A vehicle line-crossing prediction method based on yaw angular velocity

Abstract

The present invention relates to a method for predicting vehicle crossing the line based on yaw angular velocity, comprising the following steps: 1) obtaining vehicle speed and yaw angular velocity; 2) determining prediction time according to vehicle speed; 3) calculating predicted trajectory length according to prediction time and vehicle speed; 4) calculating lateral displacement L1 at the end of the predicted trajectory and lateral displacements L2 and L3 at the corresponding ends of the predicted trajectory on the left lane line and the right lane line according to the predicted trajectory length; 5) judging whether the vehicle is likely to cross the line according to the calculation result of step 4). The present invention determines the prediction time based on the reaction time required for the driver to take over, and then uses the yaw angular velocity that most directly represents the vehicle's motion state to predict the short-term motion trajectory of the vehicle, with strong real-time performance and accurate prediction results; judging whether the vehicle is likely to cross the line according to the intersection of the predicted trajectory and the lane lines on both sides, if so, an alarm is triggered to remind the driver to take over the vehicle control, thereby improving the safety of automatic driving and the user experience.

Description

Technical Field

[0001] The present invention belongs to the technical field of automobile intelligence, and in particular relates to a vehicle line-crossing prediction method based on yaw angular velocity. Background Art

[0002] With the rapid development of automobile intelligence, more and more cars are equipped with autonomous driving functions. At present, most scenarios of vehicle autonomous driving are aimed at lane keeping, that is, the autonomous driving system obtains environmental information around the vehicle through a variety of sensors and constructs a lane line equation based on the environmental information. The lane line equation is used to obtain the lane line trajectory on both sides of the vehicle, and the position of the vehicle is determined by the lane lines on the left and right of the lane, thereby controlling the vehicle to maintain centering in the lane. When the sensor fails or the collected environmental information is incorrect, it will affect the normal operation of the autonomous driving system, and it is easy for the vehicle to cross the line or run out of the lane, leading to traffic accidents.

[0003] In order to avoid the above situation and improve the safety of autonomous driving, it is necessary to predict in advance whether the vehicle is likely to cross the line. When there is a possibility of crossing the line, notify the driver in advance to take over the vehicle control, so as to ensure driving safety. Chinese patent CN201911065468.8 Vehicle crossing line detection method, device, mobile terminal and storage medium, this scheme judges whether the vehicle has crossed the line based on image detection, mainly for automatic judgment of road violations, and cannot be predicted in advance; Chinese patent CN201810320836.8 A vehicle crossing line violation advance warning device, this scheme uses the steering wheel angle to predict the vehicle's driving trajectory, and can predict the vehicle's crossing line to a certain extent, but because the actual movement of the vehicle lags behind the rotation of the steering wheel, this leads to a large error between the predicted result of the vehicle's driving trajectory and the actual situation, which is easy to cause false reminders and no reminders, and the reliability and practicality are poor, and it is difficult to meet the use requirements. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the technical problem to be solved by the present invention is to provide a vehicle crossing line prediction method based on yaw angular velocity, so as to solve the problem of poor reliability of the current vehicle crossing line prediction method and achieve the effect of improving the safety and user experience of autonomous driving.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A vehicle line crossing prediction method based on yaw angular velocity comprises the following steps:

[0007] 1) Get the current speed and yaw rate of the vehicle;

[0008] 2) Determine the predicted time based on vehicle speed;

[0009] 3) Calculate the predicted trajectory length based on the predicted time and vehicle speed;

[0010] 4) Calculate the lateral displacement L1 of the end of the predicted trajectory according to the vehicle speed, yaw rate and predicted trajectory length, and calculate the lateral displacements L2 and L3 of the corresponding predicted trajectory ends on the left lane line and the right lane line respectively according to the predicted trajectory length;

[0011] 5) According to the calculation result of step 4), determine whether the vehicle may cross the line. If yes, execute step 6); if not, jump to step 1);

[0012] 6) Alarm to remind the driver to take over the vehicle.

