Control method, system, device and electronic equipment of autonomous vehicle

Abstract

The present disclosure provides a control method, system and device of an autonomous vehicle and an electronic device, relates to the field of artificial intelligence, in particular to the field of autonomous driving, computer vision, image processing and the like. The specific implementation scheme is: in the case that the current trip of the autonomous vehicle ends, in response to sensing the limbs of a vehicle-riding object in the autonomous vehicle, identifying a limb posture combination of the vehicle-riding object; calling a target limb posture corresponding to the limb posture in the limb posture combination from a target limb posture combination, wherein the target limb posture combination includes target limb postures sorted in a calling order; matching the limb posture in the limb posture combination with the called target limb posture to obtain a matching result; and determining whether to control a target door of the autonomous vehicle to be unlocked and opened based on the matching result.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, and in particular to a control method, system, device and electronic device for autonomous vehicles in the fields of autonomous driving, computer vision, image processing and other related technologies. Background Technology

[0002] Currently, autonomous vehicles on the market all reuse the physical handles inside the car to open the door or press a physical button to start the automatic door, or rely on the unlocking method of private cars (passenger cars).

[0003] However, using physical handles or buttons to open doors exposes passengers to a larger skin contact area, making it difficult to avoid residual active viruses or bacteria from the previous journey. Even with automatic ultraviolet or alcohol disinfection, there are still unsanitary areas and reduced operational efficiency. Summary of the Invention

[0004] This disclosure provides a control method, system, device, and electronic device for an autonomous vehicle.

[0005] According to one aspect of this disclosure, a control method for an autonomous vehicle is provided. The method may include: upon the termination of the current journey of the autonomous vehicle, in response to sensing the limbs of a passenger in the autonomous vehicle, identifying a combination of limb postures of the passenger; retrieving a target limb posture corresponding to a limb posture in a target limb posture combination, wherein the target limb posture combination includes target limb postures ordered according to the retrieval order, and the correspondence between limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order; matching the limb postures in the limb posture combination with the retrieved target limb postures to obtain a matching result; and based on the matching result, determining whether to control the target door of the autonomous vehicle to unlock and open.

[0006] According to another aspect of this disclosure, another method for controlling an autonomous vehicle is also provided. This method may include: detecting the end of the current trip of the autonomous vehicle; displaying operation guidance information on the vehicle's in-vehicle interface, wherein the operation guidance information is used to guide the operation of a passenger in the autonomous vehicle to control the unlocking and opening of the vehicle's doors; sensing the limb of the passenger remaining in response to the operation guidance information; displaying limb posture prompt information on the in-vehicle interface, wherein the limb posture prompt information is used to prompt the passenger to perform a target limb posture retrieved from a target limb posture combination; displaying on the in-vehicle interface a matching result between the limb posture in the limb posture combination performed by the passenger based on the limb posture prompt information and the retrieved corresponding target limb posture, wherein the limb posture combination includes multiple limb postures performed sequentially by the passenger in an execution order, and the correspondence between the limb postures in the limb posture combination and the target limb posture in the target limb posture combination is determined based on the execution order and the retrieval order of the multiple target limb postures; and displaying on the in-vehicle interface a control result for unlocking and opening the doors of the autonomous vehicle based on the matching result.

[0007] According to another aspect of this disclosure, a control system for an autonomous vehicle is also provided. The system may include: a sensing module for sensing the limbs of a passenger in the autonomous vehicle upon the termination of the current journey of the autonomous vehicle; an image acquisition device for acquiring combinations of limb postures of the passenger, wherein the limb posture combinations include multiple limb postures executed sequentially by the passenger in an execution order; a processor for retrieving target limb postures corresponding to the limb postures in the target limb posture combinations, wherein the target limb posture combinations include target limb postures ordered in the retrieval order, and the correspondence between the limb postures in the limb posture combinations and the target limb postures in the target limb posture combinations is determined based on the execution order and the retrieval order; matching the limb postures in the limb posture combinations with the retrieved target limb postures to obtain a matching result; and a controller for determining, based on the matching result, whether to control the unlocking and opening of the doors of the autonomous vehicle.

[0008] According to another aspect of this disclosure, a control device for an autonomous vehicle is also provided. The device may include: an identification unit, configured to, upon the termination of the current trip of the autonomous vehicle, identify a combination of limb postures of a passenger in the autonomous vehicle in response to sensing the limbs of a passenger in the autonomous vehicle, wherein the combination of limb postures includes multiple limb postures performed sequentially by the passenger in an execution order; a retrieval unit, configured to retrieve a target limb posture corresponding to a limb posture in a target limb posture combination, wherein the target limb posture combination includes target limb postures ordered in the retrieval order, and the correspondence between limb postures in a limb posture combination and target limb postures in a target limb posture combination is determined based on the execution order and the retrieval order; a matching unit, configured to match the limb postures in the limb posture combination with the retrieved target limb postures to obtain a matching result; and a determination unit, configured to, based on the matching result, determine whether to control the unlocking and opening of the doors of the autonomous vehicle.

[0009] According to one aspect of this disclosure, an electronic device is also provided. The electronic device may include: 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 control method for an autonomous vehicle according to embodiments of this disclosure.

[0010] According to another aspect of this disclosure, a non-transitory computer-readable storage medium storing computer instructions is also provided, wherein the computer instructions are used to cause a computer to execute the control method of an autonomous vehicle according to embodiments of this disclosure.

[0011] According to another aspect of this disclosure, a computer program product is also provided, which may include a computer program that, when executed by a processor, implements the control method for an autonomous vehicle according to embodiments of this disclosure.

[0012] According to another aspect of this disclosure, an autonomous vehicle is also provided, which may include the aforementioned electronic equipment.

[0013] 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

[0014] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0015] Figure 1 This is a flowchart of a control method for an autonomous vehicle according to an embodiment of the present disclosure;

[0016] Figure 2 This is a flowchart of another control method for an autonomous vehicle according to an embodiment of the present disclosure;

[0017] Figure 3 This is a schematic diagram of a sensing module according to an embodiment of the present disclosure;

[0018] Figure 4(a) is a flowchart of an object-sensing gesture recognition method according to an embodiment of the present disclosure;

[0019] Figure 4(b) is a schematic diagram of a gesture recognition according to an embodiment of the present disclosure;

[0020] Figure 4(c) is a schematic diagram of a gesture recognition according to an embodiment of the present disclosure;

[0021] Figure 4(d) is a schematic diagram of a prompt message according to an embodiment of the present disclosure;

[0022] Figure 4(e) is a schematic diagram of another prompting message according to an embodiment of the present disclosure;

[0023] Figure 5 This is a schematic diagram of a control system for an autonomous vehicle according to an embodiment of the present disclosure;

[0024] Figure 6 This is a schematic diagram of a control device for an autonomous vehicle according to an embodiment of the present disclosure;

[0025] Figure 7 This is a schematic diagram of a control device for another autonomous vehicle according to an embodiment of the present disclosure;

[0026] Figure 8 This is a block diagram of an electronic device for a control method of an autonomous vehicle according to an embodiment of the present disclosure. Detailed Implementation

[0027] 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.

[0028] This disclosure provides a control method for autonomous vehicles. Figure 1 This is a flowchart of a control method for an autonomous vehicle according to an embodiment of the present disclosure, such as... Figure 1 As shown, an implementation scheme for the control method of this autonomous vehicle may include at least the following implementation steps:

[0029] Step S102: When the current trip of the autonomous vehicle ends, in response to sensing the limbs of the passenger in the autonomous vehicle, identify the limb posture combination of the passenger, wherein the limb posture combination includes the limb postures performed by the passenger in the order of execution.

[0030] In the technical solution provided in step S102 of this application, the autonomous vehicle can be an operational vehicle, such as a shared mobility operational vehicle. The current journey can refer to the process of the autonomous vehicle from leaving the station to entering the station. When the current journey of the autonomous vehicle ends, it can respond to the sensed limbs of the passenger in the autonomous vehicle and identify the limb posture of the passenger. There can be multiple limb postures, and each identified limb posture forms a combination of limb postures. The passenger can be a passenger in the autonomous vehicle, and the limbs can be limb parts that the passenger can flexibly perform, such as the passenger's hand or arm. This is just an example and no specific limitation is made on the limbs. The identification of the combination of limb postures can be done through screenshots, images, or videos, etc. This is just an example and no specific limitation is made on the identification method. The combination of limb postures can include the limb postures that the passenger performs sequentially according to the execution order. The body posture can be a sequence of actions performed by the limbs of the passenger, such as hand gestures or arm postures. For example, if the limb is the hand, the body posture can be the passenger's fingers spread or clenched fist; if the limb is the arm, the body posture can be the arm crossed or bent. It should be noted that any body posture that is convenient for the passenger to perform is within the scope of this embodiment, and no specific limitations are made here.

[0031] Optionally, upon completion of the current journey of the autonomous vehicle, the limbs of passengers in the vehicle can be sensed by information sensing devices within the vehicle. These information sensing devices can be infrared sensors or infrared sensing modules that include infrared sensors. When the information sensing device senses the limbs of a passenger, in response to the sensed limbs, the limb postures of the passenger in the autonomous vehicle can be identified to obtain the limb postures performed sequentially by the passenger in the execution order, thus forming a combination of limb postures.

[0032] It should be noted that when picking up passengers, autonomous vehicles typically authenticate the passenger's identity to avoid safety hazards caused by boarding the wrong vehicle. Once authentication is successful, the door automatically unlocks, without requiring gesture recognition and posing no safety risks. However, if the door is unlocked and opened immediately after passenger authentication when they disembark, it could easily lead to accidental door activation. Therefore, in this embodiment, when the current journey of the autonomous vehicle ends and the passenger is preparing to disembark, the vehicle senses the passenger's limbs and identifies their limb postures to obtain a set of limb postures. Based on the limb postures in the set, the vehicle can accurately determine the passenger's intention to disembark, thereby automatically unlocking and opening the door, thus improving the safety of the autonomous vehicle during operation.

