Automatic parking method and device, storage medium and electronic equipment

Abstract

The invention discloses an automatic parking method and device, a storage medium and electronic equipment. The automatic parking method comprises the following steps: in response to an automatic parking instruction, determining an initial vehicle pose of a vehicle and an initial parking space pose of a target parking space; determining an initial parking track based on the initial vehicle pose and the initial parking space pose; determining a parking track correction area based on the initial parking space pose; controlling the vehicle to park based on the initial parking track; in response to the condition that the vehicle runs to the parking track correction area, determining a corrected parking space pose of the target parking space; determining a corrected parking track based on the initial parking space pose, the corrected parking space pose and the initial parking track; and controlling the vehicle to park based on the corrected parking track. In the automatic parking process, the parking space posture and the parking track are corrected through the method, the accuracy of the parking posture and the parking efficiency can be improved, and parking to the target parking space can be achieved automatically, accurately and rapidly.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent driving technology, and in particular to an automatic parking method, device, storage medium, and electronic device. Background Art

[0002] As an important branch of intelligent driving, automatic parking technology relies on sensors (such as ultrasonic radar, millimeter-wave radar, and cameras) and algorithmic decision-making to achieve autonomous parking of the vehicle.

[0003] Typically, the starting position of an automated parking system is at a certain distance and angle from the target parking space. This results in a discrepancy between the observed position of the target parking space, as perceived by sensors (such as cameras), and its actual position. Parking based on this observed position can lead to a deviation between the final parking position and the ideal parking position corresponding to the actual parking space. Therefore, the observed position of the target parking space must be updated during the parking process, and the parking trajectory must be corrected to ensure the accuracy of the final parking position. Summary of the Invention

[0004] In order to solve the above technical problems, the present disclosure provides an automatic parking method, device, storage medium and electronic device, which correct the parking position and parking trajectory during the parking process, thereby improving the accuracy of the parking position and parking efficiency.

[0005] A first embodiment of the present disclosure provides an automatic parking method, comprising:

[0006] In response to an automatic parking instruction, determining an initial vehicle pose of the vehicle and an initial parking space pose of a target parking space;

[0007] Determining an initial parking trajectory based on the initial vehicle posture and the initial parking space posture;

[0008] Determining a parking trajectory correction area based on the initial parking space posture;

[0009] Controlling parking of the vehicle based on the initial parking trajectory;

[0010] In response to the vehicle traveling into the parking trajectory correction area, determining a corrected parking space posture of the target parking space;

[0011] determining a revised parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory;

[0012] The vehicle is controlled to park based on the corrected parking trajectory.

[0013] According to a second aspect of the present disclosure, there is provided an automatic parking device, comprising:

[0014] a first determining module, configured to determine an initial vehicle posture of the vehicle and an initial parking posture of a target parking space in response to an automatic parking instruction;

[0015] a second determining module, configured to determine an initial parking trajectory based on the initial vehicle posture and the initial parking space posture;

[0016] a third determining module, configured to determine a parking trajectory correction area based on the initial parking space posture;

[0017] a first control module, configured to control parking of the vehicle based on the initial parking trajectory;

[0018] a fourth determining module, configured to determine a corrected parking position of the target parking space in response to the vehicle traveling into the parking trajectory correction area;

[0019] a fifth determining module, configured to determine a revised parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory;

[0020] A second control module is configured to control parking of the vehicle based on the corrected parking trajectory.

[0021] A third aspect embodiment of the present disclosure provides a computer-readable storage medium, which stores a computer program for executing the automatic parking method provided by the first aspect embodiment.

[0022] An embodiment of a fourth aspect of the present disclosure provides an electronic device comprising: a processor; a memory for storing processor-executable instructions; and a processor for reading executable instructions from the memory and executing the instructions to implement the automatic parking method provided in the embodiment of the first aspect.

[0023] A fifth aspect embodiment of the present disclosure provides a computer program product. When the instructions in the computer program product are executed by a processor, the automatic parking method provided by the first aspect embodiment is executed.

[0024] The automatic parking method provided by the embodiment of the present disclosure determines the initial vehicle posture and the initial parking posture of the target parking space upon receiving an automatic parking instruction; plans an initial parking trajectory based on the initial vehicle posture and the initial parking posture, and controls the vehicle to start parking based on the initial parking trajectory. When the vehicle is detected to have driven into the parking trajectory correction area corresponding to the target parking space, the posture of the target parking space is corrected to obtain a corrected parking posture; then, a corrected parking trajectory is determined based on the initial parking posture, the corrected parking posture, and the initial parking trajectory; and finally, the vehicle is controlled to park based on the corrected parking trajectory. This method corrects the parking posture and the parking trajectory during the parking process, which can improve the accuracy of the parking posture and the parking efficiency, and achieve automatic and precise parking at the target parking space. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 is a scene diagram applicable to the automatic parking method provided by an exemplary embodiment of the present disclosure;

[0026] Figure 2 is a scene diagram applicable to an automatic parking method provided by another exemplary embodiment of the present disclosure;

[0027] Figure 3 is a flowchart of an automatic parking method provided by an exemplary embodiment of the present disclosure;

[0028] Figure 4 is a flowchart of an automatic parking method provided by another exemplary embodiment of the present disclosure;

[0029] Figure 5 is a flowchart of an automatic parking method provided by yet another exemplary embodiment of the present disclosure;

[0030] Figure 6 is a schematic diagram of a parking trajectory in an automatic parking method provided by an exemplary embodiment of the present disclosure;

[0031] Figure 7 is a flowchart of an automatic parking method provided by another exemplary embodiment of the present disclosure;

[0032] Figure 8 is a flowchart of an automatic parking method provided by yet another exemplary embodiment of the present disclosure;

[0033] Figure 9 is a flowchart of an automatic parking method provided by another exemplary embodiment of the present disclosure;

[0034] Figure 10 is a flowchart of an automatic parking method provided by yet another exemplary embodiment of the present disclosure;

[0035] Figure 11is a flowchart of an automatic parking method provided by another exemplary embodiment of the present disclosure;

[0036] Figure 12 is a schematic diagram of a collision detection point corresponding to a reference trajectory point provided by an exemplary embodiment of the present disclosure;

[0037] Figure 13 1 is a schematic diagram of the structure of an automatic parking device provided by an exemplary embodiment of the present disclosure;

[0038] Figure 14 is a schematic structural diagram of an automatic parking device provided by another exemplary embodiment of the present disclosure;

[0039] Figure 15 It is a schematic diagram of the composition structure of an electronic device provided by an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0040] To explain the present disclosure, example embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. It should be understood that the present disclosure is not limited to the example embodiments.

