A longitudinal decision planning method for autonomous driving safely through a busbar

Abstract

The application provides a longitudinal decision planning method for automatically driving through a busbar safely, comprising the following steps: S1, determining whether the current automatic driving mode meets the opening of a busbar decision planning function, and if yes, opening the busbar decision planning function and entering a standby condition determination state; S2, receiving a necessary condition signal based on visual positioning, navigation positioning and map information, determining whether the function activation is established, and if yes, opening a speed and acceleration planning calculation state for safe passing; and S3, calculating different accelerations corresponding to different distance intervals, and outputting the minimum value of the calculated target acceleration to a control module to complete vehicle acceleration control. According to the actual demand of automatic driving, the planning method is improved, so that the acceleration planning algorithm can better adapt to the upgrade iteration of the automatic driving function, and the platform adaptability and safety performance of the function are improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of automobile electrical appliances, and particularly relates to a specific scene automatic driving longitudinal decision planning method, which is mainly applicable to the application of the automatic driving function in the dangerous scene of passing through a highway merging lane. BACKGROUND

[0002] In recent years, the automatic driving function has attracted much attention from the outside world, especially from major automobile manufacturers. The basic principle of automatic driving acceleration planning is mainly based on real-time road information and vehicle surrounding target information to plan real-time target acceleration, so as to ensure that the vehicle can complete automatic driving functions such as following driving and cruising driving on the actual road.

[0003] When passing through a highway merging port, the prior art has an automatic driving control method for vehicle merging, which introduces part of the road map signal into the decision planning process. In the case that the vehicle is in the merging area and is about to pass through the exit or entrance of the road, the relative position, speed and other parameters of the vehicle and other vehicles are calculated, and it is calculated whether the vehicle will collide with other vehicles under the condition of maintaining the current speed of the vehicle, and corresponding measures are taken based on the calculation result, so that the vehicle can safely pass through the merging area.

[0004] However, the existing method has the limitation that the relationship information between the road merging lane and the vehicle motion is not fully considered for advance acceleration planning to improve the safety and driving confidence of passing through the merging lane. SUMMARY

[0005] The present application provides a longitudinal decision planning method for automatic driving safe passing through a merging lane, which mainly solves the problem of target acceleration planning by only considering the relative relationship between the merging road intersection vehicle and the vehicle, and reasonably plans the target speed and acceleration decision in advance when passing through the merging lane.

[0006] The technical scheme of the present application is as follows:

[0007] A longitudinal decision planning method for automatic driving safe passing through a merging lane, comprising:

[0008] S1, whether the current automatic driving mode meets the opening of the merging lane decision planning function, if yes, the merging lane decision planning function is opened, and the standby condition judgment state is entered.

[0009] S2, based on visual positioning, navigation positioning and map information, receiving necessary condition signals including whether there is a merging lane, the type of the merging lane, the type of the lane where the vehicle is located, the distance between the vehicle and the front merging lane, and the speed of the vehicle, etc. information, determining whether the function activation is established, if yes, the speed and acceleration planning calculation state of safe passing is started.

[0010] S3, calculate different deceleration corresponding to different distance interval, and output the minimum value of the target acceleration to the control module to complete the vehicle acceleration control.

[0011] The distance here refers to the actual distance from the vehicle to the junction.

[0012] The minimum value refers to the deceleration corresponding to the junction distance, compared with the deceleration value calculated under other scenarios in the current running state, who is the minimum is output. For example: the junction deceleration is-0.5 m / s² according to the calculation, at this time there are other vehicles in front of the vehicle cutting in, the deceleration calculated under this working condition is-0.2 m / s², and the road speed limit gives the deceleration 0 m / s², then the final output is the minimum value-0.5 m / s²

[0013] Further, the conditions and logic for determining whether the function in S2 is activated are:

[0014] Condition one: simultaneously satisfy: ① left junction in front, ② vehicle is in the leftmost lane of the main road, ③ vehicle distance from junction distance≤set distance, ④ vehicle speed≥set speed.

[0015] Condition two: simultaneously satisfy: ⑥ right junction in front, ⑦ vehicle is in the rightmost lane of the main road, ③ vehicle distance from junction distance≤set distance, ④ vehicle speed≥set speed.

[0016] When conditions one and two are not established, the function is not activated, and the input signal period determination is continued.

[0017] When conditions one and two are established, it is determined that the external signal is incorrect, and the risk warning information is output to prompt the automatic driving system to make overall decision.

