Methods, apparatus, devices, and storage media for controlling autonomous vehicles

Abstract

According to embodiments of this disclosure, a method, apparatus, device, and storage medium for controlling an autonomous vehicle are provided. The method includes: determining, by comparing perceived data of the autonomous vehicle with reference map data, a new target fence in a traffic environment associated with the autonomous vehicle, as indicated by the perceived data; determining a target shape representation corresponding to the target fence based on the perceived confidence level of the target fence, the target shape representation being determined from a plurality of candidate shape representations corresponding to different levels of precision; and determining a traffic strategy for the autonomous vehicle based on the target shape representation corresponding to the target fence. In this manner, embodiments of this disclosure can improve the driving safety of autonomous vehicles.

Description

Technical Field

[0001] The exemplary embodiments disclosed herein generally relate to the field of computers, and particularly to a method, apparatus, device, computer-readable storage medium, and computer program product for controlling autonomous vehicles. Background Technology

[0002] With the rapid development of computer technology, autonomous vehicles can predict the behavior trajectory of obstacles, plan the vehicle's driving trajectory, and issue instructions to the vehicle's actuators to ultimately reach the destination safely and smoothly.

[0003] During autonomous vehicle operation, there may be discrepancies between actual road conditions and the map data used by the autonomous vehicle, which may pose risks to the autonomous vehicle's operation. Summary of the Invention

[0004] In a first aspect of this disclosure, a method for controlling an autonomous vehicle is provided. The method includes: determining, by comparing perceived data of the autonomous vehicle with reference map data, a new target fence in a traffic environment associated with the autonomous vehicle, as indicated by the perceived data; determining a target shape representation corresponding to the target fence based on a perceived confidence level of the target fence, the target shape representation being determined from a plurality of candidate shape representations corresponding to different accuracies; and determining a passage strategy for the autonomous vehicle based on the target shape representation corresponding to the target fence.

[0005] In a second aspect of this disclosure, an apparatus for controlling an autonomous vehicle is provided. The apparatus includes: a comparison module configured to determine, by comparing perceived data of the autonomous vehicle with reference map data, a new target fence in a traffic environment associated with the autonomous vehicle, as indicated by the perceived data; a first determination module configured to determine a target shape representation corresponding to the target fence based on a perceived confidence level of the target fence, the target shape representation being determined from a plurality of candidate shape representations corresponding to different accuracies; and a second determination module configured to determine a passage strategy for the autonomous vehicle based on the target shape representation corresponding to the target fence.

[0006] In a third aspect of this disclosure, a computing device is provided. The device includes at least one processing unit and at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit. When executed by the at least one processing unit, the instructions cause the device to perform the method of the first aspect.

[0007] In a fourth aspect of this disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program that can be executed by a processor to implement the method of the first aspect.

[0008] In a fifth aspect of this disclosure, a computer program product is provided. The computer program product includes computer-executable instructions that, when executed by a processor, implement the method of the first aspect.

[0009] It should be understood that the content described in this summary section is not intended to limit the key or essential features of the embodiments of this disclosure, nor is it intended to restrict the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0010] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0011] Figure 1 A schematic diagram of an example environment in which embodiments of the present disclosure can be implemented is shown;

[0012] Figure 2 A flowchart illustrating an example process for controlling an autonomous vehicle according to some embodiments of the present disclosure is shown;

[0013] Figure 3 A flowchart illustrating an example process for controlling an autonomous vehicle according to some embodiments of the present disclosure is shown;

[0014] Figures 4A to 4E A schematic diagram of an example reference scenario of a fence object according to some embodiments of the present disclosure is shown;

[0015] Figure 5 A schematic structural block diagram of an example device for controlling an autonomous vehicle according to certain embodiments of the present disclosure is shown; and

[0016] Figure 6 A block diagram of an apparatus capable of implementing several embodiments of the present disclosure is shown. Detailed Implementation

[0017] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0018] It should be noted that the headings of any section / subsection provided herein are not limiting. Various embodiments are described throughout this document, and embodiments of any type may be included under any section / subsection. Furthermore, embodiments described in any section / subsection may be combined in any way with any other embodiments described in the same section / subsection and / or different sections / subsections.

[0019] In the description of embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit and implicit definitions may also be included below. The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0020] The embodiments of this disclosure may involve user data, data acquisition, and / or use. All of these aspects comply with applicable laws, regulations, and relevant provisions. In the embodiments of this disclosure, all data collection, acquisition, processing, manipulation, forwarding, and use are conducted with the user's knowledge and confirmation. Accordingly, in implementing the embodiments of this disclosure, the type, scope of use, and usage scenarios of any data or information that may be involved should be communicated to the user and their authorization obtained in accordance with relevant laws and regulations through appropriate means. The specific methods of notification and / or authorization may vary depending on the actual situation and application scenario, and the scope of this disclosure is not limited in this respect.

[0021] In this specification and the embodiments, any processing of personal information will be carried out only under the premise of legality (such as obtaining the consent of the personal information subject, or being necessary for the performance of a contract), and will only be carried out within the scope stipulated or agreed upon. A user's refusal to process personal information other than that necessary for basic functions will not affect the user's use of basic functions.

[0022] As briefly mentioned earlier, during autonomous vehicle operation, discrepancies may arise between actual road conditions and the map data used by the vehicle. Specifically, this could include the addition of barriers to the road or barriers being knocked askew into the road. Furthermore, some autonomous vehicles may not use map data that is updated in real time, compromising its accuracy. All of these situations can lead to risks associated with autonomous vehicle operation.

