Lane changing data processing method and device, equipment and storage medium

Abstract

The present application relates to the technical field of automatic driving, and discloses a lane changing data processing method, device and equipment and a storage medium, which are used for improving the calculation accuracy of the lane changing success rate of an automatic driving system. The lane changing data processing method comprises the following steps: when receiving automatic driving road test data, performing screening processing on the automatic driving road test data based on each lane changing request to obtain at least one data block; extracting each data block to obtain at least one lane changing time period of each data block; judging whether there is an interactive vehicle in the target lane of the host vehicle in each lane changing time period; if there is an interactive vehicle in the target lane of the host vehicle in the target lane changing time period, determining the lane changing result information of the target lane changing time period, and determining the number of lane changing successes and the number of lane changing failures of each data block according to the lane changing result information of each lane changing time period; and calculating the lane changing success rate of the automatic driving road test data based on the number of lane changing successes and the number of lane changing failures of each data block.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, and particularly relates to a lane changing data processing method and device, equipment and a storage medium. BACKGROUND

[0002] Automatic driving lane changing technology is one of the key technologies for realizing automatic driving, and is a lane changing decision based on the automatic driving vehicle itself and the surrounding environment to determine whether the automatic driving vehicle changes lanes.

[0003] The interaction between the automatic driving vehicle based on the automatic driving system and the obstacles of the target lane when changing lanes is a common scenario. At present, whether the lane changing of the automatic driving vehicle is successful mainly depends on the human judgment of the safety officer in the road test of automatic driving, thereby resulting in low calculation accuracy of the lane changing success rate of the automatic driving system. SUMMARY

[0004] The present application provides a lane changing data processing method, device, equipment and storage medium, which is used for improving the calculation accuracy of the lane changing success rate of the automatic driving system.

[0005] The first aspect of the present application provides a lane changing data processing method, comprising: when receiving automatic driving road test data, performing screening processing on the automatic driving road test data based on each lane changing request to obtain at least one data block, the data block comprising at least one lane changing request; extracting each data block to obtain at least one lane changing time period of each data block; judging whether there is an interactive vehicle in the target lane of the host vehicle in each lane changing time period; if there is an interactive vehicle in the target lane of the host vehicle in the target lane changing time period, determining the lane changing result information of the target lane changing time period, and determining the number of lane changing successes and the number of lane changing failures of each data block according to the lane changing result information of each lane changing time period, the lane changing result information being used to indicate whether the host vehicle successfully changes lanes; and calculating the lane changing success rate of the automatic driving road test data based on the number of lane changing successes and the number of lane changing failures of each data block.

[0006] In an implementation, if there is an interactive vehicle in the target lane of the host vehicle in the target lane-changing time period, the lane-changing result information of the target lane-changing time period is determined, and the number of successful lane changes and the number of failed lane changes of each data block are determined according to the lane-changing result information of each lane-changing time period, including: if there is an interactive vehicle in the target lane of the host vehicle in the target lane-changing time period, the target lane-changing time period is extracted to obtain at least one effective interactive lane-changing frame; the effective interactive lane-changing time length is calculated based on the at least one effective interactive lane-changing frame; if the interactive lane-changing time length is greater than or equal to a first preset threshold and less than or equal to a second preset threshold, the lateral displacement of the host vehicle in the target lane-changing time period is calculated, and the lane state information of the host vehicle is obtained; the lane-changing result information of the target lane-changing time period is determined based on the lateral displacement and the lane state information; if the interactive lane-changing time length is greater than the second preset threshold, it is determined that the host vehicle fails to change lanes, and the lane-changing result information of the target lane-changing time period is obtained; the number of successful lane changes and the number of failed lane changes of each data block are determined according to the lane-changing result information of each lane-changing time period.

[0007] In an implementation, the lane-changing result information of the target lane-changing time period is determined based on the lateral displacement and the lane state information, including: if the lateral displacement is greater than or equal to a preset displacement, and the lane state information indicates that the host vehicle has driven to the target lane, it is determined that the host vehicle has completed lane changing, and the corresponding lane-changing completion time is recorded; when the host vehicle completes lane changing, it is judged whether there is an effective interactive lane-changing frame in a preset time range earlier than the lane-changing completion time; if there is an effective interactive lane-changing frame in the preset time range, it is determined that the host vehicle successfully changes lanes, and the lane-changing result information of the target lane-changing time period is obtained; if the lateral displacement is less than the preset displacement, or the lane state information indicates that the host vehicle has not driven to the target lane, it is determined that the host vehicle fails to change lanes, and the lane-changing result information of the target lane-changing time period is obtained.

[0008] In an implementation, after the lane-changing success rate of the automatic driving road test data is calculated based on the number of successful lane changes and the number of failed lane changes of each data block, the first preset threshold is sequentially valued in a preset numerical range to obtain a plurality of first preset thresholds; the lane-changing success rate of a plurality of automatic driving road test data is calculated based on each first preset threshold in the plurality of first preset thresholds to obtain a plurality of lane-changing success rates of each automatic driving road test data; the lane-changing success rate mean and the lane-changing success rate standard deviation of each automatic driving road test data are calculated based on the plurality of lane-changing success rates of each automatic driving road test data; and the target automatic driving road test data is determined based on the lane-changing success rate mean and the lane-changing success rate standard deviation of each automatic driving road test data.

[0009] In an implementable embodiment, the determining the target automatic driving road test data based on the lane change success rate mean and the lane change success rate standard deviation of each automatic driving road test data comprises: calculating a mean difference based on a lane change success rate mean of first data and a lane change success rate mean of second data, the first data and the second data being any two of the plurality of automatic driving road test data; calculating a target error based on a lane change success rate standard deviation of the first data, a lane change success rate standard deviation of the second data, and a preset error propagation formula; and determining the automatic driving road test data with a larger lane change success rate mean as the target automatic driving road test data if the mean difference is greater than the target error.

[0010] In an implementable embodiment, the screening the automatic driving road test data based on each lane change request to obtain at least one data block comprising at least one lane change request comprises: determining whether there is a lane change request in a preset time period of each data frame in the automatic driving road test data, the preset time period of each data frame taking the time of each data frame as the middle time; determining a target frame as a data frame with a lane change request in the preset time period if there is a lane change request in the preset time period of the data frame; and combining a plurality of target frames with continuous time into one data block to obtain at least one data block of the automatic driving road test data, the data block comprising at least one lane change request.

[0011] In an implementable embodiment, the extracting each data block to obtain at least one lane change time period of each data block comprises: selecting a candidate time period corresponding to each lane change request in each data block to obtain at least one candidate time period of each data block; determining a target candidate time period as the lane change time period of the data block if the number of the at least one candidate time period of the data block is one; determining whether adjacent two of the at least one candidate time period of the data block meet a preset merging condition if the number of the at least one candidate time period of the data block is at least two, the preset merging condition being that the time interval between the adjacent two candidate time periods is less than or equal to a preset time length, the lane change direction of the host vehicle is unchanged, the host vehicle is in a non-continuous lane change driving state, and the host vehicle is in a non-takeover driving state in a former one of the adjacent two candidate time periods; and combining the adjacent two candidate time periods into one lane change time period to obtain the lane change time period of the data block if the adjacent two candidate time periods meet the preset merging condition, until each data block obtains the corresponding at least one lane change time period.

[0012] In an implementation, the judging whether the target lane of the host vehicle in each lane-changing time period has an interactive vehicle comprises: judging whether the target lane of the host vehicle in each data frame has a front vehicle and a rear vehicle, each lane-changing time period comprising a plurality of data frames; determining a data frame having a front vehicle and a rear vehicle as an interactive lane-changing frame; if a proportion of the interactive lane-changing frames in any lane-changing time period is greater than or equal to a preset proportion, determining that the target lane of the host vehicle in the any lane-changing time period has an interactive vehicle; and if the proportion of the interactive lane-changing frames in the any lane-changing time period is less than the preset proportion, determining that the target lane of the host vehicle in the any lane-changing time period does not have an interactive vehicle.

