Parking starting point positioning method and system
By obtaining and integrating multiple sensor data in the automatic parking system, building and updating maps, and real-time correction of the starting point of the parking trajectory, the problem of insufficient automatic parking positioning accuracy in underground parking lots is solved, and the system's adaptability and User experience has promoted the intelligent and automated development of automatic parking technology.
Patent Information
- Application Number
- CN202411378141.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-30
- Publication Date
- 2024-11-19
AI Technical Summary
In complex environments such as underground parking lots, it is difficult for the automatic parking system to accurately determine the starting point of the vehicle's outbound exit, resulting in insufficient positioning accuracy, poor stability and poor user experience, and the existing technology is difficult to adapt to environmental changes.
By obtaining the vehicle's environmental information and motion status information, building the current map, and using a variety of sensor data (such as cameras and ultrasonic sensors) for data fusion and feature extraction, correcting the starting point of the parking trajectory in real time, and dynamically updating the map to adapt to the environment change.
Improve the accuracy and reliability of parking starting point positioning, enhance the environmental adaptability of the system, simplify the user's parking and outbound processes, reduce positioning errors, improve user experience, and promote the intelligence of the automatic parking system. and automation processes.
Smart Images

Figure CN118977703A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of automatic parking, and in particular to a parking starting point positioning method and system. Background Art
[0002] With the rapid development of autonomous driving technology, automatic parking functions are popular among users. In some automatic parking scenarios, the accuracy of the starting point positioning is directly related to the user experience of the automatic parking function. If the starting point positioning is inaccurate, it may cause the vehicle to be unable to smoothly exit or enter the warehouse, or even cause safety problems such as collisions. For example, in the special environment of underground parking lots (basements), under the influence of various obstacles, dim light, unclear signs, and frequent flow of vehicles and personnel, how the automatic parking system can accurately and reliably determine the vehicle's exit starting position and improve the accuracy of the exit starting point positioning is an important problem that needs to be solved. Summary of the invention
[0003] In view of the defects in the prior art, the purpose of the present invention is to provide a parking starting point positioning method and system to improve the parking starting point positioning accuracy and reliability, thereby enhancing the user experience when using the automatic parking function.
[0004] According to a first aspect of the present invention, a method for locating a parking starting point is provided, the method comprising:
[0005] Obtain the vehicle's environmental information and motion status information at the current driving point, and determine the parking trajectory;
[0006] Build a current map of the parking lot;
[0007] Calculating the position information of the vehicle at the current driving point according to the environmental information, the motion state information and the current map;
[0008] The starting point position of the parking track is corrected according to the environmental information and the position information of the current driving point.
[0009] Preferably, the step of acquiring the environment information and motion state information of the vehicle at the current driving point includes:
[0010] Performing fusion processing and feature extraction processing on a variety of environmental data collected in real time by the vehicle's sensor device to obtain the environmental information, wherein the environmental information includes parking space information, drivable area information and obstacle information;
[0011] The motion state information includes the speed and position information of the vehicle.
[0012] Preferably, the multiple environmental data include image data collected by a camera and obstacle distance data collected by an ultrasonic sensor.
[0013] Preferably, the constructing a current map of the parking lot includes: constructing a current map of the parking lot according to the environmental information and the parking trajectory.
[0014] Preferably, the correction process for the starting point position of the parking trajectory according to the environmental information and the position information of the current driving point includes:
[0015] Determine the initial feature point of the starting point position according to the environmental information;
[0016] Obtaining a current calculated value of the starting point according to the initial feature point of the starting point and the position information of the current driving point;
[0017] When the difference between the current calculated value and the calculated value at the previous moment reaches or exceeds a preset threshold, the calculated value at the previous moment is corrected to obtain the corrected starting point position.
[0018] Preferably, when the starting point of the parking trajectory is located at a parking space, the initial feature points include line information, corner point information and a parking space ID of the parking space.
[0019] According to a second aspect of the present invention, a parking starting point positioning system is provided, the system comprising:
[0020] A map building module, used to build a current map of the parking lot;
[0021] A processing module, used to obtain environmental information and motion state information of the vehicle at the current driving point, and determine the parking trajectory; calculate the position information of the vehicle at the current driving point according to the environmental information, the motion state information and the current map;
[0022] The starting point positioning module is used to correct the starting point position of the parking trajectory according to the environmental information and the position information of the current driving point.
