Lane change identification method and device, equipment, storage medium and product

Abstract

The invention discloses a lane changing recognition method, device and equipment, a storage medium and a product, and relates to the technical field of intelligent driving, the method comprises the steps that driving map topological relation information and a driving scene target data frame are acquired, and the driving map topological relation information comprises lane information; determining a current lane and an undetermined lane according to the driving scene target data frame and the lane information; acquiring historical frame data and future frame data corresponding to the target scene data frame; and performing lane changing identification on the vehicle based on the historical frame data, the future frame data, the current lane and the to-be-determined lane, and through the method, firstly determining the current lane and the to-be-determined lane of the vehicle, and then identifying the lane changing behavior of the vehicle, thereby improving the accuracy of identifying the lane changing behavior of the vehicle by the intelligent driving system.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology and, in particular, to a lane change recognition method, device, equipment, storage medium and product. Background Art

[0002] One of the core functions of intelligent driving systems is to accurately identify and predict the behavior of traffic participants, especially the lane changing behavior of vehicles, which is crucial for vehicle intention recognition and trajectory prediction.

[0003] Existing technologies for lane change recognition usually rely on vehicle sensor data and simple behavioral pattern recognition. These methods can provide basic lane change judgment in some cases, but are often limited by sensor accuracy and environmental factors such as bad weather or occlusion, resulting in low recognition accuracy. In addition, existing technologies often lack a deep understanding and utilization of the complex relationship between lanes, resulting in insufficient lane change recognition capabilities in multi-lane, complex traffic scenarios.

[0004] This makes it difficult for existing technologies to accurately identify the lane-changing behavior of vehicles in complex traffic flows, especially in urban road environments with many lanes and heavy traffic. In addition, existing technologies are also insufficient in tracking and predicting vehicle behavior between consecutive frames, and cannot effectively handle the rapid lane-changing behavior of vehicles in a short period of time, which may cause the intelligent driving system to make wrong decisions in an emergency, increasing safety risks.

[0005] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0006] The main purpose of this application is to provide a lane change recognition method, device, equipment, storage medium and computer program product, aiming to improve the accuracy of the intelligent driving system in identifying the vehicle's lane change behavior.

[0007] To achieve the above objectives, the present application proposes a lane change recognition method, the method comprising: Acquire driving map topology relationship information and driving scene target data frame, wherein the driving map topology relationship information includes lane information; Determine a current lane and a pending lane according to the driving scene target data frame and the lane information; Acquire historical frame data and future frame data corresponding to the target scene data frame; Based on the historical frame data, the future frame data, the current lane and the pending lane, lane change recognition is performed on the vehicle.

[0008] In one embodiment, the driving scene target data frame includes a driving scene current data frame and a driving scene predecessor data frame, and the step of determining the current lane and the pending lane according to the driving scene target data frame and the lane information includes: According to a preset lane binding algorithm and the lane information, respectively calculating a current projection distance of the vehicle on each lane in a current data frame of the driving scene, and a front-drive projection distance of the vehicle on each lane in a front-drive data frame of the driving scene; Respectively screening out the minimum current projection distance among the current projection distances and the minimum front-drive projection distance among the front-drive projection distances; The lane corresponding to the minimum current projection distance is determined as the current lane, and the lane corresponding to the front-drive projection distance is determined as the pending lane.

[0009] In one embodiment, the step of performing lane change recognition on the vehicle based on the historical frame data, the future frame data, the current lane and the pending lane includes: Obtaining a lane sequence number of the current lane and a lane sequence number of the pending lane; Determining whether the lane sequence number of the current lane is the same as the lane sequence number of the pending lane; If the lane sequence number of the current lane is different from the lane sequence number of the pending lane, determining the current lane sequence and the pending lane sequence according to the lane sequence number of the current lane and the lane sequence number of the pending lane; Calculating the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence based on a preset lane distance threshold and the historical frame data and the future frame data; Whether the vehicle changes lanes is identified according to the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence.

[0010] In one embodiment, the step of calculating the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence based on a preset lane distance threshold and the historical frame data and the future frame data comprises: Determining a first identified lane set on the pending lane sequence and a second identified lane set on the current lane sequence according to a preset lane distance threshold; The stability of the vehicle on the pending lane sequence is determined based on the historical frame data and the first identified lane set, and the stability of the vehicle on the current lane sequence is determined based on the future frame data and the second identified lane set.

[0011] In one embodiment, the historical frame data includes a plurality of historical frames, the future frame data includes a plurality of future frames, and the steps of determining the stability of the vehicle on the pending lane sequence based on the historical frame data and the first identified lane set, and determining the stability of the vehicle on the current lane sequence based on the future frame data and the second identified lane set include: Identifying, in the historical frame data, the number of times the vehicle is located in the first identified lane set and the number of times the vehicle appears in the historical frame; Identifying, in the future frame data, the number of times the vehicle is located in the second identified lane set and the number of times the vehicle appears in the future frame; Determining the stability of the vehicle on the pending lane sequence according to a ratio of the number of times the vehicle is located in the first identified lane set to the number of times the vehicle appears in the historical frame; The stability of the vehicle on the current lane sequence is determined according to a ratio of the number of times the vehicle is located in the second identified lane set to the number of times the vehicle appears in the future frame.

