase driving range by actively reducing control precision in low-battery scenarios; and can compensate for the risks caused by human reaction delays by actively increasing the safety distance in driver distraction scenarios, thus achieving a dynamic balance between energy efficiency and safety and improving the safety and practicality of automatic parking under different operating conditions.

[0128] To make parking control safer and more reliable under different road conditions, this disclosure provides an optional embodiment, the specific implementation of which is described below.

[0129] In this embodiment of the disclosure, the parking control method further includes: determining the real-time slope angle of the target vehicle based on multi-source sensing data; controlling the power system of the target vehicle to output reverse torque based on the real-time slope angle, and controlling the electronic parking brake system of the target vehicle to preload electronic parking force.

[0130] It is understandable that when a vehicle is parked on a slope, if only the conventional braking system is relied upon, the vehicle may roll back after parking. Therefore, this embodiment detects the real-time slope angle and controls the power system to output a reverse torque opposite to the direction of the gravity component to counteract the downward trend. On the other hand, it controls the electronic parking brake system to preload the parking force, further improving the safety of parking on slopes.

[0131] Optionally, a slope threshold can be set. If the real-time slope angle exceeds the threshold, it is determined that there is a risk of vehicle rollback, and subsequent control is executed; if it does not exceed the threshold, the normal parking procedure is followed. It should be noted that the slope threshold can be the same as or different from the aforementioned preset slope threshold, and this disclosure does not limit it in this way.

[0132] Reverse torque can be a torque output by the powertrain that is opposite to the direction of the vehicle's current motion. For example, when a vehicle is going uphill and trying to roll backward, the motor outputs a forward torque to counteract the downward force of gravity; when a vehicle is going downhill and trying to roll forward, the motor outputs a backward torque. This torque is usually small and is used to keep the vehicle stationary rather than to propel it forward.

[0133] Optionally, controlling the output of reverse torque by the powertrain includes: calculating the reverse torque required to keep the vehicle stationary based on the real-time slope angle and the target vehicle's mass data. Specifically, based on the real-time slope angle, the component of gravity along the slope can be calculated, and then, combined with the wheel radius and transmission ratio, converted into the reverse torque that the motor should output. Further, the vehicle domain controller sends a torque command to the powertrain, causing the motor to output this reverse torque to counteract the downward force and keep the target vehicle stationary.

[0134] The electronic parking brake system can respond to the commands of the parking control module and achieve parking braking by controlling the rear wheel brake calipers or motor, and has the ability to respond quickly and preload.

[0135] Pre-loaded electronic parking brake force can be achieved by the EPB system applying a certain amount of braking force (not a complete lock-up) before the target vehicle comes to a complete stop or is about to stop. This shortens the time from when the driver releases the brake pedal to when the parking brake actually takes effect, and at the same time makes the parking process smoother. Optionally, this force can be 20% to 50% of the rated parking force, for example, it can be 30% of the rated parking force.

[0136] Optionally, the electronic parking brake system of the target vehicle is preloaded with an electronic parking force, including: based on the real-time slope angle, the component of gravity sliding down the slope can be determined, and then, combined with the braking efficiency of the braking system and the clamping force coefficient of the brake calipers, the target preload force to be applied by the electronic parking brake system is calculated. Further, the parking control module sends a preload command to the electronic parking brake system controller, causing the brake calipers to clamp the brake disc in advance with the target preload force, establishing the predetermined clamping force before the target vehicle comes to a complete stop, accelerating the parking response, and effectively preventing the vehicle from rolling back on the slope.

[0137] Optionally, controlling the electronic parking brake system of the target vehicle to preload the electronic parking force and controlling the power system to output reverse torque can be performed simultaneously or sequentially according to a preset priority.

[0138] The parking control method described above utilizes the coordinated control of the reverse torque of the power system triggered by the real-time slope angle and the preload force of the electronic parking brake system. The reverse torque provides an immediate response to counteract the gravitational force, while the preload force of the electronic parking brake system reduces parking response delay, making automatic parking equally reliable on sloping roads.