[0023] Preferably, the processing module includes a data processing module and a positioning calculation module;
[0024] The data processing module is used to obtain the environmental information and motion state information of the vehicle at the current driving point and determine the parking trajectory; the positioning calculation module is used to calculate the position information of the vehicle at the current driving point according to the environmental information, the motion state information and the current map; or,
[0025] The data processing module is used to obtain the environmental information of the vehicle at the current driving point; the positioning calculation module is used to obtain the motion state information of the vehicle at the current driving point and determine the parking trajectory; the position information of the vehicle at the current driving point is calculated based on the environmental information, the motion state information and the current map;
[0026] Preferably, the data processing module is used to perform fusion processing and feature extraction processing on a variety of environmental data collected in real time by the vehicle's sensor device to obtain the environmental information, which includes parking space information, drivable area information and obstacle information;
[0027] The motion state information includes the speed and position information of the vehicle.
[0028] Preferably, the map construction module is used to construct a current map of the parking lot according to the environmental information and the parking trajectory;
[0029] The starting point positioning module is used to determine the initial feature point of the starting point position according to the environmental information; obtain the current calculated value of the starting point position according to the initial feature point of the starting point position and the position information of the current driving point; when the difference between the current calculated value and the calculated value at the previous moment reaches or exceeds a preset threshold, correct the calculated value at the previous moment to obtain the corrected starting point position;
[0030] When the starting point of the parking trajectory is located at a parking space, the initial feature points include line information, corner point information and a parking space ID of the parking space.
[0031] Compared with the prior art, the beneficial effects of the present invention include:
[0032] 1. Improve the starting point positioning accuracy and reliability: It can accurately and reliably determine the starting point of the vehicle in the underground parking lot, thereby improving the positioning accuracy and reliability of the memory parking system.
[0033] 2. Improve user experience: Through precise positioning and starting point identification, the user's parking and exit processes are simplified, reducing the user's troubles in finding vehicles and exit starting points, and improving user experience.
[0034] 3. Enhance environmental adaptability: In the face of various environmental changes that may exist in underground parking lots (such as parking space adjustments, new obstacles, etc.), it can perceive and adapt to these changes in real time to ensure accurate positioning services in various environments.
[0035] 4. Realize intelligence and automation: Promote the intelligence and automation process of the memory parking system, reduce human intervention, and improve the efficiency and stability of system operation. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:
[0037] Figure 1A schematic diagram of a parking scenario provided in an embodiment of the present application;
[0038] Figure 2 A schematic flow chart of a parking starting point positioning method provided in an embodiment of the present application;
[0039] Figure 3 Another schematic flow chart of the parking starting point positioning method provided in an embodiment of the present application;
[0040] Figure 4 A schematic block diagram of a parking starting point positioning system provided in an embodiment of the present application;
[0041] Figure 5 Another schematic block diagram of the parking starting point positioning system provided in an embodiment of the present application;
[0042] Figure 6 A schematic block diagram of an automatic parking system provided in an embodiment of the present application. DETAILED DESCRIPTION
[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.
[0044] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.
[0045] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in the subsequent figures. In addition, all directional indications in this application (such as up, down, left, right, front, back, bottom...) are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the figures). If the specific posture changes, the directional indication will also change accordingly. Furthermore, the descriptions involving "first", "second", etc. in the application are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated.
[0046] As a special application in autonomous driving, memory parking has become one of the standard features of modern smart cars. Memory parking requires learning and memorizing parking scenarios. The driver first drives the vehicle to park in or out of the parking space. During this process, the memory parking system first uses the vehicle's sensors to learn an entry or exit route (i.e., parking trajectory), and then automatically controls the vehicle based on the memory to park the vehicle in the designated parking space or out of the parking space.
[0047] Take the scenario of a vehicle parking out of a parking space as an example. Figure 1 As shown, the numbers 1, 2, and 3 represent the parking space ID information of the three parking spaces respectively. The driver drives the vehicle out of parking space 2 along the direction of the learning route. During this process, the vehicle's memory parking system is used to learn a parking trajectory for leaving the warehouse, and the starting position of the parking trajectory is in parking space 2.
[0048] In some special parking environments, such as underground parking lots (also known as basements), memory parking systems face many challenges. Traditional positioning methods, such as GPS positioning, often do not work accurately in underground parking lots. Therefore, how to improve the accuracy of the starting point positioning is an important issue that needs to be solved.
[0049] The challenges of underground memory parking exit starting point positioning technology include the following aspects:
[0050] 1. Poor environmental adaptability: The underground parking environment is complex and changeable, with a large number of obstructions, reflective surfaces, and light changes, which makes the existing visual recognition technology and sensors easily interfered during positioning, resulting in reduced positioning accuracy. In addition, different parking lot layouts and vehicle parking conditions also bring challenges to positioning.