[0012] In one embodiment, the step of identifying whether the vehicle changes lanes according to the stability of the vehicle in the current lane sequence and the stability of the vehicle in the pending lane sequence includes: Determining whether the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence are both greater than a preset threshold; If the stability of the vehicle in the current lane sequence and the stability of the vehicle in the pending lane sequence are both greater than a preset threshold, it is determined that the vehicle has changed lanes.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a lane change recognition device, the lane change recognition device comprising: A first acquisition module is used to acquire driving map topological relationship information and driving scene target data frames, wherein the driving map topological relationship information includes lane information and lane-to-lane topological relationship information; A determination module, which determines a current lane and a pending lane according to the driving scene target data frame and the lane information; A second acquisition module acquires historical frame data and future frame data corresponding to the target scene data frame; The recognition module performs lane change recognition on the vehicle based on the historical frame data, the future frame data, the current lane and the pending lane.

[0014] In addition, to achieve the above-mentioned purpose, the present application also proposes a lane change recognition device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the lane change recognition method described above.

[0015] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the lane change recognition method described above are implemented.

[0016] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the lane change recognition method described above are implemented.

[0017] One or more technical solutions proposed in the present application obtain driving map topological relationship information and driving scene target data frames, wherein the driving map topological relationship information includes lane information; determine the current lane and the pending lane according to the driving scene target data frame and the lane information; obtain historical frame data and future frame data corresponding to the target scene data frame; and identify lane changes of the vehicle based on the historical frame data, future frame data, and the current lane and the pending lane. The above method first determines the current lane and the pending lane of the vehicle, and then identifies the lane change behavior of the vehicle, thereby improving the accuracy of the intelligent driving system in identifying the lane change behavior of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 A flowchart diagram of the first embodiment of the lane change recognition method of the present application is provided; Figure 2 A flowchart diagram of the second embodiment of the lane change recognition method of the present application is provided; Figure 3 A schematic diagram of the relationship between the preceding and succeeding lanes in the lane change recognition method of this application; Figure 4The lane change recognition method of this application is based on topological search of lane schematic diagrams on the same lane sequence; Figure 5 This is a schematic diagram of the module structure of the lane change recognition device according to an embodiment of the present application; Figure 6 Schematic diagram of the device structure of the hardware operating environment involved in the lane change recognition method in the embodiment of the present application.

[0021] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0022] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0023] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0024] The main solution of the embodiment of the present application is to propose an algorithm for realizing lane change recognition of intelligent agents in a high-precision map scenario. The algorithm is implemented by the following steps: First, the map topology at the frame level is constructed to provide lane information (lane_info) and lane-to-lane relationship information (lane_to_lane) for each frame. Lane information includes trajectory point coordinates, sub-segment midline point coordinates, starting point, end point, length, direction, etc., while lane relationship information covers the adjacent lanes on the left and right sides of the lane, and the preceding and succeeding lanes.

[0025] Secondly, each frame of the scene data is processed and lane change recognition is performed for each vehicle agent in the frame. The lane number of the vehicle in the current frame and the previous frame is determined by the vehicle lane binding algorithm. If the numbers are inconsistent, the vehicle is considered a candidate for lane change.

[0026] Then, the future frame number threshold, historical frame number threshold and distance threshold are set, and the map topology relationship is used to quickly query the lane's previous and subsequent lane information. Based on this information, a lane sequence list related to the lanes in the current and previous frames of the vehicle is constructed.

[0027] Next, the proportion of vehicles in the relevant lane sequences in the historical frames and the proportion of vehicles in these lane sequences in the future frames are calculated. By comparing these proportions with the set stability threshold, it is determined whether the vehicle has actually changed lanes.

[0028] Finally, if the stability of the vehicle in the lane sequence before and after the lane change exceeds the stability threshold, the vehicle is judged to have changed lanes. This process not only improves the accuracy of lane change recognition, but also enhances the robustness of the algorithm in complex traffic scenarios by utilizing high-precision maps and lane topology relationships.

[0029] Existing technologies for lane change recognition usually rely on vehicle sensor data and simple behavioral pattern recognition. These methods can provide basic lane change judgment in some cases, but are often limited by sensor accuracy and environmental factors such as bad weather or occlusion, resulting in low recognition accuracy. In addition, existing technologies often lack a deep understanding and utilization of the complex relationship between lanes, resulting in insufficient lane change recognition capabilities in multi-lane, complex traffic scenarios.

[0030] The defects of existing technologies are mainly manifested in the lack of accuracy and robustness of lane change recognition. Due to the lack of in-depth analysis and utilization of lane topology, existing technologies are difficult to accurately identify the lane change behavior of vehicles in complex traffic flows, especially in urban road environments with many lanes and heavy traffic. In addition, existing technologies are also insufficient in tracking and predicting vehicle behavior between consecutive frames, and cannot effectively handle the rapid lane change behavior of vehicles in a short period of time, which may cause the intelligent driving system to make wrong decisions in emergency situations, increasing safety risks.

[0031] In order to solve the shortcomings of existing technologies in lane change recognition accuracy and robustness, this solution proposes a new lane change recognition algorithm. By constructing high-precision lane topology relationships and utilizing the continuity of lane sequences, this solution can more accurately identify and predict vehicle lane change behaviors, thereby improving the navigation capability and safety of intelligent driving systems in complex traffic environments.