[0041] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0042] Application Overview

[0043] With the rapid development of intelligent driving technology, the functions provided by intelligent driving vehicles are becoming increasingly rich. Automated parking systems utilize the intelligent driving system onboard intelligent driving vehicles to autonomously park vehicles without the driver having to directly control the steering wheel, accelerator, or brakes, thus achieving automatic parking.

[0044] Automatic parking systems primarily rely on perception modules (such as ultrasonic radar, lidar, onboard cameras, and position sensors) to detect environmental data, including the parking space's position (location, size, and angle), obstacle distance, and other data, as well as vehicle data. The planning module then plans a parking trajectory based on this data. The control module generates steering, power, and braking control commands based on the planned parking trajectory. Based on these commands, the control module controls the vehicle's steering, forward movement, reverse movement, and braking operations to complete automatic parking. This demonstrates that automatic parking systems can reduce the burden of manual operation, and automated parking features are becoming increasingly popular.

[0045] In existing technologies, automatic parking systems can be used to park in good lighting conditions and conventional parking spaces, which has advantages, especially for novice drivers, such as reducing the risk of scratches, shortening parking time, and reducing driver operations. However, due to the influence of actual environments, the automatic parking functions in existing technologies still have the following drawbacks:

[0046] 1) Strong dependence on the environment: The recognition accuracy of blurred parking space lines and low obstacles is low, and it is greatly affected by environmental factors such as light and weather, which affects parking efficiency and may even cause scratches, making it impossible to ensure safe and stable parking.

[0047] 2) Inaccurate parking: When the parking function is activated, the vehicle is far away from the target parking space, and there is a certain distance and angle between the starting position and the target parking space. Due to reasons such as line of sight obstruction and perspective, there may be certain errors in the collected environmental data, resulting in a certain gap between the observed position and the actual position of the target parking space. Parking is performed based on the parking trajectory planned based on the observed position, resulting in a deviation between the final parking position and the ideal parking position, making it impossible to park accurately.

[0048] Existing automated parking methods plan a parking trajectory based solely on environmental data sensed at the start of parking, and then complete parking based on this trajectory. However, due to the starting position being far from the target parking space, perceived environmental data can be biased, resulting in a deviation between the final parking position and the ideal parking position. Alternatively, during the parking process, the parking trajectory is continuously adjusted based on real-time environmental perception data. This requires a significant amount of computational data, impacting parking efficiency. Furthermore, frequent adjustments to the parking trajectory make it difficult to ensure consistent trajectory, potentially leading to additional steering wheel vibration and reduced ride comfort.

[0049] In order to solve the defects of low parking efficiency and poor accuracy existing in the automatic parking function of related technologies, the embodiment of the present disclosure provides an automatic parking method. During the automatic parking process, when it is monitored that the vehicle is driving into the parking trajectory correction area corresponding to the target parking space, the posture of the target parking space is corrected to obtain the corrected parking space posture, and then the parking trajectory is corrected. The parking of the vehicle is controlled based on the corrected parking trajectory, which can improve the accuracy of the parking posture and parking efficiency, and realize automatic, accurate and fast parking at the target parking space.

[0050] Exemplary Systems

[0051] First, the application scenarios of the present disclosure are introduced. When an intelligent driving vehicle (hereinafter referred to as the vehicle) reaches a destination and needs to park, the driver activates the automatic parking function and determines the target parking space. The intelligent driving system responds to the user's operation and generates and executes the automatic parking command.

[0052] See also Figure 1First, the initial vehicle pose at the moment the automatic parking function is activated is determined based on the perceived vehicle data. The initial parking pose of the target parking space is determined based on the perceived environmental data. Based on the initial vehicle pose and initial parking pose, an initial parking trajectory is determined. The vehicle is then controlled to begin parking based on the initial parking trajectory.

[0053] In the embodiment of the present disclosure, after the target parking space is determined, the following settings are made according to the location and size of the target parking space: Figure 1 The parking trajectory correction area is shown. During parking, the system monitors whether the vehicle enters the parking trajectory correction area. If the vehicle enters the parking trajectory correction area, the system determines the corrected parking position for the target parking space based on the environmental data perceived by the vehicle within the correction area. Because the vehicle is closer to the target parking space than at the initial moment, the perceived environmental data has less deviation. The corrected parking position determined based on this less-deviated environmental data is more accurate, providing data support for precise parking.

[0054] After the revised parking position of the target parking space is determined, a revised parking trajectory is determined based on the initial parking position, the revised parking position, and the initial parking trajectory, and the vehicle is controlled to continue parking based on the revised parking trajectory.

[0055] In the disclosed embodiment, after the target parking space is determined, one or more parking trajectory correction areas may be set according to the location and size of the target parking space.

[0056] For example, you can set up a Figure 1 The system displays a parking trajectory correction area. When a vehicle enters the correction area, it determines a corrected parking position based on the vehicle's perceived environmental data within the correction area. Based on the initial parking position, the corrected parking position, and the initial parking trajectory, it determines a corrected parking trajectory and controls the vehicle to complete parking based on the corrected parking trajectory.

[0057] For example, you can set up the following outside and inside the target parking space: Figure 2 The two parking trajectory correction areas shown. When the vehicle is detected entering the first parking trajectory correction area outside the target parking space, the first corrected parking position is determined based on the environmental data perceived by the vehicle in the first parking trajectory correction area. Based on the initial parking position, the first corrected parking position, and the initial parking trajectory, the first corrected parking trajectory is determined, and the vehicle is controlled to continue parking based on the first corrected parking trajectory. When the vehicle is detected entering the second parking trajectory correction area within the target parking space, the second corrected parking position is determined based on the environmental data perceived by the vehicle in the second parking trajectory correction area. Based on the first corrected parking position, the second corrected parking position, and the first corrected parking trajectory, the second corrected parking trajectory is determined, and the vehicle is controlled to complete parking along the second corrected parking trajectory.

[0058] In actual applications, more parking trajectory correction areas can be set. The location of the parking trajectory correction area only needs to meet the condition that the vehicle can enter the parking trajectory correction area during parking. The specific location and size are not specifically limited in the embodiments of the present disclosure.

