Global Speed Planning Method, Global Speed Planning Device and Planning System

Abstract

The present application provides a global speed planning method, a global speed planning device and a planning system. The method includes a first planning step of using a speed planning algorithm to perform speed planning on the total running duration, departure time, global driving path and section speed limit information of the received vehicle to obtain an initial global trajectory; an acquisition step of controlling the vehicle to travel according to the initial global trajectory to obtain the actual running duration when the vehicle reaches the target path point; a first comparison step of, in the case of being less than or equal to a predetermined threshold, performing the acquisition step at least once until the vehicle reaches the destination; a second comparison step of, in the case of being greater than the predetermined threshold, calculating the remaining total running duration and the remaining global path, and sequentially performing the first planning step and the acquisition step at least once until the vehicle reaches the destination, solving the problem that it is difficult for the prior art to determine the global trajectory of the vehicle with less calculation while considering the on-time arrival requirement of the vehicle.

Description

Technical Field

[0001] The present application relates to the technical field of autonomous driving, and in particular, to a global speed planning method, a global speed planning device, a computer-readable storage medium, a processor, and a planning system. Background Art

[0002] As a key technology of autonomous driving, global speed planning mainly determines the speed of traveling along this optimal path according to the optimal path information output by the global planning and the speed limit information of the path segments.

[0003] Existing global speed planning schemes require that the obtained speed curve not only needs to satisfy the acceleration and deceleration performance of the vehicle and the speed limit of the road, but also needs to ensure that the obtained speed curve is a continuous and smooth curve. For the determination method of the above smooth speed curve, most of them are based on dynamic programming or other heuristic optimization methods, and these methods all have the problem of large computational complexity. In addition, in automated systems such as intelligent passenger transportation bus systems that are scheduled according to timing, the performance requirements for intelligent vehicles to arrive on time are often very high, but existing global speed planning schemes rarely consider this performance requirement.

[0004] Therefore, there is an urgent need for a method that can determine the optimal global speed curve of the vehicle with less computational complexity while considering the vehicle's demand for arriving on time. Summary of the Invention

[0005] The main purpose of the present application is to provide a global speed planning method, a global speed planning device, a computer-readable storage medium, a processor, and a planning system to solve the problem that it is difficult to determine the global trajectory of the vehicle with less computational complexity while considering the vehicle's demand for arriving on time in the prior art.

[0006] According to one aspect of an embodiment of the present invention, there is provided a global speed planning method, including: a first planning step of using a speed planning algorithm to perform speed planning on the received total running duration, departure time, global driving path, and road section speed limit information of the vehicle to obtain an initial global trajectory, where the initial global trajectory is used to represent the speed information and time information of each path point, and the path point is a point divided at a predetermined interval on the global driving path; an acquisition step of controlling the vehicle to drive according to the initial global trajectory to obtain the actual running duration of the vehicle reaching the target path point, where the target path point is one of the multiple path points; a first comparison step of, when the difference between the actual running duration and the path point planning duration is less than or equal to a predetermined threshold, performing the acquisition step at least once until the vehicle reaches the destination, where the path point planning duration is the running duration of the initially planned target path point; a second comparison step of, when the difference between the actual running duration and the path point planning duration is greater than the predetermined threshold, calculating the remaining total running duration and the remaining global path, and sequentially performing the first planning step and the acquisition step at least once until the vehicle reaches the destination.

[0007] Optionally, using a speed planning algorithm to perform speed planning on the received total running duration, departure time, global driving path, and road section speed limit information of the vehicle to obtain an initial global trajectory includes: dividing the global driving path according to a predetermined rule to obtain a plurality of path segments, where the predetermined rule is a rule determined according to the forward and backward movement of the vehicle; obtaining at least two key points on each of the path segments, where the key point is one of the multiple path points; preprocessing at least two of the key points on each of the path segments to obtain target key points on each of the path segments; and performing trapezoidal speed planning according to the target key points on each of the path segments, the total running duration, the departure time, the global driving path, and the road section speed limit information to obtain the initial global trajectory of the vehicle.

[0008] Optionally, preprocess at least two of the key points on each of the path segments to obtain target key points on each of the path segments, including: for consecutive first and second key points on a target path segment, determine a first speed and a second speed of the first key point, and a third speed of the second key point, where the target path segment is one of the multiple path segments, and both the first key point and the second key point are the key points; in a case where the first speed is greater than the second speed and the second speed is greater than the third speed, calculate a first distance for decelerating from the second speed to the third speed based on a maximum deceleration, and determine the target key point on the target path segment based at least on the first distance; in a case where the second speed is greater than the first speed and the first speed is greater than the third speed, calculate a second distance for decelerating from the first speed to the third speed based on the maximum deceleration, and determine the target key point on the target path segment based at least on the second distance.

[0009] Optionally, determining the target key point on the target path segment based at least on the first distance includes: calculating a distance between the first key point and the second key point to obtain a target key distance; in a case where the first distance is greater than the target key distance, determining the second key point as the target key point, determining a fourth speed of the second key point as the first speed, and deleting the first key point; in a case where the first distance is less than or equal to the target key distance, determining both the first key point and the second key point as the target key points.

[0010] Optionally, determining the target key point on the target path segment based at least on the second distance includes: calculating a distance between the first key point and the second key point to obtain a target key distance; in a case where the second distance is greater than the target key distance, determining the second key point as the target key point, determining a fourth speed of the second key point as the first speed, and deleting the first key point; in a case where the second distance is less than or equal to the target key distance, determining both the first key point and the second key point as the target key points.

[0011] Optionally, perform trapezoidal speed planning based on the target key points on each of the path segments, the total running duration, the departure time, the global driving path, and the speed limit information of the road segments to obtain the initial global trajectory of the vehicle, including: a second planning step of performing trapezoidal speed planning on the target key points on each of the path segments at least according to the speed limit information of the road segments to obtain the global speed curve of the global driving path; a first determination step of determining the global trajectory duration according to the global speed curve and the global driving path; a second determination step of determining the initial global trajectory according to the global trajectory duration, the total running duration, and the departure time.

[0012] Optionally, determining the initial global trajectory according to the global trajectory duration, the total running duration, and the departure time includes: in the case where the global trajectory duration is greater than the total running duration, generating a first planning failure message and sending the first planning failure message to the scheduling system; in the case where the global trajectory duration is less than or equal to the total running duration, calculating the difference between the global trajectory duration and the total running duration to obtain a target duration difference, and determining the initial global trajectory of the vehicle at least according to the target duration difference and the departure time.

[0013] Optionally, determining the initial global trajectory of the vehicle at least according to the target duration difference and the departure time includes: in the case where the target duration difference is less than or equal to the duration difference threshold, adding the speed information and time information to each of the path points according to the global speed curve and the departure time to obtain the initial global trajectory of the vehicle; in the case where the target duration difference is greater than the duration difference threshold, correcting the speed limit information of the road segments according to the target duration difference, and performing the second planning step, the first determination step, and the second determination step at least once until the target duration difference is less than or equal to the duration difference threshold.

[0014] Optionally, performing trapezoidal speed planning on the target key points on each of the path segments at least according to the speed limit information of the road segments to obtain the global speed curve of the global driving path includes: a third planning step of performing kinematic speed planning at least based on the speed information of the target key points on each of the path segments to obtain a plurality of initial path speed curves, where one path segment corresponds to one initial path speed curve; an optimization step of performing optimization processing on each of the initial path speed curves to obtain an initial global speed curve; an iterative step of performing iterative processing on the initial global speed curve based on the speed limit information of the road segments to obtain the global speed curve.

[0015] Optionally, optimize each of the initial path speed curves to obtain an initial global speed curve, including: constructing an optimization function based at least on the speed information of two adjacent path points to obtain a plurality of optimization objective functions; using the gradient descent method to optimize and solve the plurality of optimization objective functions to obtain the initial global speed curve.

[0016] Optionally, based on the road section speed limit information, perform iterative processing on the initial global speed curve to obtain the global speed curve, including: determining the actual path running duration of an alternative path point according to the initial global speed curve, where the alternative path point is one of the plurality of path points; calculating the difference between the actual path running duration and the planned path running duration to obtain a path duration difference, where the planned path running duration is the running duration of the planned alternative path point; in the case where the path duration difference is greater than the predetermined threshold, correcting the road section speed limit information of the alternative path point according to the corrected speed limit, to obtain corrected road section speed limit information, and performing the third planning step, the optimization step, and the iterative step at least once until the difference between the global trajectory duration of the global speed curve and the total running duration satisfies a preset range, where the corrected speed limit is determined based at least on the initial corrected speed limit, the actual path running duration, and the planned path running duration.

[0017] According to another aspect of the embodiments of the present invention, there is also provided a global speed planning device, including: a planning unit, configured to perform a first planning step, using a speed planning algorithm to perform speed planning on the received total running duration, departure time, global driving path, and road section speed limit information of a vehicle, to obtain an initial global trajectory, where the initial global trajectory is used to represent the speed information and time information of each path point, and the path point is a point divided at a predetermined interval on the global driving path; an acquisition unit, configured to perform an acquisition step, controlling the vehicle to travel according to the initial global trajectory to obtain the actual running duration of the vehicle reaching a target path point, where the target path point is one of the plurality of path points; a first comparison unit, configured to perform a first comparison step, in the case where the difference between the actual running duration and the path point planning duration is less than or equal to a predetermined threshold, performing the acquisition step at least once until the vehicle reaches the destination, where the path point planning duration is the running duration of the initially planned target path point; a second comparison unit, configured to perform a second comparison step, in the case where the difference between the actual running duration and the path point planning duration is greater than the predetermined threshold, calculating the remaining total running duration and the remaining global path, and sequentially performing the first planning step and the acquisition step at least once until the vehicle reaches the destination.

[0018] According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, where the computer-readable storage medium includes a stored program, and wherein the program executes any one of the above-mentioned global speed planning methods.

[0019] According to yet another aspect of the embodiments of the present invention, there is also provided a processor, where the processor is used to run a program, and wherein when the program runs, it executes any one of the above-mentioned global speed planning methods.

