SLAM-based enhanced parking assistance using reusable training results for nearly identical parking spaces

Abstract

A SLAM-based parking assist method includes determining that a host vehicle is entering a corridor adjacent a target parking space and determining potential reusability of at least one of a plurality of previously determined training results of a training process of the target parking space. The training result comprises training identification data of the potential parking space. The method further includes receiving identification data of the target parking space, identifying which of the plurality of training results is reusable based on the identification data of the target parking space to select an identified training result, comparing the identification data of the target parking space to identification data of potential parking spaces of the identified training result, and the identification data of the target parking space is loaded to parking assistance based on SLAM.

Description

Technical Field

[0001] The present disclosure relates to an enhanced parking assistance system and method based on simultaneous localization and mapping (SLAM) using reusable training results for nearly identical parking spaces. Background Art

[0002] This section generally presents the background of the present disclosure. The work of the presently designated inventors, to the extent it is described in this section and to the extent it may not qualify as prior art at the time of filing, is neither expressly nor impliedly admitted as prior art to the present disclosure.

[0003] SLAM-based parking assistance can be used to guide a vehicle into a parking space, and such assistance can be provided in the form of guidance information for the driver and / or enhancement of the automatic parking function. However, in some cases, whether the parking process is performed by the driver or the system, a lot of training is required to guide the vehicle into the parking space using SLAM-based parking assistance. Therefore, it is desirable to minimize the training overhead of SLAM-based parking assistance. Summary of the invention

[0004] The present disclosure describes a SLAM-based enhanced parking assistance method that uses reusable training results for nearly identical parking spaces to minimize training overhead. The method includes determining that a host vehicle is entering a corridor adjacent to a target parking space. The method also includes determining the potential reusability of multiple training results of a training process for a target parking space in response to determining that the host vehicle enters a corridor adjacent to the target parking space. Each of the multiple training results includes training identification data for a potential parking space. The method also includes receiving identification data of the target parking space in response to determining the potential reusability of multiple training results of a training process for the target parking space. The method also includes identifying which of the multiple training results is reusable based on the identification data of the target parking space to select the identified training result. The method also includes comparing the identification data of the target parking space with the identification data of the potential parking space of the identified training result to determine whether the difference between the identification data of the target parking space and the identification data of the potential parking space of the identified training result is within a preset threshold. The method also includes loading the identification data of the target parking space into the SLAM-based parking assistance in response to determining that the difference between the identification data of the target parking space and the identification data of the potential parking space identified by the training results is within a preset threshold. The method described in this paragraph reuses the training results from a previous training session to minimize the computing resources used by the SLAM-based parking assistance, rather than performing a training session for each visited parking space. Therefore, the method described in this paragraph minimizes the training overhead of the SLAM-based parking assistance of the vehicle by leveraging the reusability of the training results of other parking facilities with nearly identical parking spaces, thereby improving the vehicle technology.

[0005] In some aspects of the present disclosure, the method also includes providing visual guidance to a vehicle occupant to complete parking using SLAM-based parking assistance and identification data of a target parking space. The method may include controlling the movement of the host vehicle to reach a target parking position in the target parking space. The SLAM-based parking assistance uses a SLAM feature set to locate and map the host vehicle. The method may also include detecting a new variant of the SLAM feature data set and adjusting the identified training results using the new variant of the SLAM feature data set. The identification data of the target parking space includes geometric attributes of the target parking space, and the identification data of the potential parking space of the identified training results includes geometric attributes of the potential parking space of the identified training results. The geometric attributes of the target parking space include a corridor width of a corridor adjacent to the target parking space, a chassis ramp height of a ramp interconnecting the corridor and the target parking space, a parking space interior width of the target parking space, and a parking space exterior width of the target parking space. The method may also include determining whether the geometric attributes of the target parking and the geometric attributes of the potential parking space of the identified training structure are within a predetermined geometric threshold.

[0006] The present disclosure also describes an enhanced parking assistance system based on SLAM that uses reusable training results for almost identical parking spaces. The system includes multiple sensors inside the vehicle. The system also includes a controller that communicates with the sensors. The controller is programmed to perform the above method.