[0032] The present application provides a solution, by obtaining driving map topological relationship information and a driving scene target data frame, the driving map topological relationship information includes lane information; determining the current lane and the pending lane according to the driving scene target data frame and the lane information; obtaining historical frame data and future frame data corresponding to the target scene data frame; based on the historical frame data, future frame data and the current lane and the pending lane, performing lane change recognition on the vehicle, the above method first determines the current lane and the pending lane of the vehicle, and then identifies the lane change behavior of the vehicle, thereby improving the accuracy of the intelligent driving system in identifying the lane change behavior of the vehicle.

[0033] It should be noted that the execution subject of this embodiment may be a computing service device with lane change recognition, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a lane change recognition system, etc. The following takes the lane change recognition system as an example to illustrate this embodiment and the following embodiments.

[0034] Based on this, the embodiment of the present application provides a lane change recognition method, referring to Figure 1 , Figure 1 Schematic diagram of the flow chart of the first embodiment of the lane change recognition method of the present application.

[0035] In this embodiment, the lane change recognition method includes steps S1000 to S4000: Step S1000: Acquire driving map topology relationship information and driving scene target data frame, wherein the driving map topology relationship information includes lane information; It should be noted that in the field of intelligent driving, lane change recognition of traffic participating intelligent entities (vehicles, agents) is particularly important. On the one hand, the behavior of a vehicle changing lanes is itself one of the two core issues in the prediction field (the two core issues are the recognition of vehicle intentions and the prediction of vehicle trajectories). On the other hand, the lane change behavior close to the ego vehicle (referred to as ego) will have an important impact on the driving decision of the ego vehicle. This vehicle can be identified as a key obstacle. When training a prediction model, special processing is usually performed on the moment when the vehicle changes lanes, so that the prediction model can better predict the vehicle's lane change behavior during reasoning. When producing intelligent driving data sets, accurately identifying vehicles with lane change behavior is a relatively important issue in intelligent driving data sets.

[0036] In addition, it should be noted that the topological relationship information of the driving map includes lane information, which is crucial for constructing the position relationship of vehicles in different lanes. Lane information includes not only the geometric attributes of the lane, such as the track point coordinates of the lane, the coordinates of the midline point of the sub-segment, the starting point, the end point, the length and the direction, but also the relationship attributes between lanes, such as the left adjacent lane, the right adjacent lane, the preceding adjacent lane and the succeeding adjacent lane. These information together constitute the basic data support for the recognition of vehicle lane change behavior.

[0037] In addition, it should be noted that the purpose of determining the current lane and the pending lane is to identify whether the vehicle has changed lanes. The implementation of this step depends on the vehicle lane binding algorithm. By calculating the position of the vehicle in the current frame and the previous frame, the system can determine the projection distance of the vehicle on the lane and select the lane corresponding to the minimum projection distance, thereby determining the current lane of the vehicle and the lane that may be changed into.

[0038] For example, in a specific implementation, the system may parse high-precision map data to obtain detailed coordinate information for each lane in the traffic scene, and calculate through a lane binding algorithm that the vehicle is in lane 3 in the current frame and lane 2 in the previous frame.

[0039] It should also be noted that for each driving scene target data frame, it contains driving map topology information, where the driving map topology information can include lane information lane_info and lane-to-lane relationship information lane_to_lane, etc. lane_info is a dictionary, where the key value is the lane number laneid, and the value value is the specific information of the lane, which is also composed of a dictionary, so that the detailed information or required information of the lane can be quickly obtained through the lane number laneid.

[0040] The specific information of the lane mainly includes the track point coordinate information point_coords of the lane, which contains the coordinate information of all points in the lane; the centerline point coordinate information center_coods of the lane sub-segment, where the sub-segment is a segment composed of two adjacent coordinate points on the lane; the starting point coordinate information start of the lane; the end point coordinate information end of the lane; the length information length of the lane; the heading information heading of the lane, indicating the orientation of the lane in the odom coordinate system; the length information lengths of the lane sub-segment, the heading information headings of the lane sub-segment, and the number of the lane sub-segments.

[0041] lane_to_lane is a dictionary, where the key value is the lane number laneid, and the value value is the information related to the lane. It is also composed of a dictionary, including: the lane's left adjacent lane information left, the lane's right adjacent lane information right, the lane's previous adjacent lane information prev, and the lane's next adjacent lane information next. Through the constructed relationship between lanes, you can quickly obtain detailed information about the adjacent lanes of a specified lane.

[0042] Step S2000: determining a current lane and a pending lane according to the driving scene target data frame and the lane information; It should be noted that in this embodiment, the historical frame data and future frame data are obtained in order to more comprehensively analyze the dynamic behavior of the vehicle. The historical frame data provides the location and behavior information of the vehicle in the past period of time, while the future frame data includes the possible movement trajectory of the vehicle predicted by the system. Combining the information of the current lane and the pending lane, the system can conduct in-depth analysis and prediction of the vehicle's lane change behavior.

[0043] Alternatively, this step can be implemented by setting a specific time window. For example, the system can choose to consider the data of the next 15 frames and the data of the past 15 frames. Within this time window, the system will analyze the frequency of the vehicle's appearance in the current lane sequence and the pending lane sequence to evaluate the vehicle's stability in the front and rear lanes, and further determine whether the vehicle has changed lanes.

[0044] For example, in a specific implementation, the system may collect the position data of the vehicle in the past 30 frames and the future 30 frames, and use this data to observe whether the vehicle has a continuous lane change trend.