[0051] 2. Insufficient positioning accuracy: Current technologies often have positioning errors when memorizing the starting point of parking and exiting the garage, especially in complex underground parking environments. This error may cause the vehicle to be unable to accurately find the predetermined starting point when exiting the garage, thus affecting the efficiency and safety of parking.
[0052] 3. Stability issues: Due to the uncertainty of underground parking environments, existing technologies may fail to locate or become unstable when running for a long time or facing environmental changes. This will affect the user's parking experience and may cause trouble for parking lot management.
[0053] 4. Dependence on specific conditions: Existing memory parking exit starting point positioning technology often depends on specific conditions, such as fixed parking spaces, specific markers or light conditions, etc. This limits the scope of application of the technology and makes it impossible to achieve accurate positioning in some cases.
[0054] For parking lots with large environmental changes (such as parking layout adjustments, new obstacles, etc.), map matching may have errors; in addition, for areas with poor lighting conditions or unclear markers, feature point extraction and matching may also be affected. If the map is collected and saved first, and the map is not updated in real time, any changes in the environment will cause problems with map matching, resulting in positioning failure.
[0055] Specifically, the aspects involved in the starting point positioning of underground memory parking include:
[0056] 1. Environmental perception and recognition: The underground parking environment is complex, including various obstacles, signs, lights and other factors, which may affect the memory parking system's judgment of the vehicle's position. Therefore, how to effectively use sensors, cameras and other equipment to accurately perceive and recognize the vehicle's surrounding environment is the key to solving the problem.
[0057] 2. Memory and update of location information: The memory parking system needs to be able to memorize the location information of the vehicle at different time points and accurately call it when needed. However, since the underground parking environment may change (such as the movement of other vehicles, the reconstruction of the parking lot, etc.), the memory parking system needs to be able to update the location information in real time to ensure accuracy.
[0058] 3. Accuracy of the starting point of the exit: The accuracy of the starting point of the exit is directly related to the user experience of the memory parking system. If the positioning is inaccurate, it may cause the vehicle to fail to exit the warehouse smoothly, or even cause safety problems such as collision.
[0059] Figure 2 A schematic flow chart of a parking starting point positioning method provided in an embodiment of the present application is shown. As a possible implementation, the method can be executed in an intelligent driving vehicle. The method includes:
[0060] S210: Obtaining environmental information and motion status information of the vehicle at the current driving point, and determining a parking trajectory.
[0061] For example, the vehicle can be Figure 1 For the vehicle shown in FIG. 1 , when the vehicle is parked out of the parking space (out of the garage), the starting position of the parking trajectory is at parking space 2, and the end point of the parking trajectory is at the target position in the drivable area outside the parking space. Figure 1 The vehicle is traveling in the opposite direction. At this time, the end point of the parking trajectory is at the parking space, and the starting point of the parking trajectory is at the target position outside the parking space. The parking trajectory is the driving trajectory of the vehicle from the starting position to the target position. The current driving point of the vehicle starts from the starting position to the target position.
[0062] During the parking process, starting from the starting point of the parking trajectory, the environmental information and motion state information of the vehicle at each point on the parking trajectory (ie, the current driving point) are obtained in real time.
[0063] Smart driving vehicles are usually equipped with various sensor devices, including but not limited to: GPS, lidar, millimeter wave radar, camera, ultrasonic sensor, global positioning system, inertial measurement unit, wheel odometer, etc.
[0064] The vehicle's sensor device can be used to obtain environmental information around the vehicle, including parking space information, drivable area information, and obstacle information, such as dynamic obstacle information and static obstacle information in the vehicle's surrounding environment. Dynamic obstacles include moving vehicles, pedestrians, animals, etc., and static obstacles include stationary vehicles, steps, pits, pillars, etc., which are not specifically limited in the embodiments of this application.
[0065] As a possible implementation method, the data collected by high-definition cameras and ultrasonic sensors can be fused and processed to extract richer environmental information around the vehicle.
[0066] The vehicle's motion state information may include, but is not limited to, information such as the vehicle's speed and acceleration, as well as the vehicle's position information, which is used to characterize the vehicle's position and posture. The parking trajectory may be determined based on the vehicle's real-time position from the starting position to the target position.
[0067] S220: Construct a current map of the parking lot.
