A vehicle lateral trajectory based obstacle risk speed limiting method

Abstract

The application discloses a kind of barrier risk speed limit method based on vehicle lateral trajectory, through calculating self-vehicle map speed limit or artificial speed limit information, take the small value in map speed limit or consider speed limit as the final target speed limit value, generate the target speed limit speed curve of self-vehicle based on this target speed limit value;Through the method that lateral trajectory planning and speed planning decouple of space-time trajectory planning, according to the predicted curve of barrier and planning real-time lateral trajectory, the speed limit of barrier is calculated;According to the lateral distance and barrier speed obtained in the foregoing steps, the speed limit value to barrier at k time is calculated;Judge whether self-vehicle needs to carry out speed limit deceleration, and calculate speed limit deceleration, regenerate speed limit speed curve.The application can obtain speed limit value according to the lateral position information of barrier predicted line, and jointly map, artificial speed limit curve of self-vehicle, quickly calculate the risk speed limit deceleration of barrier.

Description

Technical Field

[0001] This invention relates to a method for limiting speed due to obstacle risk based on the lateral trajectory of a vehicle, belonging to the field of autonomous driving technology for unmanned vehicles in ports. Background Technology

[0002] Autonomous driving technology is a product of the deep integration of the automotive industry with artificial intelligence, the Internet of Things, high-performance computing, and next-generation information technology. It is a key direction for the intelligent development of current automotive mobility and a commanding height of automotive technology in various countries. Autonomous technology is the product of the fusion of various technologies, including perception, fusion, localization, planning, and control. These technologies work together to achieve a safe, comfortable, energy-efficient, and highly efficient autonomous driving system for intelligent vehicles. Planning is responsible for generating the vehicle's real-time trajectory. This trajectory typically includes real-time lateral trajectory and longitudinal speed planning. Speed ​​planning generates the vehicle's ideal speed and acceleration information based on the current state of the vehicle and obstacles. During speed planning, if the vehicle's real-time lateral trajectory has a collision relationship with an obstacle's trajectory, the vehicle can yield normally. However, for obstacles whose predicted trajectories do not have a collision relationship with the vehicle's real-time lateral trajectory, risk-limiting is required to make speed planning more human-like.

[0003] Current risk-based rate limiting methods are mostly based on data-driven approaches, sampling, and optimization, which are widely used in engineering. However, these sampling and data-driven methods are computationally intensive, consume a lot of CPU resources, and are not compatible with current hardware. To address this shortcoming, this invention discloses a risk-based rate limiting method that requires less computation while still meeting security requirements. Summary of the Invention

[0004] The technical problem to be solved by this invention is to overcome the shortcomings of the prior art and provide a risk-based speed limiting method based on the lateral trajectory of a vehicle. Based on the real-time lateral trajectory of the vehicle and the predicted trajectory of obstacles, risk-based speed limiting is achieved for obstacles where there is no collision relationship between the obstacle and the lateral trajectory of the vehicle.

[0005] To achieve the above objectives, the specific technical solution of this invention is as follows: A method for obstacle risk speed limiting based on vehicle lateral trajectory, which, based on the real-time lateral trajectory of the vehicle and the predicted trajectory of the obstacle, achieves obstacle risk speed limiting for situations where there is no collision relationship between the obstacle and the vehicle's lateral trajectory, including the following steps.

[0006] Step 1: Calculate the vehicle's map speed limit or human-mandated speed limit information, take the smaller value between the map speed limit and the human-mandated speed limit as the final target speed limit value, and generate the vehicle's target speed limit curve based on this target speed limit value.

[0007] Step 2: Using a decoupling method of spatiotemporal trajectory planning, lateral trajectory planning, and velocity planning, calculate the speed limit of the obstacle based on the predicted curve of the obstacle and the planned real-time lateral trajectory; for each obstacle, obtain its predicted trajectory from 0 to 8 seconds, which includes the obstacle's position information; calculate the obstacle's position at time k and the lateral distance of the vehicle's trajectory.

[0008] Step 3: Based on the lateral distance and obstacle speed obtained in the previous steps, where the obstacle speed is the obstacle information provided by the perception module, calculate the speed limit value for the obstacle at time k.

[0009] Step 4: Repeat steps 2 and 3 to obtain all speed limit values ​​of obstacles from 0 to 8 seconds.

[0010] Step 5: Determine whether the vehicle needs to reduce speed by limiting its speed, calculate the speed limit deceleration, and regenerate the speed limit curve.

