Travel trajectory generation method and device, storage medium and electronic equipment

Abstract

The present disclosure relates to a driving track generation method and device, a storage medium and an electronic device, and relates to the technical field of vehicles. The method comprises: predicting a first driving track of a first vehicle on a current ramp according to first historical driving data of the first vehicle on a first historical ramp and first real-time driving data of the first vehicle on the current ramp; in the case that a second vehicle exists within a preset range of the first vehicle, predicting a second driving track of the second vehicle on the current ramp according to second historical driving data of the second vehicle on a second historical ramp and second real-time driving data of the second vehicle on the current ramp; and displaying the first driving track and the second driving track. The driving track generation method provided by the present disclosure can assist the driver in driving through the ramp and improve the driving safety of the driver.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of vehicles, in particular to a driving trajectory generation method and device, a storage medium and an electronic device. BACKGROUND

[0002] When a vehicle drives near a ramp area of a highway, since the road near the ramp is relatively complex and various roads converge near the ramp, the driver needs to receive traffic sign information, and also needs to consider left and right vehicle information and road conditions to comprehensively operate the vehicle, such as acceleration, deceleration, and lane changing.

[0003] However, if the remaining vehicles near the ramp do not drive according to traffic regulations, suddenly change lanes, overtake, or reverse, it is easy to cause a rear-end accident, which greatly reduces the driving safety of the current vehicle. SUMMARY

[0004] The purpose of the present disclosure is to provide a driving trajectory generation method and device, a storage medium and an electronic device to solve the above technical problems.

[0005] To achieve the above purpose, a first aspect of an embodiment of the present disclosure provides a driving trajectory generation method, which comprises:

[0006] According to the first historical driving data of the first vehicle on the first historical ramp and the first real-time driving data of the first vehicle on the current ramp, the first driving trajectory of the first vehicle on the current ramp is predicted;

[0007] In the case that there is a second vehicle within a preset range of the first vehicle, according to the second historical driving data of the second vehicle on the second historical ramp and the second real-time driving data of the second vehicle on the current ramp, the second driving trajectory of the second vehicle on the current ramp is predicted;

[0008] The first driving trajectory and the second driving trajectory are displayed.

[0009] Optionally, in the case that there is a second vehicle within a preset range of the first vehicle, according to the second historical driving data of the second vehicle on the second historical ramp and the second real-time driving data of the second vehicle on the current ramp, the second driving trajectory of the second vehicle on the current ramp is predicted, comprising:

[0010] In the case that there is the second vehicle within the preset range of the first vehicle, the number of the second vehicles is determined;

[0011] in response to the number of the second vehicles being less than or equal to a preset number, predicting the second driving trajectory of the second vehicle according to the second historical driving data and the second real-time driving data of the second vehicle by using a first model.

[0012] Optionally, the method further comprises:

[0013] in response to the number of the second vehicles being greater than the preset number, predicting the second driving trajectory of the first target vehicle according to the second historical driving data and the second real-time driving data of the first target vehicle by using the first model, and predicting the second driving trajectory of the second target vehicle according to the second historical driving data and the second real-time driving data of the second target vehicle by using a second model;

[0014] the first target vehicle is a second vehicle closest to the first vehicle among the second vehicles, and the second target vehicle is a second vehicle other than the first target vehicle among the second vehicles.

[0015] Optionally, the first historical driving data comprises first historical lane line features and first historical driving information, and the first real-time driving data comprises current lane line features and first real-time driving information; and the predicting the first driving trajectory of the first vehicle on the current ramp according to the first historical driving data of the first vehicle on the first historical ramp and the first real-time driving data of the first vehicle on the current ramp comprises:

[0016] predicting the first driving trajectory of the first vehicle on the current ramp according to the first historical lane line features of the first historical ramp, the first historical driving information of the first vehicle driving on the first historical ramp, the current lane line features of the current ramp, and the first real-time driving information of the first vehicle driving on the current ramp.

[0017] Optionally, the second historical driving data comprises second historical lane line features and second historical driving information, and the second real-time driving data comprises current lane line features and second real-time driving information; and the predicting the second driving trajectory of the second vehicle on the current ramp according to the second historical driving data of the second vehicle on the second historical ramp and the second real-time driving data of the second vehicle on the current ramp comprises:

[0018] predicting the second driving trajectory of the second vehicle on the current ramp according to the second historical lane line features of the second historical ramp, the second historical driving information of the second vehicle driving on the second historical ramp, the current lane line features of the current ramp, and the second real-time driving information of the second vehicle driving on the current ramp.

[0019] Optionally, the method further comprises:

[0020] determining a collision risk value between the first vehicle and the second vehicle;

[0021] in response to the collision risk value of the first vehicle and the second vehicle being greater than a preset risk value, determining the second vehicle greater than the preset risk value as a dangerous vehicle;

[0022] in a case where a distance between the first vehicle and the dangerous vehicle, a distance between the first vehicle and a lane line, and an acceleration of the first vehicle satisfy a constraint condition, controlling a steering module and a braking module according to first real-time driving data of the first vehicle corresponding to a target acceleration mean value and a target acceleration variance.

[0023] Optionally, the determining the collision risk value between the first vehicle and the second vehicle comprises:

[0024] determining a target basic event occurred by the first vehicle according to the first real-time driving data of the first vehicle and self-vehicle information of the first vehicle;

[0025] determining a structural importance degree of the target basic event according to an accident tree;

[0026] determining the collision risk value according to the structural importance degree.