[0068] When a vehicle enters a parking lot, it can build a map of the parking spaces and roads inside the parking lot. For example, during the driving process of the vehicle, a map of the parking lot can be dynamically built as the current map based on the vehicle's real-time driving trajectory and the data collected by the sensor device. The current map can be saved for a long time or deleted when it is not needed to avoid wasting storage resources.
[0069] For example, during the driving process of the vehicle, map information is continuously updated, including newly discovered obstacles, changed lane lines, etc., and the current map of the parking lot is obtained by comparing and integrating it with the pre-stored static map or the static map downloaded from the vehicle network to improve the accuracy and completeness of the map.
[0070] S230: Calculate the position information of the vehicle at the current driving point according to the environmental information, the motion state information and the current map.
[0071] For example, based on the fused data, combined with known current map information and the vehicle's motion state information, a positioning algorithm can be used to calculate the precise position of the vehicle at the current driving point.
[0072] S240: Correcting the starting point position of the parking trajectory according to the environmental information and the position information of the current driving point.
[0073] During parking, when the vehicle reaches the current driving point at the current moment, the starting point position of the parking trajectory at the current moment is determined, and the starting point position determined at the current moment is compared and corrected with the starting point position determined at the previous moment to obtain a corrected starting point position.
[0074] As a possible implementation method, the current starting point of the parking trajectory is determined by combining the dynamic current map and the vehicle's location information at the current driving point, and the starting point is determined by comparing and correcting it with the historical starting point at the previous moment to reduce the positioning error. For example, if the positioning error exceeds a preset threshold, re-positioning is triggered or the user is prompted to perform manual correction to determine the starting point.
[0075] The embodiments of the present application make technical innovations based on the existing sensors of the vehicle, solve the problem of parking starting point positioning, improve positioning accuracy and reliability, enhance user experience, enhance the environmental adaptability of the automatic parking system, and promote the intelligent and automated development of the automatic parking system. The realization of these goals will help improve the efficiency and safety of parking lot management and provide users with a more convenient and efficient parking experience.
[0076] As a possible implementation method, after the starting point position of the parking trajectory is obtained in step S240, the parking starting point position and the end point target position are used as the starting point and end point of the path planning. The parking trajectory is planned in real time according to the starting point position, the end point target position and the dynamic obstacle information. According to the real-time planned parking trajectory, a control instruction is generated to drive the vehicle to perform the parking operation.
[0077] As a possible implementation method, step S210 includes: fusing and feature extracting a variety of environmental data collected in real time by the vehicle's sensor device to obtain environmental information, the environmental information including parking space information, drivable area information and obstacle information, and the motion status information including the vehicle's speed and posture information.
[0078] For example, the sensing device includes a camera, an ultrasonic sensor, and a wheel speed sensor, and the various environmental data include image data collected by the camera and obstacle distance data collected by the ultrasonic sensor. The image data of the underground parking lot is collected by a high-definition camera, including lane lines, parking signs, parking space lines, parking space IDs, ground locks, pillars, drivable areas, etc., which are not limited here. The ultrasonic sensor is used to detect the distance between the vehicle and surrounding obstacles and provide close-range environmental perception information.
[0079] Without the need for additional sensors, high-precision and reliable starting point positioning can be achieved by using only cameras, ultrasonic sensors and wheel speed sensors, which reduces costs and has good economy.
[0080] In other implementations, in addition to the camera, ultrasonic sensor and wheel speed sensor, other types of sensors may be added, or other sensor combinations may be used to collect environmental data and motion status information, without limitation.
[0081] Two low-cost sensors, cameras and ultrasonic sensors, are used to collect environmental data. After data denoising and data fusion technology, the data from different sensors are synchronized in time and aligned in space to obtain more comprehensive and accurate environmental information, reduce the impact of fuzzy and uncertain perception of the vehicle's surroundings, improve the accuracy of environmental perception, and ensure the accuracy, reliability and adaptability of parking starting point positioning.
[0082] An algorithm is used to extract and identify features of the fused data, and the recognition accuracy and stability of environmental features are improved by training and optimizing the model. This algorithm is, for example, a deep learning algorithm.
[0083] For example, semantic segmentation algorithms are used to process image data, classify and save various data, and time-stamp and spatially align the data. Finally, different types of features are extracted through calculation.
[0084] As a possible implementation manner, step S220 includes: constructing a current map of the parking lot according to the environmental information and the parking trajectory.
[0085] The embodiment of the present application proposes a dynamic map construction and update mechanism, which dynamically constructs and updates the map information of the underground parking lot according to the vehicle movement trajectory and real-time sensor data to adapt to environmental changes and improve positioning accuracy.