[0033] Step S104: Retrieve the target limb posture corresponding to the limb posture in the target limb posture combination. The target limb posture combination includes target limb postures sorted according to the retrieval order. The correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order.

[0034] In the technical solution provided by step S104 of this application, a target limb posture corresponding to the limb posture in the target limb posture combination can be retrieved from the target limb posture combination based on the limb posture in the limb posture combination. The target limb posture combination can be used to match the limb postures in the limb posture combination to determine whether the limb postures in the limb posture combination meet the execution requirements. It can include target limb postures sorted according to the retrieval order, and can be a pre-stored posture limb combination, such as a pre-set posture limb combination for unlocking a car door.

[0035] Optionally, target limb postures that are similar to or have similar postures to the limb postures in the target limb posture combination can be retrieved from the target limb posture combination. Alternatively, target limb postures that correspond to the limb postures in the identified limb posture combination can be retrieved from the target limb posture combination based on the execution order of the limb postures in the identified limb posture combination.

[0036] For example, identifying the body postures of a passenger includes body postures with fingers spread and body postures with fingers clenched. Target body postures that are similar to or similar to those with fingers spread can be extracted from the target body posture combination, resulting in target body posture A corresponding to fingers spread, and target body postures that are similar to or similar to those with fingers clenched, resulting in target body posture B corresponding to fingers clenched. Since the execution order in the body posture group is fingers spread, then fingers clenched, the extraction order in the target body posture combination is target body posture A, then target body posture B.

[0037] For another example, the desired body posture of the passenger can be displayed on the in-vehicle display. When the infrared sensing module detects the passenger's body posture, it can identify the posture and obtain a combination of body postures. The target body postures can then be retrieved sequentially from the target body posture combinations according to their execution order.

[0038] Step S106: Match the limb postures in the limb posture combination with the retrieved target limb postures to obtain the matching results.

[0039] In the technical solution provided by step S106 of this application, a combination of limb postures is identified, and a target limb posture corresponding to the limb posture is extracted from the target limb posture combination based on the execution order of the limb postures in the combination. The limb postures in the combination and the corresponding target limb postures can be matched to obtain a matching result. The matching result can indicate whether the limb posture and the target limb posture are successfully matched or not. It can be used to characterize the similarity between the limb postures in the combination and the target limb posture; for example, it can be a similarity of 80%. It should be noted that this is only an example and does not impose specific limitations on the form of the matching result.

[0040] For example, the body postures of a passenger might be either fingers spread or fingers clenched. We can extract target body posture A (fingers spread) and target body posture B (fingers clenched) from the target body posture combination. We can then match fingers spread with target body posture A and fingers clenched with target body posture B to obtain the matching results.

[0041] Step S108: Based on the matching results, determine whether to control the target door of the autonomous vehicle to unlock and open.

[0042] In the technical solution provided by step S108 of this application, based on the matching result, it can be determined whether to control the target door of the autonomous vehicle to unlock and open. For example, if the matching result indicates that the limb posture in the limb posture combination matches the corresponding target limb posture, it can be determined that the target door of the autonomous vehicle needs to be controlled to unlock and open. If the matching result indicates that the limb posture in the limb posture combination does not match the corresponding target limb posture, it can be determined that the target door of the autonomous vehicle does not need to be controlled to unlock and open. The target door can be a pre-set door for unlocking and opening, such as the right rear door of the vehicle.

[0043] Optionally, when unlocking the vehicle, only the target door needs to be opened; it is not necessary to open all the doors. The target door can be a pre-set door to be opened based on actual usage. For example, when the autonomous vehicle is a commercial vehicle, from a driving safety perspective, passengers are not allowed to sit in the front passenger seat, and the left rear door is for entry only; therefore, the target door could be the right rear door. It should be noted that there are no specific restrictions on the location of the target door; the number and location of the target doors can be changed according to actual circumstances.

[0044] For example, the limbs of a passenger can be identified to obtain the passenger's first limb posture. A target limb posture corresponding to the first posture can be retrieved from a set of target limb posture combinations. The first limb posture can be matched with its corresponding target limb posture. If the match fails, matching subsequent limb postures stops, and the target door of the autonomous vehicle is not unlocked. If the match succeeds, the passenger's limbs can be identified to obtain the passenger's second limb posture, and this second limb posture can be matched with its corresponding target limb posture. If the second limb posture fails to match its corresponding target limb posture, matching subsequent limb postures stops, and the target door of the autonomous vehicle is not unlocked. If the match succeeds, the passenger's limbs can be identified again, and the next limb posture can be judged until the last limb posture is also successfully matched with its corresponding target limb posture. At this point, the target door of the autonomous vehicle can be unlocked. If any matching result fails during the matching process, the target door of the autonomous vehicle will not be unlocked or unlocked. The limb postures obtained from each recognition of the passenger's limbs constitute the combination of the passenger's limb postures.

[0045] Through steps S102 to S108, when the current journey of the autonomous vehicle ends, in response to sensing the limbs of the passenger in the autonomous vehicle, a combination of limb postures of the passenger is identified. This combination includes limb postures performed sequentially by the passenger in the execution order. A target limb posture corresponding to the limb posture in the target limb posture combination is retrieved from the target limb posture combination. This target limb posture combination includes target limb postures ordered according to the retrieval order, and the correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order. The limb postures in the limb posture combination are matched with the retrieved target limb postures to obtain a matching result. Based on the matching result, it is determined whether to control the target door of the autonomous vehicle to unlock and open. In other words, this embodiment of the present disclosure, when the current journey of the autonomous vehicle ends, achieves a contactless unlocking and opening scheme for the vehicle door from the inside by sensing the limbs of the passenger in the autonomous vehicle. Among them, the strategy of recognizing combinations of body postures can avoid the safety hazards of misoperation, enabling shared mobility vehicles to operate more safely, thereby achieving the technical effect of improving the safety of autonomous vehicle operation and solving the technical problem of low safety of autonomous vehicle operation.

[0046] The method described in this embodiment will now be described in further detail.

[0047] As an optional embodiment, step S106, matching the limb postures in the limb posture combination with the retrieved target limb postures to obtain the matching result includes: detecting the duration of the limb postures in the limb posture combination; and matching the limb postures in the limb posture combination with the retrieved target limb postures based on the detected duration to obtain the matching result.

[0048] In this embodiment, when the limbs of a passenger in an autonomous vehicle are sensed, the system can identify combinations of limb postures and detect the duration of each posture within the combination. Based on the detected duration, the system can match the limb postures in the combination with retrieved target limb postures to obtain a matching result. The duration can be the duration of a limb posture, such as 1 second or 2 seconds; this is merely an example and no specific limitation is made on the duration.

[0049] Optionally, identifying combinations of limb postures allows for the retrieval of target limb postures corresponding to those postures from the target limb posture combinations. The limb postures can be matched with the target limb postures in terms of posture and duration to obtain a matching result.

[0050] For example, by identifying a combination of limb postures and detecting the duration of the limb postures in the combination, the limb postures can be matched with the limbs, postures, and durations of the retrieved target limb postures to obtain the matching results.

[0051] As an optional embodiment, based on the detected duration, the limb postures in the limb posture combination are matched with the retrieved target limb postures to obtain the matching result. The matching result includes: in response to the similarity between the limb postures in the limb posture combination and the retrieved target limb postures being greater than the similarity threshold, and the detected duration meeting the target duration of the retrieved target limb postures, it is determined that the limb postures in the limb posture combination and the retrieved target limb postures are successfully matched.

[0052] In this embodiment, the similarity between the limb postures in the limb posture combination and the retrieved target limb posture can be determined. If the similarity between the limb postures in the limb posture combination and the retrieved target limb posture is greater than a similarity threshold, it can be further determined whether the duration of the limb posture meets the target duration of the retrieved target limb posture. If the duration of the limb posture is not less than the target duration of the retrieved target limb posture, it can be determined that the duration of the limb posture meets the target duration of the retrieved target limb posture. If the duration of the limb posture is less than the target duration of the retrieved target limb posture, it can be determined that the duration of the limb posture does not meet the target duration of the retrieved target limb posture. In response to the similarity between the limb postures in the limb posture combination and the retrieved target limb posture being greater than a similarity threshold, and the detected duration meeting the target duration of the retrieved target limb posture, it can be determined that the limb postures in the limb posture combination and the retrieved target limb posture are successfully matched. The similarity threshold can be a pre-stored threshold, such as a value preset according to actual conditions. For example, to improve control accuracy, a larger similarity threshold, such as 98%, can be designed. The target duration can be a pre-stored time length, such as a preset time length, such as 2 seconds, 3 seconds, etc. It should be noted that the above figures are only illustrative and do not impose specific restrictions on the size of the similarity threshold and the target duration.

[0053] For example, setting a similarity threshold of 90%, the target body posture is two hands clenched into fists with a target duration of 4 seconds. The body posture combination identified as a passenger is two hands clenched into fists and both hands with extended index fingers; the duration of the clenched fists is detected as 5 seconds. The similarity between the clenched fists and the corresponding target body posture is determined, and it is also determined whether the duration of the clenched fists meets the target duration of the retrieved target body posture. If the similarity between the clenched fists and the retrieved target body posture is 95%, which is greater than the similarity threshold, and the duration of the clenched fists is 5 seconds, which meets the target duration, then it can be determined that the body posture in the body posture combination successfully matches the retrieved target body posture.

