An intelligent automobile automatic driving method

Abstract

The application discloses an automatic driving method of an intelligent automobile, and comprises the following steps: selecting a target vehicle based on a traffic intersection where a host vehicle is located, determining whether a collision occurs according to whether there is a space-time overlap, if it is determined that a collision will occur, then determining whether potential collision can be avoided by speed control without changing a path, and if it is not feasible, then re-performing path planning and speed planning. The application can avoid potential collision at the intersection, improve the safety of the intelligent automobile passing through the intersection, and ensure certain traffic efficiency and riding comfort.

Description

Technical Field

[0001] This invention belongs to the field of intelligent vehicle autonomous driving decision control technology, and specifically relates to an intelligent vehicle autonomous driving method. Background Technology

[0002] In recent years, the "electrification, intelligentization, connectivity, and sharing" of automobiles have received widespread attention worldwide, leading to rapid technological development and being seen as a significant strategic opportunity for the development of China's automotive industry. Autonomous driving has become the main development direction for intelligentization and connectivity in the current automotive field.

[0003] Crossing intersections is a crucial and typical scenario for current research in autonomous driving. In complex and dynamic urban intersections, conflicts inevitably arise between different traffic participants, either temporally or spatially. Numerous collision warning systems have emerged to address this, employing various sensors, algorithms, and control strategies to identify potential collision risks and take measures to mitigate or avoid collisions.

[0004] Existing collision warning systems can only handle relatively simple scenarios and perform poorly in complex situations such as intersections without traffic lights. Furthermore, most collision warning systems use risk values ​​as the basis for their decisions, but the level of risk cannot fully represent whether a collision will occur. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention proposes an intelligent vehicle autonomous driving method. This method aims to detect potential collisions at intersections in complex traffic conditions without traffic lights, and to plan and control the vehicle's speed and path, thereby avoiding potential collisions and improving the safety of intelligent vehicles passing through intersections.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0007] The present invention provides a method for intelligent vehicle autonomous driving, characterized by comprising the following steps:

[0008] Step 1: Based on the traffic intersection where the main vehicle is located, establish a target area centered on the intersection, select target vehicles within the target area, and determine the conflict area based on the intersection driving paths of the main vehicle and target vehicles, while also considering the control accuracy of the main vehicle path following algorithm.

[0009] Step 2: Predict the time when the main vehicle and other vehicles that will have a spatial conflict with the main vehicle enter and leave the conflict area, and calculate whether there is a time overlap. If there is an overlap, it is determined that a collision will occur; otherwise, it is determined that a collision will not occur.

[0010] Step 3: If a collision is determined to occur, calculate whether there is a feasible speed solution for safe obstacle avoidance without changing the path. If there is, control the speed of the main vehicle to avoid the potential collision; otherwise, replan the driving path and speed of the main vehicle at the intersection.

[0011] If it is determined that no collision will occur, the main vehicle will continue along the original intersection travel path.

[0012] The intelligent vehicle autonomous driving method of the present invention is also characterized in that the conflict area in step 1 is determined by the following steps:

[0013] Step 1.1: Divide a rectangular area centered on the intersection and set it as the target area. The target area includes the intersection area, a section before entering the intersection, and a section after leaving the intersection. Take any vertex of the rectangular area as the origin o, and the two sides connected to the origin o are the x-axis and y-axis, respectively, to establish a rectangular coordinate system oxy.

[0014] The master vehicle considers all other vehicles in all lanes within the target area that are heading towards or at the intersection as target vehicles, and treats each target vehicle and the master vehicle as a two-dimensional directed bounding box. The size of the two-dimensional directed bounding box is the minimum envelope rectangle of the projection of all points on the vehicle onto the ground when the vehicle is stationary. Let the number of target vehicles be denoted as n.

[0015] Step 1.2: When the main vehicle enters the target area, based on the main vehicle's starting position and target position, preliminarily plan the main vehicle's intersection driving route and driving speed;

[0016] Step 1.3: Based on the intersection driving path and driving speed planned by the main vehicle, the maximum tracking error of the path tracking obtained by the path tracking algorithm is calculated by using the fitting curve δ=f(flag,c,v) of the behavior pattern flag, path parameter c and driving speed v. Then, combined with the two-dimensional directed bounding box of the main vehicle and the intersection driving path, the future driving area Ω0 of the main vehicle at the intersection is predicted.