[0018] When one of conditions one and two is established, it is determined that the function activation condition is established, and the speed and acceleration planning calculation state of safe passing is started.

[0019] Further, the S3 calculates different deceleration corresponding to different distance is divided into:

[0020] ① When the distance is between B and A, the vehicle adopts the speed reduction strategy of reducing to X;

[0021] ② When the distance is between C and B, the vehicle adopts the speed reduction strategy of reducing to Y;

[0022] ③ When the distance is between D and C, the vehicle adopts the speed reduction strategy of reducing to Z;

[0023] Wherein, A>B>C>D, and the distance difference between A, B, C, D is above 80m, X>Y>Z, and the speed difference between X, Y, Z is above 10 km / h.

[0024] From the above technical solutions, the method is aimed at the deficiencies and defects in the existing converging road intersection passing acceleration planning algorithm. In the normal driving working condition of automatic driving, the lane position information (overtaking lane, middle lane, right lane, etc.) of the vehicle, the front road information (such as converging distance, left converging / right converging), etc. are introduced. According to the actual road conditions (slope, speed limit, etc.) and the actual driving speed of the vehicle, comprehensive planning and calculation are carried out to obtain an optimal acceleration planning value for automatic driving control target. According to the actual needs of automatic driving, the planning method is improved, so that the acceleration planning algorithm can better adapt to the upgrade iteration of automatic driving function, and the platform adaptability and safety performance of the function are improved. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 . Automatic driving control architecture diagram;

[0026] Figure 2 . Automatic driving acceleration planning diagram proposed by the method;

[0027] Figure 3 . Actual converging road scene diagram. DETAILED DESCRIPTION

[0028] The technology of the present application will be further described in detail below in combination with the drawings:

[0029] As shown in Figure 1 , it is the automatic driving function architecture diagram mentioned in the present application. In the figure, the acceleration planning is an important module in the automatic driving controller, which is responsible for completing the target acceleration calculation of the vehicle during operation.

[0030] As shown in Figure 2 , it is the specific implementation method of the volume converging road longitudinal decision planning method in the present application

[0031] Step 1, determine whether the current automatic driving mode meets the opening of the function. If the function only has ACC function, the converging road decision planning function is not opened. If IACC / HWA is opened, the converging road decision planning function is opened, and the standby condition judgment state is entered.

[0032] Step 2, based on visual positioning, navigation positioning and map information, the system starts to receive necessary condition signals at any time, including whether there is a converging road, the type of converging road (left and right direction), the type of lane where the vehicle is located (rightmost, center, leftmost), the distance between the vehicle and the front converging road, and the vehicle speed, etc. Information is logically judged.

[0033] Condition one: simultaneously satisfy: ① left merging lane in front, ② the vehicle is in the leftmost lane of the main road, ③ the vehicle is within 300m from the merging lane, ④ the vehicle speed is greater than or equal to 75km / h.

[0034] Condition two: simultaneously satisfy: ⑥ right merging lane in front, ⑦ the vehicle is in the rightmost lane of the main road, ③ the vehicle is within 300m from the merging lane, ④ the vehicle speed is greater than or equal to 75km / h.

[0035] When neither condition one nor two is established, the function activation is not established, and the system continues to maintain the input signal periodic determination.

[0036] When both conditions one and two are established, it is determined that the external signal is incorrect, and the risk warning information is output to prompt the automatic driving system to make an overall decision.

[0037] When one of conditions one and two is established, it is determined that the function activation condition is established, and the speed and acceleration planning calculation state for safe passing is started.

[0038] Step 3, according to the overall decision-making idea, different distance intervals correspond to different deceleration planning strategies. When entering the merging lane safe passing interval distance, the following condition determination is entered:

[0039] ① When the distance is between 200m and 300m, the vehicle adopts a deceleration strategy to reduce the speed to 105km / h. The deceleration calculation strategy is calculated based on the vehicle kinematics in the absolute coordinate system, and the output deceleration range is constrained between -0.5m / s² and 0m / s² to ensure comfort. If the vehicle speed is less than or equal to 105km / h, the current speed strategy is maintained, and the output acceleration is 0m / s².

[0040] ② When the distance is between 100m and 200m, the vehicle adopts a deceleration strategy to reduce the speed to 90km / h. The deceleration calculation strategy is calculated based on the vehicle kinematics in the absolute coordinate system, and the output deceleration range is constrained between -0.8m / s² and 0m / s² to ensure comfort. If the vehicle speed is less than or equal to 90km / h, the current speed strategy is maintained, and the output acceleration is 0m / s².