[0011] Furthermore, in step 1, the map speed limit is obtained through map interface data, and the manual speed limit is obtained through the maximum vehicle speed set on the visual display screen.

[0012] Furthermore, in step 1, the map elements contain speed limit information. The speed limit information is extracted from the map data, including the speed limit values ​​for each road segment. Combined with the speed limit information set by the driver or the system, all the information is integrated to form a speed limit curve that changes with time or location.

[0013] Furthermore, the specific content of step 1 is as follows:

[0014] The smaller of the map-based speed limit or the manually set speed limit is taken as the final target speed limit, denoted as v. gui v init Given the current speed of the vehicle, a soft For the vehicle speed to reach the speed limit v gui acceleration or deceleration, t gui For the vehicle speed according to a soft The acceleration or deceleration reaches v gui Time required: t gui =(v gui -v init ) / a soft ;

[0015] S gui For v gui The speed limit curve has a time dimension of 8 seconds, and the speed limit v is shown at each time point on the curve. n Calculate according to the following formula:

[0016]

[0017] By following the steps above, we can obtain the speed limit v corresponding to each time t from 0 to 8 seconds. t If we set the position s at time 0 to 0, then the position s at each time point can be calculated using the following formula:

[0018]

[0019] Furthermore, in step 2, the obstacle's position information includes x, y, theta, l, w, and v, where x and y are the obstacle's position coordinates in the Cartesian coordinate system, theta is the angle in the Cartesian coordinate system, l is the obstacle's length, w is the obstacle's width, and v is the obstacle's velocity.

[0020] Furthermore, step 2 includes the following specific details:

[0021] Step 21: The obstacle prediction curve duration is set to 8 seconds. The obstacle prediction curve includes position information and prediction time t. The bounding box of the obstacle is obtained through its position coordinates, length, and width, and its bounding box at time t is denoted as B. t ;

[0022] Step 22, based on the rectangle B of the obstacle at time t. t Calculate the lateral position of the obstacle at time t, such as Figure 3 As shown, the calculation method is not protected by this invention, and various calculation methods can be used;

[0023] Step 23, based on the rectangle B of the obstacle at time t. t Calculate the speed limit of obstacles; the projected speed of the obstacle on the vehicle's trajectory can be obtained via v. obj_lon =v*cos(delta) theta Obtained based on horizontal distance l dis_t Calculate the speed limit increment v using a table lookup and empirical method. delta_t The speed limit imposed on the vehicle by the obstacle at time t can be calculated using the following formula:

[0024] v limit_t =v obj_lon +v delta_t

[0025] At this point, the speed limit range s of the obstacle is

[0026] Furthermore, in step 3, step 2 is repeated to calculate all speed limit values ​​for the obstacle prediction line from 0 to 8 seconds, with a sampling interval of 1 second; the dataset is as follows:

[0027] {[s lowet_0 s upper_0 vlimit_0 ],[s lowet_1 s upper_1 v limit_1 ],…[s lowe s upper_8 v limit_8 ]}.

[0028] Furthermore, the specific steps of step 4 are as follows:

[0029] Step 41, count all l values ​​within the range of 0-8s. dis_t , where l dis_t B at time t t The nearest lateral distance to the vehicle's trajectory box; if l dis_t If the minimum distance is greater than 3 meters, the obstacle can be ignored and the speed limit does not need to be triggered.

[0030] Step 42, for the speed limit curve S calculated in Step 1 gui Sampling is performed at 1-second intervals to calculate the vehicle's position and speed limit information [S]. gui_t v gui_t Using the same time k, obtain [S] gui_k v gui_k ] and [s lower_k ,s upper_k ,v limit_k If condition s is met lower_k = gui_k <=s upper_k And v gui_k >v limit_k If k = m, it means that the obstacle needs to trigger the risk speed limit. Based on this, we can judge all sampling times within the range of 0-8s. If this condition is met, then the risk speed limit needs to be triggered at this time k. Let this time be m, and the data after time m does not need to be judged. If this condition is not met at any sampling time, then the obstacle does not need to trigger the risk speed limit.

[0031] Step 43, calculate the risk speed limit deceleration; based on step 42, the time when the risk speed limit is triggered is m, and the next sampling time is denoted as n. Then, between times m and n, the vehicle and the obstacle reach the same speed and position, and the time when the obstacle and the vehicle reach the same speed is denoted as t. r The deceleration a of the vehicle r The following formula is given:

[0032]

[0033] Equation (3) is obtained by simplifying equations (1) and (2).