[0007] The present disclosure also describes a tangible, non-transitory machine-readable medium comprising machine-readable instructions that, when executed by a processor, cause the processor to perform the above method.

[0008] Further areas of applicability will become apparent from the description provided herein.It should be understood that the description and specific examples are intended for purposes of illustration only and are not intended to limit the scope of the present disclosure.

[0009] The above features and advantages and other features and advantages of the presently disclosed systems and methods are apparent from the detailed description, including the claims and exemplary embodiments, when taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The present disclosure will be more fully understood from the detailed description and accompanying drawings, in which:

[0011] Figure 1 is a schematic diagram of a vehicle including a SLAM-based enhanced parking assistance system using reusable training results for nearly identical parking spaces.

[0012] Figure 2 yes Figure 1 Schematic diagram of the vehicle shown in a parking facility.

[0013] Figure 3 A flow chart of a method for training a SLAM-based enhanced parking assistance system using reusable training results for nearly identical parking spaces.

[0014] Figure 4 Flowchart of a SLAM-based enhanced parking assistance method using reusable training results of training sessions for nearly the same parking spaces during real-time sessions.

[0015] Figure 5 is a flow chart of a method for determining whether a target parking space is nearly identical to a potential parking space of a training result of a training session. DETAILED DESCRIPTION

[0016] Reference will now be made in detail to several examples of the present disclosure illustrated in the accompanying drawings. Whenever possible, the same or similar reference numerals are used in the drawings and description to refer to the same or similar components or steps.

[0017] refer to Figure 1, the host vehicle 10 generally includes a body 12 and a plurality of wheels 14 coupled to the body 12. The host vehicle 10 may be an autonomous vehicle. In the illustrated embodiment, the host vehicle 10 is depicted as a sedan in the illustrated embodiment, but it should be understood that other vehicles including trucks, coupes, sport utility vehicles (SUVs), recreational vehicles (RVs), etc. may also be used.

[0018] The system 20 may be part of or work with the host vehicle 10. The system 20 may be referred to as an enhanced SLAM-based parking assistance system that uses reusable training results for nearly identical parking spaces, and may include a controller 34. The controller 34 includes at least one processor 44 and a non-transitory computer-readable storage device or medium 46. The processor 44 may be a custom or commercially available processor, a central processing unit (CPU), a graphics processing unit (GPU), an auxiliary processor in a plurality of processors associated with the vehicle controller 34, a semiconductor-based microprocessor (in the form of a microchip or chipset), a macroprocessor, a combination thereof, or generally a device for executing instructions. The computer-readable storage device or medium 46 may include, for example, volatile and non-volatile storage devices in a read-only memory (ROM), a random access memory (RAM), and a keep-alive memory (KAM). KAM is a persistent or non-volatile memory that can be used to store various operating variables when the processor 44 is powered off. The computer readable storage device or medium 46 may be implemented using a variety of storage devices, such as a programmable read only memory (PROM), an electrical PROM (EPROM), an electrically erasable PROM (EEPROM), flash memory, or other electrical, magnetic, optical, or combination storage device capable of storing data, some of which represents executable instructions, for use by the controller 34 in controlling the vehicle 10. The controller 34 of the vehicle 10 may be referred to as a vehicle controller and may be programmed to perform the methods 100, 200, and 300 described in detail below. Figure 3-5 ).

[0019] The instructions may include one or more separate programs, each of which includes an ordered list of executable instructions for implementing logical functions. When executed by the processor 44, the instructions receive and process signals from sensors, perform logic, calculations, methods and / or algorithms for automatically controlling components of the vehicle 10, and generate control signals to automatically control components of the vehicle 10 based on the logic, calculations, methods and / or algorithms. Although Figure 1 A single controller 34 is shown, but embodiments of the host vehicle 10 may include multiple controllers 34 that communicate via a suitable communication medium or combination of communication mediums and cooperate to process sensor signals, execute logic, calculations, methods and / or algorithms, and generate control signals to automatically control functions of the vehicle 10.

