Pick-up processing method and device for autonomous driving taxi

Abstract

The application discloses a method and device for picking up a passenger in an automatic driving taxi. The method comprises: analyzing positioning information; detecting and analyzing the feature information of pedestrians on the current road when it is determined that the passenger is within a predetermined range; identifying the current position information of the passenger when the analysis result indicates that a stopping behavior is detected and the duration of the stopping behavior exceeds a preset duration; determining the parking position of the vehicle according to the position information; controlling the vehicle to park according to the route corresponding to the parking position, so that the right rear door is aligned with the current position information; then, obtaining the unlocking information initiated by the passenger device; after the information is verified, triggering the right rear door to open, so that the passenger gets on the vehicle. The application solves the technical problem that the parking mode of the unmanned taxi is not flexible enough in the related art, the taxi can only be parked at a preset station and cannot be parked at a non-predetermined station, so that the passenger needs to walk to the designated station to get on the vehicle, thereby affecting the riding experience.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving vehicle control, in particular to a method and device for picking up a passenger in an automatic driving taxi. BACKGROUND

[0002] Traditional robotaxi operation modes usually rely on preset pickup points, and passengers need to book a vehicle in advance through a mobile application and wait at a designated location. This mode limits the flexibility of passengers to a certain extent, especially when passengers are not sure of the specific pickup location or the weather conditions are bad. In this case, passengers may need to walk to the designated pickup point in the rain and snow to wait for the arrival of the robotaxi, which not only reduces the passenger experience, but also may prolong the waiting time of the passengers due to the uncertainty of the outdoor environment.

[0003] Most of the robotaxis on the market currently adopt a fixed station pickup mode, that is, passengers need to input or select a pickup point in the application, and the robotaxi will go to the designated location to pick up the passenger according to the information. Although this mode simplifies vehicle scheduling to a certain extent, it also has many problems. For example, in a busy street, there may be pedestrians, obstacles, etc. between the passenger and the station, affecting the passenger's quick finding of the vehicle; in bad weather conditions, passengers often do not want to walk far and hope that the vehicle can approach actively; in the night or poor visibility, the identification difficulty between the passenger and the vehicle increases, which may cause inaccurate positioning of the passenger and affect the pickup experience.

[0004] In addition, the existing robotaxi technology mainly relies on the passenger's mobile positioning information for passenger identification, and realizes pickup by matching the passenger's location and the vehicle's location. However, this method has limitations in actual application, for example, when the mobile signal is unstable, the passenger's positioning information may be inaccurate; in a multi-passenger scenario, the vehicle has difficulty in determining which passenger is the passenger for this booking, which may cause pickup errors.

[0005] At present, there is no effective solution to the above problems. SUMMARY

[0006] The embodiments of the present application provide a method and device for picking up a passenger in an automatic driving taxi, to at least solve the technical problem that the stopping mode of the robotaxi is not flexible enough in the related art, and the robotaxi can only stop at a preset station and cannot stop at a non-predefined station, resulting in the passenger needing to walk to the designated station to pick up the vehicle, which affects the pickup experience.

[0007] According to an aspect of an embodiment of the present application, a method for pickup processing of an autonomous taxi is provided, comprising: after receiving positioning information from a passenger device, analyzing the positioning information to determine whether a passenger is located within a predetermined range around a preset starting point, wherein the preset starting point is a pickup starting point set by the passenger through the passenger device; when it is determined that the passenger is located within the predetermined range, detecting feature information of pedestrians on a current road through a sensor device of the autonomous taxi; analyzing the feature information to obtain an analysis result; when the analysis result indicates that a stopping behavior is detected and a duration of the stopping behavior exceeds a preset duration, identifying current position information of the passenger corresponding to the stopping behavior; determining a parking position of the autonomous taxi according to a longitudinal coordinate corresponding to the current position information; controlling the autonomous taxi to park according to a route corresponding to the parking position, so that a right rear door of the autonomous taxi is aligned with the current position information; after the right rear door is aligned with the current position information, obtaining unlocking information initiated by the passenger device; after the unlocking information is verified, triggering the right rear door of the autonomous taxi to open, so that the passenger gets on the autonomous taxi.

[0008] Optionally, determining that the passenger is located in the predetermined range comprises: analyzing the positioning information to obtain a current distance between the passenger device and the preset starting point, and determining that the passenger is located in the predetermined range when the current distance is less than a preset distance within a predetermined time window; receiving a first prompt message initiated by the passenger device indicating arrival at the preset starting point, and determining that the passenger is located in the predetermined range.

[0009] Optionally, the sensor device comprises a vision device and a laser radar device, and the feature information of the pedestrians on the current road is detected through the sensor device of the autonomous taxi, comprising: detecting a passenger stopping action in a preset area on the right side of the autonomous taxi through the vision device to obtain pedestrian images of the preset area on the right side, and extracting features of the pedestrians from the pedestrian images to obtain limb actions of the pedestrians, wherein the limb actions at least include arm and hand actions, head gaze directions; detecting point cloud data of the pedestrians through the laser radar device; and fusing the limb actions and the point cloud data to obtain the feature information.

[0010] Optionally, analyzing the feature information to obtain an analysis result comprises: analyzing the feature information to obtain actions of the pedestrians and a duration of the actions to obtain the analysis result.

[0011] Optionally, the method further comprises: generating a second prompt message to prompt the passenger that the stopping behavior is effective when the analysis result indicates that the stopping behavior is detected and the duration of the stopping behavior exceeds a first preset duration.

[0012] Optionally, the method further comprises: stopping the autonomous taxi from stopping according to the route corresponding to the stop position when a total duration of the passenger continuously moving away from the autonomous taxi is greater than or equal to a second preset duration, or when the identification displayed on the display component of the autonomous taxi disappears, wherein the identification is an identifier allocated to the passenger by the autonomous taxi, and the identifier is used to track the behavior and position of the passenger.

[0013] Optionally, the method further comprises: continuing to stop at the preset starting point after stopping according to the route corresponding to the stop position.

[0014] Optionally, the method further comprises: obtaining the signal strength of the Bluetooth signal of the passenger device according to a predetermined period during the process of controlling the autonomous taxi to stop according to the route corresponding to the stop position, and continuing to control the autonomous taxi to stop according to the route corresponding to the stop position when it is determined that the signal strength increases in turn according to the collection order; and stopping according to the route corresponding to the stop position, and continuing to stop at the preset starting point after stopping according to the route corresponding to the stop position when it is determined that the signal strength decreases in turn according to the collection order within a predetermined period.

[0015] Optionally, the method further comprises: generating a temporary state during the process of controlling the autonomous taxi to stop according to the route corresponding to the stop position so that the right rear door of the autonomous taxi is aligned with the current position information; switching the order processing flow of the autonomous taxi to the temporary state; and switching from the temporary state to the preset starting point state after determining that the right rear door is opened.

[0016] According to another aspect of the embodiments of the present application, a pickup processing device of an autonomous taxi is also provided, comprising: a first determination unit configured to analyze positioning information received from a passenger device to determine whether a passenger is located within a predetermined range around a preset starting point, wherein the preset starting point is a pickup starting point set by the passenger through the passenger device; a first detection unit configured to detect feature information of pedestrians on a current road through a sensor device of the autonomous taxi when it is determined that the passenger is located within the predetermined range; a first analysis unit configured to analyze the feature information to obtain an analysis result; a first identification unit configured to identify current position information of the passenger corresponding to a stop behavior when the analysis result indicates that the stop behavior is detected and a duration of the stop behavior exceeds a preset duration; a second determination unit configured to determine a parking position of the autonomous taxi according to a longitudinal coordinate corresponding to the current position information; a first control unit configured to control the autonomous taxi to park according to a route corresponding to the parking position, so that a right rear door of the autonomous taxi is aligned with the current position information; a first acquisition unit configured to acquire unlocking information initiated by the passenger device after the right rear door is aligned with the current position information; and a first triggering unit configured to trigger the right rear door of the autonomous taxi to open after the unlocking information is verified, so that the passenger gets on the autonomous taxi.

[0017] Optionally, the first determination unit comprises: a first analysis module configured to analyze the positioning information to obtain a current distance between the passenger device and the preset starting point, and determine that the passenger is located within the predetermined range when the current distance is less than a preset distance within a predetermined time window; and a first determination module configured to determine that the passenger is located within the predetermined range when a first prompt message indicating that the passenger device arrives at the preset starting point is received.