[0023] Embodiments of this disclosure propose a scheme for controlling an autonomous vehicle. The scheme includes: determining, by comparing perceived data from the autonomous vehicle with reference map data, a new target fence in the traffic environment associated with the autonomous vehicle, as indicated by the perceived data; determining a target shape representation corresponding to the target fence based on the perceived confidence level of the target fence, the target shape representation being determined from multiple candidate shape representations corresponding to different levels of precision; and determining a passage strategy for the autonomous vehicle based on the target shape representation corresponding to the target fence.

[0024] In this manner, embodiments of the present disclosure can determine, based on the perceived confidence level of the target fence, how to describe the target fence with a target shape representation of a certain precision, and then determine the passage plan to be taken under that precision target shape representation. This plan can improve the driving safety of autonomous vehicles.

[0025] The following section provides a detailed description of various example implementations of this scheme, with reference to the accompanying drawings.

[0026] Example Environment

[0027] Figure 1 A schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented is shown. For example... Figure 1 As shown, example environment 100 may include electronic devices 110 deployed on autonomous vehicle 120.

[0028] In some embodiments, electronic device 110 communicates with remote device 130 to provide services. Electronic device 110 can be any type of mobile terminal, fixed terminal, or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, handheld computers, portable gaming terminals, VR / AR devices, personal communication system (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, positioning devices, television receivers, radio receivers, e-book devices, gaming devices, or any combination thereof, including accessories and peripherals of these devices or any combination thereof. In some embodiments, electronic device 110 can also support any type of user-facing interface (such as "wearable" circuitry).

[0029] Remote device 130 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and big data and artificial intelligence platforms. Remote device 130 may include, for example, computing systems / servers such as mainframes, edge computing nodes, and computing devices in a cloud environment, etc.

[0030] A communication connection can be established between the remote device 130 and the electronic device 110. This communication connection can be established via wired or wireless means. The communication connection may include, but is not limited to, Bluetooth, mobile network, Universal Serial Bus (USB), and Wireless Fidelity (WiFi) connections; the embodiments of this disclosure are not limited in this respect. In the embodiments of this disclosure, the remote device 130 and the electronic device 110 can achieve signaling interaction through their communication connection.

[0031] It should be understood that the structure and function of the various elements in environment 100 are described for illustrative purposes only and do not imply any limitation on the scope of this disclosure.

[0032] The following description will continue with reference to the accompanying drawings, which will provide some exemplary embodiments of this disclosure.

[0033] Example process

[0034] The following combination Figure 2 , Figure 3 To describe the specific process used to control the autonomous vehicle 120. Figure 2 A flowchart of an example process 200 for controlling an autonomous vehicle 120 according to some embodiments of the present disclosure is shown. Figure 3 A flowchart of an example process 300 for controlling an autonomous vehicle 120 according to some embodiments of the present disclosure is shown. Process 200 can be implemented at electronic device 110. Reference is made below. Figure 1 and Figure 3 To describe process 200.

[0035] like Figure 2 As shown in box 210, electronic device 110 determines, by comparing perception data 310 of autonomous vehicle 120 with reference map data 385, that perception data 310 indicates the addition of a target fence 320 in the traffic environment associated with autonomous vehicle 120.

[0036] It should be understood that during the operation of the autonomous vehicle 120, the autonomous vehicle 120 can detect its surrounding traffic environment through its configured sensing devices. For example, the sensing devices can detect obstacles such as fences in the traffic environment where the autonomous vehicle 120 is located. These obstacles may be temporarily placed or newly added. For the autonomous vehicle 120, these obstacles are not recorded in the map data it uses, making it difficult for the autonomous vehicle 120 to respond accurately to these obstacles, which may lead to potential risks for the autonomous vehicle 120. Therefore, the autonomous vehicle 120 needs to determine whether there are any newly added obstacles in its traffic environment through the sensing data 310 collected by the sensing devices.

[0037] The following explanation uses a fence as an example of an obstacle.

[0038] In some embodiments, the target fence 320 indicates a newly added fence in the traffic environment where the autonomous vehicle 120 is located. The target fence 320 can be determined by comparing the perceived data 310 and the reference map data 385. The reference map data 385 records the existing entities in the traffic environment and their positional relationships, and can be obtained, for example, from a high-definition map 380. The perceived data 310 reflects the real-time existence of the entities in the traffic environment and their positional relationships.

[0039] As an example, the sensed data 310 can be obtained in one or more ways. Taking the acquisition of sensed data 310 in two ways as an example, the electronic device 110 can acquire sensed data 310 using Occupation Network Technology (OCC) and an online map, respectively.

[0040] Specifically, the perception data 310 acquired by the electronic device 110 using occupancy network technology may include the location information, contour information, category information, and corresponding confidence level of the target fence 320. In some specific examples, the electronic device 110 may use a static fusion target detection method to obtain the aforementioned perception data 310. For example, the electronic device 110 may combine visual data from a camera, precise distance information from a lidar, and the near-field sensing capability of an ultrasonic sensor to obtain the aforementioned perception data 310.

[0041] The perception data 310 obtained by the electronic device 110 using the online map may include the location information, outline information, category information and corresponding confidence level of the target fence 320.

[0042] Furthermore, as an example, in the case where perceptual data 310 is acquired using Occupation Networking (OCC) technology and an online map respectively, the electronic device 110 can first fuse the perceptual data 310 acquired using OCC technology with the perceptual data 310 acquired using the online map, and then compare the fused perceptual data 310 with the reference map data 385 to determine the target fence 320. The electronic device 110 can, for example, use a union method for fusion. If the location information, contour information, and confidence level of the target fence 320 are inconsistent, the electronic device 110 can, for example, use a mean calculation method to unify them.

