Pick-up methods, devices, and autonomous vehicles based on autonomous vehicles
By analyzing environmental perception data, autonomous vehicles can proactively determine passengers' need for another ride, generate pick-up instructions, and dispatch a second vehicle. This solves the problem of passive service response after passengers disembark, improving passenger experience and vehicle utilization.
Patent Information
- Application Number
- CN202611328907.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-28
- Publication Date
- 2026-11-17
AI Technical Summary
Autonomous vehicles struggle to proactively assess a passenger's need for another ride after they disembark, leading to reactive service responses that negatively impact user experience and vehicle utilization.
By analyzing environmental perception data, the system proactively determines passengers' travel intentions and generates pick-up instructions to dispatch a second autonomous vehicle, achieving fully automated service throughout the journey.
It improves the passenger experience, especially by proactively providing assistance in emergencies, and increases the utilization rate of autonomous vehicles.
Smart Images

Figure CN123113534A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of artificial intelligence technology, and in particular to the fields of autonomous driving, deep learning, and electric vehicles, specifically to a ride-sharing method, device, and autonomous vehicle based on an autonomous vehicle. Background Technology
[0002] With the development of autonomous driving technology, autonomous vehicles can be used as ride-hailing vehicles to provide passengers with tasks such as booking and delivery.
[0003] However, after the trip ends, the status of disembarking passengers is unknown, making it difficult to continue providing rides to passengers who need to travel again, thus affecting the user experience. Summary of the Invention
[0004] This disclosure provides a ride-hailing method, apparatus, and autonomous vehicle based on an autonomous vehicle.
[0005] According to one aspect of this disclosure, a ride-hailing method based on an autonomous vehicle is provided, comprising: in response to detecting that a person leaves a first autonomous vehicle, analyzing environmental perception data of the first autonomous vehicle to obtain the person's ride intention; wherein the environmental perception data includes data affecting the person's behavior; if the ride intention indicates that the person has a need to ride again, generating a ride-hailing instruction, wherein the ride-hailing instruction is used to instruct a second autonomous vehicle to pick up the person; and sending the ride-hailing instruction to the second autonomous vehicle to schedule the second autonomous vehicle to pick up the person.
[0006] According to another aspect of this disclosure, a ride-hailing device based on an autonomous vehicle is provided, comprising: an analysis module, configured to analyze environmental perception data of the first autonomous vehicle in response to detecting that a person has left the first autonomous vehicle, to obtain the person's ride intention; wherein the environmental perception data includes data affecting the person's behavior; a generation module, configured to generate a ride-hailing instruction when the ride intention indicates that the person has a need to ride again, wherein the ride-hailing instruction is used to instruct a second autonomous vehicle to pick up the person; and a scheduling module, configured to send the ride-hailing instruction to the second autonomous vehicle to schedule the second autonomous vehicle to pick up the person.
[0007] According to another aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method described above.
[0008] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause a computer to perform the methods described above.
[0009] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method described above.
[0010] According to another aspect of this disclosure, an autonomous vehicle is provided, configured to implement the above-described autonomous vehicle-based pick-up method.
[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description
[0012] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:
[0013] Figure 1 The illustration schematically shows an exemplary system architecture applicable to a ride-hailing method and apparatus based on an autonomous vehicle, according to embodiments of the present disclosure;
[0014] Figure 2 A flowchart illustrating a ride-hailing method based on an autonomous vehicle according to an embodiment of the present disclosure is shown schematically.
[0015] Figure 3 The illustration schematically depicts a scenario of a ride-hailing method based on an autonomous vehicle according to an embodiment of the present disclosure;
[0016] Figure 4 A flowchart illustrating a specific embodiment of a ride-hailing method based on an autonomous vehicle according to the present disclosure is shown schematically.
[0017] Figure 5 This illustration schematically depicts a passenger pick-up and drop-off scenario involving associated buildings according to a specific embodiment of this disclosure;
[0018] Figure 6 A block diagram of a ride-hailing device based on an autonomous vehicle according to an embodiment of the present disclosure is shown schematically.
[0019] Figure 7 A schematic block diagram of an example electronic device that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation
[0020] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.
[0021] As autonomous driving technology enters the stage of large-scale commercial operation, operating autonomous vehicles no longer faces the simple issue of technical feasibility, but rather how to provide an experience and safety guarantees that approach or even surpass those of human driving. For human-driven ride-hailing services, drivers can autonomously determine a passenger's intentions by observing them in the rearview mirror and confirming verbally when they disembark, and proactively provide assistance based on this judgment, such as honking the horn to remind them, waiting for confirmation, or picking them up again. However, the state of autonomous vehicles after a passenger disembarks can be likened to a state of "blindness," where sensors no longer focus on the passenger, and the autonomous vehicle is completely "unaware" (figuratively) of the previous passenger's condition.
[0022] However, due to various factors, passengers who disembark may have a need to ride again, making it difficult for autonomous vehicles to provide this service. For example, the technology requires passengers to actively request the vehicle to wait, and then waits, making it difficult to mimic the actions of a real "driver" to proactively provide service. Alternatively, the technology may rely on passenger instructions in specific scenarios to passively confirm whether to continue driving. For instance, if a passenger disembarks at a non-pre-determined location, the vehicle might leave after receiving a message indicating they will not ride again, but wait at the original location after receiving a message indicating they will continue. If the passenger does not respond, the vehicle cannot start, impacting vehicle utilization. This method's judgment remains at the current trip stage; it only ends the current trip after receiving a passenger response, rather than assessing the passenger's intention to ride again after the current trip has ended, and it does not involve other autonomous vehicles.
[0023] Furthermore, both of these methods are essentially passive, responsive services—taking action only when a passenger explicitly requests assistance, failing to provide help when a passenger encounters difficulties but does not actively seek assistance. For example, in situations where a passenger is passively unable to respond (such as being unwell, having a dead phone battery, or being in a dangerous environment), or in situations where a passenger is actively unable to respond (such as not realizing they have forgotten their luggage or not realizing they need assistance), passively responsive autonomous vehicles are unlikely to proactively provide help and may even delay rescue efforts by waiting for a passenger's response.
[0024] In view of this, this disclosure provides a method, apparatus, and autonomous vehicle for picking up passengers. After a passenger finishes their current trip, the system proactively uses environmental perception data from the first autonomous vehicle to perceive the passenger's state and travel intention. If the passenger requests another ride, a second autonomous vehicle is dispatched to assist. Throughout the process, the system reuses environmental perception data from the current trip, proactively determines travel intention, proactively generates pick-up instructions, and proactively dispatches vehicles. This fully automated process simulates a real "driver" proactively providing service, even in emergencies, thus improving the passenger experience. Furthermore, the first autonomous vehicle does not need to wait for the passenger's response after disembarking, further increasing the utilization rate of the autonomous vehicle.
[0025] Figure 1 This illustration schematically depicts an exemplary system architecture applicable to a ride-hailing method and apparatus based on autonomous vehicles, according to embodiments of this disclosure. It should be noted that... Figure 1 The examples shown are merely examples of system architectures that can be applied to the embodiments of this disclosure, in order to help those skilled in the art understand the technical content of this disclosure, but do not mean that the embodiments of this disclosure cannot be used in other devices, systems, environments or scenarios.