[0054] As an optional embodiment, based on the detected duration, the limb postures in the limb posture combination are matched with the retrieved target limb postures to obtain the matching result, including: in response to the similarity between the limb postures in the limb posture combination and the retrieved target limb postures not being greater than a similarity threshold, and / or the detected duration not meeting the target duration of the retrieved target limb postures, it is determined that the limb postures in the limb posture combination and the retrieved target limb postures have failed to match.

[0055] In this embodiment, the similarity between the limb postures in the limb posture combination and the retrieved target limb posture, as well as the duration of the limb postures in the limb posture combination, are determined. If the similarity between the limb postures in the limb posture combination and the retrieved target limb posture is not greater than a similarity threshold, and / or the detected duration does not meet the target duration of the retrieved target limb posture, then it can be determined that the limb postures in the limb posture combination and the retrieved target limb posture have failed to match.

[0056] Optionally, the limb postures in the limb posture combination can be matched with the target limb postures in the retrieved target limb posture combination. The similarity between the limb postures and the corresponding target limb postures can be matched, and it can be determined whether the duration of the limb posture meets the target duration of the target limb posture. If the similarity between the limb posture and the corresponding target limb posture's limb movement is not greater than a similarity threshold, and the detected duration does not meet the target duration of the retrieved target limb posture, then it can be determined that the limb postures in the limb posture combination and the retrieved target limb posture have failed to match. Alternatively, if the similarity between the limb posture and the corresponding target limb posture's limb movement is not greater than a similarity threshold, but the detected duration meets the target duration of the retrieved target limb posture, then it can also be determined that the limb postures in the limb posture combination and the retrieved target limb posture have failed to match. Alternatively, if the similarity between the limb posture and the corresponding target limb posture's limb movement is greater than a similarity threshold, but the detected duration does not meet the target duration of the retrieved target limb posture, then it can also be determined that the limb postures in the limb posture combination and the retrieved target limb posture have failed to match.

[0057] In this embodiment of the disclosure, not only is the limb posture determined, but the duration of the limb posture is also judged, thereby avoiding false triggering and improving the accuracy of control.

[0058] For example, the duration of the clenched fist pose in a combination of body postures is detected to be 4 seconds, while the target duration of the corresponding target body posture in the target combination is 5 seconds. Matching the clenched fist pose with the target body posture shows a similarity of 95%, which is greater than the similarity threshold of 80%. However, the duration of the clenched fist pose is less than the target duration, failing to meet the target duration of the retrieved target body posture. Therefore, in response to the detected duration of the body posture in the combination of body postures not meeting the target duration of the retrieved target body posture, it can be determined that the clenched fist pose has failed to match the retrieved target body posture.

[0059] As an optional embodiment, step S104, retrieving the target limb posture corresponding to the limb posture in the target limb posture combination, includes: in response to the limb posture in the limb posture combination being the first limb posture, retrieving the target limb posture corresponding to the limb posture in the target limb posture combination; or, in response to the limb posture in the limb posture combination not being the first limb posture, and the previous limb posture in the limb posture combination successfully matching the target limb posture corresponding to the previous limb posture in the target limb posture combination, retrieving the target limb posture corresponding to the limb posture in the target limb posture combination.

[0060] In this embodiment, if a limb posture in a limb posture combination is determined to be the primary limb posture, then in response to this, the target limb posture corresponding to the limb posture in the target limb posture combination can be directly retrieved from the target limb posture combination. If the limb posture in the limb posture combination is not the primary limb posture, it must first be determined that the previous limb posture in the limb posture combination matches the target limb posture corresponding to the previous limb posture in the target limb posture combination before the target limb posture corresponding to the limb posture in the target limb posture combination is retrieved.

[0061] For example, when a limb posture is not the first (third) position in a limb posture combination, it is necessary to first determine whether the first limb posture in the limb posture combination matches the target limb posture corresponding to the first limb posture in the target limb posture combination. If the first limb posture fails to match the corresponding target limb posture, the matching of subsequent limb postures can be stopped; if the first limb posture matches the corresponding target limb posture, the target limb posture corresponding to the second limb posture in the target limb posture combination can be retrieved and matched. If the second limb posture in the limb posture combination fails to match the retrieved target limb posture corresponding to the second limb posture, the matching of subsequent limb postures can be stopped; if the second limb posture matches the corresponding target limb posture, the matching of the third limb posture in the limb posture combination can be continued, and the target limb posture corresponding to the non-first (third) limb posture in the target limb posture combination can be retrieved.

[0062] As an optional embodiment, in response to a successful match between a limb posture in the limb posture combination and a retrieved target limb posture, and the limb posture in the limb posture combination is not the last limb posture, a target limb posture corresponding to the next limb posture in the limb posture combination is retrieved from the target limb posture combination; the next limb posture in the limb posture combination and the target limb posture corresponding to the retrieved next limb posture are matched to obtain a matching result.

[0063] In this embodiment, when the limb posture in the limb posture combination successfully matches the retrieved target limb posture, and the limb posture in the limb posture combination is not the last limb posture, the target limb posture corresponding to the next limb posture in the limb posture combination can be retrieved from the target limb posture combination; the next limb posture in the limb posture combination and the target limb posture corresponding to the retrieved next limb posture are matched to obtain the matching result.

[0064] Optionally, in response to a successful match between a limb posture in the limb posture combination and the retrieved target limb posture, and the limb posture in the limb posture combination is not the last limb posture, a target limb posture corresponding to the next limb posture in the limb posture combination is retrieved from the target limb posture combination. The next limb posture in the limb posture combination and the retrieved target limb posture are matched. If the limb posture in the matched limb posture combination is still not the last limb posture, the target limb posture corresponding to the next limb posture in the current limb posture combination can be retrieved from the target limb posture combination. The next limb posture in the limb posture combination and the retrieved target limb posture are matched until the limb posture in the limb posture combination is the last limb posture. Then, the last limb posture is matched with the retrieved target limb posture corresponding to the last limb posture, and the final matching result is obtained.

[0065] In this embodiment of the disclosure, the limb posture group can be a set of multiple limb postures. By matching multiple limb postures in the limb posture group, the door will be unlocked and opened only when multiple limb postures are successfully matched. The combination strategy avoids the safety hazards caused by the passenger's misoperation and ensures the safety of the passenger.

[0066] For example, a combination of body postures might include a first posture of clasped hands, a second posture of open hands, and a third posture of both hands with index fingers extended. The target body posture corresponding to clasped hands can be retrieved from the target body posture combination and matched with its corresponding posture. If the match fails, matching the second posture is unnecessary. If the match succeeds, the target body posture corresponding to open hands can be retrieved from the target body posture combination and matched with its corresponding posture. If matching open hands with its corresponding posture fails, matching the third posture is unnecessary. If matching open hands with its corresponding posture succeeds, since open hands is not the last posture, the next posture after open hands—the posture with both index fingers extended—can be retrieved from the target body posture combination and matched with its corresponding posture. Since extending both index fingers is the last limb posture, the matching can be completed after both index fingers are extended, and the matching result can be obtained.

[0067] As an optional embodiment, step S108, determining whether to control the door of the autonomous vehicle to unlock and open based on the matching result includes: in response to the failure of the limb posture in the limb posture combination to match the retrieved target limb posture, prohibiting the control of the door of the autonomous vehicle to unlock and open.

[0068] In this embodiment, in response to the failure of the limb posture in the limb posture combination to match the retrieved target limb posture, the door of the autonomous vehicle is prohibited from being unlocked or opened.

[0069] Optionally, the limb postures in the limb posture combination are matched with the corresponding target limb postures. If the limb postures in the limb posture combination fail to match the retrieved target limb postures, the door of the autonomous vehicle is prevented from being unlocked or opened.

[0070] Optionally, if an incorrect body posture is detected during the matching process, the vehicle will fail to unlock or open, and the doors of the autonomous vehicle will be prevented from being unlocked or opened.

[0071] For example, it determines whether the first limb posture in a limb posture combination matches the target limb posture retrieved from the target limb posture combination. If the first limb posture fails to match the corresponding target limb posture, matching subsequent limb postures can be stopped, and unlocking / unlocking the doors of the autonomous vehicle can be prohibited. If the first limb posture matches the corresponding target limb posture, the target limb posture corresponding to the second limb posture in the target limb posture combination can be retrieved, and the second limb posture can be matched with the corresponding target limb posture. If the second limb gesture in the limb gesture combination fails to match the target limb gesture corresponding to the second limb gesture, the matching of subsequent limb gestures can be stopped, and the unlocking and opening of the doors of the autonomous vehicle can be prohibited. If the second limb gesture successfully matches the corresponding target limb gesture, the matching of the third limb gesture in the limb gesture combination can continue. The target limb gesture corresponding to the non-first (third) limb gesture in the limb gesture combination can be retrieved from the target limb gesture combination, and the third limb gesture can be matched with the corresponding target limb gesture. If the matching fails, the unlocking and opening of the doors of the autonomous vehicle can be prohibited.

[0072] As an optional embodiment, in response to the failure of matching a limb posture in a limb posture combination with the retrieved target limb posture, a failure prompt message is output. The failure prompt message is used to prompt the passenger to re-execute the limb posture that matches the retrieved target limb posture; re-identify the limb posture executed by the passenger based on the failure prompt message; and match the re-identified limb posture with the retrieved target limb posture to obtain a matching result.

[0073] In this embodiment, if a limb posture in the limb posture combination fails to match the retrieved target limb posture, a failure prompt message can be output. The limb posture performed by the passenger based on the failure prompt message can be re-identified, and the re-identified limb posture can be matched with the retrieved target limb posture to obtain a matching result. The failure prompt message can be used to prompt the passenger to re-perform the limb posture that matches the retrieved target limb posture. It can be text, voice, video, etc., for example, a voice message indicating "not matched," and the target limb posture can be re-displayed on the display interface. It should be noted that this is only an example and does not impose specific limitations on the content of the failure prompt message.