[0020] According to one aspect of the embodiments of the present invention, there is also provided a planning system, a scheduling system, where the scheduling system is used to send the total running duration and departure time to the global speed planning system; a global path planning system, where the global path planning system is used to send the global driving path and section speed limit information to the global speed planning system; the global speed planning system includes a global speed planning device, and the global speed planning device is used to execute any one of the above-mentioned global speed planning methods.

[0021] In the global speed planning method of the embodiments of the present invention, first, based on the speed planning algorithm, speed planning is performed on the received total running duration, departure time, global driving path, and section speed limit information of the vehicle to obtain an initial global trajectory; then, during the process of the vehicle driving along the initial global trajectory, the actual running duration of the vehicle reaching the target path point is obtained; after that, the difference between the actual running duration of the vehicle at the target path point and the path point planning duration is calculated. If the difference is less than or equal to a predetermined threshold, it indicates that the vehicle can reach the destination on time, so the vehicle can continue to be controlled to run along the initial global trajectory until the vehicle reaches the destination; if the difference is greater than the predetermined threshold, it indicates that the vehicle cannot reach the destination on time. Calculate the remaining total running duration and remaining global path of the vehicle, and then perform re-speed planning according to the remaining total running duration, remaining global path, departure time, and section speed limit information, so that the vehicle can reach the destination on time. Compared with the method of obtaining the global trajectory of the vehicle by dynamic programming or heuristic optimization method in the prior art, the global speed planning method of the present application only needs to use the speed planning algorithm to perform speed planning on the total running duration, departure time, global driving path, and section speed limit information of the vehicle to obtain the initial global trajectory of the vehicle, which ensures that the calculation amount for determining the initial global trajectory is small and does not require a high computing power platform at the vehicle specification level, ensuring that the cost of the global speed planning method of the present application is low. In addition, the global speed planning method of the present application takes into account the requirement of the vehicle to reach on time and dynamically adjusts the running strategy of the vehicle under the influence of dynamic traffic flow, ensuring that the vehicle can reach the destination on time, thus solving the problem in the prior art that it is difficult to determine the global trajectory of the vehicle with less calculation amount while considering the requirement of the vehicle to reach on time. Description of the Drawings

[0022] The accompanying drawings forming a part of this application are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0023] Figure 1 A flowchart of a global speed planning method according to an embodiment of this application is shown;

[0024] Figure 2 A schematic structural diagram of determining an initial global trajectory according to an embodiment of this application is shown;

[0025] Figure 3 A flowchart of a global speed planning method according to a specific embodiment of this application is shown;

[0026] Figure 4 A flowchart of a speed planning algorithm according to a specific embodiment of this application is shown;

[0027] Figures 5 to 13 A schematic diagram of an initial path speed curve according to a specific embodiment of this application is shown;

[0028] Figure 14 A schematic diagram of an initial global speed curve and a global speed curve according to a specific embodiment of this application is shown;

[0029] Figure 15 A schematic structural diagram of a global speed planning device according to an embodiment of this application is shown.

[0030] Among them, the above-mentioned drawings include the following reference numerals:

[0031] 100, scheduling system; 200, global path planning system; 300, global speed planning system; 400, initial global trajectory; 500, initial global speed curve; 600, global speed curve; 10, planning unit; 20, acquisition unit; 30, first comparison unit; 40, second comparison unit. Detailed Embodiments

[0032] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments may be combined with each other. The following will refer to the drawings and combine the embodiments to detail this application.

[0033] To enable those skilled in the art to better understand the solution of this application, the following will clearly and completely describe the technical solutions in the embodiments of this application in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0034] It should be noted that the terms "first", "second", etc. in the specification, claims and above-mentioned drawings of this application are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances for the embodiments of this application described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0035] As mentioned in the background art, it is difficult to determine the global trajectory of a vehicle with a small amount of calculation while considering the requirement of the vehicle to arrive on time in the prior art. To solve the above problem, in a typical embodiment of this application, a global speed planning method, a global speed planning device, a computer-readable storage medium, a processor and a planning system are provided.

[0036] According to an embodiment of this application, a global speed planning method is provided.

[0037] Figure 1 is a flowchart of the global speed planning method according to an embodiment of this application. As Figure 1 shown, the global speed planning method includes the following steps:

[0038] Step S101, the first planning step, using a speed planning algorithm, performs speed planning on the total running duration, departure time, global driving path and section speed limit information of the received vehicle to obtain an initial global trajectory. The above initial global trajectory is used to characterize the speed information and time information of each path point. The above path points are points divided at a predetermined interval on the above global driving path;

[0039] Step S102, the acquisition step, controls the above vehicle to drive according to the above initial global trajectory to obtain the actual running duration of the above vehicle to reach the target path point. The above target path point is one of the multiple above path points;

[0040] Step S103, the first comparison step. When the difference between the actual running duration and the path point planning duration is less than or equal to a predetermined threshold, execute the above acquisition step at least once until the vehicle reaches the destination. The path point planning duration is the running duration of the initially planned target path point.

[0041] Step S104, the second comparison step. When the difference between the actual running duration and the path point planning duration is greater than the predetermined threshold, calculate the remaining total running duration and the remaining global path, and sequentially execute the above first planning step and the above acquisition step at least once until the vehicle reaches the destination.

[0042] In the above global speed planning method, first, based on the speed planning algorithm, perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the received vehicle to obtain an initial global trajectory. Then, during the vehicle's driving along the initial global trajectory, obtain the actual running duration of the vehicle reaching the target path point. After that, calculate the difference between the actual running duration of the vehicle at the target path point and the path point planning duration. When the difference is less than or equal to the predetermined threshold, it indicates that the vehicle can reach the destination on time, so the vehicle can continue to be controlled to run along the initial global trajectory until it reaches the destination. When the difference is greater than the predetermined threshold, it indicates that the vehicle cannot reach the destination on time. Calculate the remaining total running duration and the remaining global path of the vehicle, and then perform speed planning again based on the remaining total running duration, the remaining global path, the departure time, and the road section speed limit information to enable the vehicle to reach the destination on time. Compared with the method of obtaining the global trajectory of the vehicle through dynamic programming or heuristic optimization methods in the prior art, the global speed planning method of this application only needs to use the speed planning algorithm to perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the vehicle to obtain the initial global trajectory of the vehicle. This ensures that the computational effort for determining the initial global trajectory is small and does not require being carried on a high-computing-power platform at the vehicle specification level, ensuring that the cost of the global speed planning method of this application is low. In addition, by considering the requirement for the vehicle to reach on time, the global speed planning method of this application dynamically adjusts the vehicle's operation strategy under the influence of dynamic traffic flow, ensuring that the vehicle can reach the destination on time, thus solving the problem in the prior art that it is difficult to determine the global trajectory of the vehicle with less computational effort while considering the requirement for the vehicle to reach on time.

[0043] In the actual application process, the above-mentioned road section speed limit information can be adjusted in real time according to different road sections. The above total running duration is the total running duration required for the vehicle to reach the destination from the starting point. The above global driving path is determined by the global path planning system. The global driving path is only the path of the vehicle's driving and does not have speed information and time information. And the initial global trajectory obtained in this application is to add the speed information and time information corresponding to each path point on the basis of the global driving path. In addition, in this application, the time information in the initial global trajectory can be the arrival time of the vehicle at each path point.

[0044] Specifically, this application does not limit the size of the above-mentioned predetermined threshold, which can be flexibly adjusted according to the actual driving conditions of the vehicle.

[0045] In a specific embodiment of this application, as Figure 2 shown, the global speed planning method of this application is applied in the global speed planning system 300. The scheduling system 100 realizes the decision-making on the destination, departure time, and total running duration of the vehicle, and the scheduling system 100 sends the determined destination to the global path planning system 200. The global path planning system 200 plans the global driving path for the vehicle to reach the destination according to the destination, starting point positioning information, and high-precision map information, and obtains the road section speed limit information along the global driving path. The global path planning system 200 sends the global driving path and road section speed limit information to the global speed planning system 300. For the global speed planning system 300, it receives the total running duration and departure time sent by the scheduling system 100, and receives the global driving path and road section speed limit information sent by the global path planning system 200. The global speed planning system 300 uses a speed planning algorithm to perform speed planning on the received total running duration, departure time, global driving path, and road section speed limit information, and obtains the speed curve and time series of the vehicle along the global path, so as to obtain the initial global trajectory 400 of the vehicle.

[0046] In another specific embodiment of this application, a global speed planning method is involved, specifically as shown in Figure 3As shown in the figure. Among them, the global speed planning method includes a punctual arrival speed planning part and a global speed rolling planning part during the vehicle operation. The punctual arrival speed planning part can meet the total operation duration requirement, the road section speed limit requirement, and the vehicle operation comfort requirement, and obtain the initial global trajectory. If the vehicle strictly follows the initial global trajectory, it can arrive at the destination on time. The global speed rolling planning part considers the influence of the dynamic traffic flow during the vehicle operation, which may cause the vehicle to arrive at a certain target path point earlier or later than expected during the execution of the initial global trajectory. If the difference between the actual operation duration of the vehicle arriving at the target path point and the planned operation duration of the path point corresponding to the target path point is greater than the predetermined threshold, the global speed planning system will execute the speed planning algorithm in a rolling manner according to the remaining total operation duration requirement, the remaining global path, etc., so as to dynamically adjust the global trajectory of the vehicle, obtain the corrected global trajectory, and then ensure that the vehicle arrives at the destination on time. The specific steps are as follows:

[0047] S1. Receive the total operation duration T sent by the dispatching system d , the departure time T s , and receive the global driving path given by the global path planning system, as well as the road section speed limit information;

[0048] S2. Adopt the speed planning algorithm, the total operation duration T d , the departure time T s , the global path and the road section speed limit information, obtain the initial global trajectory, and control the vehicle to start executing the initial global trajectory;

[0049] S3. Calculate the actual operation duration T v (k) of the vehicle arriving at the target path point and the planned operation duration T d (k) of the path point corresponding to the target path point, and calculate the difference value T v (k)-T d (k). If the difference value T v (k)-T d (k) is greater than the predetermined threshold ΔT max (that is, |T v (k)-T d (k)|>ΔT max ), then calculate the remaining total operation duration T d,r =T d -T v (k), and intercept the remaining global path. Call the speed planning algorithm again to obtain the corrected global trajectory of the remaining global path, and the vehicle starts to execute the corrected global trajectory. If the absolute value of the difference value T v (k)-T d (k) is less than or equal to the predetermined threshold ΔT max (that is, |T v (k)-Td (k)| ≤ ΔT max ),then control the vehicle to continue to execute the initial global trajectory;

[0050] S4. Determine whether the vehicle has reached the destination. If the vehicle accurately reaches the destination, end the program; otherwise, return to step S3.