[0017] Step 1.4: The master vehicle obtains the starting and target positions and speeds of each target vehicle through inter-vehicle communication, thereby predicting the future driving areas Ω1, Ω2, ..., Ω of each target vehicle at the intersection. i , ..., Ω n Ω i This represents the future driving area of ​​the i-th target vehicle at the intersection, where i∈[1,n];

[0018] Determine whether there is any spatial intersection between the future driving area Ω0 of the main vehicle and the future driving areas of each target vehicle. If there is, then the j-th target vehicle that has spatial intersection with the main vehicle is recorded as the conflicting vehicle V. j And the future driving area of ​​the main vehicle will be compared with that of the j-th conflicting vehicle V. j Future driving area Ω j The intersection area between them is defined as the j-th conflict area Φ. j =Ω0∩Ω j Otherwise, ignore the corresponding target vehicle; thus obtaining the conflict areas Φ1, Φ2, ..., Φ of all conflicting vehicles. j , ..., Φ k , where k represents the number of vehicles involved in the conflict.

[0019] The collision determination in step 2 includes:

[0020] Step 2.1: Denote the current speed and acceleration of the main vehicle as v0 and a0, respectively. Denote the distance of the main vehicle from its current position to entering and leaving the j-th conflict zone as Φ. j The path lengths are s 0k_enter and s 0k_leave Therefore, equation (1) is used to calculate the value of the main vehicle entering the j-th conflict zone Φ. j Time t 0k_enter and leaving the j-th conflict region Φ j Time t 0j_leave And form the time interval T of the main vehicle. 0j =[t 0j_enter ,t 0j_leave ]:

[0021]

[0022] Step 2.2, the j-th conflicting vehicle V j Φ for entering the j-th conflict zone is calculated according to formula (1). j Time t j_enter and leaving the j-th conflict zone Φ j Time t j_leave Thus, the j-th conflict vehicle V is formed. j Time interval T j =[t j_enter ,t j_leave ];

[0023] Step 2.3, if the time interval T of the main vehicle 0j =[t 0j_enter ,t 0j_leave ] Conflict with vehicle V of the jth type j Time interval T j =[t j_enter ,tj_leave If there is overlap, it means that the main vehicle and the j-th conflicting vehicle V are involved. j A collision will occur at the intersection; otherwise, it indicates that the main vehicle is in conflict with the j-th vehicle, V. j No collision occurs at the intersection; 1≤j≤k.

[0024] The feasible solution for the speed in step 3 is calculated and determined according to the following process:

[0025] Step 3.1: Establish the st relationship diagram of the main vehicle's intersection travel path and time, and divide the path area in the st relationship diagram into the pre-intersection area, the middle-intersection area, and the post-intersection area;

[0026] The j-th conflict region Φ j Displayed as rectangles on the st relationship diagram, the x-coordinates of the left and right boundaries of the j-th rectangle are t and t, respectively. j_enter and t k_leave The ordinates of the upper and lower boundaries of the j-th rectangle are s and s, respectively. k_enter and s k_leave ;

[0027] Step 3.2: Take the origin of the ST relationship diagram as the starting point of the two-dimensional directed bounding box of the main vehicle, and take the length of the two-dimensional directed bounding box of the main vehicle as the length of the line segment of the main vehicle on the ST relationship diagram. Thus, the starting position of the main vehicle is represented by the position of the line segment l on the negative half axis of the S-axis of the ST relationship diagram.

[0028] Step 3.3: Record the maximum speed of the main vehicle as v. max Based on the fitted curve δ=f(flag,c,v), the maximum permissible speed v of the main vehicle at the intersection according to the initially planned intersection travel path is obtained. limit ;

[0029] Step 3.4: In the ST relationship diagram, determine the feasible planning area of ​​the main vehicle ST curve.

[0030] Step 3.5: Within the feasible planning area, determine the ST curve of the main vehicle.