[0059] A greater number of parking trajectory correction zones results in more frequent parking posture corrections, which in turn increases the number of parking trajectory corrections, ensuring a more accurate parking posture. A smaller number of parking trajectory correction zones results in fewer parking posture corrections, which in turn reduces the amount of data processed, allowing for faster parking to the target space and improving parking efficiency. In practice, the number of parking trajectory correction zones can be set to balance parking posture accuracy and efficiency, improving parking posture accuracy while ensuring efficiency, enabling automatic, precise, and rapid parking to the target space.

[0060] Exemplary Methods

[0061] Figure 3 FIG is a flow chart of an automatic parking method provided by an exemplary embodiment of the present disclosure. The present disclosure embodiment can be applied to electronic devices, such as Figure 3 As shown, the automatic parking method provided by the embodiment of the present disclosure may include the following steps:

[0062] Step S301 : In response to an automatic parking instruction, determining an initial vehicle posture of the vehicle and an initial parking posture of a target parking space.

[0063] The automatic parking method provided by the embodiments of the present disclosure can be applied to an automatic parking system in a vehicle, and can specifically be executed by an automatic parking device in the automatic parking system.

[0064] In one embodiment, when a vehicle reaches its destination and needs to park, the driver activates the automatic parking function. The automatic parking device, in response to the user's operation, captures images of the vehicle's surroundings from the vehicle's onboard camera in the perception module and identifies available parking spaces. Based on the identification results, the available spaces are displayed on the vehicle's central control screen. The driver selects one of the multiple available spaces displayed on the central control screen as the target parking space. The automatic parking device, using the perception module and sensor data, determines the vehicle's initial position and the initial parking position of the target parking space.

[0065] The sensors may include, but are not limited to, ultrasonic radar, lidar, vehicle-mounted camera, positioning module, inertial measurement unit (IMU), steering angle sensor, etc. The data collected by these sensors include vehicle data and environmental data.

[0066] The automatic parking device processes the vehicle data collected by the sensor through the perception module to determine the initial vehicle posture of the vehicle. For example, the initial vehicle posture may include the initial position and initial yaw angle of the vehicle when the automatic parking function is activated.

[0067] The automatic parking device senses and processes the environmental data collected by the sensor through the perception module to determine the initial parking position of the target parking space. For example, the initial parking position may include the initial position and initial angle of the target parking space.

[0068] In the disclosed embodiment, the initial position of the vehicle may be the coordinate position of the midpoint of the rear axle of the vehicle in the world coordinate system when the automatic parking function is activated, and the initial yaw angle (YawAngle) is the angle at which the vehicle rotates around its own vertical axis (Z axis) when the automatic parking function is activated.

[0069] In the disclosed embodiment, the initial position of the parking space can be the coordinate position of the target point in the parking space identified when the automatic parking function is activated in the world coordinate system. For example, the target point can be the point corresponding to the midpoint of the rear axle of the vehicle on the target parking space after the parking is expected to be completed, that is, the target position of the vehicle to be parked. The coordinate position of the target point can be calculated based on the coordinate position of the preset point in the identified parking space in the world coordinate system and the size of the target parking space. For example, the initial angle of the parking space can be the angle between the long side of the parking space and the passage identified when the automatic parking function is activated. For example, the parking angle of a common vertical parking space is 90° (the angle between the long side of the parking space and the passage is 90°), the parking angle of a side parking space is 0° (the angle between the long side of the parking space and the passage is 180°), and the parking angle of an oblique parking space is the angle between the long side of the parking space and the passage.

[0070] Step S302: determining an initial parking trajectory based on the initial vehicle posture and the initial parking space posture.

[0071] The automatic parking device can plan an initial parking trajectory based on the initial vehicle posture and initial parking space posture through a planning module. Specifically, the planning module can perform trajectory planning based on the initial position and initial yaw angle of the vehicle included in the initial vehicle posture, the initial position and initial angle of the target parking space included in the initial parking space posture, and observed environmental data (such as obstacle information), to obtain the initial parking trajectory. For example, the initial parking trajectory can be the planned trajectory points of the center of the vehicle's rear axle moving during the parking process.

[0072] Step S303: determining a parking trajectory correction area based on the initial parking space posture.

[0073] For example, after determining the initial position and initial angle of the parking space, the parking trajectory correction area can be determined based on the initial position and initial angle, as well as a preset size (a*b). For example, the area (a*b) outside the entrance and exit lines of the target parking space can be used as the parking trajectory correction area.

[0074] In some embodiments, a single parking trajectory correction area can be provided, such as one outside the target parking space or one inside the target parking space. Multiple parking trajectory correction areas can also be provided, such as one outside the target parking space and one inside the target parking space, or one outside the target parking space and two inside the target parking space. When there are multiple parking trajectory correction areas, the sizes of the multiple parking trajectory correction areas can be the same or different. The specific number, location, and size of the parking trajectory correction areas in the disclosed embodiments are not specifically limited.

[0075] Step S304: Control the vehicle to park based on the initial parking trajectory.

[0076] The automatic parking system controls the vehicle's parking process using control commands generated by a control module. Specifically, the control module generates steering, power, and braking control commands based on the initial parking trajectory. Based on these commands, the control module controls the vehicle to execute corresponding steering, forward (or reverse), and braking operations, and the vehicle begins parking.

[0077] Step S305 : In response to the vehicle driving into the parking trajectory correction area, determining a corrected parking space posture of the target parking space.

[0078] After the vehicle begins parking, the automatic parking system continuously monitors whether the vehicle enters the parking trajectory correction area. Specifically, it monitors whether the center point of the vehicle's rear axle is within the parking trajectory correction area. If the center point of the vehicle's rear axle is detected to be within the parking trajectory correction area, the conditions for correcting the parking trajectory are determined to be met, and the perception module re-perceives the corrected parking position of the target parking space. The corrected parking position can include the corrected position and angle of the target parking space.

[0079] For example, the corrected position of the target parking space can be a preset point (such as a position of a vehicle) of the target parking space identified based on the environmental data collected at that moment when the vehicle drives into the parking trajectory correction area. Figure 1 The coordinate position of the upper left corner of the target parking space in the world coordinate system. The correction angle of the target parking space can be the angle between the normal side of the target parking space and the channel, as determined by the environmental data collected at the time the vehicle enters the parking trajectory correction area.

[0080] Since the distance between the vehicle and the target parking space is closer when the vehicle reaches the parking trajectory correction area than when parking begins, the deviation of the collected environmental data is smaller. The corrected parking space posture determined based on the environmental data with smaller deviation is more accurate, thereby improving parking accuracy.

[0081] Step S306 : determining a revised parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory.