[0051] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0052] In an embodiment of the present application, a speed planning algorithm is adopted to perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the received vehicle to obtain the initial global trajectory of the vehicle, including: dividing the above global driving path according to a predetermined rule to obtain a plurality of path segments, where the above predetermined rule is a rule determined according to the forward and backward of the above vehicle; obtaining at least two key points on each of the above path segments, where the above key points are one of the above plurality of path points; preprocessing at least two of the above key points on each of the above path segments to obtain the target key points on each of the above path segments; performing trapezoidal speed planning according to the above target key points on each of the above path segments, the above total running duration, the above departure time, the above global driving path, and the above road section speed limit information to obtain the above initial global trajectory of the vehicle. In this embodiment, according to the road section speed limit information, the key points on each path segment are preprocessed, so as to eliminate the keys that will cause the planning to fail, and it can be ensured that the obtained target key points can successfully perform global speed planning. Then, trapezoidal speed planning is performed using the target key points on each path segment, the total running duration, the departure time, the global driving path, and the road section speed limit information to obtain the initial global trajectory, which ensures that the obtained initial global trajectory is more reasonable and can further meet the requirement of arriving at the destination on time.

[0053] Specifically, the above predetermined rule is a rule determined according to the forward and backward of the vehicle. The forward and backward of the vehicle can be determined according to the direction of the vehicle's steering wheel.

[0054] In the actual application process, there may be more than two key points on each path segment, and each two key points can determine a path segment with a single speed limit. For some path segments where the right-side speed (i.e., the third speed) requirement of the second key point cannot be achieved even by deceleration, these path segments cannot obtain a feasible planned speed. Therefore, in another embodiment of the present application, at least two of the above-mentioned key points on each of the above-mentioned path segments are preprocessed to obtain the target key points on each of the above-mentioned path segments, including: for the consecutive first key point and second key point on the target path segment, determining the first speed and second speed of the first key point, and the third speed of the second key point, where the target path segment is one of the multiple above-mentioned path segments, and the first key point and the second key point are both the above-mentioned key points; in the case where the first speed is greater than the second speed and the second speed is greater than the third speed, based on the maximum deceleration, calculating the first distance from the second speed to the third speed, and determining the above-mentioned target key point on the target path segment at least according to the first distance; in the case where the second speed is greater than the first speed and the first speed is greater than the third speed, based on the above-mentioned maximum deceleration, calculating the second distance from the first speed to the third speed, and determining the above-mentioned target key point on the target path segment at least according to the second distance, so as to ensure that the vehicle can be successfully globally speed-planned according to the determined target key points subsequently.

[0055] Specifically, the first speed of the first key point can be the left-side speed v0 of the first key point. The second speed of the first key point can be the right-side speed v1 of the first key point. The third speed of the second key point is the right-side speed v2 of the second key point.

[0056] In another embodiment of the present application, determining the above-mentioned target key point on the target path segment at least according to the first distance includes: calculating the distance between the first key point and the second key point to obtain the target key distance; in the case where the first distance is greater than the target key distance, determining the second key point as the above-mentioned target key point, determining the fourth speed of the second key point as the first speed, and deleting the first key point; in the case where the first distance is less than or equal to the target key distance, determining both the first key point and the second key point as the above-mentioned target key points. In this embodiment, in the case where the first distance is greater than the target key distance, it indicates that even according to the maximum deceleration, the requirement of the third speed (i.e., the right-side speed) of the second key point cannot be achieved. Therefore, the fourth speed of the second key point (i.e., the left-side speed of the second key point) can be determined as the first speed of the first key point (i.e., the left-side speed of the first key point), so as to ensure that the vehicle can be successfully globally speed-planned according to the target key points subsequently.

[0057] In order to further ensure that the vehicle can be successfully globally speed-planned according to the target key points subsequently. In another embodiment of the present application, the target key points on the target path segment are determined at least according to the above second distance, including: calculating the distance between the first key point and the second key point to obtain the target key distance; in the case where the second distance is greater than the target key distance, determining the second key point as the target key point, determining the fourth speed of the second key point as the first speed, and deleting the first key point; in the case where the second distance is less than or equal to the target key distance, determining both the first key point and the second key point as the target key points.

[0058] In the actual application process, for two consecutive key points on the target path segment, namely the first key point and the second key point, assume that v0 is the left-side speed of the first key point, v1 is the right-side speed of the first key point, and v2 is the right-side speed of the second key point.

[0059] If v0 > v1 and v1 > v2, then decelerate from v1 to v2 at the maximum deceleration a min The first distance S min1 The expression is:

[0060]

[0061] If the first distance S min1 is greater than the target key distance (the distance between the first key point and the second key point), then determine the second key point as the target key point, assign the fourth speed (left-side speed) of the second key point as v0, and delete the first key point; otherwise, do not perform any processing (i.e., determine both the first key point and the second key point as the target key points).

[0062] If v1 > v0 and v0 > v2, then decelerate from v0 to v2 at the maximum deceleration a min The second distance S min2 The expression is:

[0063]

[0064] If the second distance S min2 is greater than the target key distance (the distance between the first key point and the second key point), then determine the second key point as the target key point, assign the fourth speed (left-side speed) of the second key point as v0, and delete the first key point; otherwise, do not perform any processing (i.e., determine both the first key point and the second key point as the target key points).

[0065] In an embodiment of the present application, trapezoidal speed planning is performed based on the above-mentioned target key points on each of the above-mentioned path segments, the above-mentioned total running duration, the above-mentioned departure time, the above-mentioned global driving path, and the above-mentioned road section speed limit information to obtain the above-mentioned initial global trajectory of the vehicle, including: a second planning step of performing the above-mentioned trapezoidal speed planning on the above-mentioned target key points on each of the above-mentioned path segments at least according to the above-mentioned road section speed limit information to obtain the global speed curve of the above-mentioned global driving path; a first determination step of determining the global trajectory duration according to the above-mentioned global speed curve and the above-mentioned global driving path; and a second determination step of determining the above-mentioned initial global trajectory according to the above-mentioned global trajectory duration, the above-mentioned total running duration, and the above-mentioned departure time. In this embodiment, the initial global trajectory is determined according to the global trajectory duration, the total running duration, and the departure time, so as to ensure that the subsequent vehicle can reach the destination on time according to the initial global trajectory.

[0066] Specifically, in the present application, there is no limitation on the specific method for determining the global trajectory duration according to the global speed curve and the global driving path, and it can be any feasible method in the prior art.

[0067] In another embodiment of the present application, determining the above-mentioned initial global trajectory according to the above-mentioned global trajectory duration, the above-mentioned total running duration, and the above-mentioned departure time includes: in the case where the above-mentioned global trajectory duration is greater than the above-mentioned total running duration, generating a first planning failure message and sending the above-mentioned first planning failure message to the scheduling system; in the case where the above-mentioned global trajectory duration is less than or equal to the above-mentioned total running duration, calculating the difference between the above-mentioned global trajectory duration and the above-mentioned total running duration to obtain a target duration difference, and determining the above-mentioned initial global trajectory of the vehicle at least according to the above-mentioned target duration difference and the above-mentioned departure time. In this embodiment, when the global trajectory running duration is greater than the total running duration, it indicates that even if the vehicle travels at the road section speed limit, it cannot reach the destination on time, that is, the global speed planning fails this time, and the first planning failure message can be sent to the scheduling system so that the scheduling system can re-make a decision on the total running duration. When the global trajectory duration is less than or equal to the total running duration, the initial global trajectory is determined according to the target duration difference, which can further ensure that the vehicle can reach the destination more punctually.

[0068] In order to further ensure that the vehicle can reach the destination more punctually, in another embodiment of the present application, the initial global trajectory of the vehicle is determined at least according to the above-mentioned target duration difference and the above-mentioned departure time, including: when the above-mentioned target duration difference is less than or equal to the duration difference threshold, according to the above-mentioned global speed curve and the above-mentioned departure time, speed information and time information are added to each of the above-mentioned path points to obtain the above-mentioned initial global trajectory of the vehicle; when the above-mentioned target duration difference is greater than the duration difference threshold, the above-mentioned road section speed limit information is corrected according to the above-mentioned target duration difference, and the above-mentioned second planning step, the above-mentioned first determination step, and the above-mentioned second determination step are executed at least once until the above-mentioned target duration difference is less than or equal to the above-mentioned duration difference threshold.

[0069] Specifically, in the present application, the magnitude of the above-mentioned duration difference threshold is not limited, and it can be flexibly adjusted according to the driving conditions of the vehicle.

[0070] In a specific embodiment of the present application, a method for determining the initial global trajectory of a vehicle is involved, specifically Figure 4 as shown. The specific steps for determining the initial global trajectory are as follows:

[0071] s1. Obtain the total running duration T d , departure time T s , global driving path, path speed limit information v max , maximum vehicle acceleration a that meets the comfort requirements max and maximum deceleration a min ;

[0072] s2. Divide the global driving path according to a predetermined rule (i.e., a rule for switching between forward and backward) to obtain multiple path segments. Then, speed planning is performed on each path segment respectively. During the process of performing speed planning on each path segment, except that the starting speed of the first path segment is the current driving speed of the vehicle, the starting speed and the ending speed of the remaining path segments are both set to 0;

[0073] s3. For each path segment, according to the road section speed limit information of each path segment, key points are obtained, and the number of key points for each path segment needs to be greater than or equal to 2. The information of each key point includes the speed on the left side of the key point (or the speed that the vehicle can actually reach), the speed limit on the right side, and the serial number in the global path (since the key point is a point among multiple path points, the serial number of the key point is the serial number of the corresponding path point). Since the path segment between two adjacent key points is a path segment with a single speed limit, the starting speed, ending speed, and speed limit value of this path segment are determined by the speed on the left side of the first key point, the speed limit on the right side of the second key point, and the speed on the right side of the first key point respectively;

[0074] S4. Preprocess the key points obtained for each path segment to obtain the target key points on each path segment. The purpose of preprocessing is to eliminate the key points that may cause the planning to fail.

