Obstacle intention recognition method and device, electronic equipment and storage medium

Abstract

The invention provides an obstacle intention recognition method and device, electronic equipment and a storage medium. The method comprises the steps that working condition parameters of a current frame in a self-vehicle cruise scene are acquired in real time; determining values of a plurality of intention judgment parameters based on the working condition parameters of the current frame; determining the driving intention of the obstacle based on the values of the plurality of intention judgment parameters; and performing post-processing on the determined driving intention of the obstacle based on the engineering demand in the self-vehicle cruising scene. Namely, according to the embodiment of the invention, the values of the intention judgment parameters can be determined through the working condition parameters obtained in real time, the driving intention of the obstacle is determined from multiple aspects according to the values of the intention judgment parameters, and the driving intention is post-processed, so that whether the obstacle has the lane changing intention or not is determined; and the accuracy of obstacle intention recognition of the obstacle under the full working condition of the cruise scene is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent driving, and particularly to an obstacle intention recognition method, device, electronic device and computer-readable storage medium. Background Art

[0002] With the continuous development and popularization of autonomous driving technology, users' requirements for the safety and comfort of autonomous driving technology are constantly increasing. The R & D costs, computing power costs, scenario coverage, stability and other performance indicators of autonomous driving technology required by demand parties are also getting higher and higher. In particular, the recognition of obstacles in the cruise scenario is particularly important.

[0003] In related technologies, the following two methods are usually used to recognize obstacles in the cruise scenario.

[0004] One is a rule-based trajectory prediction method, which also uses rule logic to predict the intention of obstacles in the cruise scenario. It recognizes the intention of obstacles from a single level through some single frames or simple parameters. Most of these methods do not combine actual engineering requirements and cannot cover all working conditions, resulting in relatively low accuracy of obstacle intention recognition.

[0005] The other is a neural network-based trajectory prediction method, which uses a model to train a cruise data set and uses the trained model to perform online inference of the lane-changing intention of obstacles. Although this method eliminates the cumbersome rule logic, it is restricted by many aspects, resulting in unsatisfactory online inference of the lane-changing intention of obstacles in the cruise scenario.

[0006] Therefore, how to improve the accuracy of obstacle recognition under all working conditions in the cruise scenario is a problem to be solved at present. Summary of the Invention

[0007] The present invention provides an obstacle intention recognition method, device, electronic device and computer-readable storage medium to at least solve the problem of low accuracy of obstacle recognition under all working conditions in the cruise scenario in related technologies. The technical solution of the present invention is as follows:

[0008] According to the first aspect of the embodiments of the present invention, an obstacle intention recognition method is provided, including:

[0009] Obtaining real-time working condition parameters of the current frame in the self-vehicle cruise scenario;

[0010] Determining values of a plurality of intention judgment parameters based on the working condition parameters of the current frame;

[0011] Determining the driving intention of the obstacle based on the values of the plurality of intention judgment parameters;

[0012] Based on the engineering requirements in the ego vehicle cruise scenario, post-process the driving intention of the determined obstacle.

[0013] Optionally, obtaining the values of multiple intention judgment parameters based on the working condition parameters of the current frame includes:

[0014] Based on the working condition parameters of the current frame, search for multiple calibration tables in the calibration library;

[0015] According to the query results of each calibration table, obtain the corresponding values of multiple intention judgment parameters.

[0016] Optionally, determining the driving intention of the obstacle based on the values of the multiple intention judgment parameters includes:

[0017] Select parameter values from the multiple intention judgment parameters, and based on the selected parameter values, judge whether the obstacle has a lane-changing intention from the time level, distance level, speed level, angle level, and turn signal level respectively;

[0018] When it is judged from any one of the time level, distance level, speed level, angle level, and turn signal level that the obstacle has a lane-changing intention, determine that the driving intention of the obstacle is a lane-changing intention.

[0019] Optionally, determining the driving intention of the obstacle based on the values of the multiple intention judgment parameters further includes:

[0020] When the obstacle is driving in a curve with a curvature greater than a set threshold, judge whether the obstacle has a lane-changing intention from the distance level and the turn signal level;

[0021] When it is judged from both the distance level and the turn signal level that the obstacle has a lane-changing intention, determine that the driving intention of the obstacle is a lane-changing intention.

[0022] Optionally, selecting parameter values from the multiple intention judgment parameters, and based on the selected parameter values, judging whether the obstacle has a lane-changing intention from the time level, distance level, speed level, angle level, and turn signal level respectively includes:

[0023] Select single-frame parameter values and multi-frame parameter values related to the time level from the multiple intention judgment parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the time level, judge whether the obstacle has the intention to change lanes quickly;

[0024] Select parameter values related to the distance level from the multiple intention judgment parameters, and based on the selected single-frame parameter values related to the distance level, judge whether the distance of the obstacle from the left and right lane boundaries meets the requirements for lane change;

[0025] Select a single-frame parameter value related to the speed level from the multiple intention judgment parameters, and based on the selected single-frame parameter value related to the speed level, determine whether the obstacle changes lanes at a relatively fast lateral speed;

[0026] Select a single-frame parameter value and a multi-frame parameter value related to the angle level from the multiple intention judgment parameters, and based on the selected single-frame parameter value and multi-frame parameter value related to the angle level,

[0027] Determine whether there is a large angle difference between the obstacle and the current lane, and judge whether the judgment result based on the angle difference meets the requirement of changing lanes;

[0028] Select a single-frame parameter value and a multi-frame parameter value related to the turn signal level from the multiple intention judgment parameters, and based on the selected single-frame parameter value and multi-frame parameter value related to the turn signal level,

[0029] Judge whether the turn signal information of the obstacle meets the requirement of changing lanes.

[0030] Optionally, the post-processing of the determined driving intention of the obstacle based on the engineering requirements in the self-vehicle cruise scenario includes:

[0031] In the scenario where the obstacle drives along the line for a long time or straddles the line, if the determined driving intention of the obstacle is a lane-changing intention, then the lane-changing intention of the obstacle is forcibly corrected to a straight-going intention; or

[0032] In the scenario where the obstacle has completed a lane change or has just turned at an intersection, if the determined driving intention of the obstacle is a lane-changing intention, then the lane-changing intention of the obstacle is forcibly corrected to a straight-going intention; or

[0033] In the scenario where the obstacle merges with other vehicles on the on-ramp, if the determined driving intention of the obstacle is a straight-going intention, then the straight-going intention of the obstacle is adjusted to a lane-changing intention; or

[0034] In the high-speed scenario, when there is a low-speed obstacle or a stationary obstacle in front of the obstacle, if the determined driving intention of the obstacle is a straight-going intention, then the straight-going intention of the obstacle is urgently adjusted to a lane-changing intention; or

[0035] In the scenario of approaching an intersection, when the obstacle needs to cut in and change lanes, and the obstacle makes a low-speed and large-angle lane change, if the determined driving intention of the obstacle is a straight-going intention, then the straight-going intention of the obstacle is urgently adjusted to a lane-changing intention; or

[0036] When the lane-changing intention of the obstacle meets the set number of frames, determine that the driving intention of the obstacle is a lane-changing intention.