[0139] The above embodiments illustrate the parking control method in detail. The following will combine... Figure 2 The specific implementation process of the above parking control method is explained in detail.

[0140] Figure 2 The diagram shown is a schematic flowchart of a parking control method provided in another embodiment of this disclosure. Figure 2 As shown, the parking control method of this disclosure includes the following steps.

[0141] S210: Acquire multi-source perception data of the target vehicle.

[0142] First, when the user activates the automatic parking function, the in-vehicle human-machine interaction system publishes a standardized event to the vehicle's SOA event bus.

[0143] Then, the parking control module within the vehicle domain controller (ZCU) responds. Specifically, upon receiving the event, the parking control module enters a parking preparation state and synchronously invokes the following SOA services to obtain data and make decisions: sensor availability verification service, power management service, driver monitoring status service, and navigation service. The sensor availability verification service can be used to verify whether the ADAS forward-facing camera, surround-view camera, V2X communication module, and IMU are all online and fault-free.

[0144] The power management service can be used to obtain the State of Charge (SOC). If the SOC is less than the low battery threshold (e.g., 20%), the parking control module relaxes the control precision of parking control, increasing it from 2° to ±5°. Simultaneously, if a ramp parking space is detected, it automatically excludes such spaces, prohibiting ramp parking and selecting only level parking spaces to prevent ineffective ramp anti-rollover control due to insufficient power. The driver monitoring status service obtains the driver's current attention score. If the driver's attention is below the preset attention threshold, the safety distance is adjusted from 0.3 meters to 0.5 meters, triggering trajectory adjustment or emergency braking earlier; simultaneously, manual fine-tuning is disabled, preventing the driver from manually intervening in the parking trajectory via the steering wheel or touchscreen to prevent accidents caused by accidental operation while distracted. The navigation service obtains the current parking lot type and specific parking space specifications from the high-precision map to automatically match the parking specifications corresponding to the parking scenario (e.g., "perpendicular parking space - Chinese standard - allowable deviation 2°") for subsequent target parking position pose calculation.

[0145] Finally, after completing the above service calls and initial decisions, the parking control module sends an activation command to the underlying perception components to enter the high-precision perception mode.

[0146] Specifically, the ADAS forward-facing camera switches to wide-angle mode (FOV 120°) to expand the lateral field of view and better capture the heading angles of adjacent vehicles on both sides of the parking space. The surround-view camera increases image resolution to the highest level to improve the accuracy of parking space boundary line recognition. The V2X communication module activates V2V monitoring mode to receive the headings of adjacent vehicles. The IMU increases the sampling frequency to 100Hz to more accurately capture changes in vehicle attitude, providing high-frequency input for subsequent hill-start compensation and anti-rollover control.

[0147] For example, when a target vehicle enters the underground parking garage of a shopping mall, a surround-view camera identifies the parking space boundary. The heading angles of the three surrounding vehicles are 2°, 4°, and 3°, respectively, and the IMU reports that the road slope is 0° (horizontal).

[0148] Based on the information collected by the aforementioned underlying perception components, multi-source perception data is determined. This multi-source perception data includes geometric constraint information of the parking space boundary and the orientation information of neighboring vehicles.

[0149] S220 calculates the average heading angle of surrounding vehicles based on their orientation information, and uses this as a dynamic reference orientation.

[0150] Specifically, the arithmetic mean of the heading angles of the three surrounding vehicles is determined as the average heading angle of the surrounding vehicles, which serves as a dynamic reference orientation.

[0151] S230, based on geometric constraint information, determines the direction of the static parking line of the target parking space.

[0152] Specifically, based on the fused image of the ADAS forward-looking camera and the surround-view system, the Hough transform algorithm is used to extract the geometric principal direction of the parking space line as the direction of the static parking space line.

[0153] S240 determines the target parking position based on the static parking line direction, dynamic reference orientation, and parking regulations.

[0154] The target parking position includes the target parking angle.

