Method and device for determining stop point of autonomous vehicle and storage medium

Abstract

The invention discloses a stop point determination method and device of an automatic driving vehicle and a storage medium. The method for determining the stop point of the autonomous vehicle comprises the following steps: determining an initial positioning point according to starting point address information input by a user, and controlling the autonomous vehicle to move towards the initial positioning point; within a target range determined based on the initial positioning point, identifying a target image signal by using a plurality of image sensors configured on the autonomous vehicle; the target image signal is an image signal which comes from the actual position of the user and contains target information of the user; and if the target image signal is identified, determining a corrected positioning point based on the target image signal, and determining a parking point of the autonomous vehicle based on the corrected positioning point. The user experience can be improved.

Description

Technical Field

[0001] This application relates to the field of autonomous driving, and in particular to a method, apparatus and storage medium for determining the stopping point of an autonomous vehicle. Background Technology

[0002] The development of autonomous driving technology has greatly enriched the operational models in the passenger transport sector. Taking robotaxi as an example, it is becoming a viable operational model. In this model, passengers can place an order through a mobile application (APP), which will then call an autonomous vehicle to pick them up at their designated location. However, in related technologies, the actual stopping location of the autonomous vehicle often deviates from the user's actual location. Since autonomous vehicles are often driverless, there may be no driver on board, making it difficult for a driver to manually correct this discrepancy. This results in inefficient passenger experiences and a poor user experience. Summary of the Invention

[0003] To address the aforementioned technical problems, this application provides a method for determining the stopping point of an autonomous vehicle, a device for determining the stopping point of an autonomous vehicle, an electronic device, and a computer-readable storage medium. The technical solution is as follows: According to a first aspect of this application, a method for determining the stopping point of an autonomous vehicle is provided, the method comprising: The system determines the initial location point based on the starting address information input by the user, and controls the autonomous vehicle to move towards the initial location point. Within the target area determined based on the initial positioning point, a number of image sensors configured on the autonomous vehicle are used to identify the target image signal; the target image signal is an image signal from the actual location of the user and contains the user's target information; If the target image signal is detected, a corrected positioning point is determined based on the target image signal, and a stopping point for the autonomous vehicle is determined based on the corrected positioning point.

[0004] According to a second aspect of this application, a stopping point determination device for an autonomous vehicle is provided, the device comprising: The control unit is used to determine the initial positioning point based on the starting address information input by the user, and to control the autonomous vehicle to move toward the initial positioning point; The identification unit is used to identify target image signals within a target area determined based on the initial positioning point using a plurality of image sensors configured on the autonomous vehicle; the target image signals are image signals from the actual location of the user and containing the user's target information; The determining unit is configured to, if a target image signal is detected, determine a corrected positioning point based on the target image signal, and determine the stopping point of the autonomous vehicle based on the corrected positioning point.

[0005] According to a third aspect of this application, an electronic device is provided, the electronic device comprising: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method as described in the first aspect.

[0006] According to a fourth aspect of this application, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method as described in the first aspect.

[0007] The technical solution provided in this application involves the following steps: When an autonomous vehicle moves towards a location point (initial location point) determined by the user's input starting address information, within the target area determined based on the initial location point, several image sensors configured on the autonomous vehicle identify the target image signal, which is an image signal from the user's actual location and contains the user's target information. If the target image signal is identified, the location point is corrected, i.e., a corrected location point is determined based on the target image signal, and the stopping point of the autonomous vehicle is determined based on the corrected location point. This allows for automatic correction of the location point to reduce the deviation between the actual stopping point of the autonomous vehicle and the actual location of the passenger when there is a deviation between the initial location point and the user's actual location, enabling passengers to ride the autonomous vehicle more efficiently and improving the user experience.

[0008] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0010] Figure 1 This is a schematic diagram of an autonomous vehicle parking scenario in related technologies; Figure 2 This is a flowchart illustrating a method for determining the stopping point of an autonomous vehicle according to an embodiment of this application; Figure 3This is a schematic diagram of an autonomous vehicle parking scenario according to an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a stopping point determination device for an autonomous vehicle according to an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0011] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art should fall within the scope of protection of this application.