[0045] Specifically, lane change recognition is performed for each vehicle agent in the frame frame. For the vehicle agent in the current frame frame, the lane where the vehicle is located in the current frame frame is calculated through the vehicle lane binding algorithm, and the current lane number is recorded as cur_frame_lane_id. The lane corresponding to cur_frame_lane_id is also the current lane of the vehicle. The lane where the vehicle is located in the previous frame pre_frame of the current frame is further calculated. The lane number of the previous frame is recorded as pre_frame_lane_id, and the lane corresponding to pre_frame_lane_id is also the pending lane. If cur_frame_lane_id is equal to pre_frame_lane_id, it means that the vehicle agent has not changed lanes. If the values ​​of cur_frame_lane_id and pre_frame_lane_id are not equal, the vehicle may have changed lanes, but it is only a candidate for lane change. Whether the vehicle has changed lanes needs further identification and confirmation.

[0046] In a feasible implementation manner, step S2000 may include steps S2100 to S2300: Step S2100: Calculating the current projection distance of the vehicle on each lane in the current data frame of the driving scene and the front-drive projection distance of the vehicle on each lane in the front-drive data frame of the driving scene, respectively, according to a preset lane binding algorithm and the lane information; Step S2200: filtering out the minimum current projection distance among the current projection distances and the minimum front-drive projection distance among the front-drive projection distances respectively; Step S2300: Determine that the lane corresponding to the minimum current projection distance is the current lane, and the lane corresponding to the front-drive projection distance is the pending lane.

[0047] It should be noted that the lane binding algorithm is a key technology used to determine the position of the vehicle in the driving scene. It is achieved by calculating the projection distance of the vehicle on each lane in the current data frame and the previous data frame. The current projection distance refers to the vertical distance of the vehicle from the center line of the lane or a reference point in the current frame, while the previous projection distance is a similar distance of the vehicle in the previous frame. This calculation method allows the system to accurately identify the changes in the vehicle's lane position between consecutive frames.

[0048] In addition, the purpose of screening the minimum projection distance is to find the lane that the vehicle is most likely to be traveling in. By comparing the projection distances in the current frame and the preceding frame, the system can identify the lane change of the vehicle between the two frames. The lane corresponding to the minimum current projection distance is considered to be the lane the vehicle is currently in, while the lane corresponding to the minimum preceding projection distance may be the lane that the vehicle is about to change to, that is, the pending lane.

[0049] In addition, determining the current lane and pending lane is the basis of the lane change recognition process. This not only involves accurate judgment of the vehicle's current position, but also involves the prediction of possible lane change behaviors in the future. Through this process, the system can establish a relationship model between the vehicle and the lane, laying a solid foundation for subsequent lane change behavior analysis.

[0050] For example, in one feasible implementation, the system first uses the lane binding algorithm to process the current data frame and the preceding data frame of the driving scene, and calculates the projection distance of the vehicle on all lanes. Then, the system selects the minimum value of these projection distances, identifies the lane corresponding to the minimum current projection distance as the vehicle's current lane, and the lane corresponding to the minimum preceding projection distance is tentatively designated as the target lane that the vehicle may change lanes to. This process not only takes into account the vehicle's immediate position, but also predicts its potential lane-changing behavior, providing important information for the decision-making of the intelligent driving system.

[0051] Step S3000: Acquire historical frame data and future frame data corresponding to the target scene data frame; It should be noted that the key to lane change recognition is to evaluate the stability of the vehicle in different lane sequences. By comparing the stability of the vehicle in the current lane sequence and the pending lane sequence, the system can determine whether the vehicle has the intention and behavior to change lanes. This evaluation process involves in-depth analysis of historical frame data and future frame data.

[0052] In addition, the stability is calculated by counting the number of times a vehicle appears in a specific lane sequence and the total number of times it appears. If this ratio is greater than a preset threshold, it indicates that the vehicle's driving behavior in the lane sequence is stable, thereby providing support for the identification of lane change behavior based on the stability data.

[0053] For example, in a specific embodiment, the system finds that in analyzing the 15 frames of data before and after the current frame, a total of 30 frames of data, the vehicle was located in the third lane in 25 frames and in other lanes in 5 frames, which indicates that the vehicle has a high degree of stability in the third lane and a low possibility of changing lanes.

[0054] Step S4000: Based on the historical frame data, the future frame data, the current lane and the pending lane, lane change recognition is performed on the vehicle.

[0055] It should be noted that the final lane change recognition is based on a comprehensive evaluation of the vehicle's stability in the current lane sequence and the pending lane sequence. If the vehicle's stability in both lane sequences is higher than the preset threshold, the system will determine that the vehicle has changed lanes.

[0056] In addition, this judgment process is dynamic, which means that the system will continuously adjust the recognition results of the vehicle's lane change behavior based on the real-time updated frame data. This dynamic evaluation mechanism can improve the accuracy and real-time performance of lane change recognition, thereby providing more reliable decision support for the intelligent driving system.

[0057] For example, in a specific implementation, the system determines through analysis that the stability of the vehicle in the current lane is 0.9, and the stability in the pending lane is 0.8, both exceeding the preset threshold of 0.75, so the system determines that the vehicle has indeed changed lanes.

[0058] In a feasible implementation manner, step S4000 may include steps S4100 to S4500: Step S4100: Obtaining the lane sequence number of the current lane and the lane sequence number of the pending lane; Step S4200: determining whether the lane sequence number of the current lane is the same as the lane sequence number of the pending lane; Step S4300: if the lane sequence number of the current lane is different from the lane sequence number of the pending lane, determining the current lane sequence and the pending lane sequence according to the lane sequence number of the current lane and the lane sequence number of the pending lane; It should be noted that the lane sequence number is a number used to uniquely identify the lane sequence of each lane, which plays a vital role in the lane change recognition process. By obtaining the sequence numbers of the current lane and the pending lane, the system can accurately identify the possible lane change behavior of the vehicle and further analyze the behavior pattern of the vehicle in different lane sequences.