[0018] Optionally, the first detection unit comprises: a first detection module configured to detect a passenger stop behavior in a preset area on the right side of the autonomous taxi through a visual device to obtain a pedestrian image of the preset area on the right side and extract features of the pedestrian image to obtain limb actions of the pedestrian, wherein the limb actions at least include arm and hand actions, and head gaze direction; a second detection module configured to detect point cloud data of the pedestrian through a laser radar device; and a first fusion module configured to fuse the limb actions and the point cloud data to obtain the feature information.

[0019] Optionally, the first analysis unit comprises: a second analysis module configured to analyze the feature information to obtain actions of the pedestrian and a duration of the actions, and obtain the analysis result.

[0020] Optionally, the getting-on processing apparatus further comprises a first prompting module configured to generate a second prompting message to prompt the passenger that the stopping behavior is effective when the analysis result indicates that the stopping behavior is detected and the duration of the stopping behavior exceeds a first preset duration.

[0021] Optionally, the getting-on processing apparatus further comprises a first stopping module configured to stop the autonomous taxi from stopping according to the route corresponding to the stop position when a total duration of the passenger continuously moving away from the autonomous taxi is greater than or equal to a second preset duration, or when the identification displayed on the display component of the autonomous taxi disappears, wherein the identification is an identifier allocated to the recognized passenger by the autonomous taxi, and the identifier is used to track the behavior and position of the passenger.

[0022] Optionally, the getting-on processing apparatus further comprises a first adjusting module configured to continue to stop at the preset starting point after stopping according to the route corresponding to the stop position.

[0023] Optionally, the getting-on processing apparatus further comprises a first control module configured to acquire the signal strength of the Bluetooth signal of the passenger device according to a predetermined period during the process of controlling the autonomous taxi to stop according to the route corresponding to the stop position, and continue to control the autonomous taxi to stop according to the route corresponding to the stop position when it is determined that the signal strength increases in turn according to the acquisition order; and a second stopping module configured to continue to stop at the preset starting point after stopping according to the route corresponding to the stop position when it is determined that the signal strength decreases in turn according to the acquisition order within a predetermined period.

[0024] Optionally, the getting-on processing apparatus further comprises a first alignment module configured to generate a temporary state during the process of controlling the autonomous taxi to stop according to the route corresponding to the stop position so that the right rear door of the autonomous taxi is aligned with the current position information; a first switching module configured to switch the order processing flow of the autonomous taxi to the temporary state; and a second switching module configured to switch from the temporary state to the preset starting point state after it is determined that the right rear door is opened.

[0025] According to an aspect of an embodiment of the present application, there is provided an autonomous taxi using the getting-on processing method of the autonomous taxi according to any one of the above.

[0026] According to an aspect of the embodiments of the present application, a processor is provided for running a program, wherein the program performs the pickup processing method of the autonomous taxi according to any one of the above aspects when running.

[0027] According to an aspect of the embodiments of the present application, a computer program product is provided, comprising computer instructions for performing the pickup processing method of the autonomous taxi according to any one of the above aspects when executed by a processor.

[0028] According to an aspect of the embodiments of the present application, a computer readable storage medium is provided, comprising a stored program, wherein the program performs the pickup processing method of the autonomous taxi according to any one of the above aspects when executed by a processor.

[0029] In the embodiments of the present application, after receiving the positioning information from the passenger device, the positioning information is analyzed to determine whether the passenger is located within a predetermined range around a preset starting point, wherein the preset starting point is a pickup starting point set by the passenger through the passenger device; when it is determined that the passenger is located within the predetermined range, the sensor device of the autonomous taxi is used to detect feature information of pedestrians on the current road; the feature information is analyzed to obtain an analysis result; when the analysis result indicates that a stopping behavior is detected and the duration of the stopping behavior exceeds a preset duration, the current position information of the passenger corresponding to the stopping behavior is identified; the current position information is used to determine the stopping position of the autonomous taxi; the autonomous taxi is controlled to stop at the route corresponding to the stopping position, so that the right rear door of the autonomous taxi is aligned with the current position information; after the right rear door is aligned with the current position information, unlocking information initiated by the passenger device is obtained; after the unlocking information is verified, the right rear door of the autonomous taxi is triggered to open, so that the passenger can get on the vehicle. Through the above technical solution, when the passenger is at a non-predetermined station position or in adverse weather conditions, the passenger can call the autonomous taxi by waving his hand at a non-preset station position, the vehicle can dynamically adjust the stopping position according to the actual needs of the passenger, and it is not necessary to strictly follow the fixed station, thereby reducing the waiting and walking time of the passenger in adverse weather or at a non-station position, significantly improving the convenience and riding experience of the passenger, achieving the technical effects of increasing the dynamic stopping flexibility, improving the passenger identification accuracy, enhancing the flexibility and safety of the order processing flow, and improving the riding convenience in adverse weather conditions, thereby solving the technical problem that the autonomous taxi stopping mode is not flexible enough in the related art, the autonomous taxi can only stop at a preset station, and the passenger needs to walk to the designated station for pickup, affecting the riding experience. BRIEF DESCRIPTION OF DRAWINGS

[0030] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0031] Figure 1 is a hardware structure block diagram of a mobile terminal of an automatic driving taxi pick-up processing method according to an embodiment of the present application;

[0032] Figure 2 is a flow chart of an automatic driving taxi pick-up processing method according to an embodiment of the present application;

[0033] Figure 3 is a schematic diagram of an automatic driving taxi pick-up processing device according to an embodiment of the present application.

[0034] Among the above-mentioned drawings, the following reference signs are included:

[0035] 102, processor; 104, memory; 106, transmission device; 108, input / output device. DETAILED DESCRIPTION

[0036] In order to make the technical personnel of the present application better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor should belong to the scope of protection of the present application.

[0037] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0038] As described in the background section, the stopping methods of driverless taxis in related technologies are not flexible enough, only allowing them to stop at preset stations and unable to stop at non-predetermined locations. This forces passengers to walk to designated stations to board, impacting the riding experience. To address these issues and improve passenger experience, this invention proposes a new robotaxi boarding process: passengers flag down the vehicle, which is then identified and marked by the vehicle's perception system. The vehicle control module then stops based on the marked passenger's location, prioritizing stopping the right rear door in front of the passenger. This solution aims to improve passenger convenience, reduce waiting time and walking distance in inclement weather, and improve passenger identification accuracy and recall rate through the fusion of visual and lidar perception technologies, ensuring enhanced riding safety and experience. Specifically, this invention provides a method and apparatus for handling boarding in autonomous taxis, an autonomous taxi, a computer-readable storage medium, a processor, and a computer program product.

[0039] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0040] The methods and embodiments provided in this invention can be executed on a mobile terminal, a computer terminal, or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of handling boarding in an autonomous taxi according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0041] The memory 104 can be used to store computer programs, such as software programs of application software and modules, such as the computer program corresponding to the pickup processing method of the autonomous taxi in the embodiments of the present application. The processor 102 can execute various functional applications and data processing, i.e., implement the above method, by running the computer program stored in the memory 104. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include a memory remotely arranged with respect to the processor 102, which can be connected to the mobile terminal through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The specific examples of the above network can include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet in a wireless manner.

[0042] Embodiment 1

[0043] According to the embodiments of the present application, a method embodiment of the pickup processing method of the autonomous taxi is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in a different order.

[0044] Figure 2 is a flowchart of the pickup processing method of the autonomous taxi according to the embodiments of the present application, as shown in Figure 2 The method includes the following steps:

[0045] In step S202, after receiving the positioning information from the passenger device, the positioning information is analyzed to determine whether the passenger is located within a predetermined range around a preset starting point, wherein the preset starting point is a pickup starting point set by the passenger through the passenger device.

[0046] In this embodiment, when the autonomous taxi approaches the starting point of the passenger's scheduled trip, the vehicle will start to prepare and pay attention to the possible passenger waving behavior within a range of 100 meters before and after the pickup location.

[0047] In this embodiment, after the passenger sets the pickup starting point (i.e., the preset starting point) through the mobile application or the smart device, the system immediately starts monitoring the passenger's location. Here, based on the continuous positioning data sent by the passenger's device, the algorithm calculates the relative position of the passenger to the preset starting point to determine whether the passenger is within the predetermined range of ±100 meters around the starting point.