[0043] As an example, the electronic device 110 can first determine the target area in the reference map data 385 corresponding to the location of the target fence 320, and then compare the fused perception data 310 related to the target fence 320 with the target area to determine whether the target fence 320 is a newly added entity object.

[0044] Once the electronic device 110 determines that the target fence 320 is a newly added entity, it can simultaneously determine the location information, category information, change category, changed shape information, and perception confidence level 350 of the target fence 320. Among them, the perception confidence level 350 is the fused confidence level, and the change category can specifically correspond to the situation of adding a fence, replacing a fence, or incorrectly placing a fence.

[0045] In frame 220, electronic device 110 determines a target shape representation corresponding to target fence 320 based on the perceived confidence 350 of target fence 320. The target shape representation is determined from multiple candidate shape representations corresponding to different precisions.

[0046] In some embodiments, candidate shape representations can indicate shape information of the target fence 320. Candidate shape representations of different precision indicate different degrees of fit between the indicated shape information and the actual shape of the target fence 320. Higher precision candidate shape representations indicate a higher degree of fit between the indicated shape information and the actual shape of the target fence 320. Multiple candidate shape representations can, for example, be obtained from sensing data 310 associated with the target fence 320.

[0047] For example, consider two candidate shape representations for the same target fence 320: a circle and a rectangle. The candidate shape representation presented as a rectangle has higher precision than the candidate shape representation presented as a circle.

[0048] It should be understood that the autonomous vehicle 120's perception capability of the target fence 320 varies depending on the precision of the target shape representation, and correspondingly, the decisions made by the autonomous vehicle 120 will also differ. For driving safety reasons, the autonomous vehicle 120 should try to avoid unknown obstacles by detouring. However, when facing known obstacles, the autonomous vehicle 120 can make more refined decisions. Nevertheless, this is contingent on the accuracy of the information related to the obstacle.

[0049] Based on this, the electronic device 110 can determine the target shape representation corresponding to the target fence 320 based on the perceived confidence level 350 of the target fence 320. In some embodiments, the electronic device 110 can determine the target shape representation based on the following process: in response to the perceived confidence level 350 reaching a threshold, the electronic device 110 determines the target shape representation based on the first shape representation 330 corresponding to the first precision.

[0050] Specifically, when there are multiple candidate shape representations, the precision corresponding to each candidate shape representation is also divided into multiple precision levels. When the electronic device 110 determines that the perception confidence level 350 is greater than the threshold, it can determine the first precision with the highest precision level, and then determine the first shape representation 330 corresponding to the first precision. The first shape representation 330 is the target shape representation.

[0051] In some embodiments, the electronic device 110 may also determine the target shape representation based on the following process: in response to a perceived confidence level 350 being less than a threshold, the electronic device 110 determines the target shape representation based on a second shape representation 340 corresponding to a second precision. Wherein, the first precision is higher than the second precision.

[0052] Specifically, when there are multiple candidate shape representations, the precision corresponding to each candidate shape representation is also divided into multiple precision levels. When the electronic device 110 determines that the perception confidence level 350 is less than a threshold, it can determine the second precision with the lowest precision level, and then determine the second shape representation 340 corresponding to the second precision. The second shape representation 340 is the target shape representation.

[0053] Furthermore, in some embodiments, two candidate shape representations may be provided, namely a first shape representation 330 and a second shape representation 340. As an example, the electronic device 110 may determine whether to use the first shape representation 330 or the second shape representation 340 as the target shape representation based on the relationship between the perceived confidence 350 of the target fence 320 and a threshold.

[0054] Specifically, when the perceived confidence level 350 reaches a threshold, the electronic device 110 determines the first shape representation 330 as the target shape representation. Conversely, when the perceived confidence level 350 is less than the threshold, the electronic device 110 determines the second shape representation 340 as the target shape representation. The threshold can be, for example, 90%, and can also be adaptively adjusted according to actual conditions.

[0055] To give a specific example, the first shape 330 could be a geometry, and the second shape 340 could be a polygon. Although both geometry and polygon describe and represent the shape of the target fence 320 using points, lines, and surfaces, geometry is closer to the actual shape of the target fence 320.

[0056] Additionally, the first shape representation 330 can also be associated with semantic information to indicate the first shape representation 330 regarding the fence object. For example, polygon can display the shape of the target fence; geometry can not only display the shape of the target fence, but also carry attribute information such as the word "fence".

[0057] Additionally, the electronic device 110 can also consider the detection accuracy of the target fence 320 when determining the target shape representation based on the perception confidence 350.

[0058] In some embodiments, the electronic device 110 may determine the detection accuracy based on historical detection data associated with the fence object. As an example, the detection accuracy may be the ratio of the number of times a new fence was correctly detected in the historical detection data to the total number of times a new fence was detected in the historical detection data.

[0059] In some embodiments, taking two candidate shape representations, namely a first shape representation 330 and a second shape representation 340, as an example, the specific process by which the electronic device 110 determines the target shape representation can be as follows: in response to the perception confidence 350 reaching a threshold and the detection accuracy reaching an accuracy threshold, the target shape representation is determined based on the first shape representation 330 corresponding to the first accuracy; and in response to the perception confidence 350 being less than the threshold or the detection accuracy being less than the accuracy threshold, the target shape representation is determined based on the second shape representation 340 corresponding to the second accuracy. The accuracy threshold can be adaptively set according to actual conditions.

[0060] Additionally, in cases where there are multiple candidate shape representations, once the electronic device 110 has determined the target shape representation, the accuracy of the target shape representation can be improved by updating the target shape representation.

[0061] Specifically, in some embodiments, the electronic device 110 may update the target shape representation based on the following process:

[0062] First, the electronic device 110 determines whether the perceived confidence 350 at the target time meets the preset conditions based on the change of the perceived confidence 350 within a predetermined time period.