[0034] A*t r 2 +B*t​r +C=0 (3)

[0035] in

[0036]

[0037] Formula (3) is about t r The two linear equations in two variables can be solved to find t. r ; Remember v r For t r The vehicle speed corresponding to the current moment;

[0038]

[0039] a r =v r / t r (5)

[0040] The required deceleration a can be obtained by solving formulas (4) and (5). r .

[0041] Furthermore, in step 5, the time dimension of the obstacle speed limit curve is 8 seconds, and the speed limit v at each time point in the curve is... e_t Calculate according to the following formula:

[0042]

[0043] By following the steps above, we can obtain the speed limit v corresponding to each time t from 0 to 8 seconds. e_t .

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0045] This invention can obtain speed limit values ​​based on the lateral position information of obstacle prediction lines, and, in conjunction with the vehicle's map and manually set speed limit curves, quickly calculate the risk speed limit deceleration of obstacles. It has the advantages of low resource consumption, small computational load, and meeting safety requirements. Attached Figure Description

[0046] Figure 1 This is a flowchart of the present invention.

[0047] Figure 2 This is a diagram showing the positional relationship of obstacles at time t in the invention.

[0048] Figure 3 This is a diagram showing the positional relationship of the obstacle's real-time lateral trajectory of the vehicle in this invention. Detailed Implementation

[0049] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings:

[0050] This embodiment proposes a risk-based speed limiting method based on the lateral trajectory of a vehicle, the process of which is as follows: Figure 1 As shown, it includes the following steps:

[0051] Step 1: Generate a speed curve based on the speed limit set by the map or by human intervention.

[0052] The smaller of the map-based speed limit or the manually set speed limit is taken as the final target speed limit. The map-based speed limit can be obtained from the map's interface data, while the manually set speed limit is the maximum speed set through a visual display. Based on this target speed limit, a target speed curve for the vehicle is generated, denoted as v. gui v init Given the current speed of the vehicle, a soft For the vehicle speed to reach the speed limit v gui acceleration or deceleration, t gui For the vehicle speed according to a soft The acceleration and deceleration reach v gui Time required: t gui =(v gui -v init ) / a soft ;

[0053] S gui For v gui The speed limit curve has a time dimension of 8 seconds, and the speed limit v is shown at each time point on the curve. n Calculate according to the following formula:

[0054]

[0055] By following the steps above, we can obtain the speed limit v corresponding to each time t from 0 to 8 seconds. t If we set the position s at time 0 to 0, then the position s at each time step can be calculated using the following formula:

[0056]

[0057] Step 2: Calculate the speed limit of the obstacle based on the predicted curve of the obstacle and the lateral trajectory of the vehicle.

[0058] Step 21: The obstacle prediction curve duration is set to 8 seconds. The obstacle prediction curve includes position information x, y, theta, l, w, and v, where x and y are the obstacle's position coordinates in the Cartesian coordinate system, theta is the angle in the Cartesian coordinate system, l is the obstacle's length, w is the obstacle's width, and v is the obstacle's velocity. The prediction time t is also used to obtain the obstacle's bounding box based on its position coordinates, length, and width, and the bounding box at time t is denoted as B.t ,like Figure 2 As shown.

[0059] Step 22, based on the rectangle B of the obstacle at time t. t Calculate the lateral position of the obstacle at time t, such as Figure 3 As shown, various methods can be used, which are not protected by this invention; obstacle B t The positional relationship with the real-time lateral trajectory of the vehicle is as follows: Figure 3 As shown, where s lower_t For obstacle B t The position of the vehicle corresponding to the front of the vehicle aligned with the rectangular frame of its trajectory, s upper_t For obstacle B t The position of the vehicle corresponding to the alignment of its front and rear within the rectangle of its trajectory, delta. theata Let the theta at time t be the obstacle and the s on the lateral trajectory of the vehicle. lower_t The difference between the trajectory theta of the vehicle and the value of l dis_t B at time t t The closest lateral distance to the vehicle's trajectory box.

[0060] Step 23, based on the rectangle B of the obstacle at time t. t Calculate the speed limit of obstacles; the projected speed of the obstacle on the vehicle's trajectory can be obtained via v. obj_lon =v*cos(delta) theta Obtained based on horizontal distance l dis_t Calculate the speed limit increment v using a table lookup and empirical method. delta_t The speed limit imposed on the vehicle by the obstacle at time t can be calculated using the following formula:

[0061] v limit_t =v obj_lon +v delta_t

[0062] At this point, the speed limit range s of the obstacle is

[0063] Step 3: Repeat step 2 to calculate all speed limit values ​​for the obstacle prediction line from 0 to 8 seconds, with a sampling interval of 1 second; the dataset is as follows:

[0064] {[s lowet_0 s upper_0 v limit_0 ],[s lower_1 s upper_1 v limit_1 ],…[s lowe s upper_8 V vlimit_8 ]}.