[0012] The development of autonomous driving technology has greatly enriched the operational models in the passenger transport sector. Taking robotaxis as an example, they are becoming an achievable operational model. Figure 1 As shown, in this operating model, passengers can place an order through an application (APP) on their mobile phone. Based on the order, the mobile phone can call an autonomous vehicle to pick up the passenger at the location specified by the passenger.

[0013] However, in related technologies, autonomous vehicles strictly follow GPS-based navigation routes to fixed points to wait for passengers. The actual stopping point (the fixed point) often deviates from the user's actual location. Firstly, when a user places an order through a mobile app, they input the starting point's address (either via map selection or manual typing). This input address may not be accurate and may not perfectly match the user's actual location. Secondly, GPS positioning often has errors of several meters to tens of meters, so the indicated location may also deviate from the passenger's actual location. Both of these factors can lead to discrepancies between the autonomous vehicle's actual stopping point (e.g., ...). Figure 1 The "boarding point" shown often deviates significantly from the user's actual location. Since autonomous vehicles are often driverless, this deviation is difficult to correct manually, making it inefficient for passengers to board. Users often need to travel to the actual boarding point themselves, resulting in low operational efficiency and a poor user experience.

[0014] It is worth noting that the above description of autonomous vehicle parking scenarios in related technologies is only an illustrative example. In actual applications, other parking scenarios may exist, and no specific limitations are made for them.

[0015] To address the aforementioned issues, this application provides a method for determining the stopping point of an autonomous vehicle, enabling passengers to board autonomous vehicles more efficiently and improving the user experience. For example... Figure 2 As shown, the method includes the following steps: S201. Determine the initial positioning point based on the starting address information input by the user, and control the autonomous vehicle to move towards the initial positioning point.

[0016] S202. Within the target area determined based on the initial positioning point, the target image signal is identified using a number of image sensors configured on the autonomous vehicle.

[0017] The target image signal is an image signal originating from the actual location of the user and containing the user's target information.

[0018] S203. If the target image signal is detected, a corrected positioning point is determined based on the target image signal, and a stopping point for the autonomous vehicle is determined based on the corrected positioning point.

[0019] The technical solution provided in this application involves the following steps: When an autonomous vehicle moves toward a location point (initial location point) determined by the user's input starting address information, within the target range determined based on the initial location point, several sensors configured on the autonomous vehicle identify the interaction signal, i.e., the target interaction signal, from the user's actual location. If the target interaction signal is identified, the location point is corrected, i.e., a corrected location point is determined based on the target interaction signal, and the stopping point of the autonomous vehicle is determined based on the corrected location point. This allows for automatic correction of the location point to reduce the deviation between the actual stopping point of the autonomous vehicle and the actual location of the passenger when there is a deviation between the initial location point and the user's actual location, enabling passengers to ride the autonomous vehicle more efficiently and improving the user experience.

[0020] The aforementioned starting address information can be entered in various ways. For example, when a user places an order through an app on their device, they need to provide the starting and ending points of the order route, i.e., the starting and ending address information. Another example is that the starting address information can be provided by the user manually typing it in. Yet another example is that the starting address information can also be entered based on the user's current location determined by GPS after triggering the app's automatic location control. As yet another example, the aforementioned apps often provide navigation maps where users can select points; the starting address information can be entered by the user after selecting the target location on the map.

[0021] It is worth noting that the above description of the input method for starting address information is only an example. In actual applications, other input methods may exist, and no specific limitations are made for them.

[0022] As an example, a user places an order through an app on a terminal device. The autonomous vehicle can receive the order through the terminal device, obtain the starting address information based on the order information, determine the initial positioning point based on the starting address information, and then use the initial positioning point as the initial navigation target to make an initial movement towards the initial positioning point according to the navigation route.