[0037] Optionally, the method further includes: establishing multiple calibration tables in the calibration library in advance in the following manner:

[0038] Obtain the working condition parameters and intention judgment parameters in the historical cruise scenario within a set time period;

[0039] Create multiple calibration tables based on the working condition parameters and intention judgment parameters;

[0040] Store the multiple calibration tables into the calibration library to obtain the multiple calibration tables in the calibration library.

[0041] Optionally, the working condition parameters include: obstacle type, obstacle speed, and the relative speed and Euclidean distance between the obstacle and the vehicle; the intention judgment parameters include: single-frame parameters and multi-frame parameters;

[0042] The creating multiple calibration tables based on the working condition parameters and the intention judgment parameters includes:

[0043] Divide the relative speed between the vehicle and the obstacle into N segments, where N is a natural number;

[0044] According to the obstacle type, based on the divided relative speed of the N segments, using the obstacle speed and Euclidean distance as basic parameters, respectively construct calibration tables for the single-frame parameters and multi-frame parameters in the intention judgment parameters.

[0045] According to the second aspect of the embodiments of the present invention, there is provided an obstacle intention recognition device, including:

[0046] An acquisition module, configured to acquire the working condition parameters of the current frame in the vehicle cruise scenario in real time;

[0047] A first determination module, configured to determine the values of multiple intention judgment parameters based on the working condition parameters of the current frame;

[0048] A second determination module, configured to determine the driving intention of the obstacle based on the values of the multiple intention judgment parameters;

[0049] A post-processing module, configured to perform post-processing on the determined driving intention of the obstacle based on the engineering requirements in the vehicle cruise scenario.

[0050] Optionally, the first determination module includes:

[0051] A query module, configured to query multiple calibration tables in the calibration library based on the working condition parameters of the current frame;

[0052] A parameter determination module, configured to obtain the values of the corresponding multiple intention judgment parameters according to the query results of each calibration table.

[0053] Optionally, the second determination module includes:

[0054] The first intention judgment module is used to select parameter values from the multiple intention judgment parameters, and based on the selected parameter values, judge whether the obstacle has a lane-changing intention from the time level, distance level, speed level, angle level, and turn signal level respectively;

[0055] The first intention determination module is used to determine the

[0056] driving intention of the obstacle as a lane-changing intention when it is judged from any one of the time level, distance level, speed level, angle level, and turn signal level that the obstacle has a lane-changing intention.

[0057] Optionally, the second determination module further includes:

[0058] The second intention judgment module is used to judge whether the obstacle has a lane-changing intention from the distance level and the turn signal level when the obstacle is driving in a curve with a curvature greater than a set threshold;

[0059] The second intention determination module is used to determine the driving intention of the obstacle as a lane-changing intention when it is judged from both the distance level and the turn signal level that the obstacle has a lane-changing intention.

[0060] Optionally, the first intention judgment module includes:

[0061] The first lane-changing intention judgment module is used to select single-frame parameter values and multi-frame parameter values related to the time level from the multiple intention judgment parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the time level, judge whether the obstacle has an intention to change lanes quickly;

[0062] The second lane-changing intention judgment module is used to select parameter values related to the distance level from the multiple intention judgment parameters, and based on the selected single-frame parameter values related to the distance level,

[0063] judge whether the distance of the obstacle from the left and right lane boundaries meets the requirements for lane change;

[0064] The third lane-changing intention judgment module is used to select single-frame parameter values related to the speed level from the multiple intention judgment parameters, and based on the selected single-frame parameter values related to the speed level,

[0065] judge whether the obstacle changes lanes at a relatively fast lateral speed;

[0066] The fourth lane-changing intention judgment module is used to select single-frame parameter values and multi-frame parameter values related to the angle level from the multiple intention judgment parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the angle level, judge whether there is a large angle difference between the obstacle and the current lane, and judge whether the judgment result based on the angle difference meets the lane-changing requirement;

[0067] The fifth lane-changing intention judgment module is used to select single-frame parameter values and multi-frame parameter values related to the turn signal level from the multiple intention judgment parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the turn signal level, judge whether the turn signal information of the obstacle meets the lane-changing requirement.

[0068] Optionally, the post-processing module includes at least one of the following:

[0069] The first correction module is used to forcibly correct the lane-changing intention of the obstacle to a straight-going intention when the obstacle is driving along the line for a long time or straddling the line, and if the determined driving intention of the obstacle is a lane-changing intention;

[0070] The second correction module is used to forcibly correct the lane-changing intention of the obstacle to a straight-going intention when the obstacle has completed a lane change or has just turned at an intersection, and if the determined driving intention of the obstacle is a lane-changing intention;

[0071] The first adjustment module is used to adjust the straight-going intention of the obstacle to a lane-changing intention when the obstacle merges with other vehicles on the ramp, and if the determined driving intention of the obstacle is a straight-going intention;

[0072] The second adjustment module is used to urgently adjust the straight-going intention of the obstacle to a lane-changing intention when there is a low-speed obstacle or a stationary obstacle in front of the obstacle in a high-speed scenario, and if the determined driving intention of the obstacle is a straight-going intention;

[0073] The third adjustment module is used to urgently adjust the straight-going intention of the obstacle to a lane-changing intention when the obstacle needs to cut in and change lanes in a scenario approaching an intersection, and the obstacle is a low-speed large-angle lane change, and if the determined driving intention of the obstacle is a straight-going intention;

[0074] The sending module is used to issue that the driving intention of the obstacle is a lane-changing intention when the lane-changing intention of the obstacle meets the set number of frames.

[0075] Optionally, the device further includes: a building module for pre-building multiple calibration tables in the calibration library.

[0076] Optionally, the building module includes:

[0077] A parameter acquisition module, configured to acquire operating condition parameters and intention judgment parameters in a historical cruising scenario within a set time period;

[0078] A creation module, configured to create a plurality of calibration tables based on the operating condition parameters and intention judgment parameters;

[0079] A storage module, configured to store the plurality of calibration tables into a calibration library to obtain the plurality of calibration tables in the calibration library.