[0026] like Figure 1 As shown, the system architecture 100 according to this embodiment may include at least one autonomous vehicle 101, a network 102, and a server 103. The network 102 serves as a medium for providing a communication link between the at least one autonomous vehicle 101 and the server 103. The network 102 may include various connection types, such as wired and / or wireless communication links, etc.
[0027] At least one of the autonomous vehicles 101 can interact with the server 103 via network 102 to receive or send messages, etc. Various communication client applications may be installed on at least one autonomous vehicle 101, such as knowledge reading applications, web browser applications, search applications, instant messaging tools, email clients, and / or social media platform software, etc. (for example only).
[0028] At least one autonomous vehicle 101 may be a vehicle equipped with autonomous driving capabilities, including but not limited to vehicles that simultaneously support autonomous driving and manual driving capabilities, and vehicles that only support autonomous driving capabilities. At least one autonomous vehicle 101 may include a first autonomous vehicle and a second autonomous vehicle.
[0029] Server 103 can be a server that provides various services, such as a backend management server that makes decisions or synchronizes information regarding the driving behavior of at least one autonomous vehicle 101. The backend management server can analyze and process received requests and other data, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to at least one autonomous vehicle 101.
[0030] A server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system. It solves the shortcomings of traditional physical hosts and VPS services (Virtual Private Server, or VPS for short) in terms of high management difficulty and weak business scalability. A server can also be a server for a distributed system, or a server that incorporates blockchain technology.
[0031] It should be noted that the ride-hailing method and device based on autonomous vehicles provided in this disclosure can generally be executed by autonomous vehicles, and correspondingly, the device can also be installed in autonomous vehicles. Alternatively, the ride-hailing method and device based on autonomous vehicles provided in this disclosure can also be executed by server 103, and correspondingly, the device can generally also be installed in server 103.
[0032] It should be understood that Figure 1 The number of autonomous vehicles, networks, and servers shown is merely illustrative. Any number of autonomous vehicles, networks, and servers can be included depending on implementation needs.
[0033] In the embodiments disclosed herein, the collection, storage, use, processing, transmission, provision, disclosure, and application of user personal information comply with relevant laws and regulations, necessary confidentiality measures have been taken, and they do not violate public order and good morals.
[0034] In the technical solution disclosed herein, the user's authorization or consent is obtained before acquiring or collecting the user's personal information.
[0035] Figure 2 A flowchart illustrating a ride-hailing method based on an autonomous vehicle according to an embodiment of the present disclosure is shown schematically. Figure 2 As shown, the method includes operations S210~S230.
[0036] In operation S210, in response to detecting that an object leaves the first autonomous vehicle, the environmental perception data of the first autonomous vehicle is analyzed to obtain the object's intention to ride in the vehicle; wherein, the environmental perception data includes data that affects the object's behavior.
[0037] In operation S220, if the passenger intending to ride has a need for another ride, a pick-up instruction is generated, wherein the pick-up instruction is used to instruct the second autonomous vehicle of the passenger.
[0038] In operation S230, a pick-up command is sent to the second autonomous vehicle to schedule the pick-up target for the second autonomous vehicle.
[0039] The "object" can refer to passengers riding in autonomous vehicles. Passengers can create trips through various means such as online booking, and the autonomous vehicles will execute those trips. In this embodiment, a trip can be automatically created and executed by the autonomous vehicle at the same time as the pick-up instruction is generated.
[0040] Both the first and second automated vehicles can be understood as automated vehicles executing the trip. The first automated vehicle can refer to the automated vehicle the passenger is riding in before the determination of the passenger's intention to ride; the second automated vehicle specifically refers to the automated vehicle dispatched when the passenger has a need for another ride. The period between the passenger entering and leaving the first automated vehicle can be considered the execution of the trip, and for the passenger currently riding in the first automated vehicle, the passenger leaving the first automated vehicle can be considered the end of the current trip.
[0041] In some embodiments, the presence or absence of an object can be detected through various methods such as image detection and sensor detection. For example, object presence detection of in-vehicle images can detect whether an object exists inside the first autonomous vehicle; changes in detection data from weight sensors or infrared sensors can detect whether an object exists inside the first autonomous vehicle; and information such as the door opening and closing status and frequency detected by door opening and closing sensors can detect whether an object exists inside the first autonomous vehicle.
[0042] Environmental perception data includes data that influences the behavior of objects. Environmental perception data can include data perceived around and inside the first autonomous vehicle, such as in-vehicle environmental perception data and external environmental perception data. Environmental perception data can be collected through sensors installed in the first autonomous vehicle or sensors in other autonomous vehicles. For example, in-vehicle and external environmental perception data can be collected using sensors on the first autonomous vehicle; external environmental perception data can also be collected using sensors from other autonomous vehicles.
[0043] Passenger intent can indicate whether a person has a need to travel again. It is understandable that after a passenger disembarks, various environmental factors may influence their behavior, causing them to change course.
[0044] For example, a passenger may change their original travel intention after disembarking, such as changing their original intention to get off the bus or to re-board. This situation is often sudden and may be influenced by various environmental factors such as severe weather, unexpected traffic incidents, or lost items. Therefore, embodiments of this disclosure can analyze environmental perception data to determine the travel intention affected by environmental factors.
[0045] Environmental perception data can be analyzed in various ways to determine whether an individual might have a need to take the vehicle again due to the influence of the environmental perception data, thus obtaining the individual's travel intention. For example, environmental perception data can be analyzed using at least one of rule matching, intention recognition, and model prediction.
[0046] The pick-up instruction may include identification information of the second autonomous vehicle to indicate which second autonomous vehicle is intended to carry passengers who require a second ride. In addition, the pick-up instruction may also include various information such as pick-up priority, current traffic conditions, estimated pick-up time, and estimated waiting time.
[0047] If the person indicating a ride intention has a need for another ride, a pick-up instruction can be generated directly without waiting for the person's active command or feedback, and a second autonomous vehicle can be dispatched to pick them up using the pick-up instruction. If the person indicating a ride intention does not have a need for another ride, then there is no need to generate a pick-up instruction. In some embodiments, the second autonomous vehicle can be determined and the pick-up instruction generated using environmental perception data and / or the current vehicle dispatch situation.
[0048] Understandably, the second autonomous vehicle can be different from the first autonomous vehicle; the second autonomous vehicle can also be the same as the first autonomous vehicle. If there are no other autonomous vehicles near the passenger, the first autonomous vehicle will be the fastest autonomous vehicle to pick up the passenger.
[0049] In the embodiments of this disclosure, after a passenger disembarks and ends their current journey, the system can proactively perceive the passenger's status and travel intentions using the environmental perception data from the first autonomous vehicle used for the current journey. If the passenger requests another ride, a second autonomous vehicle can be dispatched to assist. Throughout the process, the reuse of environmental perception data from the current journey, proactive determination of travel intentions, proactive generation of pick-up instructions, and proactive vehicle dispatching, all automated, simulate a real driver proactively providing service. Even in emergency situations, assistance can be proactively offered, improving the travel experience. Furthermore, the first autonomous vehicle does not need to wait for the passenger's response after disembarking, thus increasing the utilization rate of the autonomous vehicle.