[0155] First, a weighted fusion is used to generate an initial parking angle based on the dynamic reference orientation and the static parking line direction. Then, in accordance with parking regulations, if the deviation between the fused initial parking angle and the parking line direction exceeds the allowable deviation of the parking regulations, the parking line direction is forcibly used as the target parking angle.

[0156] Furthermore, to avoid exceeding the final attitude limits due to sensor errors or environmental disturbances, a certain alignment tolerance range (e.g., ±0.5°) can be set based on the target parking angle (θ_target) to obtain the final target parking angle (θ_final), thereby improving the robustness of parking control. For example, the final target parking angle can be expressed as θ_final = θ_target ± 0.5° Furthermore, the real-time slope angle of the target vehicle can be determined based on the slope information; if the real-time slope angle exceeds the preset slope threshold, the slope compensation angle can be determined based on the real-time slope angle; and the target parking angle can be corrected based on the slope compensation angle.

[0157] Based on the real-time slope angle (θ_slope) measured by IMU, when the real-time slope angle exceeds a preset slope threshold (e.g., 2°), the slope compensation angle is determined based on the slope compensation coefficient and the real-time slope angle. For example, the slope compensation angle can be calculated using the following formula: θ_comp = c × θ_slope, where c represents the slope compensation coefficient, used to adjust the compensation intensity. For example, c can be set to 0.4.

[0158] The sum of the slope compensation angle and the target parking angle is determined as the final output target parking angle (θ_output) to counteract the tendency of vehicle body deflection caused by gravity. That is, if θ_slope > 2°, then θ_comp = 0.4 × θ_slope, and the final output target parking angle is θ_output = θ_final + θ_comp.

[0159] S250 generates a target parking trajectory containing a first path point and a second path point based on the target parking pose.

[0160] First, the basic module of the Automatic Parking Assist (APA) system is invoked to plan a parking trajectory, using the target parking position as a constraint. This parking trajectory consists of a series of waypoints arranged at 0.2-meter intervals.

[0161] Then, mark two key points on the parking trajectory: the first path point, located 0.5 meters from the final target position, marks the entry point for the fine-tuning alignment stage; the second path point, the final stopping point within the target parking space, signifies the completion of parking upon the vehicle's arrival.

[0162] At the same time, a safety check is performed: the automatic emergency braking module is invoked to perform dynamic collision prediction to ensure that the distance between any point on the parking trajectory and the obstacle is greater than or equal to 0.3 meters; if it is less than this value, the parking trajectory is automatically replanned.

[0163] S260 performs parking control on the target vehicle so that the target vehicle travels along the target parking trajectory.

[0164] To control the target vehicle into the normal parking phase, the vehicle speed can be set to 1.5 km / h, the steering system can be set to take over the steering wheel, and the control accuracy can be set to ±0.5°.

[0165] The vehicle domain controller coordinates the EPS, power system, and braking system to track the vehicle segment by segment according to the target parking trajectory path until the vehicle reaches the first path point.

[0166] Throughout the parking process, the AEB module remains in a standby state in the background, continuously monitoring the dynamic collision risk of obstacles to the vehicle. By fusing perception layer data in real time, the AEB module performs dynamic collision prediction on all path points along the target parking trajectory. When it detects that the distance between any path point and an obstacle is below a safe threshold, or when an obstacle (such as a pedestrian or non-motorized vehicle) unexpectedly intrudes into the parking area, the AEB module can autonomously trigger a full-vehicle braking command within 100 milliseconds without waiting for the driver's response, achieving emergency braking of the entire vehicle.

[0167] During the parking process, when the target vehicle reaches the first path point, the vehicle speed is reduced, and at least two levels of steering fine-tuning control are executed sequentially.

[0168] In this process, the steering angle of the subsequent fine-tuning is smaller than that of the previous level, so that the target vehicle can travel to the second path point.

[0169] Specifically, when the vehicle reaches the first path point, it switches to fine control mode, which manifests as follows: the powertrain reduces the vehicle speed from 1.5 km / h to 0.8 km / h (approximately 0.22 m / s) to reduce the effects of inertia. The steering system performs three levels of fine adjustments, with each level lasting for a specific time before a closed-loop correction is made based on sensor feedback.