[0080] Optionally, the operating condition parameters acquired by the parameter acquisition module include: obstacle type, obstacle speed, and relative speed and Euclidean distance between the obstacle and the vehicle; and the intention judgment parameters acquired include: single-frame parameters and multi-frame parameters;

[0081] The creation module includes:

[0082] A division module, configured to divide the relative speed between the vehicle and the obstacle into N segments, where N is a natural number;

[0083] A construction module, configured to construct calibration tables corresponding to the single-frame parameters and multi-frame parameters in the intention judgment parameters respectively according to the obstacle type, based on the relative speed of the N segments after division, with the obstacle speed and Euclidean distance as basic parameters.

[0084] According to a third aspect of an embodiment of the present invention, there is provided an electronic device, including:

[0085] A processor;

[0086] A memory for storing executable instructions of the processor;

[0087] Wherein, the processor is configured to execute the instructions to implement the obstacle intention recognition method as described above.

[0088] According to a fourth aspect of an embodiment of the present invention, there is provided a computer-readable storage medium, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enabling the electronic device to execute the obstacle intention recognition method as described above.

[0089] According to a fifth aspect of an embodiment of the present invention, there is provided a computer program product, including a computer program or instructions, and when the computer program or instructions are executed by a processor of an electronic device, implementing the obstacle intention recognition method as described above.

[0090] The technical solutions provided by the embodiments of the present invention at least bring the following beneficial effects:

[0091] In an embodiment of the present invention, the working condition parameters of the current frame in the self-vehicle cruise scenario are obtained in real time; the values of multiple intention judgment parameters are determined based on the working condition parameters of the current frame; the driving intention of the obstacle is determined based on the values of the multiple intention judgment parameters; and post-processing is performed on the determined driving intention of the obstacle based on the engineering requirements in the self-vehicle cruise scenario. That is to say, in the embodiment of the present invention, through the working condition parameters obtained in real time, the values of multiple intention judgment parameters can be determined, and based on the values of the multiple intention judgment parameters, the driving intention of the obstacle can be determined from multiple aspects, and post-processing is performed on the driving intention, so as to determine whether the obstacle has a lane-changing intention, ensuring the accuracy of obstacle intention recognition under all working conditions in the cruise scenario and providing strong support for downstream decision-making and planning.

[0092] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present invention, and are used together with the specification to explain the principles of the present invention, and do not constitute an improper limitation of the present invention. To more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0094] Figure 1 is a flowchart of an obstacle intention recognition method provided by an embodiment of the present invention.

[0095] Figure 2 is a schematic diagram of an application example of an obstacle intention recognition method provided by an embodiment of the present invention.

[0096] Figure 3 is a block diagram of an obstacle intention recognition device provided by an embodiment of the present invention.

[0097] Figure 4 is a block diagram of a first determination module provided by an embodiment of the present invention.

[0098] Figure 5 is a block diagram of a second determination module provided by an embodiment of the present invention.

[0099] Figure 6 is a block diagram of a first intention judgment module provided by an embodiment of the present invention.

[0100] Figure 7It is a block diagram of a post-processing module provided by an embodiment of the present invention.

[0101] Figure 8 It is a block diagram of an electronic device provided by an embodiment of the present invention.

[0102] Figure 9 It is a block diagram of a device for obstacle intention recognition provided by an embodiment of the present invention. Detailed implementation manners

[0103] In order to enable those of ordinary skill in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0104] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order different from those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0105] Figure 1 It is a flowchart of a method for obstacle intention recognition provided by an embodiment of the present invention. As Figure 1 shown, the method includes the following steps:

[0106] Step 101: Real-time obtain the working condition parameters of the current frame in the self-vehicle cruise scenario;

[0107] Step 102: Determine the values of multiple intention judgment parameters based on the working condition parameters of the current frame;

[0108] Step 103: Determine the driving intention of the obstacle based on the values of the multiple intention judgment parameters;

[0109] Step 104: Post-process the determined driving intention of the obstacle based on the engineering requirements in the self-vehicle cruise scenario.

[0110] The obstacle intention recognition method described in the present invention can be applied to terminals, servers, automatic lane change assistance systems in intelligent driving, autonomous driving control systems, etc., without limitation here. The terminal implementation devices can be electronic devices such as in-vehicle terminals, vehicle control platforms, industrial computers, etc. The server can be an independent server, or a server cluster, or a server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, intermediate services, domain name services, security services, content delivery networks, or a big data and artificial intelligence platform, etc., without limitation here.

[0111] The following combines Figure 1 to detail the specific implementation steps of an obstacle intention recognition method provided by an embodiment of the present invention.

[0112] In step 101, the working condition parameters of the current frame in the self-vehicle cruise scenario are obtained in real time.

[0113] In this step, the self-vehicle information in the cruise scenario can be collected in real time through in-vehicle sensors. The self-vehicle information includes the working condition parameters of the current frame, etc. Of course, in practical applications, it can also include other parameter information, which is not limited in this embodiment.

[0114] Among them, the working condition parameters can include but are not limited to the following parameters:

[0115] 1) Relative speed: The relative speed between the obstacle and the self-vehicle directly affects the performance of the self-vehicle. When the speed of the obstacle is greater than the speed of the self-vehicle, the lane change intention should be more radical; when the speed of the self-vehicle is greater than the speed of the obstacle, the lane change intention should be more conservative.

[0116] 2) Euclidean distance: The Euclidean distance between the obstacle and the self-vehicle directly affects the performance of the self-vehicle. When the obstacle is relatively close to the self-vehicle, a wrong cut-in intention may cause the self-vehicle to brake by mistake. Therefore, the calibration parameters should be more conservative at this time.

[0117] 3) Obstacle type: The embodiment of the present invention can cover obstacles of three major categories: non-motor vehicles (cyclist), cars, and large vehicles (truck / bus). Different types of obstacles will have different behaviors when changing lanes. For example, the limit value of the lateral speed of a car-type vehicle is relatively large, while that of a large vehicle is relatively small. That is to say, generally, car-type vehicles are more flexible in changing lanes and have a shorter lane change time than large vehicles.

[0118] 4) Obstacle speed: The behavior of an obstacle when changing lanes is different in different vehicle speed segments. For example, in a high-speed working condition, the angle of the steering wheel, the value of yawrate, and the lateral speed are relatively small when changing lanes. Corresponding to the distribution of parameters, the distribution of calibration parameters is also not the same in different vehicle speed segments. Therefore, it is necessary to segment the vehicle speed in the early stage.

[0119] In step 102, values of multiple intention judgment parameters are determined based on the working condition parameters of the current frame.

[0120] In this step, based on the working condition parameters of the current frame, multiple calibration tables in the calibration library can be searched; and according to the query results of each calibration table, values of the corresponding multiple intention judgment parameters are obtained.