[0074] In this embodiment, in order to provide a better experience for passengers, when the limb posture combination of the passenger fails to match the retrieved target limb posture, a failure prompt message can be output to remind the passenger that they only need to re-execute the currently required predetermined limb posture, instead of starting a new round of detection process from scratch.

[0075] For example, when a limb posture in a combination fails to match the retrieved target limb posture, a failure message can be output and displayed on the vehicle's interface to remind the passenger to re-execute the limb posture that matches the retrieved target limb posture. The system can then re-identify the limb posture executed by the passenger based on the failure message and match it with the retrieved target limb posture to obtain a matching result.

[0076] As an optional embodiment, step S108, determining whether to control the door of the autonomous vehicle to unlock and open based on the matching result includes: in response to each limb posture in the limb posture combination successfully matching with the retrieved corresponding target limb posture, controlling the door of the autonomous vehicle to unlock and open.

[0077] In this embodiment, when each limb posture in the limb posture combination successfully matches the corresponding target limb posture, the door of the autonomous vehicle can be controlled to unlock and open in response to the successful match between each limb posture in the limb posture combination and the corresponding target limb posture.

[0078] Optionally, the doors of the autonomous vehicle can only be unlocked and opened if each limb posture in the limb posture combination successfully matches the corresponding target limb posture, thereby improving the accuracy of door control and avoiding false triggering.

[0079] As an optional embodiment, step S102, in response to sensing the limbs of a passenger in the autonomous vehicle, identifies the combination of limb postures of the passenger, including: detecting the duration for which the limbs of the passenger remain in front of the sensing module, wherein the sensing module is used to sense the limbs of the passenger; and in response to the detected duration being greater than a duration threshold, identifying the combination of limb postures of the passenger.

[0080] In this embodiment, when the limbs of a passenger in the autonomous vehicle are sensed, the limb posture of the passenger can be identified, and the duration for which the passenger's limbs remain in front of the sensing module can be detected. When it is determined that the duration of limb lingering exceeds a time threshold, it is determined that the passenger is in a state of wanting to get off the vehicle, and the combination of the passenger's limb postures can be identified. The sensing module can be an infrared sensor or infrared perception module, etc., used for sensing limbs. The time threshold can be a preset time length.

[0081] Optionally, when the limbs of a passenger in the autonomous vehicle are sensed, the system can detect whether the duration of the limb's stay exceeds a time threshold. If the duration exceeds the time threshold, it can be determined that the passenger intends to get off the vehicle, rather than that the event was mistakenly triggered. In response to the duration exceeding the time threshold, the passenger's limb posture can be identified to obtain a combination of limb postures.

[0082] For example, suppose the duration threshold is 3 seconds. The sensing module can detect when a passenger's hand remains in contact with another person's hand for more than 3 seconds. In response to this detected duration exceeding the threshold, the gesture recognition software algorithm in the AI ​​camera can be activated to identify the limbs and obtain a combination of limb postures. It should be noted that the above detection or recognition methods are merely illustrative and do not impose any specific limitations on the detection or recognition methods.

[0083] As an optional embodiment, the light information of the environment in which the autonomous vehicle is located is detected; in response to the light information being less than a light information threshold, a supplementary lighting device is activated; in response to the light information adjusted by the supplementary lighting device being greater than the light information threshold, the limbs of the passengers in the autonomous vehicle are sensed.

[0084] In this embodiment, the light information of the environment in which the autonomous vehicle is located can be detected, and it can be determined whether the light information of the environment in which the autonomous vehicle is located is less than a light information threshold. In response to the light information of the environment in which the autonomous vehicle is located being less than the light information threshold, a supplementary lighting device can be activated to adjust the light information. The supplementary lighting device can be used to adjust the light information in the vehicle. For example, the supplementary lighting device can be a small light panel or an infrared supplementary light installed in the vehicle; this is only an example and no specific limitation is made on the type of supplementary lighting device. The light information can be used to characterize the lighting conditions of the environment in the vehicle, for example, it can be the light intensity. The light information threshold can be a preset value.

[0085] Optionally, when the vehicle is in a relatively dark environment, the light information of the environment in which the autonomous vehicle is located is less than the light information threshold. In this case, limb recognition is prone to problems due to unclear recognition images, leading to recognition errors. To avoid the above problems, embodiments of this disclosure can detect the light information of the environment in which the autonomous vehicle is located. When the light information is less than the light information threshold, in response to the light information being less than the light information threshold, a supplementary lighting device can be activated to adjust the light information, thereby achieving the purpose of accurately recognizing limb postures.

[0086] In this embodiment of the disclosure, by utilizing supplementary lighting equipment, the body posture of passengers can be recognized even in low-light environments such as at night, thereby improving the accuracy of body movement recognition.

[0087] Optionally, if the light information is greater than the light information threshold, the limbs of passengers in the autonomous vehicle can be sensed by the sensing module. If the light information is not greater than the light information threshold, the light information of the vehicle's environment can be adjusted by a supplementary lighting device until the adjusted light information is greater than the light information threshold. Then, the limbs of passengers in the autonomous vehicle can be sensed by the sensing module (e.g., a camera).

[0088] As an optional embodiment, in response to sensing the limbs of the passenger, limb posture prompting information is output, wherein the limb posture prompting information is used to prompt the passenger to perform a target limb posture corresponding to the execution sequence.

[0089] In this embodiment, when the sensing module detects the limbs of a passenger, it can output limb posture prompts in response to the detected limbs. These limb posture prompts can be in the form of voice, images, text, etc.

[0090] For example, when the system senses the limbs of a passenger, it can output a limb posture prompt message: "Start matching, please extend your hands" and display it on the vehicle's screen in the form of a video animation. The passenger can then make the corresponding limb posture according to the limb posture prompt message.

[0091] It should be noted that body posture prompts can be one of the following: voice, image, or text information, or a combination of these three types of information. No specific restrictions are placed on the content of the body posture prompts here.

[0092] As an optional embodiment, in response to a limb posture configuration operation, a target limb posture is configured to obtain a target limb posture combination.

[0093] In this embodiment, the target limb posture combination can be obtained in response to a limb posture configuration operation, or the target limb posture can be configured in response to the limb posture configuration operation to obtain the target limb posture combination. The target limb posture can be an unlock gesture.

[0094] Optionally, in response to the limb posture configuration operation, a custom unlock gesture can be obtained to obtain the configured target limb posture combination.

[0095] For example, in response to a body posture configuration operation, a target body posture for unlocking the car door can be defined, a combination of target body postures can be obtained, and the target body posture can be displayed on a screen in the vehicle to remind the passenger of the process and details of unlocking the car door.

[0096] As an optional embodiment, the limbs include the hands of the passenger, and the limb postures include the gestures of the passenger.

[0097] In this embodiment, the limbs may include the hands of the passenger, and the limb posture may include the gestures of the passenger. However, it should be noted that the embodiments disclosed herein are not limited to hands and gestures, and appropriate limbs may be selected according to the actual situation.

[0098] This disclosure provides a method combining object perception and gesture recognition to enable contactless unlocking and opening of vehicle doors from the inside, making shared mobility vehicles safer, more hygienic, and more efficient to operate.

[0099] In this embodiment, upon completion of the current journey of the autonomous vehicle, a contactless unlocking and opening of the vehicle door from the inside is achieved by sensing the limbs of passengers within the vehicle. The strategy of recognizing combinations of limb postures mitigates the safety risks of accidental operation, enabling safer operation of shared mobility vehicles and thus improving the technical safety of autonomous vehicle operation, solving the technical problem of low safety in autonomous vehicle operation.

[0100] This disclosure also provides a control method for an autonomous vehicle. Figure 2 This is a flowchart of another control method for an autonomous vehicle according to an embodiment of the present disclosure, such as... Figure 2 As shown, an implementation scheme for the control method of this autonomous vehicle may include at least the following implementation steps:

[0101] Step S202: The current trip of the autonomous vehicle is detected to have ended. Operation guidance information is displayed on the vehicle's in-vehicle interface. The operation guidance information is used to guide the passengers in the autonomous vehicle to control the unlocking and opening of the vehicle's doors.

[0102] In the technical solution provided by step S202 of this application, when the current trip of the autonomous vehicle is detected to have ended, operation guidance information can be displayed on the vehicle's in-vehicle interface. This operation guidance information can be voice, text, or other forms of information, and can be used to guide the actions of passengers in the autonomous vehicle, enabling the vehicle to unlock and open its doors based on the passengers' actions.

[0103] For example, a small smart sensor screen could be embedded in the passenger-side headrest of an autonomous vehicle as an in-vehicle interface. When the vehicle's speed reaches 0, the current journey can be considered complete, and pre-set operation instructions can be displayed on the in-vehicle interface. It should be noted that the location of the in-vehicle interface described above is merely an example and does not impose any specific limitations on its placement.

[0104] Step S204: Sensing the limbs of the passenger who remain in the operation guidance information, and displaying limb posture prompt information on the vehicle interface. The limb posture prompt information is used to prompt the passenger to perform the target limb posture retrieved from the target limb posture combination.

[0105] In the technical solution provided in step S204 of this disclosure, when a limb of a passenger is sensed to remain in response to the operation guidance information, limb posture prompts can be displayed on the in-vehicle interface. These limb posture prompts can be used to prompt the passenger to perform a target limb posture extracted from a target limb posture combination, and can be prompts in the form of voice, video, or images.

[0106] Step S206: Display on the vehicle interface the matching result between the limb postures in the limb posture combination executed by the passenger based on the limb posture prompt information and the corresponding target limb postures retrieved. The limb posture combination includes multiple limb postures executed sequentially by the passenger in the execution order. The correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order of the multiple target limb postures.