[0082] Exemplarily, after the corrected parking space posture is determined, the consistency relationship between the corrected position of the target parking space and the initial position and the consistency relationship between the corrected angle of the target parking space and the initial angle are determined.

[0083] If the consistency relationship between the revised position and the initial position indicates that the revised position is inconsistent with the initial position, or if the consistency relationship between the revised angle and the initial angle indicates that the revised angle is inconsistent with the initial angle, it indicates that there is a deviation in the initial parking posture, and the initial parking trajectory planned based on the initial parking posture also has a deviation. For example, the automatic parking device can correct the initial parking trajectory based on the initial parking posture and the corrected parking posture to obtain a corrected parking trajectory.

[0084] When the consistency relationship between the corrected position and the initial position indicates that the corrected position is consistent with the initial position, and the consistency relationship between the corrected angle and the initial angle indicates that the corrected angle is consistent with the initial angle, it means that there is no deviation in the initial parking space posture, and there is no deviation in the initial parking trajectory planned based on the initial parking space posture, and no correction is required. At this time, the initial parking trajectory is used as the corrected parking trajectory.

[0085] Step S307: Control the vehicle to park based on the corrected parking trajectory.

[0086] The automatic parking system controls the vehicle to continue parking based on control commands generated by the control module. Specifically, the control module generates steering, power, and braking control commands based on the revised parking trajectory. Based on these newly generated control commands, the control module controls the vehicle to execute corresponding steering, forward (or reverse), and braking operations. Parking is determined to be complete when the vehicle reaches the end of the revised parking trajectory.

[0087] In the disclosed embodiment, the automatic parking device, in response to an automatic parking instruction, determines the vehicle's initial vehicle posture and the initial parking posture of the target parking space; determines an initial parking trajectory based on the initial vehicle posture and the initial parking posture; determines a parking trajectory correction area based on the initial parking posture; and then controls the vehicle to begin parking based on the initial parking trajectory. When the vehicle is detected driving into the parking trajectory correction area, the corrected parking posture of the target parking space is determined; based on the initial parking posture, the corrected parking posture, and the initial parking trajectory, a corrected parking trajectory is determined; and finally, the vehicle is controlled to park based on the corrected parking trajectory. This method corrects the parking posture and parking trajectory during the parking process, which can improve the accuracy of the parking posture and parking efficiency, and achieve automatic and precise parking in the target parking space.

[0088] In some embodiments, in the above Figure 3 Based on the embodiment shown, step S306 "determining a revised parking trajectory based on the initial parking space posture, the revised parking space posture and the initial parking trajectory" may include: Figure 4 The following steps are shown:

[0089] Step S3061: Determine a first parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory.

[0090] The initial parking pose is the pose obtained by the vehicle sensing and identifying the target parking space at a longer distance, while the revised parking pose is the pose obtained by sensing and identifying the target parking space at a closer distance. Because the distance between the vehicle and the target parking space is closer when sensing the revised parking pose than when sensing the revised pose, the perceived environmental data has a smaller deviation, and the recognized revised parking pose is closer to the actual pose of the target parking space.

[0091] The automatic parking device can determine the length and direction that the vehicle's parking point needs to move from the initial position to the revised position based on the revised position of the target parking space in the revised parking space posture and the initial position of the target parking space in the initial parking space posture, and move the initial parking trajectory based on the movement length and direction to obtain a first parking trajectory.

[0092] In one implementation, the first parking trajectory can be based on Figure 5 The following steps are shown to determine:

[0093] Step S501: Determine a translation vector based on the initial parking space posture and the revised parking space posture.

[0094] Figure 6 is a schematic diagram of a parking trajectory in an automatic parking method provided by an exemplary embodiment of the present disclosure, such as Figure 6As shown, the initial position of the target parking space in the initial parking space pose is recorded as A, and the corrected position of the target parking space in the corrected parking space pose is recorded as B. Then the translation vector determined by the automatic parking device based on the initial parking space pose and the corrected parking space pose can be Figure 6 The vector shown

[0095] Step S502 : performing translation processing on the initial parking trajectory based on the translation vector to obtain a first parking trajectory.

[0096] See also Figure 6 , the position of the vehicle when it drives to the parking trajectory correction area is recorded as C, and the initial parking trajectory is l1. The initial position of the parking space in the initial parking space posture is A, and the corrected position of the parking space in the corrected parking space posture is B, then the translation vector is The initial parking trajectory is l1. Based on the translation vector The initial parking trajectory l1 is translated, that is, all the trajectory points on the initial parking trajectory l1 are translated This is equivalent to converting the parking trajectory corresponding to the initial position of the target parking space to the corrected position to obtain the first parking trajectory. For example, the first parking trajectory can be Figure 6 l3 shown.

[0097] In some embodiments, since the vehicle has been Figure 6 The initial position E shown is based on the initial parking trajectory l1 and is then translated to the current position C. The vehicle will not be retracing the trajectory it has already traveled during subsequent parking. Therefore, during translation, the initial parking trajectory l1 can be translated to obtain the first parking trajectory l3. Alternatively, the vehicle can simply translate the untraveled segment CA of the initial parking trajectory l1, without further translation of the already traveled segment l0. The corresponding first parking trajectory l3 is DB.

[0098] Step S3062: Fusing the initial parking trajectory and the first parking trajectory to obtain a fused parking trajectory.

[0099] Since trajectory correction occurs during the vehicle parking process, in order to ensure smooth parking of the vehicle, it is necessary to ensure that the end point of the vehicle parking trajectory moves smoothly from the end point of the initial parking trajectory l1 to the end point of the first parking trajectory l3. That is, when correcting the trajectory, it is necessary to consider the smoothness of the trajectory and the consistency of the trajectory at each moment with the trajectory at the previous moment. In the embodiment of the present disclosure, trajectory correction mainly includes two stages: the first stage is coarse path fusion, and the second stage is path smoothing optimization. For example, the fused parking trajectory obtained by coarse path fusion can be as follows Figure 6 l2 shown.

[0100] In the disclosed embodiment, since the trajectory l0 in the initial parking trajectory l1 has already been traveled, it is not necessary to integrate the traveled trajectory l0 during the coarse path fusion. In fact, the CA segment trajectory and the DB segment trajectory are fused, and the resulting fused parking trajectory l2 is the CB segment trajectory.

[0101] Step S3063: Optimize the fused parking trajectory to obtain a corrected parking trajectory.