[0121] For example, among the working condition parameters in this embodiment, the obstacle type is the car type. Correspondingly, the Euclidean distance is 20 meters, the relative speed is 10 m / s, and the obstacle speed is 10 m / s. Based on this, by searching multiple calibration tables, the corresponding multiple intention judgment parameters can be obtained. The multiple intention judgment parameters include: single-frame parameters and multi-frame parameters. For example, there are 17 multiple intention judgment parameters, including 6 single-frame parameters and 11 multi-frame parameters. Of course, in specific applications, it is not limited to these 17 intention judgment parameters, and other parameters can also be adaptively included. This embodiment does not make any restrictions.

[0122] Among them, the 6 single-frame parameters can include:

[0123] Parameter 1: The time for the centroid of the single-frame obstacle to reach the left and right boundary lines of the lane.

[0124] Parameter 2: The lateral distance from the front of the left and right sides of the centroid of the single-frame obstacle to the lane boundary line.

[0125] Parameter 3: The lateral speed of the centroid of the single-frame obstacle relative to the current lane where it is located.

[0126] Parameter 4: The difference between the theta of the centroid of the single-frame obstacle and the theta of the lane where it is located.

[0127] Parameter 5: The lateral distance of the centroid of the single-frame obstacle from the center line of the lane where it is located.

[0128] Parameter 6: The turn signal information of the single-frame obstacle.

[0129] The 11 multi-frame parameters include:

[0130] Parameter 7: The average value of the lateral distances from the front of the left and right sides of the obstacle in the historical frames 0 to 2 to the lane boundary line.

[0131] Parameter 8: The average value of the lateral distances from the front of the left and right sides of the obstacle in the historical frames 3 to 5 to the lane boundary line.

[0132] Parameter 9: The average value of the lateral distances from the front of the left and right sides of the obstacle in the historical frames 6 to 8 to the lane boundary line.

[0133] Parameter 10: The lateral speed of the obstacle in the historical frames 0 to 4 relative to the current lane where it is located.

[0134] Parameter 11: Mean value of the difference between the theta of the obstacle in the historical 0-2 frames and the theta of the lane where it is located.

[0135] Parameter 12: Mean value of the difference between the theta of the obstacle in the historical 3-5 frames and the theta of the lane where it is located.

[0136] Parameter 13: Mean value of the difference between the theta of the obstacle in the historical 6-8 frames and the theta of the lane where it is located.

[0137] Parameter 14: Lateral distance of the obstacle in the historical 0-1 frames from the center line of the lane where it is located.

[0138] Parameter 15: Lateral distance of the obstacle in the historical 2-3 frames from the center line of the lane where it is located.

[0139] Parameter 16: Lateral distance of the obstacle in the historical 4-5 frames from the center line of the lane where it is located.

[0140] Parameter 17: Turn signal information of the obstacle in the historical 0-2 frames.

[0141] Optionally, in this embodiment, multiple calibration tables in the calibration library are pre-established, and are specifically established in the following manner:

[0142] First, obtain the working condition parameters and intention judgment parameters in the historical cruise scenarios within a set time period; secondly, create multiple calibration tables based on the working condition parameters and intention judgment parameters; finally, store the multiple calibration tables into the calibration library to obtain the multiple calibration tables in the calibration library.

[0143] Among them, the obtained working condition parameters may include: obstacle type, obstacle speed, and relative speed and Euclidean distance between the obstacle and the vehicle; the intention judgment parameters may include: single-frame parameters and multi-frame parameters; specifically, the above 6 single-frame parameters and 11 multi-frame parameters are taken as examples.

[0144] The creating of multiple calibration tables based on the working condition parameters and the intention judgment parameters includes: dividing the relative speed between the vehicle and the obstacle into N segments, where N is a natural number; according to the obstacle type, based on the divided relative speed of the N segments, using the obstacle speed and Euclidean distance as basic parameters, respectively construct calibration tables for the single-frame parameters and multi-frame parameters in the intention judgment parameters.

[0145] Among them, in this embodiment, the relative speed between the obstacle and the vehicle is divided into N segments in a segmented form. In this embodiment, it is taken as an example of being divided into 6 segments, and the thresholds of each segment are taken as -10m / s, -5m / s, 0, 5m / s, 10m / s respectively, but it is not limited to this in specific implementation:

[0146] The 6 segments are: relative speed less than -10 m / s; greater than -10 m / s and less than -5 m / s; greater than -5 m / s and less than 0; greater than 0 and less than 5 m / s; greater than 5 m / s and less than 10 m / s; greater than 10 m / s.

[0147] Based on the above division, the number of calibration tables in the calibration library is: 3 * 6 * 17 = 306. Among them, 3 represents the number of obstacle types, 6 represents the number of segments of relative speed, and 17 represents the number of calibration tables constructed with the obstacle speed and Euclidean distance as the basis parameters. It should be noted that the calibration table in this embodiment is a two-dimensional calibration table.

[0148] In step 103, based on the values of the multiple intention judgment parameters, the driving intention of the obstacle is determined.

[0149] In this step, first, parameter values are selected from the multiple intention judgment parameters. Based on the selected parameter values, it is judged whether the obstacle has a lane-changing intention from the time level, distance level, speed level, angle level, and turn signal level respectively; second, when it is judged from any one of the time level, distance level, speed level, angle level, and turn signal level that the obstacle has a lane-changing intention, the driving intention of the obstacle is determined to be a lane-changing intention.

[0150] Among them, the selection of parameter values from the multiple intention judgment parameters and the judgment of whether the obstacle has a lane-changing intention from the time level, distance level, speed level, angle level, and turn signal level based on the selected parameter values include:

[0151] Select single-frame parameter values and multi-frame parameter values related to the time level from the multiple intention judgment parameters. Based on the selected single-frame parameter values and multi-frame parameter values related to the time level, it is judged whether the obstacle has the intention of quickly changing lanes;

[0152] Select parameter values related to the distance level from the multiple intention judgment parameters. Based on the selected single-frame parameter values related to the distance level, it is judged whether the distance of the obstacle from the left and right lane boundaries meets the requirements for lane-changing;

[0153] Select single-frame parameter values related to the speed level from the multiple intention judgment parameters. Based on the selected single-frame parameter values related to the speed level, it is judged whether the obstacle changes lanes at a relatively fast lateral speed;

[0154] Select single-frame parameter values and multi-frame parameter values related to the angle level from the multiple intention judgment parameters. Based on the selected single-frame parameter values and multi-frame parameter values related to the angle level,

[0155] Determine whether there is a large angular difference between the obstacle and the current lane, and determine whether the requirement for lane change is met based on the judgment result of the angular difference;

[0156] Select the single-frame parameter value and multi-frame parameter value related to the turn signal level from the multiple intention judgment parameters, and based on the selected single-frame parameter value and multi-frame parameter value related to the turn signal level,

[0157] Judge whether the turn signal information of the obstacle meets the requirement for lane change.