[0020] The host vehicle 10 also includes one or more sensors 16 coupled to the body 12. The sensors 16 sense observable conditions of the external environment and / or the internal environment of the host vehicle 10. As non-limiting examples, the sensors 16 may include one or more cameras, one or more light detection and ranging (LIDAR) sensors, one or more proximity sensors, one or more cameras, one or more ultrasonic sensors, one or more thermal imaging sensors, and / or other sensors. Each sensor 16 is configured to generate a signal representing a sensed observable condition (i.e., sensor data) of the external environment and / or the internal environment of the vehicle 10.

[0021] The vehicle 10 includes a user interface 23, which may be a touch screen in the dashboard. The user interface 23 may include, but is not limited to, an alarm, such as one or more speakers 27 that provide audible sounds in a vehicle seat or other object, tactile feedback, one or more displays 29, one or more microphones (e.g., a microphone array), and / or other devices suitable for notifying a vehicle user of the host vehicle 10. The user interface 23 is in electronic communication with the controller 34 and is configured to receive input from a vehicle occupant (e.g., a vehicle operator or a vehicle passenger). For example, the user interface 23 may include a touch screen and / or buttons configured to receive input from a person. Therefore, the controller 34 is configured to receive input from a user via the user interface 23. The user interface 23 is also configured to output messages or notifications via the display 29 and / or the speaker 27. For example, the user interface 23 may provide visual guidance to the vehicle occupant using SLAM-based parking assistance and identification data of the target parking space to complete parking.

[0022] refer to Figure 1 and Figure 2 , the vehicle 10 may include a SLAM-based parking assistance system 20. The system 20 may be activated when the host vehicle 10 enters a parking facility 50. The system 20 minimizes the training overhead of SLAM-based parking assistance by leveraging the reusability of training results from other parking facilities with nearly identical parking spaces. The parking facility 50 has a plurality of parking spaces 52 and a corridor 54 leading to one or more of the parking spaces 52. A ramp 58 interconnects the corridor 54 with a parking space chassis 64 of the parking spaces 52. One of the parking spaces 52 is considered a target parking space. The “target parking space” is a parking space 52 in the parking facility 50 where a vehicle occupant of the host vehicle 10 selects to park. Therefore, the target parking space is adjacent to the corridor 54 of the parking facility 50.

[0023] System 20 can be used for parking facilities with mechanical parking systems or other parking spaces with hard perimeters. In the present disclosure, the term "mechanical parking system" refers to a parking system that uses a mechanical device to move and store cars in a limited space. Mechanical parking systems can be divided into two categories: semi-automatic and fully automatic. Semi-automatic parking systems require human intervention to drive or guide the car into the machine, while fully automatic parking systems do not require any human assistance and can transport the car from the entrance to the parking space by itself. Mechanical parking systems are designed to save space, reduce emissions, improve safety, and improve driver convenience. Some examples of mechanical parking systems are turntable parking systems, lifting and traversing parking systems, rack and track parking systems, and tower parking systems. The turntable parking system includes a circular structure that can rotate the car vertically and horizontally. The lifting and traversing parking system is a modular system that can move the car up, down, left or right to create or enter a parking space. The rack and track parking system is a fully automatic system that uses a lift and a cart to transfer the car vertically and horizontally to a designated space. The tower parking system is a fully automated system that stacks cars vertically in a tower-like structure.

[0024] The system 20 uses a dataset of parking spaces with hard perimeters as a typical use case. During operation, the system 20 obtains recognition data of parking spaces 52 in training and real-time sessions of SLAM-based parking assistance. Therefore, the system 20 obtains target recognition data of the target parking space 52 in real time. The recognition data of the target parking space 52 may include geometric attributes of the parking facility 50, such as a corridor width 56 of a corridor 54 adjacent to the target parking space 52, a chassis ramp height of a ramp 58 interconnecting the corridor 54 and the target parking space, a parking space interior width 62 of the target parking space 52, and a parking space exterior width 60 of the target parking space 52. The reusability of the training results of the training session is determined based on the difference between the real-time target recognition data of the target parking space 52 and the training recognition data of the parking space of the previous training session. In order to facilitate the reuse of the training recognition data, the training results are customized according to the SLAM feature points selected to be related to the parking space perimeter that is hardly affected by other parked vehicles 11. Therefore, instead of performing a training session for each visited parking space 52 , the system 20 reuses the training results of previous training sessions to minimize the computational resources used by the SLAM-based parking assistance.