[0050] Figure 3The illustration schematically depicts a scenario of a ride-hailing method based on an autonomous vehicle according to an embodiment of the present disclosure. For example... Figure 3 As shown, object 302 can be passenger A. After object 302 disembarks from the first autonomous vehicle 301, the first autonomous vehicle 301 or the server analyzes the environmental perception data and discovers that the entrance to a building is 100m ahead of the disembarkation location. At this point, it can be determined that the passenger's intention to ride indicates a need for another ride. The first autonomous vehicle 301 or the server can automatically generate a pick-up instruction and dispatch a second autonomous vehicle 303 located near the disembarkation location. In this scenario, a notification can be pushed to object 302's terminal device, or the second autonomous vehicle 303 can verbally announce to object 302, "The entrance to a building is 100m ahead. I'll accelerate and take you to the entrance," to mimic a real driver's observation of the environment and proactive service. Object 302 can respond to the second autonomous vehicle 303, such as through the terminal device or directly, whether they need to continue riding. For passengers who respond to requests to ride again, the second autonomous vehicle 303, which has already been dispatched (e.g., has arrived at the drop-off location according to the pick-up instructions), can pick up the passenger and proceed to the building entrance; for passengers who respond to requests that they do not need to ride again, the second autonomous vehicle 303, which has already been dispatched, can end the current dispatch and complete other journeys.
[0051] In one specific embodiment, environmental perception data can be collected directly by sensors after the passenger disembarks. In another specific embodiment, the world model used by the autonomous vehicle for driving decisions can also be reused. A world model is a neural network system capable of modeling environmental dynamics; given the current state and actions, it can predict the evolution of future states. For example, although the first autonomous vehicle's current journey for the passenger ends after disembarking, the first autonomous vehicle still needs to make driving decisions. This means that even after the passenger disembarks, the first autonomous vehicle continues to collect data and use it as input to the world model to obtain environmental perception data through scene reconstruction. This environmental perception data can be used not only for the first autonomous vehicle's own decision-making but also to determine the passenger's intention in this embodiment of the disclosure.
[0052] The world model can store all data collected by the autonomous vehicle during the trip, as well as the vehicle's decision-making information. After a passenger disembarks, the data stored in the world model can be directly retrieved to reconstruct the surrounding environment, disembarkation location, and passenger movement trajectory. This eliminates the need to re-collect current environmental perception data through sensors; the model can directly determine the passenger's intention to travel based on the existing model data and the potential impact on their behavior upon disembarkation. For example, scene reconstruction can detect whether a passenger remains in place due to sudden physical discomfort or lingers around the disembarkation point instead of heading to their destination due to a sudden change in weather.
[0053] According to embodiments of this disclosure, the environmental perception data includes external environmental perception data of the area outside the vehicle where the object is located; analyzing the environmental perception data of the first autonomous vehicle to obtain the object's riding intention includes: performing environmental analysis on the external environmental perception data to obtain environmental analysis results, wherein the environmental analysis results are used to indicate whether the environment outside the vehicle affects the object's behavior; if it is determined that the environmental analysis results indicate that the environment outside the vehicle affects the object's behavior, determining a riding intention to indicate that the object has a need to ride again.
[0054] The area outside the vehicle where the object is located can refer to the area within a preset range around the passenger's alighting point. The preset range can be pre-defined based on the perception coverage of the vehicle's sensors, road traffic safety standards, and other benchmarks. For example, in a typical urban road scenario, the area outside the vehicle can be set as the roadside area within a 5-meter radius centered on the passenger's alighting point.
[0055] External environment perception data may include at least one of the following: image data, meteorological data, lighting data, obstacle data, and safety data of the area outside the vehicle. Environmental analysis of the external environment perception data can be determined based on the type of data contained in the sample environmental perception data, or it can be determined based on preset abnormal events that can affect passenger behavior.
[0056] For example, meteorological data can be directly collected by the meteorological sensors of the first autonomous vehicle, or obtained by image analysis of image data collected by visual perception sensors. Meteorological data can be used to determine whether the current weather conditions will affect passengers staying in the area outside the vehicle, such as sudden rainfall or abnormal temperatures. Illumination data can be directly collected by the photosensors on the first autonomous vehicle, or obtained by brightness feature analysis of the image data collected by the first autonomous vehicle. Illumination data can be used to determine the visibility of the area outside the vehicle where the passenger is currently located, such as whether it is a low-visibility environment at night, or whether there is strong direct sunlight or large-scale backlighting causing poor visibility, thus determining whether the area outside the vehicle is unsuitable for passengers to stay for an extended period. Obstacle data can be directly collected by infrared sensors, or obtained by obstacle detection of image data. Obstacle data can be used to determine whether the area outside the vehicle where the passenger is located obstructs passage, such as whether they are staying in a pedestrian-restricted area or an unsafe area. Safety data for the area outside the vehicle can be determined by analyzing image data and / or the disembarkation location to determine whether there are factors outside the vehicle that could affect passenger safety.
[0057] The environmental analysis results correspond to the environmental analysis of the perceived external environment data. For example, the environmental analysis results can be determined based on whether the current weather will affect passengers in the area outside the vehicle, the visibility conditions in the area where passengers are currently located, whether there are any obstacles to passage in the area where passengers are located, and whether there are any factors in the area outside the vehicle that affect passenger safety. If the current weather will affect passengers, visibility will affect passengers, the area outside the vehicle will affect passenger passage, or there are any factors that affect passenger safety, then the environmental analysis results indicate that the environment outside the vehicle will affect the behavior of the passengers; otherwise, the environmental analysis results indicate that the environment outside the vehicle will not affect the behavior of the passengers.
[0058] If the environment influences passenger behavior, it can be determined that the passenger's intention to take the ride is to have a need for another ride. If the environment does not influence passenger behavior, it can be determined that the passenger's intention to take the ride is not to have a need for another ride, and there is no need to generate a pick-up instruction.
[0059] In the embodiments of this disclosure, by analyzing the perception data of the external environment after a passenger gets off the vehicle, it is possible to determine whether the passenger will be affected by the external environment after getting off the vehicle, which may lead to the passenger having a need to take the vehicle again. This accurately identifies the need to take the vehicle again due to the temporary change of the passenger's itinerary caused by the external environment. Without relying on the passenger to actively initiate a call request, the pick-up service is automatically triggered, reducing the safety hazards of passengers waiting outside the vehicle to call a car in adverse environments, and improving the user service experience and intelligent service level of autonomous vehicles.
[0060] According to embodiments of this disclosure, environmental analysis is performed on external environment perception data to obtain environmental analysis results, including: identifying weather types from the external environment perception data to obtain weather types; when the weather type is a first type, determining environmental analysis results to indicate the behavior of objects affected by the environment in the external area, wherein the first type includes types in which the affected objects take evasive actions; when the weather type is a second type, determining environmental analysis results to indicate that the environment in the external area does not affect the behavior of objects, wherein the second type includes types in which the affected objects do not take evasive actions.
[0061] Weather type is used to indicate weather conditions in the area outside the vehicle. For example, weather type can be identified from image data collected by visual perception sensors; alternatively, weather type can be identified from meteorological data such as lidar point cloud data, temperature, humidity, visibility, and precipitation intensity. In some embodiments, weather type identification is achieved using a pre-trained environmental classification neural network model or a large model.