[0048] For example, the autonomous vehicle (i.e., the autonomous taxi) can obtain real-time GPS or Beidou satellite positioning data through the passenger's smartphone (i.e., the passenger's device) or other positioning devices to locate the passenger. After receiving the positioning information, it is compared with the coordinates of the preset starting point, and the straight-line distance between the two or the walking distance estimated by a more complex path planning algorithm is calculated to determine whether the passenger is within the predetermined range around the preset starting point. If the passenger is indeed within the predetermined range near the preset starting point, the system considers that the passenger has the possibility of hailing a car, and further verifies whether the relevant stop logic needs to be activated. Conversely, if the passenger is far away from the preset starting point, the system considers that the behavior recognition of hailing a car may not be applicable in the current situation, and thus the corresponding response program will not be started.

[0049] By implementing this control method, the passenger's location is pre-screened to ensure that only passengers within a reasonable range have the opportunity to be recognized and responded to, avoiding ineffective processing of irrelevant information far away, and improving system operation efficiency and resource utilization. When the passenger approaches the preset starting point, the system can enter a ready state in advance to capture and respond to the passenger's hailing action more timely, reducing the passenger's waiting time for the vehicle.

[0050] Step S204, when it is determined that the passenger is within the predetermined range, the characteristic information of the pedestrians on the current road is detected by the sensor equipment of the autonomous taxi.

[0051] In this embodiment, the autonomous vehicle detects and recognizes the specific stopping action of the passenger by combining visual and laser radar technologies. Visual perception usually refers to image data captured by a camera, while laser radar (LiDAR) provides high-precision three-dimensional point cloud data. Visual perception helps to recognize the details of the action, while laser radar provides additional spatial positioning and object boundary information, which complement each other to ensure that the system can accurately distinguish the differences between the passenger and other pedestrians. The system focuses on pedestrians near the right side of the road and the rightmost lane, as the stopping and hailing action usually occurs on the right side of the vehicle, which is convenient for the passenger to get on the vehicle directly, and the rightmost lane is the ideal position for the vehicle to stop, reducing traffic interference. The system will pay special attention to the arm and hand movements of the pedestrians, as well as the gaze direction of the head.

[0052] Among them, the autonomous vehicle is usually equipped with various sensors, including but not limited to high-definition cameras, laser radars (LiDAR), millimeter wave radars, and ultrasonic sensors. These sensors each provide different types of data, such as visual images, three-dimensional environment modeling, speed and direction of moving objects, and other information. By integrating the information of these sensors, a composite pedestrian detection and behavior recognition system is established.

[0053] By implementing this control method, the autonomous vehicle can more intelligently and accurately identify and respond to the passenger's stop-and-ride request, improving user experience while ensuring the safety and effectiveness of the operation. Such a technical solution has important practical application value in the field of driverless taxis, especially in complex urban environments, which can effectively improve the service quality and operational efficiency of the vehicle.

[0054] Step S206, analyze the feature information to obtain an analysis result.

[0055] In this embodiment, the system defines several explicit stop-and-ride actions, including raising the palm with the arm, waving left and right and up and down with the extended arm, and raising the thumb diagonally with a single arm. Although the recognition of head posture is technically difficult, it can enhance the confidence of the system as an auxiliary judgment factor.

[0056] This method uses computer vision and machine learning techniques. The system can extract key features of pedestrians from the video stream captured by the camera, such as body shape, posture, facial expression, and hand gestures. The laser radar provides the precise three-dimensional position and motion trajectory of the pedestrian relative to the vehicle. The extracted pedestrian feature information is input into a deep learning model for processing. The model is trained on a large labeled data set and can recognize specific stop-and-ride gestures.

[0057] By implementing this control method, the system can accurately distinguish between different individuals on the road and identify specific hand signals to avoid misidentifying other non-passengers, improving the accuracy of the service. Rapid analysis of sensor data enables the vehicle to make immediate stop decisions, reducing decision-making delays and improving vehicle scheduling and operational efficiency.

[0058] Step S208, when the analysis result indicates that the stop-and-ride behavior is detected and the duration of the stop-and-ride behavior exceeds the preset duration, identify the current location information of the passenger corresponding to the stop-and-ride behavior.

[0059] In this embodiment, the action duration needs to reach at least 1 second, and this threshold setting is to filter out short and occasional actions to ensure that the system's identification of passenger actions is stable and intentional, rather than random or unintentional body movements. The passenger's phone location is at least once, and the distance to the station has been ≤100 meters within any 5-second (slider interval) in the past.

[0060] This method achieves precise position tracking of pedestrians by continuously observing their arm and hand movements and verifying the duration of the behavior against pre-set criteria, integrating data from both visual perception systems (i.e., cameras) and lidar. The visual system provides details about the posture and movement of the action, while the lidar supplements high-precision position information, including the exact position and movement trajectory of the pedestrian in three-dimensional space.

[0061] By implementing this control method, the system can accurately lock onto passengers signaling for a pickup, avoiding misidentification of unrelated pedestrians and reducing ineffective stops for the vehicle, thereby improving the smoothness of the entire service process and passenger satisfaction. Precise passenger identification and dynamic position updates help the vehicle to be more efficiently dispatched and stopped, avoiding delays caused by inaccurate position information and improving the operational efficiency and response speed of the entire autonomous taxi service.

[0062] Step S210: Determine the stopping position of the autonomous taxi based on the longitudinal coordinate corresponding to the current position information.

[0063] In this embodiment, when a stop behavior is detected, the autonomous vehicle will obtain the position and mark the passenger, and according to the longitudinal coordinate or distance, give the target point for stopping.

[0064] This method is based on the analysis of the longitudinal coordinate of the passenger's current precise position to determine the best stopping point for the vehicle. The longitudinal coordinate in this scenario refers to the position of the passenger in the direction of the road (i.e., the direction of vehicle advancement) relative to the pre-set starting point or the point where the stop behavior is detected.

[0065] By implementing this control method, the autonomous taxi can achieve immediate response to passenger requests and precise adjustment of the stopping position. First, the vehicle stops in front of the passenger, greatly reducing the walking distance before getting on the vehicle, especially in non-stop situations or bad weather, this function significantly improves the convenience and comfort of the ride. Second, by dynamically adjusting the stopping position, the autonomous vehicle can avoid wasting time at places where it does not need to stop, reducing ineffective stops and waiting, thereby improving overall dispatch efficiency and vehicle turnover rate. Finally, based on precise coordinate data to determine the stopping point, ensuring the safety of the stopping operation, avoiding traffic accidents or legal disputes that may be caused by improper stopping, providing a safer traffic environment for passengers and road participants.

[0066] Step S212: Control the autonomous taxi to stop according to the route corresponding to the stopping position, so that the right rear door of the autonomous vehicle aligns with the current position information.

[0067] In this embodiment, after receiving the hand signal of the passenger and determining the precise location of the passenger, the autonomous taxi adjusts its driving trajectory, with the ultimate goal of aligning the right rear door of the vehicle with the location of the passenger, within an error range of ±0.5 meters. That is, the vehicle will stop as accurately as possible in front of the passenger, with the right rear door positioned right next to the passenger, facilitating the passenger's boarding. The error range of ±0.5 meters mentioned here refers to the maximum allowed deviation distance between the actual stopping position of the vehicle and the target stopping position (i.e., the actual location of the passenger).

[0068] This method is a collaborative work of the vehicle control system and the path planning algorithm. After confirming the specific location of the passenger (especially the longitudinal coordinate), the central processing unit (CPU) of the vehicle synchronizes this information with the autonomous driving module. Subsequently, based on the passenger location information and the current state of the vehicle, the autonomous driving algorithm generates a new driving route, with the endpoint being the designated stopping position. When planning the route, the algorithm not only considers the shortest straight-line distance, but also ensures the safety and legality of the path, avoiding situations such as reverse driving and crossing obstacles. When the vehicle travels according to the planned route, its control system continuously adjusts steering, acceleration, and braking operations to ensure that the vehicle can stop smoothly and accurately in place. During the stopping process, the positioning system of the vehicle (such as the combination of GPS and IMU, laser radar ranging, etc.) continuously feeds back the distance of the vehicle relative to the destination, until the right rear door aligns with the passenger's location, with an error range of ±0.5 meters.

[0069] By implementing this control method, the autonomous taxi can achieve highly personalized response to passenger requests in complex urban environments. Precise stopping of the vehicle reduces the inconvenience for passengers before boarding, especially for passengers with limited mobility, carrying large luggage, or traveling in adverse weather conditions. This function greatly improves the convenience and comfort of riding.