[0025] Figure 31 is a flow chart of a method 100 for training a SLAM-based enhanced parking assistance system using reusable training results for nearly identical parking spaces. The method 100 may be referred to as a training process and begins at box 102. At box 102, the controller 34 uses sensor data obtained from one or more sensors 16 (e.g., cameras) and / or through crowdsourcing to determine that the host vehicle 10 is entering a corridor 54 adjacent to a target parking space 52. Also, at box 102, when the host vehicle 10 enters the corridor 54 adjacent to the target parking space 52, the controller 34 determines the potential reusability of the training results that are about to occur for the current training session. As a non-limiting example, the primary criterion for determining the potential reusability of the training results for SLAM-based parking assistance is whether all parking spaces 52 in the parking facility 50 are nearly identical because the parking spaces 52 are part of a mechanical parking system or otherwise have hard perimeters. The method 100 then continues to box 104.

[0026] At block 104, the controller 34 obtains primary identification data of the target parking space 52 using the sensor 16 (e.g., a camera) and / or through crowdsourcing. The primary identification data of the target parking space may be defined by the relevant geometric attributes of the target parking space 52. As non-limiting examples, the relevant geometric attributes of the target parking space 52 may include a corridor width 56 of a corridor 54 adjacent to the target parking space 52, a chassis ramp height of a ramp 58 interconnecting the corridor 54 and the target parking space, a parking space interior width 62 of the target parking space 52, a parking space exterior width 60 of the target parking space 52, and an offset position of the target parking space 52 in the corridor direction between the nearest opposite parking spaces 52 across the corridor 54. If a camera is used to obtain the primary identification data of the target parking space 52, when the host vehicle 10 is within a predetermined distance (e.g., less than 7.5 meters and greater than 3 meters) from the target parking space 52, the camera frame group is selected as a camera view frame with an appropriate spatial interval. Next, the method 100 proceeds to block 106.

[0027] At box 106, the vehicle occupant parks the host vehicle 10 at what is considered a near-perfect parking position in the target parking space 52. The near-perfect parking position may be a position where the host vehicle 10 is located in the center of the target parking space 52. It is contemplated that the vehicle occupant may command the host vehicle 10 to autonomously park at what is considered a near-perfect parking position in the target parking space 52. At this point, the camera (i.e., sensor 16) should capture images showing exterior features (e.g., other parked vehicles 11) near the target parking space 52 for subsequent steps. The method 100 then continues to box 108.

[0028] At box 108, the controller 34 suggests which exterior features (captured by the camera) are associated with the hard perimeter of the target parking space 52 that is not affected by other parked vehicles 11. The controller 34 may exclude some unqualified features. The method 100 then continues to box 110. At box 110, the controller 34 commands the vehicle occupant to confirm the feature set suggested by the controller 34 at box 108, for example, through the user interface 23. The features suggested and confirmed by the vehicle occupant will then be used in the real-time parking session.

[0029] Figure 4 2 is a flow chart of a method 200 for enhanced SLAM-based parking assistance that uses reusable training results for nearly identical parking spaces 52 during a real-time session. The method 200 begins at box 202. At box 202, the controller 34 uses sensor data obtained from one or more sensors 16 (e.g., cameras) and / or through crowdsourcing to determine that the host vehicle 10 is entering a corridor 54 adjacent to the target parking space 52. In addition, at box 202, when the host vehicle 10 enters the corridor 54 near the target parking space 52, the controller 34 determines the potential reusability of training results from a previous training session. As a non-limiting example, the primary criterion for determining the potential reusability of training results for SLAM-based parking assistance is whether all parking spaces 52 in the parking facility 50 are nearly identical because the parking spaces 52 are part of a mechanical parking system or otherwise have a hard perimeter. The method 200 then continues to box 204.