[0063] As an example, after the electronic device 110 determines the target shape representation, it can continuously acquire the perception confidence 350 of the target fence 320 through perception data 310 within a predetermined time period after determining the target shape representation. The predetermined time period can be defined, for example, by time or distance. That is, the duration of the predetermined time period can be defined, or the time required to travel a certain distance can be defined. In a specific example, the predetermined time period could be the time it takes for the autonomous vehicle 120 to travel from its current location to a position 15m away from the target fence 320. The predetermined time period can be adaptively adjusted according to actual conditions.

[0064] Furthermore, considering that the perception confidence level 350 can affect the accuracy of the target shape representation—that is, the higher the perception confidence level 350, the higher the accuracy of the target shape representation—the electronic device 110 only needs to detect a higher perception confidence level 350 within a predetermined time period to improve the accuracy of the target shape representation. The moment when the perception confidence level 350 increases compared to the perception confidence level 350 corresponding to the target shape representation is the target moment.

[0065] As an example, the degree to which the perceived confidence level 350 at the target time is improved compared to the perceived confidence level 350 corresponding to the target shape representation can also affect the accuracy of the target shape representation. The requirement for the degree of improvement in the perceived confidence level 350 is the preset condition. The preset condition could be, for example, that the perceived confidence level 350 at the target time improves to a confidence threshold, or that the percentage improvement of the perceived confidence level 350 at the target time reaches a percentage threshold. The percentage improvement is the ratio of the difference between the perceived confidence level 350 at the target time and the perceived confidence level 350 corresponding to the target shape representation to the perceived confidence level 350 corresponding to the target shape representation. Both the confidence threshold and the percentage threshold can be adaptively set according to actual circumstances.

[0066] Then, in response to the target time's perception confidence 350 meeting the preset conditions, the electronic device 110 updates the target shape representation to improve the accuracy of the target shape representation.

[0067] As an example, when the perceived confidence level 350 at the target time meets a preset condition, the electronic device 110 can update the target shape representation to the candidate shape representation corresponding to the next accuracy level. Additionally, the electronic device 110 can also set multiple different confidence thresholds or percentage thresholds. When the perceived confidence level 350 at the target time reaches different confidence thresholds or percentage thresholds, the electronic device 110 can update the target shape representation to the candidate shape representation of the corresponding accuracy level.

[0068] Taking the candidate shape representation as an example, there are two types: the first shape representation 330 and the second shape representation 340. When the electronic device 110 determines that the target shape representation is the second shape representation 340, if the perception confidence 350 of the target at that time meets the preset conditions, the electronic device 110 will update the target shape representation to the first shape representation 330.

[0069] In frame 230, electronic device 110 determines passage strategy 370 for autonomous vehicle 120 based on the target shape representation corresponding to target fence 320.

[0070] In some embodiments, the target shape representations corresponding to different accuracies can correspond to different traffic strategies 370, enabling the autonomous vehicle 120 to drive more safely. Specifically, the electronic device 110 can determine the traffic scenario corresponding to the traffic environment in response to the target shape representation having an accuracy higher than a preset accuracy, and then determine a traffic strategy 370 that matches the traffic scenario from a set of preset strategies.

[0071] As an example, the traffic scene corresponding to the traffic environment determined by the electronic device 110 can indicate the positional relationship between the target fence 320 and at least one reference road element. The reference road element can be various entities on the road, such as the autonomous vehicle 120, intersections, road boundaries, etc. Accordingly, the positional relationship between the target fence 320 and at least one reference road element can include, for example, the position of the target fence 320, the positional relationship between the target fence 320 and the autonomous vehicle 120, and the positional relationship between the target fence 320 and the intersection.

[0072] As an example, when the electronic device 110 determines a traffic strategy 370 matching a traffic scenario from a set of preset strategies, it can first determine a reference scenario 360 matching the traffic scenario, and then determine the traffic strategy 370 corresponding to the reference scenario 360. The reference scenario 360 can be, for example, a fence within the target area, a fence on the shoulder, a fence inside the shoulder or outside the lane boundary line, a fence at a curve at an intersection, a fence on the lane line, or a fence occupying lane space. See [reference needed] for details. Figures 4A to 4E The scene diagram shown is illustrated below. Figure 4A Scene 400A shown represents the fence within the target area; Figure 4B Scene 400B shown is a fence on the road shoulder; Figure 4C The scenario 400C shown is a fence inside the shoulder or outside the lane boundary line; Figure 4D Scene 400D shown is a fence at a bend in an intersection; Figure 4E The scene shown, 400E, is a fence positioned on the lane line.

[0073] The following section uses two reference scenarios 360—a fence on a lane line or a fence occupying lane space—as examples to further explain the process of determining the corresponding traffic strategy 370.

[0074] For a reference scenario 360 such as a fence positioned on a lane line, the matching traffic scenario can indicate that the target fence 320 is on a lane line and within the area corresponding to the intersection, or it can indicate that the target fence 320 is on a lane line and outside the area corresponding to the intersection. As an example, the area corresponding to the intersection can be defined using a distance threshold.

[0075] In a specific example, in response to a traffic scenario indicating that the target fence 320 is on the first lane and its distance from the intersection does not exceed a first threshold, the electronic device 110 can update the lane topology associated with the first lane to indicate that lane changing is prohibited on the first lane; then, based on the updated lane topology, a traffic strategy 370 is determined. The first threshold is a distance threshold, for example, 100m. The first threshold can be adaptively adjusted according to actual conditions.

[0076] It should be understood that since the target fence 320 is located within the area corresponding to the intersection, the autonomous vehicle 120 needs to consider whether it will affect the originally planned trajectory when making the decision to change lanes.