[0013] Furthermore, in step 2), the prediction time is determined according to the vehicle speed as follows: when the vehicle speed is 0-25 km / h, the prediction time is 1.4 s; when the vehicle speed is 25-90 km / h, the prediction time is inversely proportional to the vehicle speed, and the prediction time linearly changes from 1.4 s to 1.2 s; when the vehicle speed is greater than 90 km / h, the prediction time is 1.2 s.

[0014] Furthermore, in step 3), the predicted trajectory length is the product of the predicted time and the vehicle speed.

[0015] Further, in step 4), the lateral displacement L1 at the end of the predicted trajectory is calculated using the predicted trajectory equation as follows:

[0016]

[0017] Among them, y is the lateral displacement of the predicted trajectory in the vehicle coordinate system, x is the predicted trajectory length, v is the vehicle speed, and w is the yaw angular velocity.

[0018] Further, in step 4), the lateral displacements L2 and L3 on the lane line corresponding to the end of the predicted trajectory are calculated using the lane line equation as follows:

[0019] The left lane line equation:

[0020] y=a l0 +a l1 x+a l2 x 2 +a l3 x 3

[0021] Among them, y is the lateral displacement of the lane line in the vehicle coordinate system, x is the predicted trajectory length, and a l0 ,a l1 ,a l2 ,a l3 is the left lane line equation coefficient.

[0022] The right lane line equation:

[0023] y=a r0 +a r1 x+a r2 x 2 +a r3 x 3

[0024] Among them, y is the lateral displacement of the lane line in the vehicle coordinate system, x is the predicted trajectory length, and a r0 ,a r1 ,a r2 ,a r3 is the coefficient of the right lane line equation.

[0025] Furthermore, in step 5), the operation of judging whether the vehicle may cross the line is as follows: comparing L1 with L2 and L3 respectively, if L2<L1<L3, judging that the vehicle may not cross the line, if L1<L2 or L3<L1, judging that the vehicle may cross the line.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] The vehicle lane crossing prediction method based on yaw angular velocity described in the present invention is implemented based on a vehicle equipped with an automatic driving system, and is easy to implement and deploy. Starting from the reaction time required for the driver to take over, the prediction time is determined, and then the yaw angular velocity that most directly represents the vehicle's motion state is used to predict the vehicle's short-term motion trajectory, which has strong real-time performance and accurate prediction results. It is determined whether the predicted trajectory intersects with the lane lines on both sides obtained by the lane line equation. If so, the vehicle is likely to cross the line or run out of the lane if it continues to travel in the current state. At this time, an alarm is sounded to remind the driver to take over vehicle control, thereby improving the safety of automatic driving and the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A flowchart of a vehicle crossing line prediction method based on yaw angular velocity according to an embodiment;

[0029] Figure 2 A schematic diagram of a vehicle curve line crossing scenario described in the embodiment; DETAILED DESCRIPTION

[0030] The specific implementation modes of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0031] Example:

[0032] See also Figure 1 , a vehicle line-crossing prediction method based on yaw angular velocity comprises the following steps:

[0033] 1) Obtain the current vehicle speed and yaw rate; in this embodiment, the vehicle speed is provided by a vehicle speed sensor, and the yaw rate is obtained from a chassis-related controller and transmitted to the ADAS controller via the vehicle bus.

[0034] 2) Determine the prediction time based on the vehicle speed. For the driver to take over in an emergency, the most important thing is to give the driver enough time to react. This requires that the predicted time of crossing the line must be greater than a certain driver reaction time to be of practical significance. On the other hand, as the prediction time becomes longer, the accuracy of the prediction result will decrease. With the real-time active control adjustment of the autonomous driving vehicle, a prediction time that is too long will be meaningless.