[0170] Specifically, the first-level fine-tuning control accuracy can be ±0.8°, and the duration can be 0.3 seconds, for larger corrections; the second-level fine-tuning control accuracy can be ±0.4°, and the duration can be 0.3 seconds, for fine correction based on feedback, reducing deviations; the third-level fine-tuning control accuracy can be ±0.2°, and the duration can be 0.1 seconds, for final precise alignment, eliminating residual errors.

[0171] After each level of fine-tuning, sensors (such as IMU and surround-view cameras) measure the current actual heading angle, allowing the parking control module to compare it with the target parking pose and determine the correction direction and magnitude for the next level of fine-tuning. The steering angle of each subsequent level is smaller than that of the previous level, achieving incremental convergence.

[0172] After three levels of fine-tuning, the target vehicle continued to drive to the second path point and completed the parking maneuver.

[0173] Optionally, after the vehicle enters the target parking space, the parking completion phase begins, including the following steps.

[0174] First, the results are verified. If the error between the actual heading angle of the target vehicle and the target parking position is less than or equal to 1.5°, it is marked as compliant parking. Subsequently, the HMI system synchronously displays the parking results through the instrument panel, central control screen, and Augmented Reality Head-Up Display (AR-HUD), including the orientation error value (e.g., 0.9°) and a comparison diagram of the virtual parking space line and the actual front arrow of the vehicle, intuitively presenting the parking accuracy to the user.

[0175] For example, the HMI system is built on a shared event bus and a unified user interface (UI) framework. By interfacing with the vehicle's SOA event bus, it can subscribe to and respond to various events issued by the parking control module in real time. These events include parking status, safety warnings, system feedback, and driver interaction requests.

[0176] Secondly, a buffer period begins.

[0177] Specifically, the driver releases control of the steering wheel but maintains a speed of 0.8 km / h for 5 seconds, allowing for minor adjustments to the vehicle's attitude. If the driver's operation lasts longer than 0.3 seconds, the amount of manual correction will be recorded for subsequent analysis.

[0178] Then, the buffer period ends, and the execution state resumes.

[0179] Specifically, the steering system disengages from control and restores EPS power assist; ADAS switches back to normal mode; V2X enters low-power monitoring mode; and the electronic parking brake is activated, completing the vehicle's status recovery.

[0180] Optionally, after parking is completed, all relevant data (including vehicle position, final posture, battery level, driver monitoring information, perception data, etc.) are encrypted and uploaded to the vehicle data platform. The platform uses aggregated anonymization processing to continuously optimize two core knowledge bases using a large amount of parking data while protecting user privacy: one is a parking standard knowledge base, such as identifying "a parking space line on the B2 floor of a shopping mall has a 3° deviation", so that subsequent vehicles can be pre-set with corresponding compensation; the other is a fine-tuning control parameter base, such as "when the slope is >4°, the slope compensation coefficient is increased to 0.5" to improve control accuracy in high-slope scenarios.

[0181] Furthermore, based on the optimized knowledge base and parameters, OTA upgrade packages can be generated and pushed to the entire fleet via over-the-air download technology, realizing an evolutionary closed loop from single-vehicle intelligence to collective intelligence.

[0182] Those skilled in the art will understand that the above embodiments are merely illustrative examples. Without departing from the core ideas of this disclosure, reasonable modifications can be made to the order of steps, parameter values, and adjustment strategies, all of which fall within the protection scope of this disclosure.

[0183] Figure 3 The diagram shown is a structural schematic of a parking control device provided in an embodiment of this disclosure. Figure 3 As shown, the parking control device provided in this embodiment includes: an acquisition module 301, a determination module 302, and a control module 303.