[0075] S5. Perform trapezoidal speed planning based on the above target key points, total running duration, global driving path, and road section speed limit information on each path segment to obtain the global speed curve. Then, based on the global speed curve and the global driving path, determine the global trajectory duration. If the global trajectory duration is greater than the total running time, generate the first planning failure information and send the first planning failure information to the dispatching system so that the dispatching system can re-make a decision on the total running duration; if the global trajectory duration is less than or equal to the total running duration, calculate the difference between the global trajectory duration and the total running duration to obtain the target duration difference.

[0076] S6. In the case where the target duration difference is greater than the duration difference threshold, correct the corresponding road section speed limit information according to the target duration difference, and after correcting the corresponding road section speed limit information, re-execute from step S1 until the above target duration difference is less than or equal to the above duration difference threshold to obtain the initial global trajectory. In the case where the target duration difference is less than or equal to the duration difference threshold, add speed information to each path point according to the global speed curve to obtain the above initial global trajectory of the vehicle.

[0077] In another embodiment of the present application, perform the above trapezoidal speed planning on the above target key points on each of the above path segments at least according to the above road section speed limit information to obtain the global speed curve of the above global driving path, including: a third planning step, perform kinematic speed planning based at least on the speed information of the above target key points on each of the above path segments and the corresponding above road section speed limit information to obtain a plurality of initial path speed curves, where one of the above path segments corresponds to one of the above initial path speed curves; an optimization step, perform optimization processing on each of the above initial path speed curves to obtain an initial global speed curve; an iteration step, perform iteration processing on the above initial global speed curve to obtain the above global speed curve. In this embodiment, perform optimization processing on the obtained plurality of initial path speed curves, which ensures that the obtained initial global speed curve can meet the requirements for the smoothness of the speed curve and also makes the comfort better. Then perform iteration processing on the initial global speed curve, which ensures that the obtained global speed curve can meet the requirement of arriving on time.

[0078] In a specific embodiment of the present application, before kinematic speed planning is performed based at least on the speed information of the target key points on each path segment and the corresponding road section speed limit information to obtain multiple initial path speed curves, it is also possible to determine whether the planning of each path segment can be completed. The specific process is as follows: Select two target key points (i.e., the first target key point and the second target key point) of a certain path segment. Assume that v0 is the speed on the left side of the first target key point, v1 is the speed on the right side of the first target key point, and v2 is the speed on the right side of the second target key point. If v1 is not equal to the speed on the left side of the second target key point, since there are two different maximum speeds between the two target key points, the target key points are set incorrectly and the planning fails. If v0 > v1 + 0.05, since v0 is the vehicle speed on the right side after the planning of the previous path segment or the actual initial vehicle speed, it cannot be greater than the maximum vehicle speed of this path segment, so the planning fails. If neither of the above two situations is satisfied, kinematic speed planning is started.

[0079] In another specific embodiment of the present application, the kinematic speed planning takes the speed information of the target key points corresponding to a certain path segment (this speed information is the speed v0 on the left side of the first target key point, the speed v1 on the right side of the first target key point, and the speed v2 on the right side of the second target key point), the maximum acceleration a max and the maximum deceleration a min as inputs and outputs the initial path speed curve of this path segment. The following is an introduction by cases:

[0080] Case 1:

[0081] When |v0 - v1| < 0.01 and v2 ≥ v1, let the path length corresponding to the first target key point be S0, and the path length corresponding to the second target key point be S2. In this case, continuing to maintain v1 can meet the operation requirements. At this time, sparse speed path points are generated between the two target key points, and v1 is used as the speed of this path segment. The obtained initial path speed curve is as follows Figure 5 as shown.

[0082] Case 2:

[0083] When |v0 - v1| < 0.01 and v2 < v1, let the path length corresponding to the first target key point be S0, and the path length corresponding to the second target key point be S2. At this time, the distance between the two target key points is S2 - S0, and according to the maximum deceleration a max the shortest distance S mind to decelerate to v2 can be calculated as:

[0084]

[0085] If S2 - S0 < S mind, the minimum deceleration distance is not satisfied and the planning fails; if S2 - S0 > S mind , take v2 as the final speed after planning. When the vehicle maintains v1 and runs to S1, it starts to decelerate, and S1 = S2 - S mind , then during the deceleration to v2, let s be the path length where the vehicle is currently located, and the expression for generating sparse speed path points is

[0086]

[0087] Then, based on the first target key point, the second target key point, and the generated sparse speed path points, obtain Figure 6 the initial path speed curve as shown in

[0088] Case 3:

[0089] There are two special cases (i) and (ii) in Case 3.

[0090] When v0 < v1 and v2 ≥ v1, let the path length corresponding to the first target key point be S1, the path length corresponding to the second target key point be S3, and s be the path length where the vehicle is currently located. Then the distance between the two target key points is S3 - S1. According to the maximum acceleration a max the shortest distance to accelerate to v2 can be calculated as S mind is:

[0091]

[0092] (i), if S3 - S1 > S mind , at this time v1 is the final speed after planning. Let the vehicle start to accelerate when it runs to S2, and S2 = S3 - S mind , and the expression for generating sparse speed path points between the two target key points at this time is:

[0093]

[0094] Then, based on the first target key point, the second target key point, and the generated sparse speed path points, obtain Figure 7 the initial path speed curve as shown in

[0095] (ii), if S3 - S1 < S mind , then it is necessary to keep accelerating between the two target key points. At this time, the expression for the planned final speed v last is:

[0096]

[0097] And the expression for generating sparse speed path points between the two target key points is:

[0098]

[0099] Then, based on the first target key point, the second target key point, and the generated sparse velocity path points, obtain Figure 8 the initial path velocity curve shown in

[0100] Case 4:

[0101] Case 4 contains two special cases (i) and (ii).

[0102] When v0 < v1 and v2 < v1 and v2 ≤ v0, let the path length corresponding to the first target key point be S0, the path length corresponding to the second target key point be S3, and s be the path length where the vehicle is currently located. Then the distance between the two key points S3 - S0, according to the maximum acceleration a max The shortest distance to accelerate to v1 can be calculated as S acc is:

[0103]

[0104] According to the maximum deceleration a min The shortest distance to decelerate to v2 can be calculated as S dec :

[0105]

[0106] (i), If (S acc + S dec ) < (S3 - S0), there is a path segment of uniform motion at this time. The expression for generating sparse velocity path points during the acceleration to v1 is:

[0107]

[0108] Travel at a uniform speed of v1 during the uniform motion process; the expression for generating sparse velocity path points during the deceleration to v2 is:

[0109]

[0110] Then, based on the first target key point, the second target key point, and the generated sparse velocity path points, obtain Figure 9 the initial path velocity curve shown in

[0111] (ii), If (S acc + S dec ) ≥ (S3 - S0), there is no path segment of uniform motion at this time. Let S′ be the path length corresponding to the start of deceleration, and the maximum speed that can be accelerated to at this time is set as v′. Then the expression for v′ is:

[0112]

[0113] Let the acceleration distance be S1, and S1 = S′ - S0, then the expression for S1 is:

[0114]

[0115] The expression for generating sparse velocity path points during the acceleration to v′ is:

[0116]

[0117] And the expression for generating sparse velocity path points during the deceleration process is:

[0118]

[0119] Then, based on the first target key point, the second target key point, and the generated sparse velocity path points, the Figure 10 initial path velocity curve shown is obtained.

[0120] Case 5:

[0121] There are three special cases (i), (ii), and (iii) in Case 5

[0122] When v0 < v1 and v2 < v1 and v2 > v0, let the path length corresponding to the first target key point be S0, the path length corresponding to the second target key point be S3, and s be the current path length of the vehicle. Then the distance between the two target key points is (S3 - S0). According to the maximum acceleration a max The shortest distance S from v0 to v1 can be calculated as acc :

[0123]

[0124] According to the maximum deceleration a min The shortest distance S from v1 to v2 can be calculated as dec :

[0125]

[0126] Then, according to the maximum acceleration a max The shortest distance S from v0 to v2 can be calculated as min :

[0127]

[0128] (i), If (S acc + S dec) < (S3 - S0), then there is a path segment with uniform motion in this case. The expression for generating sparse velocity path points during the acceleration to v1 is:

[0129]

[0130] During the uniform motion process, it travels at a constant speed of v1; the expression for generating sparse velocity path points during the deceleration to v2 is:

[0131]

[0132] Then, according to the first target key point, the second target key point, and the generated sparse velocity path points, the Figure 11 shown initial path velocity curve is obtained.

[0133] (ii) If (S acc + S dec ) ≥ (S3 - S0), there is no path segment with uniform motion in this case. Assume S′ is the path length corresponding to the start of deceleration, and the maximum speed that can be accelerated to at this time is set as v′, then the expression for v′ is:

[0134]

[0135] Let the acceleration distance be S1, S1 = S′ - S0, then the expression for S1 is:

[0136]

[0137] The expression for generating sparse velocity path points during the acceleration to v′ is:

[0138]

[0139] The expression for generating sparse velocity path points during the deceleration process is:

[0140]

[0141] Then, according to the first target key point, the second target key point, and the generated sparse velocity path points, the Figure 12 shown initial path velocity curve is obtained.

[0142] (iii) If S min ≥ (S3 - S0), then it accelerates all the time in this case. The expression for the planned final velocity v last is:

[0143]

[0144] And the expression for generating sparse velocity path points between the two target key points is:

[0145]

[0146] Then, based on the first target key point, the second target key point, and the generated sparse velocity path points, the Figure 13 shown initial path velocity curve is obtained.