[0062] The first category can include severe weather types, such as heavy rain, torrential rain, snowfall, thunderstorms, strong winds, low visibility, extreme high temperatures, and extreme low temperatures. These typically refer to weather types that cause discomfort to passengers and require them to take shelter. The second category can include various weather types other than the first, such as cloudy skies, sunny skies, light winds, and light rain—weather types that do not significantly affect passengers' normal travel and usually do not require passengers to change their planned walking routes after disembarking. It is understandable that the above weather types can also be a combination of multiple meteorological dimensions, such as high temperatures and heavy rain, or low temperatures and strong winds.
[0063] It should be noted that the first and second types mentioned above can be determined based on historical statistical data. For example, if the number of orders placed on a ride-hailing platform increases under a certain weather type, then that weather type can be identified as the first type.
[0064] In this embodiment, by identifying the weather type of the external environment perception data, the actual weather conditions after the passenger gets off the vehicle can be determined. Based on this, it can be determined whether the current weather type will affect the passenger's behavior, thereby proactively providing vehicle services to the passenger based on the weather. This makes the service logic of autonomous vehicles more in line with the dynamic travel needs of passengers in special weather conditions, avoids passengers being exposed to the outdoors for a long time due to severe weather, and improves the user experience of autonomous driving travel services.
[0065] According to embodiments of this disclosure, environmental analysis is performed on vehicle exterior environment perception data to obtain environmental analysis results, including: performing safety analysis on vehicle exterior environment perception data to obtain safety data; wherein, the safety data is used to indicate the safety level of the vehicle exterior area; and when the safety data meets abnormal safety conditions, the environmental analysis results used to indicate the behavior of environmentally impacted objects in the vehicle exterior area are determined.
[0066] Safety data includes at least one of the following: illumination in the area outside the vehicle, frequency of traffic accidents, construction area, terrain data, and noise. Abnormal safety conditions may include a first abnormal safety sub-condition, a second abnormal safety sub-condition, a third abnormal safety sub-condition, a fourth abnormal safety sub-condition, and a fifth abnormal safety sub-condition, respectively, corresponding to the illumination in the area outside the vehicle, the frequency of traffic accidents, the construction data, the terrain data, and the noise data.
[0067] For example, the first abnormal safety sub-condition could be that the light intensity outside the vehicle is less than the light intensity threshold, indicating that the area outside the vehicle is at night or in a low-visibility environment, potentially posing a safety risk to passengers due to the inability to see the road clearly. The second abnormal safety sub-condition could be that the frequency of traffic accidents is higher than the frequency threshold, indicating that the area outside the vehicle is an accident-prone area, and passengers disembarking there face the risk of traffic accidents. The third abnormal safety sub-condition could be that the area outside the vehicle is a construction zone, indicating that the potential safety risks of the construction zone may affect passenger safety. The fourth abnormal safety sub-condition could be that the slope of the terrain outside the vehicle is greater than the slope threshold, indicating that the terrain outside the vehicle is uneven, posing a risk of falls and other safety hazards to passengers walking on it. The fifth abnormal safety sub-condition could be that the noise level is greater than the noise threshold, indicating that the area outside the vehicle is adjacent to a noise source, and passengers staying there after disembarking may be significantly affected by noise, failing to meet the requirements for a comfortable riding experience. When any of the safety data meets the corresponding abnormal safety condition, the environmental analysis results used to indicate the behavior of objects affected by the environmental impact of the area outside the vehicle can be determined.
[0068] In the embodiments of this disclosure, by performing safety area analysis on the vehicle's external environment perception data, safety data is obtained. Based on various safety data affecting passenger safety, the system can autonomously determine the passenger's intention to ride and proactively provide vehicle services to the passenger. This makes the service logic of autonomous vehicles more aligned with the passenger's safe travel needs and improves the user experience of autonomous driving travel services.
[0069] In one specific embodiment, security analysis can also be performed on image data in the environmental perception data to obtain state data indicating the safety status of an object. For example, a large model can be used to perform semantic analysis on the image data, extract passenger action features, and determine the passenger's state based on these features to obtain state data. State data can be used to indicate whether a passenger is unwell or has fallen; both falling and unwell conditions can be considered abnormal states. When the state data indicates that the corresponding object is in an abnormal state, the passenger's intention is to have a need for another ride. The first autonomous vehicle or the server can automatically generate a pick-up instruction to dispatch a second autonomous vehicle to pick them up. In this embodiment, if the second autonomous vehicle arrives at the pick-up location (e.g., the passenger's drop-off point after leaving the first autonomous vehicle) according to the pick-up instruction, and the environmental perception data detects that the passenger is still in an abnormal state, an alarm can be triggered through interaction with the passenger or directly, proactively providing service without receiving a request for assistance from the passenger.
[0070] According to embodiments of this disclosure, environmental perception data includes the disembarkation location of an object leaving the first autonomous driving vehicle; analyzing the environmental perception data of the first autonomous driving vehicle to obtain the object's riding intention includes: obtaining a predetermined destination location of the first autonomous driving vehicle, wherein the predetermined destination location is determined based on the object's interactive operation; comparing the predetermined destination location and the disembarkation location to obtain a first comparison result, wherein the first comparison result is used to indicate whether the predetermined destination location and the disembarkation location are the same; and determining the riding intention based on the first comparison result.
[0071] The intended destination is set by the passenger when they initiate the ride-hailing request, for example, through an interactive interface such as a ride-hailing app. The intended destination can also be adjusted or modified at any time during the ride via the in-cabin interface within the first autonomous vehicle.
[0072] The first autonomous vehicle can store trip data, which may include various types of information, such as the planned start point, planned destination, and waypoints. The first autonomous vehicle can directly obtain the planned destination data from the trip data.
[0073] During the journey, passengers may alight at a location different from their intended destination due to unforeseen emergencies or temporary traffic disruptions. Therefore, comparing the intended destination with the actual alighting location can help analyze whether there are any abnormal circumstances surrounding a passenger's early disembarkation.
[0074] The first comparison result indicates whether the planned destination and the drop-off location are the same. If the first comparison result indicates that the planned destination and the drop-off location are the same, it means that the current trip has been completed according to the passenger's original plan, and the passenger does not have a need to take the bus again; conversely, if the first comparison result indicates that the planned destination and the drop-off location are different, it means that the passenger did not proceed according to the original planned itinerary and may have a need to take the bus again.
[0075] In one specific embodiment, if the difference between the predetermined destination location and the drop-off location exceeds a distance threshold, such as 50m, it is determined that the predetermined destination location and the drop-off location are different; conversely, if the difference between the predetermined destination location and the drop-off location does not exceed the distance threshold, it is determined that the predetermined destination location and the drop-off location are the same.
[0076] In the embodiments of this disclosure, by comparing the actual drop-off location with the predetermined destination location, it is possible to determine whether a passenger has a need to take the ride again based on whether the passenger gets off at the predetermined location. In turn, based on whether the passenger gets off at the predetermined destination location, ride services can be proactively provided to improve the ride experience.
[0077] According to embodiments of this disclosure, the environmental perception data includes in-vehicle environmental perception data of a first autonomous vehicle; analyzing the environmental perception data of the first autonomous vehicle to obtain the passenger's intention to ride includes: comparing the in-vehicle environmental perception data at a first moment and the in-vehicle environmental perception data at a second moment to obtain a second comparison result, wherein the first moment and the second moment are the moments when the passenger enters and leaves the first autonomous vehicle, respectively, and the second comparison result is used to indicate whether there are any different items inside the first autonomous vehicle; and determining the passenger's intention to ride based on the second comparison result.