[0027] According to a second aspect of an embodiment of the present disclosure, a driving trajectory generation device is provided, and the device comprises:

[0028] a first driving trajectory prediction module configured to predict a first driving trajectory of a first vehicle on a current ramp according to first historical driving data of the first vehicle on a first historical ramp and first real-time driving data of the first vehicle on the current ramp;

[0029] a second driving trajectory prediction module configured to, in a case where a second vehicle exists within a preset range of the first vehicle, predict a second driving trajectory of the second vehicle on the current ramp according to second historical driving data of the second vehicle on a second historical ramp and second real-time driving data of the second vehicle on the current ramp;

[0030] a display module configured to display the first driving trajectory and the second driving trajectory.

[0031] According to a third aspect of an embodiment of the present disclosure, a non-transitory computer readable storage medium is provided, and the medium stores a computer program, which, when executed by a processor, implements steps of the driving trajectory generation method provided by the first aspect of the present disclosure.

[0032] According to a fourth aspect of an embodiment of the present disclosure, an electronic device is provided, and the device comprises:

[0033] a memory having stored thereon a computer program;

[0034] a processor configured to execute the computer program in the memory to implement steps of the method for generating a driving trajectory according to the first aspect of the present disclosure.

[0035] The method for generating a driving trajectory according to the present disclosure can predict a first driving trajectory of a first vehicle on a current ramp through first historical driving data and first real-time driving data of the first vehicle, predict a second driving trajectory of a second vehicle on the current ramp through second historical driving data and second real-time driving data of the second vehicle, and display the first driving trajectory and the second driving trajectory to a driver of the first vehicle, thereby assisting the driver to adjust the speed and direction of the first vehicle according to the first driving trajectory of the first vehicle and the second driving trajectory of the second vehicle, to avoid collision with the second vehicle, and to improve the driving safety and driving safety of the driver.

[0036] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0037] The accompanying drawings are included to provide a further understanding of the present disclosure and constitute a part of the specification, which together with the following detailed description, serve to explain the present disclosure. In the drawings:

[0038] Figure 1 is a step flow chart of the method for generating a driving trajectory according to an example embodiment of the present disclosure.

[0039] Figure 2 is a schematic diagram of displaying a first driving trajectory and a second driving trajectory according to an example embodiment of the present disclosure.

[0040] Figure 3 is a schematic diagram of an accident occurrence tree according to an example embodiment of the present disclosure.

[0041] Figure 4 is a logic diagram of the method for generating a driving trajectory according to an example embodiment of the present disclosure.

[0042] Figure 5 is a block diagram of a driving trajectory generation device according to an example embodiment of the present disclosure.

[0043] Figure 6 is a block diagram of an electronic device according to an example embodiment of the present disclosure.

[0044] Figure 7 is a block diagram of an electronic device according to an example embodiment of the present disclosure. DETAILED DESCRIPTION

[0045] The specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure, and are not used to limit the present disclosure.

[0046] It should be noted that all actions of obtaining signals, information or data in the present disclosure are carried out in accordance with the corresponding data protection regulations and policies of the country where the device is located, and with the authorization of the owner of the corresponding device.

[0047] Please refer to Figure 1 As shown in the drawings, the present disclosure provides a driving trajectory generation method, which comprises the following steps:

[0048] In step S11, the first driving trajectory of the first vehicle on the current ramp is predicted according to the first historical driving data of the first vehicle on the first historical ramp and the first real-time driving data of the first vehicle on the current ramp.

[0049] In this step, when the first vehicle is at a first preset distance from the current ramp, the first historical driving data of the first vehicle driving on the past first historical ramp and the first real-time driving data of the first vehicle around the current ramp can be obtained, and the first driving trajectory of the first vehicle in the future first preset time length is predicted according to the first historical driving data and the first real-time driving data.

[0050] Wherein, the first preset distance can be 20m, 30m, etc., which is not limited by the present disclosure; the first preset time length can be 6s, 10s, etc., which is not limited by the present disclosure.

[0051] The first vehicle can be a current vehicle, and the first vehicle can obtain first historical driving data from a cloud platform through a V2X (Vehicle to Everything) technology. The first historical driving data is historical driving data of the first vehicle driving on a past ramp, and the first historical driving data includes first historical lane line features of the first historical ramp and first historical driving information of the first vehicle driving on the first historical ramp. The first historical lane line features are used to represent environmental information on the first historical ramp, and the first historical lane line features include lane width, lane number, boundary position of the lane, linearity, color, speed limit, lane center line, curvature, and slope near the first historical ramp. The first historical driving information is used to represent information of the first vehicle driving on the first historical ramp in the past, and the first historical driving information includes lateral and longitudinal distances between the first vehicle and the lane center line, lateral and longitudinal distances between the first vehicle and the lane edge, a geometric position of the first vehicle driving on the lane, a driving speed of the first vehicle, and driving acceleration of the first vehicle, and the driving acceleration includes lateral acceleration and longitudinal acceleration.

[0052] The first real-time driving data is driving data of the first vehicle on a current ramp, and the first real-time driving data includes current lane line features and first real-time driving information. The current lane line features refer to environmental information on the current lane, and the current lane line features have the same data types as the first historical lane line features. The current lane line features include lane width, lane number, boundary position of the lane, linearity, color, speed limit, lane center line, curvature, and slope near the current lane. The first real-time driving information refers to driving information generated by the first vehicle driving on the current ramp, and the first real-time driving information has the same data types as the first historical driving information. The first real-time driving information includes lateral and longitudinal distances between the first vehicle and the lane center line, lateral and longitudinal distances between the first vehicle and the lane edge, a geometric position of the first vehicle driving on the lane, a driving speed of the first vehicle, and driving acceleration of the first vehicle.