[0158] Optionally, determining the driving intention of the obstacle based on the values of the multiple intention judgment parameters may further include: when the obstacle is driving on a curve with a curvature greater than a set threshold (such as a curve with a large curvature, etc.), judge whether the obstacle has a lane change intention from the distance level and the turn signal level; when it is judged from both the distance level and the turn signal level that the obstacle has a lane change intention, determine that the driving intention of the obstacle is a lane change intention.

[0159] In step 104, post-process the determined driving intention of the obstacle based on the engineering requirements in the self-vehicle cruise scenario.

[0160] In this step, post-process the determined driving intention of the obstacle. In this embodiment, 6 engineering actual scenarios are taken as examples, such as: riding on the line, straightening judgment, merging scenario, emergency lane change, cutting in line lane change, and intention hysteresis and other actual scenarios. However, in specific applications, it is not limited to this.

[0161] For the actual scenario of riding on the line: when the obstacle is driving along the line or pressing the line for a long time, if the determined driving intention of the obstacle is a lane change intention, then forcefully correct the lane change intention of the obstacle to a straight driving intention; or

[0162] For the actual scenario of straightening judgment: when the obstacle has completed a lane change or just turned at an intersection, if the determined driving intention of the obstacle is a lane change intention, then forcefully correct the lane change intention of the obstacle to a straight driving intention; or

[0163] For the actual scenario of merging scenario: when the obstacle merges with other vehicles on the ramp, if the determined driving intention of the obstacle is a straight driving intention, then adjust the straight driving intention of the obstacle to a lane change intention; or

[0164] For the actual scenario of emergency lane change: in the high-speed scenario, when there is a low-speed obstacle or a stationary obstacle in front of the obstacle, if the determined driving intention of the obstacle is a straight driving intention, then urgently adjust the straight driving intention of the obstacle to a lane change intention; or

[0165] For the actual scenario of cutting in and changing lanes: when approaching an intersection, if an obstacle needs to cut in and change lanes and the obstacle makes a low-speed and large-angle lane change, and if the determined driving intention of the obstacle is a straight-ahead intention, then the straight-ahead intention of the obstacle is urgently adjusted to a lane-changing intention; or

[0166] For the actual scenario of intention hysteresis: when the lane-changing intention of an obstacle satisfies a set number of frames (such as three consecutive frames, etc.), the driving intention of the obstacle is determined to be a lane-changing intention.

[0167] In the embodiments of the present invention, working condition parameters of the current frame in the self-vehicle cruise scenario are obtained in real time; values of multiple intention judgment parameters are determined based on the working condition parameters of the current frame; the driving intention of the obstacle is determined based on the values of the multiple intention judgment parameters; and post-processing is performed on the determined driving intention of the obstacle based on the engineering requirements in the self-vehicle cruise scenario. That is to say, in the embodiments of the present invention, by obtaining working condition parameters in real time, values of multiple intention judgment parameters can be determined, and based on the values of the multiple intention judgment parameters, the driving intention of the obstacle is determined from multiple aspects, and post-processing is performed on the driving intention, so as to determine whether the obstacle has a lane-changing intention, ensuring the accuracy of obstacle intention recognition in all working conditions of the cruise scenario and providing strong support for downstream decision-making and planning.

[0168] Furthermore, in the embodiments of the present invention, multiple calibration tables are created in advance based on the working condition parameters and intention judgment parameters in the historical cruise scenario within a set time period and stored in a calibration library. When the working condition parameters of the current frame in the self-vehicle cruise scenario are obtained in real time, based on the working condition parameters of the current frame, the corresponding values of multiple intention judgment parameters are determined by querying the calibration tables in the calibration library. Based on the values of the multiple intention judgment parameters, the driving intention of the obstacle is determined from multiple aspects. That is to say, in the embodiments of the present invention, through three stages of creating calibration tables in the calibration library, intention judgment, and intention post-processing, the intention recognition of the obstacle is completed from multiple aspects, improving the accuracy of obstacle intention recognition in the cruise scenario.

[0169] Please also refer to Figure 2 , which is a schematic diagram of an application example of an obstacle intention recognition method provided by the embodiments of the present invention. This method performs intention recognition on multi-stage obstacles in the cruise scenario. Through logical settings at multiple levels in a multi-stage manner in this embodiment, the accuracy of obstacle intention recognition in all working conditions of the cruise scenario is ensured, providing strong support for downstream decision-making and planning.

[0170] As Figure 2 shown, this embodiment includes four stages:

[0171] The first stage: Parameter selection, that is, according to engineering experience, some working condition parameters and intention judgment parameters are pre-selected. Among them, the working condition parameters include: relative speed, Euclidean distance, obstacle type (taking obstacles of three major types, namely non-motor vehicle cyclist, car type, and truck / bus type as examples), and obstacle speed. Among them, taking 6 single-frame parameters and 11 multi-frame parameters as examples for the intention judgment parameters, the specific details are as described above and will not be elaborated here.

[0172] The second stage: Creating a calibration library, that is, in this stage, according to the selected working condition parameters and intention judgment parameters, multiple calibration tables are created based on engineering experience. The creation process is as described above and will not be elaborated here. When the working condition parameters of the current frame in the cruise scenario are obtained in real time, the corresponding multiple intention judgment parameters can be dynamically obtained by looking up the calibration tables.

[0173] It should be noted that each calibration table includes: obstacle type, Euclidean distance, relative speed, obstacle speed, values of intention judgment parameters, etc.

[0174] The third stage: Using the found intention judgment parameters for intention judgment, that is, in this stage, the working condition parameters of the current frame are obtained in real time from the line, and multiple calibration tables in the calibration library are searched based on the working condition parameters to obtain the values of the corresponding intention judgment parameters, and the driving intention of the obstacle is judged based on these intention judgment parameter values. In this embodiment, the intention judgment parameters are, for example, Figure 2 time, vy, and dist as in

[0175] Taking this embodiment as an example, the driving intention of the obstacle is judged from five aspects: time level, distance level, speed level, angle level, and turn signal level, but it is not limited to this in actual applications:

[0176] Time level: In this embodiment, parameter 1, parameter 3, and parameter 10 in the intention judgment parameters can be selected to judge whether the obstacle has the intention of quickly changing lanes, and if it is determined that the obstacle has the intention of quickly changing lanes, that is, the intention of changing lanes is determined.

[0177] Distance level: In this embodiment, parameter 2, parameter 3, and parameter 5 in the intention judgment parameters can be selected to judge whether the distance of the obstacle from the left and right lane boundaries meets the requirements for changing lanes, and if it is determined that the distance of the obstacle from the left and right lane boundaries meets the requirements for changing lanes, that is, the intention of changing lanes is determined.