[0031] Step 3.4 includes:

[0032] Step 3.4.1: In the area before the intersection, line segment l has a slope value v max Rise until entering the area within the intersection;

[0033] Step 3.4.2: Within the intersection area, line segment l with a slope value v limit As it ascends, if it touches any rectangle, line segment l moves horizontally to the right until it reaches the right boundary of the corresponding rectangle, then continues at a slope value v. limitAscend, repeat steps 3.4.2 until entering the area after the intersection;

[0034] Step 3.4.3: In the area after the intersection, line segment l with a slope value v max Ascend until leaving the target area; thus, the trajectory of line segment l is a broken line, which is the left boundary LB;

[0035] Step 3.4.4: Move the left boundary LB to the right in parallel until it contacts any rectangle in the area within the intersection, thus obtaining the contact point between the left boundary LB and the corresponding rectangle.

[0036] Step 3.4.5: The line segment l starts from the contact point with a slope value v limit As it descends, if it encounters any rectangle, line segment l moves horizontally to the left until it reaches the left boundary of the corresponding rectangle, then continues descending at a slope value v. limit Descend and repeat this process until you enter the area before the intersection;

[0037] Step 3.4.6: In the area before the intersection, line segment l with a slope value v max The line segment l descends until it leaves the target area, and the trajectory of the line segment l below the contact point is obtained, which is a broken line.

[0038] Step 3.4.7: Continue to move the left boundary LB horizontally to the right above the contact point until it contacts the next rectangle in the intersection area, thus obtaining a new contact point. Then, process it according to the steps 3.4.5-3.4.6 until the movement trajectory of line segment l is connected to the movement trajectory of line segment l below the previous contact point, forming the movement trajectory of line segment l below the new contact point.

[0039] Step 3.4.8: Repeat step 3.4.7 until only a diagonal line segment remains in the middle of the intersection above the contact point on the left boundary LB. From this contact point upwards, line segment l has a slope value v. limit Ascend until entering the area after the intersection; in the area after the intersection, line segment l ascends with a slope value v. max Ascend until leaving the target area, forming the trajectory of line segment l above the contact point, which is a broken line. Connect the trajectory of line segment l above the contact point with the trajectory of line segment l below the contact point to form the right boundary RB.

[0040] Step 3.4.9: Take the area formed between the left boundary LB and the right boundary RB as the feasible planning area for the main vehicle st curve.

[0041] Step 3.5 includes:

[0042] Step 3.5.1: Divide the feasible planning region into several parallelograms based on the ordinates of the vertices of the left boundary LB and the right boundary RB.

[0043] Step 3.5.2: For each parallelogram region, use a fifth-order polynomial to plan the ST curve. The first and second derivatives at the starting point of the first parallelogram region are equal to the velocity and acceleration values ​​of the main vehicle when entering the target region, respectively. The first and second derivatives at the starting points of the remaining parallelogram regions are consistent with the position, velocity, and acceleration values ​​at the ending point of the previous parallelogram region, respectively.

[0044] Step 3.5.3: Connect the planned ST curves of each parallelogram region to obtain the ST curve of the main vehicle in the feasible planning region.

[0045] The present invention provides an electronic device, including a memory and a processor, wherein the memory is used to store a program that supports the processor in executing the method, and the processor is configured to execute the program stored in the memory.

[0046] The present invention discloses a computer-readable storage medium on which a computer program is stored, wherein the computer program is executed by a processor to perform the steps of the method.

[0047] Compared with existing technologies, the beneficial effects of this invention are reflected in:

[0048] 1. This invention fully considers the two-dimensional dimensions and driving direction of the vehicle, further clarifying the shape and size of the collision area, thereby improving the accuracy of collision detection;

[0049] 2. This invention proposes a method for finding the feasible region of speed on a path-time relationship graph (ST graph), ensuring both traffic efficiency and passenger comfort while avoiding collisions;

[0050] 3. This invention fully considers the uncertainties caused by vehicle path tracking errors and has a certain degree of tolerance. Attached Figure Description

[0051] Figure 1 This is a flowchart of the process of the present invention;

[0052] Figure 2 This is a plan view of the intersection according to the present invention;