[0102] Since the fused parking trajectory l2 is roughly fused based on the initial parking trajectory l1 and the first parking trajectory l3, in order to ensure smooth parking of the vehicle, the fused parking trajectory l2 needs to be optimized to obtain a smooth corrected parking trajectory.

[0103] In this disclosed embodiment, the initial parking trajectory is translated based on the corrected position of the target parking space to obtain a first parking trajectory. A coarse path fusion is performed on the initial and first parking trajectories to quickly determine an approximate parking trajectory. The resulting fused parking trajectory is then optimized to ensure smoothness, ensuring both precise and stable parking.

[0104] In some embodiments, based on the above embodiment, step S3062 of the above embodiment, “merging the initial parking trajectory and the first parking trajectory to obtain a fused parking trajectory”, may include: Figure 7 The following steps are shown:

[0105] Step S701 : determining a first vehicle position of the vehicle when the vehicle travels to a parking trajectory correction area.

[0106] For example, when the automatic parking device performs the coarse path fusion process, to ensure smooth movement of the vehicle, the front section of the fused parking trajectory l2 may gradually move away from the initial parking trajectory l1, and the rear section of the fused parking trajectory l2 may gradually move closer to the first parking trajectory l3.

[0107] In one implementation, the timing of moving away from the initial parking trajectory l1 and approaching the first parking trajectory l3 can be determined based on the length of the trajectory. For example, the length of the remaining parking trajectory can be determined based on the current position of the vehicle and the initial position of the target parking space. Based on this, the first vehicle position of the vehicle when it reaches the parking trajectory correction area is determined. For example, the first vehicle position is Figure 6 The current position C of the vehicle is shown in FIG.

[0108] Step S702 : determining a first length of a first untraveled sub-trajectory in the initial parking trajectory based on the first vehicle position and the initial parking space posture.

[0109] For example, the automatic parking device may determine a first length of the untraveled first sub-trajectory in the initial parking trajectory l1, that is, the length of the arc CA, based on the first vehicle position C and the initial position A of the target vehicle in the initial parking position.

[0110] In some embodiments, the first length of the untraveled first sub-trajectory in the initial parking trajectory l1 may also be determined based on the difference between the length of the initial parking trajectory l1 and the length of the traveled second sub-trajectory l0.

[0111] Step S703: Determine a second length between each track point on the first sub-track and the first vehicle position.

[0112] Step S704: Determine the fusion ratio corresponding to each trajectory point based on the first length and the second length.

[0113] The automatic parking system can determine the fusion ratio for each trajectory point based on the ratio of the calculated second length to the first length. Trajectory points closer to the first vehicle position C are closer to the initial parking trajectory l1 during fusion. Trajectory points farther from the first vehicle position C are closer to the first parking trajectory l3 during fusion.

[0114] For example, the fusion ratio corresponding to the first parking trajectory l3 can be recorded as k, and the fusion ratio k can be expressed as a stage function as shown in formula (1):

[0115]

[0116] Here, i represents the ratio of the second length corresponding to each trajectory point on the first sub-trajectory to the first length, that is, the ratio of the distance between each trajectory point and the first vehicle position C to the entire trajectory length of the first sub-trajectory, 0≤i≤1.

[0117] As shown in formula (1), for the trajectory points whose ratio of the second length to the first length is less than 5%, that is, the first 5% of the trajectory points on the first sub-trajectory, the corresponding fusion ratio k=0; for the trajectory points whose ratio of the second length to the first length is greater than 95%, that is, the last 5% of the trajectory points on the first sub-trajectory, the corresponding fusion ratio k=1; for the trajectory points whose ratio of the second length to the first length is greater than or equal to 5% and less than or equal to 95%, that is, the trajectory points between 5% and 95% on the first sub-trajectory, the corresponding fusion ratio k is a value between 0 and 1.

[0118] Step S705 : Based on the fusion ratio, the first parking trajectory and the first sub-trajectory are fused to obtain a fused parking trajectory.

[0119] For example, the fusion formula for fusing the first parking trajectory l3 and the first sub-trajectory can be shown as formula (2):

[0120] P l2,i =(1-k)*P CA,i +k*P DB,i (2);

[0121] Among them, P CA,i Indicates the trajectory point on the first sub-trajectory corresponding to the ratio i, P DB,i represents the trajectory point on the first parking trajectory l3 corresponding to the ratio i, P l2,i represents the point on the fused parking trajectory l2 corresponding to the ratio i. Substituting k calculated from equation (1) into equation (2) above, the fused parking trajectory l2 is calculated.

[0122] In the disclosed embodiment, a fusion ratio is determined based on the ratio of the distance between each trajectory point and the vehicle's current position to the entire untraveled trajectory. Based on the fusion ratio, the initial parking trajectory and the first parking trajectory are fused to achieve a coarse path fusion process. This allows for rapid generation of a fused parking trajectory, and thus a corrected parking trajectory, thereby improving parking efficiency.

[0123] In some embodiments, based on the above embodiment, step S3063 of the above embodiment, “optimizing the fused parking trajectory to obtain a corrected parking trajectory”, may include: Figure 8 The following steps are shown:

[0124] Step S801: Determine multiple reference trajectory points based on the fused parking trajectory.

[0125] The automatic parking device can obtain multiple reference trajectory points by sampling from all trajectory points on the fused parking trajectory l2. For example, the number of reference trajectory points can be recorded as n, and the coordinates of the jth reference trajectory point among the n reference trajectory points can be recorded as (x j-ref ,y j-ref ). Where x j-ref Indicates the x coordinate of the jth reference point, y j-ref Represents the y-coordinate of the j-th reference point, j = 1, 2, ..., n.

[0126] In the embodiments of the present disclosure, unless otherwise specified, coordinates refer to coordinates in the world coordinate system.

[0127] Step S802: constructing an objective function and constraint conditions based on multiple reference trajectory points.

[0128] The objective function can be a cost objective function constructed based on smoothing cost, trajectory length cost, and translation cost. For example, the objective function can be denoted as cost, the smoothing cost as cost1, the trajectory length cost as cost2, and the translation cost as cost3. Then, cost = cost1 + cost2 + cost3, and the optimization goal corresponding to the objective function can be minimizing the cost.

[0129] Constraints may include multi-dimensional conditions, such as those used to limit the vehicle's travel range, the smoothness of the parking trajectory, and parking safety. For example, at least one of a curvature constraint, a target vehicle posture constraint, a collision constraint, and a position constraint may be determined as a constraint.

[0130] Step S803 : Optimize and fuse the parking trajectory based on the objective function and the constraint conditions to obtain a corrected parking trajectory.

