Lane changing behavior recognition method and device, storage medium and computer device

Abstract

The application provides a lane changing behavior recognition method and device, a storage medium and a computer device. According to the position and lane changing state of a host vehicle corresponding to each automatic driving moment, a first confidence degree of a first lane to which the host vehicle belongs at each automatic driving moment is determined to reflect the possibility of the host vehicle changing lanes from the first lane to other lanes. According to the position and lane changing state of the host vehicle corresponding to each automatic driving moment, a second confidence degree of a second lane at each automatic driving moment is determined to determine the possibility of the host vehicle changing lanes from other lanes to the second lane. The computer device can recognize the lane changing behavior of the host vehicle in the automatic driving process according to the first confidence degree and the second confidence degree of each lane at each automatic driving moment. In this way, the misjudgment caused by excessive dependence on the internal state of the planning control algorithm module can be corrected, and the lane changing behavior in the automatic driving process can be more accurately recognized, and the recognition accuracy is improved.

Description

Technical Field

[0001] This application relates to the field of autonomous driving, and in particular to a method, apparatus, storage medium and computer equipment for lane change behavior recognition. Background Technology

[0002] In autonomous driving, lane changing is a crucial aspect for evaluating the rationality of autonomous driving algorithm planning. Therefore, during simulation testing of autonomous driving algorithms, it is necessary to identify and score lane changing behaviors that occur during the simulation to determine their rationality and safety. Currently, existing technologies rely solely on information generated by the vehicle's planning and control algorithm module to determine whether the vehicle has engaged in lane changing. However, during actual driving, the vehicle may deviate from the lane centerline to bypass or overtake obstacles, or reach road forks or merges, all of which could be incorrectly identified as lane changes. Therefore, over-reliance on data from the planning and control algorithm module to identify lane changing behaviors is prone to misjudgments and suffers from low accuracy. Summary of the Invention

[0003] The purpose of this application is to address at least one of the aforementioned technical deficiencies, particularly the technical deficiency of low identification accuracy in the prior art.

[0004] In a first aspect, embodiments of this application provide a lane change behavior recognition method, the method comprising:

[0005] Obtain the location of the master vehicle and the first lane to which the master vehicle belongs at each autonomous driving moment;

[0006] For each autonomous driving moment, based on the position of the master vehicle corresponding to the autonomous driving moment and the lane-changing state of the master vehicle at the autonomous driving moment, a first confidence level corresponding to the first lane at the autonomous driving moment is determined. The first confidence level is used to reflect the possibility of the master vehicle changing lanes from the first lane to other lanes besides the first lane.

[0007] For each autonomous driving moment, if at least part of the master vehicle is traveling in the second lane during the autonomous driving moment, then based on the position of the master vehicle corresponding to the autonomous driving moment and the lane change status of the master vehicle during the autonomous driving moment, the second confidence level corresponding to the second lane during the autonomous driving moment is determined. The second confidence level is used to reflect the possibility that the master vehicle changes lanes from other lanes to the second lane.

[0008] Based on the first confidence level and the second confidence level of each lane at each of the aforementioned autonomous driving moments, the lane-changing behavior of the master vehicle during the autonomous driving process is identified.

[0009] In one embodiment, the step of determining the first confidence level of the first lane at the time of autonomous driving based on the position of the driver vehicle at the time of autonomous driving and the lane-changing state of the driver vehicle at the time of autonomous driving includes:

[0010] Based on the position of the main vehicle and the position of the center line of the first lane corresponding to the autonomous driving moment, calculate the first distance between the main vehicle and the first lane at the autonomous driving moment, and determine the first score based on the first distance;

[0011] If the lane-changing state of the master vehicle at the autonomous driving moment is a lane-changing request state, then the duration of the first state of the lane-changing request state up to the autonomous driving moment is determined, and a second score is determined based on the duration of the first state, and the first confidence level of the first lane at the autonomous driving moment is determined based on the first score and the second score.

[0012] If the lane-changing state of the master vehicle at the moment of autonomous driving is not the requested lane-changing state, then the first confidence level of the first lane at the moment of autonomous driving is determined according to the first score.

[0013] In one embodiment, the step of determining the first fraction based on the first interval includes:

[0014] Calculate the first ratio between the first spacing and the lane width of the first lane;

[0015] The first ratio is processed based on the following expression to determine the first score:

[0016] ln(1+(e-1)·a)

[0017] In the formula, e is the natural constant, and a is the first ratio.

[0018] In one embodiment, the step of determining the second confidence level of the second lane at the time of autonomous driving based on the position of the driver vehicle at the time of autonomous driving and the lane-changing state of the driver vehicle at the time of autonomous driving includes:

[0019] Based on the position of the main vehicle and the position of the center line of the second lane corresponding to the autonomous driving moment, calculate the second distance between the main vehicle and the second lane at the autonomous driving moment, and determine the third score based on the second distance;

[0020] If the lane-changing state of the master vehicle at the autonomous driving moment is the end of lane-changing state, then the duration of the second state of the end of lane-changing state up to the autonomous driving moment is determined, and a fourth score is determined based on the duration of the second state. Furthermore, the second confidence level corresponding to the second lane at the autonomous driving moment is determined based on the third score and the fourth score.

[0021] If the lane-changing state of the master vehicle at the moment of autonomous driving is not the end of lane-changing state, then the second confidence level of the second lane at the moment of autonomous driving is determined according to the third score.

[0022] In one embodiment, the step of determining the third fraction based on the second spacing includes:

[0023] Calculate a second ratio between the second distance between the main vehicle and the second lane at the time of autonomous driving and the lane width of the second lane, and determine a fifth score based on the second ratio;

[0024] Based on the second distance between the main vehicle and the second lane at at least one pre-autopilot moment, and the second distance between the main vehicle and the second lane at that autopilot moment, it is determined whether the second distance between the main vehicle and the second lane remains unchanged at multiple autopilot moments, and each pre-autopilot moment is earlier than that autopilot moment;

[0025] If the second distance between the main vehicle and the second lane remains unchanged at multiple autonomous driving moments, then the duration of the distance during which the second distance remains unchanged up to that autonomous driving moment is determined, and a sixth score is determined based on the duration of the distance, and a third score is determined based on the fifth score and the sixth score.

[0026] If the second distance between the main vehicle and the second lane at this autonomous driving moment is different from the second distance at the previous autonomous driving moment, then the fifth score is used as the third score.

[0027] In one embodiment, the sixth score is positively correlated with the duration of the distance; the step of determining the third score based on the fifth score and the sixth score includes:

[0028] If the sixth score is greater than or equal to a preset score threshold, then the sum of the fifth score and the sixth score is taken as the third score; otherwise, the fifth score is taken as the third score.