[0041] ③ When the distance is between -10m and 100m, the vehicle adopts a deceleration strategy to reduce the speed to 75km / h. The deceleration calculation strategy is calculated based on the vehicle kinematics in the absolute coordinate system, and the output deceleration range is constrained between -1m / s² and 0m / s² to ensure comfort. If the vehicle speed is less than or equal to 90km / h, the current speed strategy is maintained, and the output acceleration is 0m / s².

[0042] The final planned target acceleration is output to the control module as the minimum value of ①②③, completing the vehicle acceleration control and achieving the safe passing merging lane strategy.

[0043] In addition, the speed reduction strategy is then released, and the vehicle is no longer constrained by the function and is in a free cruise state

[0044] The present application scheme is aimed at the deficiencies and defects in the existing merging road intersection passing acceleration planning algorithm, and the planning method is improved according to the actual demand of automatic driving, so that the acceleration planning algorithm can better adapt to the upgrade iteration of the automatic driving function, and the platform adaptability and safety performance of the function are improved.

Claims

1. A longitudinal decision planning method for safe autonomous driving through a busbar, characterized in that, Comprise: S1, determine whether the current automatic driving mode meets the opening of the junction decision planning function, if it meets, open the junction decision planning function, enter the standby condition determination state; S2, based on visual positioning, navigation positioning and map information, receive the necessary condition signal, including whether there is a junction, the type of junction, the type of lane where the vehicle is located, the distance between the vehicle and the front junction and the speed of the vehicle, determine whether the function activation is established, if yes, open the speed and acceleration planning calculation state of safe passing; The conditions and logic of determining whether the function activation is established are: Condition one: simultaneously meet: left junction in front, vehicle is in the leftmost lane of the main road, vehicle distance from junction is less than or equal to the set distance, vehicle speed is greater than or equal to the set speed; Condition two: simultaneously meet: right junction in front, vehicle is in the rightmost lane of the main road, vehicle distance from junction is less than or equal to the set distance, vehicle speed is greater than or equal to the set speed; When conditions one and two are not established, the function activation is not established, continue to maintain the periodic determination of input signal; When conditions one and two are established, it is determined that the external signal is wrong, output risk warning information, prompt the automatic driving system to make overall decision; When one of conditions one and two is established, it is determined that the function activation condition is established, then open the speed and acceleration planning calculation state of safe passing; S3, calculate different accelerations corresponding to different distance intervals, and output the minimum value of the target acceleration calculated to the control module to complete the vehicle acceleration control; S3 calculates different accelerations corresponding to different distance intervals: ① When the distance is between B and A, the vehicle adopts the speed reduction strategy of reducing to X; ② When the distance is between C and B, the vehicle adopts the speed reduction strategy of reducing to Y; ③ When the distance is between D and C, the vehicle adopts the speed reduction strategy of reducing to Z; Wherein, A > B > C > D, and the distance difference between A, B, C and D is more than 80m, X > Y > Z, and the speed difference between X, Y and Z is more than 10km / h; The deceleration is calculated according to the kinematics relationship of the vehicle in the absolute coordinate system; The final planned target acceleration is output to the control module as the minimum value in ①②③ to complete the vehicle acceleration control.

2. The method for longitudinal decision planning of an autonomous vehicle to safely pass a merge according to claim 1, wherein : When the distance is between B and A, the output deceleration range is restricted between -0.8m / s² and 0m / s², if the vehicle speed is less than or equal to X, the current speed strategy is maintained, and the output acceleration is 0m / s²; When the distance is between C and B, the output deceleration range is restricted between -0.8m / s² and 0m / s², if the vehicle speed is less than or equal to Y, the current speed strategy is maintained, and the output acceleration is 0m / s²; When the distance is between D and C, the output deceleration range is restricted between -1m / s² and 0m / s², if the vehicle speed is less than or equal to Z, the current speed strategy is maintained, and the output acceleration is 0m / s².

3. The method of claim 1 or 2, wherein the method further comprises: determining a longitudinal decision plan for the autonomous vehicle to safely pass the merging traffic based on the merging traffic information and the autonomous vehicle information. In S1, it is necessary to determine whether the current automatic driving mode meets the opening of the junction decision planning function, if the function is only ACC function, the junction decision planning function is not opened, if IACC / HWA is opened, the junction decision planning function is opened.

4. The method of claim 2, wherein: When the safe passing through the busbar strategy has been reached, the speed reduction passing strategy is released, and the vehicle is no longer subject to the function constraints, and is in a free cruise state.

Images