[0131] The automatic parking device is based on the jth reference trajectory point (x j-ref ,y j-ref ), determine the minimum cost of the objective function under the constraints min . Set the objective function minimum cost min The corresponding trajectory point is determined as the target trajectory point (x, y) on the modified parking trajectory. The jth target trajectory point is the jth reference trajectory point (x j-ref ,y j-ref ) is corrected. In the above manner, n corrected trajectory points are obtained by correcting the n reference trajectory points on the fused parking trajectory l2, and the trajectory corresponding to the n corrected trajectory points is determined as the corrected parking trajectory.

[0132] In the disclosed embodiment, the constrained optimization objective function is constructed to obtain the corrected trajectory points on the corrected parking trajectory, thereby optimizing the rough path. The optimized trajectory can ensure that the vehicle is parked smoothly and accurately in the target parking space.

[0133] In some embodiments, based on the above embodiment, the step S802 in the above embodiment of “building an objective function based on multiple reference trajectory points” may include: Figure 9 The following steps are shown:

[0134] Step S901 : determining a smoothing cost and a trajectory length cost based on the distances between adjacent target trajectory points.

[0135] For example, the jth target trajectory point can be recorded as (x j ,y j), according to the distance between the j-th target trajectory point and its previous and next target trajectory points, determine the smoothing cost and trajectory length cost. For example, the smoothing cost is recorded as cost1, and the smoothing cost cost1 can be determined according to formula (3):

[0136]

[0137] In the embodiment of the present disclosure, the smoothness characteristic of the trajectory is quantified by the smoothness cost, and optimization is performed based on the smoothness cost function to generate a smooth trajectory that ensures good vehicle stability. The trajectory length cost is recorded as cost2, and the trajectory length cost cost2 can be determined according to formula (4):

[0138]

[0139] In the disclosed embodiment, the length between adjacent trajectory points is quantified by length cost, and optimization is performed based on the length cost function to generate a trajectory with the minimum total length. Shortening the trajectory length helps improve parking efficiency.

[0140] Step S902 : determining a translation cost based on the distance between the target trajectory point and a plurality of reference trajectory points.

[0141] For example, the translation cost can be determined based on the distance between the jth target trajectory point and its corresponding jth reference trajectory point. For example, the translation cost is recorded as cost3, and the translation cost cost3 can be determined according to formula (5):

[0142]

[0143] In the disclosed embodiment, the overall offset of the target trajectory point relative to the reference trajectory point is constrained by the translation cost, thereby ensuring that the corrected parking trajectory and the reference trajectory (fused parking trajectory) maintain geometric consistency.

[0144] Step S903: construct an objective function based on the smoothing cost, the trajectory length cost, and the translation cost.

[0145] For example, the automatic parking device can determine an objective function cost=cost1+cost2+cost3 based on the smoothing cost, trajectory length cost, and translation cost. The objective function cost can be minimized to obtain n target trajectory points corresponding to the minimum cost, thereby forming a corrected parking trajectory.

[0146] In the disclosed embodiments, a smoothing cost is used to quantify the smoothness of the trajectory, ensuring smoothness and stable parking of the vehicle. A length cost is used to quantify the length between adjacent trajectory points, minimizing the length of the generated revised parking trajectory and ensuring rapid parking. A translation cost is used to constrain the overall offset of the target trajectory points, ensuring consistency between the generated revised parking trajectory and the coarse trajectory. These three costs are combined for optimization to ensure that the generated revised parking trajectory meets smoothness and minimum length requirements while maintaining geometric consistency with the coarse trajectory.

[0147] In some embodiments, based on the above embodiment, the “building constraint conditions based on multiple reference trajectory points” in step S802 of the above embodiment may include: Figure 10 The following steps are shown:

[0148] Step S1001 : determining an average distance between adjacent reference trajectory points based on a plurality of reference trajectory points.

[0149] For example, the average distance between adjacent reference trajectory points is recorded as Δs, which can be calculated according to formula (6):

[0150]

[0151] Step S1002 : constructing a trajectory curvature constraint condition based on the distance between adjacent target trajectory points, the average distance between adjacent reference trajectory points, the relaxation factor of the collision constraint, and the maximum curvature constraint.

[0152] For example, the automatic parking device may consider correcting the curvature of the parking trajectory and construct a constraint condition based on the trajectory curvature, as shown in formula (7):

[0153]

[0154] In formula (7), w j is the relaxation factor of the collision constraint, △s is the average distance between adjacent reference trajectory points, κ max is the maximum curvature constraint.

[0155] Step S1003: constructing the target posture constraint conditions of the vehicle based on adjacent target trajectory points.

[0156] After parking is completed, the vehicle should theoretically be oriented parallel to the parking space. Based on this, the target posture constraint of the vehicle is constructed as shown in formula (8):

[0157]

[0158] In formula (8), θ n is the correction angle of the target parking space. It is the ratio of the vertical and horizontal coordinates of the adjacent target trajectory points at the end of the corrected parking trajectory.

[0159] Step S1004: Construct collision constraint conditions based on each reference trajectory point.

[0160] The most critical thing in trajectory planning is to avoid collision, so collision constraints need to be constructed.

[0161] In one implementation, the collision constraint can be based on Figure 11 Follow the steps shown to build:

[0162] Step S1101: Determine the collision detection point corresponding to each reference trajectory point.

[0163] In the embodiment of the present disclosure, at each correction track point on the correction parking track, set Figure 12 The six collision detection points are shown. These six collision detection points are points on the vehicle's corresponding envelope. The envelope is designed to avoid collisions. Each point on the corrected trajectory corresponds to a vehicle safety box, which is the envelope. The collision detection points on the envelope are denoted as points ABCDEF.

[0164] Step S1102 : determining a collision line corresponding to each reference trajectory point based on a collision detection point corresponding to each reference trajectory point.

[0165] Determine the occupancy grid map constructed from the environmental data perceived by the vehicle perception module, find the obstacles closest to the six collision detection points in the occupancy grid map, determine the perpendicular lines corresponding to the lines connecting the obstacles and the collision detection points, calculate the perpendicular line equations corresponding to these six points, and obtain the six collision lines corresponding to each reference trajectory point.

[0166] Step S1103 : Constructing collision constraint conditions based on the collision lines corresponding to the reference trajectory points.