[0178] Speed level: In this embodiment, parameter 2, parameter 3, parameter 7 - 9, and parameter 14 - 16 in the intention judgment parameters can be selected to judge whether the obstacle changes lanes at a relatively fast lateral speed, and if it is determined that the obstacle changes lanes at a relatively fast lateral speed, that is, the intention of changing lanes is determined.

[0179] Angle aspect: In this embodiment, parameters 2, 3, 4, 11 - 13 in the intention judgment parameters can be selected to determine whether there is a large angle difference between the obstacle and the current lane, so as to determine whether the requirement for lane change is met, and it is determined that there is a large angle difference between the obstacle and the current lane, and then it is determined that the requirement for lane change is met, that is, it is determined that there is a lane change intention.

[0180] Turn signal aspect: In this embodiment, parameters 2, 3, 6, 17 in the intention judgment parameters can be selected to determine whether the turn signal information of the obstacle meets the requirement for lane change, and it is determined that the turn signal information of the obstacle meets the requirement for lane change, that is, it is determined that there is a lane change intention.

[0181] That is to say, if it is determined that the obstacle has a lane change intention in any of the above five aspects, then it is determined that the driving intention of the obstacle is the lane change intention.

[0182] Of course, in combination with the actual engineering situation, if the obstacle is driving on a curve with a large curvature (the curvature is greater than the set threshold, and it can be considered a curve with a large curvature), only the distance aspect and the turn signal aspect are used to determine whether the driving intention of the obstacle is the lane change intention.

[0183] The fourth stage: Intention post - processing, that is, in this stage, in combination with engineering requirements, post - processing is performed on the driving intention of the obstacle determined in the previous stage.

[0184] In this step, intention post - processing can be performed on the driving intention of the obstacle determined in six actual engineering scenarios. Among them, intention post - processing can include: riding on the line, straightening judgment, merging scenario, emergency lane change, cutting - in lane change, and intention hysteresis loop. Among them,

[0185] Riding on the line: If the obstacle drives along the line or presses on the line for a long time, it indicates that actually the obstacle does not have the intention of lane change. At this time, if the driving intention determined in the third stage is the lane change intention, forced correction is required;

[0186] Straightening judgment: If the obstacle has completed a lane change or just turned at an intersection, although it is easy to judge the lane change intention from the distance aspect, in fact, the obstacle does not have the intention of lane change. At this time, if the driving intention determined in the third stage is the lane change intention, forced correction is required;

[0187] Merging scenario: In the on - ramp and other merging scenarios, in most cases, the obstacle has an obvious intention of merging into the lane. At this time, if the driving intention determined in the third stage is the straight - driving intention, the post - processing part needs to ensure that the lane change intention can be given as soon as possible;

[0188] Emergency lane change: In a highway scenario, if there are other low-speed obstacles or stationary obstacles in front of an obstacle, the obstacle will have an obstacle avoidance behavior. At this time, if the driving intention determined in the third stage is a straight-ahead intention, the post-processing part needs to ensure that a lane change intention can be given as soon as possible;

[0189] Cut-in lane change: In scenarios such as approaching an intersection, the obstacle needs to perform a cut-in lane change, which is mostly manifested as a low-speed large-angle lane change. At this time, if the driving intention determined in the third stage is a straight-ahead intention, the post-processing part needs to ensure that a lane change intention can be given as soon as possible;

[0190] Intention hysteresis loop: The lane change intention needs to meet the set number of frames (such as three consecutive frames) before it is sent.

[0191] In the embodiments of the present invention, the intention of the obstacle in the cruise scenario can be identified in multiple stages and multiple levels through the calibration table, intention judgment, and intention post-processing in the created calibration library, thereby improving the accuracy of the obstacle intention recognition in the cruise scenario.

[0192] In the embodiments of the present invention, by fully utilizing the prior information such as maps and obstacles, and meeting the engineering requirements of downstream decision-making and planning, a complete intention recognition framework with multiple stages and multiple levels is provided for the obstacles in the cruise scenario. Through the logical settings of multiple stages and multiple levels, the accuracy of the obstacle intention recognition in all working conditions of the cruise scenario is guaranteed, providing strong support for downstream decision-making and planning.

[0193] It should be noted that for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present disclosure is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential to the present invention.

[0194] Please also refer to Figure 3 , which is a block diagram of an obstacle intention recognition device provided by an embodiment of the present invention. The device includes: an acquisition module 301, a first determination module 302, a second determination module 303, and a post-processing module 304, where,

[0195] The acquisition module 301 is configured to acquire the working condition parameters of the current frame in the self-vehicle cruise scenario in real time;

[0196] The first determination module 302 is configured to determine the values of multiple intention judgment parameters based on the working condition parameters of the current frame;

[0197] The second determination module 303 is configured to determine the driving intention of the obstacle based on the values of the multiple intention judgment parameters;

[0198] The post-processing module 304 is configured to post-process the determined driving intention of the obstacle based on the engineering requirements in the self-vehicle cruise scenario.

[0199] Optionally, in another embodiment, based on the above embodiment, the first determination module 302 includes: a query module 401 and a parameter determination module 402, and its structural block diagram is as Figure 4 shown, where

[0200] The query module 401 is configured to search for multiple calibration tables in the calibration library based on the working condition parameters of the current frame;

[0201] The parameter determination module 402 is configured to obtain the values of the corresponding multiple intention judgment parameters according to the query results of each calibration table.

[0202] Optionally, in another embodiment, based on the above embodiment, the second determination module 303 includes: a first intention judgment module 501 and a first intention determination module 502, and its structural block diagram is as Figure 5 shown, where

[0203] The first intention judgment module 501 is configured to select parameter values from the multiple intention judgment parameters, and based on the selected parameter values, judge whether the obstacle has a lane-changing intention respectively from the time level, distance level, speed level, angle level and turn signal level;

[0204] The first intention determination module 502 is configured to determine the

[0205] driving intention of the obstacle as a lane-changing intention when it is judged from any one of the time level, distance level, speed level, angle level and turn signal level that the obstacle has a lane-changing intention.

[0206] Optionally, in another embodiment, based on the above embodiment, the second determination module further includes: a second intention judgment module and a second intention determination module, where

[0207] The second intention judgment module is configured to judge whether the obstacle has a lane-changing intention from the distance level and the turn signal level when the obstacle is driving in a curve with a curvature greater than a set threshold;

[0208] The second intention determination module is configured to determine the driving intention of the obstacle as a lane-changing intention when it is judged from both the distance level and the turn signal level that the obstacle has a lane-changing intention.