[0147] In an embodiment of the present application, the above initial path velocity curves are optimized to obtain an initial global velocity curve, including: constructing an optimization function based on the velocity information of at least two adjacent path points to obtain a plurality of optimization objective functions; using the gradient descent method to optimize and solve the plurality of optimization objective functions to obtain the above initial global velocity curve.

[0148] In the actual application process, although each initial global velocity curve obtained by using the kinematic velocity planning method can meet the requirements of the speed limit of the road section of the path segment and the constraints of the maximum acceleration and / or maximum deceleration. However, due to the maximum acceleration and / or maximum deceleration, the speed change of the obtained initial global velocity curve is not smooth, which will affect the driving comfort. To improve the driving comfort, in a specific embodiment of the present application, the gradient descent method is used to optimize each initial path velocity curve. Specifically, by establishing an optimization objective function and using the gradient descent method to find the optimal solution, a smooth initial global velocity curve between path segments is generated.

[0149] (1) Establish an optimization objective function

[0150] To achieve the purpose of smooth speed change between path segments, in a specific embodiment, an optimization objective function is established according to the speeds of adjacent path points in the path segment:

[0151]

[0152] Among them, the function f1(v) reflects the magnitude of the speed change between adjacent path points, and the function f2(v) imposes certain constraints on the speeds before and after optimization, so that the range of speed curve optimization is in the adjacent area of the original speed curve. Δv i = v i - v i-1 is the speed change value between path point i and the adjacent rear path point, Δv i+1 = v i+1 - v i is the speed change value between path point i and the adjacent front path point, v i，0 represents the speed of path point i before optimization, that is, the speed obtained by kinematic velocity planning.

[0153] (2) Use the gradient descent method to find the optimal solution

[0154] By taking the partial derivative of the objective function of path point i with respect to velocity v i to solve for the gradient value at this path point. The direction of the gradient is the direction in which the function changes fastest. The gradient of the objective function with respect to v at path point i i is obtained as follows:

[0155]

[0156] The velocity smoothed along the gradient direction is expressed as:

[0157]

[0158] where δ represents the iteration step size along the gradient direction.

[0159] In order to further ensure that the vehicle can reach the destination on time when driving along the initial global trajectory, in another embodiment of the present application, the above initial global velocity curve is iteratively processed to obtain the above global velocity curve, including: determining the actual path running duration of the alternative path point according to the above initial global velocity curve, where the alternative path point is one of the multiple above path points; calculating the difference between the actual path running duration and the planned path running duration to obtain a path duration difference, where the planned path running duration is the running duration of the planned alternative path point; in the case where the path duration difference is greater than the predetermined threshold, correcting the speed limit information of the section to which the alternative path point belongs according to the corrected speed limit velocity to obtain corrected section speed limit information, and performing the above third planning step, the above optimization step, and the above iterative step at least once until the difference between the global trajectory duration of the global velocity curve and the total running duration satisfies a preset range, where the corrected speed limit velocity is determined at least according to the initial corrected speed limit velocity, the actual path running duration, and the planned path running duration.

[0160] During the actual operation, through the above trapezoidal velocity planning, a smooth initial global velocity curve that meets the section speed limit requirements and the maximum acceleration and / or maximum deceleration constraints of the path segment can be obtained. However, using the path speed limit information corresponding to the path segment as the highest speed limit, the global trajectory duration T v (k) obtained from the initial global velocity curve will be much smaller than the predetermined threshold T d (k), that is, the vehicle will arrive at the alternative path point or the destination in advance and cannot meet the constraint of on-time operation. Therefore, in a specific embodiment of the present application, a global velocity curve that meets the operation time constraint is planned by a combined iterative method of trapezoidal velocity planning and gradient descent method. The specific flowchart can be seen in Figure 4 as shown.

[0161] Specifically, as Figure 14As shown in the figure, in the initial global velocity curve 500 smoothed using the gradient descent method, the calculation formula for the actual path running duration corresponding to the alternative path points is as follows:

[0162]

[0163] where s i and v i respectively represent the path length and velocity corresponding to path point i.

[0164] Specifically, when the difference between the actual path running duration T v (k) of the alternative path point and the planned path running duration T d (k) is greater than the predetermined threshold ΔT max , the speed limit information v max of the corresponding path segment is corrected by the initially given corrected speed limit Δv to obtain the corrected speed limit information for the path segment. The specific calculation expression for the corrected speed limit information v max is:

[0165]

[0166] n is the total number of path points. Again, trapezoidal velocity planning is performed based on the corrected speed limit information for the path segment to obtain the actual path running duration T′ v (k) corresponding to the iterated alternative path point. Again, Δv is iterated based on the difference between T′ v (k) and T d (k). Among them, the iteration step expression of Δv is:

[0167]

[0168] Then, after obtaining the corrected speed limit information for the path segment again, trapezoidal velocity planning can be used again for velocity planning until the difference between the global trajectory duration consumed by the planned global velocity curve 600 and the total running duration meets the preset range, and the global velocity curve 600 that meets the on-time operation is obtained as Figure 14 shown in the figure.

[0169] The embodiment of the present application also provides a global velocity planning device. It should be noted that the global velocity planning device in the embodiment of the present application can be used to execute the global velocity planning method provided in the embodiment of the present application. The following introduces the global velocity planning device provided in the embodiment of the present application.

[0170] Figure 15 is a schematic diagram of the global velocity planning device according to the embodiment of the present application. As Figure 15 shown, the global velocity planning device includes:

[0171] The planning unit 10 is used for the first planning step. It adopts a speed planning algorithm to perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the received vehicle, and obtains an initial global trajectory. The above initial global trajectory is used to represent the speed information and time information of each path point. The above path points are the points divided at a predetermined interval on the above global driving path;

[0172] The acquisition unit 20 is used for the acquisition step. It controls the above vehicle to drive according to the above initial global trajectory to obtain the actual running duration of the above vehicle reaching the target path point. The above target path point is one of the multiple above path points;

[0173] The first comparison unit 30 is used for the first comparison step. When the difference between the above actual running duration and the path point planning duration is less than or equal to a predetermined threshold, the above acquisition step is executed at least once until the above vehicle reaches the destination. The above path point planning duration is the running duration of the above target path point in the initial planning;

[0174] The second comparison unit 40 is used for the second comparison step. When the difference between the above actual running duration and the above path point planning duration is greater than the above predetermined threshold, calculate the remaining total running duration and the remaining global path, and sequentially execute the above first planning step and the above acquisition step at least once until the above vehicle reaches the above destination.

[0175] In the above global speed planning device, the planning unit is used to perform speed planning on the total running duration, departure time, global driving path, and section speed limit information of the received vehicle based on a speed planning algorithm to obtain an initial global trajectory; the acquisition unit is used to obtain the actual running duration of the vehicle reaching the target path point during the process of the vehicle driving along the initial global trajectory; the first comparison unit is used to calculate the difference between the actual running duration of the vehicle at the target path point and the path point planning duration. If the difference is less than or equal to a predetermined threshold, it indicates that the vehicle can reach the destination on time. Therefore, the vehicle can continue to be controlled to run along the initial global trajectory until the vehicle reaches the destination; the second comparison unit is used to, when the difference is greater than the predetermined threshold, it indicates that the vehicle cannot reach the target on time. Calculate the remaining total running duration and the remaining global path of the vehicle, and then perform re-speed planning based on the remaining total running duration, the remaining global path, the departure time, and the section speed limit information, so that the vehicle can reach the destination on time. Compared with the existing solution of obtaining the global trajectory of the vehicle through dynamic programming or heuristic optimization methods, the global speed planning device of the present application only needs to use a speed planning algorithm to perform speed planning on the total running duration, departure time, global driving path, and section speed limit information of the vehicle to obtain the initial global trajectory of the vehicle. This ensures that the computational complexity of determining the initial global trajectory is small and does not require being mounted on a high-computing power platform at the vehicle specification level, ensuring that the cost of the global speed planning method of the present application is low. In addition, by considering the requirement of the vehicle to reach on time, the global speed planning device of the present application dynamically adjusts the running strategy of the vehicle under the influence of dynamic traffic flow, ensuring that the vehicle can reach the destination on time, thus solving the problem in the prior art that it is difficult to determine the global trajectory of the vehicle with less computational complexity while considering the requirement of the vehicle to reach on time.

[0176] In the actual application process, the above section speed limit information can be adjusted in real time according to different sections. The above total running duration is the total running duration required for the vehicle to reach the destination from the starting point. The above global driving path is determined by the global path planning system. For the global driving path, it is only the path of the vehicle driving and does not have speed information and time information. The initial global trajectory obtained in the present application is to add the speed information and time information corresponding to each path point on the basis of the global driving path. In addition, in the present application, the time information in the initial global trajectory can be the arrival time of the vehicle at each path point.

[0177] Specifically, the present application does not limit the size of the above predetermined threshold, and it can be flexibly adjusted according to the actual driving situation of the vehicle.

[0178] In a specific embodiment of the present application, such as Figure 2As shown in the figure, the global speed planning method of the present application is applied in the global speed planning system 300. The scheduling system 100 makes decisions on the destination, departure time, and total running duration of the vehicle, and the scheduling system 100 sends the determined destination to the global path planning system 200. The global path planning system 200 plans the global driving path for the vehicle to reach the destination based on the destination, starting point location information, and high-precision map information, and obtains the speed limit information for the road sections along the global driving path. The global path planning system 200 sends the global driving path and the road section speed limit information to the global speed planning system 300. For the global speed planning system 300, it receives the total running duration and departure time sent by the scheduling system 100, and receives the global driving path and road section speed limit information sent by the global path planning system 200. The global speed planning system 300 uses a speed planning algorithm to perform speed planning on the received total running duration, departure time, global driving path, and road section speed limit information, obtaining the speed curve and time series of the vehicle along the global path, thereby obtaining the initial global trajectory 400 of the vehicle.