[0078] In-vehicle environmental perception data can be collected through cameras inside the first autonomous vehicle; for example, the data can be image data. The first moment is when the passenger enters the vehicle, at which time the in-vehicle environmental perception data will include information about the passenger's personal belongings. The second moment is when the passenger leaves the vehicle, at which time the data may include information about items left behind. Item information can be extracted from the in-vehicle environmental perception data of the first and second moments respectively, and the item information from the two moments can be compared to obtain a second comparison result.
[0079] The second comparison result is used to indicate whether a passenger has left any items in the vehicle. For example, if the second comparison result indicates that the information of two items is different, and the item information at the second moment includes a preset item, it can be determined that the passenger has left any items in the vehicle, and thus determine the passenger's intention to take the vehicle again. In fact, if there are items left by a passenger in the first autonomous vehicle, the passenger is likely to need to take the vehicle again to find the forgotten items. Therefore, in this case, a pick-up instruction can be automatically generated and a second autonomous vehicle can be dispatched to facilitate the passenger's subsequent retrieval. In this embodiment, the second autonomous vehicle can be the same as the first autonomous vehicle, that is, the same vehicle is used to return lost items; the second autonomous vehicle can also be different from the first autonomous vehicle. In this case, when the first autonomous vehicle is being cleaned manually, the second autonomous vehicle can be dispatched to the cleaning location first, and the lost items can be transferred manually at the cleaning location; then, the second autonomous vehicle can pick up the passenger carrying the lost items. The second autonomous vehicle can go to the cleaning location immediately and return the lost items immediately, or it can go to the cleaning location immediately and return the lost items according to the dispatch order.
[0080] If the second comparison result indicates that the two item information is the same, or if the second comparison result indicates that the two item information is different and the item information at the second moment does not include the preset item, it is determined that the passenger did not leave any items in the vehicle, and therefore it is determined that the passenger does not have the intention to travel again. The preset item can be a mobile phone, suitcase, bag, umbrella, or other items belonging to the luggage category; for items belonging to the garbage category, such as packaging bags, even if the item information at the two moments is different, the passenger's intention to travel can be ruled out by the fact that the item information at the second moment does not include the preset item.
[0081] In this embodiment, by comparing the in-vehicle environmental perception data at the time of passenger boarding and alighting, it is possible to determine whether the passenger has left any items behind, and thus accurately determine whether the passenger needs to take the vehicle again to retrieve the items. This eliminates the need for passengers to actively report or for manual verification of information about lost items, reducing the cost of manual intervention in the operation process and allowing scheduling preparations to be completed in advance before the passenger initiates a request, reducing the response time for subsequent pick-ups, and improving the efficiency of retrieving lost items and the service experience of autonomous vehicles.
[0082] Figure 4 A flowchart illustrating a specific embodiment of a ride-hailing method based on an autonomous vehicle according to this disclosure is shown schematically. Figure 4 As shown, this embodiment includes operations S410 to S490.
[0083] Operating the S410 to collect data.
[0084] When operating the S420, scene reconstruction is performed based on the world model.
[0085] Operation S430 acquires environmental perception data. For example, operation S410 can collect data during a passenger's ride in the first autonomous vehicle and after the passenger disembarks. After the passenger disembarks, this data is used to reconstruct the scene, and the reconstructed scene is used to acquire environmental perception data.
[0086] In operation S440, it is determined whether the disembarkation position is the same as the predetermined destination position. If the disembarkation position is the same as the predetermined destination position, operation S450 is executed; otherwise, operation S470 is executed.
[0087] In operation S450, it is determined whether the external environment necessitates another ride. For example, based on the environmental analysis results, it can be determined whether another ride is necessary. If no further ride is required, operation S460 is executed; otherwise, operation S470 is executed.
[0088] In operation S460, it is determined whether there are any items left behind by the passenger in the vehicle. For example, the presence of items left behind by the passenger can be determined using in-vehicle environmental perception data. If so, operation S470 is executed; otherwise, it indicates that multiple judgment conditions of operations S430 to S470 are not met, and it is ultimately determined that the passenger does not have the intention to ride again, and operation S490 is executed.
[0089] The S470 is operated to generate a pick-up command and dispatch a second autonomous vehicle.
[0090] When operating S480, push a pick-up notification to the target.
[0091] Operation S490, end.
[0092] It should be noted that the above embodiments can be used in combination to determine whether a passenger has a need to take the ride again from multiple dimensions. For example, if any of the following conditions are met, the intention to take the ride again can be determined: environmental analysis results indicate the behavior of objects influencing the environment outside the vehicle; the first comparison result indicates that the predetermined destination location is different from the alighting location; and the second comparison result indicates that the information of the two items is different and the item information at the second moment includes the preset item. Conversely, if the passenger's intention to take the ride again is determined to be non-existent when the external environmental perception data, the internal environmental perception data, and the alighting location are all combined, the final determination is that the passenger does not have a need to take the ride again. Specific conditions can be found above and will not be repeated here.
[0093] In one specific embodiment, passenger B booked a first autonomous vehicle at night (22:30), with the intended destination being the east gate of XX residential area. During the journey, passenger B changed their mind and disembarked at an intersection approximately 300 meters from the intended destination. While the first autonomous vehicle was operating (including the passenger's time inside and after disembarking), it continuously collected surround-view images, LiDAR point clouds, and vehicle pose data. The world model recorded data in real time, including: vehicle position, speed, and steering angle at each timestamp; surround-view camera image sequences (including keyframes from the second moment after disembarking); and the disembarking position (a deviation of 350 meters was found compared to the intended destination).
[0094] After passenger B disembarks, the first autonomous vehicle completes its current journey and automatically prepares for the next trip. Simultaneously, the first autonomous vehicle or the server can load data and reconstruct the scene using a world model. For example, the reconstructed scene is a 3D environmental model of the disembarkation location, including: a T-junction with no surrounding buildings or streetlights; illumination below the 0.5 lux threshold; and the passenger's walking direction opposite to the intended destination. Based on the reconstructed scene, environmental perception data is obtained, and the following judgments are made: if the deviation between the disembarkation location and the intended destination is 350m (> 50m threshold), there is a risk of disembarking at an unintended location; if the illumination is 0.5 lux (< 5 lux threshold), there is a risk of disembarking in an unsafe area. Both anomalies can determine that the passenger has a need for another ride, thus generating a pick-up instruction without any passenger intervention. For example, the nearest available second autonomous vehicle can be dispatched, such as one approximately 2.3km from the disembarkation location and expected to arrive in 5 minutes, and a notification can be pushed to the passenger's terminal: "We detected that you disembarked in an unsafe area; a vehicle has been arranged to pick you up."
[0095] In another embodiment, passenger C travels to a high-speed rail station in a first autonomous vehicle, arriving during a rainstorm. Passenger C hurriedly disembarks, leaving their handbag on the back seat. Using a scenario reconstruction and passenger intent determination method similar to the above embodiment, the following judgments can be made: visibility <50m, weather type is type 1 (e.g., rainstorm), indicating disembarkation due to severe weather; a second comparison result indicates that the item information at the two times is different, and the item information at the second time includes a preset item (e.g., the handbag), indicating that the passenger left something behind. Both anomalies confirm that the passenger has a need to travel again, thus a pick-up command can be generated to dispatch a second autonomous vehicle and push a notification to the passenger's terminal: "You have left something in the vehicle, and the current weather is severe. A vehicle has been arranged to return your item and pick you up to shelter from the rain." This embodiment automatically decides and executes the pick-up operation after detecting the dual anomalies of severe weather and left-behind items, without waiting for the passenger's response.