[0077] Specifically, electronic device 110 can determine traffic strategy 370 based on the updated lane topology through the following process: if changing lanes would affect the originally planned trajectory, it means that autonomous vehicle 120 cannot change lanes and can only travel in its current lane and pass through the intersection. At this time, electronic device 110 determines traffic strategy 370 to travel in its current lane.

[0078] Conversely, if changing lanes does not affect the originally planned trajectory, it indicates that there are multiple lanes that can pass through the intersection. In this case, the electronic device 110 determines the traffic strategy 370 to change lanes in advance to other lanes that can pass through the intersection. In addition, when the autonomous vehicle 120 fails to change lanes, the electronic device 110 determines the traffic strategy 370 to proceed according to the current lane.

[0079] In a specific example, in response to a traffic scenario indicating that the target fence 320 is on the second lane and the distance to the intersection is greater than a second threshold, the electronic device 110 can control the autonomous vehicle 120 to change to the first lane, which is unrelated to the target fence 320. The second threshold is a distance threshold, for example, 100m. The second threshold can be adaptively adjusted according to actual conditions.

[0080] Additionally, in response to the failure of the autonomous vehicle 120 to change to the first lane, the electronic device 110 can determine a second lane to be driven in based on obstacle information and / or traffic flow information of the candidate lane. The candidate lane can be a lane unrelated to the target fence 320.

[0081] If the autonomous vehicle 120 fails to change lanes to the first lane, it can also change lanes to other lanes. When determining a second lane from at least one candidate lane, the electronic device 110 can base its determination on obstacle information and / or traffic flow information. Specifically, the electronic device 110 can determine a lane without obstacles as the second lane from at least one candidate lane, or a lane with less traffic from at least one candidate lane, or a lane with both obstacles and less traffic from at least one candidate lane.

[0082] Furthermore, before the electronic device 110 controls the autonomous vehicle 120 to change to the first lane, the electronic device 110 can also determine the first lane from at least one candidate lane in the manner described above.

[0083] Additionally, when the electronic device 110 determines a first lane from at least one candidate lane, if there is no candidate lane unrelated to the target fence 320, the traffic strategy 370 determined by the electronic device 110 is to control the autonomous vehicle 120 to change to an obstacle-free lane.

[0084] For a reference scenario 360 such as a fence occupying lane space, the matching traffic scenario can indicate that the target fence 320 is located within the target lane and within the area corresponding to the intersection, or it can indicate that the target fence 320 is within the target lane and outside the area corresponding to the intersection. As an example, the area corresponding to the intersection can also be divided using a distance threshold.

[0085] In a specific example, in response to a traffic scenario where the target fence 320 is located within the target lane and the distance to the intersection does not exceed a third threshold, electronic device 110 can determine the passage strategy 370 of autonomous vehicle 120 based on the passable space of the target lane. Here, the target lane is the lane currently occupied by autonomous vehicle 120. The third threshold is a distance threshold, for example, 100m. The third threshold can be adaptively adjusted according to actual conditions.

[0086] It should be understood that the situation where the target fence 320 is located within the target lane may have occurred due to a vehicle being knocked off course. In this case, the target fence 320 obstructs the normal passage space of its lane. For the autonomous vehicle 120, the electronic equipment 110 needs to further determine the passable space of the target lane in order to determine the passage strategy 370.

[0087] As an example, electronic device 110 can determine traffic strategy 370 based on the following process: in response to the size of the passable space being smaller than a preset size, controlling autonomous vehicle 120 to change to another lane. The preset size can be, for example, 2.7m, or can be adaptively adjusted according to actual conditions.

[0088] Additionally, when electronic device 110 fails to control autonomous vehicle 120 to change to another lane, electronic device 110 may adopt the following traffic strategy 370: control autonomous vehicle 120 to stop in the target lane and initiate an assistance request. As an example, electronic device 110 may initiate an assistance request to remote device 130.

[0089] Furthermore, considering that the determined traffic strategy 370 may affect the originally planned trajectory, the electronic device 110 can also determine the traffic strategy 370 based on the following process: in response to the size of the passable space reaching a preset size, determine whether the target lane is the only lane for the autonomous vehicle 120 to pass through the intersection; and then determine the traffic strategy 370 for the above two situations respectively.

[0090] Specifically, in response to the fact that the target lane is the only lane for passing through the intersection, the electronic device 110 can control the autonomous vehicle to travel through the passable space to pass through the intersection. In other words, if changing lanes would affect the originally planned trajectory, it means that the target lane is the only lane for passing through the intersection. The autonomous vehicle 120 cannot change lanes and can only travel through the intersection in its own lane. At this time, the electronic device 110 determines the passage strategy 370 to control the autonomous vehicle 120 to travel through the passable space to pass through the intersection, that is, to control the autonomous vehicle 120 to slowly push the target fence 320 forward within the target lane to pass through the intersection.

[0091] Conversely, if the target lane is not the only lane through the intersection, the electronic device 110 can control the autonomous vehicle 120 to change to another lane used to pass through the intersection. In other words, if changing lanes does not affect the originally planned trajectory, it means that there are multiple lanes that can pass through the intersection. At this time, the electronic device 110 determines the passage strategy 370 to control the autonomous vehicle 120 to change to another lane used to pass through the intersection. In addition, when the autonomous vehicle 120 fails to change lanes, the electronic device 110 can control the autonomous vehicle 120 to travel through the passable space to pass through the intersection, that is, control the autonomous vehicle 120 to slowly push the target fence 320 forward within the target lane to pass through the intersection.