[0184] The acquisition module 301 is configured to acquire multi-source perception data of the target vehicle, including geometric constraint information of the parking space boundary and the orientation information of neighboring vehicles; the determination module 302 is configured to determine the target parking pose of the target vehicle based on the multi-source perception data and parking specifications matching the parking scenario in which the target vehicle is located; the control module 303 is configured to perform whole-vehicle cooperative parking control based on the target parking pose to enable the target vehicle to drive into the target parking space.

[0185] In some embodiments, the determining module 302 is configured to: calculate the average heading angle of the surrounding vehicles based on the adjacent vehicle orientation information of the surrounding vehicles, as a dynamic reference orientation, the dynamic reference orientation representing the overall parking orientation of the surrounding vehicles; determine the static parking space line direction of the target parking space based on geometric constraint information; and determine the target parking position posture based on the static parking space line direction, the dynamic reference orientation, and the parking specifications.

[0186] In some embodiments, the determining module 302 is configured to fuse the dynamic reference orientation with the static parking line direction to obtain the target parking angle; if the target parking angle exceeds the included angle specification threshold, correct the target parking angle based on the static parking line direction; and determine the target parking pose based on the corrected target parking angle.

[0187] In some embodiments, the determining module 302 is configured to: determine the real-time slope angle of the target vehicle based on slope information; determine the slope compensation angle based on the real-time slope angle if the real-time slope angle exceeds a preset slope threshold; and correct the target parking posture based on the slope compensation angle.

[0188] In some embodiments, the control module 303 is configured to generate a target parking trajectory based on the target parking pose, the target parking trajectory including a first path point and a second path point; perform parking control on the target vehicle to make the target vehicle travel along the target parking trajectory; and, when the target vehicle travels to the first path point, sequentially perform at least two levels of steering fine-tuning control, wherein the steering angle of the later level of fine-tuning is smaller than that of the earlier level, so as to make the target vehicle travel to the second path point.

[0189] In some embodiments, the control module 303 is further configured to perform dynamic collision detection on the target parking trajectory based on the automatic braking system of the target vehicle; adjust the target parking trajectory if the distance between any path point on the target parking trajectory and an obstacle is less than a safe distance; and / or perform emergency braking on the target vehicle.

[0190] In some embodiments, the control module 303 is further configured to acquire vehicle status data of the target vehicle, the vehicle status data including battery status information and / or driver attention information of the target vehicle; adjust the control precision of parking control based on the battery status information; and / or adjust the safety distance based on the driver attention information.

[0191] In some embodiments, the control module 303 is further configured to determine the real-time slope angle of the target vehicle based on multi-source sensing data; based on the real-time slope angle, control the power system of the target vehicle to output reverse torque, and control the electronic parking brake system of the target vehicle to preload electronic parking force.

[0192] In one embodiment of this disclosure, a parking control module is also provided, which is deployed in the vehicle domain controller. The parking control module interacts with other functional modules deployed in the vehicle domain controller through a standardized service interface to implement the parking control method of the above embodiment.

[0193] Figure 4 The diagram shown is a structural schematic of a vehicle provided according to an embodiment of this disclosure. Figure 4As shown, vehicle 40 includes one or more processors 401 and memory 402. Processor 401 may be a central processing unit (CPU) or other processing unit with data processing capabilities and / or instruction execution capabilities, and can control other components in vehicle 40 to perform desired functions. Memory 402 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and processor 401 may execute program instructions to implement the parking control methods of the various embodiments of this disclosure described above and / or other desired functions. Various contents, such as port allocation algorithms, load balancer resource instances, and development machine resource instances, may also be stored in the computer-readable storage medium.

[0194] In one example, vehicle 40 may further include an input device 403 and an output device 404, these components being interconnected via a bus system and / or other forms of connection mechanisms (not shown). The input device 403 may include, for example, a keyboard, a mouse, etc. The output device 404 may output various information to the outside, including port allocation algorithms, load balancer resource instances, development machine resource instances, etc. The output device 404 may include, for example, a display, speakers, a printer, and a communication network and its connected remote output devices, etc.

[0195] In addition to the methods and apparatus described above, embodiments of this disclosure may also be computer program products, including computer program instructions that, when executed by a processor, cause the processor to perform the steps of the parking control methods according to various embodiments of this disclosure as described above.