[0096] In another embodiment, after passenger D arrives at the predetermined destination at night, the first autonomous vehicle completes its formation and departs. Through post-event scene reconstruction and backtracking of the world model, it can be confirmed whether passenger D is in a safe area or has entered a building based on safety data. If an anomaly is detected, such as the passenger not entering a building or being approached by a stranger, a pick-up instruction for a second ride can be automatically generated and a second autonomous vehicle can be dispatched to pick up the passenger.
[0097] In the embodiments of this disclosure, by comprehensively analyzing multi-dimensional environmental perception data and determining passenger intent across multiple dimensions, it is possible to more comprehensively and accurately capture passengers' potential repeat travel needs, avoiding omissions or misjudgments due to single-dimensional judgment, thereby improving the passenger travel experience. Furthermore, since the first autonomous vehicle does not need to wait for passenger response after disembarking, determining passenger intent across multiple dimensions does not increase the waiting time of the first autonomous vehicle, thus improving the utilization rate of the autonomous vehicle.
[0098] According to an embodiment of this disclosure, determining a passenger's intention to ride a vehicle based on a first comparison result includes: in response to determining that the first comparison result indicates a difference between a predetermined destination location and a drop-off location, determining the object's direction of movement based on an object behavior image in environmental perception data; and determining the object's intention to ride a vehicle based on the relationship between the direction of movement and the predetermined destination location.
[0099] Considering that it is common for passengers to get off at locations other than the intended destination, and that passengers have various intentions, when the intended destination and the drop-off location are different, other data in the environmental perception data can be combined to further accurately determine the passenger's intention, so as to avoid misjudgment of the passenger's intention and avoid affecting the vehicle dispatching efficiency.
[0100] The object behavior image can be an image sequence of the object's actions after it gets out of the vehicle. For example, the object behavior image can be image data from the vehicle's external environment perception data, or it can be a partial image data including the object extracted from the aforementioned image data.
[0101] Passenger behavior features can be extracted from images of passenger behavior, and the direction of passenger movement can be determined based on these features. After determining the direction of movement, the image coordinate system can be matched with the location coordinate system to determine the relationship between the mode of movement and the predetermined destination.
[0102] The relationship between the direction of movement and the predetermined endpoint position can include: the direction of movement is towards the predetermined endpoint position, or the direction of movement is away from the predetermined endpoint position.
[0103] In one specific embodiment, if the direction of movement is closer to the predetermined destination, it indicates that the passenger's current journey may still end at the original predetermined destination, and they will need to travel to that destination again. Therefore, it can be determined that the passenger has a need for another ride. In this case, the first autonomous vehicle can be selected as the second autonomous vehicle to connect the current journey for a second ride.
[0104] In another specific embodiment, if the direction of movement is away from the predetermined destination, it indicates that the destination of the passenger's current trip has changed and the passenger may no longer need to go to the predetermined destination. Therefore, it can be determined that the passenger does not have a need to take the ride again.
[0105] In this embodiment, when the predetermined destination and the drop-off location are different, the relationship between the passenger's actual direction of movement and the predetermined destination is combined to infer the potential correlation between the passenger's actual drop-off behavior and the predetermined destination. This can more accurately identify the passenger's real need for another ride, further reduce the ineffective scheduling of autonomous vehicles, and improve the overall efficiency and user experience of autonomous driving travel services.
[0106] In another specific embodiment, determining the passenger's travel intention based on the relationship between the direction of movement and the predetermined destination location includes: in response to determining that the direction of movement is a direction close to the predetermined destination location, determining a travel intention to indicate that the passenger has a need to travel again; in response to determining that the direction of movement is a direction away from the predetermined destination location, determining the passenger's travel intention based on the relationship between the direction of movement and associated buildings and the entrance traffic data of the associated buildings, wherein the associated buildings include buildings associated with the drop-off location.
[0107] The associated building can be a building within a preset range centered on the drop-off location. Entrance traffic data is used to indicate the entrance access requirements of the associated building. Entrance traffic data can include at least one of the following: at least one entrance to the associated building, the location of the entrance, and the vehicle access requirements and pedestrian communication requirements corresponding to each entrance. For example, if the associated building is a school, the entrance traffic data could include: the school has an east gate, a south gate, a west gate, and a north gate; the east gate is closed to vehicles; and the west gate is near the school's parking lot and XX Road.
[0108] When the direction of movement is away from the intended destination, the passenger's travel intention may no longer be related to the intended destination, but rather to the facilities around the current drop-off location. For example, if the passenger wants to go to a related building near the drop-off location, their movement direction will be towards that related building, and entrance traffic data can help determine the traffic demand at the entrance the passenger is heading to.
[0109] Based on the relationship between the direction of movement and associated buildings, as well as the entrance traffic data of the associated buildings, determining the passenger's intention to travel can include: if the direction of movement is towards the associated buildings, the entrance to which the direction of movement is directed and the entrance traffic data can be determined, and the passenger's intention to travel can be determined based on the above information.
[0110] Figure 5 The illustration schematically depicts a passenger pick-up and drop-off scenario involving associated buildings according to a specific embodiment of this disclosure. For example... Figure 5 As shown, passenger 502 can disembark from the first autonomous vehicle 501 at the intersection, and the associated building 503 is a school. If the passenger's direction of movement is P1, and the traffic data at entrance A indicates that vehicles are not allowed to enter, then the passenger may walk into the associated building without generating a new ride request, thus confirming that they do not have a ride-hailing intention. If the passenger's direction of movement is P2, and the traffic data at entrance B indicates that vehicles are allowed to enter, then it can be predicted in advance that the passenger may have a ride-hailing intention to enter the associated building again, confirming their ride-hailing intention and generating a pick-up instruction autonomously, thus completing vehicle dispatch preparation in advance.
[0111] In another specific embodiment, if there are multiple buildings associated with the drop-off location, that is, when there are multiple associated buildings, the direction of movement of the associated building can be selected from the multiple associated buildings, and then the entrance traffic data of the associated building can be used to determine the intention to take the bus. For specific operations, please refer to the above, and will not be repeated here.
[0112] In the embodiments of this disclosure, when the direction of movement is away from the predetermined destination, by introducing the relationship between the associated building and the direction of movement, as well as the entrance traffic data of the associated building, it is possible to combine the passenger's actual drop-off location with the surrounding real environment and the passenger's behavioral interaction with the real environment, and determine whether the environment affects the passenger's travel intention based on the environment. This improves the accuracy of travel intention judgment based on real behavior, reduces invalid pick-up and drop-off scheduling caused by misjudgment, and improves the operational efficiency and user experience of autonomous driving travel services.
[0113] According to embodiments of this disclosure, when the person indicating the intention to ride has a need to ride again, a pick-up instruction is generated, including: acquiring traffic data at the drop-off location, pose data and trip data of at least one autonomous vehicle, wherein the at least one autonomous vehicle includes a first autonomous vehicle; determining a second autonomous vehicle from the at least one autonomous vehicle based on the traffic data, pose data and trip data; and generating a pick-up instruction based on the identification information of the second autonomous vehicle and the drop-off location.