[0179] In another specific embodiment of the present application, a global speed planning method is involved, as specifically shown in Figure 3 the figure. Among them, the global speed planning method includes a punctual arrival speed planning part and a global speed rolling planning part during the vehicle operation. The punctual arrival speed planning part can meet the total running duration requirement, road section speed limit requirement, and vehicle operation comfort requirement, obtaining the initial global trajectory. If the vehicle strictly follows the initial global trajectory, it can arrive at the destination on time. The global speed rolling planning part considers that the vehicle is affected by the dynamic traffic flow during operation, which may cause the vehicle to arrive at a certain target path point either late or early during the execution of the initial global trajectory. If the difference between the actual running duration of the vehicle reaching the target path point and the path point planned running duration of this target path point is greater than a predetermined threshold, the global speed planning system will roll and execute the speed planning algorithm according to the remaining total running duration requirement, remaining global path, etc., thereby dynamically adjusting the global trajectory of the vehicle to obtain a corrected global trajectory, and further ensuring that the vehicle arrives at the destination on time. The specific steps are as follows:

[0180] S1. Receive the total running duration T sent by the scheduling system d , the departure time T s , and receive the global driving path given by the global path planning system, as well as the road section speed limit information;

[0181] S2. Use the speed planning algorithm for the total running duration T d , the departure time T s , the global path, and the road section speed limit information to obtain the initial global trajectory, and control the vehicle to start executing the initial global trajectory;

[0182] S3. Calculate the actual running duration T for the vehicle to reach the target waypoint v (k) The planned running duration T of the waypoint corresponding to the target waypoint d (k) The difference T v (k) - T d (k). If the difference T v (k) - T d (k) The absolute value of is greater than the predetermined threshold ΔT max (i.e., |T v (k) - T d (k)| > ΔT max ), then calculate the remaining total running duration T d,r = T d - T v (k), and intercept the remaining global path. Call the speed planning algorithm again to obtain the corrected global trajectory of the remaining global path, and the vehicle starts to execute the corrected global trajectory. If the difference T v (k) - T d (k) The absolute value of is less than or equal to the predetermined threshold ΔT max (i.e., |T v (k) - T d (k)| ≤ ΔT max ), then control the vehicle to continue executing the initial global trajectory;

[0183] S4. Determine whether the vehicle has reached the destination. If the vehicle accurately reaches the destination, end the program; otherwise, return to step S3.

[0184] In an embodiment of the present application, the above-mentioned planning unit includes a division module, an acquisition module, a preprocessing module, and a planning module. Among them, the above-mentioned division module is used to divide the above-mentioned global driving path according to a predetermined rule to obtain a plurality of path segments, and the above-mentioned predetermined rule is a rule determined according to the forward and backward movement of the above-mentioned vehicle; the above-mentioned acquisition module is used to acquire at least two key points on each of the above-mentioned path segments, and the above-mentioned key point is one of the above-mentioned multiple path points; the above-mentioned preprocessing module is used to preprocess at least two of the above-mentioned key points on each of the above-mentioned path segments to obtain the target key points on each of the above-mentioned path segments; the above-mentioned planning module is used to perform trapezoidal speed planning according to the above-mentioned target key points on each of the above-mentioned path segments, the above-mentioned total running duration, the above-mentioned departure time, the above-mentioned global driving path, and the above-mentioned road section speed limit information to obtain the above-mentioned initial global trajectory of the above-mentioned vehicle. In this embodiment, according to the road section speed limit information, the key points on each path segment are preprocessed, so as to eliminate the keys that will cause the planning to fail, and it can be ensured that the obtained target key points can successfully perform global speed planning. Then, the target key points on each path segment, the total running duration, the departure time, the global driving path, and the road section speed limit information are used for trapezoidal speed planning to obtain the initial global trajectory, which ensures that the obtained initial global trajectory is relatively reasonable and can further meet the requirement of arriving at the destination on time.

[0185] Specifically, the above-mentioned predetermined rule is a rule determined according to the forward and backward movement of the vehicle. The forward and backward movement of the vehicle can be determined according to the direction of the vehicle's steering wheel.

[0186] In the actual application process, there may be more than two key points in each path segment, and each two key points can determine a path segment with a single speed limit. For some path segments where the right-side speed (i.e., the third speed) requirement of the second key point cannot be achieved even by deceleration, these path segments cannot obtain a feasible planned speed. Therefore, in another embodiment of the present application, the above preprocessing module includes a first determination sub-module, a first calculation sub-module, and a second calculation sub-module. Among them, the first determination sub-module is used to determine the first speed and the second speed of the first key point, and the third speed of the second key point for the consecutive first key point and second key point on the target path segment, where the target path segment is one of the multiple path segments, and both the first key point and the second key point are the above key points; the first calculation sub-module is used to calculate the first distance from the second speed to the third speed based on the maximum deceleration when the first speed is greater than the second speed and the second speed is greater than the third speed, and determine the target key point on the target path segment at least according to the first distance; the second calculation sub-module is used to calculate the second distance from the first speed to the third speed based on the maximum deceleration when the second speed is greater than the first speed and the first speed is greater than the third speed, and determine the target key point on the target path segment at least according to the second distance, so as to ensure that the vehicle can be successfully globally speed-planned according to the determined target key points subsequently.

[0187] Specifically, the first speed of the first key point can be the left-side speed v0 of the first key point. The second speed of the first key point can be the right-side speed v1 of the first key point. The third speed of the second key point is the right-side speed v2 of the second key point.

[0188] In another embodiment of the present application, the first calculation sub-module includes a third calculation sub-module, a second determination sub-module, and a third determination sub-module. Among them, the third calculation sub-module is used to calculate the distance between the first key point and the second key point to obtain a target key distance; the second determination sub-module is used to determine the second key point as the target key point when the first distance is greater than the target key distance, and determine the fourth speed of the second key point as the first speed, and delete the first key point; the third determination sub-module is used to determine both the first key point and the second key point as the target key points when the first distance is less than or equal to the target key distance. In this embodiment, when the first distance is greater than the target key distance, it indicates that even at the maximum deceleration, the requirement of the third speed (i.e., the right speed) of the second key point cannot be met. Therefore, the fourth speed of the second key point (i.e., the left speed of the second key point) can be determined as the first speed of the first key point (i.e., the left speed of the first key point), which ensures that the vehicle can be successfully globally speed-planned according to the target key points subsequently.

[0189] To further ensure that the vehicle can be successfully globally speed-planned according to the target key points subsequently. In another embodiment of the present application, the second calculation sub-module includes a fourth calculation sub-module, a fourth determination sub-module, and a fifth determination sub-module. Among them, the fourth calculation sub-module is used to calculate the distance between the first key point and the second key point to obtain a target key distance; the fourth determination sub-module is used to determine the second key point as the target key point when the second distance is greater than the target key distance, and determine the fourth speed of the second key point as the first speed, and delete the first key point; the fifth determination sub-module is used to determine both the first key point and the second key point as the target key points when the second distance is less than or equal to the target key distance.

[0190] In the actual application process, for two consecutive key points on the target path segment, namely the first key point and the second key point, assume that v0 is the left speed of the first key point, v1 is the right speed of the first key point, and v2 is the right speed of the second key point.

[0191] If v0 > v1 and v1 > v2, then at the maximum deceleration a min The first distance S min1 from v1 to v2 decelerating is

[0192]

[0193] If the first distance S min1If it is greater than the target key distance (the distance between the first key point and the second key point), then the second key point is determined as the target key point, the fourth speed (left speed) of the second key point is assigned as v0, and the first key point is deleted; otherwise, no processing is performed (that is, the first key point and the second key point are determined as the target key points).

[0194] If v1 > v0 and v0 > v2, then at the maximum deceleration a min The second distance S for decelerating from v0 to v2 min2 The expression is as follows:

[0195]

[0196] If the second distance S min2 is greater than the target key distance (the distance between the first key point and the second key point), then the second key point is determined as the target key point, the fourth speed (left speed) of the second key point is assigned as v0, and the first key point is deleted; otherwise, no processing is performed (that is, the first key point and the second key point are determined as the target key points).

[0197] In an embodiment of the present application, the above planning module includes a first planning sub-module, a sixth determination sub-module, and a seventh determination sub-module. Among them, the first planning sub-module is used for the second planning step, and at least according to the above road section speed limit information, performs the above trapezoidal speed planning on the above target key points on each of the above path segments to obtain the global speed curve of the above global driving path; the sixth determination sub-module is used for the first determination step, and determines the global trajectory duration according to the above global speed curve and the above global driving path; the seventh determination sub-module is used for the second determination step, and determines the above initial global trajectory according to the above global trajectory duration, the above total running duration, and the above departure time. In this embodiment, the initial global trajectory is determined according to the global trajectory duration, the total running duration, and the departure time, so as to ensure that the subsequent vehicle can reach the destination on time according to the initial global trajectory.

[0198] Specifically, in the present application, there is no limitation on the specific method for determining the global trajectory duration according to the global speed curve and the global driving path, and it can be any feasible method in the prior art.

[0199] In another embodiment of the present application, the above-mentioned seventh determination sub-module includes a first generation sub-module and a fifth calculation sub-module. Among them, the first generation sub-module is configured to generate first planning failure information when the global trajectory duration is greater than the total running duration, and send the first planning failure information to the scheduling system; the fifth calculation sub-module is configured to calculate the difference between the global trajectory duration and the total running duration to obtain a target duration difference when the global trajectory duration is less than or equal to the total running duration, and determine the initial global trajectory of the vehicle at least based on the target duration difference and the departure time. In this embodiment, when the global trajectory running duration is greater than the total running duration, it indicates that even if the vehicle travels at the section speed limit, it cannot reach the destination on time, that is, the global speed planning fails this time. The first planning failure information can be sent to the scheduling system so that the scheduling system can re-make a decision on the total running duration. When the global trajectory duration is less than or equal to the total running duration, the initial global trajectory is determined according to the target duration difference, which can further ensure that the vehicle can reach the destination more punctually.

[0200] In order to further ensure that the vehicle can reach the destination more punctually, in another embodiment of the present application, the above-mentioned fifth calculation sub-module includes a second generation sub-module and a first correction sub-module. Among them, the second generation sub-module is configured to add the speed information and time information to each of the above path points according to the global speed curve and the departure time when the target duration difference is less than or equal to the duration difference threshold, to obtain the initial global trajectory of the vehicle; the first correction sub-module is configured to correct the section speed limit information according to the target duration difference when the target duration difference is greater than the duration difference threshold, and execute the second planning step, the first determination step and the second determination step at least once until the target duration difference is less than or equal to the duration difference threshold.