[0059] In addition, it should be noted that determining whether the sequence numbers of the current lane and the pending lane are the same is a key step in identifying lane change behavior. If the sequence numbers are the same, it means that the vehicle has not changed lanes; if they are different, the system needs to further analyze the stability of the vehicle in the two lane sequences.

[0060] Step S4400: Calculating the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence based on a preset lane distance threshold and the historical frame data and the future frame data; Step S4500: Identify whether the vehicle changes lanes based on the stability of the vehicle in the current lane sequence and the stability of the vehicle in the pending lane sequence.

[0061] Furthermore, the current lane sequence and the pending lane sequence are determined based on the lane sequence numbers. The system will use the lane relationship information stored in the high-precision map, such as the previous and subsequent relationships of the lanes, to construct these two lane sequences. This step is crucial for understanding the vehicle's driving path and predicting its possible lane change behavior.

[0062] In addition, the stability of the vehicle in the current lane sequence and the pending lane sequence is calculated by analyzing the historical frame data and the future frame data. The system will evaluate the ratio of the number of times the vehicle appears in these lane sequences to the total number of times according to the preset lane distance threshold to determine the stability of the vehicle.

[0063] Finally, based on the vehicle's stability in the two lane sequences, the system can identify whether the vehicle has changed lanes. If the vehicle's stability in both lane sequences is high, the system will determine that the vehicle has not changed lanes; if the stability difference is significant, the system will determine that the vehicle has changed lanes.

[0064] For example, in one feasible implementation, the system first obtains the serial numbers of the vehicle's current lane and the lane it may change into. By comparing the two serial numbers, the system finds that they are different, indicating that the vehicle may change lanes. Subsequently, the system constructs the current lane sequence and the pending lane sequence based on the topological relationship information of the lanes, and calculates the stability of the vehicle on these two sequences based on historical frame data and future frame data. By analyzing these stability indicators, the system ultimately identifies that the vehicle has indeed changed lanes, and provides accurate information for the decision-making of the intelligent driving system.

[0065] The present embodiment provides a lane change recognition method, which obtains driving map topological relationship information and a driving scene target data frame, wherein the driving map topological relationship information includes lane information; determines a current lane and a pending lane according to the driving scene target data frame and the lane information; obtains historical frame data and future frame data corresponding to the target scene data frame; and performs lane change recognition on the vehicle based on the historical frame data, future frame data, and the current lane and the pending lane. The above method first determines the current lane and the pending lane of the vehicle, and then recognizes the lane change behavior of the vehicle, thereby improving the accuracy of the intelligent driving system in recognizing the lane change behavior of the vehicle.

[0066] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 2 , step S4400 of the lane change recognition method further includes steps S4410 to S4420: Step S4410: determining a first identified lane set on the pending lane sequence and a second identified lane set on the current lane sequence according to a preset lane distance threshold; It should be noted that, in this embodiment, the lane distance threshold is a key parameter used to determine the recognition range of the vehicle in the lane sequence. The system uses this threshold to define which lanes should be included in the recognition lane sets of the pending lane sequence and the current lane sequence. The first recognition lane set and the second recognition lane set are the basis for the vehicle lane change behavior analysis, and they include lanes connected to the pending lane and the current lane within a certain distance.

[0067] Additionally, the purpose of determining these two identified lane sets is to more accurately assess the stability of the vehicle in the lane sequence. The system will consider the connection between lanes and their distances relative to the vehicle to construct a range of lanes that the vehicle may travel. This step is crucial for understanding the vehicle's driving path and predicting its lane change behavior.

[0068] For example, in a specific embodiment, the system identifies all lanes connected to the pending lane and within the threshold range according to a preset lane distance threshold, forming a first identified lane set. Similarly, the system also identifies lanes connected to the current lane and within the same distance threshold range, forming a second identified lane set.

[0069] More specifically, in this embodiment, a future frame number threshold consider_point_future_num and a history frame number threshold consider_point_history_num (for example, both are 15), as well as a lane distance threshold (check_dist_threshold) may be set. Based on the pre-built frame-level map topology, lane_to_lane can be used to quickly query the lane's predecessor lane (the predecessor lane is the lane that is connected to the lane and is searched in the opposite direction of the lane) and subsequent lane information (the subsequent lane is the lane that is connected to the lane and is searched in the same direction as the lane). Figure 3 As shown in the figure, the relationship between the preceding lane and the following lane is shown. Lane 1 is the preceding lane of lane 2, and lanes 3 and 4 are the following lanes of lane 2. The length, coordinates and other information of any lane can be quickly obtained through the lane information lane_info.

[0070] Based on the predecessor and successor relationship between lanes in the frame-level map topology and the distance threshold (check_dist_threshold), a list pre_frame_lane_pre_list consisting of all lanes whose pre_frame_lane_id is in the same lane sequence and whose distance to pre_frame_lane_id is less than the threshold check_dist_threshold can be obtained, which corresponds to the first identified lane set on the pending lane sequence.