[0209] Optionally, in another embodiment, which is based on the above embodiment, the first intention determination module 501 includes: a first lane change intention determination module 601, a second lane change intention determination module 602, a third lane change intention determination module 603, a fourth lane change intention determination module 604, and a fifth lane change intention determination module 605. The structural block diagram is as shown in Figure 6 shown, where

[0210] The first lane change intention determination module 601 is configured to select single-frame parameter values and multi-frame parameter values related to the time level from the multiple intention determination parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the time level, determine whether the obstacle has the intention to quickly change lanes;

[0211] The second lane change intention determination module 602 is configured to select parameter values related to the distance level from the multiple intention determination parameters, and based on the selected single-frame parameter values related to the distance level,

[0212] determine whether the distance of the obstacle from the left and right lane boundaries meets the requirements for lane change;

[0213] The third lane change intention determination module 603 is configured to select single-frame parameter values related to the speed level from the multiple intention determination parameters, and based on the selected single-frame parameter values related to the speed level,

[0214] determine whether the obstacle changes lanes at a relatively fast lateral speed;

[0215] The fourth lane change intention determination module 604 is configured to select single-frame parameter values and multi-frame parameter values related to the angle level from the multiple intention determination parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the angle level, determine whether there is a large angle difference between the obstacle and the current lane, and determine whether the judgment result based on the angle difference meets the requirements for lane change;

[0216] The fifth lane change intention determination module 605 is configured to select single-frame parameter values and multi-frame parameter values related to the turn signal level from the multiple intention determination parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the turn signal level, determine whether the turn signal information of the obstacle meets the requirements for lane change.

[0217] Optionally, in another embodiment, which is based on the above embodiment, the post-processing module 304 includes at least one of a first correction module 701, a second correction module 702, a first adjustment module 703, a second adjustment module 704, a third adjustment module 705, and a sending module 706. In this embodiment, taking all of the above modules as an example, the structural block diagram is as shown in Figure 7 shown, where

[0218] The first correction module 701 is configured to, when an obstacle drives along a line for a long time or straddles a line, if the determined driving intention of the obstacle is a lane-changing intention, forcibly correct the lane-changing intention of the obstacle to a straight-going intention;

[0219] The second correction module 702 is configured to, when an obstacle has completed a lane change or is at a scene where it has just turned at an intersection, if the determined driving intention of the obstacle is a lane-changing intention, forcibly correct the lane-changing intention of the obstacle to a straight-going intention;

[0220] The first adjustment module 703 is configured to, when an obstacle merges with other vehicles on an on-ramp, if the determined driving intention of the obstacle is a straight-going intention, adjust the straight-going intention of the obstacle to a lane-changing intention;

[0221] The second adjustment module 704 is configured to, in a highway scene, when there is a low-speed obstacle or a stationary obstacle in front of an obstacle, if the determined driving intention of the obstacle is a straight-going intention, urgently adjust the straight-going intention of the obstacle to a lane-changing intention;

[0222] The third adjustment module 705 is configured to, in a scene approaching an intersection, when an obstacle needs to cut in and change lanes, and the obstacle makes a low-speed and large-angle lane change, if the determined driving intention of the obstacle is a straight-going intention, urgently adjust the straight-going intention of the obstacle to a lane-changing intention;

[0223] The sending module 706 is configured to, when the lane-changing intention of an obstacle meets a set number of frames, send down that the driving intention of the obstacle is a lane-changing intention.

[0224] Optionally, in another embodiment, based on the above embodiment, the device further includes: a building module, configured to pre-build a plurality of calibration tables in a calibration library:

[0225] Optionally, the building module includes:

[0226] A parameter acquisition module, configured to acquire working condition parameters and intention judgment parameters in a historical cruise scene within a set time period;

[0227] A creation module, configured to create a plurality of calibration tables based on the working condition parameters and intention judgment parameters;

[0228] A storage module, configured to store the plurality of calibration tables into the calibration library to obtain a plurality of calibration tables in the calibration library.

[0229] Optionally, in another embodiment, based on the above embodiment, the working condition parameters obtained by the parameter acquisition module include: obstacle type, obstacle speed, and the relative speed and Euclidean distance between the obstacle and the vehicle; and the intention judgment parameters obtained include: single-frame parameters and multi-frame parameters;

[0230] The creation module includes:

[0231] A division module, configured to divide the relative speed between the host vehicle and the obstacle into N segments, where N is a natural number;

[0232] A construction module, configured to construct calibration tables for the single-frame parameters and multi-frame parameters in the intention judgment parameters respectively according to the obstacle type, based on the relative speed of the N segments after division, with the obstacle speed and Euclidean distance as basic parameters.

[0233] Optionally, an embodiment of the present invention further provides an electronic device, including:

[0234] A processor;

[0235] A memory for storing executable instructions of the processor;

[0236] Wherein, the processor is configured to execute the instructions to implement the obstacle intention recognition method as described above.

[0237] Optionally, an embodiment of the present invention further provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device can execute the obstacle intention recognition method as described above.

[0238] Optionally, an embodiment of the present invention further provides a computer program product, including a computer program or instructions. When the computer program or instructions are executed by the processor of the electronic device, the obstacle intention recognition method as described above is implemented.

[0239] Regarding the device in the above embodiment, the specific manners in which each module performs operations have been described in detail in the embodiment related to the method, and will not be elaborated here.

[0240] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0241] Figure 8 It is a block diagram of an electronic device 800 provided by an embodiment of the present invention. As shown in the figure, it includes a processor 801, a communication interface 802, a memory 803, and a communication bus 804. Among them, the processor 801, the communication interface 802, and the memory 803 complete communication with each other through the communication bus 804;

[0242] The memory 803 is used to store executable instructions of the processor;

[0243] The processor 801 is used to implement the above-mentioned method when executing the executable instructions on the memory 803.

[0244] Among them, the communication bus in this embodiment can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, only a thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0245] The communication interface is used for communication between the above-mentioned electronic device and other devices.

[0246] The memory may include a Random Access Memory (RAM), or may also include a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located far from the aforementioned processor.

[0247] The above-mentioned processor may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0248] In yet another embodiment provided by the present invention, a computer-readable storage medium is further provided. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the obstacle intention recognition method as described above. For example, the computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0249] In yet another embodiment provided by the present invention, a computer program product is further provided, including a computer program or instructions. When the computer program or instructions are executed by a processor of an electronic device, the obstacle intention recognition method as described above is implemented.

[0250] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that the computer can access, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium may be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

[0251] Figure 9 It is a block diagram of a device 900 for obstacle intention recognition provided by an embodiment of the present invention. For example, the device 900 may be provided as a server. Referring to Figure 9 , the device 900 includes a processing component 922, which further includes one or more processors, and memory resources represented by a memory 932 for storing instructions executable by the processing component 922, such as application programs. The application programs stored in the memory 932 may include one or more modules each corresponding to a set of instructions. In addition, the processing component 922 is configured to execute instructions to perform the above method.