[0092] In a specific example, in response to a traffic scenario where the target fence 320 is located within the target lane and the distance to the intersection is greater than a fourth threshold, electronic device 110 can determine the passage strategy 370 of autonomous vehicle 120 based on the passable space of the target lane. The fourth threshold is a distance threshold, for example, 100m. The fourth threshold can be adaptively adjusted according to actual conditions.

[0093] Similar to the previous traffic scenario, when determining the passage strategy 370 in this traffic scenario, the electronic device 110 also needs to conduct further analysis based on the size relationship between the size of the passable space and the preset size.

[0094] As an example, when the size of the passable space is smaller than the preset size, the specific process by which the electronic device 110 determines the passage strategy 370 can be the same as the process of determining the passage strategy 370 of the previous traffic scenario, and will not be elaborated on here.

[0095] Furthermore, in response to the passage space reaching a preset size, the electronic device 110 can control the autonomous vehicle 120 to change to another lane for passing through the intersection. In addition, when the autonomous vehicle 120 fails to change lanes, the electronic device 110 can control the autonomous vehicle 120 to travel through the passage space to pass through the intersection, that is, control the autonomous vehicle 120 to slowly push the target fence 320 forward in the target lane to pass through the intersection.

[0096] Furthermore, the electronic device 110 can respond to situations where the accuracy corresponding to the target shape representation is lower than a preset accuracy by controlling the autonomous vehicle 120 to detour around the target fence 320 or controlling the autonomous vehicle 120 to stop within the current lane. The preset accuracy can be adaptively adjusted according to actual conditions.

[0097] Taking two candidate shape representations, namely a first shape representation 330 and a second shape representation 340, as an example, when the target shape representation is the first shape representation 330, the accuracy corresponding to the target shape representation is higher than the preset accuracy. In this case, the electronic device 110 can determine a traffic strategy 370 matching the traffic scenario from a set of preset strategies based on the traffic scene indicated by the traffic environment. When the target shape representation is the second shape representation 340, the accuracy corresponding to the target shape representation is lower than the preset accuracy. In this case, the electronic device 110 can control the autonomous vehicle 120 to bypass the target fence 320 or control the autonomous vehicle 120 to stop within the current lane.

[0098] Additionally, after determining the target shape representation, the electronic device 110 can also send the target shape representation to the remote device 130 so that the remote device 130 can update the high-precision map 380.

[0099] In some embodiments, the remote device 130 can be configured as follows:

[0100] First, a map is constructed using the target shape representation to generate a layer representation corresponding to the target fence 320. Then, the layer representation corresponding to the target fence 320 is sent to at least one autonomous vehicle 120 to add map data related to the target fence 320. The layer representation can be, for example, a geometry layer.

[0101] The remote device 130 can quickly synchronize map data that has not been updated on the high-precision map 380 to at least one autonomous vehicle 120 by using the layer representation generated by the target shape representation, so as to improve the map data before the release of a new version of the high-precision map 380, thereby facilitating the autonomous vehicle 120 to make decisions during driving.

[0102] Additionally, when sending layer representations to at least one autonomous vehicle 120, the remote device 130 may first determine the map data that the target autonomous vehicle 120 needs to supplement based on the timestamp of the target autonomous vehicle 120, then fuse at least one layer representation corresponding to the map data that needs to be supplemented, and finally send the fused layer representation to the target autonomous vehicle 120. As an example, when fusing at least one layer representation, the remote device 130 may fuse at least one layer representation corresponding to the same target fence 320.

[0103] Additionally, remote device 130 can compile high-precision map 380 using annotations generated based on multiple target shape representations to update high-precision map 380 and release a new version of high-precision map 380. Through the new version of high-precision map 380 released by remote device 130, autonomous vehicle 120 can obtain a download link to download the new version of high-precision map 380 when an update is detected.

[0104] Example devices and equipment

[0105] Figure 5 A schematic structural block diagram of a device 500 for controlling an autonomous vehicle according to certain embodiments of the present disclosure is shown. The device 500 may be implemented as or included in an autonomous vehicle 110. Various modules / components in the device 500 may be implemented by hardware, software, firmware, or any combination thereof.

[0106] As shown in the figure, the device 500 includes a comparison module 510 configured to determine, by comparing the perception data of the autonomous vehicle with reference map data, a new target fence in the traffic environment associated with the autonomous vehicle indicated by the perception data; a first determination module 520 configured to determine the target shape representation corresponding to the target fence based on the perception confidence of the target fence, wherein the target shape representation is determined from multiple candidate shape representations corresponding to different accuracies; and a second determination module 530 configured to determine the autonomous vehicle's passage strategy based on the target shape representation corresponding to the target fence.

[0107] In some embodiments, determining the target shape representation corresponding to the target fence based on the perceived confidence of the target fence includes: determining the target shape representation based on the first shape representation corresponding to the first precision in response to the perceived confidence reaching a threshold; or determining the target shape representation based on the second shape representation corresponding to the second precision in response to the perceived confidence being less than the threshold, wherein the first precision is higher than the second precision.

[0108] In some embodiments, the first shape representation is also associated with semantic information to indicate that the first shape representation is associated with a fence object.

[0109] In some embodiments, the system further includes an update module configured to determine whether the perceived confidence at a target time meets a preset condition based on the change in perceived confidence over a predetermined period; and to update the target shape representation in response to the perceived confidence at the target time meeting the preset condition, so as to improve the accuracy of the target shape representation.

[0110] In some embodiments, determining the passage strategy of an autonomous vehicle based on the target shape representation corresponding to the target fence includes: determining the traffic scenario corresponding to the traffic environment in response to the accuracy of the target shape representation being higher than a preset accuracy; and determining the passage strategy that matches the traffic scenario from a set of preset strategies.

[0111] In some embodiments, the traffic scene indicates the positional relationship between the target fence and at least one reference road element.