[0053] Figure 3 This is a path-time relationship diagram (ST relationship diagram) of the present invention;

[0054] Figure 4This is a path and time planning diagram (ST planning diagram) for the main vehicle of the present invention. Detailed Implementation

[0055] In this example, a method for autonomous driving of an intelligent vehicle is described in [reference needed]. Figure 1 ,include:

[0056] Step 1: Based on the traffic intersection where the main vehicle is located, establish a target area centered on the intersection, select target vehicles within the target area, and determine the conflict area based on the intersection travel paths of the main vehicle and target vehicles, while also considering the uncertainty of the main vehicle path tracking algorithm.

[0057] In practice, the method for determining the conflict zone is as follows:

[0058] Step 1.1: Divide a rectangular area centered on the intersection and set it as the target area. The target area includes the intersection area, a section before entering the intersection, and a section after leaving the intersection. Take any vertex of the rectangular area as the origin o, and the two sides connected by the origin o are the x-axis and y-axis, respectively, to establish a rectangular coordinate system oxy.

[0059] The main vehicle considers all other cars in all lanes within the target area that are heading towards the intersection or are currently at the intersection as target vehicles, and treats each target vehicle and the main vehicle as a two-dimensional directed bounding box. The size of the two-dimensional directed bounding box is the minimum envelope rectangle of the projection of all points on the vehicle onto the ground when the vehicle is stationary. Let the number of target vehicles be denoted as n.

[0060] In this embodiment, see Figure 2 The main vehicle is approaching the intersection, and the remaining vehicles are numbered as Car 1, Car 2, Car 3, Car 4, Car 5, Car 6, Car 7, Car 8, Car 9, and Car 10. A rectangular region P1P2P3P4 is defined centered on the intersection. This region should include the intersection area, a section before entering the intersection, and a section after leaving the intersection. This region is designated as the target area. All other vehicles in the lanes within the target area that are approaching or currently at the intersection are considered target vehicles. In the diagram, the target vehicles include Car 1, Car 2, Car 4, Car 7, Car 8, and Car 10, totaling 6. The main vehicle and the 6 target vehicles are approximated as two-dimensional directed bounding boxes. The size of the two-dimensional directed bounding box is the smallest envelope rectangle of the projections of all points on the vehicle onto the ground when the vehicle is stationary.

[0061] Step 1.2: When the main vehicle enters the target area, based on the main vehicle's starting position and target position, initially plan the main vehicle's intersection driving route and speed. The driving route is in... Figure 2 The image shows a combination of two straight lines and a curve.

[0062] Step 1.3: Based on the planned intersection driving path and driving speed of the main vehicle, use the curve of the maximum error of path tracking and the behavior mode, path parameters and driving speed obtained by the path tracking algorithm, δ=f(flag,c,v), where δ represents the maximum error of path tracking, flag represents the behavior mode of the vehicle (straight or turning), c represents the parameters of the path (curvature, length, etc.), and v represents the driving speed of the vehicle.

[0063] Based on the predetermined intersection driving path and driving speed of the main vehicle, the maximum tracking error is calculated. Then, combined with the two-dimensional oriented bounding box and driving path of the main vehicle, the area Ω0 that the main vehicle will pass through at the intersection is obtained.

[0064] Step 1.4: The main vehicle obtains the driving intention of each target vehicle through inter-vehicle communication, or predicts the area that each target vehicle will pass through at the intersection based on the location and speed information of each target vehicle collected by the roadside unit: Ω1, Ω2, Ω3, Ω4, Ω5, Ω6, Ω n Let n represent the area that the nth target car will pass through at the intersection, where 1 ≤ n ≤ 6;

[0065] Determine if the driving area of ​​the main vehicle intersects with the driving area of ​​the target vehicle, i.e., determine if Ω0∩(Ω1∪Ω2∪Ω3∪Ω4∪Ω5∪Ω6) is equal to Ω0∩(Ω1∪Ω2∪Ω3∪Ω4∪Ω5∪Ω6). If If the main vehicle is determined not to collide with the target vehicle at the intersection, it continues driving along the predetermined path and speed. If not... It is then determined that the main vehicle will intersect with the target vehicle at the intersection. Figure 2 The four vehicles that will spatially intersect with the main vehicle at the intersection are vehicle 1, vehicle 2, vehicle 4, and vehicle 8, denoted as vehicle V. k The intersecting areas are designated as conflict areas, denoted as Φ1, Φ2, Φ4, Φ8, and Φ... k =Ω0∩Ω k k = 1, 2, 4, 8;