[0029] In one embodiment, the step of identifying the lane-changing behavior of the master vehicle during the autonomous driving process based on the first confidence level and the second confidence level corresponding to each lane at each of the autonomous driving moments includes:

[0030] For each autonomous driving moment, if the first confidence level of the third lane at that autonomous driving moment is greater than or equal to the first preset confidence level threshold, and the second confidence level of the fourth lane adjacent to the third lane at that autonomous driving moment is greater than or equal to the second preset threshold, then it is determined that the driver vehicle has a lane-changing behavior at that autonomous driving moment.

[0031] In one embodiment, the method further includes:

[0032] For each autonomous driving moment in which lane changing occurs, calculate the product between the first confidence level of the third lane and the second confidence level of the fourth lane in that autonomous driving moment.

[0033] The lane change behavior corresponding to the largest product among the various products is taken as the target lane change behavior, and the start time and end time of the lane change corresponding to the target lane change behavior are obtained respectively.

[0034] In one embodiment, the step of obtaining the lane change start time and lane change end time corresponding to the target lane change behavior includes:

[0035] Determine the target third lane and target fourth lane corresponding to the target lane change behavior;

[0036] In each of the target first confidence levels corresponding to the target third lane, the first target first confidence level that is greater than or equal to the first preset confidence level threshold is determined according to the time sequence of each of the autonomous driving moments, and the autonomous driving moment corresponding to the first target first confidence level is taken as the lane change start moment corresponding to the target lane change behavior.

[0037] The maximum second confidence level is determined among the various second confidence levels corresponding to the target fourth lane, and the automatic driving time corresponding to the maximum second confidence level is taken as the lane change end time corresponding to the target lane change behavior.

[0038] Secondly, embodiments of this application provide a lane change behavior recognition device, the device comprising:

[0039] The location acquisition module is used to acquire the location of the master vehicle and the first lane to which the master vehicle belongs at each autonomous driving moment;

[0040] The first confidence level acquisition module is used to determine the first confidence level of the first lane at each autonomous driving moment based on the position of the master vehicle and the lane-changing state of the master vehicle at that autonomous driving moment. The first confidence level is used to reflect the possibility of the master vehicle changing lanes from the first lane to other lanes besides the first lane.

[0041] The second confidence level acquisition module is used to determine the second confidence level of the second lane at each autonomous driving moment if at least part of the master vehicle is traveling in the second lane at that autonomous driving moment, based on the position of the master vehicle corresponding to that autonomous driving moment and the lane change state of the master vehicle at that autonomous driving moment. The second confidence level is used to reflect the possibility that the master vehicle changes lanes from other lanes to the second lane.

[0042] The identification module is used to identify the lane-changing behavior of the master vehicle during the autonomous driving process based on the first confidence level and the second confidence level corresponding to each lane at each autonomous driving moment.

[0043] Thirdly, embodiments of this application provide a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the lane change behavior recognition method described in any of the above embodiments.

[0044] Fourthly, embodiments of this application provide a computer device, including: one or more processors, and a memory;

[0045] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the lane change behavior recognition method described in any of the above embodiments.

[0046] In the lane change behavior recognition method, apparatus, storage medium, and computer device of this application, the computer device can determine the first confidence level of the first lane to which the main vehicle belongs at each autonomous driving moment based on the position and lane change status of the main vehicle at each autonomous driving moment, and then determine the probability of the main vehicle changing lanes from the currently traveling first lane to other lanes besides the first lane. The computer device can also determine the second confidence level of the second lane at each autonomous driving moment based on the position and lane change status of the main vehicle at each autonomous driving moment, and then determine the probability of the main vehicle changing lanes from other lanes besides the second lane to the second lane. After obtaining the first and second confidence levels of each lane at each autonomous driving moment, the computer device can identify the lane change behavior of the main vehicle during autonomous driving. Thus, lane change behavior can be automatically identified based on the first confidence level of the main vehicle leaving the currently traveling first lane and the confidence level of the main vehicle entering the second lane, thereby correcting misjudgments caused by over-reliance on the internal state of the planning and control algorithm module, and thus more accurately identifying lane change behavior during autonomous driving, improving recognition accuracy. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart illustrating a lane change behavior recognition method in one embodiment;

[0049] Figure 2 This is a flowchart illustrating the steps for obtaining the location of the master vehicle and the first lane to which the master vehicle belongs at each autonomous driving moment in one embodiment.

[0050] Figure 3 This is a flowchart illustrating the step of determining the first confidence level in one embodiment;

[0051] Figure 4 This is a flowchart illustrating the step of determining the second confidence level in one embodiment;

[0052] Figure 5 This is a flowchart illustrating the step of determining the third fraction based on the second spacing in one embodiment;

[0053] Figure 6 This is a schematic diagram of the lane change behavior recognition device in one embodiment;

[0054] Figure 7This is a schematic diagram of the structure of a computer device in one embodiment. Detailed Implementation

[0055] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0056] In one embodiment, this application provides a lane change behavior recognition method. The following embodiments illustrate this method applied to a computer device. It is understood that the computer device refers to a device with data processing capabilities, and may be, but is not limited to, a computer, a personal laptop, a single server, a server cluster, etc. This application does not impose specific limitations in this regard. Figure 1 As shown, the method specifically includes the following steps:

[0057] S102: Obtain the position of the master vehicle and the first lane to which the master vehicle belongs at each autonomous driving moment.

[0058] Here, "autonomous driving moment" refers to the moment when the vehicle is controlled by the autonomous driving algorithm. For example, it could be any moment when the autonomous driving algorithm controls the vehicle to drive on a real road, or any simulation test moment when the autonomous driving algorithm is being simulated. "First lane" refers to the lane to which the vehicle belongs, that is, the lane the vehicle is currently traveling in. It can be understood that as the position of the vehicle changes, the first lane to which the vehicle belongs in different autonomous driving moments can be the same or different.

[0059] It should be noted that this application can obtain the position of the master vehicle at each autonomous driving moment and the first lane to which the master vehicle belongs at each autonomous driving moment in any implementation method, and this application does not impose any specific restrictions on this.

[0060] In one embodiment, the computer device can follow Figure 2 The steps shown implement S102. For example... Figure 2 As shown, computer devices can perform the following steps:

[0061] S202: Obtain the master vehicle's position, orientation angle, and surrounding line segment information at each autonomous driving moment. The line segment information refers to the lane information corresponding to the first lane, including but not limited to road waypoints and lane relationship information.