[0201] Specifically, in the present application, the size of the above duration difference threshold is not limited, and it can be flexibly adjusted according to the driving conditions of the vehicle.

[0202] In a specific embodiment of the present application, a method for determining the initial global trajectory of a vehicle is involved, specifically Figure 4 as shown. Among them, the specific steps for determining the initial global trajectory are as follows:

[0203] s1. Obtain the total running duration T d , the departure time T s , the global driving path, the path speed limit information v max , the maximum acceleration a of the vehicle that meets the comfort requirements max and the maximum deceleration a min;

[0204] S2. Divide the global driving path according to a predetermined rule (i.e., a rule for switching forward or backward) to obtain multiple path segments. Then, perform speed planning on each path segment respectively. During the process of performing speed planning on each path segment, except that the starting speed of the first path segment is the current driving speed of the vehicle, the starting speeds and ending speeds of the remaining path segments are both set to 0;

[0205] S3. For each path segment, obtain key points according to the speed limit information of each path segment, and the number of key points for each path segment should be greater than or equal to 2. The information of each key point includes the speed on the left side of the key point (or the speed that the vehicle can actually reach), the speed limit on the right side, and the serial number in the global path (since the key point is a point among multiple path points, the serial number of the key point is the serial number of the corresponding path point). Since the path segment between two adjacent key points is a path segment with a single speed limit, the starting speed, ending speed, and speed limit value of this path segment are determined by the speed on the left side of the first key point, the speed limit on the right side of the second key point, and the speed on the right side of the first key point respectively;

[0206] S4. Preprocess the key points obtained for each path segment to obtain the target key points on each path segment. The purpose of preprocessing is to eliminate the key points that may cause the planning to fail;

[0207] S5. Perform trapezoidal speed planning according to the above target key points, total running duration, global driving path, and speed limit information of each path segment to obtain the global speed curve. Then, determine the global trajectory duration according to the global speed curve and the global driving path. If the global trajectory duration is greater than the total running time, generate a first planning failure message and send the first planning failure message to the scheduling system so that the scheduling system can re - make a decision on the total running duration; when the global trajectory duration is less than or equal to the total running duration, calculate the difference between the global trajectory duration and the total running duration to obtain the target duration difference;

[0208] S6. When the target duration difference is greater than the duration difference threshold, correct the corresponding speed limit information according to the target duration difference, and after correcting the corresponding speed limit information, restart from step S1 until the above target duration difference is less than or equal to the above duration difference threshold to obtain the initial global trajectory. When the target duration difference is less than or equal to the duration difference threshold, add speed information to each path point according to the global speed curve to obtain the above initial global trajectory of the vehicle.

[0209] In another embodiment of the present application, the above-mentioned first planning sub-module includes a second planning sub-module, an optimization sub-module, and an iterative processing sub-module. Among them, the second planning sub-module is used for the third planning step, and kinematic speed planning is performed at least based on the speed information of the target key points on each of the above-mentioned path segments and the corresponding speed limit information of the above-mentioned road segments, to obtain a plurality of initial path speed curves, where one of the above-mentioned path segments corresponds to one of the above-mentioned initial path speed curves; the optimization sub-module is used for the optimization step to perform optimization processing on each of the above-mentioned initial path speed curves to obtain an initial global speed curve; the iterative processing sub-module is used for the iterative step to perform iterative processing on the above-mentioned initial global speed curve to obtain the above-mentioned global speed curve. In this embodiment, the obtained plurality of initial path speed curves are optimized, so as to ensure that the obtained initial global speed curve can meet the requirements for the smoothness of the speed curve, and at the same time, the comfort can be better. Then, iterative processing is performed on the initial global speed curve, so as to ensure that the obtained global speed curve can meet the requirement of arriving on time.

[0210] In a specific embodiment of the present application, before performing kinematic speed planning at least based on the speed information of the target key points on each path segment and the corresponding speed limit information of the road segment to obtain a plurality of initial path speed curves, it is also possible to determine whether the planning of each path segment can be completed. The specific process is as follows: Select two target key points of a certain path segment (i.e., the first target key point and the second target key point). Assume that v0 is the speed on the left side of the first target key point, v1 is the speed on the right side of the first target key point, and v2 is the speed on the right side of the second target key point. If v1 is not equal to the speed on the left side of the second target key point, then because there are two different maximum speeds between the two target key points, the target key points are set incorrectly and the planning fails. If v0 > v1 + 0.05, then since v0 is the speed on the right side of the previous path segment after planning or the actual initial speed, it cannot be greater than the maximum speed of this path segment, so the planning fails. If neither of the above two situations is satisfied, then kinematic speed planning is started.

[0211] In another specific embodiment of the present application, the kinematic speed planning takes the speed information of the target key points corresponding to a certain path segment (this speed information is the speed v0 on the left side of the first target key point, the speed v1 on the right side of the first target key point, and the speed v2 on the right side of the second target key point), the maximum acceleration a max and the maximum deceleration a min as inputs, and outputs the initial path speed curve of this path segment. The following is an introduction by cases:

[0212] Case 1:

[0213] When |v0 - v1| < 0.01 and v2 ≥ v1, let the path length corresponding to the first target key point be S0, and the path length corresponding to the second target key point be S2. In this case, continuing to maintain v1 can meet the running requirements. At this time, sparse speed path points are generated between the two target key points, and v1 is used as the speed of this path segment. The obtained initial path speed curve is as follows Figure 5 as shown

[0214] Case 2:

[0215] When |v0 - v1| < 0.01 and v2 < v1, let the path length corresponding to the first target key point be S0, and the path length corresponding to the second target key point be S2. At this time, the distance between the two target key points is S2 - S0, and according to the maximum deceleration a max the shortest distance to decelerate to v2 can be calculated as S mind which is

[0216]

[0217] If S2 - S0 < S mind , it does not meet the shortest deceleration distance and the planning fails; if S2 - S0 > S mind , v2 is used as the final speed after planning. The vehicle maintains v1 until it reaches S1 and then starts to decelerate. S1 = S2 - S mind . Then, during the deceleration process to v2, assuming s is the path length where the vehicle is currently located, the expression for generating sparse speed path points is

[0218]

[0219] Then, according to the first target key point, the second target key point, and the generated sparse speed path points, the Figure 6 initial path speed curve as shown is obtained

[0220] Case 3:

[0221] Case 3 contains two special cases (i) and (ii).

[0222] When v0 < v1 and v2 ≥ v1, let the path length corresponding to the first target key point be S1, and the path length corresponding to the second target key point be S 3，s which is the path length where the vehicle is currently located. Then the distance between the two target key points is S3 - S1. According to the maximum acceleration a max the shortest distance to accelerate to v2 can be calculated as S mind which is

[0223]

[0224] (i). If S3 - S1 > S mind , at this time, v1 is the planned final velocity. Assume that the vehicle starts to accelerate when it reaches S2, and S2 = S3 - S mind . At this time, the expression for generating sparse velocity path points between the two target key points is:

[0225]

[0226] Then, based on the first target key point, the second target key point, and the generated sparse velocity path points, obtain Figure 7 the initial path velocity curve shown.

[0227] (ii). If S3 - S1 < S mind , then it is necessary to keep accelerating between the two target key points. At this time, the expression for the planned final velocity v last is:

[0228]

[0229] And the expression for generating sparse velocity path points between the two target key points is:

[0230]

[0231] Then, based on the first target key point, the second target key point, and the generated sparse velocity path points, obtain Figure 8 the initial path velocity curve shown.

[0232] Case 4:

[0233] There are two special cases (i) and (ii) in Case 4.

[0234] When v0 < v1 and v2 < v1 and v2 ≤ v0, assume that the path length corresponding to the first target key point is S0, the path length corresponding to the second target key point is S3, and s is the current path length of the vehicle. Then the distance between the two key points S3 - S0. According to the maximum acceleration a max the shortest distance S acc to accelerate to v1 can be calculated as:

[0235]

[0236] According to the maximum deceleration a min the shortest distance S dec to decelerate to v2 can be calculated as:

[0237]

[0238] (i). If (S acc + S dec) < (S3 - S0), and there is a path segment with uniform speed at this time. The expression for generating sparse speed path points during the acceleration to v1 is:

[0239]

[0240] During the uniform speed process, it travels at a speed of v1; the expression for generating sparse speed path points during the deceleration to v2 is:

[0241]

[0242] Then, according to the first target key point, the second target key point, and the generated sparse speed path points, obtain Figure 9 The initial path speed curve shown.

[0243] (ii) If (S acc + S dec ) ≥ (S3 - S0), there is no path segment with uniform speed at this time. Let S′ be the path length corresponding to the start of deceleration, and let the maximum speed that can be accelerated to at this time be v′, then the expression for v′ is:

[0244]

[0245] Let the acceleration distance be S1, S1 = S′ - S0, then the expression for S1 is:

[0246]

[0247] The expression for generating sparse speed path points during the acceleration to v′ is:

[0248]

[0249] And the expression for generating sparse speed path points during the deceleration process is:

[0250]

[0251] Then, according to the first target key point, the second target key point, and the generated sparse speed path points, obtain Figure 10 The initial path speed curve shown.

[0252] Case 5:

[0253] There are three special cases (i), (ii), and (iii) in Case 5

[0254] When v0 < v1, v2 < v1, and V2 > V0, let the path length corresponding to the first target key point be S0, the path length corresponding to the second target key point be S3, and s be the path length where the vehicle is currently located. Then the distance between the two target key points is (S3 - S0). According to the maximum acceleration a max The shortest distance from v0 to v1 can be calculated as S acc is:

[0255]

[0256] According to the maximum deceleration a min The shortest distance from v1 to v2 can be calculated as S dec is:

[0257]

[0258] Then, according to the maximum acceleration a max The shortest distance S from v0 to v2 can be calculated min is:

[0259]

[0260] (i) If (S acc + S dec ) < (S3 - S0), then there is a path segment of uniform motion in this case. The expression for generating sparse velocity path points during the acceleration to v1 is:

[0261]

[0262] During the uniform motion, it travels at a constant speed of v1; the expression for generating sparse velocity path points during the deceleration to v2 is:

[0263]

[0264] Then, based on the first target key point, the second target key point, and the generated sparse velocity path points, Figure 11 the initial path velocity curve shown in

[0265] (ii) If (S acc + S dec ) ≥ (S3 - S0), there is no path segment of uniform motion in this case. Assume S′ is the path length corresponding to the start of deceleration, and the maximum speed that can be accelerated to at this time is set as v′. Then the expression for v′ is:

[0266]

[0267] Let the acceleration distance be S1, S1 = S′ - S0. Then the expression for S1 is:

[0268]

[0269] The expression for generating sparse velocity path points during the acceleration to v' is:

[0270]

[0271] The expression for generating sparse velocity path points during the deceleration process is:

[0272]

[0273] Then, based on the first target key point, the second target key point, and the generated sparse velocity path points, obtain Figure 12 the initial path velocity curve shown.

