Method, system and device for navigating autonomous vehicle to pass intersection

Abstract

The invention relates to a method (200) for navigating an autonomous vehicle through an intersection, comprising: obtaining trajectory information (220) of a tracked object; navigating the autonomous vehicle during a first navigation cycle using the trajectory information (230); verifying the tracked object in real time (240); updating the trajectory information based on a verification result of the tracked object (250); and navigating the autonomous vehicle (260) using the updated trajectory information during a second navigation cycle, where the second navigation cycle is a next navigation cycle after the first navigation cycle.

Description

Technical Field

[0001] The present invention relates to the technical field of navigating an autonomous driving vehicle through an intersection, and in particular to a method, system, device and readable storage medium for navigating an autonomous driving vehicle through an intersection by following the trajectory of an object. Background Art

[0002] The successful design and deployment of autonomous vehicles (AVs) requires finding the best road / lane-level path from the AV's current location to a specified destination in a map that includes the road network. Routes often require crossing intersections, which include common areas that vehicles can approach and pass from multiple directions. Such intersections are one of the more challenging scenarios for AVs to handle, as decisions about when to enter an intersection must take into account traffic controls (i.e., signs or traffic signals) and the predicted actions of other road users.

[0003] When following the target vehicle 120 through an intersection, many complex situations arise. For example, when an object (e.g., another vehicle 130 other than the target vehicle, a cyclist, a pedestrian, an animal, etc.) cuts into the original driving trajectory of the AV 110 at the intersection, it causes a change in the driving trajectory, thereby causing the steering wheel of the AV to shake (see Figure 1 ). For example, when the target vehicle 120 is lost while crossing the intersection (i.e., the vehicle 120' is not detected), the AV autonomous driving function is terminated, causing the AV 110 to stop at the intersection (see Figure 2 ). Summary of the invention

[0004] In view of the above problems in the prior art, the embodiments of the present application specification provide a method, system, device and readable storage medium for navigating an autonomous driving vehicle through an intersection by following the trajectory of an object, which improves the stability of the autonomous driving vehicle through the intersection and reduces the steering wheel jitter of the autonomous driving vehicle.

[0005] According to one aspect of an embodiment of the present invention, a method for navigating an autonomous driving vehicle through an intersection is provided, comprising:

[0006] Obtaining trajectory information of the tracked object;

[0007] navigating the autonomous vehicle during a first navigation cycle using the trajectory information;

[0008] Verifying the tracked object in real time;

[0009] updating the trajectory information based on the verification result of the tracked object; and

[0010] The automated driving vehicle is navigated during a second navigation cycle using the updated trajectory information, wherein the second navigation cycle is a next navigation cycle after the first navigation cycle.

[0011] In the context of this application, the term "intersection" may include, for example, a crossroad, a T-junction, a Y-junction, etc. The term "navigation cycle" refers to the period during which the autonomous driving vehicle navigates according to the planned trajectory, for example, 0.1 seconds, because the planned trajectory needs to be constantly updated to adapt to the real-time changing road conditions. Those skilled in the art will appreciate that the "navigation cycle" may also be other time periods besides 0.1 seconds.

[0012] A "tracked object" may be, for example, a target vehicle when planning a trajectory. A "new object" may be, for example, other vehicles other than the target vehicle, cyclists, pedestrians, animals, etc.

[0013] In one example, verifying the tracked object may include:

[0014] determining whether the trajectory of the new object intersects the navigation path of the autonomous vehicle, and

[0015] determining whether the distance from the new object to the autonomous vehicle is less than the distance from the tracked object to the autonomous vehicle,

[0016] Among them, if it is determined that the trajectory of the new object intersects with the navigation path of the autonomous driving vehicle and the distance from the new object to the autonomous driving vehicle is less than the distance from the tracked object to the autonomous driving vehicle, the new object is determined as a candidate object.

[0017] In this example, verifying the tracked object may further include: determining whether the candidate object exists, and determining whether the tracked object exists.

[0018] In this example, if the candidate object exists and the tracked object does not exist, the method further includes: determining the candidate object as the final tracked object during the second navigation cycle as the verification result.

[0019] In this example, the method may also include determining whether the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is within the left and right boundaries of the width of the autonomous driving vehicle.

[0020] In this example, if it is determined that the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is not within the left and right boundaries of the width of the autonomous driving vehicle, the method further includes:

[0021] determining the tracked object as the final tracked object during the second navigation cycle as the verification result,

[0022] Otherwise, the candidate object is determined as the final tracked object during the second navigation cycle as the verification result.

[0023] In this example, if the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is still navigated using the trajectory information of the tracked object during the second navigation cycle, or if the verification result is that the candidate object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is navigated using the historical trajectory of the candidate object during the second navigation cycle.

[0024] In another example, verifying the tracked object may include determining whether the tracked object exists.

[0025] In this example, if it is determined that the tracked object exists, the method further includes: determining the tracked object as the final tracked object during the second navigation cycle as the verification result; or

[0026] If it is determined that the tracked object does not exist, the method further includes comparing the time when the tracked object does not exist with a predetermined time.

[0027] In this example, if the tracked object does not exist for a period greater than the predetermined time, the method further includes: no longer taking the tracked object as the final tracked object during the second navigation cycle.

[0028] In this example, if the time during which the tracked object does not exist is less than the predetermined time, the method further includes:

[0029] Re-searching for objects around the autonomous vehicle; and

[0030] It is identified whether there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle.

[0031] In this example, if the list of objects being searched has an object that is the same as the tracked object during the first navigation cycle, the method further includes:

[0032] determining the tracked object as the final tracked object during the second navigation period as the verification result;

[0033] Otherwise, the tracked object is no longer regarded as the final tracked object during the second navigation cycle.

[0034] According to another aspect of an embodiment of the present invention, a system for navigating an autonomous driving vehicle through an intersection is provided, comprising:

[0035] A memory for storing instruction codes; and

[0036] at least one processor coupled to the memory for executing the stored instruction code, the instruction code being configured to:

[0037] Obtaining trajectory information of the tracked object;

[0038] navigating the autonomous vehicle during a first navigation cycle using the trajectory information;

[0039] Verifying the tracked object in real time;

[0040] updating the trajectory information based on the verification result of the tracked object; and

[0041] The automated driving vehicle is navigated during a second navigation cycle using the updated trajectory information, wherein the second navigation cycle is a next navigation cycle after the first navigation cycle.