[0066] Step 2: Predict the time when the main vehicle and other vehicles that are in spatial conflict with the main vehicle enter and leave the conflict area, and calculate whether there is overlap. If there is overlap, it is determined that a collision will occur; otherwise, it is determined that a collision will not occur.

[0067] The specific method for determining whether a collision has occurred is as follows:

[0068] Step 2.1: The current speed and acceleration of the main vehicle are denoted as v0 and a0, respectively. The current position of the main vehicle is Φ, which represents the distance from entering to leaving the conflict zone. k The path lengths are s 0k_enter and s 0k_leave The main vehicle enters and leaves the conflict zone.k The time is denoted as t. 0k_enter and t 0k_leave This forms the time interval T. 0k =[t 0k_enter ,t 0k_leave ], where t 0k_enter and t 0k_leave The calculation formula is as shown in equation (1):

[0069]

[0070] In equation (1), s 0k_enter The two-dimensional directed bounding box of the main vehicle represents the region of friction Φ. k The path length from the initial contact point to the current position of the main vehicle, s 0k_leave The two-dimensional directed bounding box of the main vehicle represents the region of friction Φ. k The path length from the point of complete detachment to the current position of the main vehicle, t 0k_enter This indicates that the two-dimensional directed bounding box of the main vehicle initially collides with the conflict region Φ. k At the moment of initial contact, t 0k_leave The two-dimensional directed bounding box of the main vehicle represents the region of friction Φ. k The moment of complete detachment;

[0071] Step 2.2, Vehicle V k The current velocity and acceleration are denoted as v. k and a k Vehicle V k Current location distance from entering and leaving the conflict zone Φ k The path lengths are s k_enter and s k_leave Vehicle V k Entering and leaving the conflict zone Φ k The time is denoted as t. k_enter and t k_leave This forms the time interval T. k =[t k_enter ,t k_leave ], where t k_enter and t k_leave The calculation formula is as follows:

[0072]

[0073] In equation (2), s k_enter Indicates vehicle V k The two-dimensional oriented bounding box just meets the conflict region Φ k The point of initial contact to vehicle V k The path length at the current location, s k_leave Indicates vehicle Vk The two-dimensional oriented bounding box just meets the conflict region Φ k The position of complete separation from vehicle V k The path length at the current location, t k_enter Indicates vehicle V k The two-dimensional oriented bounding box just meets the conflict region Φ k At the moment of initial contact, t k_leave Indicates vehicle V k The two-dimensional oriented bounding box just meets the conflict region Φ k The moment of complete detachment;

[0074] Step 2.3: If all time intervals are relative to T 0k and T k If there is no overlap, it is determined that the main vehicle will not collide with the target vehicle at the intersection, and the main vehicle continues to travel along the predetermined path and speed; if at least one time interval T 0i and T i If there is overlap (1≤i≤k), it is determined that the main vehicle will collide with the target vehicle at the intersection;

[0075] Step 3: If a collision is determined to occur, then calculate whether the potential collision can be avoided by controlling the speed without changing the path. If it is feasible, then control the speed of the main vehicle to avoid the potential collision. If it is not feasible, then re-plan the path and speed.

[0076] If it is determined that no collision will occur, the main vehicle will proceed according to the established plan.

[0077] The specific method for avoiding potential collisions by controlling speed without changing the path, provided that a collision is determined to occur, is as follows:

[0078] Step 3.1, see Figure 3 Establish a path-time relationship diagram for the main vehicle (referred to as the ST relationship diagram), and divide the area in the diagram into the pre-intersection area, the middle area of ​​the intersection area, and the post-intersection area according to the different paths.