[0062] S204: Based on the location of the master vehicle at each autonomous driving moment, the road waypoints within a preset distance range from the master vehicle are taken as the target waypoints for that autonomous driving moment. For example, if the preset distance range is 2 meters, then all waypoints within a 2-meter radius of the master vehicle's location can be taken as target waypoints.

[0063] S206: Based on the waypoint orientation angles of the target waypoints corresponding to each autonomous driving time, determine multiple candidate waypoints for each autonomous driving time. For each autonomous driving time, the computer device can use the N target waypoints with the smallest waypoint orientation angles at that autonomous driving time as candidate waypoints for that autonomous driving time, where N is a positive integer greater than 1. For example, 2 to 3 target waypoints with the smallest waypoint orientation angles can be used as candidate waypoints.

[0064] S208: Based on multiple alternative waypoints at each autonomous driving moment, determine the optimal waypoint for each autonomous driving moment, and determine the first lane to which the master vehicle belongs at that autonomous driving moment based on the optimal waypoint for each autonomous driving moment.

[0065] Specifically, for each autonomous driving moment, the computer device can use the target waypoints of the previous M autonomous driving moments and the target waypoints of the next M autonomous driving moments as voting waypoints, where M is a positive integer, for example, M can be 5. For each voting waypoint in the same autonomous driving moment, the computer device can use a depth-first search algorithm to search a certain distance (e.g., 10 meters) along the direction of travel of the main vehicle to determine whether the candidate waypoints for that autonomous driving moment are located near the voting waypoint. If so, it votes for the candidate waypoints near the voting waypoint. After searching based on the voting waypoints for the same autonomous driving moment, the waypoint with the most votes among the candidate waypoints for the same autonomous driving moment is taken as the best waypoint for that autonomous driving moment, and the road to which the best waypoint is located is the lane to which the main vehicle belongs in the corresponding autonomous driving moment.

[0066] In this embodiment, the computer device determines multiple alternative waypoints for each autonomous driving moment and votes based on the target waypoints corresponding to multiple autonomous driving moments before and after to determine the best waypoint from multiple alternative waypoints. The road to which the best waypoint belongs is taken as the first lane to which the master vehicle belongs at that autonomous driving moment. This allows for accurate determination of the first lane to which the master vehicle currently belongs in complex road relationships (such as forked roads and merging roads), thereby further improving the accuracy of subsequent lane change behavior recognition.

[0067] S104: For each autonomous driving moment, based on the position of the master vehicle corresponding to the autonomous driving moment and the lane-changing state of the master vehicle at the autonomous driving moment, determine the first confidence level of the first lane at the autonomous driving moment. The first confidence level is used to reflect the possibility of the master vehicle changing lanes from the first lane to other lanes besides the first lane.

[0068] The lane change status can refer to whether the autonomous driving algorithm controls the vehicle to change lanes, and the computer equipment can obtain this information from the planning and control algorithm module of the autonomous driving algorithm. In one embodiment, the lane change status can include a lane change request status and a lane change termination status. The lane change request status can be the state in which the autonomous driving algorithm attempts to control the master vehicle to change lanes, and the lane change termination status can be the state in which the lane change behavior has ended.

[0069] Specifically, for each autonomous driving moment, the computer device can determine the first confidence level of the first lane currently occupied by the driver vehicle at that autonomous driving moment based on the driver vehicle's position and lane change status at that autonomous driving moment, and then determine the possibility of the driver vehicle changing lanes from the first lane to other lanes at that autonomous driving moment.

[0070] Furthermore, in some maps, lanes may branch or merge, resulting in the same lane corresponding to different lane markings at different locations. Therefore, computer equipment can combine road segment information to determine whether the initial confidence level corresponding to different lane markings needs to be inherited. For example, if the road segment information determines that two lane markings correspond to lanes that are connected sequentially, then the latter lane can inherit the initial confidence level corresponding to the former lane. This sequential relationship can be determined based on the direction of travel of the main vehicle; for example, as the main vehicle travels in its direction, it will pass through the preceding and following lanes in sequence.

[0071] In one embodiment, in S104, the computer device can execute for each autonomous driving moment. Figure 3 S302 to S306 are shown to obtain the first confidence level of the first lane at each autonomous driving time. Figure 3 As shown, for each autonomous driving moment, the computer device can perform the following steps:

[0072] S302: Based on the position of the main vehicle and the position of the center line of the first lane corresponding to the autonomous driving moment, calculate the first distance between the main vehicle and the first lane at the autonomous driving moment, and determine the first score based on the first distance.

[0073] Specifically, the computer device can determine a first score based on the degree to which the master vehicle is far from the first lane. In determining the first score, the computer device can calculate the distance between the master vehicle and the first lane to which the master vehicle belongs, based on the master vehicle's position at the current autonomous driving moment and the lane centerline position of the first lane at the current autonomous driving moment; this distance is the first distance. In one embodiment, the computer device can calculate the minimum distance between the master vehicle's center point position and the lane centerline of the first lane based on the master vehicle's center point position, the master vehicle's position, and the lane centerline position of the first lane, to obtain the first distance.

[0074] After determining the first distance, the computer device can determine a first score based on the first distance, such that the first score reflects the degree to which the master vehicle is far from its first lane at the current moment of autonomous driving. In one embodiment, the first score and the first distance can be positively correlated, that is, the larger the first distance, the larger the first score.

[0075] Furthermore, in one example, in determining the first fraction, the computer device can calculate a first ratio between the first spacing and the lane width of the first lane, and process the first ratio based on ln(1+(e-1)·a) to obtain the first fraction. Here, e is a natural constant, and a is the first ratio. For example, when the width of the first lane is 2 meters, if the first spacing is 0.5 meters, the first fraction can be ln(1+(e-1)*0.25) = 0.35. If the first spacing is 1 meter, the first fraction can be ln(1+(e-1)*0.5) = 0.62. If the first spacing is 1.5 meters, the first fraction can be ln(1+(e-1)*0.75) = 0.82. Thus, as the first gap increases, the growth rate of the first score can slow down, making the changes in the first score more consistent with the actual lane-changing situation. This allows the determination of the first confidence level based on the first score to more accurately reflect the possibility of the main vehicle changing lanes from the first lane to other lanes besides the first lane.

[0076] S304: If the lane change state of the master vehicle at the autonomous driving moment is a lane change request state, then determine the duration of the first state of the lane change request state up to the autonomous driving moment, determine a second score based on the duration of the first state, and determine the first confidence level of the first lane at the autonomous driving moment based on the first score and the second score.

[0077] Specifically, lane change status can reflect lane change intention. If the lane change status of the master vehicle at the current moment of autonomous driving is a lane change request status, it indicates that the master vehicle is more likely to change lanes from the first lane to other lanes besides the first lane, and the probability of lane change is related to the duration of the lane change request status.