[0013] In a second aspect, the application provides a processing device for lane-changing data, comprising: a screening module configured to, when receiving automatic driving road test data, screen the automatic driving road test data based on each lane-changing request to obtain at least one data block, the data block comprising at least one lane-changing request; an extraction module configured to extract each data block to obtain at least one lane-changing time period of each data block; a judgment module configured to judge whether the target lane of the host vehicle in each lane-changing time period has an interactive vehicle; a first determination module configured to, if the target lane of the host vehicle in a target lane-changing time period has an interactive vehicle, determine lane-changing result information of the target lane-changing time period, and determine a number of successful lane changes and a number of failed lane changes of each data block according to lane-changing result information of each lane-changing time period, the lane-changing result information being used to indicate whether the host vehicle successfully changes lanes; and a first calculation module configured to calculate a lane-changing success rate of the automatic driving road test data based on the number of successful lane changes and the number of failed lane changes of each data block.

[0014] In an implementation, the first determination module comprises: an extraction unit configured to, if the target lane of the host vehicle in a target lane-changing time period has an interactive vehicle, extract the target lane-changing time period to obtain at least one effective interactive lane-changing frame; a first calculation unit configured to calculate based on the at least one effective interactive lane-changing frame to obtain an effective interactive lane-changing duration; a second calculation unit configured to, if the interactive lane-changing duration is greater than or equal to a first preset threshold and less than or equal to a second preset threshold, calculate a lateral displacement of the host vehicle in the target lane-changing time period and obtain lane state information of the host vehicle; a first determination unit configured to determine lane-changing result information of the target lane-changing time period based on the lateral displacement and the lane state information; a second determination unit configured to, if the interactive lane-changing duration is greater than the second preset threshold, determine that the host vehicle fails to change lanes to obtain lane-changing result information of the target lane-changing time period; and a third determination unit configured to determine the number of successful lane changes and the number of failed lane changes of each data block according to lane-changing result information of each lane-changing time period.

[0015] In an implementation, the first determining unit is specifically configured to: if the lateral displacement is greater than or equal to a preset displacement, and the lane state information indicates that the host vehicle drives to the target lane, determine that the host vehicle completes lane changing, and record a lane changing completion time; when the host vehicle completes lane changing, determine whether there is an effective interactive lane changing frame within a preset time range earlier than the lane changing completion time; if there is an effective interactive lane changing frame within the preset time range, determine that the host vehicle successfully changes lanes, and obtain lane changing result information of the target lane changing time period; if the lateral displacement is less than the preset displacement, or the lane state information indicates that the host vehicle does not drive to the target lane, determine that the host vehicle fails to change lanes, and obtain the lane changing result information of the target lane changing time period.

[0016] In an implementation, the processing device of the lane changing data further includes: a value taking module configured to take values of the first preset threshold in a preset value range to obtain a plurality of first preset thresholds; a second calculation module configured to calculate lane changing success rates of a plurality of automatic driving road test data based on each first preset threshold in the plurality of first preset thresholds to obtain a plurality of lane changing success rates of each automatic driving road test data; a third calculation module configured to calculate based on the plurality of lane changing success rates of each automatic driving road test data to obtain a lane changing success rate mean and a lane changing success rate standard deviation of each automatic driving road test data; and a second determining module configured to determine a target automatic driving road test data based on the lane changing success rate mean and the lane changing success rate standard deviation of each automatic driving road test data.

[0017] In an implementation, the second determining module is specifically configured to: calculate based on a lane changing success rate mean of first data and a lane changing success rate mean of second data to obtain a mean difference, the first data and the second data being any two of the plurality of automatic driving road test data; calculate based on a lane changing success rate standard deviation of the first data, a lane changing success rate standard deviation of the second data, and a preset error propagation formula to obtain a target error; and if the mean difference is greater than the target error, determine an automatic driving road test data with a larger lane changing success rate mean as the target automatic driving road test data.

[0018] In an implementation, the screening module is specifically configured to: determine whether there is a lane changing request within a preset time period of each data frame in the automatic driving road test data, the preset time period of each data frame taking a time of each data frame as a middle time; if there is a lane changing request within the preset time period of a data frame, determine a data frame with a lane changing request within a preset time period as a target frame; and combine a plurality of target frames with continuous time into a data block to obtain at least one data block of the automatic driving road test data, the data block including at least one lane changing request.

[0019] In an implementation, the extraction module is specifically configured to: select candidate time periods corresponding to each lane change request in each data block to obtain at least one candidate time period of each data block; if the number of the at least one candidate time period of the data block is one, determine the target candidate time period as the lane change time period of the data block; if the number of the at least one candidate time period of the data block is at least two, determine whether two adjacent candidate time periods in the at least two candidate time periods meet a preset merging condition, the preset merging condition being that a time interval between the two adjacent candidate time periods is less than or equal to a preset time length, a lane change direction of the host vehicle is unchanged, the host vehicle is in a non-continuous lane change driving state, and the host vehicle is in a non-takeover driving state in a former one of the two adjacent candidate time periods; if the two adjacent candidate time periods meet the preset merging condition, combine the two adjacent candidate time periods into one lane change time period to obtain the lane change time period of the data block, until each data block obtains at least one corresponding lane change time period.

[0020] In an implementation, the determination module is specifically configured to: determine whether a target lane of the host vehicle in each data frame has a front vehicle and a rear vehicle, each lane change time period including a plurality of data frames; determine a data frame having the front vehicle and the rear vehicle as an interactive lane change frame; if a proportion of the interactive lane change frames in any lane change time period is greater than or equal to a preset proportion, determine that the target lane of the host vehicle in the any lane change time period has an interactive vehicle; if the proportion of the interactive lane change frames in the any lane change time period is less than the preset proportion, determine that the target lane of the host vehicle in the any lane change time period does not have the interactive vehicle.

[0021] The third aspect of the present application provides a lane change data processing device, comprising: a memory and at least one processor, the memory storing instructions; the at least one processor calling the instructions in the memory to enable the lane change data processing device to perform the lane change data processing method described above.

[0022] The fourth aspect of the present application provides a computer readable storage medium, the computer readable storage medium storing instructions, when the instructions are run on a computer, enabling the computer to perform the lane change data processing method described above.

[0023] In the technical solution provided by this invention, when autonomous driving road test data is received, the autonomous driving road test data is filtered based on each lane change request to obtain at least one data block, and each data block includes at least one lane change request; each data block is extracted to obtain at least one lane change time period for each data block; it is determined whether there is an interactive vehicle in the target lane of the master vehicle in each lane change time period; if there is an interactive vehicle in the target lane of the master vehicle in the target lane change time period, the lane change result information of the target lane change time period is determined, and the number of successful lane changes and the number of failed lane changes for each data block are determined based on the lane change result information of each lane change time period, and the lane change result information is used to indicate whether the master vehicle has successfully changed lanes; the lane change success rate of the autonomous driving road test data is calculated based on the number of successful lane changes and the number of failed lane changes for each data block. In this embodiment of the invention, when autonomous driving road test data is received, the autonomous driving road test data is filtered based on each lane change request to obtain at least one data block. Each data block is extracted to obtain at least one lane change time period. If there is an interacting vehicle in the target lane of the main vehicle during the target lane change time period, the lane change result information of the target lane change time period is determined. Based on the lane change result information of each lane change time period, the number of successful lane changes and the number of failed lane changes for each data block are determined. Based on the number of successful lane changes and the number of failed lane changes for each data block, the lane change success rate of the autonomous driving road test data is calculated. This avoids the problem of low accuracy in calculating the lane change success rate of the autonomous driving system due to human judgment, thereby avoiding the problem of low accuracy in evaluating the lane change interaction capability of the autonomous driving system, improving the accuracy of calculating the lane change success rate of the autonomous driving system, and improving the accuracy of evaluating the lane change interaction capability of the autonomous driving system. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of an embodiment of the lane change data processing method of the present invention;

[0025] Figure 2 This is a schematic diagram of another embodiment of the lane change data processing method in this invention;

[0026] Figure 3 This is a schematic diagram of another embodiment of the lane change data processing method in this invention;

[0027] Figure 4 This is a schematic diagram of an embodiment of the present invention in which there are vehicles behind the target lane of the main vehicle;

[0028] Figure 5 This is a schematic diagram of one embodiment of the lane change data processing device in this invention;

[0029] Figure 6 This is a schematic diagram of another embodiment of the lane change data processing device in this invention;

[0030] Figure 7 An embodiment of a processing device for lane change data in the embodiments of the present application is shown. DETAILED DESCRIPTION

[0031] The present application provides a lane change data processing method, device, equipment and storage medium, which is used to improve the calculation accuracy of the lane change success rate of the automatic driving system.