[0079] The conflict area Φ k The rectangles are represented on the graph as Q1, Q2, Q3, and Q4, with the x-coordinates of their left and right boundaries being t, respectively. k_enter and t k_leave The ordinates of the upper and lower boundaries of the rectangle are s and s respectively. k_enter and s k_leave That is, the rectangle represents the time t. k_enter and t k_leave Within the interval, the target vehicle will occupy s of the main vehicle's predetermined path. k_enter to s k_leave Section of road.

[0080] Step 3.2: Take the origin of the ST relationship diagram as the starting point of the two-dimensional directed bounding box of the main vehicle, and take the length of the two-dimensional directed bounding box of the main vehicle as the length of the line segment of the main vehicle on the ST relationship diagram. Thus, the starting position of the main vehicle is represented by the position of the line segment l on the negative half axis of the S-axis of the ST relationship diagram; the main vehicle is represented by the line segment l parallel to the S-axis.

[0081] Step 3.3, the maximum speed of the main vehicle is denoted as v. max Within the expected path tracking error, the maximum permissible speed of the main vehicle traveling along the predetermined trajectory at the intersection is obtained according to the formula δ=f(flag,c,v), denoted as v. limit .

[0082] Step 3.4: In the st relationship diagram, determine the feasible planning region of st.

[0083] In the area before the intersection, line segment l has a slope value v. max Rise until you enter the area at the intersection;

[0084] In the area of ​​the intersection, line segment l has a slope value v. limit Ascending, encountering rectangle Q1 symbolizing conflict, line segment l moves horizontally to the right to reach the right boundary of rectangle Q1, continuing at a slope value v limit Ascending, encountering rectangle Q2 symbolizing conflict, line segment l moves horizontally to the right to reach the right boundary of rectangle Q2, continuing at a slope value v. limit Ascend, reaching the area after the intersection; in the area after the intersection, line segment l ascends with a slope value v. max Ascend until leaving the target area. The trajectory of line segment l is a broken line, with the left boundary LB.

[0085] LB moves parallel to the right, making contact with rectangle Q3, which symbolizes conflict in the intersection area, and then moves downwards from that point of contact:

[0086] Line segment l with slope value v limit Descending, entering the area before the intersection, in the area before the intersection, line segment l descends with a slope value v max Descend until leaving the target area; above this point of contact, LB continues to move horizontally to the right, contacting the rectangle Q4 symbolizing conflict; from this point of contact downwards, line segment l descends with a slope value v. limit Descending, line segment l contacts rectangle Q3, then moves horizontally to the left to reach the left boundary of rectangle Q3, connecting with the previous trajectory; from this point of contact upwards, line segment l moves with a slope value v limit Ascend until entering the area after the intersection; in the area after the intersection, line segment l ascends with a slope value v. max Ascend until leaving the target area. The trajectory of line segment l is a broken line, with the right boundary RB.

[0087] See Figure 4The feasible planning region is divided into several parallelograms based on the ordinates of the vertices of the left boundary LB and the right boundary RB. For each parallelogram region, a fifth-order polynomial is used to plan the st curve. The first and second derivatives at the starting point of the first parallelogram region are equal to the velocity and acceleration values ​​of the main vehicle when entering the target region, respectively. The first and second derivatives at the starting points of the remaining parallelogram regions are consistent with the position, velocity, and acceleration values ​​at the ending point of the previous parallelogram region, respectively. The st curves planned by each parallelogram region are connected to obtain the st planning curve of the main vehicle in the feasible planning region.

[0088] In this embodiment, an electronic device includes a memory and a processor. The memory stores a program that supports the processor in executing the above-described method, and the processor is configured to execute the program stored in the memory.

[0089] In this embodiment, a computer-readable storage medium stores a computer program, which is executed by a processor to perform the steps of the above method.