[0053] In this step, since the first real-time driving data is obtained, the ramp features of the current ramp where the first vehicle is located and the first driving information of the first vehicle driving on the current ramp are recorded in the first real-time driving data. Therefore, the first vehicle can be predicted to drive at multiple positions near the current ramp through a model, and the multiple positions are connected into a line to obtain a first driving trajectory of the first vehicle driving on the current ramp within a first preset time in the future.

[0054] In this process, the first real-time driving data reflects the current ramp road environment and the current position change of the first vehicle, but when predicting the first driving trajectory of the first vehicle in the future, only the first real-time driving data of the first vehicle in the current driving condition is used for prediction, and the accuracy of the obtained first driving trajectory is insufficient.

[0055] For example, when the first vehicle is driving on the current ramp, if the first driving trajectory is predicted by the first real-time driving data, the obtained first driving trajectory may be that the first vehicle drives along the left lane after passing the ramp, but according to the past driving habits of the driver of the first vehicle, the first vehicle will continue to drive along the middle lane after passing the ramp, resulting in insufficient accuracy of the predicted first driving trajectory.

[0056] In order to improve the accuracy of predicting the first driving trajectory, the present disclosure introduces first historical driving data, which contains first historical driving information of the driver of the first vehicle driving on a plurality of first historical ramps in the past, reflecting the driving behavior habits of the driver of the first vehicle, so that the first historical driving data and the first real-time driving data are used as inputs of the model, which can more accurately predict the first driving trajectory.

[0057] In step S12, in the case that there is a second vehicle in the preset range of the first vehicle, the second driving trajectory of the second vehicle on the current ramp is predicted according to the second historical driving data of the second vehicle on the second historical ramp and the second real-time driving data of the second vehicle on the current ramp.

[0058] In this step, when the sensor device such as radar or camera on the first vehicle determines that there is a second vehicle in the preset range of the first vehicle, the second driving trajectory of the second vehicle in the future second preset time period can be predicted according to the second historical driving data and the second real-time driving data.

[0059] Wherein, the preset range refers to a circular area formed with the first vehicle as the center and the second preset distance as the radius, and the second preset distance can be 6m, 7m, 8m, 10m, etc., which is not limited in the present disclosure; the second preset time period can be equal to or different from the first preset time period.

[0060] Wherein, since the cloud platform stores historical driving data of a plurality of vehicles on different roads, the first vehicle can also obtain the second historical driving data of the second vehicle from the cloud platform through V2X technology; since the first vehicle and the second vehicle are interconnected, the first vehicle can also obtain the second real-time driving data of the second vehicle; and the second driving trajectory of the second vehicle in the future second preset time period is predicted according to the second real-time driving data and the second historical driving data.

[0061] The second historical driving data is historical driving data of the second vehicle driving on a historical ramp in the past, and the data type of the second historical driving data is the same as that of the first historical driving data. The second real-time driving data is real-time data of the first vehicle driving on the current ramp, and the data type of the second real-time driving data is the same as that of the first real-time driving data.

[0062] In this step, the second historical driving data and the second real-time driving data are combined as the input of the model, and a more accurate second driving trajectory can be obtained according to the same principle as step S12.

[0063] In step S13, the first driving trajectory and the second driving trajectory are displayed.

[0064] In this step, after obtaining the first driving trajectory and the second driving trajectory, the first driving trajectory of the vehicle and the second driving trajectory of the remaining vehicles can be displayed in the form of a guide line on the HMI (Human Machine Interface) interface of the first vehicle. For example, please refer to FIG. 2. Figure 2 As shown in FIG. 2, there are a left lane and a middle lane on the ramp, and the remaining vehicles drive on the left lane, and the vehicle drives on the middle lane. According to the driving trajectory generation method provided by the present disclosure, the first driving trajectory of the vehicle driving on the middle lane and the second driving trajectory of the remaining vehicles driving on the left lane can be predicted.

[0065] According to the driving trajectory generation method provided by the present disclosure, the first driving trajectory of the first vehicle on the current ramp can be predicted by the first historical driving data and the first real-time driving data of the first vehicle. The second driving trajectory of the second vehicle on the current ramp can be predicted by the second historical driving data and the second real-time driving data of the second vehicle. The first driving trajectory and the second driving trajectory are presented to the driver of the first vehicle, so as to assist the driver to adjust the speed and direction of the first vehicle according to the first driving trajectory of the vehicle and the second driving trajectory of the remaining vehicles, avoid collision with the remaining vehicles, and improve the driving safety of the driver.

[0066] In the present disclosure, the first driving trajectory of the first vehicle can be predicted by using a first model, for example, a data-driven model (Long Short-Term Memory-Convolutional Neural Network-Short-Term Memory, LSTM-CNN-LSTM model). The prediction of the data-driven model is relatively accurate, and the first historical driving data with the driving habits of the driver and the first real-time driving data reflecting the real-time operation of the driver can be combined to obtain a first driving trajectory with high accuracy on the current ramp in the initial driving. In the case where the accuracy of the obtained first driving trajectory is high, the driver can avoid the second vehicle according to the first driving trajectory, thereby improving the driving safety.

[0067] When predicting the second driving trajectory and the first driving trajectory, different models can be used to predict different driving trajectories, specifically including the following steps:

[0068] In step S21, when the second vehicle exists in the preset range of the first vehicle, the number of the second vehicles is determined.

[0069] In this step, whether the second vehicle exists in the preset range of the first vehicle can be detected by using a sensor device such as a radar or a camera. When the second vehicle exists, the number of the second vehicles is determined.

[0070] In step S22, in response to the number of the second vehicles being less than or equal to a preset number, the first model is used to predict the second driving trajectory according to the second historical driving data and the second real-time driving data.