[0078] To accurately determine the first confidence level and more accurately identify lane-changing behavior during autonomous driving, when the driver vehicle's lane-changing state at the current autonomous driving moment is a requested lane-changing state, the computer device can use the duration of the requested lane-changing state up to that autonomous driving moment as the first state duration, and determine the second score based on the first state duration. For example, if the autonomous driving algorithm switches the lane-changing state from a non-requested lane-changing state to a requested lane-changing state at the first moment, and the requested lane-changing state continues until the current autonomous driving moment, which is the second moment, then the first state duration is the difference between the second moment and the first moment.

[0079] In one embodiment, the second score can be positively correlated with the duration of the first state; that is, the longer the duration of the first state, the higher the second score. In another embodiment, when the duration of the first state is less than or equal to a first duration threshold, the second score is linearly correlated with the duration of the first state; otherwise, the second score is the first maximum preset score. For example, if the first duration threshold is 10 seconds and the first maximum preset score is 0.3 points, when the duration of the first state is within the range [0, 10], the second score is linearly correlated with the duration of the first state, that is, accumulating 0.03 points per second. When the duration of the first state is greater than 10 seconds, the second score is 0.3 points.

[0080] Given a second score, the computer device can determine a first confidence level for the first lane at the current moment of autonomous driving, based on the first and second scores. In one embodiment, the computer device can use the sum of the first and second scores as the first confidence level.

[0081] S306: If the lane-changing state of the master vehicle at the moment of autonomous driving is not the requested lane-changing state, then the first confidence level of the first lane at the moment of autonomous driving is determined according to the first score.

[0082] Specifically, if the lane change state of the main vehicle at the current autonomous driving moment is not a requested lane change state, the computer device can directly determine the first confidence level of the first lane at the current autonomous driving moment based on the first score, for example, using the first score as the first confidence level.

[0083] In this embodiment, the computer device can determine a first score based on a first distance between the driver vehicle and the first lane, so that the first score can reflect the degree to which the driver vehicle is moving away from the first lane. The computer device determines a first confidence level based on the first score and whether the lane change state is in a requested lane change state, thereby recognizing the accurate determination of the first confidence level by comprehensively considering the degree to which the driver vehicle is moving away and the intention to change lanes. This allows the first confidence level to more accurately reflect the possibility of the driver vehicle changing lanes from the first lane to other lanes, and thus to more accurately identify lane change behavior during the autonomous driving process.

[0084] S106: For each of the autonomous driving moments, if at least part of the master vehicle is traveling in the second lane during the autonomous driving moment, then based on the position of the master vehicle corresponding to the autonomous driving moment and the lane-changing state of the master vehicle during the autonomous driving moment, a second confidence level corresponding to the second lane during the autonomous driving moment is determined. The second confidence level is used to reflect the possibility that the master vehicle changes lanes from other lanes other than the second lane to the second lane.

[0085] The second lane is different from the first lane. For each autonomous driving moment, the computer can determine whether the driver vehicle is partially traveling in the second lane. If so, it indicates that during that autonomous driving moment, part of the driver vehicle is traveling in the first lane, and the other part is traveling in the second lane; that is, the driver vehicle is driving on the line between the first and second lanes. In this case, the computer can determine the second confidence level of the second lane based on the driver vehicle's position and lane-changing status at that autonomous driving moment, thus determining the probability that the driver vehicle will change lanes from any lane other than the second lane to the second lane during that autonomous driving moment.

[0086] Similarly, computer equipment can combine road segment information to determine whether the second confidence level corresponding to different lane markings needs to be inherited. The specific description of the inheritance of the second confidence level can be compared with the relevant description of the inheritance of the first confidence level above, and will not be repeated here.

[0087] In one embodiment, in S106, the computer device can execute for each autonomous driving moment. Figure 4 S402 to S406 are shown to obtain the second confidence level of the first lane at each autonomous driving time. For example... Figure 4 As shown, for each autonomous driving moment, the computer device can perform the following steps:

[0088] S402: Based on the position of the main vehicle and the position of the center line of the second lane corresponding to the autonomous driving moment, calculate the second distance between the main vehicle and the second lane at the autonomous driving moment, and determine the third score based on the second distance.

[0089] Specifically, the second spacing reflects the degree to which the driver vehicle is close to the second lane. Furthermore, by combining the second spacing under different autonomous driving times, the changes in the spacing between the driver vehicle and the second lane can be determined. The computer device can calculate the spacing between the driver vehicle and the second lane based on the driver vehicle's position at the current autonomous driving time and the centerline position of the second lane at the current autonomous driving time; this spacing is the second spacing. In one embodiment, the computer device can use the minimum distance between the center point of the driver vehicle and the centerline of the second lane as the second spacing.

[0090] The computer equipment can determine a third score based on the second distance, such that the third score at least reflects the degree to which the main vehicle is close to the second lane.

[0091] In one embodiment, such as Figure 5 As shown, the steps for determining the third fraction based on the second interval may include:

[0092] S502: Calculate the second ratio between the second distance between the main vehicle and the second lane at the time of automatic driving and the lane width of the second lane, and determine the fifth score based on the second ratio.

[0093] It is understood that computer devices can determine the fifth fraction based on the second ratio in any manner, and this application does not impose any specific restrictions on this.

[0094] S504: Based on the second distance between the main vehicle and the second lane at at least one pre-autopilot moment, and the second distance between the main vehicle and the second lane at that autopilot moment, determine whether the second distance between the main vehicle and the second lane remains unchanged at multiple autopilot moments, wherein each pre-autopilot moment is earlier than that autopilot moment.

[0095] Here, the P preceding autonomous driving moments can be the P preceding autonomous driving moments of the currently processed autonomous driving moment. In other words, the P preceding autonomous driving moments are sequentially adjacent, and the latest driving moment among the P preceding autonomous driving moments is the autonomous driving moment preceding the currently processed autonomous driving moment. Here, P is a positive integer.

[0096] Specifically, the computer device can determine the second spacing corresponding to P previous autonomous driving moments and the second spacing corresponding to the current autonomous driving moment, and determine whether the second spacing remains unchanged continuously in the current autonomous driving moment and at least one previous autonomous driving moment based on the determined (P+1) second spacings.

[0097] S506: If the second distance between the main vehicle and the second lane remains unchanged at multiple autonomous driving moments, then determine the duration of the distance during which the second distance remains unchanged up to the autonomous driving moment, determine a sixth score based on the duration of the distance, and determine a third score based on the fifth score and the sixth score.