[0035] Therefore, the present invention starts with the prediction time and selects a reasonable prediction time to determine the length of the short-term motion trajectory of the vehicle; in this embodiment, the prediction time is determined according to the vehicle speed as follows: when the vehicle speed is 0-25 km / h, the prediction time is 1.4s; when the vehicle speed is 25-90 km / h, the prediction time is inversely proportional to the vehicle speed, and the prediction time changes linearly from 1.4s to 1.2s; when the vehicle speed is greater than 90 km / h, the prediction time is 1.2s;

[0036] In this way, the predicted trajectory is longer when driving at high speed, and shorter when driving at low speed, which is in line with the safety assessment expectations. In addition, when selecting the prediction time, it can also be adjusted according to the situation outside the lane. When there are obstacles such as guardrails and other vehicles outside the lane, the prediction time can be appropriately extended to give the driver more sufficient reaction time. There is no restriction here.

[0037] 3) Calculate the predicted trajectory length according to the predicted time and vehicle speed; in this embodiment, the trajectory length refers to the length in the direction of the vehicle's driving speed, which is obtained by directly multiplying the vehicle speed and the predicted time.

[0038] 4) Calculate the lateral displacement L1 of the end of the predicted trajectory according to the vehicle speed, yaw rate and predicted trajectory length, and calculate the lateral displacements L2 and L3 of the corresponding predicted trajectory ends on the left lane line and the right lane line respectively according to the predicted trajectory length;

[0039] In this embodiment, the lateral displacement at the end of the predicted trajectory is calculated using the predicted trajectory equation as follows:

[0040]

[0041] Where y is the lateral displacement of the predicted trajectory in the vehicle coordinate system, x is the length of the predicted trajectory, v is the vehicle speed, and w is the yaw angular velocity;

[0042] The derivation process of the predicted trajectory equation is as follows:

[0043] Known displacement formula:

[0044]

[0045] Since the coordinate system of the present invention is a vehicle coordinate system, the initial velocity can be offset during the movement. Therefore, in the vehicle coordinate system, formula ① changes as follows:

[0046]

[0047] Applying the displacement formula to the lateral motion of the vehicle yields:

[0048]

[0049] Where y is the lateral displacement of the vehicle, a y is the lateral acceleration of the vehicle;

[0050] There is also a formula for the lateral acceleration of the vehicle:

[0051] a y =v×w ④

[0052] Where, v is the vehicle speed, w is the yaw rate;

[0053] Combining ③④, we get:

[0054]

[0055] In this embodiment, the lane line equation is used to calculate the lateral displacements L2 and L3 on the lane line corresponding to the end of the predicted trajectory as follows:

[0056] The left lane line equation:

[0057] y=a l0 +a l1 x+a l2 x 2 +a l3 x 3

[0058] Among them, y is the lateral displacement of the lane line in the vehicle coordinate system, x is the predicted trajectory length, and a l0 ,a l1 ,a l2 ,a l3 is the left lane line equation coefficient.

[0059] The right lane line equation:

[0060] y=a r0 +a r1 x+a r2 x 2 +a r3 x 3

[0061] Among them, y is the lateral displacement of the lane line in the vehicle coordinate system, x is the predicted trajectory length, and a r0 ,a r1 ,a r2 ,a r3 is the coefficient of the right lane line equation.

[0062] 5) According to the calculation result of step 4), determine whether the vehicle may cross the line. If yes, execute step 6); if not, jump to step 1);

[0063] For vehicle crossing the curve, please refer to Figure 2 As shown, if the position of the end of the predicted trajectory is not between the left and right lane lines, then there must be an intersection between the predicted trajectory and the lane line. If the vehicle continues to drive in the current state, it is very likely to cross the line and run out of the lane; correspondingly, in this embodiment, the operation of judging whether the vehicle is likely to cross the line is as follows: compare L1 with L2 and L3 respectively. If L2<L1<L3, it is judged that the vehicle is not likely to cross the line. If L1<L2 or L3<L1, it is judged that the vehicle is likely to cross the line.

[0064] 6) Alarm to remind the driver to take over the vehicle.