[0071] When the number of the second vehicles is less than or equal to the preset number, the first model can be used to predict the second driving trajectory, so that the obtained second driving trajectory of the second vehicle is more accurate.

[0072] The preset number can be 2 or 3, which is set according to specific conditions, and the present disclosure does not limit it.

[0073] In step S23, in response to the number of the second vehicles being greater than the preset number, the first model is used to predict the second driving trajectory of the first target vehicle according to the second historical driving data and the second real-time driving data of the first target vehicle; and the second model is used to predict the second driving trajectory of the second target vehicle according to the second historical driving data and the second real-time driving data of the second target vehicle. The first target vehicle is the second vehicle closest to the first vehicle among the plurality of second vehicles, and the second target vehicle is the second vehicle excluding the first target vehicle among the plurality of second vehicles.

[0074] The prediction accuracy of the first model is higher than that of the second model, for example, the second model can be a Markov prediction model, but the prediction data received by the first model is more than that received by the second model, the computing power of the first model is more occupied, and the prediction speed of the predicted driving trajectory is relatively slow.

[0075] Therefore, in the case that the number of second vehicles is greater than the preset number, if the second driving trajectories of all second vehicles are predicted using the first model, the speed of predicting the second driving trajectory will be slow, and the second driving trajectory cannot be displayed in time. In order to balance the prediction accuracy and prediction speed of the second driving trajectory.

[0076] In the case that the number of second vehicles is greater than the preset number, from the plurality of second vehicles, the first model with higher accuracy is used to predict the second driving trajectory of the first target vehicle closest to the first vehicle, so that the driver of the first vehicle can avoid the first target vehicle with higher risk according to the second driving trajectory of the first target vehicle, and improve the driving safety of the driver of the first vehicle; from the plurality of second vehicles, the second model with faster prediction speed is used to predict the second driving trajectory of the second target vehicle except the first target vehicle, so that the second driving trajectory of the second target vehicle can be displayed in time.

[0077] In steps S21 to S23, in the process of predicting the first driving trajectory and the second driving trajectory by using the first model, the first driving trajectory and the second driving trajectory can be predicted by the following method.

[0078] In the process of predicting the first driving trajectory of the first vehicle by using the first model, the first historical driving data and the first real-time driving data of the first vehicle are normalized and preprocessed to obtain a high-dimensional feature vector, and then the high-dimensional feature vector is input into a feature embedding layer to map the high-dimensional feature vector into a low-dimensional dense feature vector. Then, the low-dimensional feature vector is input into the first model as a time series model, and a mean square error function is used as a loss function of the first model to update the first model, so that the first model outputs the first driving trajectory of the first vehicle in the future first preset time length.

[0079] In the first model is used to predict the second driving track of the first target vehicle, the second historical driving data and the second real-time driving data are normalized and pretreated to obtain a high-dimensional feature vector, and then the high-dimensional feature vector is input into the feature embedding layer to map the high-dimensional feature vector into a low-dimensional dense feature vector; the low-dimensional feature vector is input into the first model as a time series model, and a mean square error function is used as the loss function of the first model to update the first model, so that the first model outputs the second driving track of the second vehicle in the future second preset time length.

[0080] In the second model is used to predict the second driving track of the second target vehicle, the decimeter-level positioning information of the second target vehicle and the second real-time driving information of the second target vehicle can be obtained first. Since the second real-time driving information includes the lateral and longitudinal distances of the second target vehicle from the center line of the lane, the lateral and longitudinal distances of the second target vehicle from the edge of the lane, the geometric position of the second target vehicle in the lane, the driving speed of the second target vehicle, and the driving acceleration (including lateral acceleration and longitudinal acceleration) of the second target vehicle, the driving intention of the driver of the second target vehicle can be determined according to the positioning information and the real-time information of the second target vehicle. The driving intention includes the driving intention of the driver to drive straight along the current lane, to turn left from the current lane to the left lane, to turn right from the current lane to the right lane, etc. After determining the driving intention of the driver, the target lane to be reached by the second target vehicle can be determined in combination with the real-time driving information of the second target vehicle. Then, according to the driving acceleration in the second real-time driving information, the longitudinal distance of the second target vehicle to reach the target lane within the second preset time length is determined to obtain the termination position of the second target vehicle on the target lane after driving for the second preset time length. Finally, an iterative algorithm is used to generate an optimal action sequence of the second target vehicle from the current position to the termination position, and the optimal action sequence generates the second driving track of the second target vehicle.

[0081] The optimal action sequence includes a plurality of position points of the second target vehicle moving smoothly from the current position to the termination position. The positioning information of the second target vehicle is different from the geometric position of the second target vehicle in the lane. The positioning information refers to the position of the second target vehicle in the world coordinate system, and the geometric position refers to the position of the second target vehicle in the lane.

[0082] The second target vehicle decimeter-level positioning information can be obtained by using a scheme in the related art, which combines a high-definition map, a GNSS (Global Navigation Satellite System) sensor, an RTK (Real-time Kinematic) service, an ADAS (Advanced Driving Assistance System) camera, an IMU (Inertial Measurement Unit), and a vehicle CAN (Controller Area Network) signal.

[0083] The driving trajectory generation method provided in the present disclosure can use the first model to predict the first driving trajectory, so that the obtained first driving trajectory is more accurate, and the driver of the first vehicle can determine the future driving route of the vehicle according to the first driving trajectory with high accuracy, so as to more safely avoid the second vehicle. When the number of second vehicles is small, the first model is used to predict the second driving trajectory of the second vehicle, which also improves the accuracy of predicting the second driving trajectory. When the number of second vehicles is large, the first model is used to predict the second driving trajectory of the first target vehicle closest to the first vehicle, so that the driver can accurately avoid the first target vehicle closest to the vehicle according to the accurate first driving trajectory and the second driving trajectory, thereby improving the driving safety of the driver. The second model is further used to predict the second driving trajectory of the second target vehicle, which can timely display the second driving trajectory of the second target vehicle, and the second target vehicle is far away from the first vehicle, so the accuracy of the second driving trajectory displayed in time is relatively low, and it will not cause danger to the first vehicle.