[0114] Traffic data is used to indicate the current traffic status and passage conditions at the drop-off location. For example, traffic data may include traffic flow data, parking space data, temporary traffic control data, road type, whether U-turns are permitted, and distance to intersections around the drop-off location.
[0115] Position data includes the autonomous vehicle's current position, direction of travel, vehicle posture, and lane information. Trip data includes data required and recorded by the autonomous vehicle to execute the current trip, such as the predetermined start point, predetermined destination, and operating status (e.g., whether it is idle).
[0116] At least one autonomous vehicle may include autonomous vehicles around the drop-off location, and at least one autonomous vehicle includes at least the first autonomous vehicle.
[0117] Based on traffic data, the current lane information and traffic conditions for the current time period can be determined, and vehicles that do not meet the current traffic conditions and / or are not in an idle state can be removed from at least one autonomous vehicle to obtain at least one candidate vehicle; based on the pose data of each candidate vehicle, the travel distance of the candidate vehicle to the drop position can be calculated, including the additional distance caused by the U-turn operation; the autonomous vehicle with the shortest travel distance can be determined as the second autonomous vehicle.
[0118] For example, if a passenger needs to pick up another passenger, and the first autonomous vehicle is located on a one-way street where U-turns and stops are not allowed, the autonomous vehicle located on the opposite side of the road, which needs to cross several medians to make a U-turn to reach the drop-off point, will travel a longer distance than the autonomous vehicle located on the same-direction road. Therefore, if there is an autonomous vehicle on the same-direction road, it can be used as the second autonomous vehicle.
[0119] After identifying the second autonomous vehicle, a pick-up instruction can be generated by combining the identification information of the second autonomous vehicle and the drop-off location. In other embodiments, the pick-up instruction may also include the location for determining the intent, such as when determining the passenger's intent, the passenger has already moved 20m relative to the drop-off location; the pick-up instruction may also include information on the basis for determining the passenger's intent, such as weather type, safety data, external environment perception data, internal environment perception data, drop-off at a non-predetermined location, drop-off in inclement weather, drop-off in an unsafe area, or items left behind by the passenger.
[0120] In the embodiments of this disclosure, by combining pose data, traffic data, and trip data, a second autonomous vehicle with a better position for the current drop-off location can be quickly determined. This ensures that the vehicle can successfully complete the pick-up and drop-off task while minimizing passenger waiting time and improving vehicle dispatching efficiency and passenger experience for autonomous driving travel services.
[0121] Figure 6 A block diagram of a ride-hailing device based on an autonomous vehicle according to an embodiment of the present disclosure is shown schematically. Figure 6 As shown, the ride-hailing device 600 based on an autonomous vehicle includes: an analysis module 610, used to analyze the environmental perception data of the first autonomous vehicle in response to detecting that a person has left the first autonomous vehicle, to obtain the person's ride intention; wherein, the environmental perception data includes data affecting the person's behavior; a generation module 620, used to generate a ride-hailing instruction when the ride intention indicates that the person has a need to ride again, wherein the ride-hailing instruction is used to instruct a second autonomous vehicle to pick up the person; and a scheduling module 630, used to send the ride-hailing instruction to the second autonomous vehicle to schedule the second autonomous vehicle to pick up the person.
[0122] The environmental perception data includes the external environmental perception data of the area outside the vehicle where the object is located. The analysis module 610 includes: a first analysis submodule, used to perform environmental analysis on the external environmental perception data to obtain environmental analysis results, wherein the environmental analysis results are used to indicate whether the environment outside the vehicle affects the object's behavior; and a first determination submodule, used to determine, when the environmental analysis results indicate that the environment outside the vehicle affects the object's behavior, an intention to take the vehicle that indicates the object has a need to take the vehicle again.
[0123] According to an embodiment of this disclosure, the first analysis submodule includes: a first identification unit, configured to identify the weather type of the external environment perception data to obtain the weather type; a second identification unit, configured to determine an environmental analysis result indicating the behavior of an object affected by the environment in the external area when the weather type is a first type, wherein the first type includes the type in which the affected object makes an avoidance behavior; and a third identification unit, configured to determine an environmental analysis result indicating that the environment in the external area does not affect the behavior of an object when the weather type is a second type, wherein the second type includes the type in which the object does not make an avoidance behavior.
[0124] According to an embodiment of this disclosure, the first analysis submodule includes: a first analysis unit, used to perform safety analysis on the vehicle external environment perception data to obtain safety data; wherein the safety data is used to indicate the safety level of the vehicle external area; and a second analysis unit, used to determine the environmental analysis result used to indicate the behavior of the environmental impact object in the vehicle external area when the safety data meets abnormal safety conditions.
[0125] According to embodiments of this disclosure, environmental perception data includes the disembarkation location of an object leaving the first autonomous driving vehicle; the analysis module 610 includes: an acquisition submodule for acquiring a predetermined destination location of the first autonomous driving vehicle, wherein the predetermined destination location is determined based on the object's interactive operation; a first comparison submodule for comparing the predetermined destination location and the disembarkation location to obtain a first comparison result, wherein the first comparison result is used to indicate whether the predetermined destination location and the disembarkation location are the same; and a first determination submodule for determining the riding intention based on the first comparison result.
[0126] According to an embodiment of this disclosure, the first determining submodule includes: a first determining unit, configured to determine the moving direction of an object based on an object behavior image in environmental perception data in response to determining that the first comparison result indicates that the predetermined destination location and the alighting location are different; and a second determining unit, configured to determine the object's travel intention based on the relationship between the moving direction and the predetermined destination location.
[0127] According to an embodiment of this disclosure, the second determining unit includes: a first determining subunit, configured to determine, in response to determining that the direction of movement is a direction close to a predetermined destination, an intention to ride a vehicle indicating that the subject has a need to ride again; and a second determining subunit, configured to determine, in response to determining that the direction of movement is a direction away from the predetermined destination, the subject's intention to ride a vehicle based on the relationship between the direction of movement and associated buildings and entrance traffic data of the associated buildings, wherein the associated buildings include buildings associated with the alighting location.
[0128] According to embodiments of this disclosure, the environmental perception data includes in-vehicle environmental perception data of the first autonomous vehicle; the analysis module 610 includes: a second comparison submodule, used to compare the in-vehicle environmental perception data at a first moment and the in-vehicle environmental perception data at a second moment to obtain a second comparison result, wherein the first moment and the second moment are the moments when the object enters and leaves the first autonomous vehicle, respectively, and the second comparison result is used to indicate whether there are different items in the first autonomous vehicle; and a second determination submodule, used to determine the riding intention based on the second comparison result.
[0129] According to an embodiment of this disclosure, the generation module 620 includes: a data acquisition submodule, used to acquire traffic data at the drop-off location, pose data and trip data of at least one autonomous vehicle, wherein the at least one autonomous vehicle includes a first autonomous vehicle; a vehicle determination submodule, used to determine a second autonomous vehicle from the at least one autonomous vehicle based on the traffic data, pose data and trip data; and a generation submodule, used to generate a pick-up instruction based on the identification information of the second autonomous vehicle and the drop-off location.