[0071] Specifically, Figure 4 Assume that pre_frame_lane_id is Figure 4 Lane 6 in the list, then the predecessor lane of lane 6 is 5. If the length of lane 5 is less than the distance threshold (check_dist_threshold), then continue to search based on lane 5. There are two predecessor lanes of lane 5, lane 3 and lane 4. Lane 3 and lane 4 are processed separately. If the sum of the length of lane 3 and the length of lane 5 is less than the distance threshold (check_dist_threshold), then further search is performed based on lane 3. The predecessor lane of lane 3 is 1. If the sum of the lengths of lanes 1, 3, and 5 is greater than the distance threshold (check_dist_threshold), then no further search is performed based on lane 1. For lane 4, if the sum of the lengths of lane 4 and lane 5 is greater than the distance threshold (check_dist_threshold), no further search is performed based on lane 4. Finally, the pre_frame_lane_pre_list that is in the same lane sequence as pre_frame_lane_id and meets the distance threshold condition can be obtained, which includes lane 6, lane 5, lane 3, lane 1, and lane 4. This method can be used to search for pre_frame_lane_pre_list in complex lane situations including merging.

[0072] Similarly, based on the predecessor and successor relationship between lanes in the frame-level map topology and the distance threshold (check_dist_threshold), a list cur_frame_lane_pre_list consisting of all lanes whose cur_frame_lane_id is in the same lane sequence and whose distance to cur_frame_lane_id is less than the threshold check_dist_threshold can be obtained, which corresponds to the second identified lane set on the current lane sequence.

[0073] Step S4420: Determine the stability of the vehicle on the pending lane sequence based on the historical frame data and the first identified lane set, and determine the stability of the vehicle on the current lane sequence based on the future frame data and the second identified lane set.

[0074] It should be noted that the calculation of stability is based on historical frame data and future frame data, which provide the driving information of the vehicle in the past and future. By analyzing the frequency of the vehicle's appearance in the first and second lane sets, the system can evaluate the stability of the vehicle in these two lane sequences.

[0075] Additionally, the implementation of this step involves tracking the position of the vehicle in the historical and future frames, and counting its distribution in the two identified lane sets. The stability of the vehicle on the pending lane sequence reflects the stability of the lane sequence to which the vehicle belongs before the lane change, while the stability of the vehicle on the current lane sequence reflects the stability of the lane sequence to which the vehicle belongs after the lane change.

[0076] For example, in a specific implementation, the system calculates the number of times the vehicle appears in the first identified lane set by analyzing the historical frame data, and compares it with the total number of times the vehicle appears in the historical frames to obtain the stability of the vehicle in the pending lane sequence. Similarly, the system also analyzes the future frame data and calculates the stability of the vehicle in the second identified lane set, thereby evaluating the stability of the vehicle in the current lane sequence. In this way, the system can accurately determine whether the vehicle has changed lanes.

[0077] In a feasible implementation, step S4420 may include steps S4421 to S4424: Step S4421: Identify the number of times the vehicle is located in the first identified lane set and the number of times the vehicle appears in the historical frame in the historical frame data; Step S4422: Identify the number of times the vehicle is located in the second identified lane set and the number of times the vehicle appears in the future frame data; Step S4423: determining the stability of the vehicle on the pending lane sequence according to the ratio of the number of times the vehicle is located in the first identified lane set to the number of times the vehicle appears in the historical frame; Step S4424: Determine the stability of the vehicle on the current lane sequence based on the ratio of the number of times the vehicle is located in the second identified lane set to the number of times the vehicle appears in the future frame.

[0078] It should be noted that identifying the position of the vehicle in the historical frame data and the future frame data is the core part of the lane change recognition algorithm. This process involves counting the frequency of the vehicle's appearance in a specific lane set to evaluate the vehicle's stability. Stability is an indicator that reflects the stability of the vehicle in a certain lane sequence. It is calculated based on the ratio of the number of times the vehicle appears in a specific lane set to the total number of appearances.

[0079] In addition, by identifying the number of times a vehicle is in the first identified lane set, the system can determine the distribution of vehicles in historical frames. This provides a basis for evaluating the stability of the vehicle on the pending lane sequence. Similarly, identifying the number of times a vehicle is in the second identified lane set helps the system understand the driving trend of the vehicle in future frames, thereby evaluating the stability of the vehicle on the current lane sequence.

[0080] In addition, the implementation of these steps requires the system to have efficient data processing capabilities, to be able to quickly extract vehicle location information from a large amount of frame data, and to perform accurate statistics and analysis. This process requires not only the accuracy of the algorithm, but also the high operating efficiency of the algorithm to meet the needs of real-time or near real-time lane change recognition.

[0081] For example, in one feasible implementation, the system first analyzes the distribution of vehicles in historical frame data, counts the frequency of vehicles appearing in the first identified lane set, and calculates the ratio of this frequency to the number of times the vehicle appears in all historical frames, thereby obtaining the stability of the vehicle in the pending lane sequence. Then, the system performs the same analysis on the distribution of vehicles in future frame data and obtains the stability of the vehicle in the current lane sequence. By comparing and comprehensively evaluating these two stabilities, the system can accurately determine whether the vehicle has changed lanes, as well as the stability and continuity of the lane change behavior. This overall solution not only improves the accuracy of lane change recognition, but also provides strong data support for the decision-making of the intelligent driving system.

[0082] More specifically, the ratio of the vehicle in the lane sequence where cur_frame_lane_id is located and the vehicle in the lane sequence where pre_frame_lane_id is located in several frames of historical frame data (consider_point_history_num) is calculated, and the ratio of the vehicle in the lane sequence where cur_frame_lane_id is located and the vehicle in the lane sequence where pre_frame_lane_id is located in several frames of future frames (consider_point_future_num) is calculated.