[0107] In the technical solution provided by step S206 of this application, the matching result between the limb postures in the combination of limb postures performed by the passenger based on the limb posture prompt information and the corresponding target limb posture can be obtained, and the matching result can be displayed on the vehicle interface. The combination of limb postures may include multiple limb postures performed sequentially by the passenger in the order of execution.

[0108] For example, the body posture prompt information is "both hands clenched". By obtaining the body posture of the passenger, the corresponding target body posture can be retrieved from the target body posture combination, and the matching result between the passenger's body posture and the target body posture can be determined. The matching result can be displayed on the in-vehicle interface.

[0109] Step S208: Display the control results of unlocking and opening the doors of the autonomous vehicle based on the matching results on the vehicle interface.

[0110] In the technical solution provided by step S208 disclosed in this application, the control result of unlocking and opening the doors of the autonomous vehicle based on the matching result can be displayed on the vehicle interface. For example, if the matching result is successful, the control result of "controlling the door to unlock and open" can be displayed on the vehicle interface; if the matching result is unsuccessful, the control result of "matching failed, unable to control the door to unlock and open" can be displayed on the vehicle interface.

[0111] It should be noted that the content displayed on the in-vehicle interface is for illustrative purposes only, and no specific restrictions are imposed on the content displayed on the in-vehicle interface.

[0112] Through steps S202 to S208, the current journey of the autonomous vehicle is detected to have ended. Operation guidance information is displayed on the vehicle's in-vehicle interface, guiding the passenger's actions to control the unlocking and opening of the vehicle's doors. The system senses the passenger's limbs remaining in response to the operation guidance information and displays limb posture prompts on the in-vehicle interface, prompting the passenger to execute a target limb posture retrieved from a target limb posture combination. The in-vehicle interface displays the matching result between the limb postures executed by the passenger based on the limb posture prompts and the corresponding retrieved target limb postures. The limb posture combination includes multiple limb postures executed sequentially by the passenger, and the correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order of the multiple target limb postures. Finally, the in-vehicle interface displays the control result of unlocking and opening the vehicle's doors based on the matching result. This achieves the technical effect of improving the safety of autonomous vehicle operation and solves the technical problem of low safety in autonomous vehicle operation.

[0113] The above technical solutions of the present disclosure will be further illustrated below with reference to preferred embodiments.

[0114] With the intelligent development of road and vehicle collaboration, unmanned shared mobility vehicles have gradually become more widespread. However, during use, shared mobility vehicles are prone to hygiene and epidemic prevention issues due to contact with surfaces that may harbor viruses and bacteria. Therefore, disinfection of vehicles after the end of their operation has become an urgent problem to be solved.

[0115] Currently, shared mobility vehicles on the market rely on physical door handles or physical buttons to activate automatic doors, as well as unlocking methods based on private car or passenger car keys. However, these methods easily result in a large skin contact area, making it difficult to avoid residual active viruses or bacteria from the previous journey. Even if the vehicle automatically disinfects with ultraviolet light or alcohol, there are still unsanitary areas and reduced operational efficiency; for example, ultraviolet light requires a longer exposure time to achieve a sterilization effect.

[0116] When autonomous shared mobility vehicles arrive at a station, the ability to unlock and open the doors quickly and without contact is a rigid requirement. Therefore, there is an urgent need for a more hygienic and user-friendly contactless method to unlock the rear doors upon arrival, replacing the traditional physical button control for opening the doors.

[0117] This disclosure provides a method combining object perception and gesture recognition. Figure 3 This is a schematic diagram of a sensing module according to an embodiment of the present disclosure, such as... Figure 3 As shown, a small sensing module, such as a smart sensor screen, is embedded in the passenger-side headrest. This smart sensor screen can have a built-in infrared sensor (such as an infrared sensing module) to detect hand movement immediately. It can also be paired with an Artificial Intelligence (AI) camera for gesture recognition. Additionally, an infrared auxiliary light can be installed, combined with AI gesture recognition algorithms, to quickly recognize hand movements even in low-light conditions such as at night. This enables contactless unlocking and opening of the car door from the inside, improving the safety of autonomous vehicles and addressing the technical issue of low safety in autonomous vehicle operation.

[0118] The following section will further introduce the method of combining object perception with gesture recognition.

[0119] Figure 4(a) is a flowchart of an object perception gesture recognition method according to an embodiment of the present disclosure. As shown in Figure 4(a), the implementation scheme of the object perception gesture recognition method in the autonomous vehicle may include at least the following implementation steps:

[0120] Step S402: Define the door unlocking gesture and inform the passenger of the unlocking process through the software.

[0121] In this embodiment, an unlocking gesture can be preset on the vehicle door. This gesture can be a single gesture or a set of gestures. The unlocking process can be communicated to the passenger via software. The passenger can be any person inside the vehicle. The software can be from an in-vehicle display (or a smart sensor screen within the vehicle); the location and type of the software are not specifically limited here.

[0122] For example, Figure 4(b) is a schematic diagram of a gesture recognition according to an embodiment of the present disclosure. As shown in Figure 4(b), images and text can be displayed on the vehicle's smart sensor screen to inform passengers of the unlocking process and the required body posture. For instance, as shown in Figure 4(b), the text "Please make a fist with one hand" and a demonstration image of making a fist with one hand can be displayed on the smart sensor screen. The unlocking process can be communicated to passengers through text, images, voice, etc. It should be noted that the above-mentioned communication methods are only illustrative examples and do not impose specific limitations on the communication methods.

[0123] For example, unlocking gestures for car doors can be predefined. When a passenger is in the car, they can activate a mini-program on the in-vehicle display (e.g., an in-vehicle tablet interface), or when a passenger wants to get out of the car, they can confirm the unlocking process by clicking on a ride-hailing app on their mobile phone.

[0124] It should be noted that this disclosure can also recognize other parts of the passenger's face, head, etc., such as recognizing the passenger's face, blinking, tilting their head up, and opening their mouth. However, since actions such as facing the face, blinking, tilting their head up, and opening their mouth are common actions of the passenger, they are easy to accidentally unlock the door. Therefore, considering operational safety and in order to improve the accuracy of unlocking the door, this disclosure focuses on recognizing gestures to achieve the purpose of unlocking the door.

[0125] Step S404: Sensing the duration of hand lingering.

[0126] In this embodiment, the duration of a passenger's hand movement can be sensed using an infrared sensing device installed in the vehicle.

[0127] Optionally, the duration of hand contact can be determined. If the duration exceeds a time threshold, step S406 can be implemented; if the duration does not exceed the time threshold, the passenger can be informed of the unlocking process via a mini-program, for example, through voice or visual explanation. The time threshold can be a preset duration, such as 3 seconds.

[0128] Step S406, obtain gesture 1.

[0129] In this embodiment, by setting a time threshold, the sensed gesture will only be acquired when the hand stays for a longer period of time than the time threshold, thereby reducing the probability of accidental touch and improving the accuracy of recognition.

[0130] Optionally, if the duration of the perceived hand lingering exceeds a time threshold, it can be determined that the passenger wants to get off the vehicle, and the passenger's current gesture 1 can be obtained. Gesture 1 can be obtained by acquiring the fingertip position and tracking the movement trajectory, or by capturing a dynamic image of the hand. It should be noted that the gesture acquisition methods here are only illustrative and do not impose specific limitations on the gesture acquisition methods.

[0131] For example, Figure 4(c) is a schematic diagram of gesture recognition according to an embodiment of the present disclosure. As shown in Figure 4(c), the passenger makes a corresponding action according to the instructions on the smart sensing display screen. The duration of the passenger's hand lingering can be sensed by an infrared sensing device installed in the vehicle. When the duration of the passenger's hand lingering exceeds a time threshold, the gesture recognition software algorithm can be activated to obtain the passenger's gesture.

[0132] Step S408: Recognize and compare gesture 1.

[0133] In this embodiment, the acquired gesture 1 can be recognized and compared. If the comparison is successful, the passenger can be informed via voice, video, image, or other means that the recognition is successful and the passenger can proceed to the recognition of gesture 2.

[0134] For example, once the gesture recognition software algorithm is activated, it can obtain the fingertip position of the passenger at this moment and track the movement trajectory of the finger to obtain gesture 1, which can then be recognized and compared.

[0135] For example, a time threshold of 3 seconds can be set, and the unlocking action is an initial gesture of five fingers spread out, which is held for 3 seconds. The infrared sensing device senses the duration of the passenger's hand lingering. When the duration of the passenger's hand lingering exceeds 3 seconds, the gesture of the passenger can be obtained by capturing a dynamic image of the hand. The obtained gesture is then compared and judged to determine whether the passenger's first action meets the requirement of five fingers spread out and lingering for 3 seconds. If it does, step S410 is executed.

[0136] Step S410, obtain gesture 2.

[0137] In this embodiment, the car door will only open if the passenger performs both gestures. However, if the passenger does not perform either gesture, the door unlocking will fail, and it will be necessary to re-sensor the passenger's hand and restart the recognition of a series of gestures; or, it will be necessary to re-recognize the passenger's current gesture.

[0138] Optionally, when the passenger's gesture 1 is successfully matched, the passenger's gesture 2 can be obtained by capturing a dynamic image of the hand.

[0139] This embodiment of the disclosure reduces misjudgments and improves recognition accuracy by recognizing a combination of multiple gestures. In other words, it avoids safety hazards caused by misoperation through gesture combination strategies, achieving precise control of the car door opening and closing. The multiple gestures can be a combination of two or three gestures; the number and actions of the combined gestures are not specifically limited here.

[0140] Step S412: Recognize and compare gesture 2.

[0141] In this embodiment, the acquired gesture 2 can be recognized and compared. If the comparison is successful, the passenger can be notified via voice, video, image, or other means that the recognition was successful and the car door can be unlocked.