[0042] In one example, the instruction code may also be configured to:

[0043] determining whether the trajectory of the new object intersects the navigation path of the autonomous vehicle, and

[0044] determining whether the distance from the new object to the autonomous vehicle is less than the distance from the tracked object to the autonomous vehicle,

[0045] Among them, if it is determined that the trajectory of the new object intersects with the navigation path of the autonomous driving vehicle and the distance from the new object to the autonomous driving vehicle is less than the distance from the tracked object to the autonomous driving vehicle, the new object is determined as a candidate object.

[0046] In this example, the instruction code may also be configured to: determine whether the candidate object exists, and determine whether the tracked object exists.

[0047] In this example, if the candidate object exists and the tracked object does not exist, the instruction code is further configured to: determine the candidate object as the final tracked object during the second navigation cycle as the verification result.

[0048] In this example, the instruction code may also be configured to determine whether the lateral distance of the candidate object's historical trajectory relative to the navigation path of the autonomous driving vehicle is within the left and right boundaries of the width of the autonomous driving vehicle.

[0049] In this example, if it is determined that the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is not within the left and right boundaries of the width of the autonomous driving vehicle, the instruction code is further configured to:

[0050] determining the tracked object as the final tracked object during the second navigation cycle as the verification result,

[0051] Otherwise, the candidate object is determined as the final tracked object during the second navigation cycle as the verification result.

[0052] In this example, if the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is still navigated using the trajectory information of the tracked object during the second navigation cycle, or if the verification result is that the candidate object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is navigated using the historical trajectory of the candidate object during the second navigation cycle.

[0053] In another example, the instruction code may be further configured to: determine whether the tracked object exists.

[0054] In this example, if it is determined that the tracked object exists, the instruction code is further configured to: determine the tracked object as the final tracked object during the second navigation cycle as the verification result; or

[0055] If it is determined that the tracked object does not exist, the instruction code is further configured to: compare the time when the tracked object does not exist with a predetermined time.

[0056] In this example, if the tracked object does not exist for a period greater than the predetermined time, the instruction code is further configured to: no longer regard the tracked object as the final tracked object during the second navigation cycle.

[0057] In this example, if the time during which the tracked object is absent is less than the predetermined time, the instruction code is further configured to:

[0058] Re-searching for objects around the autonomous vehicle; and

[0059] It is identified whether there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle.

[0060] In this example, if the list of objects being searched has an object that is the same as the tracked object during the first navigation cycle, the instruction code is further configured to:

[0061] determining the tracked object as the final tracked object during the second navigation period as the verification result;

[0062] Otherwise, the tracked object is no longer regarded as the final tracked object during the second navigation cycle.

[0063] According to another aspect of an embodiment of the present invention, there is provided a device for navigating an autonomous driving vehicle through an intersection, comprising:

[0064] A trajectory information acquisition unit, used to obtain trajectory information of the tracked object;

[0065] a navigation unit for navigating the autonomous driving vehicle during a first navigation cycle using the trajectory information;

[0066] a verification unit, configured to verify the tracked object in real time; and

[0067] a trajectory updating unit, configured to update the trajectory information based on a verification result of the tracked object;

[0068] The navigation unit is further used to navigate the autonomous driving vehicle during a second navigation cycle using the updated trajectory information, wherein the second navigation cycle is a next navigation cycle after the first navigation cycle.

[0069] In one example, the verification unit may also be used to:

[0070] determining whether the trajectory of the new object intersects the navigation path of the autonomous vehicle, and

[0071] determining whether the distance from the new object to the autonomous vehicle is less than the distance from the tracked object to the autonomous vehicle,

[0072] Among them, if it is determined that the trajectory of the new object intersects with the navigation path of the autonomous driving vehicle and the distance from the new object to the autonomous driving vehicle is less than the distance from the tracked object to the autonomous driving vehicle, the new object is determined as a candidate object.

[0073] In this example, the verification unit may also be configured to: determine whether the candidate object exists, and determine whether the tracked object exists.

[0074] In this example, if the candidate object exists and the tracked object does not exist, the verification unit is further configured to: determine the candidate object as the final tracked object during the second navigation cycle as the verification result.

[0075] In this example, the verification unit may also be used to determine whether the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is within the left and right boundaries of the width of the autonomous driving vehicle.

[0076] In this example, if it is determined that the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is not within the left and right boundaries of the width of the autonomous driving vehicle, the verification unit is further configured to:

[0077] determining the tracked object as the final tracked object during the second navigation cycle as the verification result,

[0078] Otherwise, the candidate object is determined as the final tracked object during the second navigation cycle as the verification result.

[0079] In this example, if the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is still navigated using the trajectory information of the tracked object during the second navigation cycle, or if the verification result is that the candidate object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is navigated using the historical trajectory of the candidate object during the second navigation cycle.

[0080] In another example, the verification unit may also be used to determine whether the tracked object exists.

[0081] In this example, if it is determined that the tracked object exists, the verification unit is further configured to: determine the tracked object as the final tracked object during the second navigation cycle as the verification result; or

[0082] If it is determined that the tracked object does not exist, the verification unit is further configured to compare the time when the tracked object does not exist with a predetermined time.

[0083] In this example, if the tracked object does not exist for a period greater than the predetermined time, the verification unit is further configured to: no longer regard the tracked object as the final tracked object during the second navigation cycle.

[0084] In this example, if the tracked object is absent for less than the predetermined time, the verification unit is further configured to:

[0085] Re-searching for objects around the autonomous vehicle; and

[0086] It is identified whether there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle.

[0087] In this example, if there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle, the verification unit is further configured to:

[0088] determining the tracked object as the final tracked object during the second navigation period as the verification result;

[0089] Otherwise, the tracked object is no longer regarded as the final tracked object during the second navigation cycle.

[0090] According to another aspect of an embodiment of the present invention, there is provided a machine-readable storage medium having executable instructions stored thereon, wherein the executable instructions, when executed, cause the machine to execute the method described in the preceding paragraphs.

[0091] From the above, it can be seen that the solution of the embodiment of the present application improves the stability of the autonomous driving vehicle passing through the intersection and reduces the steering wheel vibration of the autonomous driving vehicle. The solution improves the usability of the autonomous driving or assisted driving function, the stability of control, and the customer experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] The features, characteristics, advantages and benefits of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings.