[0070] Step S214: After the right rear door aligns with the current position information, obtain the unlocking information initiated by the passenger device.

[0071] In this embodiment, when the system identifies and responds to the passenger's hand signal to board, and the vehicle successfully stops near the passenger, the order management system of the vehicle will automatically switch to a special state - passenger stopping after stopping. This state is specifically designed to solve the process connection problem in the case of the passenger's hand signal to board, allowing the passenger to unlock the door and immediately board without additional waiting when the vehicle has not yet arrived at the preset starting position.

[0072] The method utilizes wireless communication technology between the passenger's mobile device (such as a smartphone) and the vehicle to achieve remote unlocking operation. Once the right rear door of the vehicle is accurately aligned with the location of the passenger, the vehicle's control system begins to listen for an unlock signal from the passenger's device. This operation relies on various communication modules installed on the vehicle, such as Bluetooth, Wi-Fi, or cellular network, which can receive the unlock instruction from the passenger's device within a certain range. Usually, this unlock instruction is sent by the passenger operating a special ride-hailing application on their own mobile device. After receiving the unlock request, the vehicle performs a series of verifications, including checking whether the signal source matches the pre-registered passenger information and confirming whether the alignment accuracy meets the unlocking conditions. After verification, the vehicle's electronic lock system will be activated, allowing the passenger to open the door and get on the vehicle smoothly.

[0073] By implementing this control method, even if the vehicle has not completely arrived at the planned starting location, the passenger can send an unlock instruction through the mobile application to open the door in advance. This not only reduces the waiting time of the passenger in bad weather or emergency situations, improves the convenience of taking the vehicle, but also effectively alleviates the risk of system misjudgment, as the system has the opportunity to perform a second confirmation before the passenger actually unlocks and gets on the vehicle, ensuring the accuracy of service docking.

[0074] Step S216, after the unlocking information is verified, the right rear door of the autonomous taxi is triggered to open, allowing the passenger to get on the vehicle.

[0075] In this embodiment, if the system finds that the originally identified "passenger" is not the actual intended passenger or the passenger changes the decision to get on the vehicle during the subsequent confirmation process, the vehicle can quickly exit the "passenger stops after stopping" state, restart the normal pickup process, and go to the next intended starting point or perform other tasks without causing confusion or unnecessary waiting time due to the already unlocked vehicle state.

[0076] After the unlocking information initiated by the passenger's device is successfully received by the vehicle, the system immediately starts the verification program for the unlock request, which involves checking the legality of the signal source and the accuracy of the location information. During the verification process, the vehicle control system compares the passenger ID carried by the unlock signal with the passenger information of the pre-booked trip in the system to confirm the legality of the request source. At the same time, by continuously tracking the passenger's moving path and real-time location, the system checks whether the right rear door of the vehicle is indeed aligned with the passenger's location information to ensure the correctness of the door opening direction and avoid the situation where the passenger needs to go around to the other side of the vehicle or find the door. Once the unlocking information is verified without error, i.e., the passenger's identity and the accuracy of the boarding location are confirmed, the vehicle's electronic door lock system will be activated, the right rear door will be unlocked and automatically opened, providing a seamless and safe boarding path for the passenger.

[0077] By implementing this control method, the autonomous taxi can provide passengers with a more efficient, convenient and safe pickup experience. When the passenger successfully unlocks the door through his own mobile device, not only does he save the step of manually unlocking the door, but he also avoids the inconvenience of finding the correct door in bad weather such as rain and snow. More importantly, this process greatly improves the safety of the pickup, because only when the system fully verifies the legality of the unlocking request and confirms that the right rear door is accurately aligned with the passenger's location information, will the door automatically open. This design avoids safety hazards caused by unauthorized personnel or misoperation, ensuring that only passengers who have booked a trip can enter the vehicle and enjoy the service.

[0078] As can be seen from the above, in this embodiment, after receiving the positioning information from the passenger device, the positioning information is analyzed to determine whether the passenger is located within a predetermined range around the preset starting point, wherein the preset starting point is the pickup starting point set by the passenger through the passenger device; when it is determined that the passenger is located within the predetermined range, the feature information of the pedestrians on the current road is detected through the sensor device of the autonomous taxi; the feature information is analyzed to obtain an analysis result; the stop position of the autonomous taxi is determined according to the longitudinal coordinate corresponding to the current position information; the autonomous taxi is controlled to stop according to the route corresponding to the stop position, so that the right rear door of the autonomous vehicle is aligned with the current position information; after the right rear door is aligned with the current position information, unlocking information initiated by the passenger device is obtained; after the unlocking information is verified, the right rear door of the autonomous taxi is triggered to open, so that the passenger can get on the vehicle. The purpose of allowing the passenger to call the autonomous taxi by waving his hand at a non-predefined station location in bad weather conditions, the vehicle can dynamically adjust the stop position according to the actual needs of the passenger, without having to strictly follow the fixed station, reducing the waiting and walking time of the passenger in bad weather or at a non-station location, significantly improving the convenience of the passenger and the riding experience, thereby achieving the technical effects of increasing the dynamic stopping flexibility, improving the passenger identification accuracy, enhancing the flexibility and safety of the order processing flow, and improving the riding convenience in bad weather conditions.

[0079] Therefore, the technical scheme provided by the above embodiments of the present application solves the technical problem in the related art that the autonomous taxi has insufficient flexibility in stopping, can only stop at a predefined station, and cannot stop at a non-predefined station location, resulting in the passenger needing to walk to the designated station for pickup, affecting the riding experience.

[0080] According to the above embodiment of the present application, determining that the passenger is located in the predetermined range comprises: analyzing the positioning information to obtain a current distance between the passenger device and the preset starting point, and determining that the passenger is located in the predetermined range when the current distance is less than the preset distance within a predetermined time window; and determining that the passenger is located in the predetermined range when a first prompt message of reaching the preset starting point initiated by the passenger device is received.

[0081] In this embodiment, it is clear that the geographic range in which the autonomous taxi starts to respond to the "passenger waving" stopping behavior is within 100 meters before and after the preset passenger pickup site starting position. To improve the accuracy of vehicle stopping and passenger identification, two pre-judgment conditions are proposed, and the vehicle will start the identification process of waving and stopping if any of the conditions is met. The two conditions are: within any 5 seconds in the past, the passenger's mobile phone positioning at least once records that the distance between it and the preset pickup site is not more than 100 meters. This means that even if the passenger does not directly click "arrived at the pickup point", as long as its actual position is close enough to the site, the autonomous taxi will identify its waving action and prepare to stop; the passenger actively clicks the "I have arrived at the pickup point" button on the APP of the car-hailing software, directly telling the vehicle that he is in the pickup preparation state. If neither of the above two situations occurs, that is, the passenger is neither within 100 meters of the site nor has confirmed arrival through the APP, even if the vehicle captures the waving action, it will not start the stopping process.

[0082] This method intelligently judges whether the passenger is in the effective position of the autonomous taxi that can respond to the waving pickup request by analyzing the positioning information of the passenger device (such as a smart phone), combining time window and preset distance conditions. Specifically, the system will continuously track and analyze the GPS or Beidou positioning data of the passenger device, calculate the distance between the current location of the passenger and the preset starting point. If this distance remains within the preset distance (for example, 100 meters) within any continuous predetermined time window (for example, 5 seconds), the system will determine that the passenger is located in the predetermined range that can be responded to, thereby activating the identification mechanism of the waving pickup behavior.

[0083] By implementing this control method, the autonomous taxi can significantly improve its service quality and operational efficiency in complex urban environments. Accurate judgment of the passenger's position not only avoids the vehicle's misresponse to distant or irrelevant waving actions, reducing invalid stopping attempts, but also ensures that when the passenger actually waves near the preset starting point, the vehicle can quickly and accurately respond to the passenger's needs, reducing the passenger's waiting time and improving the riding experience.

[0084] According to the above embodiment of the present application, the sensor device comprises a visual device and a laser radar device, and the feature information of the pedestrian on the current road is detected by the sensor device of the autonomous taxi, comprising: detecting the passenger stopping action of the right side preset area of the autonomous taxi by the visual device to obtain the pedestrian image of the right side preset area, and extracting the features of the pedestrian image to obtain the limb action of the pedestrian, wherein the limb action at least includes: arm and hand action, head gaze direction; detecting the point cloud data of the pedestrian by the laser radar device; and fusing the limb action and the point cloud data to obtain the feature information.