[0065] The vehicle lane crossing prediction method based on yaw angular velocity described in the present invention is implemented based on a vehicle equipped with an automatic driving system, and is easy to implement and deploy. Starting from the reaction time required for the driver to take over, the prediction time is determined, and then the yaw angular velocity that most directly represents the vehicle's motion state is used to predict the vehicle's short-term motion trajectory, which has strong real-time performance and accurate prediction results. It is determined whether the predicted trajectory intersects with the lane lines on both sides obtained by the lane line equation. If so, the vehicle is likely to cross the line or run out of the lane if it continues to travel in the current state. At this time, an alarm is sounded to remind the driver to take over vehicle control, thereby improving the safety of automatic driving and the user experience.

[0066] The method described in the present invention is implemented in software. The corresponding software is recommended to be deployed inside the controller of ADAS (generally referring to advanced driver assistance system, i.e., automatic driving system). The vehicle speed sensor transmits the vehicle speed to the ADAS controller through the vehicle bus. The yaw angular velocity is obtained from the chassis-related controller and transmitted to the ADAS controller through the vehicle bus. The lane line equation is an existing technology in the field of automatic driving. The corresponding input parameters can be adjusted according to the lane line equations adopted by different manufacturers. When it is determined that there is a possibility of crossing the line, the alarm system of the automatic driving system is used to issue an alarm prompt to remind the driver to take over the vehicle control and ensure the safe driving of the vehicle.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A vehicle crossing line prediction method based on yaw angular velocity, characterized in that: The steps include: 1) Get the current speed and yaw rate of the vehicle; 2) Determine the predicted time based on vehicle speed; 3) Calculate the predicted trajectory length based on the predicted time and vehicle speed; 4) Calculate the lateral displacement L1 of the end of the predicted trajectory according to the vehicle speed, yaw rate and predicted trajectory length, and calculate the lateral displacements L2 and L3 of the corresponding predicted trajectory ends on the left lane line and the right lane line respectively according to the predicted trajectory length; 5) According to the calculation result of step 4), determine whether the vehicle is likely to cross the line. If yes, execute step 6); if not, jump to step 1); 6) Alarm to remind the driver to take over the vehicle; In step 2), the prediction time is determined according to the vehicle speed as follows: when the vehicle speed is 0-25 km / h, the prediction time is 1.4 s; when the vehicle speed is 25-90 km / h, the prediction time is inversely proportional to the vehicle speed, and the prediction time changes linearly from 1.4 s to 1.2 s; when the vehicle speed is greater than 90 km / h, the prediction time is 1.2 s.

2. The vehicle crossing line prediction method based on yaw angular velocity according to claim 1, characterized in that: In step 3), the predicted trajectory length is the product of the predicted time and the vehicle speed.

3. The vehicle crossing line prediction method based on yaw angular velocity according to claim 2, characterized in that: In step 4), the lateral displacement L1 at the end of the predicted trajectory is calculated using the predicted trajectory equation as follows: Among them, y is the lateral displacement of the predicted trajectory in the vehicle coordinate system, x is the predicted trajectory length, v is the vehicle speed, and w is the yaw angular velocity.

4. The vehicle crossing line prediction method based on yaw angular velocity according to claim 1, characterized in that: In step 4), the lateral displacements L2 and L3 on the lane line corresponding to the end of the predicted trajectory are calculated using the lane line equation as follows: The left lane line equation: y=a l0 +a l1 x+a l2 x 2 +a l3 x 3 Among them, y is the lateral displacement of the lane line in the vehicle coordinate system, x is the predicted trajectory length, and a l0 ,a l1 ,a l2 ,a l3 is the left lane line equation coefficient; The right lane line equation: y=a r0 +a r1 x+a r2 x 2 +a r3 x 3 Among them, y is the lateral displacement of the lane line in the vehicle coordinate system, x is the predicted trajectory length, and a r0 ,a r1 ,a r2 ,a r3 is the coefficient of the right lane line equation.

5. The vehicle crossing line prediction method based on yaw angular velocity according to claim 1, characterized in that: In step 5), the operation of judging whether the vehicle may cross the line is as follows: compare L1 with L2 and L3 respectively. If L2<L1<L3, it is judged that the vehicle may not cross the line. If L1<L2 or L3<L1, it is judged that the vehicle may cross the line.

Images