Claims

1. A method for autonomous driving of an intelligent vehicle, characterized in that, Includes the following steps: Step 1: Based on the traffic intersection where the main vehicle is located, establish a target area centered on the intersection, select target vehicles within the target area, and determine the conflict area based on the intersection driving paths of the main vehicle and target vehicles, while also considering the control accuracy of the main vehicle path following algorithm. Step 2: Predict the time when the main vehicle and other vehicles that will have a spatial conflict with the main vehicle enter and leave the conflict area, and calculate whether there is a time overlap. If there is an overlap, it is determined that a collision will occur; otherwise, it is determined that a collision will not occur. Step 3: If a collision is determined to occur, calculate whether there is a feasible speed solution for safe obstacle avoidance without changing the path. If there is, control the speed of the main vehicle to avoid the potential collision; otherwise, replan the driving path and speed of the main vehicle at the intersection. If it is determined that no collision will occur, the main vehicle will continue to travel along the original intersection route; Step 3.1: Establish the st relationship diagram of the main vehicle's intersection travel path and time, and divide the path area in the st relationship diagram into the pre-intersection area, the middle-intersection area, and the post-intersection area; The j-th conflict region Displayed as rectangles on the ST relationship diagram, the x-coordinates of the left and right boundaries of the j-th rectangle are respectively... and The ordinates of the upper and lower boundaries of the j-th rectangle are respectively and ; Step 3.2: Using the origin of the ST relationship diagram as the starting point of the main vehicle's two-dimensional directed bounding box, and the length of the main vehicle's two-dimensional directed bounding box as the length of the main vehicle's line segment on the ST relationship diagram, the line segment on the negative half-axis of the ST relationship diagram is then used. The position indicates the starting position of the main vehicle; Step 3.3: Record the maximum speed of the main vehicle as... And based on the fitted curve The maximum permissible speed of the main vehicle at the intersection, according to the initially planned intersection route, is obtained. ; Step 3.4: In the ST relationship diagram, determine the feasible planning area of ​​the main vehicle's ST curve; Step 3.5: Within the feasible planning area, determine the ST curve of the main vehicle.

2. The intelligent vehicle automatic driving method according to claim 1, characterized in that, The conflict area in step 1 is determined according to the following steps: Step 1.1: Divide a rectangular area centered on the intersection and set it as the target area. The target area includes the intersection area, a section before entering the intersection, and a section after leaving the intersection. Take any vertex of the rectangular area as the origin o, and the two sides connected to the origin o are the x-axis and y-axis, respectively, to establish a rectangular coordinate system oxy. The master vehicle considers all other vehicles in all lanes within the target area that are heading towards or at the intersection as target vehicles, and treats each target vehicle and the master vehicle as a two-dimensional directed bounding box. The size of the two-dimensional directed bounding box is the minimum envelope rectangle of the projection of all points on the vehicle onto the ground when the vehicle is stationary. Let the number of target vehicles be denoted as n. Step 1.2: When the main vehicle enters the target area, based on the main vehicle's starting position and target position, preliminarily plan the main vehicle's intersection driving route and driving speed; Step 1.3: Based on the planned intersection driving path and speed of the main vehicle, use the maximum error of the path tracking obtained by the path tracking algorithm. With behavioral patterns Path parameters and driving speed Fitted curve The maximum tracking error of the vehicle's initial planned intersection travel path is calculated. Then, combined with the vehicle's two-dimensional directed bounding box and the intersection travel path, the future travel area of ​​the vehicle at the intersection is predicted. ; Step 1.4: The master vehicle obtains the starting and target positions and speeds of each target vehicle through inter-vehicle communication, thereby predicting the future driving area of ​​each target vehicle at the intersection. , This represents the future driving area of ​​the i-th target vehicle at the intersection. ; Determine the future driving area of ​​the main vehicle If there is any spatial intersection with the future driving area of ​​each target vehicle, then the j-th target vehicle that spatially intersects with the main vehicle is recorded as the conflicting vehicle. And the future driving area of ​​the main vehicle will be compared with that of the j-th conflicting vehicle. Future driving area The intersection area between them is defined as the j-th conflict area. Otherwise, ignore the corresponding target vehicle; This yields the conflict zones of all conflicting vehicles. , Indicates the number of vehicles involved in the conflict.