[0085] In this embodiment, the stopping behavior is defined as a series of common actions in line with human daily habits, such as raising the arm above the shoulder, waving the palm, or raising the thumb, which are natural actions that people will make when they wave to stop a car on the street. In order to ensure that the passenger is aware of and can use this service, the APP of the car-hailing software designs a special reminder entrance. In the passenger's interface, this reminder entrance will explicitly inform the passenger: "You can directly stop the car by such behavior, and the vehicle will stop for you, improving the pickup experience".

[0086] This method involves the efficient cooperation of the visual device and the laser radar device. The visual device is mainly used to capture and analyze the pedestrian image of the right side preset area of the autonomous taxi, and its core task is to accurately identify the limb action of the pedestrian, including but not limited to the waving of the arm, the waving of the hand, and the gaze direction of the head, which are important features of traditional stopping behavior. At the same time, the laser radar device generates point cloud data of the pedestrian by emitting laser pulses and receiving reflected signals, which provides accurate position and contour information of the pedestrian in three-dimensional space, and is an effective supplement to visual information.

[0087] For example, by recognizing the passenger stopping action, the key passenger stopping action can be recognized by the visual + laser radar perception fusion method: a. target detection range: lock the pedestrian near the right side of the road and the rightmost lane; b. key posture extraction: arm and hand action, head gaze direction. Among them, the specific actions include but are not limited to: i. raise the arm above the shoulder, wave the palm up and down; ii. stretch out the arm and wave it left and right / up and down; iii. stretch out one arm obliquely and raise the thumb; iv. The head gaze direction is only used as an auxiliary judgment due to accuracy issues, and the key is the above-mentioned key limb action. In addition, the duration of the above-mentioned first three whole actions is greater than or equal to 1 second.

[0088] By implementing this control method, it ensures that the vehicle can respond to the passenger's boarding request in a timely and accurate manner, even without a driver, and can understand the road user's stop gesture as a human driver, increasing the flexibility of the vehicle and the convenience of the passenger. Secondly, by fusing visual and lidar data, the system can effectively filter out background pedestrians or other irrelevant motion interference, reducing the misidentification rate and improving the passenger experience and vehicle operation efficiency. Furthermore, the accurate identification of limb movements, especially arm and hand movements and head gaze direction, enables the vehicle to more intelligently determine the true intention of the passenger, avoiding ineffective stops caused by misinterpretation of behavior, and further improving the accuracy of service.

[0089] According to the above embodiment of the present application, the feature information is analyzed to obtain an analysis result, including: analyzing the feature information to obtain the motion of the pedestrian and the duration of the motion, and obtaining the analysis result.

[0090] In this embodiment, the system stipulates that these key limb movements must last at least 1 second, and this time threshold is set to filter out brief, unintentional limb movements and avoid misjudgment. A 1-second time window can capture the complete stop motion sequence and effectively avoid unnecessary vehicle stop decisions caused by instantaneous motion.

[0091] This method comprehensively utilizes the data collected by the visual equipment and lidar equipment mounted on the autonomous taxi, and through advanced machine learning and artificial intelligence algorithms, it analyzes the limb movements in the pedestrian image in detail. The algorithm focuses on identifying arm movements, hand waving and head direction, and these elements are evaluated to determine whether the pedestrian is performing a specific stop behavior. At the same time, the system monitors the duration of these key postures to ensure that they reach a preset threshold of at least 1 second to improve the accuracy and stability of the identification and avoid false positives caused by brief or accidental movements.

[0092] By implementing this control method, the recognition accuracy of the autonomous taxi for the passenger's stop and boarding intention is significantly improved, and the vehicle's response mechanism for stopping is optimized. This mechanism ensures that only when the system is sure that the pedestrian is performing a clear stop motion, will the vehicle path be adjusted to approach and stop at the target pedestrian, avoiding responses to unrelated pedestrians or false movements, reducing the number of false stops and ineffective stops, and improving operational efficiency.

[0093] According to the above embodiment of the present application, the boarding processing method further includes: when the analysis result indicates that a stop behavior is detected and the duration of the stop behavior exceeds the first preset duration, generating a second prompt message to prompt the passenger that the stop behavior is effective.

[0094] In this embodiment, to enhance the interaction and experience with passengers in a road environment, and to let passengers know that the autonomous vehicle has "seen" the stop, the LED screen in front of the autonomous vehicle will display the corresponding text "found stop, stopping" and turn on the double flash in advance.

[0095] This method automatically triggers the generation of a specific second prompt message when the intelligent perception system of the autonomous taxi identifies that the pedestrian has indeed performed a stop in accordance with the regulations, and the duration of this behavior exceeds the first preset threshold (for example, 1 second) through analysis of feature information. The logic behind this process is the application of a two-way confirmation mechanism: on the one hand, the system ensures the reliability and authenticity of the stop signal by continuously observing and analyzing the pedestrian's actions; on the other hand, by sending a second prompt message to the passenger, such as displaying "found stop, stopping" information on the vehicle's LED display screen or sending a reminder through the mobile app, the passenger is explicitly informed that their stop behavior has been identified and taken effect, and the vehicle will respond soon.

[0096] By implementing this control method, the passenger's ride experience and satisfaction with autonomous taxi services are significantly improved. By sending a second prompt message in a timely manner, passengers can clearly understand that the vehicle is responding to their request to get on the bus, which not only reduces the passenger's anxiety and waiting time, but also increases the transparency and interactivity of the entire service process. For passengers, this is similar to direct communication with human drivers, making autonomous driving services feel more human.

[0097] According to the above embodiment of the present application, the boarding process method further comprises: during the process of controlling the autonomous taxi to stop at the stop position according to the corresponding route, if it is detected that the total duration of the passenger continuously moving away from the autonomous taxi is greater than or equal to a second preset duration; or, when the identification displayed in the display component of the autonomous taxi disappears, stop stopping according to the route corresponding to the stop position, wherein the identification is an identifier allocated by the autonomous taxi to the identified passenger, and the identifier is used to track the behavior and position of the passenger.

[0098] In this embodiment, when the vehicle detects that the person originally marked as the target passenger has been away from the vehicle for more than 2 seconds during the stopping process, or the visual and laser radar sensors can no longer "see" the passenger due to obstruction and other reasons (i.e., the marked ID disappears), the system will automatically cancel the current stopping attempt.

[0099] The method designs a dynamic verification mechanism to monitor the relative position changes between the passenger and the vehicle in real time. During the process of the vehicle preparing to stop based on the previously identified passenger position information, the system continuously monitors the passenger's behavior and position, especially whether the passenger is moving towards the vehicle or maintaining sufficient proximity. If the system detects that the passenger has been continuously moving away from the vehicle for more than a second preset time (for example, 2 seconds), it may indicate that the passenger has changed his intention to get on the vehicle or encountered an emergency, causing him to be out of the vehicle's preset stopping range. In addition, the system also pays attention to whether the specific identifier assigned to the passenger on the display component of the autonomous taxi disappears. This identifier is an important sign for the vehicle to track and confirm the passenger's identity, and once it disappears, it may mean that the passenger has left the camera's field of view or the sensor can no longer accurately track his position. In this case, in order to ensure the safety and efficiency of operation, the system will immediately stop the current stopping operation and execute an emergency plan, such as re-planning the route to the original designated stop point or finding the next suitable stopping position.

[0100] By implementing this control method, it has a significant positive effect on improving the operational efficiency of autonomous taxis and passenger experience. First of all, it greatly enhances the flexibility and adaptability of the vehicle's stopping operation. In the face of passenger behavior uncertainty, such as the passenger temporarily deciding not to ride, sudden physical discomfort or being distracted by other emergencies, the vehicle can quickly respond, avoiding wasting time in no one getting on the location, and reducing the impact on other traffic participants. Secondly, this mechanism also improves the safety of the service, especially avoiding the potential risk of trying to stop when the passenger is obviously far away from the vehicle, reducing the likelihood of traffic accidents. For passengers, the existence of this mechanism means that even when they change their intention to get on the vehicle, the vehicle can adjust the stopping strategy in time, reducing the passengers' anxiety and waiting time, and improving the overall travel experience. Finally, by reducing invalid stopping and waiting, this mechanism can also reduce the energy consumption of the vehicle, improve its operational efficiency in busy urban environments, and make positive contributions to environmental protection and resource conservation.