[0032] The terms "first", "second", "third", "fourth" and the like in the description, claims, and drawings of the present application, and those above (if any) are used to distinguish similar objects and are not necessarily used to describe a particular sequential or chronological order. It should be understood that the data thus used can be interchanged, where appropriate, so that the embodiments described herein can be carried out in sequences other than those illustrated or described herein. Furthermore, the terms "comprise" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, processes, methods, systems, products, or devices that include a series of steps or units not necessarily limited to those clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0033] For ease of understanding, the specific flow of the embodiments of the present application is described below. Please refer to Figure 1 One embodiment of the lane change data processing method in the embodiments of the present application includes:

[0034] 101, when receiving the automatic driving road test data, performing screening processing on the automatic driving road test data based on each lane change request to obtain at least one data block, the data block including at least one lane change request;

[0035] It can be understood that the execution subject of the present application can be a lane change data processing device, and can also be a terminal, which is not limited here. The embodiments of the present application take the terminal as the execution subject for example.

[0036] The automatic driving road test data is the data recorded during the road test based on the host vehicle of the automatic driving system, wherein the automatic driving road test data includes surrounding vehicle information, a plurality of lane change requests, information whether a safety officer takes over the host vehicle, positioning information and lane information of the host vehicle, the positioning information of the host vehicle includes host vehicle position information and host vehicle speed, and the surrounding vehicle information includes surrounding vehicle position information and surrounding vehicle speed of each surrounding vehicle.

[0037] By way of example and not limitation, the triggering condition of the lane change request can be: (1) the global path planning in the automatic driving system requires the host vehicle to turn right at the front, and the lane change request is triggered to control the host vehicle to drive to the right turn lane when the host vehicle is a preset distance from the intersection, (2) when the vehicles in the current lane where the host vehicle is located are driving slowly and the vehicles in the adjacent lane are driving faster, the lane change request is triggered to control the host vehicle to drive to the adjacent lane, (3) when the host vehicle needs to overtake the slow vehicle in front, the lane change request is triggered to control the host vehicle to overtake, and the triggering condition of the lane change request can also be the triggering condition of other scenarios, as long as the host vehicle has a lane change request.

[0038] 102, extracting each data block to obtain at least one lane change time period of each data block;

[0039] In a possible implementation, each data block is extracted to obtain at least one lane change time period of each data block, specifically including: (1) the terminal selects candidate time periods corresponding to each lane change request in each data block to obtain at least one candidate time period of each data block; (2) the terminal determines at least one lane change time period of each data block based on at least one candidate time period of each data block.

[0040] The starting time point in the candidate time period is the time point corresponding to the data frame in which the lane change request changes from none to one, for example, the first data frame has no lane change request, and the second data frame has a lane change request, so the time point corresponding to the second data frame is determined as the starting time point. In the presence of the starting time point, the time point corresponding to the data frame in which the lane change request changes from one to none, the lane change direction changes to the opposite direction, or the automatic driving state changes to the manual driving state is determined as the end time point in the candidate time period, for example, the first data frame has a lane change request, and the second data frame has no lane change request, so the time point corresponding to the second data frame is determined as the end time point. The lane change direction changes to the opposite direction, which can be understood as the host vehicle overtaking the front vehicle and then changing lanes back to the original lane, which is regarded as two lane changes.

[0041] The lane change request is a Boolean (BOOL) data, and the value of the lane change request changes from true to false, that is, from one to none.

[0042] The automatic driving state is also a Boolean (BOOL) data, and the value of the automatic driving state changes from true to false, that is, from the automatic driving state to the manual driving state.

[0043] The automatic driving state changes to the manual driving state, which can be understood as the safety officer in the host vehicle determining that the current time cannot change lanes, and thus manually taking over the host vehicle to prevent the host vehicle from actively changing lanes.

[0044] 103, determining whether there is an interactive vehicle in the target lane of the host vehicle in each lane change time period;

[0045] In an implementation, determining whether the target lane of the host vehicle in each lane-changing time period has an interactive vehicle includes: (1) determining, by the terminal, whether the target lane of the host vehicle in each data frame has a front vehicle and a rear vehicle, each lane-changing time period including a plurality of data frames; (2) determining, by the terminal, a data frame having a front vehicle and a rear vehicle as an interactive lane-changing frame; and (3) determining, by the terminal, whether the target lane of the host vehicle in each lane-changing time period has an interactive vehicle based on the interactive lane-changing frame of each lane-changing time period.

[0046] The target lane is calculated by the automatic driving system during automatic driving of the host vehicle, and lane information of the target lane is recorded in the automatic driving road test data.

[0047] 104. If the target lane of the host vehicle in the target lane-changing time period has an interactive vehicle, determining lane-changing result information of the target lane-changing time period, and determining the number of successful lane changes and the number of failed lane changes of each data block according to the lane-changing result information of each lane-changing time period, the lane-changing result information being used to indicate whether the host vehicle successfully changes lanes;

[0048] In an implementation, if the target lane of the host vehicle in the target lane-changing time period has an interactive vehicle, determining lane-changing result information of the target lane-changing time period includes: (1) if the target lane of the host vehicle in the target lane-changing time period has an interactive vehicle, extracting, by the terminal, the target lane-changing time period to obtain at least one valid interactive lane-changing frame; (2) calculating, by the terminal, an effective interactive lane-changing duration based on the at least one valid interactive lane-changing frame; and (3) determining, by the terminal, the lane-changing result information of the target lane-changing time period based on the interactive lane-changing duration.

[0049] For example, the 5th frame to the 15th frame are all valid interactive lane-changing frames, and if the interval duration between adjacent two frames is 0.1 second, the interactive lane-changing duration is (15-5)*0.1=1 second.

[0050] 105. Calculating a lane-changing success rate of the automatic driving road test data based on the number of successful lane changes and the number of failed lane changes of each data block.

[0051] In an implementation, calculating a lane-changing success rate of the automatic driving road test data based on the number of successful lane changes and the number of failed lane changes of each data block includes: (1) adding, by the terminal, the number of successful lane changes and the number of failed lane changes of each data block to obtain a total number of lane changes of the automatic driving road test data; (2) adding, by the terminal, the number of successful lane changes of each data block to obtain a total number of successful lane changes of the automatic driving road test data; and (3) calculating, by the terminal, the lane-changing success rate of the automatic driving road test data based on the total number of successful lane changes and the total number of lane changes.

[0052] It can be understood that the number of lane changing successes can be 0, and the number of lane changing failures can also be 0.

[0053] The total number of lane changes and the total number of successful lane changes of the automatic driving road test data are counted through each data block, which can reduce the operation time length of the terminal, improve the processing efficiency of the lane change data, and improve the calculation efficiency of the lane change success rate.

[0054] In the embodiment of the application, when the automatic driving road test data is received, the automatic driving road test data is filtered based on each lane change request to obtain at least one data block, each data block is extracted to obtain at least one lane change time period of each data block, if there is an interactive vehicle in the target lane of the target vehicle in the target lane change time period, the lane change result information of the target lane change time period is determined, and the number of successful lane changes and the number of failed lane changes of each data block are determined according to the lane change result information of each lane change time period, the lane change success rate of the automatic driving road test data is calculated based on the number of successful lane changes and the number of failed lane changes of each data block, thereby avoiding the problem of low calculation accuracy of the lane change success rate of the automatic driving system due to human judgment, avoiding the problem of low evaluation accuracy of the lane change interaction ability of the automatic driving system, improving the calculation accuracy of the lane change success rate of the automatic driving system, and improving the evaluation accuracy of the lane change interaction ability of the automatic driving system.