[0086] In step S230, based on more comprehensive and accurate environmental information, motion status information and the current map, the position information of the vehicle at the current driving point is calculated, and the position information of the current driving point is also more accurate.
[0087] As a possible implementation manner, step S240 specifically includes:
[0088] The initial feature points of the starting position are determined according to the environmental information. For example, when the starting position of the parking trajectory is located at a parking space, the initial feature points include line information, corner point information and parking space ID of the parking space.
[0089] According to the initial feature point of the starting point and the position information of the current travel point, the current calculated value of the starting point is obtained.
[0090] When the difference between the current calculated value and the calculated value at the previous moment reaches or exceeds the preset threshold, the calculated value at the previous moment is corrected to obtain a corrected starting point position.
[0091] For example, when the vehicle is at the current driving point, the coordinate value of the starting point in the current map (i.e., the current calculated value) is obtained. The coordinate value of the vehicle's driving point at the previous moment is the calculated value of the starting point at the previous moment. The difference between the current calculated value and the calculated value at the previous moment can be a distance difference and / or an angle difference, or other calculated values based on the coordinate value, without limitation. (Please confirm whether the inventor is correct)
[0092] The embodiment of the present application is based on a starting point positioning algorithm based on multi-sensor data fusion and dynamic maps, and achieves accurate and reliable starting point positioning by combining vehicle position information, map information and environmental characteristics.
[0093] The following Figure 1 Taking the vehicle exit shown in the figure as an example, the parking starting point positioning method of the embodiment of the present application is specifically described. The self-driving vehicle first uses the vehicle's sensors to learn an exit route. During the learning process, the starting point position of the underground parking lot memory parking is determined.
[0094] like Figure 1 As shown, when the vehicle is parked in parking space 2, the sensors on the vehicle body cannot detect the parking space line information and parking space ID information of the parking space. The vehicle is stationary, there is no real-time kinematic (RTK) measurement technology, and the camera cannot collect parking space information and parking space ID.
[0095] In an embodiment of the present application, the vehicle uses a camera to perceive parking space information during the process of leaving the warehouse, and continuously updates the information of the four corner points of the starting parking space and the parking space ID through calculation and processing and saves them in the starting point positioning module as feature points for starting point positioning.
[0096] Figure 3 A schematic flow chart of a parking starting point positioning method provided in an embodiment of the present application is shown, comprising the following steps:
[0097] S310 data acquisition: Use high-definition cameras to capture image information of underground parking lots, including lane lines, parking signs, parking space lines, parking space IDs, ground locks, pillars, drivable areas, etc. Use ultrasonic sensors to detect the distance between the vehicle and surrounding obstacles, and provide close-range environmental perception information.
[0098] S320 multi-source data fusion and feature extraction: synchronize the time and space of the data collected by the high-definition camera and ultrasonic sensor, and reduce the noise of the data. After fusing the collected environmental data, use the deep learning algorithm to extract richer environmental feature information around the vehicle.
[0099] S330 Dynamic Map Construction and Update: Based on the vehicle's motion trajectory and sensor data, a dynamic map of the underground parking lot is constructed in real time. During the vehicle's driving, map information is continuously updated, including newly discovered obstacles, changed lane lines, etc. By comparing and integrating with pre-stored static maps, the accuracy and completeness of the map are improved.
[0100] S340 Positioning Calculation: Based on the fused data, combined with known map information and the vehicle's motion status, the positioning algorithm is used to calculate the vehicle's current precise position. The positioning error is reduced by comparing and correcting the position information at the previous moment.
[0101] S350 Starting point positioning: Combine dynamic map and vehicle location information to determine the starting point of memory parking. Verify the accuracy of positioning by comparing with the user's input or preset starting point. If the positioning error exceeds the preset threshold, repositioning is triggered or the user is prompted to perform manual correction.
[0102] The embodiments of the present application can improve the accuracy and stability of the positioning of the starting point of underground memory parking, reduce positioning errors, and improve parking efficiency. It is suitable for application prospects in various underground parking lots and autonomous driving scenarios. The advantages of the embodiments of the present application include the following aspects:
[0103] 1. High positioning accuracy and stability: By integrating data from multiple sensors such as cameras and ultrasonic sensors, the embodiments of the present application can obtain more comprehensive and accurate environmental information, thereby improving the accuracy of starting point positioning. The application of deep learning algorithms further enhances the stability of feature recognition and positioning, and can maintain high positioning accuracy even in complex underground parking environments.