[0101] According to the above embodiment of the application, the method for getting on the vehicle further comprises: after stopping at the preset starting point, continuing to stop at the preset starting point.

[0102] In this embodiment, when the system automatically cancels the current stopping attempt, it executes the predetermined stopping logic, i.e., returns to the originally planned stop point for picking up passengers.

[0103] The core of this method is the "alternative plan" mechanism, which means that when the current task (i.e. dynamic stop for passengers) encounters obstacles or cannot continue to execute, the autonomous driving system will seamlessly switch to the previously planned backup route or operation process. The system pre-computes the optimal path to the preset starting point based on map data, traffic conditions and station information, and when the dynamic stop attempt is blocked, this path becomes the default navigation target for the vehicle.

[0104] By implementing this control method, the reliability and passenger confidence of autonomous taxi service are significantly enhanced. In practice, this means that even if the passenger fails to arrive on time or the vehicle cannot safely stop due to environmental factors, the vehicle will not be stranded on the way or blindly looking for a new stopping point, but will follow the pre-set route to return to the preset starting point in an orderly manner, waiting for the next passenger or proceeding to the next dispatch. This processing method not only ensures the efficient completion of each trip, but also reduces passenger anxiety and dissatisfaction caused by abnormal situations, improving the overall ride experience.

[0105] According to the above embodiment of the present application, the pickup processing method further comprises: during the process of controlling the autonomous taxi to stop at the stop position according to the route corresponding to the stop position, acquiring the signal strength of the Bluetooth signal of the passenger device according to a predetermined period, and when it is determined that the signal strength increases in turn according to the acquisition order, continuing to control the autonomous taxi to stop at the stop position according to the route corresponding to the stop position; when it is determined that the signal strength decreases in turn according to the acquisition order within a predetermined period, then after stopping at the stop position according to the route corresponding to the stop position, continue to stop at the preset starting point.

[0106] In this embodiment, in addition to the visual and lidar recognition, the system also uses the change in Bluetooth signal strength to determine whether the target passenger is approaching the vehicle during the autonomous taxi stop. As the vehicle stops and gradually approaches the passenger, the Bluetooth signal between the vehicle and the passenger's phone will strengthen as the distance between them decreases. The system will periodically (every 1 second) check this signal strength, and if the signal consistently strengthens, this is a sign that the vehicle is approaching the passenger correctly, and the system will determine that the stop is in progress and going in the right direction, so it will continue to execute the stop procedure until the vehicle's right rear door is aligned with the passenger. Conversely, if the Bluetooth signal strength shows a weakening trend for 2 seconds during the stop, this indicates that the distance between the vehicle and the passenger is increasing, meaning that the passenger may not be at the location where the vehicle is stopping. This signal weakening in this case is considered a warning of a failed stop, and the system will immediately abort the current special stop operation, as this means that the vehicle cannot accurately stop in front of the passenger, or the passenger may have moved from the original location. To avoid getting stuck in an infinite loop after identifying the wrong passenger or failing a stop attempt, the system designs a "catch-all" strategy. If the above cancellation conditions are triggered during the stop, the vehicle will continue to monitor the roadside hailing behavior on the way to the original preset station. If a qualified hailing action is identified again, the system will update the target passenger's ID and restart the stop process. However, to prevent this process from repeating indefinitely and affecting overall order-taking efficiency and service quality, the system limits the number of times this re-identification and stop attempt can be made to a maximum of 2 times. If both attempts fail to successfully stop or identify the correct passenger, the system will completely stop the hailing stop and only perform the normal stop procedure to the preset station.

[0107] When the vehicle begins to stop in the direction of the identified passenger, the system will collect the Bluetooth signal strength from the passenger's device at a set time interval (e.g. every 1 second). If the signal strength shows an upward trend over time, i.e. each round of signal collection is stronger than the previous one, this means that the distance between the vehicle and the passenger is gradually decreasing, and the system will determine that the stop is proceeding correctly, so it will continue to execute the current stop instructions until the vehicle's right rear door is perfectly aligned with the passenger's location. However, if the signal strength shows a downward trend over a preset period of time (e.g. for 2 consecutive seconds), i.e. each time the signal strength is weaker than the previous one, this indicates that the distance between the vehicle and the passenger is increasing, which may mean that the passenger is moving or the vehicle is not stopping in the correct direction of the passenger. At this time, the system will immediately abort the stop operation and no longer follow the current stop location route. Subsequently, the vehicle will automatically return to the original route planning and drive to the preset pickup station.

[0108] By implementing this control method, it plays an important role in improving the pick-up experience of passengers and the overall operational efficiency of autonomous taxis. It not only provides a way for the vehicle to verify the passenger's location in real time, ensuring the accuracy of the stop and reducing the waiting time caused by positioning errors, but also enhances the system's ability to respond to unexpected situations, such as sudden movements of passengers or slight deviations of the vehicle positioning system. When the vehicle detects a decrease in signal strength, it can quickly adjust its strategy to avoid ineffective stops, prevent possible passenger loss and dissatisfaction, and reduce potential interference with other vehicles and pedestrians on the road, thereby improving the safety of the overall traffic environment.

[0109] According to the above embodiment of the present application, the pick-up processing method further comprises: generating a temporary state during the process of controlling the autonomous taxi to stop at the stop position corresponding to the route, so that the right rear door of the autonomous vehicle is aligned with the current position information; switching the order processing flow of the autonomous taxi to the temporary state; switching from the temporary state to the preset starting point state after determining that the right rear door is opened.

[0110] In this embodiment, when the autonomous taxi recognizes the passenger's hand signal and completes the stop, the vehicle management system will not immediately enter the traditional "arrival starting point" state, but will first enter a newly added temporary state - "passenger stops after stopping". In this temporary state, although the vehicle has not been officially marked as "arrived at the starting point", the passenger can already send a command to unlock the vehicle door through the mobile APP, and then get on the vehicle.

[0111] This method is based on real-time response to passenger behavior and dynamic adjustment of vehicle position. In the activation of the temporary state, the order processing flow of the vehicle will automatically switch to this state. At this time, the vehicle has not yet fully reached the preset "arrival starting point" state, as it may have stopped at a position closer to the passenger than the planned station. Once the vehicle stops in place and the passenger successfully unlocks and opens the right rear door to get on the vehicle, the system's intelligent logic will recognize this behavior as a clear signal that the passenger wants to get on the vehicle. At this time, the temporary state will automatically end, and the order processing flow of the vehicle will officially switch to the "arrival at the preset starting point" state, indicating that the passenger has successfully gotten on the vehicle, and the vehicle can begin to execute subsequent travel instructions, such as confirming passenger information and updating the travel state.

[0112] By implementing this control method, it allows passengers to immediately start the pick-up process as soon as the vehicle stops, without waiting for any additional confirmation or state change, thereby significantly shortening the waiting time of passengers and improving the immediacy and convenience of getting on the vehicle.

[0113] From the above, in the embodiment of the application, by limiting the aforementioned constraint condition, on the premise of improving the behavior recognition accuracy: in the process of vehicle parking, the passenger waves to stop the vehicle → the vehicle perception system recognizes and marks → gives the vehicle control module a new tendency of stopping point → tries to park the right rear door of the vehicle in front of the passenger. Of course, more positioning and interaction information will be used in the process to improve the accuracy and recall rate, and thus improve the riding experience.

[0114] It should be noted that, for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the action sequence described, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0115] From the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for causing a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the method of each embodiment of the present application.

[0116] Embodiment 2

[0117] According to the embodiment of the present application, an automatic taxi pickup processing device for implementing the automatic taxi pickup processing method is also provided, Figure 3 is a schematic diagram of the automatic taxi pickup processing device according to the embodiment of the present application, as Figure 3 shown, the device includes a first determination unit 301, a first detection unit 303, a first analysis unit 305, a first identification unit 307, a second determination unit 309, a first control unit 311, a first acquisition unit 313, and a first triggering unit 315. The device will be described in detail below.

[0118] The first determination unit 301 is configured to analyze the positioning information after receiving the positioning information from the passenger device to determine whether the passenger is located within a predetermined range around a preset starting point, wherein the preset starting point is a pickup starting point set by the passenger through the passenger device.

[0119] The first detection unit 303 is configured to detect feature information of pedestrians on the current road through a sensor device of the autonomous taxi when it is determined that the passenger is within the predetermined range.