[0083] For historical frames, based on the frame-level map topology lane-to-lane relationship and the distance threshold (check_dist_threshold), a list cur_frame_lane_pre_list consisting of all lanes whose cur_frame_lane_id is in the same lane sequence and whose distance to cur_frame_lane_id is less than the threshold check_dist_threshold can be obtained.

[0084] Taking the current frame into account, consider_point_history_num frames, determine the number of times the vehicle falls on the lane in pre_frame_lane_pre_list (pre_num_1), the number of times the vehicle falls on the lane in cur_frame_lane_pre_list (pre_num_2), and the number of times the vehicle appears (pre_num_3). Then, pre_num_1 divided by pre_num_3 (here rate_prev = pre_num_1 / pre_num_3) indicates the stability of the vehicle in the lane sequence (i.e., the pending lane sequence) before changing lanes.

[0085] For future frames, based on the subsequent relationship between lanes in the frame-level map topology and the distance threshold (check_dist_threshold), a list pre_frame_lane_next_list consisting of all lanes whose pre_frame_lane_id is in the same lane sequence and whose distance to pre_frame_lane_id is less than the threshold check_dist_threshold can be obtained.

[0086] Similarly, when searching for the subsequent lanes of the lane in question, based on the subsequent relationship between lanes in the frame-level map topology and the distance threshold (check_dist_threshold), a list cur_frame_lane_next_list consisting of all lanes whose cur_frame_lane_id is in the same lane sequence and whose distance to cur_frame_lane_id is less than the threshold check_dist_threshold can be obtained.

[0087] Consider the future number of frames of consider_point_future_num frames in the current frame, determine the number of times the vehicle falls on the lane in pre_frame_lane_next_list next_num_1, the number of times the vehicle falls on the lane in cur_frame_lane_next_list next_num_2, and the number of times the vehicle appears is next_num_3. Then next_num_2 divided by next_num_3 (here rate_next = next_num_2 / next_num_3) indicates the stability of the vehicle in the lane sequence (that is, the current lane sequence) after changing lanes.

[0088] In a feasible implementation, step S4500 may include steps S4510 to S4520: Step S4510: determining whether the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence are both greater than a preset threshold; Step S4520: If the stability of the vehicle in the current lane sequence and the stability of the vehicle in the pending lane sequence are both greater than a preset threshold, it is determined that the vehicle has changed lanes.

[0089] It should be noted that the stability of the vehicle is a quantitative indicator used to measure the coherence and consistency of the vehicle's driving in a specific lane sequence. In the lane change recognition algorithm, this indicator is crucial because it is directly related to the judgment of whether the vehicle has changed lanes. The preset threshold is an empirical value used to define the standard for vehicle driving stability. If the stability of the vehicle in both the current lane sequence and the pending lane sequence exceeds this threshold, it indicates that the vehicle's driving in these two lane sequences is stable, and it can be inferred that the vehicle may have completed the lane change.

[0090] Additionally, the logic for determining whether a vehicle has changed lanes is based on the comparison of the vehicle's stability in the two lane sequences with a preset threshold. This judgment process is the core of the algorithm's decision-making, which determines the final lane change recognition result. If the vehicle's stability is lower than the preset threshold, it may indicate that the vehicle has not completed the lane change, or its driving path is not stable enough to make a clear lane change judgment.

[0091] In addition, the implementation of this overall solution requires accurate data collection and efficient data processing capabilities. The system must be able to accurately track and record the position changes of the vehicle in consecutive frames and quickly calculate the stability of the vehicle in order to make accurate lane change decisions in a timely manner. This capability is crucial for intelligent driving systems because it directly affects the safety and reliability of the system.

[0092] For example, in one feasible implementation, the system first calculates the stability of the vehicle in the current lane sequence and the pending lane sequence. These two values ​​are obtained by analyzing the vehicle's position data in historical frames and future frames. The system sets a preset threshold for stability of 0.75, which means that only when the vehicle's stability in both lane sequences reaches or exceeds 75% will the system determine that the vehicle has changed lanes. In this way, the system can accurately identify the vehicle's lane change behavior and provide key decision support for intelligent driving.

[0093] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the lane change recognition method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0094] Please refer to Figure 5 The present application further provides a lane change recognition device 50, the lane change recognition device comprising: A first acquisition module 51 is configured to acquire driving map topology information and driving scene target data frames, wherein the driving map topology information includes lane information and lane topology information; A determination module 52 determines a current lane and a pending lane according to the driving scene target data frame and the lane information; A second acquisition module 53 acquires historical frame data and future frame data corresponding to the target scene data frame; The recognition module 54 performs lane change recognition on the vehicle based on the historical frame data, the future frame data, the current lane and the pending lane.

[0095] The lane change recognition device provided by the present application adopts the lane change recognition method in the above embodiment, which can solve the technical problem of lane change recognition. Compared with the prior art, the beneficial effects of the lane change recognition device provided by the present application are the same as the beneficial effects of the lane change recognition method provided by the above embodiment, and other technical features in the lane change recognition device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0096] The present application provides a lane change recognition device, which includes: at least one processor; and a memory that is communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the lane change recognition method in the above-mentioned embodiment one.