3. The intelligent vehicle automatic driving method according to claim 2, characterized in that, The collision determination in step 2 includes: Step 2.1: Record the current speed and acceleration of the main vehicle as follows: and The distance of the main vehicle from its current position to entering and leaving the j-th conflict zone. The path lengths are respectively and Thus, equation (1) is used to calculate the number of times the main vehicle enters the j-th conflict zone. time and leaving the j-th conflict region time And form the time interval of the main vehicle. : (1) Step 2.2, the j-th conflicting vehicle Calculate the entry into the j-th conflict zone according to formula (1). time and leaving the j-th conflict zone time Thus, the j-th conflict vehicle is formed. time interval ; Step 2.3, if the main vehicle's time interval Conflict with vehicle j time interval If there is overlap, it means that the main vehicle is in conflict with the j-th vehicle. A collision will occur at the intersection; otherwise, it indicates that the main vehicle is colliding with the j-th vehicle. No collision occurred at the intersection; .

4. The intelligent vehicle automatic driving method according to claim 3, characterized in that, Step 3.4 includes: Step 3.4.1: In the area before the intersection, line segment With slope value Rise until you enter the area inside the intersection; Step 3.4.2: Within the intersection area, line segments With slope value Ascending, if it touches any rectangle, then the line segment... After moving horizontally to the right and reaching the right boundary of the corresponding rectangle, continue moving at the slope value. Ascend, repeat steps 3.4.2 until entering the area after the intersection; Step 3.4.3: In the area after the intersection, line segments With slope value Ascend until leaving the target area; thus obtaining the line segment. The trajectory of the movement is a broken line, which is the left boundary LB; Step 3.4.4: Move the left boundary LB to the right in parallel until it contacts any rectangle in the area within the intersection, thus obtaining the contact point between the left boundary LB and the corresponding rectangle. Step 3.4.5, the line segment Starting from the contact point, with the slope value As it descends, if it encounters any rectangle, the line segment... After moving horizontally to the left and reaching the left boundary of the corresponding rectangle, continue moving at the slope value. Descend and repeat this process until you enter the area before the intersection; Step 3.4.6: In the area before the intersection, line segments With slope value Descend until leaving the target area, obtaining a line segment. The trajectory of motion below the point of contact is a broken line; Step 3.4.7: Continue moving the left boundary LB horizontally to the right along the broken line portion above the contact point until it contacts the next rectangle within the intersection area, thus obtaining a new contact point. Then, process according to steps 3.4.5-3.4.6 until the line segment... The trajectory of the motion is the line segment below the previous contact point. The trajectories of their movements are connected to form a line segment. The trajectory of motion below the new point of contact; Step 3.4.8: Repeat step 3.4.7 until only a diagonal line segment remains in the middle of the intersection above the contact point on the left boundary LB. From this contact point upwards, the line segment... With slope value Ascend until entering the area after the intersection; in the area after the intersection, the line segment... With slope value Ascend until leaving the target area, forming a line segment. The trajectory of motion above the contact point is a broken line, and the line segment The trajectory and line segment above the contact point The motion trajectories below the contact point are connected to form the right boundary RB; Step 3.4.9: Take the area formed between the left boundary LB and the right boundary RB as the feasible planning area for the main vehicle st curve.

5. The intelligent vehicle automatic driving method according to claim 4, characterized in that, Step 3.5 includes: Step 3.5.1: Divide the feasible planning region into several parallelograms based on the ordinates of the vertices of the left boundary LB and the right boundary RB. Step 3.5.2: For each parallelogram region, use a fifth-order polynomial to plan the ST curve. The first and second derivatives at the starting point of the first parallelogram region are equal to the velocity and acceleration values ​​of the main vehicle when entering the target region, respectively. The first and second derivatives at the starting points of the remaining parallelogram regions are consistent with the position, velocity, and acceleration values ​​at the ending point of the previous parallelogram region, respectively. Step 3.5.3: Connect the planned ST curves of each parallelogram region to obtain the ST curve of the main vehicle in the feasible planning region.

6. An electronic device, comprising a memory and a processor, characterized in that, The memory is used to store a program that supports the processor in executing the method of any one of claims 1-5, the processor being configured to execute the program stored in the memory.

7. A computer-readable storage medium storing a computer program thereon, characterized in that, The computer program is executed by the processor to perform the steps of any of the methods described in claims 1-5.

Images