[0055] Please refer to Figure 2 Another embodiment of the lane change data processing method in the embodiment of the application includes:

[0056] 201、When the automatic driving road test data is received, it is determined whether there is a lane change request in the preset time period of each data frame in the automatic driving road test data, and the preset time period of each data frame takes the time of each data frame as the middle time;

[0057] As an example but not limited, the preset time period can be 4 seconds, and can also be 5 seconds. The specific preset time period can be set according to the actual application scene.

[0058] Because the lane change decision of different data frames may change, the same lane change request may correspond to multiple continuous data frames, or may correspond to multiple interval data frames. It can be understood that the vehicle speed of the adjacent lane is faster than the speed of the host vehicle, but the degree of fastness is just floating around the threshold value of whether to trigger the lane change request, so that some data frames have lane change requests and some data frames do not have lane change requests, thereby causing the same lane change request to correspond to multiple interval data frames. Therefore, it is only necessary to satisfy that there is a lane change request in the preset time period of the data frame.

[0059] 202、if there is a lane change request in the preset time period of the data frame, the data frame in which there is a lane change request in the preset time period is determined as the target frame;

[0060] It can be understood that there is a lane change request in the preset time period of the data frame, and the data frame may have a lane change request or may not have a lane change request, so that the target frame is set as the data frame with a lane change request.

[0061] 203、a plurality of target frames in time sequence are combined into one data block, and at least one data block of the automatic driving road test data is obtained, and the data block includes at least one lane change request;

[0062] For example, if there is at least one data frame with a lane change request in the preset time period of a certain data frame, the certain data frame is recorded as 1 (i.e. the target frame). If there is no lane change request in the preset time period of a certain data frame, the certain data frame is recorded as 0, such as the sequence: 01111100000011111110. It can be understood that the data frame with a lane change request and all data frames in the preset time period of the data frame with a lane change request are marked as 1. Therefore, the sequence 01111100000011111110 includes two consecutive 1s, i.e. corresponding to two data blocks.

[0063] 204、extracting each data block to obtain at least one lane change time period of each data block;

[0064] The execution steps of step 204 are the same as those of step 102, and will not be repeated here.

[0065] 205、judging whether there is an interactive vehicle in the target lane of the host vehicle in each lane change time period;

[0066] The execution steps of step 205 are the same as those of step 103, and will not be repeated here.

[0067] 206、if there is an interactive vehicle in the target lane of the host vehicle in the target lane change time period, extracting the target lane change time period to obtain at least one effective interactive lane change frame;

[0068] In a feasible implementation, (1) the terminal judges whether each data frame in the target lane change time period meets a preset rule, and the preset rule is that there is an interactive vehicle in the target lane of the host vehicle in the data frame, and at least one of the speed of the host vehicle and the speed of the interactive vehicle is greater than zero; (2) if any data frame meets the preset rule, the terminal determines any data frame as an effective interactive lane change frame, and at least one effective interactive lane change frame is obtained.

[0069] At least one of the speed of the host vehicle and the speed of the interactive vehicle is greater than zero, which can be understood as excluding the scenario in which the host vehicle and the interactive vehicle are temporarily parked due to road prohibition or other factors, and the host vehicle cannot perform lane changing.

[0070] By determining the effective interactive lane changing frame, the misjudgment of determining that lane changing fails in some special scenarios in which lane changing cannot be performed is excluded, and the calculation accuracy of the lane changing success rate is improved.

[0071] 207、Based on at least one effective interactive lane changing frame, an effective interactive lane changing duration is calculated;

[0072] For example, the 5th frame to the 15th frame are all effective interactive lane changing frames, and if the interval duration between adjacent two frames is 0.1 seconds, then the interactive lane changing duration is: (15-5)*0.1=1 second.

[0073] 208、If the interactive lane changing duration is greater than or equal to a first preset threshold and less than or equal to a second preset threshold, then the lateral displacement of the host vehicle in the target lane changing time period is calculated, and the lane state information of the host vehicle is obtained;

[0074] By way of example but not limitation, the first preset threshold can be 2 seconds, and can also be 3 seconds, and the second preset threshold can be 20 seconds, and can also be 30 seconds. Here, no limitation is made, as long as the first preset threshold is less than the second preset threshold.

[0075] It should be noted that if the interactive lane changing duration is less than the first preset threshold, the target lane changing time period is removed and is not counted as lane changing success or lane changing failure.

[0076] In a possible implementation, the lateral displacement of the host vehicle in the target lane changing time period is calculated, specifically including: (1) the terminal obtains the center point of the host vehicle in the target lane changing time period; (2) the terminal performs displacement calculation based on the center point of the host vehicle and a preset direction to obtain the lateral displacement of the host vehicle in the target lane changing time period, and the preset direction is the direction in which the center point is perpendicular to the lane line of the target lane.

[0077] The center point of the host vehicle is used to indicate the geometric center of the host vehicle.

[0078] By calculating the lateral displacement of the host vehicle, it can be determined whether the host vehicle is performing lane changing, so as to further judge whether the host vehicle succeeds in lane changing, and the calculation accuracy of the lane changing success rate is improved.

[0079] 209、Based on the lateral displacement and the lane state information, lane changing result information of the target lane changing time period is determined;

[0080] Specifically, (1) if the lateral displacement is greater than or equal to the preset displacement, and the lane state information indicates that the host vehicle drives to the target lane, it is determined that the lane changing of the host vehicle is completed, and the corresponding lane changing completion time is recorded; (2) when the lane changing of the host vehicle is completed, it is judged whether there is an effective interactive lane changing frame within a preset time range earlier than the lane changing completion time; (3) if there is an effective interactive lane changing frame within the preset time range, it is determined that the lane changing of the host vehicle is successful, and the lane changing result information of the target lane changing time period is obtained; (4) if the lateral displacement is less than the preset displacement, or the lane state information indicates that the host vehicle does not drive to the target lane, it is determined that the lane changing of the host vehicle fails, and the lane changing result information of the target lane changing time period is obtained.

[0081] Among them, the lane changing result information is used to indicate whether the host vehicle successfully changes lanes.

[0082] By way of example, and not limitation, the preset displacement can be 1.5 meters, and can also be 2 meters. The specific preset displacement can be set according to the actual application scenario. The preset time range can be 2 seconds, and can also be 3 seconds. The specific preset time range can be set according to the actual application scenario.

[0083] It should be noted that if there is no effective interactive lane changing frame within the preset time range, the target lane changing time period is removed and is not counted as lane changing success or failure.

[0084] By determining the completion of the lane changing of the host vehicle through the lateral displacement and the lane state information, and excluding the scenario of being unable to change lanes according to the effective interactive lane changing frame, the calculation accuracy of the lane changing success rate can be improved.

[0085] 210, if the interactive lane changing duration is greater than the second preset threshold, it is determined that the host vehicle fails to change lanes, and the lane changing result information of the target lane changing time period is obtained;

[0086] It can be understood that when the interactive lane changing duration is greater than the second preset threshold, whether the host vehicle completes the lane changing or not, it is determined that the host vehicle fails to change lanes.

[0087] 211, according to the lane changing result information of each lane changing time period, the number of successful lane changes and the number of failed lane changes of each data block are determined;

[0088] Since each data block includes at least one lane changing time period, and the lane changing result information indicates whether the host vehicle successfully changes lanes, the number of successful lane changes and the number of failed lane changes of each data block can be determined according to the lane changing result information of each lane changing time period. It can be understood that the number of successful lane changes can be 0, and the number of failed lane changes can also be 0.

[0089] 212, based on the number of successful lane changes and the number of failed lane changes of each data block, the lane changing success rate of the automatic driving road test data is calculated.

[0090] The execution step of step 212 is the same as that of step 105, and thus is not described herein again.