[0093] Figure 1 An exemplary scenario of an autonomous vehicle 110 passing through an intersection is shown;

[0094] Figure 2 Another exemplary scenario is shown when the autonomous driving vehicle 110 passes through an intersection;

[0095] Figure 3 A moving object trajectory estimation device 1 according to an embodiment of the present invention is shown;

[0096] Figure 4 An overall flow chart of a method 200 for navigating an autonomous driving vehicle 110 through an intersection according to an embodiment of the present invention is shown;

[0097] Figure 5An exemplary scenario of an autonomous driving vehicle 110 passing through an intersection according to an embodiment of the present invention is shown;

[0098] Figure 6 Shown for Figure 5 Flowchart of the operations performed by the embodiment of the present invention;

[0099] Figure 7 Another exemplary scenario of an autonomous driving vehicle 110 passing through an intersection according to another embodiment of the present invention is shown;

[0100] Figure 8 Shown for Figure 7 Flowchart of the operations performed by the embodiment of the present invention;

[0101] Fig. 9 A schematic diagram showing a system 500 for navigating an autonomous vehicle through an intersection according to one embodiment of the present invention; and

[0102] Fig.10 A schematic diagram of an apparatus 600 for navigating an autonomous driving vehicle through an intersection according to an embodiment of the present invention is shown.

[0103] Reference numerals list :

[0104] 110: Autonomous vehicles

[0105] 120: Target vehicle (as tracked object)

[0106] 120': Lost (target) vehicle

[0107] 125: Navigation path (for autonomous vehicles)

[0108] 130: (inserted) new vehicle

[0109] 135: New vehicle (predicted driving) trajectory

[0110] 137: New Vehicle History Track

[0111] 200: Method for navigating an autonomous vehicle through an intersection

[0112] 210: Start

[0113] 220: Get the trajectory information of the tracked object

[0114] 230: Navigating the autonomous vehicle during the first navigation cycle using trajectory information

[0115] 240: Real-time verification of tracked objects

[0116] 250: Update the trajectory information based on the verification results of the tracked object

[0117] 260: Navigating the autonomous vehicle during a second navigation cycle using the updated trajectory information

[0118] 270: End

[0119] 310: Determine whether the trajectory of the new object intersects the navigation path of the autonomous vehicle

[0120] 320: Determine whether the distance from the new object to the autonomous driving vehicle is less than the distance from the tracked object to the autonomous driving vehicle 330: Determine the new object as a candidate object

[0121] 340: Determine whether the candidate object and the tracked object exist

[0122] 350: Determine the candidate object as the final tracked object during the second navigation cycle

[0123] 360: Determine whether the lateral distance of the candidate object's historical trajectory relative to the autonomous vehicle's navigation path is within the left and right boundaries of the autonomous vehicle's width

[0124] 370: Determine the tracked object as the final tracked object during the second navigation cycle

[0125] 410: Determine whether the tracked object exists

[0126] 420: Determine the tracked object as the final tracked object during the second navigation cycle

[0127] 430: Compare the time when the tracked object is not present with the predetermined time

[0128] 440: The tracked object is no longer considered as the final tracked object during the second navigation cycle

[0129] 450: Re-searching for objects around autonomous vehicles

[0130] 460: Identify whether there is an object in the list of searched objects that is the same as the object tracked during the first navigation cycle

[0131] 470: Determine the tracked object as the final tracked object during the second navigation cycle

[0132] 500: System for navigating autonomous vehicles through intersections

[0133] 510: Memory

[0134] 520: At least one processor

[0135] 600: Device for navigating an autonomous vehicle through an intersection

[0136] 610: Trajectory information acquisition unit

[0137] 620: Navigation unit

[0138] 630: Verification Unit

[0139] 640: Trajectory update unit DETAILED DESCRIPTION

[0140] Hereinafter, various embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0141] like Figure 5 and Figure 7 As shown, when the autonomous driving vehicle 110 follows the trajectory 125 of the target vehicle (i.e., the “tracked object”) 120 through the intersection, in one scenario, a new vehicle (i.e., the “new object”) 130 cuts into the original driving trajectory of the autonomous driving vehicle 110 (see FIG. Figure 5 ), in another scenario, the autonomous driving vehicle 110 loses tracking of the target vehicle 120 (see Figure 7 ). Both scenarios will affect the autonomous driving vehicle passing through the intersection. To this end, the present invention aims to provide a method, system, device and readable storage medium for navigating an autonomous driving vehicle through an intersection by following the trajectory of an object, which improves the stability of the autonomous driving vehicle passing through the intersection and reduces the steering wheel vibration of the autonomous driving vehicle.

[0142] Now refer to the attached Figure 4 , the method 200 of the present invention for navigating an autonomous driving vehicle through an intersection, after starting 210, preferably in step 220, obtains trajectory information of the tracked object.

[0143] According to one embodiment of the present application, the autonomous driving vehicle 110 may include a moving object trajectory estimation device 1, which includes a plurality of networked electronic control units (ECUs) that may be communicatively coupled via a bus structure to perform control functions and information sharing, including executing control routines locally or in a distributed manner. Figure 3 As shown, the mobile object trajectory estimation device 1 has an object detection electronic control unit (ECU) 5, a position calculation ECU 6, an observable object extraction ECU 7, and an object trajectory prediction ECU 8. The ECUs each perform their own control and are configured by, for example, a central processing unit (CPU), a read-only memory (ROM), a random access memory (RAM), an input signal circuit, an output signal circuit, a power supply circuit, etc. The object detection ECU 5 is connected to the camera 2 and the laser radar 3. The position calculation ECU 6 is connected to the global positioning system (GPS) receiver 4.

[0144] Camera 2 may be a monocular camera, a stereo camera, an infrared camera, etc., and is used to obtain the situation around the autonomous driving vehicle 110 by capturing images of objects such as other vehicles, pedestrians, roadside objects, etc.

[0145] The laser radar 3 emits a laser beam around the autonomous driving vehicle 110 while scanning in the horizontal direction of the autonomous driving vehicle 110, and receives waves reflected from the surface of another vehicle (such as vehicles 120, 130) or a pedestrian to detect the distance, direction and speed of the other vehicle or pedestrian.