[0084] In a possible implementation, the collision risk value between the first vehicle and the second vehicle can be determined based on an accident occurrence tree, and the steering module and the braking module on the first vehicle can be controlled based on the collision risk value to avoid the second vehicle. Specifically, the following steps are included:

[0085] In step S31, a target basic event occurring to the first vehicle is determined according to the first real-time driving data of the first vehicle, the self-vehicle information of the first vehicle, and the positioning information of the first vehicle.

[0086] In step S32, the structural importance of the target basic event is determined according to the accident occurrence tree.

[0087] In this step, please refer to Figure 3The structure diagram of the shown accident tree, in which vehicle collision is taken as the top event T of the accident tree, the defect events constituting the top event T are composed of intermediate events and basic events.

[0088] According to the analysis of the factors affecting the risk of vehicle collision, the intermediate events are as follows:

[0089] A1: dynamic characteristics of the model vehicle; A2: initial spacing error of the vehicle; A3: system delay; A4: initial speed of the vehicle; A5: human factors; A6: emergency avoidance speed; A7: delay.

[0090] These intermediate events are further subdivided into various basic events, and the various basic events are as follows:

[0091] x1: vehicle body length; x2: vehicle body width; x3: vehicle yaw acceleration; x4: insufficient navigation accuracy; x5: execution error; x6: path planning algorithm failure; x7: network transmission delay; x8: mechanical failure; x9: decision error; x10: personnel reaction delay; x11: performance limitation.

[0092] Among them, the ego information of the first vehicle includes the length, width and height of the first vehicle and the yaw acceleration of the vehicle, which is used as the basis for judging whether the basic event X1, the basic event X2 and the basic event X3 occur. When the vehicle body length of the first vehicle exceeds the first preset safety value, the basic event X1 occurs; when the vehicle body width of the first vehicle exceeds the second preset safety value, the basic event X2 occurs; when the vehicle yaw acceleration exceeds the preset limit, the basic event X3 occurs.

[0093] Among them, the obtained positioning information can be combined with the sensors such as cameras and radars of the first vehicle to judge whether the obtained positioning information meets the requirements, as the basis for the occurrence of the basic event X4. When the positioning information does not meet the decimeter level, the basic event X4 occurs.

[0094] Whether the loss function in the first model exceeds a preset value, for example, 0.6, can be used as the basis for determining whether the basic event X6 occurs. When the specific value of the loss function exceeds 0.6, it is determined that the path planning algorithm fails.

[0095] The staff can determine whether the basic event X7, the basic event X8 and the basic event X11 occur according to the first real-time driving data of the first vehicle, and the specific judgment rules can be determined according to the working experience of the staff. The present disclosure does not make specific elaboration.

[0096] The probability distribution of the basic event X9 and the basic event X10 can be determined according to related research, and whether the basic event X9 and the basic event X10 occur can be determined according to the probability distribution.

[0097] Wherein, the logical expression of the fault tree used by the present disclosure is as follows:

[0098] T = A1*A2*A4*A6*A7 = (x1+x2+x3+A5)*(x4+x5)*(x4+x8)*(x9+x 10 11 )*(A5+A3) = (x1+x2+x3+x9+x 10 )*(x4+x5)*(x4+x8)*(x9+x 10 11 )*(x9+x 10 +x6+x7) (1)

[0099] The logical expression of the success tree of the fault tree obtained by using the dual rule of Boolean algebra is as follows:

[0100]

[0101] In formula (2), composed of 5 minimum path sets, and the structural importance of each basic event can be obtained according to the 5 minimum path sets.

[0102] The calculation formula of the structural importance of basic event x i is as follows:

[0103]

[0104] In formula (3), I Φ (i) is the structural importance, the greater the value of I Φ (i), the greater the importance of basic event x i , and x i ∈G r is that the basic event x i belongs to a minimum path set G r , and n represents the number of events contained in the minimum path set where the basic event x i is located.

[0105] The structural importance of various basic events obtained by the above formula (3) is as follows:

[0106] I Φ (1) = 0.0625; I Φ (2) = 0.0625; I Φ (3) = 0.0625; I Φ (4) = 1; I Φ (5) = 0.5; I Φ (6) = 0.25; I Φ (7) = 0.25; I Φ ​​(8) = 0.5; I Φ (9) = 0.4375; I Φ (10) = 0.4375; I Φ (11) = 0.25.

[0107] When a plurality of target basic events are determined to occur according to the first real-time driving data of the first vehicle, the self-vehicle information of the first vehicle, and the positioning information of the first vehicle, the structural importance degrees corresponding to the plurality of target basic events can be determined from the above plurality of structural importance degrees.

[0108] In step S33, the collision risk value is determined according to the structural importance degrees.

[0109] The structural importance degrees corresponding to the plurality of target basic events can be added to obtain the collision risk value between the first vehicle and the second vehicle.

[0110] For example, after determining that the basic event x1, the basic event x4, and the basic event x8 occur, the structural importance degrees I Φ (1), I Φ (4), and I Φ (8) are added to obtain a collision risk degree of 1.5625.

[0111] In step S34, in response to the collision risk value between the first vehicle and the second vehicle being greater than a preset risk value, the second vehicle greater than the preset risk value is determined as a dangerous vehicle.