[0065] Step 4: Calculate whether the obstacle speed limit is triggered and calculate the speed limit deceleration.

[0066] Step 41, count all l values ​​within the range of 0-8s. dis_t , where l dis_t B at time t t The nearest lateral distance to the vehicle's trajectory box; if l dis_t If the minimum value is greater than 3 meters, the obstacle can be ignored and speed limit does not need to be triggered. The 3-meter value is an empirical value and is not protected by this invention.

[0067] Step 42, for the speed limit curve S calculated in Step 1 gui Sampling is performed at 1-second intervals to calculate the vehicle's position and speed limit information [S]. gui_t v gui_t Using the same time k, obtain [S] gui_k v gui_k ] and [s lower_k s uppre_k v limit_k If condition s is met lower_k = gui_k <=s upper_k And v gui_k >v limit_k If k = m, it means that the obstacle needs to trigger the risk speed limit. Based on this, we can judge all sampling times within the range of 0-8s. If this condition is met, then the risk speed limit needs to be triggered at this time k. Let this time be m, and the data after time m does not need to be judged. If this condition is not met at any sampling time, then the obstacle does not need to trigger the risk speed limit.

[0068] Step 43, calculate the risk speed limit deceleration; based on step 42, the time when the risk speed limit is triggered is m, and the next sampling time is denoted as n. Then, between times m and n, the vehicle and the obstacle reach the same speed and position, and the time when the obstacle and the vehicle reach the same speed is denoted as t. r The deceleration a of the vehicle r The following formula is given:

[0069]

[0070] Equation (3) is obtained by simplifying equations (1) and (2).

[0071] A*t r 2 +B*t r +C=0 (3)

[0072] in

[0073] ​

[0074] Formula (3) is about t r The two linear equations in two variables can be solved to find t. r ; Remember v r For t r The vehicle speed corresponding to the current moment;

[0075]

[0076] a r =v r / t r (5)

[0077] The required deceleration a can be obtained by solving formulas (4) and (5). r .

[0078] Step 5: Calculate the velocity curve based on the deceleration obtained in Step 4.

[0079] The obstacle speed limit curve has a time dimension of 8 seconds, and the speed limit v is at each time point in the curve. e_t Calculate according to the following formula:

[0080]

[0081] By following the steps above, we can obtain the speed limit v corresponding to each time t from 0 to 8 seconds. e_t .

[0082] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the specific embodiments described above. The specific embodiments and descriptions in the specification are merely for further illustrating the principles of the invention. Various changes and modifications can be made to the present invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the claims and their equivalents.

Claims

1. A method for limiting speed due to obstacle risk based on the lateral trajectory of a vehicle, characterized in that: Based on the vehicle's real-time lateral trajectory and the predicted trajectory of obstacles, a risk-based speed limit is implemented for obstacles where there is no collision relationship between the obstacle and the vehicle's lateral trajectory. This includes the following steps. Step 1: Calculate the vehicle's map speed limit or manual speed limit information, take the smaller value between the map speed limit and the manual speed limit as the final target speed limit value, and generate the vehicle's target speed limit curve based on this target speed limit value. Step 2: Using a decoupling method of spatiotemporal trajectory planning, lateral trajectory planning, and velocity planning, calculate the speed limit of the obstacle based on the predicted curve of the obstacle and the planned real-time lateral trajectory; for each obstacle, obtain its predicted trajectory from 0 to 8 seconds, which includes the obstacle's position information; calculate the obstacle's position at time k and the lateral distance of the vehicle's trajectory. Step 3: Based on the lateral distance and obstacle speed obtained in the previous steps, where the obstacle speed is the obstacle information provided by the perception module, calculate the speed limit value for the obstacle at time k. Step 4: Repeat steps 2 and 3 to obtain all speed limit values ​​of obstacles from 0 to 8 seconds. The specific steps are as follows: Step 41, count all values ​​within the range of 0-8 seconds. ,in At time t The nearest lateral distance to the vehicle's trajectory box; if If the minimum distance is greater than 3 meters, the obstacle can be ignored and the speed limit does not need to be triggered. Step 42, Apply the speed limit curve calculated in Step 1. Sampling is performed at 1-second intervals to calculate the vehicle's position and speed limit information. Using the same time k, obtain [ ]and[ If the conditions are met. =< <= and > This indicates that the obstacle needs to trigger the risk speed limit; This is used to determine all sampling times within the range of 0-8s. If this condition is met, then the risk speed limit needs to be triggered at this time k. This time is denoted as m, and data after time m does not need to be judged. If this condition is not met at any sampling time, then the obstacle does not need to trigger the risk speed limit. Step 43, calculate the risk speed limit deceleration; based on step 42, the time when the risk speed limit is triggered is denoted as m, and the next sampling time is denoted as n. Then, between times m and n, the vehicle and the obstacle reach the same speed and position. The time when the obstacle and the vehicle reach the same speed is denoted as... The deceleration of the vehicle The following formula is given: (1) (2) Equation (3) is obtained by simplifying equations (1) and (2). (3) in , , C = ， Formula (3) is about The two linear equations in two variables can be solved. ;remember for The vehicle speed corresponding to the current moment; (4) (5) The required deceleration can be obtained by solving formulas (4) and (5). ; Step 5: Determine whether the vehicle needs to reduce speed by limiting its speed, calculate the speed limit deceleration, and regenerate the speed limit curve.