[0146] The GPS receiver 4 receives GPS satellite signals to determine the position of the autonomous driving vehicle 110, and detects the positioning of the autonomous driving vehicle 110 based on the received GPS satellite signals. The GPS receiver 4 outputs the determined position of the autonomous driving vehicle 110 to the position calculation ECU 6.

[0147] The object detection ECU 5 is a surrounding information acquisition device for acquiring information about the surroundings of the autonomous driving vehicle 110, acquires the image signal output by the camera 2 and the signals of a plurality of other vehicles (e.g., vehicles 120, 130) output by the laser radar 3, and detects a plurality of other vehicles. Then, the object detection ECU 5 outputs the detected other vehicles to the position calculation ECU 6.

[0148] The position calculation ECU 6 is connected to the object detection ECU 5, and can specify an object from a plurality of other vehicles detected by the object detection ECU 5. For example, from a plurality of oncoming vehicles traveling on an oncoming lane, a vehicle closest to the autonomous driving vehicle 110 can be selected. In addition, the position calculation ECU 6 calculates the absolute position of the target vehicle based on the information about the specified vehicle (referred to as the "target vehicle or tracked object") output by the GPS receiver 4 and the absolute position of the autonomous driving vehicle 110. Then, the position calculation ECU 6 outputs the calculated absolute position for the target vehicle to the observable object extraction ECU 7.

[0149] The observable object extraction ECU 7 is connected to the position calculation ECU 6 and the map information storage device 9. In the map information storage device 9, road information or map information including information on structures around the road is stored.

[0150] The observable object extraction ECU 7 as the identification information acquisition means extracts observable objects from the target vehicle based on the absolute position of the target vehicle output from the position calculation ECU 6 and the map information around the automatic driving vehicle 110 output from the map information storage device 9. Then, the observable object extraction ECU 7 outputs information on the extracted observable objects of the target vehicle to the object trajectory prediction ECU 8.

[0151] The object trajectory prediction ECU 8 as a trajectory estimation device generates a trajectory predicted by each observable object based on the information about the observable object from the target vehicle extracted by the observable object extraction ECU 7, and predicts the trajectory of the target vehicle or the tracked object based on the generated result. The object trajectory prediction ECU 8 then outputs the predicted trajectory of the tracked object to the output section 10. The output section 10 determines the trajectory of the autonomous driving vehicle 110 in response to the result of the predicted trajectory of the tracked object, for example, and automatically controls the steering actuator.

[0152] Thus, in this embodiment, the autonomous driving vehicle 110 obtains the trajectory information of the tracked object (e.g., the target vehicle 120) from the moving object trajectory estimation device 1. In another embodiment, the autonomous driving vehicle 110 may obtain the trajectory information of the tracked object (e.g., the target vehicle 120) in other ways.

[0153] If the trajectory information is obtained in step 220, the method 200 jumps to step 230 and uses the trajectory information to navigate the autonomous driving vehicle 110 during the first navigation cycle. In the context of the present application, a "navigation cycle" refers to a period during which the autonomous driving vehicle navigates according to the planned trajectory, for example, 0.1 seconds, because the planned trajectory needs to be constantly updated to adapt to real-time changing road conditions. Those skilled in the art will appreciate that the "navigation cycle" can also be a time period other than 0.1 seconds. Therefore, the "first navigation cycle" refers to the first 0.1 seconds of navigation after the trajectory information is obtained. Based on the trajectory information obtained, known autonomous driving or assisted driving technologies can be used to navigate the autonomous driving vehicle 110.

[0154] During navigation of the autonomous driving vehicle 110, the method 200 jumps to step 240 to verify the tracked object in real time. Then, in step 250, the trajectory information is updated based on the verification result of the tracked object.

[0155] Below, steps 240 - 250 are described in detail in combination with two specific scenarios.

[0156] In one example, referring to Figure 5-6 To describe the specific operation of the present invention when a new vehicle (ie, a “new object”) 130 cuts into the original driving trajectory of the autonomous driving vehicle 110 .

[0157] In this example, the step 240 of verifying the tracked object may include determining the trajectory 135 of the new vehicle 130 at block 310 (see Figure 1) intersects with the navigation path 125 of the autonomous driving vehicle 110, and at block 320 determines whether the distance from the new vehicle 130 to the autonomous driving vehicle 110 is less than the distance from the tracked object (e.g., the target vehicle 120) to the autonomous driving vehicle 110. The trajectory 135 of the new vehicle 130 may be obtained by, for example, predicting the running trajectory of the new vehicle 130 by the object trajectory prediction ECU 8. Further, other ECUs (not shown) of the autonomous driving vehicle 110 may calculate whether the trajectory 135 of the new vehicle 130 intersects with the navigation path 125 of the autonomous driving vehicle 110. In addition, the distance from the new vehicle 130 to the autonomous driving vehicle 110 and the distance from the target vehicle 120 to the autonomous driving vehicle 110 may be determined by, for example, the position calculation ECU 6. Further, other ECUs (not shown) of the autonomous driving vehicle 110 may calculate which vehicle is closest to the autonomous driving vehicle 110.

[0158] If the trajectory 135 of the new vehicle 130 intersects the navigation path 125 of the autonomous driving vehicle 110, it is determined that the new vehicle 130 cuts into the driving trajectory of the autonomous driving vehicle 110. In this case, if the distance from the new vehicle 130 to the autonomous driving vehicle 110 is less than the distance from the target vehicle 120 to the autonomous driving vehicle 110, the new vehicle 130 (i.e., the new object) is determined as a candidate object or vehicle at block 330.

[0159] In this example, the step 240 of verifying the tracked object may also preferably include: determining whether the candidate object (i.e., the new vehicle 130) exists, and determining whether the tracked object (e.g., the target vehicle 120) exists at box 340. This may be performed, for example, by the object detection ECU 5. Alternatively, objects around the autonomous driving vehicle 110 may be detected by other known methods.

[0160] If the candidate object (i.e., the new vehicle 130) exists and the tracked object does not exist (e.g., the target vehicle 120 is lost), the candidate object (i.e., the new vehicle 130) is determined as the final tracked object during the second navigation cycle as the verification result at box 350. Therefore, returning to step 250, it is determined how to update the trajectory information according to the determined final tracked object, that is, it is determined which object or vehicle's trajectory information is used to navigate the autonomous driving vehicle 110 during the second navigation cycle. Here, the second navigation cycle is the next navigation cycle after the first navigation cycle, for example, the second 0.1 second.