[0130] Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure, or at least part of the functions of any one or more of them, can be implemented in one module. Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be implemented by dividing them into multiple modules. Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be at least partially implemented as hardware circuitry, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a System-on-Chip, a System-on-a-Substrate, a System-on-Package, an Application-Specific Integrated Circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be at least partially implemented as computer program modules, which, when run, can perform corresponding functions.
[0131] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, a computer program product, and an autonomous vehicle.
[0132] According to an embodiment of the present disclosure, an electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the methods described above.
[0133] According to embodiments of this disclosure, an autonomous vehicle is configured to perform the methods described above. In one specific embodiment, the implementation of the autonomous vehicle may refer to an electronic device.
[0134] According to embodiments of the present disclosure, a non-transitory computer-readable storage medium stores computer instructions, wherein the computer instructions are used to cause a computer to perform the methods described above.
[0135] According to an embodiment of this disclosure, a computer program product includes a computer program that, when executed by a processor, implements the method described above.
[0136] Figure 7A schematic block diagram of an example electronic device that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.
[0137] like Figure 7 As shown, device 700 includes a computing unit 701, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 702 or a computer program loaded into random access memory (RAM) 703 from storage unit 708. The RAM 703 may also store various programs and data required for the operation of device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via bus 704. Input / output (I / O) interface 705 is also connected to bus 704.
[0138] Multiple components in device 700 are connected to I / O interface 705, including: input unit 706, such as keyboard, mouse, etc.; output unit 707, such as various types of monitors, speakers, etc.; storage unit 708, such as disk, optical disk, etc.; and communication unit 709, such as network card, modem, wireless transceiver, etc. Communication unit 709 allows device 700 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.
[0139] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above, such as a ride-hailing method based on an autonomous vehicle. For example, in some embodiments, the ride-hailing method based on an autonomous vehicle can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program can be loaded and / or installed on device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by the computing unit 701, one or more steps of the ride-hailing method based on an autonomous vehicle described above can be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform a ride-sharing method based on an autonomous vehicle by any other suitable means (e.g., by means of firmware).
[0140] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.
[0141] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0142] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0143] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).
[0144] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.
[0145] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, distributed system servers, or servers incorporating blockchain technology.
[0146] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this disclosure can be achieved, and this is not limited herein.
[0147] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.
Claims
1. A ride-hailing method based on autonomous vehicles, comprising: In response to detecting that an object leaves the first autonomous vehicle, the environmental perception data of the first autonomous vehicle is analyzed to obtain the object's intention to ride in the vehicle; wherein, the environmental perception data includes data that affects the object's behavior; If the ride intention indicates that the object has a need for another ride, a pick-up instruction is generated, wherein the pick-up instruction is used to instruct a second autonomous driving vehicle to pick up the object; The pick-up instruction is sent to the second autonomous vehicle to dispatch the second autonomous vehicle to pick up the person.
2. The method of claim 1, wherein, The environmental perception data includes external environmental perception data of the area outside the vehicle where the object is located; the step of analyzing the environmental perception data of the first autonomous vehicle to obtain the object's riding intention includes: Environmental analysis is performed on the perceived external environment data to obtain environmental analysis results, wherein the environmental analysis results are used to indicate whether the environment in the external area affects the object's behavior; If the environmental analysis results indicate that the environment outside the vehicle affects the object's behavior, a travel intention is determined to indicate that the object has a need to travel again.
3. The method of claim 2, wherein, The environmental analysis of the perceived external environment data to obtain environmental analysis results includes: The weather type is identified by performing weather type identification on the vehicle external environment perception data; When the weather type is a first type, the environmental analysis results used to indicate the environmental influence of the external area on the object's behavior are determined, wherein the first type includes types that influence the object to make avoidance behavior; When the weather type is the second type, the environmental analysis result used to indicate that the environment outside the vehicle does not affect the behavior of the object is determined, wherein the second type includes types that do not affect the object to make avoidance behavior.
4. The method of claim 2 or 3, wherein, The environmental analysis of the perceived external environment data to obtain environmental analysis results includes: Safety data is obtained by performing safety analysis on the perceived external environment data, wherein the safety data is used to indicate the safety level of the external area; If the safety data meets the abnormal safety conditions, determine the environmental analysis results used to indicate the influence of the environment in the external area on the object's behavior.
5. The method according to any one of claims 1 to 4, wherein, The environmental perception data includes the location where the object exits the first autonomous vehicle; the analysis of the environmental perception data of the first autonomous vehicle to obtain the object's travel intention includes: Obtain the predetermined destination location of the first autonomous vehicle, wherein the predetermined destination location is determined based on the interactive operation of the object; By comparing the predetermined destination location and the drop-off location, a first comparison result is obtained, wherein the first comparison result is used to indicate whether the predetermined destination location and the drop-off location are the same; Based on the first comparison result, the travel intention is determined.
6. The method of claim 5, wherein, Determining the travel intention based on the first comparison result includes: In response to determining that the first comparison result indicates that the predetermined endpoint location and the disembarkation location are different, the moving direction of the object is determined based on the object behavior image in the environmental perception data; Based on the relationship between the direction of movement and the predetermined destination, the passenger's travel intention is determined.
7. The method of claim 6, wherein, Determining the passenger's travel intention based on the relationship between the direction of movement and the predetermined destination includes: In response to determining that the direction of movement is closer to the predetermined destination, the travel intention is determined to indicate that the subject has a need to travel again; In response to determining that the direction of movement is away from the predetermined destination, the passenger's travel intention is determined based on the relationship between the direction of movement and associated buildings and the entrance traffic data of the associated buildings, wherein the associated buildings include buildings associated with the drop-off location.
8. The method according to any one of claims 1 to 7, wherein, The environmental perception data includes the in-vehicle environmental perception data of the first autonomous vehicle; the step of analyzing the environmental perception data of the first autonomous vehicle to obtain the passenger's intention to ride includes: A second comparison result is obtained by comparing the in-vehicle environment perception data at a first moment and the in-vehicle environment perception data at a second moment, wherein the first moment and the second moment are the moments when the object enters and leaves the first autonomous vehicle, respectively, and the second comparison result is used to indicate whether there are different items in the first autonomous vehicle. Based on the second comparison result, the travel intention is determined.
9. The method according to any one of claims 1 to 8, wherein, If the travel intention indicates that the person has a need for another ride, a pick-up instruction is generated, including: The system acquires traffic data at the drop-off location, pose data and trip data of at least one autonomous vehicle, wherein the at least one autonomous vehicle includes the first autonomous vehicle. Based on the traffic data, the pose data, and the trip data, the second autonomous vehicle is determined from the at least one autonomous vehicle; The pick-up instruction is generated based on the identification information of the second autonomous vehicle and the drop-off location.
10. A ride-hailing device based on an autonomous vehicle, comprising: An analysis module is used to analyze the environmental perception data of the first autonomous vehicle in response to detecting that an object has left the first autonomous vehicle, and to obtain the object's intention to ride in the vehicle; wherein, the environmental perception data includes data that affects the object's behavior; A generation module is configured to generate a pick-up instruction when the ride intention indicates that the object has a need for another ride, wherein the pick-up instruction is used to instruct a second autonomous driving vehicle to pick up the object; The scheduling module is used to send the pick-up instruction to the second autonomous vehicle to schedule the second autonomous vehicle to pick up the object.
11. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-9.
12. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-9.
13. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-9.
14. An autonomous vehicle configured to implement the method according to any one of claims 1-9.