[0030] At block 204, the controller 34 obtains primary identification data of the target parking space 52 using the sensor 16 (e.g., a camera) and / or through crowdsourcing. The primary identification data of the target parking space may be defined by the relevant geometric attributes of the target parking space 52. As non-limiting examples, the relevant geometric attributes of the target parking space 52 may include a corridor width 56 of a corridor 54 adjacent to the target parking space 52, a chassis ramp height of a ramp 58 interconnecting the corridor 54 and the target parking space, a parking space interior width 62 of the target parking space 52, a parking space exterior width 60 of the target parking space 52, and an offset position of the target parking space 52 in the corridor direction between the nearest opposite parking spaces 52 across the corridor 54. If a camera is used to obtain the primary identification data of the target parking space 52, when the host vehicle 10 is within a predetermined distance (e.g., less than 7.5 meters and greater than 3 meters) from the target parking space 52, the camera frame group is selected as a camera view frame with appropriate spatial spacing. Next, the method 200 proceeds to block 206.

[0031] At block 206, controller 34 identifies which of the plurality of training results is reusable based on the identification data of the target parking space to select the identified training result. Controller 34 may identify the training result based on the geographic location of target parking space 52. Controller 34 then loads the primary identification data of the potential parking space of the identified training result. Method 200 then proceeds to block 208.

[0032] At box 208, the controller 34 compares the identification data of the target parking space with the identification data of the potential parking space identified as a result of the training to determine whether the difference between the identification data of the target parking space and the identification data of the potential parking space identified as a result of the training is within a preset threshold. If the difference between the identification data of the target parking space and the identification data of the potential parking space identified as a result of the training is not within the preset threshold, the method 200 proceeds to box 210. If the difference between the identification data of the target parking space and the identification data of the potential parking space identified as a result of the training is within the preset threshold, the method 200 proceeds to box 212. Figure 5 Block 208 is described in more detail in the associated description.

[0033] At box 210, the controller 34 issues a notification to the vehicle occupant, for example, via the user interface 23. The notification indicates that the training results of the previous training session do not match the current target parking space 52. Additionally, if the vehicle occupant authorizes such action, the controller 34 begins a training session as described above with respect to the method 100.

[0034] At box 212, the controller 34 loads the identification data of the target parking space into the SLAM-based parking aid. The controller 34 then commands the user interface 23 to provide visual guidance to the vehicle occupant to complete parking using the SLAM-based parking aid and the identification data of the target parking space. Alternatively, if the host vehicle 10 is an autonomous vehicle, the controller 34 controls the movement of the host vehicle 10 to reach the target parking position in the target parking space 52. Then, the method 200 continues to box 214.

[0035] At block 214, the controller 34 detects new variants of the SLAM feature dataset, if applicable. Next, the controller 34 adjusts the identified training results using the new variants of the SLAM feature dataset.

[0036] Figure 5is a flow chart of a method 300 for determining whether a target parking space 52 is substantially identical to a potential parking space of a training result of a training session. The method 300 details block 208 of the method 200 and begins at block 302. At block 302, the controller 34 selects a possible reusable training result (i.e., an identified training result) of a training session. Such a possible reusable training result has not yet been evaluated. The method 300 then continues to block 304.

[0037] At block 304, the controller 34 determines (e.g., calculates) a difference between the relevant geometric attributes of the target parking space 52 and the geometric attributes of the identified potential parking space of the training result of the training session to determine whether the difference is within a predetermined geometric threshold. If the difference between the relevant geometric attributes of the target parking space 52 and the geometric attributes of the identified potential parking space of the training result is not within the predetermined geometric threshold, the method 300 continues to block 306. If the difference between the relevant geometric attributes of the target parking space 52 and the geometric attributes of the identified potential parking space of the training result is within the predetermined geometric threshold, the method 300 continues to block 308.