[0161] If both the candidate object (ie, the new vehicle 130) and the tracked object (eg, the target vehicle 120) are present, then at block 360 the historical trajectory 137 (see Figure 5 ) is the lateral distance relative to the navigation path 125 of the autonomous driving vehicle 110 at the left and right boundaries of the (vehicle) width of the autonomous driving vehicle 110 (e.g. Figure 5 Here, the historical trajectory 137 of the new vehicle 130 is obtained by, for example, tracking the new vehicle 130 by the moving object trajectory estimation device 1, and the trajectory 135 of the new vehicle 130 is obtained by, for example, predicting the driving trajectory of the new vehicle 130 by the object trajectory prediction ECU 8. The navigation path 125 of the autonomous driving vehicle 110 may be, for example, the driving path of the autonomous driving vehicle 110 during the first navigation cycle, in which case the historical trajectory 137 of the new vehicle 130 may be the center line of the driving trajectory of the new vehicle 130 during the time period corresponding to the first navigation cycle. The left and right boundaries of the vehicle width of the autonomous driving vehicle 110 are known, and therefore, various methods may be used to calculate whether (the center line of) the historical trajectory 137 of the new vehicle 130 is within the historical trajectory 137 of the new vehicle 130.

[0162] Next, if it is determined that the lateral distance of the historical trajectory 137 of the new vehicle 130 relative to the navigation path 125 of the autonomous driving vehicle 110 is not within the left and right boundaries of the width of the autonomous driving vehicle 110, the method further includes:

[0163] At block 370, determining the tracked object (ie, the target vehicle 120) as the final tracked object during the second navigation cycle as the verification result;

[0164] Otherwise, at block 350 , the candidate object (ie, the new vehicle 130 ) is determined as the final tracked object during the second navigation cycle as the verification result.

[0165] Regardless of whether the verification result determines the target vehicle 120 or the new vehicle 130 as the final tracked object during the second navigation cycle, step 250 will determine how to update the trajectory information according to the determined final tracked object, as described above.

[0166] In one example, referring to Figure 7-8 The specific operation of the present invention will be described when the autonomous driving vehicle 110 loses tracking of the target vehicle 120 (ie, the vehicle 120' cannot be detected).

[0167] In this example, the step 240 of verifying the tracked object may include determining whether the tracked object exists at block 410. For example, the existence of the target vehicle 120 may be determined by the object detection ECU 5. Alternatively, objects (such as the target vehicle 120) around the autonomous driving vehicle 110 may be detected by other known methods.

[0168] If it is determined that the tracked object (e.g., target vehicle 120) exists, the tracked object (e.g., target vehicle 120) is determined as the final tracked object during the second navigation cycle as the verification result at box 420. Therefore, returning to step 250, it is determined how to update the trajectory information based on the determined final tracked object, that is, it is determined that the trajectory information of, for example, the target vehicle 120 will still be used to navigate the autonomous driving vehicle 110 during the second navigation cycle. Here, the second navigation cycle is the next navigation cycle after the first navigation cycle, for example, the second 0.1 second.

[0169] Alternatively, if it is determined that the tracked object (e.g., target vehicle 120) does not exist (i.e., vehicle 120' is not detected), the time when the tracked object (e.g., target vehicle 120) does not exist is compared with a predetermined time at box 430. For example, once the object detection ECU 5 or otherwise fails to detect the target vehicle 120, the autonomous driving vehicle 110 may also include a timer (not shown), which will start timing at this time and use the recorded time as the time when the target vehicle 120 does not exist. The predetermined time may be pre-set when the vehicle leaves the factory, or may be manually set by an engineer, for example, when the vehicle is being repaired. The predetermined time may be, for example, 1 second. In other examples, the predetermined time may also be set to be greater than 1 second or less than 1 second.

[0170] If the time during which the tracked object (e.g., the target vehicle 120) does not exist (e.g., the time recorded by the timer since the object detection ECU 5 cannot detect the target vehicle 120) is less than the predetermined time (e.g., 1 second), the objects around the autonomous driving vehicle 110 are searched again at box 450, and preferably, it is possible to identify in the list of searched objects whether there is the same object as the tracked object (e.g., the target vehicle 120) during the first navigation cycle. For example, the operation of re-searching the surrounding objects can be performed by the object detection ECU 5, and a list of searched objects can be generated. Here, each object or vehicle detected by the object detection ECU 5 is assigned an ID. Furthermore, other ECUs (not shown) of the autonomous driving vehicle 110 can identify in the list of searched objects whether there is the same vehicle as the target vehicle 120, for example, by comparing the IDs.

[0171] If the time during which the tracked object (e.g., target vehicle 120) does not exist (e.g., the time recorded by the timer starting from when the object detection ECU 5 fails to detect the target vehicle 120) is greater than the predetermined time (e.g., 1 second), the target vehicle 120 is no longer used as the (final) tracked object of the subsequent (i.e., during the second navigation cycle) of the autonomous driving vehicle 110 at block 440. In this case, the autonomous driving vehicle 110 can, for example, re-search the surrounding objects or vehicles through the object detection ECU 5 to re-determine the tracked object or vehicle, and then perform subsequent navigation.

[0172] If there is no object in the list of searched objects that is the same as the tracked object (e.g., target vehicle 120) during the first navigation cycle, then the target vehicle 120 is no longer used as the (final) tracked object of the subsequent (i.e., during the second navigation cycle) of the autonomous driving vehicle 110 at box 440. In this case, the autonomous driving vehicle 110 can also re-search the surrounding objects or vehicles, for example, through the object detection ECU 5, so as to re-determine the tracked object or vehicle, and then perform subsequent navigation.

[0173] If there is an object in the list of objects being searched that is the same as the tracked object (e.g., target vehicle 120) during the first navigation cycle, the tracked object is determined as the final tracked object during the second navigation cycle at block 470. That is, the autonomous driving vehicle 110 is still navigated according to the trajectory information of the target vehicle 120 during the second navigation cycle.