[0091] In an implementation, after the lane changing success rate of the automatic driving road test data is calculated based on the number of lane changing successes and the number of lane changing failures of each data block, the method further includes: (1) the terminal sequentially takes values of the first preset threshold within a preset value range to obtain a plurality of first preset thresholds; (2) the terminal calculates the lane changing success rate of the plurality of automatic driving road test data based on each first preset threshold in the plurality of first preset thresholds to obtain a plurality of lane changing success rates of each automatic driving road test data; (3) the terminal calculates based on the plurality of lane changing success rates of each automatic driving road test data to obtain a lane changing success rate mean and a lane changing success rate standard deviation of each automatic driving road test data; and (4) the terminal determines the target automatic driving road test data based on the lane changing success rate mean and the lane changing success rate standard deviation of each automatic driving road test data.

[0092] It should be noted that each automatic driving road test data corresponds to an automatic driving system version.

[0093] The preset value range can be set according to actual application scenarios.

[0094] For example, the first preset threshold is 2 seconds, and the preset value range is 1 second to 6 seconds, so 6 first preset thresholds can be obtained, and the lane changing success rate of each automatic driving road test data is calculated by the 6 first preset thresholds to obtain 6 lane changing success rates of each automatic driving road test data.

[0095] The lane changing success rate mean is used to indicate the average value of the plurality of lane changing success rates.

[0096] The formula of the lane changing success rate standard deviation is: Wherein, S is used to represent the lane changing success rate standard deviation, n is used to represent the number of lane changing success rates, X i is used to represent the lane changing success rate, is used to represent the lane changing success rate mean.

[0097] The first preset threshold can also be other preset data, such as a second preset threshold, a preset displacement, a preset time range, a preset time period, and a preset ratio.

[0098] It can be understood that a plurality of preset data can be sequentially valued at the same time, for example, the first preset threshold is 2 seconds, the preset value range of the first preset threshold is 1 second to 9 seconds, the second preset threshold is 10 seconds, and the preset value range of the second preset threshold is 1 second to 25 seconds, so 9*25=225 lane changing success rates can be obtained.

[0099] In an implementable embodiment, the target automatic driving road test data is determined based on the lane change success rate mean and the lane change success rate standard deviation of each automatic driving road test data, specifically comprising: (1) the terminal calculates the mean difference based on the lane change success rate mean of the first data and the lane change success rate mean of the second data, the first data and the second data being any two of the plurality of automatic driving road test data; (2) the terminal calculates the target error based on the lane change success rate standard deviation of the first data, the lane change success rate standard deviation of the second data and a preset error propagation formula; and (3) if the mean difference is greater than the target error, the terminal determines the automatic driving road test data with the larger lane change success rate mean as the target automatic driving road test data.

[0100] The preset error propagation formula is: Wherein, d is used to represent the target error, a is used to represent one lane change success rate standard deviation, and b is used to represent another lane change success rate standard deviation.

[0101] For example, the lane change success rate mean of the first data is 0.42, and the lane change success rate mean of the second data is 0.62, so the mean difference is 0.2, the lane change success rate standard deviation of the first data is 0.05, and the lane change success rate standard deviation of the second data is 0.04, i.e. the target error is 0.064, so the mean difference 0.2 is greater than the target error 0.064, and the terminal determines the second data as the target automatic driving road test data.

[0102] By determining the target automatic driving road test data based on the lane change success rate mean and the lane change success rate standard deviation, it can be determined whether the difference between the lane change success rates of different versions of the automatic driving system is greater than the error, and the lane change interaction capabilities of different versions of the automatic driving system are compared, thereby improving the evaluation accuracy of the lane change interaction capability of the automatic driving system.

[0103] In the embodiment of the application, when the automatic driving road test data is received, the automatic driving road test data is filtered based on each lane change request to obtain at least one data block, each data block is extracted to obtain at least one lane change time period of each data block, if there is an interactive vehicle in the target lane of the target vehicle in the target lane change time period, the lane change result information of the target lane change time period is determined, and the number of successful lane changes and the number of failed lane changes of each data block are determined according to the lane change result information of each lane change time period, the lane change success rate of the automatic driving road test data is calculated based on the number of successful lane changes and the number of failed lane changes of each data block, thereby avoiding the problem of low calculation accuracy of the lane change success rate of the automatic driving system due to human judgment, thereby avoiding the problem of low evaluation accuracy of the lane change interaction capability of the automatic driving system, improving the calculation accuracy of the lane change success rate of the automatic driving system, and improving the evaluation accuracy of the lane change interaction capability of the automatic driving system.

[0104] Referring to Figure 3 Another embodiment of the method for processing lane-changing data in the embodiments of the present application comprises:

[0105] 301. When receiving the automatic driving road test data, the automatic driving road test data is filtered based on each lane-changing request to obtain at least one data block, and the data block comprises at least one lane-changing request;

[0106] The execution steps of step 301 are the same as those of step 101, which will not be repeated here.

[0107] 302. In each data block, a candidate time period corresponding to each lane-changing request is selected to obtain at least one candidate time period of each data block;

[0108] The starting time of the candidate time period is the time corresponding to the data frame in which the lane-changing request changes from no to yes, for example, the first data frame has no lane-changing request, and the second data frame has a lane-changing request, so the time corresponding to the second data frame is determined as the starting time. In the case of the starting time, the time corresponding to the data frame in which the lane-changing request changes from yes to no, the lane-changing direction changes to the opposite direction, or the automatic driving state changes to the manual driving state is determined as the ending time of the candidate time period.

[0109] 303. If the number of at least one candidate time period of the data block is one, the target candidate time period is determined as the lane-changing time period of the data block;

[0110] It can be understood that when the data block has only one candidate time period, the candidate time period corresponds to one lane-changing request.

[0111] 304. If the number of at least one candidate time period of the data block is at least two, it is determined whether the adjacent two candidate time periods in the at least two candidate time periods meet a preset merging condition, the preset merging condition is that the time interval between the adjacent two candidate time periods is less than or equal to a preset time length, the lane-changing direction of the host vehicle does not change, the host vehicle is in a non-continuous lane-changing driving state, and the host vehicle is in a non-takeover driving state in the former one of the adjacent two candidate time periods;

[0112] Since the lane-changing request may be no->yes->no->yes->...->no, which repeatedly jumps, that is, the host vehicle hesitates whether to change lanes, therefore, it is necessary to merge the adjacent two candidate time periods that meet the preset merging condition.

[0113] By way of example and not limitation, the preset time length can be 2 seconds, and can also be 3 seconds, which is not limited here.

[0114] The lane-changing direction of the host vehicle is unchanged, which means that the host vehicle has a lane-changing request in the same direction in both candidate time periods, but the host vehicle fails to change lanes in the first candidate time period and still stays in the original lane, and the host vehicle continues to change lanes in the second candidate time period.

[0115] The host vehicle is in a non-continuous lane-changing driving state, which means that the host vehicle does not change lanes continuously.

[0116] 305、If the two adjacent candidate time periods meet the preset merging condition, the two adjacent candidate time periods are combined into one lane-changing time period, and the lane-changing time period of the data block is obtained, until each data block obtains at least one corresponding lane-changing time period;

[0117] By merging the two candidate time periods that meet the preset merging condition, it can be prevented that the hesitant lane-changing scene belonging to the same lane-changing time period is divided into multiple lane-changing time periods.

[0118] 306、Judge whether the target lane of the host vehicle in each data frame has a front vehicle and a rear vehicle, and each lane-changing time period includes multiple data frames;

[0119] In a possible implementation, the judgment of whether the target lane of the host vehicle in each data frame has a front vehicle and a rear vehicle specifically includes: (1) the terminal obtains the center point of the host vehicle and the center point of each obstacle in each data frame, and each obstacle is in the target lane of the host vehicle; (2) the terminal calculates the first distance and the second distance between the host vehicle and each obstacle in any data frame based on the center point of the host vehicle and the center point of each obstacle in the data frame, the first distance is used to indicate the distance perpendicular to the target lane direction, and the second distance is used to indicate the distance along the target lane direction; (3) if the first distance is less than or equal to the first preset distance, the second distance is less than or equal to the second preset distance, and the obstacle is located in front of the host vehicle in any data frame, the terminal determines that the target lane of the host vehicle in the data frame has a front vehicle; (4) if the first distance is less than or equal to the first preset distance, the second distance is less than or equal to the second preset distance, and the obstacle is located behind the host vehicle in any data frame, the terminal determines that the target lane of the host vehicle in the data frame has a rear vehicle; (5) if the first distance is greater than the first preset distance or the second distance is greater than the second preset distance in any data frame, the terminal determines that the target lane of the host vehicle in the data frame does not have a front vehicle and a rear vehicle.