[0167] Consider that the vehicle is located on the same side (i.e., inside) of the six vertical lines in reality, such as Figure 12 As shown, the collision constraint condition is constructed by the six points being on the current side of the vertical line, thereby forming a collision constraint. For example, the collision constraint condition is constructed as shown in formula (9):

[0168]

[0169] Among them, x aj +p a y aj 、x bj +p b y bj 、x cj +p c ycj 、x dj +p d y dj 、x ej +p e y ej and x fj +p f y fj are the perpendicular line equations of the six collision detection points ABCDEF corresponding to the j-th reference trajectory point. a 、p b 、p c 、p d 、p e 、p f and p, are the slopes of the lines.

[0170] In the disclosed embodiment, the constructed collision constraint is a linear constraint. The constraint function is constructed by considering that each reference point is located on the same side of the perpendicular line of the six collision detection points. This can ensure the convexity of the optimization problem and ensure that the local optimal solution is the global optimal solution. It can improve the optimization efficiency and quickly optimize the corrected parking trajectory.

[0171] Step S1005: constructing position constraints based on each reference trajectory point.

[0172] For example, the distance between each corrected trajectory point on the corrected parking trajectory and its corresponding reference trajectory point should be within a certain range. Based on this, a position constraint condition is constructed, as shown in formula (10):

[0173]

[0174] Among them, x l 、y l 、x u and y u The preset range value.

[0175] Step S1006 : determining a constraint condition based on at least one of a curvature constraint condition, a target posture constraint condition of the vehicle, a collision constraint condition, and a position constraint condition.

[0176] The objective function is optimized based on the at least one constraint condition to obtain a corrected parking trajectory.

[0177] In the disclosed embodiment, the objective function can be optimized based on various constraints, such as trajectory smoothness, posture rationality, collision safety, and reference trajectory accuracy, to achieve a unified parking trajectory of safety, efficiency, comfort, and stability, and realize automatic, precise, and rapid parking at the target parking space.

[0178] Exemplary devices

[0179] Figure 13 FIG. 1 is a schematic diagram of the structure of an automatic parking device provided by an exemplary embodiment of the present disclosure. Figure 13 As shown, the automatic parking device 1300 may include:

[0180] A first determining module 1301 is configured to determine an initial vehicle posture of the vehicle and an initial parking posture of a target parking space in response to an automatic parking instruction;

[0181] A second determining module 1302 is configured to determine an initial parking trajectory based on the initial vehicle posture and the initial parking space posture;

[0182] A third determining module 1303 is configured to determine a parking trajectory correction area based on the initial parking space posture;

[0183] A first control module 1304 is configured to control parking of the vehicle based on the initial parking trajectory;

[0184] a fourth determining module 1305, configured to determine a corrected parking position of the target parking space in response to the vehicle traveling into the parking trajectory correction area;

[0185] a fifth determining module 1306, configured to determine a revised parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory;

[0186] The second control module 1307 is configured to control the vehicle to park based on the corrected parking trajectory.

[0187] See also Figure 14 In some embodiments, the fifth determining module 1306 may include:

[0188] a determining unit 13061, configured to determine a first parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory;

[0189] a fusion processing unit 13062, configured to fuse the initial parking trajectory and the first parking trajectory to obtain a fused parking trajectory;

[0190] The optimization processing unit 13063 is configured to optimize the fused parking trajectory to obtain the corrected parking trajectory.

[0191] In some embodiments, the determining unit 13061 is further configured to:

[0192] Determining a translation vector based on the initial parking space pose and the revised parking space pose;

[0193] The initial parking trajectory is translated based on the translation vector to obtain the first parking trajectory.

[0194] In some embodiments, the fusion processing unit 13062 is further configured to:

[0195] determining a first vehicle position of the vehicle when the vehicle travels into the parking trajectory correction area;

[0196] Determining a first length of a first untraveled sub-trajectory of the initial parking trajectory based on the first vehicle position and the initial parking space posture;

[0197] determining a second length between each track point on the first sub-track and the first vehicle position;

[0198] Determining a fusion ratio corresponding to each of the trajectory points based on the first length and the second length;

[0199] Based on the fusion ratio, the initial parking trajectory and the first sub-trajectory are fused to obtain the fused parking trajectory.

[0200] In some embodiments, the optimization processing unit 13063 is further configured to:

[0201] determining a plurality of reference trajectory points based on the fused parking trajectory;

[0202] constructing an objective function and constraints based on the multiple reference trajectory points;

[0203] The fused parking trajectory is optimized based on the objective function and the constraint conditions to obtain the corrected parking trajectory.

[0204] In some embodiments, the optimization processing unit 13063 is further configured to:

[0205] Determine the smoothing cost and trajectory length cost based on the distance between adjacent target trajectory points;

[0206] determining a translation cost based on distances between the target trajectory point and the plurality of reference trajectory points;

[0207] The objective function is constructed based on the smoothing cost, the trajectory length cost, and the translation cost.

[0208] In some embodiments, the optimization processing unit 13063 is further configured to:

[0209] determining an average distance between adjacent reference trajectory points based on the plurality of reference trajectory points;

[0210] Constructing a trajectory curvature constraint condition based on the distance between adjacent target trajectory points, the average distance between adjacent reference trajectory points, a relaxation factor of the collision constraint, and a maximum curvature constraint;

[0211] Constructing target posture constraints of the vehicle based on adjacent target trajectory points;

[0212] Constructing collision constraint conditions based on each of the reference trajectory points;

[0213] Constructing position constraints based on each of the reference trajectory points;

[0214] The constraint condition is determined based on at least one constraint condition among the curvature constraint condition, the target posture constraint condition of the vehicle, the collision constraint condition, and the position constraint condition.

[0215] In some embodiments, the optimization processing unit 13063 is further configured to:

[0216] Determining a collision detection point corresponding to each of the reference trajectory points;

[0217] Determining a collision line corresponding to each reference trajectory point based on a collision detection point corresponding to each reference trajectory point;

[0218] The collision constraint condition is constructed based on the collision straight line corresponding to each reference trajectory point.

[0219] The beneficial technical effects corresponding to the exemplary embodiment of this device can be found in the corresponding beneficial technical effects of the above exemplary method part, which will not be repeated here.

[0220] Exemplary electronic devices

[0221] Figure 15 FIG. 1 is a schematic diagram of the structure of an electronic device provided by an exemplary embodiment of the present disclosure. Figure 15 As shown, the electronic device 1500 may include at least one processor 1501 and a memory 1502 .

[0222] The processor 1501 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 1500 to perform desired functions.

[0223] Memory 1502 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), a hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and processor 1501 may execute one or more computer program instructions to implement the automatic parking method and / or other desired functions of the various embodiments of the present disclosure described above.