[0174] Back to Figure 4 , method 200 proceeds to step 260, navigating the autonomous vehicle during a second navigation cycle using the updated trajectory information. Figure 5-6 In the illustrated embodiment, if the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the trajectory information of the tracked object is still used to navigate the autonomous driving vehicle during the second navigation cycle; or, if the verification result is that the candidate object is determined to be the final tracked object during the second navigation cycle, the historical trajectory of the candidate object is used to navigate the autonomous driving vehicle during the second navigation cycle. Figure 7-8In the illustrated embodiment, if the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the trajectory information of the tracked object is still used to navigate the autonomous vehicle during the second navigation cycle, or, as described above, the target vehicle 120 is no longer used as a subsequent tracked object of the autonomous vehicle 110. Finally, after performing the navigation task of the autonomous vehicle 110, the method 200 is explained 270.

[0175] Fig. 9 A schematic diagram of a system 500 for navigating an autonomous driving vehicle through an intersection according to an embodiment of the present invention is shown. Figure 4 , 6 , 8 method to describe system 500 in detail.

[0176] The system 500 includes a memory 510 and at least one processor 520 coupled to the memory 510. Fig. 9 The memory 510 is configured to store instruction codes. When the instruction codes are executed by at least one processor 520, the at least one processor 520 executes Figure 4 , 6 , the methods and operations described in 8.

[0177] Specifically, when the instruction code is executed, the at least one processor 520 may:

[0178] Obtaining trajectory information of the tracked object;

[0179] navigating the autonomous vehicle during a first navigation cycle using the trajectory information;

[0180] Verifying the tracked object in real time;

[0181] updating the trajectory information based on the verification result of the tracked object; and

[0182] The automated driving vehicle is navigated during a second navigation cycle using the updated trajectory information, wherein the second navigation cycle is a next navigation cycle after the first navigation cycle.

[0183] In one example, the instruction code may also be configured to:

[0184] determining whether the trajectory of the new object intersects the navigation path of the autonomous vehicle, and

[0185] determining whether the distance from the new object to the autonomous vehicle is less than the distance from the tracked object to the autonomous vehicle,

[0186] Among them, if it is determined that the trajectory of the new object intersects with the navigation path of the autonomous driving vehicle and the distance from the new object to the autonomous driving vehicle is less than the distance from the tracked object to the autonomous driving vehicle, the new object is determined as a candidate object.

[0187] In this example, the instruction code may also be configured to:

[0188] It is determined whether the candidate object exists, and it is determined whether the tracked object exists.

[0189] In this example, if the candidate object exists and the tracked object does not exist, the instruction code is further configured to: determine the candidate object as the final tracked object during the second navigation cycle as the verification result.

[0190] In this example, the instruction code may also be configured to:

[0191] Determine whether the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is within the left and right boundaries of the width of the autonomous driving vehicle.

[0192] In this example, if it is determined that the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is not within the left and right boundaries of the width of the autonomous driving vehicle, the instruction code is further configured to:

[0193] determining the tracked object as the final tracked object during the second navigation cycle as the verification result,

[0194] Otherwise, the candidate object is determined as the final tracked object during the second navigation cycle as the verification result.

[0195] In this example, if the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is still navigated using the trajectory information of the tracked object during the second navigation cycle, or

[0196] If the verification result is that the candidate object is determined to be the final tracked object during the second navigation period, the autonomous driving vehicle is navigated using the historical trajectory of the candidate object during the second navigation period.

[0197] In another example, the instruction code may be further configured to: determine whether the tracked object exists.

[0198] In this example, if it is determined that the tracked object exists, the instruction code is further configured to: determine the tracked object as the final tracked object during the second navigation cycle as the verification result; or

[0199] If it is determined that the tracked object does not exist, the instruction code is further configured to: compare the time when the tracked object does not exist with a predetermined time.

[0200] In this example, if the tracked object does not exist for a period greater than the predetermined time, the instruction code is further configured to: no longer regard the tracked object as the final tracked object during the second navigation cycle.

[0201] In this example, if the time during which the tracked object is absent is less than the predetermined time, the instruction code is further configured to:

[0202] Re-searching for objects around the autonomous vehicle; and

[0203] It is identified whether there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle.

[0204] In this example, if the list of objects being searched has an object that is the same as the tracked object during the first navigation cycle, the instruction code is further configured to:

[0205] determining the tracked object as the final tracked object during the second navigation period as the verification result;

[0206] Otherwise, the tracked object is no longer regarded as the final tracked object during the second navigation cycle.

[0207] Fig.10 A schematic diagram of an apparatus 600 for navigating an autonomous driving vehicle through an intersection according to an embodiment of the present invention is shown. Fig.10 The device 600 shown can be implemented by software, hardware, or a combination of software and hardware. Fig.10 The device 400 shown can be integrated, for example, in Fig. 9 system 500.

[0208] like Fig.10 As shown, the apparatus 600 may include a trajectory information acquiring unit 610 , a navigation unit 620 , a verification unit 630 , and a trajectory updating unit 640 .

[0209] The navigation unit 620 is configured to navigate the autonomous driving vehicle during a first navigation cycle using the trajectory information.

[0210] The verification unit 630 is configured to verify the tracked object in real time.

[0211] The trajectory updating unit 640 is configured to update the trajectory information based on the verification result of the tracked object.

[0212] In this example, the navigation unit 620 is further configured to navigate the autonomous driving vehicle during a second navigation cycle using the updated trajectory information, wherein the second navigation cycle is a next navigation cycle after the first navigation cycle.

[0213] In this example, the verification unit 630 is further configured to:

[0214] determining whether the trajectory of the new object intersects the navigation path of the autonomous vehicle, and

[0215] determining whether the distance from the new object to the autonomous vehicle is less than the distance from the tracked object to the autonomous vehicle,

[0216] Among them, if it is determined that the trajectory of the new object intersects with the navigation path of the autonomous driving vehicle and the distance from the new object to the autonomous driving vehicle is less than the distance from the tracked object to the autonomous driving vehicle, the new object is determined as a candidate object.

[0217] In this example, the verification unit 630 is further configured to:

[0218] It is determined whether the candidate object exists, and it is determined whether the tracked object exists.

[0219] In this example, if the candidate object exists and the tracked object does not exist, the verification unit 630 is further configured to: determine the candidate object as the final tracked object during the second navigation cycle as the verification result.

[0220] In this example, the verification unit 630 is further configured to determine whether the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is within the left and right boundaries of the width of the autonomous driving vehicle.

[0221] In this example, if it is determined that the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is not within the left and right boundaries of the width of the autonomous driving vehicle, the verification unit 630 is further configured to:

[0222] determining the tracked object as the final tracked object during the second navigation cycle as the verification result,

[0223] Otherwise, the candidate object is determined as the final tracked object during the second navigation cycle as the verification result.