[0142] For example, a time threshold of 3 seconds is set, and the unlocking action is an initial gesture of five fingers spread, which stays for 3 seconds and then changes to a clenched fist for 2 seconds. The infrared sensing device senses the duration of the passenger's hand lingering. When the duration of the passenger's hand lingering exceeds 3 seconds, the gesture 1 of the passenger can be obtained by capturing a dynamic image of the hand. The gesture 1 can be identified and compared to determine whether the passenger's gesture 1 meets the requirement of five fingers spread and staying for 3 seconds. If it does, gesture 2 can be obtained. It is then determined whether the passenger's gesture 2 is a clenched fist for 2 seconds. If it is, step S414 is executed; otherwise, the door unlocking / opening fails.

[0143] Step S414: Unlock the car door.

[0144] In this embodiment, the gestures of the passenger are recognized. If the passenger's gestures meet the preset gesture actions, the car door is unlocked and opened, and the detection process ends.

[0145] Optionally, the gestures of the passenger can be recognized. If the passenger's gestures meet the preset gesture actions, a matching success prompt message can be output on the display interface, and the car door can be unlocked. Figure 4(d) is a schematic diagram of a prompt message according to an embodiment of the present disclosure. As shown in Figure 4(d), after successful unlocking, the prompt message "Matching successful, the right rear door has been successfully unlocked and opened. Please take all your belongings with you. Please be aware of oncoming vehicles when getting off the vehicle. Welcome to ride with us again" can be displayed on the display interface to remind the passenger that the match has been successful and the location where they can get off the vehicle.

[0146] For example, when matching fails, the display interface can remind the passenger of the matching failure and re-match the passenger's hand. Figure 4(e) is a schematic diagram of another prompt message according to an embodiment of the present disclosure. As shown in Figure 4(e), when the passenger's matching fails, the display interface can show: "Matching failed, please make a fist with one hand again", and display an image of the gesture that the passenger needs to make.

[0147] This embodiment of the disclosure provides a hygienic and safe method for opening the vehicle doors when the unmanned shared mobility vehicle arrives at the station. It is simple to operate and requires no training for passengers to avoid misunderstandings. At the same time, it saves disinfection and cleaning time and improves operational efficiency.

[0148] This disclosure also provides a control system for an autonomous vehicle. Figure 5 This is a schematic diagram of a control system for an autonomous vehicle according to an embodiment of the present disclosure. Figure 5 As shown, the control system 500 of the autonomous vehicle may include: a sensing module 502, an image acquisition device 504, a processor 506, and a controller 508.

[0149] The sensing module 502 is used to sense the limbs of passengers in the autonomous vehicle when the current trip of the autonomous vehicle has ended.

[0150] In this embodiment, the sensing module can be used to sense the limbs of passengers in the autonomous vehicle when the current trip of the autonomous vehicle has ended. The sensing module can be an infrared sensing module. The limbs can be the hands, torso, etc. of the passenger; this is only an example and no specific limitation is made to the limbs of the passenger.

[0151] For example, a sensing module can be installed in the passenger headrest of an autonomous vehicle. When the current trip of the autonomous vehicle ends, the sensing module can be used to sense the limbs of the passengers in the autonomous vehicle.

[0152] Image acquisition device 504 is used to acquire a combination of limb postures of a passenger, wherein the combination of limb postures includes multiple limb postures performed sequentially by the passenger in an execution order.

[0153] In this embodiment, a combination of body postures of a passenger can be captured using an image acquisition device. This image acquisition device, which can be a camera, such as an AI camera, is used to capture the passenger's body postures. The combination of body postures can include multiple body postures performed sequentially by the passenger.

[0154] Processor 506 is used to retrieve target limb postures corresponding to limb postures in target limb posture combinations from target limb posture combinations, wherein the target limb posture combination includes target limb postures sorted according to the retrieval order, and the correspondence between limb postures in the limb posture combination and target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order; and to match the limb postures in the limb posture combination with the retrieved target limb postures to obtain a matching result.

[0155] In this embodiment, the processor can retrieve a target limb pose corresponding to a limb pose in a target limb pose combination. The target limb pose can be a preset combination of target limb poses ordered by retrieval. The processor can then match the limb poses in the limb pose combination with the retrieved target limb pose to obtain a matching result. This matching result can be used to characterize the matching status between the limb poses in the limb pose combination and the target limb pose.

[0156] Controller 508 is used to determine, based on the matching result, whether to control the doors of the autonomous vehicle to unlock or open.

[0157] In this embodiment, based on the matching results, the controller can control the unlocking of the doors of the autonomous vehicle.

[0158] Optionally, it can be determined whether the limb postures in the limb posture combination match the retrieved target limb posture. If the matching result indicates a match between the limb postures in the limb posture combination and the retrieved target limb posture, the door of the autonomous vehicle can be unlocked and opened via the controller. If the matching result indicates a mismatch between the limb postures in the limb posture combination and the retrieved target limb posture, the door of the autonomous vehicle can be unlocked and opened without controlling the door via the controller.

[0159] In this embodiment, the control system of the autonomous vehicle may further include a supplementary lighting device. This supplementary lighting device can be used to adjust the light information within the vehicle. For example, the supplementary lighting device can be a small light panel or an infrared supplementary light installed in the vehicle; this is merely an example and does not impose specific limitations on the type of supplementary lighting device.

[0160] Optionally, it can be determined whether the light information of the environment in which the autonomous vehicle is located is less than a light information threshold. In response to the light information of the environment in which the autonomous vehicle is located being less than the light information threshold, the light information can be adjusted by a supplementary lighting device. The light information can be used to characterize the lighting conditions of the environment in the vehicle, for example, it can be the light intensity. The light information threshold can be a preset value.

[0161] In this embodiment of the disclosure, by utilizing supplementary lighting equipment, the body posture of passengers can be recognized even in low-light environments such as at night, thereby improving the accuracy of body movement recognition.

[0162] In this embodiment, if the light information is greater than the light information threshold, the sensing module can sense the limbs of passengers in the autonomous vehicle. If the light information is not greater than the light information threshold, the light information of the vehicle's environment can be adjusted by a supplementary lighting device until the adjusted light information is greater than the light information threshold, after which the sensing module can sense the limbs of passengers in the autonomous vehicle.

[0163] This disclosure presents a control system for an autonomous vehicle. The system uses a sensing module to detect the limbs of passengers in the autonomous vehicle upon the end of the current journey. An image acquisition device captures combinations of the passengers' limb postures. A processor retrieves a target limb posture corresponding to the limb posture in the target limb posture combination and matches it with the retrieved target limb posture. Based on the matching result, the controller determines whether to unlock or open the vehicle's doors, thereby improving the safety of autonomous vehicle operation and solving the technical problem of low safety in autonomous vehicle operation.

[0164] This disclosure also provides an embodiment for performing Figure 1 The control device for an autonomous vehicle is a control method for an autonomous vehicle according to the embodiment shown.

[0165] Figure 6 This is a schematic diagram of a control device for an autonomous vehicle according to an embodiment of the present disclosure. Figure 6 As shown, the control device 600 of the autonomous vehicle may include: an identification unit 602, a retrieval unit 604, a matching unit 606, and a determination unit 608.

[0166] The recognition unit 602 is used to recognize the combination of limb postures of the passenger in the autonomous vehicle in response to sensing the limbs of the passenger in the autonomous vehicle when the current trip of the autonomous vehicle ends. The combination of limb postures includes multiple limb postures performed by the passenger in the order of execution.

[0167] The retrieval unit 604 is used to retrieve the target limb posture corresponding to the limb posture in the target limb posture combination from the target limb posture combination. The target limb posture combination includes target limb postures sorted according to the retrieval order. The correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order.

[0168] The matching unit 606 is used to match the limb postures in the limb posture combination with the retrieved target limb postures to obtain the matching result.

[0169] The determining unit 608 is used to determine, based on the matching result, whether to control the doors of the autonomous vehicle to unlock and open.

[0170] Optionally, the matching unit 606 includes: a detection module for detecting the duration of limb postures in the limb posture combination; and a first processing module for matching the limb postures in the limb posture combination with the retrieved target limb postures based on the detected duration, to obtain a matching result.

[0171] Optionally, the first processing module includes: a first determining submodule, configured to determine that the limb posture in the limb posture combination is successfully matched with the retrieved target limb posture in response to the similarity between the limb posture in the limb posture combination and the retrieved target limb posture being greater than a similarity threshold, and the detected duration meeting the target duration of the retrieved target limb posture.

[0172] Optionally, the first processing module includes: a second determining submodule, configured to determine that the matching between the limb posture in the limb posture combination and the retrieved target limb posture fails in response to the following: the similarity between the limb posture in the limb posture combination and the retrieved target limb posture is not greater than a similarity threshold, and / or the detected duration does not meet the target duration of the retrieved target limb posture.

[0173] Optionally, the retrieval unit 604 includes: a retrieval module, configured to retrieve a target limb posture corresponding to the limb posture in the target limb posture combination in response to the limb posture being the first limb posture in the limb posture combination; or, in response to the limb posture being not the first limb posture in the limb posture combination, and the previous limb posture in the limb posture combination successfully matching the target limb posture corresponding to the previous limb posture in the target limb posture combination, retrieve a target limb posture corresponding to the limb posture in the target limb posture combination in response to the limb posture being the first limb posture in the limb posture combination.

[0174] Optionally, the device further includes: a first processing unit, configured to respond to a successful match between a limb posture in the limb posture combination and a retrieved target limb posture, and the limb posture in the limb posture combination is not the last limb posture, retrieve a target limb posture corresponding to the next limb posture in the limb posture combination from the target limb posture combination; and match the next limb posture in the limb posture combination with the retrieved target limb posture to obtain a matching result.