[0098] Specifically, if the second distance remains constant during the current autonomous driving moment and at least one previous autonomous driving moment, the computer device can use the duration during which the second distance remains constant up to the current autonomous driving moment as the distance duration. For example, if the second distance corresponding to the third moment T3 is D1, and the second distance D1 remains constant until the current autonomous driving moment T4, then the distance duration is the difference between T4 and T4.

[0099] The computer device can determine the sixth fraction based on the duration of the distance, such that the sixth fraction reflects the maintenance of the second spacing. It is understood that the sixth fraction can be determined in any manner, and this application does not impose any specific limitations on it. In one example, the computer device may use e x / e 3 The sixth score is calculated using the formula, where x is the distance duration and e is the natural constant. For example, when the distance duration is 2 seconds, it indicates that the second lane spacing remains unchanged for 2 seconds up to the current autonomous driving moment, and the sixth score can be 0.367; when the distance duration is 2.5 seconds, it indicates that the second lane spacing remains unchanged for 2.5 seconds up to the current autonomous driving moment, and the sixth score can be 0.606. By using the formula in this example to determine the sixth score, the rate of increase of the sixth score changes from slow to fast as the distance duration increases, making the change of the sixth score more consistent with actual lane changing situations. This allows the determination of the second confidence level based on the sixth score to more accurately reflect the probability of the driver changing lanes from other lanes to the second lane.

[0100] Given a fifth and sixth score, the computer device can determine a third score based on these scores. In one example, the sixth score is positively correlated with the duration of the distance traveled. The computer device uses the sum of the fifth and sixth scores as the third score if the sixth score is greater than or equal to a preset score threshold, and uses the fifth score as the third score if the sixth score is less than the preset score threshold. In this example, a larger sixth score indicates a longer duration of distance traveled, meaning the vehicle continuously drove over the lane lines for a longer period, impacting driving safety. This continuous lane-crossing behavior indicates that the autonomous driving algorithm's control process has room for optimization (e.g., the autonomous driving algorithm incorrectly judges that a lane change has been completed, or the autonomous driving algorithm can be optimized to control the vehicle to completely change lanes to overtake / around obstacles). Therefore, the computer device can identify continuous lane-crossing behavior as lane-changing behavior, facilitating subsequent judgment and optimization by engineers.

[0101] S508: If the second distance between the main vehicle and the second lane at this autonomous driving moment is different from the second distance at the previous autonomous driving moment, then the fifth score is used as the third score.

[0102] If the second spacing corresponding to the current autonomous driving moment is different from the second spacing corresponding to the previous autonomous driving moment, it indicates that the second spacing has changed between the previous autonomous driving moment and the current autonomous driving moment. Therefore, the fifth score can be directly used as the third score.

[0103] In this embodiment, the second confidence level is determined by combining the distance change of the second spacing, so that the second confidence level can more accurately reflect the possibility of the master vehicle changing lanes from other lanes to the second lane, and thus can more accurately identify lane changing behavior in the autonomous driving process.

[0104] S404: If the lane-changing state of the master vehicle at the autonomous driving moment is the lane-changing end state, then determine the duration of the second state of the lane-changing end state up to the autonomous driving moment, determine the fourth score based on the duration of the second state, and determine the second confidence level of the second lane at the autonomous driving moment based on the third score and the fourth score.

[0105] Specifically, lane change status can reflect lane change intention. If the lane change status of the driver vehicle at the current moment of autonomous driving is the end of lane change, it indicates that the driver vehicle is more likely to change lanes from other lanes to the second lane, and the probability of lane change is related to the duration of the end of lane change status.

[0106] To accurately determine the second confidence level and more accurately identify lane-changing behavior during autonomous driving, when the vehicle's lane-changing state at the current autonomous driving moment is in the "end of lane-changing" state, the computer device can use the duration of the lane-changing state ending at that moment as the second state duration and determine a fourth score based on the second state duration. In one embodiment, the fourth score and the second state duration can be positively correlated. In another embodiment, when the second state duration is less than or equal to a second duration threshold, the fourth score is linearly correlated with the second state duration; otherwise, the fourth score is a second maximum preset score.

[0107] The computer device can determine the second confidence level of the second lane at the current moment of autonomous driving processing based on the third and fourth scores. In one embodiment, the computer device can use the sum of the third and fourth scores as the second confidence level.

[0108] S406: If the lane-changing state of the master vehicle at the moment of autonomous driving is not the end of lane-changing state, then the second confidence level of the second lane at the moment of autonomous driving is determined according to the third score.

[0109] In one embodiment, if the lane-changing state of the master vehicle at the current moment of autonomous driving is not the end of the lane-changing state, the computer device can directly use the third score as the second confidence level.

[0110] In this embodiment, the computer device can determine a third score based on a second distance between the driver vehicle and the second lane, so that the third score can reflect the degree to which the driver vehicle approaches the second lane. The computer device determines a second confidence level based on the third score and whether the lane change state is in the completed lane change state. This allows for a comprehensive consideration of the driver vehicle's approach and lane change intention to accurately determine the second confidence level, enabling the second confidence level to more accurately reflect the possibility of the driver vehicle changing lanes from other lanes to the second lane, and thus more accurately identify lane change behavior during autonomous driving.

[0111] S108: Based on the first confidence level and the second confidence level corresponding to each lane at each of the aforementioned autonomous driving moments, identify the lane-changing behavior of the master vehicle during the autonomous driving process.

[0112] Specifically, based on the first and second confidence levels of each lane at each autonomous driving moment, the computer equipment can determine the probability that the driver vehicle will leave its current first lane and the probability that it will change lanes to the second lane, and thus identify the driver vehicle's lane-changing behavior.

[0113] In one embodiment, S108 may include: for each of the autonomous driving moments, if the first confidence level of the third lane at that autonomous driving moment is greater than or equal to a first preset confidence level threshold, and the second confidence level of the fourth lane adjacent to the third lane at that autonomous driving moment is greater than or equal to a second preset threshold, then it is determined that the driver vehicle has a lane-changing behavior at that autonomous driving moment.

[0114] Specifically, the computer equipment can identify each pair of third and fourth lanes that simultaneously meet the following conditions:

[0115] (1) The third lane and the fourth lane are adjacent to each other on the left and right;

[0116] (2) The first confidence level of the third lane at the same autonomous driving moment is greater than or equal to the first preset confidence level threshold;

[0117] (3) The second confidence level of the fourth lane at the same autonomous driving moment is greater than or equal to the second preset confidence level threshold.