[0224] In this example, if the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is still navigated using the trajectory information of the tracked object during the second navigation cycle, or if the verification result is that the candidate object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is navigated using the historical trajectory of the candidate object during the second navigation cycle.

[0225] In another example, the verification unit 630 is further configured to: determine whether the tracked object exists.

[0226] In this example, if it is determined that the tracked object exists, the verification unit 630 is further configured to: determine the tracked object as the final tracked object during the second navigation cycle as the verification result; or

[0227] If it is determined that the tracked object does not exist, the verification unit 630 is further configured to compare the time when the tracked object does not exist with a predetermined time.

[0228] In this example, if the tracked object does not exist for a period greater than the predetermined time, the verification unit 630 is further configured to: no longer regard the tracked object as the final tracked object during the second navigation cycle.

[0229] In this example, if the time during which the tracked object does not exist is less than the predetermined time, the verification unit 630 is further configured to:

[0230] Re-searching for objects around the autonomous vehicle; and

[0231] It is identified whether there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle.

[0232] In this example, if there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle, the verification unit 630 is further configured to:

[0233] determining the tracked object as the final tracked object during the second navigation period as the verification result;

[0234] Otherwise, the tracked object is no longer regarded as the final tracked object during the second navigation cycle.

[0235] According to another embodiment of the present specification, a machine-readable storage medium is provided, on which executable instructions are stored, wherein when the executable instructions are executed, the machine executes Figure 4 , 6 , the methods and operations shown in 8.

[0236] The specific embodiments described above in conjunction with the accompanying drawings describe exemplary embodiments, but do not represent all embodiments that can be implemented or fall within the scope of protection of the claims. The term "exemplary" used throughout this specification means "used as an example, instance or illustration" and does not mean "preferred" or "having advantages" over other embodiments. For the purpose of providing an understanding of the described technology, the specific embodiments include specific details. However, these technologies can be implemented without these specific details. In some instances, in order to avoid making the concepts of the described embodiments difficult to understand, well-known structures and devices are shown in block diagram form.

[0237] The above description of the present disclosure is provided to enable any person of ordinary skill in the art to implement or use the present disclosure. Various modifications to the present disclosure will be apparent to those of ordinary skill in the art, and the general principles defined herein may be applied to other variations without departing from the scope of protection of the present disclosure. Therefore, the present disclosure is not limited to the examples and designs described herein, but is consistent with the widest range of principles and novel features disclosed herein.

Claims

1. A method (200) for navigating an autonomous vehicle through an intersection, comprising: Obtaining trajectory information of the tracked object (220); navigating the autonomous vehicle during a first navigation cycle using the trajectory information (230); Verifying the tracked object in real time (240); updating the trajectory information based on the verification result of the tracked object (250); and The automated driving vehicle (260) is navigated during a second navigation cycle using the updated trajectory information, wherein the second navigation cycle is a next navigation cycle after the first navigation cycle.

2. The method of claim 1, wherein: Verifying the tracked object includes: determining whether the trajectory of the new object intersects the navigation path of the autonomous vehicle (310), and determining whether the distance from the new object to the autonomous vehicle is less than the distance from the tracked object to the autonomous vehicle (320), If it is determined that the trajectory of the new object intersects with the navigation path of the autonomous driving vehicle and the distance from the new object to the autonomous driving vehicle is less than the distance from the tracked object to the autonomous driving vehicle, the new object is determined as a candidate object (330).

3. The method of claim 2, wherein: Verifying the tracked object further includes: It is determined whether the candidate object exists, and it is determined whether the tracked object exists (340).

4. The method of claim 3, wherein: If the candidate object exists and the tracked object does not exist, the method further includes determining the candidate object as the final tracked object during the second navigation cycle (350) as the verification result.

5. The method according to claim 2 or 3, further comprising: A determination is made as to whether the lateral distance of the candidate object's historical trajectory relative to the navigation path of the autonomous vehicle is within the left and right boundaries of the width of the autonomous vehicle (360).

6. The method of claim 5, wherein: If it is determined that the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is not within the left and right boundaries of the width of the autonomous driving vehicle, the method further includes: determining the tracked object as the final tracked object during the second navigation cycle (370) as the verification result, Otherwise, the candidate object is determined as the final tracked object during the second navigation cycle (350) as the verification result.

7. The method according to claim 4 or 6, wherein: If the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is still navigated using the trajectory information of the tracked object during the second navigation cycle, or If the verification result is that the candidate object is determined to be the final tracked object during the second navigation period, the autonomous driving vehicle is navigated using the historical trajectory of the candidate object during the second navigation period.

8. The method of claim 1, wherein: Authenticating the tracked object includes determining whether the tracked object exists (410).

9. The method of claim 8, wherein: If it is determined that the tracked object exists, the method further includes: determining the tracked object as the final tracked object during the second navigation cycle (420) as the verification result; or If it is determined that the tracked object does not exist, the method further includes comparing the time when the tracked object does not exist with a predetermined time (430).

10. The method of claim 9, wherein: If the tracked object is absent for a period greater than the predetermined time, the method further includes: no longer considering the tracked object as the final tracked object during the second navigation cycle (440).

11. The method of claim 9, wherein: If the time during which the tracked object does not exist is less than the predetermined time, the method further includes: Re-searching for objects around the autonomous driving vehicle (450); and It is identified whether there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle (460).

12. The method of claim 11, wherein: If the list of objects being searched has an object that is the same as the tracked object during the first navigation cycle, the method further comprises: determining the tracked object as a final tracked object during the second navigation cycle (470) as the verification result; Otherwise, the tracked object is no longer considered as the final tracked object during the second navigation cycle (440).

13. A system (500) for navigating an autonomous vehicle through an intersection, comprising: A memory (510) for storing instruction codes; as well as at least one processor (520) coupled to the memory for executing the stored instruction code, the instruction code being configured to: Obtaining trajectory information of the tracked object; navigating the autonomous vehicle during a first navigation cycle using the trajectory information; Verifying the tracked object in real time; updating the trajectory information based on the verification result of the tracked object; and The automated driving vehicle is navigated during a second navigation cycle using the updated trajectory information, wherein the second navigation cycle is a next navigation cycle after the first navigation cycle.