[0120] The center point of the host vehicle is used to indicate the geometric center of the host vehicle, and the center point of the obstacle is used to indicate the geometric center of the obstacle.

[0121] By way of example, and not limitation, the first preset distance can be 1.5 meters, and can also be 2 meters, and the second preset distance can be 10 meters, and can also be 15 meters.

[0122] like Figure 4 As shown, the right rectangle represents the main vehicle, which changes lanes to the left. The left rectangle represents the vehicles behind in the target lane. L1 is the first distance, and L2 is the second distance.

[0123] 307. Data frames containing both vehicles in front and behind are identified as interactive lane change frames;

[0124] It is understandable that the lane change time period includes at least one interactive lane change frame.

[0125] 308. If the proportion of interactive lane change frames in any lane change time period is greater than or equal to a preset ratio, then it is determined that there are interactive vehicles in the target lane of the main vehicle in any lane change time period.

[0126] The percentage of interactive lane change frames is used to indicate the ratio between the total number of interactive lane change frames and the total number of all data frames in the corresponding lane change time period.

[0127] As an example, and not a limitation, the preset ratio can be 60% or 70%, as long as the preset ratio is greater than 50%.

[0128] 309. If the proportion of interactive lane change frames in any lane change time period is less than a preset ratio, then it is determined that there are no interactive vehicles in the target lane of the main vehicle in any lane change time period.

[0129] For example, if the preset ratio is 60%, and the proportion of interactive lane change frames in any lane change time period is less than 60%, then the terminal determines that there are no interactive vehicles in the target lane of the main vehicle in any lane change time period.

[0130] 310. If there are interacting vehicles in the target lane of the main vehicle during the target lane change time period, determine the lane change result information of the target lane change time period, and determine the number of successful lane changes and the number of failed lane changes for each data block based on the lane change result information of each lane change time period. The lane change result information is used to indicate whether the main vehicle has successfully changed lanes.

[0131] The execution steps of step 310 are the same as those of step 104, and will not be repeated here.

[0132] 311. Calculate the lane change success rate of autonomous driving road test data based on the number of successful lane changes and the number of failed lane changes for each data block.

[0133] The execution steps of step 311 are the same as those of step 105, and will not be repeated here.

[0134] When the automatic driving road test data is received, the automatic driving road test data is filtered based on each lane change request to obtain at least one data block, each data block is extracted to obtain at least one lane change time period of each data block, if there is an interactive vehicle in the target lane of the host vehicle in the target lane change time period, the lane change result information of the target lane change time period is determined, and the lane change success times and the lane change failure times of each data block are determined according to the lane change result information of each lane change time period, the lane change success rate of the automatic driving road test data is calculated based on the lane change success times and the lane change failure times of each data block, thereby avoiding the problem that the calculation accuracy of the lane change success rate of the automatic driving system is low due to human judgment, avoiding the problem that the evaluation accuracy of the lane change interaction capability of the automatic driving system is low, improving the calculation accuracy of the lane change success rate of the automatic driving system, and improving the evaluation accuracy of the lane change interaction capability of the automatic driving system.

[0135] The processing method of the lane change data in the embodiment of the application is described above, and the processing device of the lane change data in the embodiment of the application is described below, please refer to Figure 5 The processing device of the lane change data in the embodiment of the application includes one embodiment:

[0136] The filtering module 501 is configured to filter the automatic driving road test data based on each lane change request when the automatic driving road test data is received, and obtain at least one data block, the data block including at least one lane change request;

[0137] The extraction module 502 is configured to extract each data block to obtain at least one lane change time period of each data block;

[0138] The judgment module 503 is configured to judge whether there is an interactive vehicle in the target lane of the host vehicle in each lane change time period;

[0139] The first determination module 504 is configured to determine the lane change result information of the target lane change time period if there is an interactive vehicle in the target lane of the host vehicle in the target lane change time period, and determine the lane change success times and the lane change failure times of each data block according to the lane change result information of each lane change time period, the lane change result information being used to indicate whether the host vehicle succeeds in lane change;

[0140] The first calculation module 505 is configured to calculate the lane change success rate of the automatic driving road test data based on the lane change success times and the lane change failure times of each data block.

[0141] When the automatic driving road test data is received, the automatic driving road test data is filtered based on each lane change request to obtain at least one data block, each data block is extracted to obtain at least one lane change time period of each data block, if there is an interactive vehicle in the target lane of the host vehicle in the target lane change time period, the lane change result information of the target lane change time period is determined, and the number of successful lane changes and the number of failed lane changes of each data block are determined according to the lane change result information of each lane change time period, the lane change success rate of the automatic driving road test data is calculated based on the number of successful lane changes and the number of failed lane changes of each data block, thereby avoiding the problem that the calculation accuracy of the lane change success rate of the automatic driving system is low due to human judgment, avoiding the problem that the evaluation accuracy of the lane change interaction capability of the automatic driving system is low, improving the calculation accuracy of the lane change success rate of the automatic driving system, and improving the evaluation accuracy of the lane change interaction capability of the automatic driving system.

[0142] Please refer to Figure 6 Another embodiment of the lane change data processing device in the embodiment of the application includes:

[0143] The screening module 501 is configured to, when the automatic driving road test data is received, filter the automatic driving road test data based on each lane change request to obtain at least one data block, and the data block includes at least one lane change request.

[0144] The extraction module 502 is configured to extract each data block to obtain at least one lane change time period of each data block.

[0145] The judgment module 503 is configured to judge whether there is an interactive vehicle in the target lane of the host vehicle in each lane change time period.

[0146] The first determination module 504 is configured to, if there is an interactive vehicle in the target lane of the host vehicle in the target lane change time period, determine the lane change result information of the target lane change time period, and determine the number of successful lane changes and the number of failed lane changes of each data block according to the lane change result information of each lane change time period, and the lane change result information is used to indicate whether the host vehicle successfully changes lanes.

[0147] The first calculation module 505 is configured to calculate the lane change success rate of the automatic driving road test data based on the number of successful lane changes and the number of failed lane changes of each data block.

[0148] Optionally, the first determination module 504 includes:

[0149] The extraction unit 5041 is configured to, if there is an interactive vehicle in the target lane of the host vehicle in the target lane change time period, extract the target lane change time period to obtain at least one effective interactive lane change frame.

[0150] The first calculation unit 5042 is configured to perform calculation based on the at least one valid interactive lane-changing frame to obtain an effective interactive lane-changing duration.

[0151] The second calculation unit 5043 is configured to, if the interactive lane-changing duration is greater than or equal to the first preset threshold and less than or equal to the second preset threshold, calculate a lateral displacement of the host vehicle in the target lane-changing time period and obtain lane state information of the host vehicle.

[0152] The first determination unit 5044 is configured to determine lane-changing result information of the target lane-changing time period based on the lateral displacement and the lane state information.

[0153] The second determination unit 5045 is configured to, if the interactive lane-changing duration is greater than the second preset threshold, determine that the host vehicle fails to change lanes and obtain the lane-changing result information of the target lane-changing time period.

[0154] The third determination unit 5046 is configured to determine the number of successful lane changes and the number of failed lane changes of each data block according to the lane-changing result information of each lane-changing time period.

[0155] Optionally, the first determination unit 5044 is specifically configured to:

[0156] If the lateral displacement is greater than or equal to a preset displacement and the lane state information indicates that the host vehicle travels to the target lane, it is determined that the host vehicle completes lane changing, and a lane-changing completion time point is recorded.