[0112] In some embodiments, determining a traffic strategy matching a traffic scenario from a set of preset strategies includes: updating the lane topology associated with the first lane line in response to a traffic scenario indicating that the target fence is on the first lane line and the distance to the intersection does not exceed a first threshold, so as to indicate that lane changing is prohibited on the first lane line; and determining a traffic strategy based on the updated lane topology.

[0113] In some embodiments, determining a traffic strategy that matches the traffic scenario from a set of preset strategies includes: in response to a traffic scenario indicating that a target fence is on a second lane and the distance from the intersection is greater than a second threshold, controlling the autonomous vehicle to change to a first lane that is unrelated to the target fence.

[0114] In some embodiments, in response to the failure of the autonomous vehicle to change to the first lane, a second lane to be driven is determined based on obstacle information and / or traffic flow information of a set of candidate lanes.

[0115] In some embodiments, determining a traffic strategy that matches a traffic scenario from a set of preset strategies includes: in response to a traffic scenario where a target fence is located within a target lane, determining a traffic strategy for the autonomous vehicle based on the passable space of the target lane.

[0116] In some embodiments, determining the autonomous vehicle's passage strategy based on the passable space of the target lane includes: controlling the autonomous vehicle to change to another lane in response to the size of the passable space being smaller than a preset size.

[0117] In some embodiments, in response to the failure of the autonomous vehicle to change to another lane, the autonomous vehicle is controlled to stop in the target lane and an assistance request is initiated.

[0118] In some embodiments, determining the autonomous vehicle's passage strategy based on the passable space of the target lane includes: in response to the size of the passable space reaching a preset size and the distance between the target fence and the intersection not exceeding a third threshold, determining whether the target lane is the only lane for the autonomous vehicle to pass through the intersection; and in response to the target lane being the only lane for passing through the intersection, controlling the autonomous vehicle to travel through the passable space to pass through the intersection.

[0119] In some embodiments, determining the autonomous vehicle's passage strategy based on the passable space of the target lane further includes: in response to the target lane not being the only lane for passing through the intersection, controlling the autonomous vehicle to change to another lane for passing through the intersection.

[0120] In some embodiments, determining the passage strategy of an autonomous vehicle based on the target shape representation corresponding to the target fence includes: in response to the accuracy corresponding to the target shape representation being lower than a preset accuracy, controlling the autonomous vehicle to detour through the target fence or controlling the autonomous vehicle to stop in the current lane.

[0121] In some embodiments, the system further includes a sending module configured to send a target shape representation to a remote device, the remote device being configured to: construct a map using the target shape representation to generate a layer representation corresponding to the target fence; and send the layer representation corresponding to the target fence to at least one autonomous vehicle to add map data related to the target fence.

[0122] Figure 6 A block diagram of a computing device 600 in which one or more embodiments of the present disclosure may be implemented is shown. It should be understood that... Figure 6 The computing device 600 shown is merely exemplary and should not be construed as limiting the functionality and scope of the embodiments described herein. Figure 6 The computing device 600 shown can be used to implement Figure 1 Electronic devices 110.

[0123] like Figure 6 As shown, computing device 600 is in the form of a general-purpose computing device. Components of computing device 600 may include, but are not limited to, one or more processors or processing units 610, memory 620, storage devices 630, one or more communication units 640, one or more input devices 650, and one or more output devices 660. Processing unit 610 may be a physical or virtual processor and is capable of performing various processes according to programs stored in memory 620. In a multiprocessor system, multiple processing units execute computer-executable instructions in parallel to improve the parallel processing capability of computing device 600.

[0124] Computing device 600 typically includes multiple computer storage media. Such media can be any accessible media that is accessible to computing device 600, including but not limited to volatile and non-volatile media, removable and non-removable media. Memory 620 can be volatile memory (e.g., registers, cache, random access memory (RAM)), non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. Storage device 630 can be removable or non-removable media and can include machine-readable media, such as flash drives, disks, or any other media that can be used to store information and / or data (e.g., training data for training) and can be accessed within computing device 600.

[0125] The computing device 600 may further include additional removable / non-removable, volatile / non-volatile storage media. Although not explicitly stated... Figure 6As shown, disk drives for reading from or writing to removable, non-volatile disks (e.g., "floppy disks") and optical disk drives for reading from or writing to removable, non-volatile optical disks can be provided. In these cases, each drive can be connected to a bus (not shown) via one or more data media interfaces. Memory 620 may include computer program product 625 having one or more program modules configured to perform various methods or actions of various embodiments of this disclosure.

[0126] The communication unit 640 enables communication with other computing devices via a communication medium. Additionally, the components of the computing device 600 can function as a single computing cluster or multiple computing machines capable of communicating via communication connections. Therefore, the computing device 600 can operate in a networked environment using logical connections to one or more other servers, networked personal computers (PCs), or another network node.

[0127] Input device 650 can be one or more input devices, such as a mouse, keyboard, trackball, etc. Output device 660 can be one or more output devices, such as a monitor, speaker, printer, etc. Computing device 600 can also communicate as needed with one or more external devices (not shown) via communication unit 640. These external devices, such as storage devices, display devices, etc., can communicate with one or more devices that enable user interaction with computing device 600, or with any device (e.g., network card, modem, etc.) that enables computing device 600 to communicate with one or more other computing devices. Such communication can be performed via input / output (I / O) interfaces (not shown).

[0128] According to an exemplary implementation of this disclosure, a computer-readable storage medium is provided that stores computer-executable instructions thereon, wherein the computer-executable instructions are executed by a processor to implement the methods described above. According to an exemplary implementation of this disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, which are executed by a processor to implement the methods described above.