[0112] The preset risk value can be 2 or other numerical values, which are not limited by the present disclosure.

[0113] When the collision risk value is greater than the preset risk value, it indicates that the collision risk between the first vehicle and the second vehicle is large, and at this time the second vehicle can be determined as a dangerous vehicle.

[0114] In step S35, in the case that the distance between the first vehicle and the dangerous vehicle, the distance between the first vehicle and the lane line, and the acceleration of the first vehicle satisfy the constraint condition, the first real-time driving data corresponding to the target acceleration mean and the target acceleration variance of the first vehicle is used to control the steering module and the braking module.

[0115] Please refer to Figure 4As shown, the calculation time step can be determined according to the speed interval of the first vehicle. Since the functional scenario applicable to the present disclosure is mainly the lane change of the ego vehicle at the ramp, it is assumed that the driving speed of the ego vehicle is 0-40 km / h. When the driving speed of the first vehicle is 0-10 km / h (including 0 km / h and 10 km / h), the calculation time step is 2 ms; when the driving speed of the first vehicle is 10-40 km / h (not including 10 km / h and including 40 km / h), the first time step is 0.5 ms.

[0116] The driving position, driving speed and heading angle of the first vehicle on the current ramp at the current time are obtained according to the first real-time driving data of the first vehicle and the ego information of the first vehicle; the driving position, driving speed and heading angle of the dangerous vehicle on the current ramp are obtained according to the second real-time driving data of the dangerous vehicle and the ego information of the dangerous vehicle (the ego information includes the length, width, height and heading angle of the dangerous vehicle).

[0117] After determining the calculation time step, the driving position, driving speed and heading angle of the first vehicle and the dangerous vehicle at the next time are determined; the distance between the first vehicle and the dangerous vehicle, the distance between the first vehicle and the lane line, and the acceleration of the first vehicle are determined according to the driving position, driving speed and heading angle of the first vehicle and the dangerous vehicle at the next time.

[0118] In the case where the distance between the first vehicle and the dangerous vehicle, the distance between the first vehicle and the lane line, and the acceleration of the first vehicle satisfy the constraint condition, the first real-time driving data of the first vehicle corresponding to the target acceleration mean and the target acceleration variance are used to control the steering module and the brake module to avoid collision between the first vehicle and the dangerous vehicle.

[0119] Specifically, in the case where the distance between the first vehicle and the dangerous vehicle is greater than the first safety distance, the distance between the first vehicle and the lane line is greater than the second safety distance, and the acceleration of the first vehicle is less than the preset acceleration, it is indicated that the first vehicle and the second vehicle satisfy the constraint condition, and at this time the first vehicle will not collide with the dangerous vehicle. In order to further protect the driving safety of the first vehicle, the first real-time driving data corresponding to the target acceleration mean and the target acceleration variance can be used, that is, the driving speed of the first vehicle corresponding to the target acceleration mean and the target acceleration variance is used as the input variable of the brake module (Integrated Power Brake, IPB), and the heading angle of the first vehicle corresponding to the target acceleration mean and the target acceleration variance is used as the input variable of the steering module (Electrical Power Steering, EPS), to control the brake module and the steering module respectively to avoid the dangerous vehicle.

[0120] The target acceleration mean value is the minimum acceleration mean value, and the target acceleration variance is the minimum acceleration variance, which are expressed by the following formulas:

[0121]

[0122] In formula (4), the comf value is as small as possible to avoid the dangerous vehicle when the first vehicle meets the safety distance; is the acceleration mean value; and σ is the acceleration variance.

[0123] As can be seen from formula (4), the comf value is smaller when the acceleration mean value and the acceleration variance are as small as possible. Therefore, the driving speed and the heading angle of the first vehicle corresponding to the comf value are used to control the braking module and the steering module to avoid the dangerous vehicle.

[0124] When the distance between the first vehicle and the dangerous vehicle is less than the first safety distance, the distance between the first vehicle and the lane line is less than the second safety distance, and the acceleration of the first vehicle is greater than the preset acceleration, it indicates that the first vehicle and the second vehicle do not meet the constraint condition. At this time, in order to ensure the driving safety of the first vehicle, the driving speed and the heading speed of the first vehicle can be corrected by using the constraint condition (the constraint condition includes the first safety distance, the second safety distance and the third safety distance), the target acceleration mean value and the target acceleration variance, so that the first vehicle and the dangerous vehicle meet the constraint condition and avoid collision.

[0125] Based on the same inventive concept, please refer to Figure 5 The present disclosure proposes a driving trajectory generation device 120, which includes a first driving trajectory prediction module 121, a second driving trajectory prediction module 122, and a display module 123.

[0126] The first driving trajectory prediction module 121 is configured to predict a first driving trajectory of a first vehicle on a current ramp according to first historical driving data of the first vehicle on a first historical ramp and first real-time driving data of the first vehicle on the current ramp;

[0127] The second driving trajectory prediction module 122 is configured to predict a second driving trajectory of a second vehicle on the current ramp according to second historical driving data of the second vehicle on a second historical ramp and second real-time driving data of the second vehicle on the current ramp when the second vehicle exists within a preset range of the first vehicle;

[0128] The display module 123 is configured to display the first driving trajectory and the second driving trajectory.

[0129] Optionally, the second travel trajectory prediction module 122 comprises:

[0130] a vehicle number determination module configured to determine a number of the second vehicles in a preset range of the first vehicle;

[0131] a first prediction module configured to, in response to the number of the second vehicles being less than or equal to a preset number, predict the second travel trajectory according to the second historical driving data and the second real-time driving data of the second vehicles by using a first model.