[0038] At block 306, the controller 34 checks whether there are any possible training results that have not yet been evaluated. If there are one or more training results that have not yet been evaluated, the method 300 returns to block 302. If all training results have been evaluated, the method 300 continues to block 310. The controller 34 outputs a negative output (i.e., no), and the method 200 described above continues to block 210.

[0039] At box 308, the controller 34 selects a key video frame of a potential parking space 52 associated with the selected possible reusable training result. The method 300 then continues to box 310. At box 312, the controller 34 determines (e.g., calculates) a correlation metric of the key video frames between the key video frames captured in real time and the key video frames of the training results of the training session to determine whether the image correlation is within a preset threshold. The image correlation metric can be, for example, phase correlation, high frequency component, hue and saturation spectrum correlation, and some combination of multiple correlation values, etc., which is appropriately designed to accommodate the relevant scene. If the image correlation metric is within the preset threshold, the method 300 proceeds to box 314. At box 314, the controller 34 outputs a positive output (i.e., yes) and the above method 200 proceeds to box 212. If the image correlation metric is not within the preset threshold, the method 300 proceeds to box 316.

[0040] At block 316, the controller 34 determines whether any key video frame pairs (i.e., real-time video frames and key frames of training results during the training session) have not been evaluated. If there are some key video frame pairs that have not been evaluated, the method 300 returns to block 308. If all key video frame pairs have been evaluated, the method 300 returns to block 306.

[0041] Although exemplary embodiments are described above, it is not intended that these embodiments describe all possible forms covered by the claims. The words used in the specification are descriptive rather than restrictive, and it should be understood that various changes can be made without departing from the spirit and scope of the present disclosure. As previously mentioned, the features of the various embodiments can be combined to form further embodiments of the currently disclosed system and method that may not be explicitly described or shown. Although various embodiments may have been described as having advantages or being superior to other embodiments or prior art implementations relative to one or more desired characteristics, it is recognized by those of ordinary skill in the art that one or more features or characteristics can be compromised to achieve the desired overall system properties, depending on the specific application and implementation. These properties may include, but are not limited to, cost, strength, durability, life cycle cost, marketability, appearance, packaging, size, applicability, weight, manufacturability, ease of assembly, etc. Therefore, with respect to embodiments of one or more features, embodiments described as less ideal than other embodiments or prior art implementations do not exceed the scope of the present disclosure and may be ideal for specific applications.

[0042] The drawings are in simplified form and are not drawn to exact scale. For convenience and clarity purposes only, directional terms such as top, bottom, left, right, up, above, above, below, below, rear, and front may be used with respect to the drawings. These directional terms and similar directional terms should not be construed to limit the scope of the present disclosure in any way.

[0043] Embodiments of the present disclosure are described herein. However, it should be understood that the disclosed embodiments are merely examples, and other embodiments may take various alternative forms. The drawings are not necessarily drawn to scale; certain features may be enlarged or minimized to show the details of a particular component. Therefore, the specific structural and functional details disclosed herein should not be interpreted as limiting, but merely as a representative basis for teaching those skilled in the art to adopt the currently disclosed systems and methods in different ways. As will be understood by those of ordinary skill in the art, the various features shown and described with reference to any of the accompanying drawings may be combined with the features shown in one or more other drawings to produce embodiments that are not explicitly shown or described. The combination of features shown provides representative embodiments of typical applications. However, various combinations and modifications of features consistent with the teachings of the present disclosure may be required for specific applications or implementations.

[0044] Embodiments of the present disclosure may be described herein in terms of functional and / or logical block components and various processing steps. It should be understood that such block components may be implemented by multiple hardware, software, and / or firmware components configured to perform specified functions. For example, embodiments of the present disclosure may employ various integrated circuit components, such as memory elements, digital signal processing elements, logic elements, lookup tables, etc., which may perform various functions under the control of one or more microprocessors or other control devices. In addition, those skilled in the art will appreciate that embodiments of the present disclosure may be practiced in conjunction with a variety of systems, and the systems described herein are merely exemplary embodiments of the present disclosure.