[0104] 2. Dynamic adaptability: The dynamic map construction and update mechanism proposed in the embodiment of the present application enables the system to perceive environmental changes in real time and update map information accordingly. This dynamic adaptability allows the starting point positioning to be not only based on static maps, but also adjusted according to real-time data, further improving the reliability of positioning.
[0105] 3. Intelligence and automation: Through deep learning and algorithm optimization, the embodiment of the present application realizes intelligent perception and understanding of the vehicle position and environment, thereby realizing automatic starting point positioning. This greatly reduces the burden of manual operation for users, improves parking efficiency, and brings users a more convenient and intelligent parking experience.
[0106] 4. Wide applicability: The embodiments of the present application are not only applicable to underground parking lots, but also to other similar closed or semi-closed environments, such as indoor parking lots, garages, etc. At the same time, due to the use of universal sensors, the embodiments of the present application also have strong scalability and customizability, and can be flexibly adjusted according to different scenarios and needs.
[0107] 5. Technological advancement and innovation: The embodiments of this application have significant technological advancement and innovation in sensor data fusion, deep learning algorithm application, dynamic map construction, etc., and provide new ideas and methods for the development of autonomous driving technology. This will help promote the continuous advancement and widespread application of autonomous driving technology.
[0108] The embodiment of the present application achieves high-precision and high-stability positioning of the starting point of underground memory parking by integrating multiple sensor data, applying deep learning algorithms and dynamic map construction technology. It has the advantages of intelligence, automation and wide applicability, and provides strong support for the development of autonomous driving technology.
[0109] Figure 4 A schematic block diagram of a parking starting point positioning system provided in an embodiment of the present application is shown. The parking starting point positioning system includes a map construction module 410, a processing module 420 and a starting point positioning module 430.
[0110] The map building module 410 is used to build a current map of the parking lot.
[0111] The processing module 420 is used to obtain the environmental information and motion state information of the vehicle at the current driving point and determine the parking trajectory; and calculate the position information of the vehicle at the current driving point based on the environmental information, motion state information and the current map.
[0112] The starting point positioning module 330 is used to correct the starting point position of the parking trajectory according to the environmental information and the position information of the current driving point.
[0113] Figure 5 Another schematic block diagram of the parking starting point positioning system provided in an embodiment of the present application is shown. The parking starting point positioning system includes a map construction module 510, a processing module 520 and a starting point positioning module 530. The processing module 520 includes a data processing module 521 and a positioning calculation module 522.
[0114] As a possible implementation method, the data processing module 521 is used to obtain the environmental information and motion status information of the vehicle at the current driving point and determine the parking trajectory; the positioning calculation module 522 is used to calculate the position information of the vehicle at the current driving point based on the environmental information, motion status information and the current map.
[0115] As another possible implementation method, the data processing module 521 is used to obtain the environmental information of the vehicle at the current driving point; the positioning calculation module 422 is used to obtain the motion state information of the vehicle at the current driving point and determine the parking trajectory; based on the environmental information, motion state information and the current map, the position information of the vehicle at the current driving point is calculated.
[0116] As another possible implementation, the data processing module 521 is used to perform fusion processing and feature extraction processing on the various environmental data collected by the vehicle's sensor device in real time to obtain environmental information, which includes parking space information, drivable area information, and obstacle information; the motion state information includes the speed and posture information of the vehicle. For example, the various environmental data include image data collected by a camera and obstacle distance data collected by an ultrasonic sensor.
[0117] As a possible implementation, the map construction module 510 is used to construct a current map of the parking lot according to the environmental information and the parking trajectory.
[0118] The starting point positioning module 530 is used to determine the initial feature points of the starting point position according to the environmental information; obtain the current calculated value of the starting point position according to the initial feature points of the starting point position and the position information of the current driving point; when the difference between the current calculated value and the calculated value at the previous moment reaches or exceeds the preset threshold, the calculated value at the previous moment is corrected to obtain the corrected starting point position.
[0119] As a possible implementation manner, when the starting point of the parking trajectory is located at a parking space, the initial feature points include line information, corner point information, and a parking space ID of the parking space.
[0120] Figure 6 A schematic block diagram of an automatic parking system provided in an embodiment of the present application, wherein the automatic parking system includes a parking starting point positioning system, a planning module 640, a control module 650 and a chassis control actuator 660, wherein the parking starting point positioning system includes a map construction module 610, a processing module 620 and a starting point positioning module 630, and the processing module 620 includes a data processing module 621 and a positioning calculation module 622.