[0118] If at least one pair of third and fourth lanes satisfying all the above conditions exists, it can be determined that the driver vehicle has engaged in lane-changing behavior at the moment of autonomous driving. In one embodiment, if no third and fourth lanes satisfying all the above conditions exist, it can be determined that the driver vehicle has not engaged in lane-changing behavior during autonomous driving.

[0119] Furthermore, in the event of a lane change, the computer device can determine the lane change parameters corresponding to this lane change behavior based on the first confidence level of each pair of third lanes at each autonomous driving time and / or the second confidence level of the fourth lane at each autonomous driving time, such as the start and end times of the lane change, so as to calculate the score corresponding to this lane change behavior based on the lane change parameters.

[0120] In the lane change behavior recognition method of this application, the computer device can determine the first confidence level of the first lane to which the main vehicle belongs at each autonomous driving moment based on the position and lane change status of the main vehicle at each autonomous driving moment, and then determine the probability of the main vehicle changing lanes from the currently traveling first lane to other lanes besides the first lane. The computer device can also determine the second confidence level of the second lane at each autonomous driving moment based on the position and lane change status of the main vehicle at each autonomous driving moment, and then determine the probability of the main vehicle changing lanes from other lanes besides the second lane to the second lane. After obtaining the first and second confidence levels of each lane at each autonomous driving moment, the computer device can recognize the lane change behavior of the main vehicle during autonomous driving. Thus, lane change behavior can be automatically recognized based on the first confidence level of the main vehicle leaving the currently traveling first lane and the confidence level of the main vehicle entering the second lane, thereby correcting misjudgments caused by over-reliance on the internal state of the planning and control algorithm module, and thus more accurately recognizing lane change behavior during autonomous driving, improving recognition accuracy.

[0121] In one embodiment, the lane change behavior recognition method of this application may further include the following steps:

[0122] For each autonomous driving moment in which lane changing occurs, calculate the product between the first confidence level of the third lane and the second confidence level of the fourth lane in that autonomous driving moment.

[0123] The lane change behavior corresponding to the largest product among the various products is taken as the target lane change behavior, and the start time and end time of the lane change corresponding to the target lane change behavior are obtained respectively.

[0124] Specifically, in some maps, lanes may branch or merge, resulting in the same lane corresponding to different lane markings at different locations. Therefore, a single lane change may cause the computer to identify multiple pairs of third and fourth lanes that meet the conditions, leading to multiple lane change behaviors being detected. To select the most reasonable lane change behavior from among the identified multiple lane change behaviors as the target lane change behavior, and to calculate the lane change score based on the start and end times of the target lane change behavior, the computer can calculate the product of the first confidence level of the third lane and the second confidence level of the fourth lane at the same autonomous driving time when the third lane and the fourth lane are adjacent to each other, the first confidence level of the third lane at a certain autonomous driving time is greater than or equal to the first preset confidence level threshold, and the second confidence level of the fourth lane at the same autonomous driving time is greater than or equal to the second preset confidence level threshold. This product is used to obtain the confidence level product corresponding to each lane change behavior.

[0125] After obtaining the confidence product corresponding to each lane change, the computer device can take the lane change corresponding to the maximum confidence product as the target lane change and obtain the start time and end time of the lane change corresponding to the target lane change.

[0126] In this embodiment, the most reasonable lane change behavior can be selected from multiple lane change behaviors as the target lane change behavior, so that the lane change score can be calculated based on the start time and end time of the lane change behavior, thereby reducing the amount of subsequent data processing.

[0127] In one embodiment, the step of obtaining the lane change start time and lane change end time corresponding to the target lane change behavior includes:

[0128] Determine the target third lane and target fourth lane corresponding to the target lane change behavior;

[0129] In each of the target first confidence levels corresponding to the target third lane, the first target first confidence level that is greater than or equal to the first preset confidence level threshold is determined according to the time sequence of each of the autonomous driving moments, and the autonomous driving moment corresponding to the first target first confidence level is taken as the lane change start moment corresponding to the target lane change behavior.

[0130] The maximum second confidence level is determined among the various second confidence levels corresponding to the target fourth lane, and the automatic driving time corresponding to the maximum second confidence level is taken as the lane change end time corresponding to the target lane change behavior.

[0131] Specifically, the third lane involved in the target lane change is designated as the target third lane, and the fourth lane involved is designated as the target fourth lane. The computer can determine the earliest autonomous driving time corresponding to a first confidence level greater than or equal to a first preset confidence level threshold for the target third lane at various autonomous driving times, and use this as the lane change start time. The computer can also determine the maximum second confidence level corresponding to the target fourth lane and use the autonomous driving time corresponding to this maximum second confidence level as the lane change end time. This allows for a more accurate determination of the lane change start and end times, enabling the lane change score calculated subsequently based on these times to more accurately evaluate the target lane change behavior.

[0132] The lane change behavior recognition device provided in the embodiments of this application is described below. The lane change behavior recognition device described below can be referred to in correspondence with the lane change behavior recognition method described above.

[0133] In one embodiment, this application provides a lane change behavior recognition device 600. For example... Figure 6 As shown, the device 600 includes a location acquisition module 610, a first confidence level acquisition module 620, a second confidence level acquisition module 630, and an identification module 640. Wherein:

[0134] The location acquisition module 610 is used to acquire the location of the master vehicle and the first lane to which the master vehicle belongs at each autonomous driving moment.

[0135] The first confidence level acquisition module 620 is used to determine the first confidence level of the first lane at each autonomous driving moment based on the position of the master vehicle and the lane-changing state of the master vehicle at that autonomous driving moment. The first confidence level is used to reflect the possibility of the master vehicle changing lanes from the first lane to other lanes besides the first lane.

[0136] The second confidence level acquisition module 630 is used to determine the second confidence level of the second lane at each autonomous driving moment if at least part of the master vehicle is traveling in the second lane at that autonomous driving moment, based on the position of the master vehicle corresponding to the autonomous driving moment and the lane change state of the master vehicle at that autonomous driving moment. The second confidence level is used to reflect the possibility that the master vehicle changes lanes from other lanes to the second lane.

[0137] The identification module 640 is used to identify the lane-changing behavior of the master vehicle during the autonomous driving process based on the first confidence level and the second confidence level corresponding to each lane at each autonomous driving moment.

[0138] In one embodiment, the first confidence level acquisition module 620 includes a first score determination unit, a first confidence level determination unit, and a second confidence level determination unit. The first score determination unit is used to calculate a first distance between the driver vehicle and the first lane at the autonomous driving time, based on the driver vehicle's position and the lane centerline position of the first lane at the autonomous driving time, and to determine a first score based on the first distance. The first confidence level determination unit is used to determine the duration of a first state of requesting a lane change up to the autonomous driving time when the driver vehicle's lane-changing state at the autonomous driving time is a request lane change state, and to determine a second score based on the duration of the first state, and to determine the first confidence level of the first lane at the autonomous driving time based on the first score and the second score. The second confidence level determination unit is used to determine the first confidence level of the first lane at the autonomous driving time based on the first score when the driver vehicle's lane-changing state at the autonomous driving time is not a request lane change state.