[0045] For the sake of brevity, technologies related to signal processing, data fusion, signaling, control and other functional aspects of the system (as well as the various operating components of the system) may not be described in detail herein. In addition, the connecting lines shown in the various figures contained herein are intended to represent example functional relationships and / or physical connections between the various elements. It should be noted that alternative or additional functional relationships or physical connections may exist in the embodiments of the present disclosure.

[0046] This description is merely illustrative in nature and is in no way intended to limit the present disclosure, its application or use. The broad teachings of the present disclosure can be implemented in many forms. Therefore, although the present disclosure includes specific examples, the true scope of the present disclosure should not be so limited, as other modifications will become apparent upon studying the drawings, the specification and the appended claims.

Claims

1. A parking assistance method based on SLAM, comprising: determining that the host vehicle is entering a corridor adjacent to the target parking space; In response to determining that the host vehicle enters the corridor adjacent to the target parking spot, determining potential reusability of a plurality of training results of a training process for the target parking spot, wherein each of the plurality of training results includes training identification data for a potential parking spot; In response to determining potential reusability of a plurality of training results of a training process for the target parking space, receiving identification data of the target parking space; identifying which training result of the plurality of training results is reusable based on the identification data of the target parking space to select the identified training result; comparing the identification data of the target parking space with the identification data of the identified potential parking space of the training result to determine whether a difference between the identification data of the target parking space and the identification data of the identified potential parking space of the training result is within a preset threshold; and In response to determining that the difference between the identification data of the target parking space and the identification data of the identified potential parking space of the training result is within a preset threshold, the identification data of the target parking space is loaded into the SLAM-based parking assistance. 2 . The method of claim 1 , further comprising providing visual guidance to a vehicle occupant to complete parking using the SLAM-based parking assistance and the identification data of the target parking space. 3 . The method of claim 1 , further comprising controlling movement of the host vehicle to reach a target parking position in the target parking space.

4. The method according to claim 1, wherein: The SLAM-based parking assistance uses a SLAM feature set to locate and map the host vehicle, and the method further includes: Detecting new variants of SLAM feature datasets; and The identified training results are adjusted using a new variant of the SLAM feature dataset.

5. The method according to claim 1, wherein: The recognition data of the target parking space includes geometric attributes of the target parking space, and the recognition data of the identified potential parking space of the training result includes geometric attributes of the identified potential parking space of the training result.

6. The method of claim 5, wherein the geometric attributes of the target parking space include a corridor width of a corridor adjacent to the target parking space, a chassis ramp height of a ramp interconnecting the corridor and the target parking space, a parking space interior width of the target parking space, and a parking space exterior width of the target parking space. 7 . The method of claim 5 , further comprising determining that a difference between a geometric attribute of the target parking space and a geometric attribute of the identified potential parking space is within a preset geometric threshold.

8. A parking assistance system based on SLAM, comprising: Multiple sensors; a controller in communication with the plurality of sensors, wherein the controller is programmed to: determining that the host vehicle is entering a corridor adjacent to the target parking space; In response to determining that the host vehicle enters the corridor adjacent to the target parking spot, determining potential reusability of a plurality of training results of a training process for the target parking spot, wherein each of the plurality of training results includes training identification data for a potential parking spot; In response to determining potential reusability of a plurality of training results of a training process for the target parking space, receiving identification data of the target parking space; identifying which training result of the plurality of training results is reusable based on the identification data of the target parking space to select the identified training result; comparing the identification data of the target parking space with the identification data of the identified potential parking space of the training result to determine whether a difference between the identification data of the target parking space and the identification data of the identified potential parking space of the training result is within a preset threshold; and In response to determining that the difference between the identification data of the target parking space and the identification data of the identified potential parking space of the training result is within a preset threshold, the identification data of the target parking space is loaded into the SLAM-based parking assistance. 9 . The system of claim 8 , wherein the controller is programmed to command a user interface to provide visual guidance to a vehicle occupant using the SLAM-based parking assistance and the identification data of the target parking space to complete parking.

10. The system according to claim 8, wherein: The controller is programmed to control movement of the host vehicle to reach a target parking position in the target parking space.

Images