[0132] Optionally, the travel trajectory generation device 120 comprises:

[0133] a second prediction module configured to, in response to the number of the second vehicles being greater than the preset number, predict a second travel trajectory of a first target vehicle according to second historical driving data and second real-time driving data of the first target vehicle by using the first model, and predict a second travel trajectory of a second target vehicle according to second historical driving data and second real-time driving data of the second target vehicle by using a second model;

[0134] the first target vehicle is a second vehicle closest to the first vehicle among the second vehicles, and the second target vehicle is a second vehicle other than the first target vehicle among the second vehicles.

[0135] Optionally, the first historical driving data comprises first historical lane line features and first historical driving information, and the first real-time driving data comprises current lane line features and first real-time driving information; the first travel trajectory prediction module 121 comprises:

[0136] a third prediction module configured to predict a first travel trajectory of the first vehicle on the current ramp according to first historical lane line features of the first historical ramp, first historical driving information of the first vehicle driving on the first historical ramp, current lane line features of the current ramp, and first real-time driving information of the first vehicle driving on the current ramp.

[0137] Optionally, the second historical driving data comprises second historical lane line features and second historical driving information, and the second real-time driving data comprises current lane line features and second real-time driving information; the second travel trajectory prediction module 122 comprises:

[0138] A fourth prediction module is configured to predict a second driving trajectory of the second vehicle on the current ramp according to a second historical lane line feature of the second historical ramp, second historical driving information of the second vehicle driving on the second historical ramp, a current lane line feature of the current ramp, and second real-time driving information of the second vehicle driving on the current ramp.

[0139] Optionally, the driving trajectory generation apparatus 120 comprises:

[0140] A risk determination module is configured to determine a collision risk value between the first vehicle and the second vehicle.

[0141] A dangerous vehicle determination module is configured to determine the second vehicle as a dangerous vehicle if the collision risk value between the first vehicle and the second vehicle is greater than a preset risk value.

[0142] A control module is configured to control the steering module and the braking module according to the first real-time driving data of the first vehicle corresponding to a target acceleration mean value and a target acceleration variance if a distance between the first vehicle and the dangerous vehicle, a distance between the first vehicle and a lane line, and an acceleration of the first vehicle satisfy a constraint condition.

[0143] Optionally, the risk determination module comprises:

[0144] A target basic event determination module is configured to determine a target basic event of the first vehicle according to the first real-time driving data of the first vehicle and self-vehicle information of the first vehicle.

[0145] A structural importance determination module is configured to determine a structural importance of the target basic event according to an accident occurrence tree.

[0146] A first risk determination module is configured to determine the collision risk value according to the structural importance.

[0147] As to the apparatus in the above-mentioned embodiments, the specific manners in which various modules perform operations have been described in details in the embodiments of the method, and thus will not be described in details here.

[0148] Figure 6 is a block diagram of an electronic device 700 according to an exemplary embodiment. As shown in Figure 6 the electronic device 700 can include a processor 701 and a memory 702. The electronic device 700 can also include one or more of a multimedia component 703, an input / output (I / O) interface 704, and a communication component 705.

[0149] The processor 701 is configured to control overall operations of the electronic device 700 to complete all or part of the steps of the driving trajectory generation method described above. The memory 702 is configured to store various types of data to support operations of the electronic device 700, which can include, for example, instructions for operating any application or method on the electronic device 700, and application-related data, such as contact data, transmitted and received messages, pictures, audio, video, and the like. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage, a flash memory, a magnetic disk, or an optical disk. The multimedia component 703 can include a screen and an audio component. The screen can be, for example, a touch screen, and the audio component is configured to output and / or input audio signals. For example, the audio component can include a microphone configured to receive external audio signals. The received audio signals can be further stored in the memory 702 or transmitted through the communication component 705. The audio component further includes at least one speaker configured to output audio signals. The I / O interface 704 provides an interface between the processor 701 and other interface modules, which can be a keyboard, a mouse, a button, and the like. The buttons can be virtual buttons or physical buttons. The communication component 705 is configured to perform wired or wireless communication between the electronic device 700 and other devices. The wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, and the like, or a combination of one or more of them, is not limited herein. Therefore, the corresponding communication component 705 can include a Wi-Fi module, a Bluetooth module, an NFC module, and the like.

[0150] In an exemplary embodiment, the electronic device 700 can be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic elements for performing the above-described method of generating a travel trajectory.

[0151] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the above-described method of generating a travel trajectory. For example, the computer-readable storage medium can be the above-described memory 702 including program instructions, which can be executed by the processor 701 of the electronic device 700 to complete the above-described method of generating a travel trajectory.

[0152] Figure 7 is a block diagram of an electronic device 1900 according to an exemplary embodiment. For example, the electronic device 1900 can be provided as a server. Referring to Figure 7 , the electronic device 1900 includes a processor 1922, the number of which can be one or more, and a memory 1932 for storing a computer program executable by the processor 1922. The computer program stored in the memory 1932 can include one or more modules each corresponding to a set of instructions. In addition, the processor 1922 can be configured to execute the computer program to perform the above-described method of generating a travel trajectory.

[0153] In addition, the electronic device 1900 can further include a power supply component 1926, which can be configured to perform power management of the electronic device 1900, and a communication component 1950, which can be configured to implement communication of the electronic device 1900, for example, wired or wireless communication. In addition, the electronic device 1900 can further include an input / output (I / O) interface 1958. The electronic device 1900 can operate based on an operating system stored in the memory 1932.