[0120] The first analysis unit 305 is configured to analyze the feature information to obtain an analysis result.

[0121] The first identification unit 307 is configured to identify current location information of the passenger corresponding to the stopping behavior when the analysis result indicates that the stopping behavior is detected and the duration of the stopping behavior exceeds the preset duration.

[0122] The second determination unit 309 is configured to determine a stop position of the autonomous taxi according to a longitudinal coordinate corresponding to the current location information.

[0123] The first control unit 311 is configured to control the autonomous taxi to stop at the stop position according to a route corresponding to the stop position, so that the right rear door of the autonomous vehicle is aligned with the current location information.

[0124] The first acquisition unit 313 is configured to acquire unlocking information initiated by the passenger device after the right rear door is aligned with the current location information.

[0125] The first triggering unit 315 is configured to trigger the right rear door of the autonomous taxi to open after the unlocking information is verified, so that the passenger gets on the vehicle.

[0126] It should be noted that the first determination unit 301, the first detection unit 303, the first analysis unit 305, the first identification unit 307, the second determination unit 309, the first control unit 311, the first acquisition unit 313 and the first triggering unit 315 correspond to steps S202 to S216 in the above embodiment, and the eight units have the same instances and application scenarios as the corresponding steps, but are not limited to the content disclosed in the above embodiment.

[0127] From the above, in the scheme described in the above embodiment of the present application, the first determination unit can be used to analyze the positioning information after receiving the positioning information from the passenger device to determine whether the passenger is located within the predetermined range around the preset starting point, wherein the preset starting point is the pickup starting point set by the passenger through the passenger device; the first detection unit is used to detect the feature information of the pedestrian on the current road through the sensor device of the autonomous taxi when it is determined that the passenger is located within the predetermined range; the first analysis unit is used to analyze the feature information to obtain an analysis result; the first identification unit is used to identify the current position information of the passenger corresponding to the stopping behavior when the analysis result indicates that the stopping behavior is detected and the duration of the stopping behavior exceeds the preset duration; the second determination unit is used to determine the stopping position of the autonomous taxi according to the longitudinal coordinate corresponding to the current position information; the first control unit is used to control the autonomous taxi to stop according to the route corresponding to the stopping position, so that the right rear door of the autonomous vehicle is aligned with the current position information; the first acquisition unit is used to acquire the unlocking information initiated by the passenger device after the right rear door is aligned with the current position information; and the first triggering unit is used to trigger the right rear door of the autonomous taxi to open after the unlocking information is verified, so that the passenger gets on the vehicle. Through the above scheme, when the non-predetermined station position or the adverse weather condition, the passenger is allowed to call the autonomous taxi by waving his hand at the non-preset station position, the vehicle can dynamically adjust the stopping position according to the actual needs of the passenger, and it is not necessary to strictly follow the fixed station, thereby reducing the waiting and walking time of the passenger in adverse weather or non-station position, significantly improving the convenience and riding experience of the passenger, thereby achieving the technical effects of increasing the dynamic stopping flexibility, improving the passenger identification accuracy, enhancing the flexibility and safety of the order processing flow, and improving the riding convenience in adverse weather conditions, thereby solving the technical problems that the autonomous taxi stopping mode is not flexible enough in the related art, and the autonomous taxi can only stop at the preset station, and the passenger needs to walk to the designated station to get on the vehicle, thereby affecting the riding experience.

[0128] Optionally, the first determination unit comprises: a first analysis module, configured to analyze the positioning information to obtain a current distance between the passenger device and the preset starting point, and determine that the passenger is located in the predetermined range when the current distance is less than a preset distance within a predetermined time window; and a first determination module, configured to determine that the passenger is located in the predetermined range when receiving a first prompt message of reaching the preset starting point initiated by the passenger device.

[0129] Optionally, the first detection unit comprises: a first detection module configured to detect, by the vision device, a passenger stopping action of a preset area on the right side of the autonomous taxi, to obtain a pedestrian image of the preset area on the right side, and to perform feature extraction on the pedestrian image to obtain a body action of the pedestrian, wherein the body action at least includes: arm and hand action, head gaze direction; a second detection module configured to detect, by the laser radar device, point cloud data of the pedestrian; and a first fusion module configured to fuse the body action and the point cloud data to obtain feature information.

[0130] Optionally, the first analysis unit comprises: a second analysis module configured to analyze the feature information to obtain an action of the pedestrian and a duration of the action, to obtain an analysis result.

[0131] Optionally, the pickup processing device further comprises: a first prompting module configured to generate a second prompt message to prompt that the stopping action is effective when the analysis result indicates that the stopping action is detected and the duration of the stopping action exceeds a first preset duration.

[0132] Optionally, the pickup processing device further comprises: a first stopping module configured to, in a process of controlling the autonomous taxi to stop at the route corresponding to the stop position, stop stopping at the route corresponding to the stop position if it is detected that the total duration of the passenger continuously moving away from the autonomous taxi is greater than or equal to a second preset duration, or it is detected that the identifier displayed in the display component of the autonomous taxi disappears, wherein the identifier is an identifier allocated by the autonomous taxi to the recognized passenger, and the identifier is used to track the behavior and position of the passenger.

[0133] Optionally, the pickup processing device further comprises: a first adjustment module configured to continue to stop at the preset starting point after stopping at the route corresponding to the stop position.

[0134] Optionally, the pickup processing device further comprises: a first control module configured to, in a process of controlling the autonomous taxi to stop at the route corresponding to the stop position, acquire a signal strength of a Bluetooth signal of the passenger device according to a predetermined period, and continue to control the autonomous taxi to stop at the route corresponding to the stop position when it is determined that the signal strength increases in turn according to the acquisition order; and a second stopping module configured to, when it is determined that the signal strength decreases in turn according to the acquisition order within a predetermined period, stop stopping at the route corresponding to the stop position and continue to stop at the preset starting point.

[0135] Optionally, any one of the above modules comprises: a first alignment module, configured to generate a temporary state in a process of controlling the autonomous taxi to stop at the stop position according to the route corresponding to the stop position, so that the right rear door of the autonomous vehicle is aligned with the current position information; a first switching module, configured to switch an order processing flow of the autonomous taxi to the temporary state; and a second switching module, configured to switch from the temporary state to a state of reaching a preset starting point after determining that the right rear door is opened.

[0136] According to an aspect of an embodiment of the present application, there is provided an autonomous taxi, which uses the method for picking up a passenger in the autonomous taxi according to any one of the above embodiments.

[0137] According to an aspect of an embodiment of the present application, there is provided a processor, which is configured to run a program, wherein the program performs the method for picking up a passenger in the autonomous taxi according to any one of the above embodiments when the program is running.

[0138] According to an aspect of an embodiment of the present application, there is provided a computer program product, comprising computer instructions, which perform the method for picking up a passenger in the autonomous taxi according to any one of the above embodiments when executed by a processor.

[0139] According to an aspect of an embodiment of the present application, there is provided a computer readable storage medium, comprising a stored program, wherein the program performs the method for picking up a passenger in the autonomous taxi according to any one of the above embodiments when executed by a processor.

[0140] Optionally, in the present embodiment, the computer readable storage medium can be located in any one of a group of computer terminals in a computer network, or in any one of a group of communication devices.

[0141] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: after receiving the positioning information from the passenger device, analyzing the positioning information to determine whether the passenger is located within a predetermined range around a preset starting point, wherein the preset starting point is a pickup starting point set by the passenger through the passenger device; when it is determined that the passenger is located within the predetermined range, detecting feature information of pedestrians on the current road through a sensor device of the autonomous taxi; analyzing the feature information to obtain an analysis result; when the analysis result indicates that a stopping behavior is detected and a duration of the stopping behavior exceeds a preset duration, identifying current position information of the passenger corresponding to the stopping behavior; determining a stopping position of the autonomous taxi according to a longitudinal coordinate corresponding to the current position information; controlling the autonomous taxi to stop according to a route corresponding to the stopping position, so that a right rear door of the autonomous vehicle is aligned with the current position information; after the right rear door is aligned with the current position information, obtaining unlocking information initiated by the passenger device; after the unlocking information is verified, triggering the right rear door of the autonomous taxi to open, so that the passenger gets on the vehicle.

[0142] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: analyzing the positioning information to obtain a current distance between the passenger device and the preset starting point, and determining that the passenger is located in the predetermined range when the current distance is less than a preset distance within a predetermined time window.