[0252] Device 900 may also include a power supply component 926 configured to perform power management of device 900, a wired or wireless network interface 950 configured to connect device 900 to a network, and an input / output (I / O) interface 958. Device 900 may operate based on an operating system stored in memory 932, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM or the like.

[0253] Other embodiments of the present invention will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following the general principles of the invention and including known common knowledge or conventional technical means in the technical field not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are pointed out by the following claims.

[0254] It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for obstacle intention recognition, characterized in that, Including: Obtaining the working condition parameters of the current frame in the self-vehicle cruise scenario in real time; Determining the values of multiple intention judgment parameters based on the working condition parameters of the current frame; Determining the driving intention of the obstacle based on the values of the multiple intention judgment parameters; Performing post-processing on the determined driving intention of the obstacle based on the engineering requirements in the self-vehicle cruise scenario.

2. The obstacle intention recognition method according to claim 1, wherein The obtaining the values of multiple intention judgment parameters based on the working condition parameters of the current frame includes: Searching for multiple calibration tables in the calibration library based on the working condition parameters of the current frame; Obtaining the corresponding values of multiple intention judgment parameters according to the query results of each calibration table.

3. The obstacle intention recognition method according to claim 1, wherein The determining the driving intention of the obstacle based on the values of the multiple intention judgment parameters includes: Selecting parameter values from the multiple intention judgment parameters, and based on the selected parameter values, judging whether the obstacle has a lane-changing intention respectively from the time level, distance level, speed level, angle level and turn signal level; When it is judged from any one of the time level, distance level, speed level, angle level and turn signal level that the obstacle has a lane-changing intention, determining that the driving intention of the obstacle is a lane-changing intention.

4. The obstacle intention recognition method according to claim 3, characterized in that The determining the driving intention of the obstacle based on the values of the multiple intention judgment parameters further includes: When the obstacle is driving in a curve with a curvature greater than a set threshold, judging whether the obstacle has a lane-changing intention from the distance level and the turn signal level; When it is judged from both the distance level and the turn signal level that the obstacle has a lane-changing intention, determining that the driving intention of the obstacle is a lane-changing intention.

5. The obstacle intention recognition method according to claim 3, characterized in that, The selecting parameter values from the multiple intention judgment parameters, and based on the selected parameter values, judging whether the obstacle has a lane-changing intention respectively from the time level, distance level, speed level, angle level and turn signal level includes: Selecting single-frame parameter values and multi-frame parameter values related to the time level from the multiple intention judgment parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the time level, judging whether the obstacle has an intention to change lanes quickly; Selecting parameter values related to the distance level from the multiple intention judgment parameters, and based on the selected single-frame parameter values related to the distance level, judging whether the distance of the obstacle from the left and right lane boundaries meets the requirements for lane-changing; Selecting single-frame parameter values related to the speed level from the multiple intention judgment parameters, and based on the selected single-frame parameter values related to the speed level, judging whether the obstacle changes lanes at a relatively fast lateral speed; Selecting single-frame parameter values and multi-frame parameter values related to the angle level from the multiple intention judgment parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the angle level, Judging whether there is a large angle difference between the obstacle and the current lane, and judging whether the judgment result based on the angle difference meets the requirements for lane-changing; Selecting single-frame parameter values and multi-frame parameter values related to the turn signal level from the multiple intention judgment parameters, and based on the selected single-frame parameter values and multi-frame parameter values related to the turn signal level, Judging whether the turn signal information of the obstacle meets the requirements for lane-changing.

6. The method for identifying the intention of an obstacle according to any one of claims 1 to 5, characterized in that, Based on the engineering requirements in the ego vehicle cruise scenario, post - process the determined driving intention of the obstacle, including: In the scenario where the obstacle drives along the line for a long time or straddles the line, if the determined driving intention of the obstacle is a lane - change intention, then force - correct the lane - change intention of the obstacle to a straight - driving intention; or In the scenario where the obstacle has completed a lane - change or has just turned at an intersection, if the determined driving intention of the obstacle is a lane - change intention, then force - correct the lane - change intention of the obstacle to a straight - driving intention; or In the scenario where the obstacle merges with other vehicles on the on - ramp, if the determined driving intention of the obstacle is a straight - driving intention, then adjust the straight - driving intention of the obstacle to a lane - change intention; or In the highway scenario, when there is a low - speed obstacle or a stationary obstacle in front of the obstacle, if the determined driving intention of the obstacle is a straight - driving intention, then urgently adjust the straight - driving intention of the obstacle to a lane - change intention; or In the scenario of approaching an intersection, when the obstacle needs to cut in and change lanes, and the obstacle makes a low - speed and large - angle lane - change, if the determined driving intention of the obstacle is a straight - driving intention, then urgently adjust the straight - driving intention of the obstacle to a lane - change intention; or When the lane - change intention of the obstacle meets the set number of frames, determine that the driving intention of the obstacle is a lane - change intention.

7. The method for identifying obstacle intention according to any one of claims 2 to 5, characterized in that The method further includes: establishing multiple calibration tables in the calibration library in the following manner: Obtain the operating condition parameters and intention judgment parameters in the historical cruise scenario within a set time period; Create multiple calibration tables based on the operating condition parameters and intention judgment parameters; Store the multiple calibration tables into the calibration library to obtain multiple calibration tables in the calibration library.

8. The obstacle intention recognition method according to claim 7, wherein The operating condition parameters include: obstacle type, obstacle speed, and the relative speed and Euclidean distance between the obstacle and the vehicle itself; the intention judgment parameters include: single - frame parameters and multi - frame parameters; The creating of multiple calibration tables based on the operating condition parameters and the intention judgment parameters includes: Divide the relative speed between the ego vehicle and the obstacle into N segments, where N is a natural number; According to the obstacle type, based on the relative speed of the N segments after division, with the obstacle speed and Euclidean distance as basic parameters, respectively construct calibration tables for the single - frame parameters and multi - frame parameters in the intention judgment parameters.

9. An obstacle intention recognition device, characterized in that, Including: An acquisition module, configured to acquire the operating condition parameters of the current frame in the ego vehicle cruise scenario in real time; A first determination module, configured to determine the values of multiple intention judgment parameters based on the operating condition parameters of the current frame; A second determination module, configured to determine the driving intention of the obstacle based on the values of the multiple intention judgment parameters; A post - processing module, configured to post - process the determined driving intention of the obstacle based on the engineering requirements in the ego vehicle cruise scenario.

10. An electronic device, characterized in that, Including: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the instructions to implement the obstacle intention recognition method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the obstacle intention recognition method according to any one of claims 1 to 8.

Images