[0157] When the host vehicle completes lane changing, it is determined whether there is a valid interactive lane-changing frame in a preset time range earlier than the lane-changing completion time point.

[0158] If there is a valid interactive lane-changing frame in the preset time range, it is determined that the host vehicle successfully changes lanes, and the lane-changing result information of the target lane-changing time period is obtained.

[0159] If the lateral displacement is less than the preset displacement or the lane state information indicates that the host vehicle does not travel to the target lane, it is determined that the host vehicle fails to change lanes, and the lane-changing result information of the target lane-changing time period is obtained.

[0160] Optionally, the lane-changing data processing apparatus further includes:

[0161] The value taking module 506 is configured to take values of the first preset threshold one by one in a preset value range to obtain a plurality of first preset thresholds.

[0162] The second calculation module 507 is configured to perform lane-changing success rate calculation on a plurality of automatic driving road test data based on each first preset threshold in the plurality of first preset thresholds to obtain a plurality of lane-changing success rates of the automatic driving road test data.

[0163] The third calculation module 508 is configured to calculate based on the multiple lane-changing success rates of each automatic driving road test data to obtain a lane-changing success rate mean and a lane-changing success rate standard deviation of each automatic driving road test data.

[0164] The second determination module 509 is configured to determine the target automatic driving road test data based on the lane-changing success rate mean and the lane-changing success rate standard deviation of each automatic driving road test data.

[0165] Optionally, the second determination module 509 is specifically configured to:

[0166] calculate based on the lane-changing success rate mean of the first data and the lane-changing success rate mean of the second data to obtain a mean difference, the first data and the second data being any two of the multiple automatic driving road test data;

[0167] calculate based on the lane-changing success rate standard deviation of the first data, the lane-changing success rate standard deviation of the second data and a preset error propagation formula to obtain a target error;

[0168] If the mean difference is greater than the target error, the automatic driving road test data with the greater lane-changing success rate mean is determined as the target automatic driving road test data.

[0169] Optionally, the screening module 501 is specifically configured to:

[0170] determine whether there is a lane-changing request in a preset time period of each data frame in the automatic driving road test data, the preset time period of each data frame taking a time point of each data frame as a middle time point;

[0171] If there is a lane-changing request in the preset time period of the data frame, a data frame with a lane-changing request in the preset time period is determined as a target frame;

[0172] a plurality of target frames with continuous time points are combined into one data block to obtain at least one data block of the automatic driving road test data, and the data block includes at least one lane-changing request.

[0173] Optionally, the extraction module 502 is specifically configured to:

[0174] select a candidate time period corresponding to each lane-changing request in each data block to obtain at least one candidate time period of each data block;

[0175] If the number of the at least one candidate time period of the data block is one, the target candidate time period is determined as the lane-changing time period of the data block;

[0176] If the number of the at least one candidate time period of the data block is at least two, it is judged whether two adjacent candidate time periods in the at least two candidate time periods meet a preset merging condition, the preset merging condition being that a time interval between the two adjacent candidate time periods is less than or equal to a preset time length, a lane changing direction of the host vehicle is unchanged, the host vehicle is in a driving state of non-continuous lane changing, and the host vehicle is in a non-takeover driving state in a former one of the two adjacent candidate time periods;

[0177] If the two adjacent candidate time periods meet the preset merging condition, the two adjacent candidate time periods are combined as one lane changing time period, and a lane changing time period of the data block is obtained, until each data block obtains corresponding at least one lane changing time period.

[0178] Optionally, the judging module 503 is specifically configured to:

[0179] It is judged whether a front vehicle and a rear vehicle exist in a target lane of the host vehicle in each data frame, and each lane changing time period includes a plurality of data frames.

[0180] Data frames in which the front vehicle and the rear vehicle exist are determined as interactive lane changing frames.

[0181] If a proportion of the interactive lane changing frames in any lane changing time period is greater than or equal to a preset proportion, it is determined that the target lane of the host vehicle in the any lane changing time period has an interactive vehicle.

[0182] If the proportion of the interactive lane changing frames in any lane changing time period is less than the preset proportion, it is determined that the target lane of the host vehicle in the any lane changing time period has no interactive vehicle.

[0183] In the embodiment of the application, when the automatic driving road test data is received, the automatic driving road test data is filtered based on each lane changing request to obtain at least one data block, each data block is extracted to obtain at least one lane changing time period of each data block, if an interactive vehicle exists in a target lane of the host vehicle in a target lane changing time period, a lane changing result information of the target lane changing time period is determined, and the number of successful lane changes and the number of failed lane changes of each data block are determined according to the lane changing result information of each lane changing time period, the lane changing success rate of the automatic driving road test data is calculated based on the number of successful lane changes and the number of failed lane changes of each data block, the problem that the calculation accuracy of the lane changing success rate of the automatic driving system is low due to human judgment is avoided, the problem that the evaluation accuracy of the lane changing interactive ability of the automatic driving system is low is avoided, the calculation accuracy of the lane changing success rate of the automatic driving system is improved, and the evaluation accuracy of the lane changing interactive ability of the automatic driving system is improved.

[0184] The above Figure 5 and Figure 6The processing device of the lane change data in the embodiments of the present application is described in detail from the perspective of the modular functional entity, and the processing device of the lane change data in the embodiments of the present application is described in detail from the perspective of hardware processing.

[0185] Figure 7 Fig. 7 is a schematic diagram of the structure of the processing device of the lane change data provided by the embodiments of the present application. The processing device 700 of the lane change data can be quite different due to different configurations or performances, and can include one or more processors (central processing units, CPU) 710 (for example, one or more processors) and a memory 720, and one or more storage media 730 (for example, one or more mass storage devices) storing application programs 733 or data 732. The memory 720 and the storage media 730 can be temporary storage or persistent storage. The programs stored in the storage media 730 can include one or more modules (not shown in the figure), and each module can include a series of instruction operations in the processing device 700 of the lane change data. Furthermore, the processor 710 can be configured to communicate with the storage media 730 and execute a series of instruction operations in the storage media 730 on the processing device 700 of the lane change data.

[0186] The processing device 700 of the lane change data can also include one or more power supplies 740, one or more wired or wireless network interfaces 750, one or more input and output interfaces 760, and / or one or more operating systems 731, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art can understand that the processing device of the lane change data can include more or fewer components than those shown in the figure, or some components can be combined, or different components can be arranged. Figure 7 The structure of the processing device of the lane change data shown in the figure does not constitute a limitation on the processing device of the lane change data, and can include more or fewer components than those shown in the figure, or some components can be combined, or different components can be arranged.

[0187] The present application also provides a processing device of lane change data, the computer device includes a memory and a processor, the memory has computer readable instructions stored therein, and the computer readable instructions are executed by the processor to make the processor execute the steps of the lane change data processing method in each of the above embodiments.

[0188] The present application also provides a computer readable storage medium, which can be a non-volatile computer readable storage medium or a volatile computer readable storage medium, and the computer readable storage medium has instructions stored therein, and the instructions make the computer execute the steps of the lane change data processing method when the instructions are run on the computer.