[0023] Based on the target range determined by the initial positioning point, there can be various specific implementations. Correspondingly, the timing for triggering the image sensors configured on the autonomous vehicle to identify the target image signal can also be implemented in various ways. As an example, the target range can be determined centered on the initial positioning point, and the area of ​​the target range can be less than or equal to a preset area threshold. When the autonomous vehicle is detected entering the target range, the aforementioned sensors can be used to identify the target image signal. In this embodiment, the sensors configured on the autonomous vehicle are only triggered to actively identify the target image signal when the autonomous vehicle reaches a small area near the initial positioning point, which minimizes the consumption of computational resources.

[0024] Understandably, if the target area is defined centered on the initial positioning point and its area is less than or equal to a preset area threshold, then the target area can be a connecting area defined by the initial positioning point, where, based on experience, vehicles typically pick up passengers. As an example, the preset area threshold can be determined based on experience; that is, it can be the size of a connecting area determined according to general experience. For instance, if the target area is a circular region, it can be a circular region with a radius of 50-200 meters, or a radius of 200-300 meters, centered on the initial positioning point. The target area is not limited to a circular region; it can also be a rectangular region or other shapes, without specific limitations.

[0025] As another example, a specific implementation of the target range determined based on the aforementioned initial positioning point may include: the target range may include the position point before the autonomous vehicle begins to move. Correspondingly, another specific implementation of the timing for triggering the recognition of target image signals by several image sensors configured on the autonomous vehicle may include: when it is detected that the autonomous vehicle has started moving from the aforementioned position point before moving towards the aforementioned initial positioning point, the target image signals are recognized using the aforementioned image sensors.

[0026] In this embodiment, when the autonomous vehicle receives an order and begins to move toward the initial positioning point, it triggers several image sensors to identify the target image signal. The target image signal is continuously identified throughout the entire process of reaching the initial positioning point, which can maximize the accuracy, improve the recognition coverage, and avoid missed identification.

[0027] As another example, another specific implementation of the target range determined based on the aforementioned initial positioning point may include: the target range may be determined based on the initial positioning point and any location point in the navigation route of the autonomous vehicle to the initial positioning point. Correspondingly, another specific implementation of the timing for triggering the recognition of the target image signal by several image sensors configured on the autonomous vehicle may include: when the autonomous vehicle is detected to have arrived at any location point, the target image signal is recognized by the aforementioned several image sensors.

[0028] It is worth noting that the above description of the specific implementation of the target range and the specific implementation of the timing of the recognition of the target image signal by the corresponding image sensors is only an exemplary demonstration. In practical applications, other specific implementations may exist, and no specific limitation is made here.

[0029] As an example, after triggering several image sensors configured on an autonomous vehicle to identify target image signals, these image sensors can continue to identify, or stop identifying when the identification time is greater than or equal to a preset time threshold.

[0030] The aforementioned target information can be implemented in various ways. As an example, the target information may include a preset posture signal and / or target facial feature information. The preset posture is a preset body posture made by the user, and the target facial feature information is facial feature information that matches the facial feature information pre-entered by the user. If the target image signal is recognized, the user's own position can be determined based on the target image signal, and the above-mentioned corrected positioning point can be determined based on the user's own position.

[0031] In this embodiment, the image sensor identifies whether there is a user making a preset pose and / or whose facial feature information matches the user's pre-recorded facial feature information. The target image signal determines the position of the user making the preset pose (such as waving) and / or whose facial feature information matches the user's pre-recorded facial feature information. The positioning point is then corrected based on the user's position.

[0032] The aforementioned preset body posture can have various specific implementations. As an example, a preset body posture can refer to a preset gesture made by the user, such as waving or raising an arm. Preset body postures can also be other types of body postures that can emit a summoning signal, and there are no specific limitations on this.

[0033] There are several ways to determine a user's position based on a target image signal. As an example, the position of a user who has adopted a preset posture can be determined by visual positioning based on the preset posture signal contained in the target image signal.

[0034] The image sensor described above can be implemented in various ways. As an example, the image sensor can be a camera or other types of image sensors, without any specific limitation. As another example, the number of image sensors can be one or more. When there are multiple image sensors, the target image signal can be an image signal obtained by fusing image data collected by multiple image sensors respectively.