[0121] After the parking starting point positioning system obtains the starting position of the parking trajectory, the automatic parking system plans the parking trajectory in real time according to the starting position, the end target position and the dynamic obstacle information, and generates control instructions based on the real-time planned parking trajectory to drive the vehicle to perform parking operations.
[0122] The data processing module 621 receives data collected by various sensors and performs preliminary processing, processes and analyzes the collected data using a deep learning algorithm, extracts key features, and provides data support for positioning calculations. The map construction module 610 constructs and updates a dynamic map based on the vehicle's motion trajectory and sensor data. The positioning calculation module 622 calculates the vehicle's location information using a positioning algorithm based on the results of the data processing module 621 and the map construction module 610. The starting point positioning module 630 calculates and saves the location information of the vehicle's starting point (i.e., the starting point location information).
[0123] The data processing module 621 receives and saves in real time information about obstacles around the vehicle body identified by cameras around the vehicle, including environmental information such as lane lines, parking signs, parking space lines, parking space IDs, ground locks, pillars, and drivable areas, as well as the distance and position information of obstacles around the vehicle body detected by ultrasonic sensors.
[0124] The data processing module 621 processes the image data using a semantic segmentation algorithm, classifies and saves various data, and simultaneously timestamps and spatially aligns the data. Finally, different types of environmental information are extracted through calculation and output to the positioning calculation module 622 and the starting point positioning module 630 respectively.
[0125] To achieve the above functions, the data processing module 621 may include a data denoising unit, a data storage unit, an image recognition unit, a semantic segmentation unit, a time synchronization unit and a spatial alignment unit, which perform data denoising, data storage, image recognition, semantic segmentation, time synchronization and spatial alignment processing respectively.
[0126] The positioning calculation module 622 obtains the vehicle's speed information and posture information (i.e., motion status information) through the wheel speed sensor on the vehicle side, and records the entire route trajectory of the warehouse. According to the environmental information, motion status information and the current map, the position information of the vehicle at the current driving point is calculated.
[0127] The map construction module 610 obtains environmental information and the entire route trajectory of the outbound vehicle from the positioning calculation module 622, constructs a current map of the parking lot, and then sends the current map to the positioning calculation module 622.
[0128] The starting point positioning module 630 corrects the starting point position of the entire route track out of the warehouse according to the environmental information and the position information of the current driving point, obtains the starting point position, and sends it to the positioning calculation module 622.
[0129] When parking automatically:
[0130] The positioning calculation module 622 can obtain the coordinate information of the starting point and the end point of the entire route track of the outbound shipment through calculation and coordinate conversion, which is used as a reference for automatic outbound shipment.
[0131] The planning module 640 plans an optimal reference parking trajectory according to the vehicle's motion state information and the starting and ending positions of the entire route trajectory of the vehicle leaving the warehouse, and sends it to the control module 650.
[0132] The control module 650 generates control instructions according to the reference parking trajectory planned in real time, and sends corresponding throttle, brake, steering and other control instructions to the chassis control actuator 660 to drive the vehicle to perform automatic parking.
[0133] In other implementations, the data processing module 621 receives and processes the data collected by the camera, ultrasonic sensor, and wheel speed sensor. Specifically, the data processing module 621 receives and saves in real time the obstacle information around the vehicle body identified by the cameras around the vehicle, including lane lines, parking signs, parking space lines, parking space IDs, ground locks, pillars, drivable areas and other environmental information, as well as the distance and position information of the obstacles around the vehicle body detected by the ultrasonic sensor. The vehicle's speed information and posture information (i.e., motion status information) are obtained through the wheel speed sensor on the vehicle side, and the entire route trajectory of the outbound route is recorded at the same time.
[0134] The data processing module 621 uses a semantic segmentation algorithm to process image data, classify and save various types of data, and simultaneously timestamp and spatially align the data. Finally, different types of environmental information are extracted through calculation and output to the positioning calculation module 622 and the starting point positioning module 630 respectively. The wheel speed sensor on the vehicle side obtains the vehicle's speed information and posture information (i.e., motion status information) as well as the entire route trajectory of the outbound shipment and outputs them to the positioning calculation module 622.
[0135] The positioning calculation module 622 calculates the position information of the vehicle at the current driving point based on the environmental information, the motion state information and the current map.
[0136] The map construction module 610 obtains environmental information and the entire route trajectory of the outbound vehicle from the positioning calculation module 622, constructs a current map of the parking lot, and then sends the current map to the positioning calculation module 622.