[0154] In another exemplary embodiment, a computer readable storage medium including program instructions is also provided, which when executed by a processor, implement the steps of the above-described travel trajectory generation method. For example, the non-transitory computer readable storage medium can be the above-described memory 1932 including program instructions, which can be executed by the processor 1922 of the electronic device 1900 to complete the above-described travel trajectory generation method.

[0155] In another exemplary embodiment, a computer program product is also provided, which contains a computer program capable of being executed by a programmable device, the computer program having code portions for executing the above-described travel trajectory generation method when executed by the programmable device.

[0156] The preferred embodiments of the present disclosure are described in detail above with reference to the accompanying drawings, but the present disclosure is not limited to the specific details in the above-described embodiments. Various simple modifications can be made to the technical solutions of the present disclosure within the scope of the technical concept of the present disclosure, and all these simple modifications shall fall within the protection scope of the present disclosure.

[0157] In addition, it should be noted that each specific technical feature described in the above-described specific embodiments can be combined in any appropriate manner without contradiction, and in order to avoid unnecessary repetition, the present disclosure will not make further descriptions on various possible combinations.

[0158] Furthermore, any combination of the various different embodiments of the present disclosure can also be made, as long as it does not deviate from the idea of the present disclosure, and it shall be considered as the disclosed content of the present disclosure.

Claims

1. A travel trajectory generation method characterized by comprising: The method comprises: predicting a first driving track of a first vehicle on a current ramp according to first historical driving data of the first vehicle on a first historical ramp and first real-time driving data of the first vehicle on the current ramp; in a case where a second vehicle exists within a preset range of the first vehicle, predicting a second driving track of the second vehicle on the current ramp according to second historical driving data of the second vehicle on a second historical ramp and second real-time driving data of the second vehicle on the current ramp; in a case where a number of the second vehicles is greater than a preset number, the second driving tracks of the second vehicles with different distances from the first vehicle are predicted by different models; displaying the first driving track and the second driving track.

2. The method of claim 1, wherein, The method further comprises: in a case where the second vehicle exists within the preset range of the first vehicle, determining the number of the second vehicles; in response to the number of the second vehicles being less than or equal to the preset number, predicting the second driving track according to the second historical driving data and the second real-time driving data by a first model.

3. The method of claim 2, wherein, The method further comprises: in response to the number of the second vehicles being greater than the preset number, predicting a second driving track of a first target vehicle according to second historical driving data and second real-time driving data of the first target vehicle by the first model, and predicting a second driving track of a second target vehicle according to second historical driving data and second real-time driving data of the second target vehicle by a second model; the first target vehicle is a second vehicle closest to the first vehicle among the second vehicles, and the second target vehicle is a second vehicle other than the first target vehicle among the second vehicles.

4. The method of claim 1, wherein, The first historical driving data comprises first historical lane line features and first historical driving information, and the first real-time driving data comprises current lane line features and first real-time driving information; the method of predicting the first driving track of the first vehicle on the current ramp according to the first historical driving data of the first vehicle on the first historical ramp and the first real-time driving data of the first vehicle on the current ramp comprises: predicting the first driving track of the first vehicle on the current ramp according to first historical lane line features of the first historical ramp, first historical driving information of the first vehicle on the first historical ramp, current lane line features of the current ramp, and first real-time driving information of the first vehicle on the current ramp.

5. The method of claim 1, wherein, The second historical driving data comprises second historical lane line features and second historical driving information, and the second real-time driving data comprises current lane line features and second real-time driving information; the second vehicle's second driving trajectory on the current ramp is predicted according to the second vehicle's second historical driving data on the second historical ramp and the second vehicle's second real-time driving data on the current ramp, comprising: The second vehicle's second driving trajectory on the current ramp is predicted according to the second historical lane line features of the second historical ramp, the second historical driving information of the second vehicle driving on the second historical ramp, the current lane line features of the current ramp, and the second real-time driving information of the second vehicle driving on the current ramp.

6. The method of claim 1, wherein, The method further comprises: determining a collision risk value between the first vehicle and the second vehicle; in response to the collision risk value between the first vehicle and the second vehicle being greater than a preset risk value, determining the second vehicle greater than the preset risk value as a dangerous vehicle; in the case that the distance between the first vehicle and the dangerous vehicle, the distance between the first vehicle and the lane line, and the acceleration of the first vehicle satisfy the constraint condition, controlling the steering module and the braking module according to the first real-time driving data of the first vehicle corresponding to the target acceleration mean value and the target acceleration variance.

7. The method of claim 6, wherein, The determination of the collision risk value between the first vehicle and the second vehicle comprises: determining a target basic event of the first vehicle according to the first real-time driving data of the first vehicle and the ego information of the first vehicle; determining the structural importance of the target basic event according to the fault tree; determining the collision risk value according to the structural importance.

8. A travel trajectory generation device characterized by comprising: The device comprises: a first driving trajectory prediction module configured to predict the first vehicle's first driving trajectory on the current ramp according to the first vehicle's first historical driving data on the first historical ramp and the first vehicle's first real-time driving data on the current ramp; a second driving trajectory prediction module configured to, in the case that there is a second vehicle within a preset range of the first vehicle, predict the second vehicle's second driving trajectory on the current ramp according to the second vehicle's second historical driving data on the second historical ramp and the second vehicle's second real-time driving data on the current ramp; in the case that the number of the second vehicles is greater than a preset number, the second driving trajectories of the second vehicles with different distances from the first vehicle are predicted by different models; a display module configured to display the first driving trajectory and the second driving trajectory.

9. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the steps of the method of any one of claims 1-7.

10. An electronic device, comprising: comprise: a memory having a computer program stored thereon; a processor configured to execute the computer program in the memory to implement the steps of the method of any one of claims 1-7.

Images