[0189] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0190] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0191] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing lane change data, characterized in that, The method for processing the lane change data includes: When autonomous driving road test data is received, the autonomous driving road test data is filtered based on each lane change request to obtain at least one data block, and the data block includes at least one lane change request. Extract each data block to obtain at least one lane change time period for each data block; Determine whether there are interacting vehicles in the target lane of the main vehicle during each lane change time period; If there are interacting vehicles in the target lane of the master vehicle during the target lane change time period, the lane change result information of the target lane change time period is determined, and the number of successful lane changes and the number of failed lane changes for each data block are determined based on the lane change result information of each lane change time period. The lane change result information is used to indicate whether the master vehicle has successfully changed lanes. The lane change success rate of the autonomous driving road test data is calculated based on the number of successful lane changes and the number of failed lane changes for each data block. If there are interacting vehicles in the target lane of the main vehicle during the target lane change time period, the lane change result information for the target lane change time period is determined, and the number of successful lane changes and the number of failed lane changes for each data block are determined based on the lane change result information for each lane change time period. This includes: if there are interacting vehicles in the target lane of the main vehicle during the target lane change time period, the target lane change time period is extracted to obtain at least one valid interactive lane change frame; the effective interactive lane change duration is calculated based on the at least one valid interactive lane change frame; if the interactive lane change duration is greater than or equal to a first preset threshold and less than or equal to a second preset threshold, the lateral displacement of the main vehicle during the target lane change time period is calculated, and the lane status information of the main vehicle is obtained; the lane change result information for the target lane change time period is determined based on the lateral displacement and the lane status information; if the interactive lane change duration is greater than the second preset threshold, the lane change of the main vehicle is determined to have failed, and the lane change result information for the target lane change time period is obtained; the number of successful lane changes and the number of failed lane changes for each data block are determined based on the lane change result information for each lane change time period. The step of determining the lane change result information for the target lane change time period based on the lateral displacement and the lane status information includes: if the lateral displacement is greater than or equal to a preset displacement, and the lane status information indicates that the main vehicle has traveled to the target lane, then the main vehicle is determined to have completed the lane change, and the corresponding lane change completion time is recorded; when the main vehicle completes the lane change, it is determined whether there is a valid interactive lane change frame within a preset time range earlier than the lane change completion time; if there is a valid interactive lane change frame within the preset time range, then the main vehicle is determined to have successfully changed lanes, and the lane change result information for the target lane change time period is obtained; if the lateral displacement is less than the preset displacement, or the lane status information indicates that the main vehicle has not traveled to the target lane, then the main vehicle is determined to have failed to change lanes, and the lane change result information for the target lane change time period is obtained.

2. The method for processing lane change data according to claim 1, characterized in that, After calculating the lane change success rate of the autonomous driving road test data based on the number of successful lane changes and the number of failed lane changes for each data block, the method further includes: By taking values ​​one by one from the first preset threshold within a preset numerical range, multiple first preset thresholds are obtained; Based on each of the plurality of first preset thresholds, the lane change success rate of the plurality of autonomous driving road test data is calculated to obtain the plurality of lane change success rates of each autonomous driving road test data. Based on the multiple lane change success rates of various autonomous driving road test data, the mean lane change success rate and standard deviation of the lane change success rate of each autonomous driving road test data are calculated. The target autonomous driving road test data is determined based on the mean and standard deviation of the lane change success rate of each autonomous driving road test data.

3. The method for processing lane change data according to claim 2, characterized in that, The determination of target autonomous driving road test data based on the mean and standard deviation of lane change success rates from each autonomous driving road test data includes: The mean difference is calculated based on the average lane change success rate of the first data and the average lane change success rate of the second data. The first data and the second data are any two autonomous driving road test data from the plurality of autonomous driving road test data. The target error is calculated based on the standard deviation of the lane change success rate of the first data, the standard deviation of the lane change success rate of the second data, and a preset error propagation formula. If the mean difference is greater than the target error, then the autonomous driving road test data with the larger mean lane change success rate is determined as the target autonomous driving road test data.

4. The method for processing lane change data according to claim 1, characterized in that, The autonomous driving road test data is filtered based on each lane change request to obtain at least one data block, the data block including at least one lane change request, including: Determine whether there is a lane change request within a preset time period of each data frame in the autonomous driving road test data, with the preset time period of each data frame taking the time of each data frame as the midpoint. If a lane change request exists within a preset time period of a data frame, then the data frame with the lane change request within the preset time period will be identified as the target frame. Multiple target frames that are consecutive in time are combined into a data block to obtain at least one data block of the autonomous driving road test data, and the data block includes at least one lane change request.

5. The method for processing lane change data according to claim 1, characterized in that, The step of extracting each data block to obtain at least one lane-change time period for each data block includes: In each data block, select the candidate time period corresponding to each lane change request to obtain at least one candidate time period for each data block; If the number of at least one candidate time period for a data block is one, then the target candidate time period is determined as the lane change time period of the data block. If the number of at least two candidate time periods in a data block is at least two, then determine whether two adjacent candidate time periods in the at least two candidate time periods meet the preset merging conditions. The preset merging conditions are that the time interval between two adjacent candidate time periods is less than or equal to a preset duration, the lane change direction of the master vehicle remains unchanged, the master vehicle is in a non-continuous lane change driving state, and the master vehicle is in a non-takeover driving state in the previous candidate time period of the two adjacent candidate time periods. If two adjacent candidate time periods meet the preset merging conditions, the two adjacent candidate time periods are combined into a lane change time period to obtain the lane change time period of the data block, until each data block obtains at least one corresponding lane change time period.

6. The method for processing lane change data according to any one of claims 1-5, characterized in that, Determining whether there are interacting vehicles in the target lane of the main vehicle during each lane-changing time period includes: Determine whether there are vehicles in front and behind in the target lane of the main vehicle in each data frame. Each lane change time period includes multiple data frames. Data frames containing both vehicles in front and behind are identified as interactive lane change frames. If the proportion of interactive lane change frames in any lane change time period is greater than or equal to a preset ratio, then it is determined that there are interactive vehicles in the target lane of the main vehicle in any lane change time period. If the proportion of interactive lane change frames in any lane change time period is less than the preset ratio, then it is determined that there are no interactive vehicles in the target lane of the main vehicle in any lane change time period.

7. A processing device for lane change data, characterized in that, The device for processing lane change data includes: The filtering module is used to filter the autonomous driving road test data based on each lane change request when receiving autonomous driving road test data, and obtain at least one data block, wherein the data block includes at least one lane change request. The extraction module is used to extract each data block to obtain at least one lane change time period for each data block; The judgment module is used to determine whether there are interacting vehicles in the target lane of the main vehicle during each lane change time period. The first determining module is used to determine the lane change result information of the target lane change time period if there are interacting vehicles in the target lane of the main vehicle during the target lane change time period, and to determine the number of successful lane changes and the number of failed lane changes for each data block based on the lane change result information of each lane change time period. The lane change result information is used to indicate whether the main vehicle has successfully changed lanes. The first calculation module is used to calculate the lane change success rate of the autonomous driving road test data based on the number of successful lane changes and the number of failed lane changes for each data block. The first determining module includes: an extraction unit, configured to extract the target lane change time period to obtain at least one valid interactive lane change frame if there is an interacting vehicle in the target lane of the main vehicle during the target lane change time period; a first calculation unit, configured to calculate the effective interactive lane change duration based on the at least one valid interactive lane change frame; a second calculation unit, configured to calculate the lateral displacement of the main vehicle during the target lane change time period and obtain the lane status information of the main vehicle if the interactive lane change duration is greater than or equal to a first preset threshold and less than or equal to a second preset threshold; a first determining unit, configured to determine the lane change result information of the target lane change time period based on the lateral displacement and the lane status information; a second determining unit, configured to determine the lane change failure of the main vehicle if the interactive lane change duration is greater than the second preset threshold, and obtain the lane change result information of the target lane change time period; and a third determining unit, configured to determine the number of successful lane changes and the number of failed lane changes for each data block based on the lane change result information of each lane change time period. The first determining unit is specifically used for: if the lateral displacement is greater than or equal to a preset displacement, and the lane status information indicates that the main vehicle has traveled to the target lane, then determining that the main vehicle has completed the lane change and recording the corresponding lane change completion time; when the main vehicle has completed the lane change, determining whether there is a valid interactive lane change frame within a preset time range earlier than the lane change completion time; if there is a valid interactive lane change frame within the preset time range, then determining that the main vehicle has successfully changed lanes and obtaining the lane change result information for the target lane change time period; if the lateral displacement is less than the preset displacement, or the lane status information indicates that the main vehicle has not traveled to the target lane, then determining that the main vehicle has failed to change lanes and obtaining the lane change result information for the target lane change time period.

8. A device for processing lane change data, characterized in that, The device for processing lane change data includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the lane change data processing device to perform the lane change data processing method as described in any one of claims 1-6.

9. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements the lane change data processing method as described in any one of claims 1-6.

Images