[0035] Considering that image sensors may identify multiple users and obtain the locations of multiple users when performing user location recognition (e.g., through gesture recognition), it is impossible to determine which user is the one that the autonomous vehicle needs to pick up.

[0036] Based on this, as an example, when the user's target information contained in the aforementioned target image signal includes both the aforementioned preset pose signal and the aforementioned target facial feature information, the preset pose signal can be identified first; if the preset pose signal is identified, then the identification of the target facial feature information is triggered. When the target information includes both the preset pose signal and the target facial feature information, visual identification can be used first to determine if a user is waving (preset pose), and facial recognition verification is only triggered if a user is waving, thus reducing the computational load.

[0037] As another example, the target facial feature information can be identified first; if the target facial feature information is identified, the recognition of the preset pose signal is triggered. Performing facial recognition verification first, and only triggering gesture recognition after the target face is identified, can also reduce the computational load.

[0038] As another example, the preset pose signal can also be identified, and the target facial feature information can be identified simultaneously. That is, the preset pose signal and the target facial feature information can be detected simultaneously without distinguishing their priorities.

[0039] As an example, when the target image signal contains user target information including the preset pose signal and the target facial feature information, if the preset pose signal and the target facial feature information are identified, and the user associated with the preset pose signal and the user corresponding to the target facial feature information are the same user, then the corrected positioning point is determined based on the position of the same user. In this embodiment, gesture recognition is combined with facial verification. When the users corresponding to the two are the same, the verification is passed, and the verified user is determined as a legitimate service object, and the positioning point is updated based on this object.

[0040] As another example, if the target image signal (i.e., an interactive image signal from the user's actual location containing the user's target information) is not identified, the initial positioning point is used as the stopping point for the autonomous vehicle. The situations where the target image signal is not identified include: not identifying either the preset gesture signal or the target facial feature information; or, although the preset gesture signal and the target facial feature information are identified, the user associated with the preset gesture signal and the user corresponding to the target facial feature information are different users. In this embodiment, gesture recognition is combined with facial verification. If the users corresponding to the two are inconsistent, verification fails. In this case, the positioning point is not corrected, and the vehicle directly stops at the initial positioning point.

[0041] The following is combined with Figure 3 The following is an exemplary description of a specific docking scenario in an embodiment of this application: like Figure 1 As shown, passengers (users) can place orders through an application (APP) on their mobile devices. Based on the order, the mobile phone can call an autonomous vehicle to pick up the passenger at the specified location. The autonomous vehicle first moves towards the initial positioning point in the general direction according to the navigation path. This initial positioning point is determined based on the starting address information entered by the passenger when placing the order. When the autonomous vehicle approaches the initial positioning point, it can actively trigger several image sensors configured on it to identify target image signals. If an image signal containing the user's target information from the user's actual location is identified, the positioning point is corrected based on the target image signal. The stopping point is then determined based on the corrected positioning point, and finally, the autonomous vehicle stops at the stopping point.

[0042] However, in related technologies, autonomous vehicles strictly follow GPS-based navigation routes to fixed points to wait for passengers. The actual stopping point (the fixed point) often deviates from the user's actual location. Firstly, when a user places an order through a mobile app, they input the starting point's address (either via map selection or manual typing). This input address may not be accurate and may not perfectly match the user's actual location. Secondly, GPS positioning often has errors of several meters to tens of meters, so the indicated location may also deviate from the passenger's actual location. Both of these factors can lead to discrepancies between the autonomous vehicle's actual stopping point (e.g., ...). Figure 1 The "boarding point" shown often deviates significantly from the user's actual location. Since there is no driver on the autonomous vehicle, this deviation is difficult to correct manually, making it inefficient for passengers to board autonomous vehicles. Users often need to travel to the actual stop themselves, resulting in low operational efficiency and a poor user experience.

[0043] Considering that in some scenarios, the user's current location may be a no-parking zone, or the user's current location may be an area where it is difficult to continue driving after the vehicle is parked, the default "car finds person" may not be reasonable in such cases.