[0097] Reference below Figure 6 , which shows a schematic diagram of the structure of a lane change recognition device suitable for implementing the embodiment of the present application. The lane change recognition device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The lane change recognition device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0098] like Figure 6As shown, the lane change recognition device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the lane change recognition device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the lane change recognition device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a lane change recognition device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0099] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0100] The lane change recognition device provided in the present application adopts the lane change recognition method in the above-mentioned embodiment. Compared with the prior art, the beneficial effects of the lane change recognition device provided in the present application are the same as the beneficial effects of the lane change recognition method provided in the above-mentioned embodiment, and other technical features in the lane change recognition device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0101] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0102] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0103] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the lane change recognition method in the above-mentioned embodiment.

[0104] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0105] The computer-readable storage medium may be included in the lane change recognition device; or may exist independently without being assembled into the lane change recognition device.

[0106] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0107] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0108] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0109] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above lane change recognition method, and can solve the technical problem of lane change recognition. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the lane change recognition method provided in the above embodiment, and will not be elaborated here.

[0110] The present application also provides a computer program product, including a computer program, which implements the steps of the lane change recognition method as described above when executed by a processor.

[0111] The computer program product provided by the present application can solve the technical problem of lane change recognition. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the lane change recognition method provided by the above embodiment, which will not be repeated here.

[0112] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A lane change recognition method, characterized in that: The method is applied to a vehicle, and comprises: Acquire driving map topology relationship information and driving scene target data frame, wherein the driving map topology relationship information includes lane information; Determine a current lane and a pending lane according to the driving scene target data frame and the lane information; Acquire historical frame data and future frame data corresponding to the target scene data frame; Based on the historical frame data, the future frame data, the current lane and the pending lane, lane change recognition is performed on the vehicle.

2. The method according to claim 1, characterized in that The driving scene target data frame includes a driving scene current data frame and a driving scene predecessor data frame, and the step of determining the current lane and the pending lane according to the driving scene target data frame and the lane information includes: According to a preset lane binding algorithm and the lane information, respectively calculating a current projection distance of the vehicle on each lane in a current data frame of the driving scene, and a front-drive projection distance of the vehicle on each lane in a front-drive data frame of the driving scene; Respectively screening out the minimum current projection distance among the current projection distances and the minimum front-drive projection distance among the front-drive projection distances; The lane corresponding to the minimum current projection distance is determined as the current lane, and the lane corresponding to the front-drive projection distance is determined as the pending lane.

3. The method according to claim 1, characterized in that The step of performing lane change recognition on the vehicle based on the historical frame data, the future frame data, the current lane and the pending lane comprises: Obtaining a lane sequence number of the current lane and a lane sequence number of the pending lane; Determining whether the lane sequence number of the current lane is the same as the lane sequence number of the pending lane; If the lane sequence number of the current lane is different from the lane sequence number of the pending lane, determining the current lane sequence and the pending lane sequence according to the lane sequence number of the current lane and the lane sequence number of the pending lane; Calculating the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence based on a preset lane distance threshold and the historical frame data and the future frame data; Whether the vehicle changes lanes is identified according to the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence.

4. The method according to claim 3, characterized in that The step of calculating the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence based on the preset lane distance threshold and the historical frame data and the future frame data comprises: Determining a first identified lane set on the pending lane sequence and a second identified lane set on the current lane sequence according to a preset lane distance threshold; The stability of the vehicle on the pending lane sequence is determined based on the historical frame data and the first identified lane set, and the stability of the vehicle on the current lane sequence is determined based on the future frame data and the second identified lane set.

5. The method according to claim 4, characterized in that The historical frame data includes a plurality of historical frames, the future frame data includes a plurality of future frames, and the steps of determining the stability of the vehicle on the pending lane sequence based on the historical frame data and the first identified lane set, and determining the stability of the vehicle on the current lane sequence based on the future frame data and the second identified lane set include: Identifying, in the historical frame data, the number of times the vehicle is located in the first identified lane set and the number of times the vehicle appears in the historical frame; Identifying, in the future frame data, the number of times the vehicle is located in the second identified lane set and the number of times the vehicle appears in the future frame; Determining the stability of the vehicle on the pending lane sequence according to a ratio of the number of times the vehicle is located in the first identified lane set to the number of times the vehicle appears in the historical frame; The stability of the vehicle on the current lane sequence is determined according to a ratio of the number of times the vehicle is located in the second identified lane set to the number of times the vehicle appears in the future frame.

6. The method according to claim 3, characterized in that The step of identifying whether the vehicle changes lanes according to the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence comprises: Determining whether the stability of the vehicle on the current lane sequence and the stability of the vehicle on the pending lane sequence are both greater than a preset threshold; If the stability of the vehicle in the current lane sequence and the stability of the vehicle in the pending lane sequence are both greater than a preset threshold, it is determined that the vehicle has changed lanes.

7. A lane change recognition device, characterized in that: The device comprises: A first acquisition module is used to acquire driving map topological relationship information and driving scene target data frames, wherein the driving map topological relationship information includes lane information and lane-to-lane topological relationship information; A determination module, which determines a current lane and a pending lane according to the driving scene target data frame and the lane information; A second acquisition module acquires historical frame data and future frame data corresponding to the target scene data frame; The recognition module performs lane change recognition on the vehicle based on the historical frame data, the future frame data, the current lane and the pending lane.

8. A lane change recognition device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the lane change recognition method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the lane change recognition method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the lane change recognition method according to any one of claims 1 to 6 are implemented.

Images