[0139] In one embodiment, the first score determination unit includes a first ratio calculation unit and a ratio processing unit. The first ratio calculation unit calculates a first ratio between the first spacing and the lane width of the first lane. The ratio processing unit processes the first ratio according to the following expression to determine the first score:

[0140] ln(1+(e-1)·a)

[0141] In the formula, e is the natural constant, and a is the first ratio.

[0142] In one embodiment, the second confidence level acquisition module 630 includes a second score determination unit, a third confidence level determination unit, and a fourth confidence level determination unit. The second score determination unit calculates a second distance between the driver vehicle and the second lane at the autonomous driving time based on the driver vehicle's position and the lane centerline position of the second lane at the autonomous driving time, and determines a third score based on the second distance. The third confidence level determination unit determines the duration of the second state (end of lane change) up to the autonomous driving time if the driver vehicle's lane change state at the autonomous driving time is an end of lane change state, determines a fourth score based on the duration of the second state, and determines the second confidence level of the second lane at the autonomous driving time based on the third score and the fourth score. The fourth confidence level determination unit determines the second confidence level of the second lane at the autonomous driving time based on the third score if the driver vehicle's lane change state at the autonomous driving time is not an end of lane change state.

[0143] In one embodiment, the second score determination unit includes a third score determination unit, a judgment unit, a fourth score determination unit, and a fifth score determination unit. The third score determination unit is used to calculate a second ratio between the second distance between the driver vehicle and the second lane at the time of autonomous driving and the lane width of the second lane, and to determine the fifth score based on the second ratio.

[0144] The determination unit is used to determine, based on the second distance between the driver vehicle and the second lane at at least one pre-autonomous driving moment, and the second distance between the driver vehicle and the second lane at that autonomous driving moment, whether the second distance between the driver vehicle and the second lane remains unchanged across multiple autonomous driving moments, wherein each pre-autonomous driving moment is earlier than that autonomous driving moment. The fourth score determination unit is used to determine, if the second distance between the driver vehicle and the second lane remains unchanged across multiple autonomous driving moments, the duration of the unchanged distance up to that autonomous driving moment, and to determine a sixth score based on the duration of the unchanged distance, and to determine a third score based on the fifth score and the sixth score. The fifth score determination unit is used to take the fifth score as the third score if the second distance between the driver vehicle and the second lane at that autonomous driving moment is different from the second distance at the previous autonomous driving moment.

[0145] In one embodiment, the sixth score is positively correlated with the duration of the distance. The fourth score determination unit includes a sixth score determination unit, wherein the sixth score determination unit is used to determine the third score by summing the fifth score and the sixth score if the sixth score is greater than or equal to a preset score threshold; otherwise, the fifth score is used as the third score.

[0146] In one embodiment, the identification module 640 is configured to perform the following steps: for each autonomous driving moment, if the first confidence level of the third lane at that autonomous driving moment is greater than or equal to a first preset confidence level threshold, and the second confidence level of the fourth lane adjacent to the third lane at that autonomous driving moment is greater than or equal to a second preset threshold, then it is determined that the main vehicle has lane-changing behavior at that autonomous driving moment.

[0147] In one embodiment, the device 600 further includes a product calculation module and a target lane change behavior determination module. The product calculation module calculates the product between the first confidence level of the third lane and the second confidence level of the fourth lane at each autonomous driving moment where a lane change occurs. The target lane change behavior determination module identifies the lane change behavior corresponding to the largest product among the products as the target lane change behavior and obtains the lane change start time and lane change end time corresponding to the target lane change behavior.

[0148] In one embodiment, the target lane change behavior determination module includes a target lane determination unit, a start time determination unit, and an end time determination unit. The target lane determination unit determines the target third lane and the target fourth lane corresponding to the target lane change behavior. The start time determination unit determines, among the target first confidence levels corresponding to the target third lane, the first target first confidence level that is greater than or equal to the first preset confidence level threshold, according to the chronological order of the autonomous driving times, and uses the autonomous driving time corresponding to the first target first confidence level as the lane change start time corresponding to the target lane change behavior. The end time determination unit determines, among the target fourth lane's second confidence levels, the maximum second confidence level, and uses the autonomous driving time corresponding to the maximum second confidence level as the lane change end time corresponding to the target lane change behavior.

[0149] In one embodiment, this application also provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the lane change behavior recognition method described in any of the above embodiments.

[0150] In one embodiment, this application also provides a computer device. The computer device stores computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the lane change behavior recognition method described in any of the above embodiments.

[0151] Indicatively, Figure 7 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. In one example, the computer device can be a server. (Refer to...) Figure 7 The computer device 900 includes a processing component 902, which further includes one or more processors, and memory resources represented by memory 901 for storing instructions, such as application programs, that can be executed by the processing component 902. The application programs stored in memory 901 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 902 is configured to execute instructions to perform the steps of the lane change behavior recognition method described in any of the above embodiments.

[0152] The computer device 900 may also include a power supply component 903 configured to perform power management of the computer device 900, a wired or wireless network interface 904 configured to connect the computer device 900 to a network, and an input / output (I / O) interface 905. The computer device 900 may operate on an operating system stored in memory 901, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0153] Those skilled in the art will understand that the internal structure of the computer device shown in this application is merely a block diagram of a portion of the structure related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0154] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. In this document, "a," "an," "the," "the," and "its" may also include plural forms unless the context clearly indicates otherwise. "Multiple" refers to at least two, such as 2, 3, 5, or 8, etc. "And / or" includes any and all combinations of the related listed items.

[0155] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0156] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for recognizing lane change behavior, characterized in that, The method includes: Obtain the location of the master vehicle and the first lane to which the master vehicle belongs at each autonomous driving moment; For each autonomous driving moment, based on the position of the master vehicle corresponding to the autonomous driving moment and the lane-changing state of the master vehicle at the autonomous driving moment, a first confidence level corresponding to the first lane at the autonomous driving moment is determined. The first confidence level is used to reflect the possibility of the master vehicle changing lanes from the first lane to other lanes besides the first lane. For each autonomous driving moment, if at least part of the master vehicle is traveling in the second lane during the autonomous driving moment, then based on the position of the master vehicle corresponding to the autonomous driving moment and the lane change status of the master vehicle during the autonomous driving moment, the second confidence level corresponding to the second lane during the autonomous driving moment is determined. The second confidence level is used to reflect the possibility that the master vehicle changes lanes from other lanes to the second lane. For each autonomous driving moment, if the first confidence level of the first lane at that autonomous driving moment is greater than or equal to the first preset confidence level threshold, and the second confidence level of the second lane adjacent to the first lane at that autonomous driving moment is greater than or equal to the second preset threshold, then it is determined that the driver vehicle has a lane-changing behavior at that autonomous driving moment.