[0044] Therefore, as an example, if the target image signal is identified, and the corrected location point determined based on the target image signal belongs to a no-stopping zone or an area where the autonomous vehicle would have difficulty continuing to drive after stopping, then when determining the stopping point of the autonomous vehicle based on the corrected location point, the target location point closest to the corrected location point can be used as the stopping point. This target location point does not belong to a no-stopping zone or an area where the autonomous vehicle would have difficulty continuing to drive after stopping. As another example, if the corrected location point belongs to a no-stopping zone or an area where the autonomous vehicle would have difficulty continuing to drive after stopping, the initial location point can also be directly determined as the stopping point of the autonomous vehicle.

[0045] As another example, if the final determined stop is neither the user's own location nor the location of the target terminal device, a prompt message can be sent to the user indicating the location of the stop, instructing the user to move to the final determined stop and board the vehicle. As another example, this prompt message can be sent to the user's terminal device for display, or it can be displayed on a display device configured on the autonomous vehicle. Considering that the target image signal may not necessarily be recognized, for example, if the user moves a long distance while the autonomous vehicle is driving towards the initial positioning point, and the user is still far away from the autonomous vehicle when the autonomous vehicle arrives near the initial positioning point, it may be possible that some image sensors of the autonomous vehicle will not be able to recognize the target image signal.

[0046] Based on this, as an example, if the target image signal is not identified, the initial positioning point can be used as the docking point, and the autonomous vehicle can be controlled to dock at the initial positioning point.

[0047] Considering that there may be multiple users hailing a ride near the autonomous vehicle's stop, these users may mistakenly identify the autonomous vehicle corresponding to their order, leading to a conflict in getting on the ride.

[0048] Based on this, as an example, after detecting that the autonomous vehicle has stopped at the designated stop, the user can be authenticated; if the authentication is successful, the vehicle's doors can be opened. For instance, after the vehicle stops, the identity of the user wanting to board can be verified through methods such as QR code scanning or facial recognition, and the doors will open automatically upon successful authentication.

[0049] As an example, the aforementioned autonomous vehicles could be robotaxi or autonomous ride-hailing vehicles.

[0050] The following is an exemplary description of a specific application scenario of this application: In some embodiments, where the target information includes a preset gesture signal and target facial feature information, the target facial feature information is facial feature information that matches the facial feature information pre-recorded by the user. When the autonomous vehicle recognizes a calling gesture (preset gesture signal) using several image sensors configured on it, the autonomous vehicle does not respond directly. Instead, it captures the facial features of the person waving (i.e., collects facial feature information) through a long-range biometric acquisition unit (i.e., an image sensor, such as a telephoto camera) and compares it with a pre-stored biometric template of the order user (i.e., the facial feature information pre-recorded by the user). Only when the comparison matches is the person waving confirmed as a legitimate service recipient. This precise binding of the general act of "waving" with the identity of a "specific user" prevents false responses and ensures the security and exclusivity of the service.

[0051] In some embodiments, during the gesture recognition process, the autonomous vehicle can enter an active search (call gesture) mode. The vehicle's onboard perception system (i.e., the aforementioned image sensor, such as a front-facing wide-angle camera) can continuously scan the surrounding environment and process the video stream in real time to identify image patterns that match a preset call gesture (such as an arm raised and waving). In other embodiments, when the autonomous vehicle recognizes a call gesture, an identity verification process is triggered: First, biometric data collection is performed: the autonomous vehicle uses its long-range biometric data collection unit (i.e., image sensors, such as two fixed telephoto cameras installed facing the left and right front of the vehicle, respectively) to capture directional images of the pedestrian making the waving gesture and obtain their facial feature information; then, feature comparison is performed: the autonomous vehicle compares the captured facial feature image with the target user biometric template (i.e., the user's pre-recorded facial feature information) stored in the order in real time and calculates the feature similarity; finally, consistency is determined: if the feature similarity is higher than a preset confidence threshold, the pedestrian is determined to be the target user to be picked up in this order. In other embodiments, several image sensors configured on the autonomous vehicle belong to the perception and positioning module (corresponding to the gesture recognition module and some long-distance biometric recognition modules): including a wide-angle camera (for large-scale gesture recognition) and a telephoto camera (for long-distance facial feature acquisition), which are responsible for collecting visual data of the vehicle's surrounding environment.