[0129] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatuses, devices, and computer program products implemented according to this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0130] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processing unit of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0131] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0132] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0133] Various implementations of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is chosen to best explain the principles, practical applications, or improvements to technology in the market, or to enable others skilled in the art to understand the various implementations disclosed herein.

Claims

1. A method for controlling an autonomous vehicle, comprising: By comparing the perception data of the autonomous vehicle with the reference map data, it is determined that the perception data indicates the addition of a target fence in the traffic environment associated with the autonomous vehicle. Based on the perceived confidence of the target fence, a target shape representation corresponding to the target fence is determined, wherein the target shape representation is determined from multiple candidate shape representations corresponding to different precisions; as well as Based on the target shape representation corresponding to the target fence, the passage strategy of the autonomous vehicle is determined.

2. The method according to claim 1, wherein determining the target shape representation corresponding to the target fence based on the perceived confidence of the target fence includes: In response to the perception confidence reaching a threshold, the target shape representation is determined based on a first shape representation corresponding to a first precision; or In response to the perception confidence being less than the threshold, the target shape representation is determined based on a second shape representation corresponding to the second precision, wherein the first precision is higher than the second precision.

3. The method of claim 2, wherein the first shape representation is further associated with semantic information to indicate that the first shape representation is associated with a fence object.

4. The method according to claim 1, further comprising: Based on the change of the perceived confidence level within a predetermined time period, determine whether the perceived confidence level at the target time meets the preset conditions; as well as In response to the perception confidence at the target time meeting a preset condition, the target shape representation is updated to improve the accuracy of the target shape representation.

5. The method according to claim 1, wherein determining the passage strategy of the autonomous vehicle based on the target shape representation corresponding to the target fence includes: In response to the target shape representation having a precision higher than a preset precision, the traffic scene corresponding to the traffic environment is determined; as well as The traffic strategy that matches the traffic scenario is determined from a set of preset strategies.

6. The method of claim 5, wherein the traffic scenario indicates the positional relationship between the target fence and at least one reference road element.

7. The method of claim 5, wherein determining the traffic strategy matching the traffic scenario from a set of preset strategies comprises: In response to the traffic scenario indicating that the target fence is on the first lane and the distance to the intersection does not exceed a first threshold, the lane topology associated with the first lane is updated to indicate that lane changing is prohibited on the first lane. as well as The traffic strategy is determined based on the updated lane topology.

8. The method of claim 5, wherein determining the traffic strategy matching the traffic scenario from a set of preset strategies comprises: In response to the traffic scenario indicating that the target fence is on the second lane and the distance from the intersection is greater than a second threshold, the autonomous vehicle is controlled to change to the first lane, which is unrelated to the target fence.

9. The method according to claim 8, further comprising: In response to the failure of the autonomous vehicle to change to the first lane, a second lane to be driven is determined based on obstacle information and / or traffic flow information of a set of candidate lanes.

10. The method of claim 5, wherein determining the traffic strategy matching the traffic scenario from a set of preset strategies comprises: In response to the traffic scenario where the target fence is located within the target lane, the autonomous vehicle determines its passage strategy based on the passable space of the target lane.

11. The method of claim 10, wherein determining the autonomous vehicle's traffic strategy based on the passable space of the target lane comprises: In response to the fact that the size of the passable space is smaller than a preset size, the autonomous vehicle is controlled to change to another lane.

12. The method of claim 11, further comprising: In response to the failure of the autonomous vehicle to change to the other lane, the autonomous vehicle is controlled to stop in the target lane and an assistance request is initiated.

13. The method of claim 10, wherein determining the autonomous vehicle's traffic strategy based on the passable space of the target lane comprises: In response to the fact that the size of the passable space reaches a preset size and the distance between the target fence and the intersection does not exceed a third threshold, it is determined whether the target lane is the only lane for the autonomous vehicle to pass through the intersection; as well as In response to the fact that the target lane is the only lane for passing through the intersection, the autonomous vehicle is controlled to travel through the passable space to pass through the intersection.

14. The method of claim 13, wherein determining the autonomous vehicle's traffic strategy based on the passable space of the target lane further comprises: In response to the fact that the target lane is not the only lane for passing through the intersection, the autonomous vehicle is controlled to change to another lane for passing through the intersection.

15. The method of claim 1, wherein determining the passage strategy of the autonomous vehicle based on the target shape representation corresponding to the target fence includes: In response to the target shape representation having a precision lower than a preset precision, the autonomous vehicle is controlled to either detour around the target fence or stop within the current lane.

16. The method according to claim 1, further comprising: The target shape representation is sent to a remote device configured to: The target shape is used to construct a map, thereby generating a layer representation corresponding to the target fence; as well as The layer representation corresponding to the target fence is sent to at least one autonomous vehicle to add map data related to the target fence.

17. A device for controlling an autonomous vehicle, comprising: The comparison module is configured to determine, by comparing the perception data of the autonomous vehicle with reference map data, the perception data indicating the addition of a target fence in the traffic environment associated with the autonomous vehicle. The first determining module is configured to determine a target shape representation corresponding to the target fence based on the perceived confidence of the target fence, wherein the target shape representation is determined from multiple candidate shape representations corresponding to different precisions; as well as The second determining module is configured to determine the passage strategy of the autonomous vehicle based on the target shape representation corresponding to the target fence.

18. A computing device, comprising: At least one processing unit; as well as At least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions causing the computing device to perform the method according to any one of claims 1 to 16 when executed by the at least one processing unit.

19. A computer-readable storage medium having a computer program stored thereon, the computer program being executable by a processor to implement the method according to any one of claims 1 to 16.

20. A computer program product comprising computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, implement the method according to any one of claims 1 to 16.

Images