2. The lane change behavior recognition method according to claim 1, characterized in that, The step of determining the first confidence level of the first lane at the time of autonomous driving based on the position of the master vehicle at the time of autonomous driving and the lane-changing state of the master vehicle at the time of autonomous driving includes: Based on the position of the main vehicle and the position of the center line of the first lane corresponding to the autonomous driving moment, calculate the first distance between the main vehicle and the first lane at the autonomous driving moment, and determine the first score based on the first distance; If the lane-changing state of the master vehicle at the autonomous driving moment is a lane-changing request state, then the duration of the first state of the lane-changing request state up to the autonomous driving moment is determined, and a second score is determined based on the duration of the first state, and the first confidence level of the first lane at the autonomous driving moment is determined based on the first score and the second score. If the lane-changing state of the master vehicle at the moment of autonomous driving is not the requested lane-changing state, then the first confidence level of the first lane at the moment of autonomous driving is determined according to the first score.

3. The lane change behavior recognition method according to claim 2, characterized in that, The step of determining the first score based on the first interval includes: Calculate the first ratio between the first spacing and the lane width of the first lane; The first ratio is processed based on the following expression to determine the first score: ln(1+(e-1)·a) In the formula, e is the natural constant, and a is the first ratio.

4. The lane change behavior recognition method according to claim 1, characterized in that, The step of determining the second confidence level of the second lane at the time of autonomous driving based on the position of the master vehicle at the time of autonomous driving and the lane-changing state of the master vehicle at the time of autonomous driving includes: Based on the position of the main vehicle and the position of the center line of the second lane corresponding to the autonomous driving moment, calculate the second distance between the main vehicle and the second lane at the autonomous driving moment, and determine the third score based on the second distance; If the lane-changing state of the master vehicle at the autonomous driving moment is the end of lane-changing state, then the duration of the second state of the end of lane-changing state up to the autonomous driving moment is determined, and a fourth score is determined based on the duration of the second state. Furthermore, the second confidence level corresponding to the second lane at the autonomous driving moment is determined based on the third score and the fourth score. If the lane-changing state of the master vehicle at the moment of autonomous driving is not the end of lane-changing state, then the second confidence level of the second lane at the moment of autonomous driving is determined according to the third score.

5. The lane change behavior recognition method according to claim 4, characterized in that, The step of determining the third fraction based on the second interval includes: Calculate a second ratio between the second distance between the main vehicle and the second lane at the time of autonomous driving and the lane width of the second lane, and determine a fifth score based on the second ratio; Based on the second distance between the main vehicle and the second lane at at least one pre-autopilot moment, and the second distance between the main vehicle and the second lane at that autopilot moment, it is determined whether the second distance between the main vehicle and the second lane remains unchanged at multiple autopilot moments, and each pre-autopilot moment is earlier than that autopilot moment; If the second distance between the main vehicle and the second lane remains unchanged at multiple autonomous driving moments, then the duration of the distance during which the second distance remains unchanged up to that autonomous driving moment is determined, and a sixth score is determined based on the duration of the distance, and a third score is determined based on the fifth score and the sixth score. If the second distance between the main vehicle and the second lane at this autonomous driving moment is different from the second distance at the previous autonomous driving moment, then the fifth score is used as the third score.

6. The lane change behavior recognition method according to claim 5, characterized in that, The sixth score is positively correlated with the duration of the distance; the step of determining the third score based on the fifth score and the sixth score includes: If the sixth score is greater than or equal to a preset score threshold, then the sum of the fifth score and the sixth score is taken as the third score; otherwise, the fifth score is taken as the third score.

7. The lane change behavior recognition method according to claim 1, characterized in that, The method further includes: For each autonomous driving moment in which lane changing occurs, calculate the product between the first confidence level of the first lane and the second confidence level of the second lane in that autonomous driving moment. The lane change behavior corresponding to the largest product among the various products is taken as the target lane change behavior, and the start time and end time of the lane change corresponding to the target lane change behavior are obtained respectively.

8. The lane change behavior recognition method according to claim 7, characterized in that, The steps of obtaining the lane change start time and lane change end time corresponding to the target lane change behavior respectively include: Determine the target third lane and target fourth lane corresponding to the target lane change behavior; In each target first confidence level corresponding to the target third lane, the first target first confidence level that is greater than or equal to the first preset confidence level threshold is determined according to the time sequence of each of the autonomous driving moments, and the autonomous driving moment corresponding to the first target first confidence level is taken as the lane change start moment corresponding to the target lane change behavior. The maximum second confidence level is determined among the various second confidence levels corresponding to the target fourth lane, and the automatic driving time corresponding to the maximum second confidence level is taken as the lane change end time corresponding to the target lane change behavior.

9. A lane change behavior recognition device, characterized in that, The device includes: The location acquisition module is used to acquire the location of the master vehicle and the first lane to which the master vehicle belongs at each autonomous driving moment; The first confidence level acquisition module is used to determine the first confidence level of the first lane at each autonomous driving moment based on the position of the master vehicle and the lane-changing state of the master vehicle at that autonomous driving moment. The first confidence level is used to reflect the possibility of the master vehicle changing lanes from the first lane to other lanes besides the first lane. The second confidence level acquisition module is used to determine the second confidence level of the second lane at each autonomous driving moment if at least part of the master vehicle is traveling in the second lane at that autonomous driving moment, based on the position of the master vehicle corresponding to that autonomous driving moment and the lane change state of the master vehicle at that autonomous driving moment. The second confidence level is used to reflect the possibility that the master vehicle changes lanes from other lanes to the second lane. The identification module is used to determine that the main vehicle has lane-changing behavior at each autonomous driving moment if the first confidence level of the first lane at that autonomous driving moment is greater than or equal to a first preset confidence level threshold, and the second confidence level of the second lane adjacent to the first lane at that autonomous driving moment is greater than or equal to a second preset threshold.

10. A storage medium, characterized in that, The storage medium stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the lane change behavior recognition method as described in any one of claims 1 to 8.

11. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the lane change behavior recognition method as described in any one of claims 1 to 8.

Images