[0137] The starting point positioning module 630 corrects the starting point position of the entire route track out of the warehouse according to the environmental information and the position information of the current driving point, obtains the starting point position, and sends it to the positioning calculation module 622.
[0138] Once the starting point is determined, automatic parking can be performed as needed.
[0139] The above describes the specific embodiments of the present invention. Based on the above description, relevant personnel can make various changes and modifications without departing from the technical concept of the present invention.
Claims
1. A parking starting point positioning method, characterized in that: The method comprises: Obtain the vehicle's environmental information and motion status information at the current driving point, and determine the parking trajectory; Build a current map of the parking lot; Calculating the position information of the vehicle at the current driving point according to the environmental information, the motion state information and the current map; The starting point position of the parking track is corrected according to the environmental information and the position information of the current driving point.
2. The parking starting point positioning method according to claim 1, characterized in that: The acquisition of the vehicle's environment information and motion state information at the current driving point includes: Performing fusion processing and feature extraction processing on a variety of environmental data collected in real time by the vehicle's sensor device to obtain the environmental information, wherein the environmental information includes parking space information, drivable area information and obstacle information; The motion state information includes the speed and position information of the vehicle.
3. The parking starting point positioning method according to claim 2, characterized in that: The multiple environmental data include image data collected by a camera and obstacle distance data collected by an ultrasonic sensor.
4. The parking starting point positioning method according to claim 1, characterized in that: The constructing the current map of the parking lot includes: constructing the current map of the parking lot according to the environmental information and the parking trajectory.
5. The parking starting point positioning method according to claim 1, characterized in that: According to the environmental information and the position information of the current driving point, the correction process for the starting position of the parking trajectory includes: Determine the initial feature point of the starting point position according to the environmental information; Obtaining a current calculated value of the starting point according to the initial feature point of the starting point and the position information of the current driving point; When the difference between the current calculated value and the calculated value at the previous moment reaches or exceeds a preset threshold, the calculated value at the previous moment is corrected to obtain the corrected starting point position.
6. The parking starting point positioning method according to claim 5, characterized in that: When the starting point of the parking trajectory is located at a parking space, the initial feature points include line information, corner point information and a parking space ID of the parking space.
7. A parking starting point positioning system, characterized in that: The system comprises: A map building module, used to build a current map of the parking lot; A processing module, used to obtain environmental information and motion state information of the vehicle at the current driving point, and determine the parking trajectory; calculate the position information of the vehicle at the current driving point according to the environmental information, the motion state information and the current map; The starting point positioning module is used to correct the starting point position of the parking trajectory according to the environmental information and the position information of the current driving point.
8. The parking starting point positioning system according to claim 7, characterized in that: The processing module includes a data processing module and a positioning calculation module; The data processing module is used to obtain the environmental information and motion state information of the vehicle at the current driving point and determine the parking trajectory; the positioning calculation module is used to calculate the position information of the vehicle at the current driving point according to the environmental information, the motion state information and the current map; or, The data processing module is used to obtain environmental information of the vehicle at the current driving point; The positioning calculation module is used to obtain the motion state information of the vehicle at the current driving point and determine the parking trajectory; and calculate the position information of the vehicle at the current driving point based on the environmental information, the motion state information and the current map.
9. The parking starting point positioning system according to claim 8, characterized in that: The data processing module is used to perform fusion processing and feature extraction processing on a variety of environmental data collected in real time by the vehicle's sensor device to obtain the environmental information, which includes parking space information, drivable area information and obstacle information; The motion state information includes the speed and position information of the vehicle.
10. The parking starting point positioning system according to claim 7, characterized in that: The map construction module is used to construct a current map of the parking lot according to the environmental information and the parking trajectory; The starting point positioning module is used to determine the initial feature point of the starting point position according to the environmental information; According to the initial feature point of the starting point position and the position information of the current driving point, a current calculated value of the starting point position is obtained; when the difference between the current calculated value and the calculated value at the previous moment reaches or exceeds a preset threshold, the calculated value at the previous moment is corrected to obtain the corrected starting point position; When the starting point of the parking trajectory is located at a parking space, the initial feature points include line information, corner point information and a parking space ID of the parking space.
Citation Information
Patent Citations
Memory parking system, method, terminal and cloud server
CN111439257A
Memory parking method, device and equipment with reverse parking function
CN115056767A
Vehicle starting point position confirmation method and device based on memory parking system and computer system
CN116994220A
Parking method and device, electronic equipment and readable storage medium
CN118205546A