[0143] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: the visual device and the laser radar device detect feature information of pedestrians on the current road through a sensor device of the autonomous taxi, including: detecting a passenger stopping action of the right side of the autonomous taxi through the visual device to obtain pedestrian images of the right side of the autonomous taxi, and extracting features of the pedestrians to obtain limb actions of the pedestrians, wherein the limb actions at least include: arm and hand actions, head gaze direction; detecting point cloud data of the pedestrians through the laser radar device; and fusing the limb actions and the point cloud data to obtain the feature information.

[0144] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: analyzing the feature information to obtain actions of the pedestrians and a duration of the actions, and obtaining an analysis result.

[0145] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: generating a second prompt message to prompt the passenger that the stopping behavior is effective when the analysis result indicates that the stopping behavior is detected and the duration of the stopping behavior exceeds the first preset duration.

[0146] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: stopping the automatic driving taxi from stopping according to the route corresponding to the stopping position when it is detected that the total duration of the passenger continuously moving away from the automatic driving taxi is greater than or equal to the second preset duration, or the identification displayed in the display component of the automatic driving taxi disappears, wherein the identification is an identifier allocated by the automatic driving taxi for the identified passenger, and the identifier is used to track the behavior and position of the passenger.

[0147] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: continuing to stop at the preset starting point after stopping according to the route corresponding to the stopping position.

[0148] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: obtaining the signal strength of the Bluetooth signal of the passenger device according to a predetermined period during the process of controlling the automatic driving taxi to stop according to the route corresponding to the stopping position, and continuing to control the automatic driving taxi to stop according to the route corresponding to the stopping position when it is determined that the signal strength increases in turn according to the collection order; and stopping according to the route corresponding to the stopping position, and continuing to stop at the preset starting point after stopping according to the route corresponding to the stopping position when it is determined that the signal strength decreases in turn according to the collection order within a predetermined period.

[0149] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: generating a temporary state during the process of controlling the automatic driving taxi to stop according to the route corresponding to the stopping position so that the right rear door of the automatic driving vehicle is aligned with the current position information; switching the order processing flow of the automatic driving taxi to the temporary state; and switching from the temporary state to the state of reaching the preset starting point after it is determined that the right rear door is opened.

[0150] In the above embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0151] In several embodiments provided in the present application, it should be understood that the disclosed technical contents can be implemented by other manners. Among them, the above-described device embodiments are only illustrative, for example, the division of units can be a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection between the units or modules through some interfaces, and can be electrical or other forms.

[0152] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0153] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed to multiple units. Part or all of the units can be selected to achieve the purpose of the present embodiment according to actual needs.

[0154] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0155] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the embodiments of the present application. The foregoing storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0156] The above merely is the preferred embodiment of the present application, it should be pointed out that, for ordinary skilled in the art, without departing from the principles of the present application, can also make a number of improvements and refinements, these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method for pickup processing of an autonomous taxi, characterized by, The method comprises: after receiving the positioning information from the passenger device, analyzing the positioning information to determine whether the passenger is located within a predetermined range around a preset starting point, wherein the preset starting point is a pickup starting point set by the passenger through the passenger device; when it is determined that the passenger is located within the predetermined range, detecting feature information of pedestrians on the current road through a sensor device of the autonomous taxi; analyzing the feature information to obtain an analysis result; when the analysis result indicates that a stopping behavior is detected and the duration of the stopping behavior exceeds a preset duration, identifying the current location information of the passenger corresponding to the stopping behavior; determining the stopping position of the autonomous taxi according to the longitudinal coordinate corresponding to the current location information; controlling the autonomous taxi to stop at the stopping position according to the route corresponding to the stopping position, so that the right rear door of the autonomous taxi is aligned with the current location information; after the right rear door is aligned with the current location information, obtaining unlocking information initiated by the passenger device; after the unlocking information is verified, triggering the right rear door of the autonomous taxi to open, so that the passenger gets on the vehicle; wherein the pickup processing method further comprises: generating a temporary state during the process of controlling the autonomous taxi to stop at the stopping position according to the route corresponding to the stopping position, so that the right rear door of the autonomous taxi is aligned with the current location information; switching the order processing flow of the autonomous taxi to the temporary state; switching from the temporary state to the preset starting point after determining that the right rear door is opened.

2. The method of claim 1, wherein Determining that the passenger is located in the predetermined range comprises: analyzing the positioning information to obtain the current distance between the passenger device and the preset starting point, and determining that the passenger is located in the predetermined range when the current distance is less than a preset distance within a predetermined time window; when receiving a first prompt message initiated by the passenger device to arrive at the preset starting point, determining that the passenger is located in the predetermined range.

3. The method of claim 1, wherein The sensor device comprises a visual device and a laser radar device, and the feature information of pedestrians on the current road is detected through the sensor device of the autonomous taxi, which comprises: detecting the passenger stopping action of the right side preset area of the autonomous taxi through the visual device to obtain the pedestrian image of the right side preset area, and extracting the feature of the pedestrian image to obtain the limb action of the pedestrian, wherein the limb action at least includes arm and hand action, head gaze direction; detecting the point cloud data of the pedestrian through the laser radar device; fuse the limb action and the point cloud data to obtain the feature information.

4. The method of claim 1, wherein Analyzing the feature information to obtain an analysis result comprises: analyzing the feature information to obtain the action of the pedestrian and the duration of the action to obtain the analysis result.

5. The method of claim 1, wherein The pickup processing method further comprises: When the analysis result indicates that the stopping behavior is detected and the duration of the stopping behavior exceeds a first preset duration, a second prompt message is generated to prompt that the stopping behavior is effective.

6. The method of claim 1, wherein The pickup processing method further includes: In the process of controlling the autonomous taxi to stop at the route corresponding to the stop position, if it is detected that the total duration of the passenger continuously moving away from the autonomous taxi is greater than or equal to a second preset duration, or it is detected that the identification displayed in the display component of the autonomous taxi disappears, the autonomous taxi stops stopping at the route corresponding to the stop position, wherein the identification is an identifier allocated to the recognized passenger by the autonomous taxi, and the identifier is used to track the behavior and position of the passenger.

7. The method of claim 1, wherein The pickup processing method further includes: After stopping at the route corresponding to the stop position, the autonomous taxi continues to stop at the preset starting point.

8. The method of claim 1, wherein, The pickup processing method further includes: In the process of controlling the autonomous taxi to stop at the route corresponding to the stop position, the signal strength of the Bluetooth signal of the passenger device is acquired at a predetermined period, and when it is determined that the signal strength increases in turn according to the acquisition order, the autonomous taxi continues to stop at the route corresponding to the stop position; When it is determined that the signal strength decreases in turn according to the acquisition order within a predetermined period, the autonomous taxi stops stopping at the route corresponding to the stop position and continues to stop at the preset starting point.

9. A boarding processing device of an autonomous taxi, characterized by, It includes: A first determination unit is configured to analyze the positioning information received from the passenger device to determine whether the passenger is located within a predetermined range around a preset starting point after receiving the positioning information, wherein the preset starting point is a pickup starting point set by the passenger through the passenger device; A first detection unit is configured to detect feature information of pedestrians on the current road through a sensor device of the autonomous taxi when it is determined that the passenger is located within the predetermined range; A first analysis unit is configured to analyze the feature information to obtain an analysis result; A first identification unit is configured to identify the current position information of the passenger corresponding to the stopping behavior when the analysis result indicates that the stopping behavior is detected and the duration of the stopping behavior exceeds a preset duration; A second determination unit is configured to determine a stop position of the autonomous taxi according to a longitudinal coordinate corresponding to the current position information; A first control unit is configured to control the autonomous taxi to stop at a route corresponding to the stop position, so that the right rear door of the autonomous taxi is aligned with the current position information; A first acquisition unit is configured to acquire unlocking information initiated by the passenger device after the right rear door of the autonomous taxi is aligned with the current position information; A first triggering unit is configured to trigger the right rear door of the autonomous taxi to open to allow the passenger to get on the vehicle after the unlocking information is verified. The pickup processing device further comprises: a first alignment module, configured to generate a temporary state during control of the autonomous taxi to stop at the route corresponding to the stop position, so that the right rear door of the autonomous taxi is aligned with the current position information; a first switching module, configured to switch the order processing flow of the autonomous taxi to the temporary state; and a second switching module, configured to switch from the temporary state to the preset starting point state after determining that the right rear door is opened.

Images