14. The system of claim 13, wherein: The instruction code is also configured to: determining whether the trajectory of the new object intersects the navigation path of the autonomous vehicle, and determining whether the distance from the new object to the autonomous vehicle is less than the distance from the tracked object to the autonomous vehicle, Among them, if it is determined that the trajectory of the new object intersects with the navigation path of the autonomous driving vehicle and the distance from the new object to the autonomous driving vehicle is less than the distance from the tracked object to the autonomous driving vehicle, the new object is determined as a candidate object.

15. The system of claim 14, wherein: The instruction code is also configured to: It is determined whether the candidate object exists, and it is determined whether the tracked object exists.

16. The system of claim 15, wherein: If the candidate object exists and the tracked object does not exist, the instruction code is further configured to: determine the candidate object as the final tracked object during the second navigation cycle as the verification result.

17. The system of claim 14 or 15, wherein: The instruction code is also configured to: Determine whether the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is within the left and right boundaries of the width of the autonomous driving vehicle.

18. The system of claim 17, wherein: If it is determined that the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is not within the left and right boundaries of the width of the autonomous driving vehicle, the instruction code is further configured to: determining the tracked object as the final tracked object during the second navigation cycle as the verification result, Otherwise, the candidate object is determined as the final tracked object during the second navigation cycle as the verification result.

19. The system of claim 16 or 18, wherein: If the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is still navigated using the trajectory information of the tracked object during the second navigation cycle, or If the verification result is that the candidate object is determined to be the final tracked object during the second navigation period, the autonomous driving vehicle is navigated using the historical trajectory of the candidate object during the second navigation period.

20. The system of claim 13, wherein: The instruction code is further configured to determine whether the tracked object exists.

21. The system of claim 20, wherein: If it is determined that the tracked object exists, the instruction code is further configured to: determine the tracked object as the final tracked object during the second navigation cycle as the verification result; or If it is determined that the tracked object does not exist, the instruction code is further configured to: compare the time when the tracked object does not exist with a predetermined time.

22. The system of claim 21, wherein: If the tracked object is absent for a period greater than the predetermined time, the instruction code is further configured to: no longer regard the tracked object as the final tracked object during the second navigation cycle.

23. The system of claim 21, wherein: If the tracked object is absent for a period of time shorter than the predetermined time, the instruction code is further configured to: Re-searching for objects around the autonomous vehicle; and It is identified whether there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle.

24. The system of claim 23, wherein: If the list of objects being searched contains an object that is the same as the tracked object during the first navigation cycle, the instruction code is further configured to: determining the tracked object as the final tracked object during the second navigation period as the verification result; Otherwise, the tracked object is no longer regarded as the final tracked object during the second navigation cycle.

25. An apparatus (600) for navigating an autonomous driving vehicle through an intersection, comprising: A trajectory information acquisition unit (610), used to obtain trajectory information of the tracked object; A navigation unit (620) for navigating the autonomous driving vehicle during a first navigation cycle using the trajectory information; A verification unit (630), used for verifying the tracked object in real time; as well as A trajectory updating unit (640), configured to update the trajectory information based on the verification result of the tracked object; The navigation unit is further used to navigate the autonomous driving vehicle during a second navigation cycle using the updated trajectory information, wherein the second navigation cycle is a next navigation cycle after the first navigation cycle.

26. The device of claim 25, wherein: The verification unit is also used for: determining whether the trajectory of the new object intersects the navigation path of the autonomous vehicle, and determining whether the distance from the new object to the autonomous vehicle is less than the distance from the tracked object to the autonomous vehicle, Among them, if it is determined that the trajectory of the new object intersects with the navigation path of the autonomous driving vehicle and the distance from the new object to the autonomous driving vehicle is less than the distance from the tracked object to the autonomous driving vehicle, the new object is determined as a candidate object.

27. The device of claim 26, wherein: The verification unit is also used for: It is determined whether the candidate object exists, and it is determined whether the tracked object exists.

28. The apparatus of claim 27, wherein: If the candidate object exists and the tracked object does not exist, the verification unit is further configured to: determine the candidate object as the final tracked object during the second navigation cycle as the verification result.

29. The device of claim 26 or 27, wherein: The verification unit is also used for: Determine whether the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is within the left and right boundaries of the width of the autonomous driving vehicle.

30. The apparatus of claim 29, wherein: If it is determined that the lateral distance of the historical trajectory of the candidate object relative to the navigation path of the autonomous driving vehicle is not within the left and right boundaries of the width of the autonomous driving vehicle, the verification unit is further configured to: determining the tracked object as the final tracked object during the second navigation cycle as the verification result, Otherwise, the candidate object is determined as the final tracked object during the second navigation cycle as the verification result.

31. The device according to claim 28 or 30, wherein: If the verification result is that the tracked object is determined to be the final tracked object during the second navigation cycle, the autonomous driving vehicle is still navigated using the trajectory information of the tracked object during the second navigation cycle, or If the verification result is that the candidate object is determined to be the final tracked object during the second navigation period, the autonomous driving vehicle is navigated using the historical trajectory of the candidate object during the second navigation period.

32. The apparatus of claim 25, wherein: The verification unit is also used for: Determine whether the tracked object exists.

33. The apparatus of claim 32, wherein: If it is determined that the tracked object exists, the verification unit is further used to: determine the tracked object as the final tracked object during the second navigation cycle as the verification result; or If it is determined that the tracked object does not exist, the verification unit is further configured to compare the time when the tracked object does not exist with a predetermined time.

34. The apparatus of claim 33, wherein: If the tracked object does not exist for a period greater than the predetermined time, the verification unit is further configured to: no longer regard the tracked object as the final tracked object during the second navigation cycle.

35. The apparatus of claim 33, wherein: If the tracked object is absent for a period of time shorter than the predetermined period of time, the verification unit is further configured to: Re-searching for objects around the autonomous vehicle; and It is identified whether there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle.

36. The apparatus of claim 35, wherein: If there is an object in the list of searched objects that is the same as the tracked object during the first navigation cycle, the verification unit is further configured to: determining the tracked object as the final tracked object during the second navigation period as the verification result; Otherwise, the tracked object is no longer regarded as the final tracked object during the second navigation cycle.

37. A machine-readable storage medium having executable instructions stored thereon, wherein: When the executable instructions are executed, the machine performs the method of any one of claims 1-12.

Images