[0175] Optionally, the determining unit 608 includes: a second processing module, configured to prevent the door of the autonomous vehicle from unlocking or opening in response to a failure to match the limb posture in the limb posture combination with the retrieved target limb posture.

[0176] Optionally, the device further includes: a second processing unit, configured to output a failure prompt message in response to a failure to match a limb posture in a limb posture combination with a retrieved target limb posture, wherein the failure prompt message is used to prompt the passenger to re-execute the limb posture that matches the retrieved target limb posture; re-identify the limb posture executed by the passenger based on the failure prompt message; and match the re-identified limb posture with the retrieved target limb posture to obtain a matching result.

[0177] Optionally, the determining unit 608 includes a control module, configured to control the doors of the autonomous vehicle to unlock and open in response to a successful match between each limb posture in the limb posture combination and the retrieved corresponding target limb posture.

[0178] Optionally, the retrieval unit 602 includes: a processing submodule, used to detect the duration for which a passenger's limbs remain in front of the sensing module, wherein the sensing module is used to sense the passenger's limbs; and in response to the detected duration being greater than a duration threshold, to identify a combination of the passenger's limb postures.

[0179] Optionally, the device further includes: a third processing unit for detecting light information of the environment in which the autonomous vehicle is located; activating a supplementary lighting device in response to light information being less than a light information threshold; and sensing the limbs of passengers in the autonomous vehicle in response to light information adjusted by the supplementary lighting device being greater than the light information threshold.

[0180] Optionally, the device further includes a fourth processing unit, configured to configure a target limb posture in response to a limb posture configuration operation, thereby obtaining a target limb posture combination.

[0181] This disclosure also provides another method for performing... Figure 2 The control device for an autonomous vehicle is a control method for an autonomous vehicle according to the embodiment shown.

[0182] Figure 7 This is a schematic diagram of a control device for another autonomous vehicle according to an embodiment of the present disclosure. Figure 7 As shown, the control device 700 of the autonomous vehicle may include: a first display unit 702, a second display unit 704, a third display unit 706 and a fourth display unit 708.

[0183] The first display unit 702 is used to detect the end of the current trip of the autonomous vehicle and display operation guidance information on the vehicle interface of the autonomous vehicle. The operation guidance information is used to guide the operation of the passengers in the autonomous vehicle to control the unlocking and opening of the doors of the autonomous vehicle.

[0184] The second display unit 704 is used to sense the limbs of the passenger who remain under the operation guidance information and display limb posture prompt information on the vehicle interface. The limb posture prompt information is used to prompt the passenger to perform the target limb posture retrieved from the target limb posture combination.

[0185] The third display unit 706 is used to display on the vehicle interface the matching result between the limb postures in the limb posture combination performed by the passenger based on the limb posture prompt information and the corresponding target limb posture retrieved. The limb posture combination includes multiple limb postures performed by the passenger in the order of execution. The correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order of the multiple target limb postures.

[0186] The fourth display unit 708 is used to display the control results of unlocking and opening the doors of the autonomous vehicle based on the matching results on the vehicle interface.

[0187] In the control device for an autonomous vehicle according to embodiments of this disclosure, an identification unit identifies a combination of limb postures of a passenger in the autonomous vehicle in response to sensing the limbs of the passenger when the current trip of the autonomous vehicle ends; a retrieval unit retrieves a target limb posture corresponding to the limb posture in the target limb posture combination; a matching unit matches the limb posture in the limb posture combination with the retrieved target limb posture to obtain a matching result; and a determination unit determines whether to control the door of the autonomous vehicle to unlock or open based on the matching result, thereby achieving the technical effect of improving the safety of autonomous vehicle operation and solving the technical problem of low safety of autonomous vehicle operation.

[0188] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0189] 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.

[0190] Embodiments of this disclosure provide an autonomous vehicle, wherein the autonomous vehicle may include electronic equipment. The electronic equipment may include: 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 control method of the autonomous vehicle according to embodiments of this disclosure.

[0191] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0192] According to embodiments of the present disclosure, the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to perform a method for controlling an autonomous vehicle according to embodiments of the present disclosure.

[0193] Optionally, in this embodiment, the non-volatile storage medium described above can be configured to store a computer program for performing the following steps:

[0194] S1, when the current trip of the autonomous vehicle ends, in response to sensing the limbs of the passenger in the autonomous vehicle, identify the combination of the passenger's limb postures.

[0195] S2, retrieve the target limb posture corresponding to the limb posture in the target limb posture combination from the target limb posture combination, wherein the target limb posture combination includes target limb postures sorted according to the retrieval order, and the correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order.

[0196] S3, match the limb postures in the limb posture combination with the retrieved target limb postures to obtain the matching results;

[0197] S4, based on the matching results, determines whether to control the target door of the autonomous vehicle to unlock and open.

[0198] Optionally, in this embodiment, the non-volatile storage medium may also be configured to store a computer program for performing the following steps:

[0199] S1, detect that the current trip of the autonomous vehicle has ended, and display operation guidance information on the vehicle's in-vehicle interface. The operation guidance information is used to guide the operation of the passengers in the autonomous vehicle to control the unlocking and opening of the vehicle's doors.

[0200] S2, sense the limbs of the passenger who remain in the operation guidance information, and display limb posture prompt information on the vehicle interface. The limb posture prompt information is used to prompt the passenger to perform the target limb posture retrieved from the target limb posture combination.

[0201] S3, Display on the vehicle interface the matching result between the limb postures in the limb posture combination executed by the passenger based on the limb posture prompt information and the corresponding target limb posture retrieved. The limb posture combination includes multiple limb postures executed by the passenger in the order of execution. The correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order of multiple target limb postures.

[0202] S4 displays the control results of unlocking and opening the doors of the autonomous vehicle based on the matching results on the in-vehicle interface.

[0203] Optionally, in this embodiment, the aforementioned non-transitory computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. More specific examples of 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 fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0204] According to embodiments of this disclosure, this disclosure also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0205] S1, when the current trip of the autonomous vehicle ends, in response to sensing the limbs of the passenger in the autonomous vehicle, identify the combination of the passenger's limb postures.

[0206] S2, retrieve the target limb posture corresponding to the limb posture in the target limb posture combination from the target limb posture combination, wherein the target limb posture combination includes target limb postures sorted according to the retrieval order, and the correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order.

[0207] S3, match the limb postures in the limb posture combination with the retrieved target limb postures to obtain the matching results;

[0208] S4, based on the matching results, determines whether to control the target door of the autonomous vehicle to unlock and open.

[0209] Optionally, in this embodiment, the computer program, when executed by the processor, may further perform the following steps:

[0210] S1, detect that the current trip of the autonomous vehicle has ended, and display operation guidance information on the vehicle's in-vehicle interface. The operation guidance information is used to guide the operation of the passengers in the autonomous vehicle to control the unlocking and opening of the vehicle's doors.

[0211] S2, sense the limbs of the passenger who remain in the operation guidance information, and display limb posture prompt information on the vehicle interface. The limb posture prompt information is used to prompt the passenger to perform the target limb posture retrieved from the target limb posture combination.

[0212] S3, Display on the vehicle interface the matching result between the limb postures in the limb posture combination executed by the passenger based on the limb posture prompt information and the corresponding target limb posture retrieved. The limb posture combination includes multiple limb postures executed by the passenger in the order of execution. The correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order of multiple target limb postures.

[0213] S4 displays the control results of unlocking and opening the doors of the autonomous vehicle based on the matching results on the in-vehicle interface.

[0214] Figure 8 This is a block diagram of an electronic device for a control method of an autonomous vehicle according to an embodiment of the present disclosure. 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.

[0215] like Figure 8 As shown, device 800 includes a computing unit 801, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 802 or a computer program loaded from storage unit 808 into random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0216] Multiple components in device 800 are connected to I / O interface 805, including: input unit 806, such as keyboard, mouse, etc.; output unit 804, such as various types of monitors, speakers, etc.; storage unit 808, such as disk, optical disk, etc.; and communication unit 809, such as network card, modem, wireless transceiver, etc. Communication unit 809 allows device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0217] The computing unit 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 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 801 performs the various methods and processes described above, such as method data processing methods. For example, in some embodiments, the method data processing method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the data processing method described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to perform data processing methods by any other suitable means (e.g., by means of firmware).

[0218] 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.

[0219] 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.

[0220] 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.

[0221] 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).

[0222] 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 embodiments 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.

[0223] 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, servers in distributed systems, or servers incorporating blockchain technology.

[0224] 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 disclosed in this disclosure can be achieved, and this is not limited herein.

[0225] 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 control method for an autonomous vehicle, comprising: When the current trip of the autonomous vehicle ends, in response to sensing the limbs of the passenger in the autonomous vehicle, the combination of the limb postures of the passenger is identified. The combination of limb postures includes the limb postures performed by the passenger in the order of execution. The limbs include the passenger's hands, and the limb postures include the passenger's gestures. The combination of limb postures is obtained by capturing the limb postures of the passenger through an image acquisition device. The limb postures are the sequence of actions performed by the passenger's limbs. Retrieve the target limb posture corresponding to the limb posture in the target limb posture combination from the target limb posture combination, wherein the target limb posture combination includes target limb postures sorted according to the retrieval order, and the correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order; In response to the limb posture being the first limb posture of the passenger, the first limb posture is matched with the retrieved target limb posture, wherein the retrieved target limb posture corresponds to the first limb posture; The duration of the limb postures in the limb posture combination is detected, and in response to the successful matching of the first limb posture with the retrieved target limb posture, the matching result of the limb postures in the limb posture combination and the retrieved target limb posture is determined based on the detected duration and the similarity between the remaining limb postures in the limb posture combination and the retrieved target limb posture, wherein the remaining limb postures are the limb postures in the limb posture combination other than the first limb posture; Based on the matching results, determine whether to control the target door of the autonomous vehicle to unlock and open. Specifically, determining whether to control the target door of the autonomous vehicle to unlock and open based on the matching result includes: in response to the matching result being that each limb posture in the limb posture combination successfully matches the retrieved corresponding target limb posture, controlling the target door to unlock and open.