[0052] In some embodiments, the core of the aforementioned long-range biometric recognition module is an image processing unit that receives data from a telephoto camera and performs face detection, feature extraction, and comparison algorithms. In other embodiments, the long-range biometric acquisition unit may not use dual fixed telephoto cameras, but instead employs a single telephoto camera mounted on a two-dimensional servo gimbal. Once the wide-angle camera recognizes a gesture and determines its approximate location, it can drive the gimbal to rotate, allowing the telephoto camera to aim at the target for tracking and recognition. This reduces the number of cameras required.

[0053] In some embodiments, the aforementioned biometrics may not be limited to the face. When the lighting is poor or the user is facing away from the autonomous vehicle, the system may automatically switch to using body contour features or gait features for identification and comparison.

[0054] In some embodiments, the aforementioned “preset summoning gesture” needs to be predefined and model trained, and can be a single arm raised and waved, or other postures that are easy to recognize and not easily confused with daily actions.

[0055] In some embodiments, the aforementioned pre-stored "target user biometric template" can be created by the user uploading a photo through the APP when using the service for the first time, or accumulated and generated with the user's authorization in the historical itinerary.

[0056] Corresponding to the above method embodiments, this application also provides a device for determining the stopping point of an autonomous vehicle, such as... Figure 4 As shown, the device includes: The control unit 401 is used to determine the initial positioning point based on the starting address information input by the user, and control the autonomous vehicle to move towards the initial positioning point; The identification unit 402 is used to identify target image signals within a target range determined based on the initial positioning point using a plurality of image sensors configured on the autonomous vehicle; the target image signals are image signals from the actual location of the user and containing the user's target information. The determining unit 403 is used to determine a corrected positioning point based on the target image signal if a target image signal is identified, and to determine the stopping point of the autonomous vehicle based on the corrected positioning point.

[0057] As an example, the target range is determined with the initial positioning point as the center, and the area is less than or equal to a preset area threshold; the identification unit 402 is specifically used to identify the target image signal using the plurality of image sensors when the autonomous vehicle enters the target range.

[0058] As an example, the target range includes the position point before the autonomous vehicle starts moving; the identification unit 402 is specifically used to identify the target image signal using the plurality of image sensors when it detects that the autonomous vehicle starts moving from the position point to the initial positioning point.

[0059] As an example, the target information includes a preset posture signal and / or target facial feature information, wherein the preset posture is a preset body posture made by the user, and the target facial feature information is facial feature information that matches the facial feature information pre-recorded by the user. The determining unit 403 is specifically used to determine the user's own position based on the target image signal if the target image signal is identified, and to determine the corrected positioning point based on the user's own position.

[0060] As an example, when the target information includes the preset posture signal and the target facial feature information, the recognition unit 402 is specifically used to prioritize the recognition of the preset posture signal; if the preset posture signal is recognized, the recognition of the target facial feature information is triggered, or the target facial feature information is prioritized for recognition; if the target facial feature information is recognized, the recognition of the preset posture signal is triggered, or the preset posture signal is recognized and the target facial feature information is recognized simultaneously.

[0061] As an example, the determining unit 403 is specifically used to determine the corrected positioning point based on the position of the same user if the preset posture signal and the target face feature information are identified, and the user associated with the preset posture signal and the user corresponding to the target face feature information are the same user.

[0062] As an example, the determining unit 403 is further configured to use the initial positioning point as the stopping point of the autonomous vehicle if the target image signal is not identified; wherein, the failure to identify the target image signal includes: failure to identify one of the preset posture signal and the target facial feature information, or, identification of the preset posture signal and the target facial feature information, wherein the user associated with the preset posture signal and the user corresponding to the target facial feature information are different users.