[0224] In one example, the electronic device 1500 may further include an input device 1503 and an output device 1504 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0225] The input device 1503 may also include, for example, a keyboard, a mouse, and the like.

[0226] The output device 1504 can output various information to the outside, and may include, for example, a display, a speaker, a printer, a communication network and its connected remote output devices, etc.

[0227] Of course, to simplify, Figure 15 Only some of the components related to the present disclosure in the electronic device 1500 are shown, and components such as buses, input / output interfaces, etc. are omitted. In addition, according to specific application scenarios, the electronic device 1500 may further include any other appropriate components.

[0228] Exemplary computer program products and computer-readable storage media

[0229] In addition to the above-mentioned methods and devices, embodiments of the present disclosure may also provide a computer program product, including computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the automatic parking method of various embodiments of the present disclosure described in the above-mentioned "Exemplary Method" section.

[0230] The computer program product may be written in any combination of one or more programming languages to implement the operations of the disclosed embodiments, including object-oriented programming languages such as Java, C++, and conventional procedural programming languages such as C or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0231] In addition, an embodiment of the present disclosure may also be a computer-readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the processor executes the steps of the automatic parking method of various embodiments of the present disclosure described in the above “Exemplary Method” section.

[0232] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium is, for example, but not limited to, a system, device or component comprising electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0233] The basic principles of the present disclosure have been described above in conjunction with specific embodiments. However, the advantages, strengths, and effects mentioned in this disclosure are merely illustrative and not restrictive, and should not be considered as essential to each embodiment of the present disclosure. Furthermore, the specific details disclosed above are provided for illustrative purposes and to facilitate understanding, rather than as limitations. These details do not limit the present disclosure to necessarily being implemented using these specific details.

[0234] Those skilled in the art may make various changes and modifications to the present disclosure without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is intended to include these modifications and variations.

Claims

1. An automatic parking method, comprising: In response to an automatic parking instruction, determining an initial vehicle pose of the vehicle and an initial parking space pose of a target parking space; Determining an initial parking trajectory based on the initial vehicle posture and the initial parking space posture; Determining a parking trajectory correction area based on the initial parking space posture; Controlling parking of the vehicle based on the initial parking trajectory; In response to the vehicle traveling into the parking trajectory correction area, determining a corrected parking space posture of the target parking space; determining a revised parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory; The vehicle is controlled to park based on the corrected parking trajectory.

2. The method according to claim 1, wherein The determining of the revised parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory includes: determining a first parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory; fusing the initial parking trajectory and the first parking trajectory to obtain a fused parking trajectory; The fused parking trajectory is optimized to obtain the corrected parking trajectory.

3. The method according to claim 2, wherein: The determining a first parking trajectory based on the initial parking space posture, the corrected parking space posture, and the initial parking trajectory includes: Determining a translation vector based on the initial parking space pose and the revised parking space pose; The initial parking trajectory is translated based on the translation vector to obtain the first parking trajectory.

4. The method according to claim 2, wherein: The fusing the initial parking trajectory and the first parking trajectory to obtain a fused parking trajectory includes: determining a first vehicle position of the vehicle when the vehicle travels into the parking trajectory correction area; Determining a first length of a first untraveled sub-trajectory of the initial parking trajectory based on the first vehicle position and the initial parking space posture; determining a second length between each track point on the first sub-track and the first vehicle position; Determining a fusion ratio corresponding to each of the trajectory points based on the first length and the second length; Based on the fusion ratio, the initial parking trajectory and the first sub-trajectory are fused to obtain the fused parking trajectory.

5. The method according to claim 2, wherein: The optimizing the fused parking trajectory to obtain the corrected parking trajectory includes: determining a plurality of reference trajectory points based on the fused parking trajectory; constructing an objective function and constraints based on the multiple reference trajectory points; The fused parking trajectory is optimized based on the objective function and the constraint conditions to obtain the corrected parking trajectory.

6. The method according to claim 5, wherein: The constructing of an objective function based on the multiple reference trajectory points includes: Determine the smoothing cost and trajectory length cost based on the distance between adjacent target trajectory points; determining a translation cost based on distances between the target trajectory point and the plurality of reference trajectory points; The objective function is constructed based on the smoothing cost, the trajectory length cost, and the translation cost.

7. The method according to claim 5, wherein: Based on the multiple reference trajectory points, constrained conditions are constructed, including: determining an average distance between adjacent reference trajectory points based on the plurality of reference trajectory points; Constructing a trajectory curvature constraint condition based on the distance between adjacent target trajectory points, the average distance between adjacent reference trajectory points, a relaxation factor of the collision constraint, and a maximum curvature constraint; Constructing target posture constraints of the vehicle based on adjacent target trajectory points; Constructing collision constraint conditions based on each of the reference trajectory points; Constructing position constraints based on each of the reference trajectory points; The constraint condition is determined based on at least one constraint condition among the curvature constraint condition, the target posture constraint condition of the vehicle, the collision constraint condition, and the position constraint condition.

8. The method according to claim 7, wherein: The constructing of collision constraint conditions based on each of the reference trajectory points includes: Determining a collision detection point corresponding to each of the reference trajectory points; Determining a collision line corresponding to each reference trajectory point based on a collision detection point corresponding to each reference trajectory point; The collision constraint condition is constructed based on the collision straight line corresponding to each reference trajectory point.

9. An automatic parking device comprising: a first determining module, configured to determine an initial vehicle posture of the vehicle and an initial parking posture of a target parking space in response to an automatic parking instruction; a second determining module, configured to determine an initial parking trajectory based on the initial vehicle posture and the initial parking space posture; a third determining module, configured to determine a parking trajectory correction area based on the initial parking space posture; a first control module, configured to control parking of the vehicle based on the initial parking trajectory; a fourth determining module, configured to determine a corrected parking position of the target parking space in response to the vehicle traveling into the parking trajectory correction area; a fifth determining module, configured to determine a revised parking trajectory based on the initial parking space posture, the revised parking space posture, and the initial parking trajectory; A second control module is configured to control parking of the vehicle based on the corrected parking trajectory.

10. A computer-readable storage medium storing a computer program, wherein the computer program is used to execute the automatic parking method according to any one of claims 1 to 8.

11. An electronic device, comprising: processor; a memory for storing instructions executable by the processor; The processor is configured to read the instructions from the memory and execute the instructions to implement the automatic parking method described in any one of claims 1-8.

Images