2. The obstacle risk speed limiting method based on vehicle lateral trajectory according to claim 1, characterized in that: In step 1, the map speed limit is obtained through map interface data, and the manual speed limit is obtained through the maximum vehicle speed set on the visual display screen.

3. The obstacle risk speed limiting method based on vehicle lateral trajectory according to claim 1, characterized in that: In step 1, the map elements contain speed limit information. The speed limit information is extracted from the map data, including the speed limit values ​​for each road segment. Combined with the speed limit information set by the driver or the system, all the information is integrated to form a speed limit curve that changes with time or location.

4. The obstacle risk speed limiting method based on vehicle lateral trajectory according to claim 3, characterized in that: The specific content of step 1 is as follows: The smaller value of the map speed limit or the manually set speed limit is taken as the final target speed limit, denoted as... , The current speed of the vehicle. To ensure the vehicle reaches the speed limit acceleration or deceleration For the vehicle speed according to The acceleration or deceleration reaches Time required: ; for The speed limit curve has a time dimension of 8 seconds, and the speed limit is displayed at each time point on the curve. Calculate according to the following formula: ； By following the steps above, we can obtain the speed limit corresponding to each time t from 0 to 8 seconds. If we set the position s at time 0 to 0, then the position s at each time point can be calculated using the following formula: .

5. The obstacle risk speed limiting method based on vehicle lateral trajectory according to claim 1, characterized in that: In step 2, the position information of the obstacle includes x, y, theta, l, w, and v, where x and y are the position coordinates of the obstacle in the Cartesian coordinate system, theta is the angle in the Cartesian coordinate system, l is the length of the obstacle, w is the width of the obstacle, and v is the speed of the obstacle.

6. The obstacle risk speed limiting method based on vehicle lateral trajectory according to claim 5, characterized in that: Step 2 includes the following specific contents. Step 21: The obstacle prediction curve duration is set to 8 seconds. The obstacle prediction curve includes position information and prediction time t. The bounding box of the obstacle is obtained using its position coordinates, length, and width, and its bounding box at time t is recorded as... ; Step 22, based on the rectangle of the obstacle at time t Calculate the lateral position of the obstacle at time t; Step 23, based on the rectangle of the obstacle at time t Calculate the speed limit of obstacles; The projected speed of the obstacle on the vehicle's trajectory can be obtained through Obtain; based on horizontal distance Speed ​​limit increments are calculated using a table-lookup, empirical method. The speed limit imposed on the vehicle by the obstacle at time t can be calculated using the following formula: At this point, the speed limit range s of the obstacle is [ ].

7. The obstacle risk speed limiting method based on vehicle lateral trajectory according to claim 1, characterized in that: In step 3, step 2 is repeated to calculate all speed limit values ​​for the obstacle prediction line from 0 to 8 seconds, with a sampling interval of 1 second; the dataset is as follows: {[ ],[ ], …[ ]}。 8. The obstacle risk speed limiting method based on vehicle lateral trajectory according to claim 1, characterized in that: In step 5, the time dimension of the obstacle speed limit curve is 8 seconds, and the speed limit is defined for each time interval on the curve. Calculate according to the following formula: , By following the steps above, the speed limit corresponding to each time t from 0 to 8 seconds can be obtained. .

Images