2. The method according to claim 1, wherein, Based on the detected duration and the similarity between the remaining limb postures in the limb posture combination and the target limb posture, the matching result is determined to include: In response to the similarity being greater than a similarity threshold and the detected duration satisfying the target duration of the target limb posture, the matching result is determined to be a successful match between the remaining limb posture and the target limb posture.

3. The method according to claim 1, wherein, Based on the detected duration and / or the similarity between the remaining limb postures in the limb posture combination and the target limb posture, the matching result is determined as follows: In response to the similarity not being greater than a similarity threshold, and / or the detected duration not meeting the target duration of the target limb posture, the matching result is determined to be a failure to match the remaining limb posture with the target limb posture.

4. The method according to claim 1, wherein, Retrieving the target limb posture corresponding to the limb posture in the target limb posture combination includes: In response to the limb posture being the primary limb posture in the limb posture combination, a target limb posture corresponding to the limb posture in the target limb posture combination is retrieved from the target limb posture combination; or... In response to the fact that the limb posture in the limb posture combination is not the first limb posture, and the previous limb posture in the limb posture combination is successfully matched with the target limb posture in the target limb posture combination corresponding to the previous limb posture, the target limb posture corresponding to the limb posture in the limb posture combination is retrieved from the target limb posture combination.

5. The method according to claim 1, further comprising: In response to a successful match between the limb posture in the limb posture combination and the retrieved target limb posture, and the limb posture in the limb posture combination is not the last limb posture, a target limb posture corresponding to the next limb posture of the limb posture in the limb posture combination is retrieved from the target limb posture combination. The matching result is obtained by matching the next limb posture of the limb posture in the limb posture combination with the target limb posture corresponding to the retrieved next limb posture.

6. The method according to claim 1, wherein, Based on the matching result, determining whether to control the doors of the autonomous vehicle to unlock and open includes: In response to the failure of the limb posture in the limb posture combination to match the retrieved target limb posture, the control of the door of the autonomous vehicle to unlock and open is prohibited.

7. The method according to claim 1, further comprising: In response to the failure of the limb posture in the limb posture combination to match the retrieved target limb posture, a failure prompt message is output, wherein the failure prompt message is used to prompt the passenger to re-execute the limb posture that matches the retrieved target limb posture; Re-identify the body posture of the passenger based on the failure message; The re-identified limb posture is matched with the retrieved target limb posture to obtain the matching result.

8. The method according to claim 1, wherein, In response to sensing the limbs of a passenger in the autonomous vehicle, recognizing combinations of the passenger's limb postures includes: The duration for which the passenger's limbs remain in front of the sensing module is detected, wherein the sensing module is used to sense the passenger's limbs; In response to the detected dwell time being greater than a duration threshold, the combination of body postures of the passenger is identified.

9. The method according to any one of claims 1 to 8, further comprising: Detect the light information of the environment in which the autonomous vehicle is located; In response to the light information being less than the light information threshold, the supplementary lighting device is activated; In response to the light information adjusted by the supplementary lighting device being greater than the light information threshold, the limbs of the passenger in the autonomous vehicle are sensed.

10. The method according to any one of claims 1 to 8, further comprising: In response to sensing the limbs of the passenger, a limb posture prompting information is output, wherein the limb posture prompting information is used to prompt the passenger to perform the target limb posture corresponding to the execution sequence.

11. The method according to any one of claims 1 to 8, further comprising: In response to the limb posture configuration operation, the target limb posture is configured to obtain the target limb posture combination.

12. The method according to any one of claims 1 to 8, wherein the limb includes the hand of the passenger, and the limb posture includes the gesture of the passenger.

13. A control method for an autonomous vehicle, comprising: When the current trip of the autonomous vehicle is detected to have ended, operation guidance information is displayed on the vehicle's in-vehicle interface. The operation guidance information is used to guide the operation of passengers in the autonomous vehicle to control the unlocking and opening of the vehicle's doors. The system senses the limbs of the passenger that remain in the operation guidance information and displays limb posture prompts on the vehicle interface. The limb posture prompts the passenger to perform a target limb posture retrieved from a target limb posture combination. The limbs include the passenger's hands, and the limb postures include the passenger's gestures. The limb posture combination is obtained by capturing the passenger's limb postures through an image acquisition device. The limb postures are a sequence of actions performed by the passenger's limbs. The in-vehicle interface displays the matching results between the limb postures performed by the passenger based on the limb posture prompt information in the limb posture combination and the retrieved corresponding target limb posture. The limb posture combination includes multiple limb postures performed sequentially by the passenger in an execution order. The correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order of the multiple target limb postures. The matching result is determined by matching the first limb posture with the target limb posture in response to the limb posture being the passenger's first limb posture; detecting the duration of the limb postures in the limb posture combination; and determining the similarity between the detected duration and the remaining limb postures in the limb posture combination and the retrieved target limb posture in response to the successful matching of the first limb posture with the retrieved target limb posture. The target limb posture corresponds to the first limb posture, and the remaining limb postures are the limb postures in the limb posture combination other than the first limb posture. The control result of unlocking and opening the target door of the autonomous vehicle based on the matching result is displayed on the in-vehicle interface. The control result for unlocking and opening the doors of the autonomous vehicle based on the matching result displayed on the in-vehicle interface includes: displaying on the in-vehicle interface the matching result for each limb posture in the limb posture combination being successfully matched with the corresponding target limb posture, and the control result for unlocking and opening the target door is to control the target door to unlock and open.

14. A control system for an autonomous vehicle, comprising: A sensing module is used to sense the limbs of passengers in the autonomous vehicle when the current journey of the autonomous vehicle has ended. An image acquisition device is used to acquire the combination of limb postures of the passenger, wherein the combination of limb postures includes multiple limb postures performed sequentially by the passenger in an execution order; A processor is configured to retrieve a target limb posture corresponding to a limb posture in a target limb posture combination, wherein the target limb posture combination includes target limb postures sorted according to the retrieval order, the correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order, the limb includes the hand of the passenger, the limb posture includes the gesture of the passenger, the limb posture combination is obtained by acquiring the limb posture of the passenger through an image acquisition device, and the limb posture is a sequence of actions performed by the limbs of the passenger; In response to the limb posture being the first limb posture of the passenger, the first limb posture is matched with the retrieved target limb posture, wherein the retrieved target limb posture corresponds to the first limb posture; The duration of the limb postures in the limb posture combination is detected, and in response to the successful matching of the first limb posture with the retrieved target limb posture, the matching result of the limb postures in the limb posture combination and the retrieved target limb posture is determined based on the detected duration and the similarity between the remaining limb postures in the limb posture combination and the retrieved target limb posture, wherein the remaining limb postures are the limb postures in the limb posture combination other than the first limb posture; The controller is used to determine, based on the matching result, whether to control the target door of the autonomous vehicle to unlock and open. The controller is configured to determine whether to unlock or open the target door of the autonomous vehicle based on the matching result by performing the following steps: in response to the matching result that each limb posture in the limb posture combination successfully matches the retrieved corresponding target limb posture, the controller controls the target door to unlock or open.

15. The system of claim 14, further comprising: A supplemental lighting device is used to adjust the light information in response to the light information of the environment in which the autonomous vehicle is located being less than a light information threshold. The sensing module is used to sense the limbs of the passenger in the autonomous vehicle in response to the light information adjusted by the supplementary lighting device being greater than the light information threshold.

16. A control device for an autonomous vehicle, comprising: The recognition unit is configured to, upon the termination of the current journey of the autonomous vehicle, in response to sensing the limbs of a passenger in the autonomous vehicle, recognize a combination of limb postures of the passenger, wherein the combination of limb postures includes multiple limb postures performed sequentially by the passenger in an execution order, the limbs include the passenger's hands, the limb postures include the passenger's gestures, the combination of limb postures is obtained by acquiring the limb postures of the passenger through an image acquisition device, and the limb postures are a sequence of actions performed by the passenger's limbs; A retrieval unit is configured to retrieve a target limb posture corresponding to a limb posture in a target limb posture combination, wherein the target limb posture combination includes target limb postures sorted according to the retrieval order, and the correspondence between the limb postures in the limb posture combination and the target limb postures in the target limb posture combination is determined based on the execution order and the retrieval order. A matching unit is configured to, in response to the limb posture being the first limb posture of the passenger, match the first limb posture with the retrieved target limb posture, wherein the retrieved target limb posture corresponds to the first limb posture. The duration of the limb postures in the limb posture combination is detected, and in response to the successful matching of the first limb posture with the retrieved target limb posture, the matching result of the limb postures in the limb posture combination and the retrieved target limb posture is determined based on the detected duration and the similarity between the remaining limb postures in the limb posture combination and the retrieved target limb posture, wherein the remaining limb postures are the limb postures in the limb posture combination other than the first limb posture; The determining unit is configured to determine, based on the matching result, whether to control the target door of the autonomous vehicle to unlock and open. The determining unit is configured to determine whether to control the target door of the autonomous vehicle to unlock and open based on the matching result by performing the following steps: in response to the matching result that each limb posture in the limb posture combination successfully matches the retrieved corresponding target limb posture, controlling the target door to unlock and open.

17. 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 1-13.

18. 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 1-13.

19. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-13.

20. An autonomous vehicle, including the electronic equipment as claimed in claim 17.

Images