[0196] Computer program products can be written in any combination of one or more programming languages ​​to perform the operations of embodiments of this disclosure. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0197] Furthermore, embodiments of this disclosure may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the parking control methods according to various embodiments of this disclosure described above. The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0198] It should also be noted that in the apparatus, devices, and methods of this disclosure, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions to this disclosure.

[0199] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A parking control method, characterized in that, include: Acquire multi-source perception data of the target vehicle, including geometric constraint information of the parking space boundary and the orientation information of neighboring vehicles. Based on the multi-source perception data and the parking specifications that match the parking scenario of the target vehicle, the target parking position of the target vehicle is determined. Based on the target parking position, vehicle-wide coordinated parking control is performed to enable the target vehicle to enter the target parking space.

2. The parking control method according to claim 1, characterized in that, The step of determining the target parking position of the target vehicle based on the multi-source sensing data and parking specifications matching the parking scenario of the target vehicle includes: Based on the neighboring vehicle orientation information of the surrounding vehicles, the average heading angle of the surrounding vehicles is calculated as a dynamic reference orientation, which represents the overall parking orientation of the surrounding vehicles. Based on the geometric constraint information, the direction of the static parking line of the target parking space is determined; The target parking position is determined based on the direction of the static parking space line, the dynamic reference orientation, and the parking specifications.

3. The parking control method according to claim 2, characterized in that, The parking regulations include the included angle regulation threshold; Determining the target parking position based on the direction of the static parking space line, the dynamic reference orientation, and the parking regulations includes: The target parking angle is obtained by fusing the dynamic reference orientation with the static parking space line direction; If the target parking angle exceeds the specified angle threshold, the target parking angle is corrected based on the direction of the static parking space line. The target parking pose is determined based on the corrected target parking angle.

4. The parking control method according to claim 2, characterized in that, The multi-source sensing data also includes the slope information of the target vehicle; The method further includes: Based on the slope information, the real-time slope angle of the target vehicle is determined; If the real-time slope angle exceeds a preset slope threshold, a slope compensation angle is determined based on the real-time slope angle. The target parking position is corrected based on the slope compensation angle.

5. The parking control method according to claim 1, characterized in that, The step of performing vehicle-wide cooperative parking control based on the target parking pose includes: A target parking trajectory is generated based on the target parking pose, and the target parking trajectory includes a first path point and a second path point; Parking control is applied to the target vehicle so that the target vehicle travels along the target parking trajectory; When the target vehicle reaches the first path point, the vehicle speed is reduced, and at least two levels of steering fine-tuning control are executed sequentially, wherein the steering angle of the later level of fine-tuning is smaller than that of the earlier level, so that the target vehicle reaches the second path point.

6. The parking control method according to claim 5, characterized in that, Also includes: Based on the automatic braking system of the target vehicle, dynamic collision detection is performed on the target parking trajectory; If the distance between any path point on the target parking trajectory and an obstacle is less than a safe distance, the target parking trajectory shall be adjusted, and / or emergency braking shall be applied to the target vehicle.

7. The parking control method according to claim 6, characterized in that, Also includes: Obtain the overall vehicle status data of the target vehicle, including the battery status information and / or driver attention information of the target vehicle; Based on the battery status information, adjust the control precision of the parking control. and / or The safe distance is adjusted based on the driver's attention level information.

8. The parking control method according to claim 1, characterized in that, Also includes: Based on the multi-source sensing data, the real-time slope angle of the target vehicle is determined; Based on the real-time slope angle, the power system of the target vehicle is controlled to output reverse torque, and the electronic parking brake system of the target vehicle is controlled to preload electronic parking force.

9. A parking control module, characterized in that, Deployed in the vehicle domain controller, the parking control module interacts with other functional modules deployed in the vehicle domain controller through a standardized service interface to implement the method described in any one of claims 1 to 8.

10. A vehicle, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the method of any one of claims 1 to 8 by executing the executable instructions.

Images