[0063] As an example, the determining unit 403 is also used to detect that the autonomous vehicle has stopped at the parking point, and then authenticate the user; if the authentication is successful, the unit controls the door of the autonomous vehicle to open.

[0064] This application also provides an electronic device, such as Figure 5 As shown, the electronic device includes: Processor 501; Memory 502 is used to store processor-executable instructions; The processor 501 is configured to implement the method for determining the stopping point of an autonomous vehicle as described in any of the embodiments above.

[0065] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for determining the stopping point of an autonomous vehicle as described in any of the embodiments above.

[0066] The above description is only a specific embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for determining the stopping point of an autonomous vehicle, characterized in that, The method includes: The system determines the initial location point based on the starting address information input by the user, and controls the autonomous vehicle to move towards the initial location point. Within the target area determined based on the initial positioning point, a number of image sensors configured on the autonomous vehicle are used to identify the target image signal; the target image signal is an image signal from the actual location of the user and contains the user's target information; If the target image signal is detected, a corrected positioning point is determined based on the target image signal, and a stopping point for the autonomous vehicle is determined based on the corrected positioning point.

2. The method according to claim 1, characterized in that, The target range is determined centered on the initial positioning point, and its area is less than or equal to a preset area threshold; the step of identifying the target image signal using several image sensors configured on the autonomous vehicle includes: When the autonomous vehicle is detected to have entered the target area, the target image signal is identified using the plurality of image sensors.

3. The method according to claim 1, characterized in that, The target range includes the position point before the autonomous vehicle begins to move; the identification of the target image signal using a plurality of image sensors configured on the autonomous vehicle includes: When the autonomous vehicle is detected to be moving from the location point to the initial positioning point, the target image signal is identified using the plurality of image sensors.

4. The method according to claim 1, characterized in that, The target information includes a preset posture signal and / or target facial feature information, wherein the preset posture is a preset body posture made by the user, and the target facial feature information is facial feature information that matches the facial feature information pre-recorded by the user; The step of determining the corrected positioning point based on the target image signal includes: If the target image signal is detected, the user's own position is determined based on the target image signal, and the corrected positioning point is determined based on the user's own position.

5. The method according to claim 4, characterized in that, When the target information includes the preset posture signal and the target facial feature information, the step of using a plurality of image sensors configured on the autonomous vehicle to identify the target image signal includes: The preset attitude signal is identified first; If the preset posture signal is detected, the recognition of the target facial feature information is triggered, or Prioritize the identification of the target facial feature information; If the target facial feature information is detected, the recognition of the preset pose signal is triggered, or The preset posture signal is identified, and the target facial feature information is identified simultaneously.

6. The method according to claim 5, characterized in that, The step of determining the corrected positioning point based on the target image signal includes: If the preset posture signal and the target facial feature information are identified, and the user associated with the preset posture signal and the user corresponding to the target facial feature information are the same user, then the corrected positioning point is determined based on the position of the same user.

7. The method according to claim 6, characterized in that, The method further includes: If the target image signal is not detected, the initial positioning point will be used as the stopping point for the autonomous vehicle. The unidentified target image signal includes: If neither the preset pose signal nor the target facial feature information is detected, or The preset posture signal and the target facial feature information are identified, and the user associated with the preset posture signal and the user corresponding to the target facial feature information are different users.

8. The method according to claim 1, characterized in that, The method further includes: After detecting that the autonomous vehicle has stopped at the designated stop, the user is authenticated. If authentication is successful, the doors of the autonomous vehicle will be opened.

9. A device for determining the stopping point of an autonomous vehicle, characterized in that, The device includes: The control unit is used to determine the initial positioning point based on the starting address information input by the user, and to control the autonomous vehicle to move toward the initial positioning point; The identification unit is used to identify target image signals within a target area determined based on the initial positioning point using a plurality of image sensors configured on the autonomous vehicle; the target image signals are image signals from the actual location of the user and containing the user's target information; The determining unit is configured to, if a target image signal is detected, determine a corrected positioning point based on the target image signal, and determine the stopping point of the autonomous vehicle based on the corrected positioning point.

10. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method of any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images