[0274] (iii), if S min ≥(S3 - S0), then keep accelerating in this case. The expression for the planned final velocity v last is:

[0275]

[0276] The expression for generating sparse velocity path points between the two target key points is:

[0277]

[0278] Then, based on the first target key point, the second target key point, and the generated sparse velocity path points, obtain Figure 13 the initial path velocity curve shown.

[0279] In an embodiment of the present application, the above optimization sub-module includes a construction sub-module and a solution sub-module. Among them, the above construction sub-module is used to construct an optimization function based on at least the velocity information of two adjacent path points to obtain a plurality of optimization objective functions; the above solution sub-module is used to optimize and solve the plurality of optimization objective functions by using the gradient descent method to obtain the above initial global velocity curve.

[0280] In the actual application process, although the initial global velocity curves obtained by using the kinematic velocity planning method can meet the requirements of the speed limits of the road segments of the path segments and the constraints of the maximum acceleration and / or maximum deceleration, due to the maximum acceleration and / or maximum deceleration, the velocity changes of the obtained initial global velocity curves are not smooth, which will affect the driving comfort. In order to improve the driving comfort, in a specific embodiment of the present application, the gradient descent method is used to optimize each initial path velocity curve. Specifically, by establishing an optimization objective function and using the gradient descent method to find the optimal solution, a smooth initial global velocity curve between the path segments is generated.

[0281] (1) Establish an optimization objective function

[0282] In order to achieve the purpose of smooth velocity change between each path segment, in a specific embodiment, an optimization objective function is established according to the velocities of adjacent path points in the path segment:

[0283]

[0284] Among them, the function f1(v) reflects the magnitude of the velocity change between adjacent path points, and the function f2(v) imposes certain constraints on the velocities before and after optimization, so that the range of velocity curve optimization is in the adjacent area of the original velocity curve. Δv i = v i - v i-1 is the velocity change value between path point i and the adjacent rear path point, and Δv i+1 = v i+1 - v i is the velocity change value between path point i and the adjacent front path point, and v i，0 represents the velocity of path point i before optimization, that is, the velocity obtained by kinematic velocity planning.

[0285] (2) Use the gradient descent method to find the optimal solution

[0286] By taking the partial derivative of the objective function of path point i with respect to the velocity v i to solve the gradient value at this path point, the direction of the gradient is the direction in which the function changes fastest. The gradient of the objective function at path point i with respect to v i is:

[0287]

[0288] The velocity smoothed along the gradient direction is expressed as:

[0289]

[0290] Among them, δ represents the iteration step size along the gradient direction.

[0291] In order to further ensure that the vehicle can reach the destination on time when traveling along the initial global trajectory, in another embodiment of the present application, the above iterative sub-module includes an eighth determination sub-module, a sixth calculation sub-module, and a second correction sub-module. Among them, the eighth determination sub-module is used to determine the actual path running duration of an alternative path point according to the above initial global speed curve, and the alternative path point is one of the multiple above path points; the sixth calculation sub-module is used to calculate the difference between the actual path running duration and the planned path running duration to obtain a path duration difference, where the planned path running duration is the running duration of the planned alternative path point; the second correction sub-module is used to, when the path duration difference is greater than the above predetermined threshold, correct the speed limit information of the section to which the alternative path point belongs according to the corrected speed limit, obtain corrected section speed limit information, and execute the above third planning step, the above optimization step, and the above iterative step at least once until the difference between the global trajectory duration of the global speed curve and the above total running duration meets the preset range, where the corrected speed limit is determined at least according to the initial corrected speed limit, the above actual path running duration, and the above planned path running duration.

[0292] During the actual operation process, through the above trapezoidal speed planning, a smooth initial global speed curve that meets the section speed limit requirements and the maximum acceleration and / or maximum deceleration constraints of the path segment can be obtained. However, using the path speed limit information corresponding to the path segment as the upper limit of the maximum speed, the global trajectory duration T v (k) obtained from the initial global speed curve will be much smaller than the predetermined threshold T d (k), that is, the vehicle will arrive at the alternative path point or the destination in advance and cannot meet the constraint of on-time operation. Therefore, in a specific embodiment of the present application, a combined iterative method of trapezoidal speed planning and gradient descent method is used to plan a global speed curve that meets the operation time constraint. The specific flowchart can be seen in Figure 4 as shown.

[0293] Specifically, as Figure 14 shown, in the initial global speed curve 500 smoothed using the gradient descent method, the calculation formula for the actual path running duration corresponding to the alternative path point is as follows:

[0294]

[0295] where s i and v i respectively represent the path length and speed corresponding to path point i.

[0296] Specifically, when the actual path running duration T v (k) of the alternative path point and the corresponding planned path running duration Td When the difference of (k) is greater than a predetermined threshold ΔT max the speed limit information v of the corresponding path segment is corrected by the initially given corrected speed limit Δv max to obtain the corrected speed limit information of the path segment. The specific calculation formula for the corrected speed limit information v max is as follows:

[0297]

[0298] n is the total number of path points. Again, trapezoidal speed planning is performed according to the corrected speed limit information of the path segment to obtain the actual path running duration T′ corresponding to the iterated alternative path points v (k). Again, based on T′ v (k) and T d the difference of (k) is used to iterate Δv. Among them, the iteration step formula of Δv is:

[0299]

[0300] Then, after obtaining the corrected speed limit information of the path segment again, trapezoidal speed planning can be used again for speed planning until the difference between the global trajectory duration consumed by the planned global speed curve 600 and the total running duration meets the preset range, and the global speed curve 600 that meets the on-time operation is obtained as Figure 14 shown.

[0301] The above global speed planning device includes a processor and a memory. The above planning unit, acquisition unit, first comparison unit, second comparison unit, etc. are all stored in the memory as program units, and the corresponding functions are implemented by the processor executing the above program units stored in the memory.

[0302] The processor contains a kernel, and the corresponding program units are retrieved from the memory by the kernel. One or more kernels can be set, and by adjusting the kernel parameters, the problem that it is difficult to determine the global trajectory of the vehicle with less computational effort while considering the on-time arrival requirement of the vehicle in the prior art can be solved.

[0303] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of, for example, read-only memory (ROM) or flash memory (flash RAM), and the memory includes at least one storage chip.

[0304] An embodiment of the present invention provides a computer-readable storage medium, on which a program is stored, and when the program is executed by a processor, the above global speed planning method is implemented.

[0305] An embodiment of the present invention provides a processor, which is used to run a program. When the program runs, the above-mentioned global speed planning method is executed.

[0306] In a typical embodiment of the present application, a planning system is further provided. The planning system includes a scheduling system, a global path planning system, and a global speed planning system. Among them, the above-mentioned scheduling system is used to send the total running duration and departure time to the global speed planning system; the global path planning system, the above-mentioned global path planning system is used to send the global driving path and road section speed limit information to the above-mentioned global speed planning system; the above-mentioned global speed planning system includes a global speed planning device, and the above-mentioned global speed planning device is used to execute any one of the above-mentioned global speed planning methods.

[0307] The above-mentioned planning system includes a global speed planning system, and the global speed planning system includes a global speed planning device. The above-mentioned global speed planning device is used to execute any one of the above-mentioned global speed planning methods. In the above-mentioned global speed planning method, first, based on the speed planning algorithm, speed planning is performed on the total running duration, departure time, global driving path, and road section speed limit information of the received vehicle to obtain an initial global trajectory; then, during the process of the vehicle driving along the initial global trajectory, obtain the actual running duration of the vehicle reaching the target path point; then, calculate the difference between the actual running duration of the vehicle at the target path point and the path point planning duration. If the difference is less than or equal to the predetermined threshold, it means that the vehicle can reach the destination on time, so the vehicle can continue to be controlled to run along the initial global trajectory until the vehicle reaches the destination; if the difference is greater than the predetermined threshold, it means that the vehicle cannot reach the destination on time. Calculate the remaining total running duration and remaining global path of the vehicle, and then perform re-speed planning according to the remaining total running duration, remaining global path, departure time, and road section speed limit information, so that the vehicle can reach the destination on time. Compared with the method of obtaining the global trajectory of the vehicle by dynamic programming or heuristic optimization method in the prior art, the global speed planning method of the present application only needs to use the speed planning algorithm to perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the vehicle to obtain the initial global trajectory of the vehicle, which ensures that the calculation amount for determining the initial global trajectory is small and does not need to be carried on a high computing power platform at the vehicle specification level, ensuring that the cost of the global speed planning method of the present application is low. In addition, the global speed planning method of the present application dynamically adjusts the running strategy of the vehicle under the influence of dynamic traffic flow by considering the requirement of the vehicle to reach the destination on time, ensuring that the vehicle can reach the destination on time, thus solving the problem in the prior art that it is difficult to determine the global trajectory of the vehicle with less calculation amount while considering the requirement of the vehicle to reach the destination on time.

[0308] An embodiment of the present invention provides a device, which includes a processor, a memory, and a program stored on the memory and executable on the processor. When the processor executes the program, at least the following steps are implemented:

[0309] Step S101, the first planning step: Using a speed planning algorithm, perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the received vehicle to obtain an initial global trajectory. The initial global trajectory is used to represent the speed information and time information of each path point, and the path point is a point divided at a predetermined interval on the global driving path;

[0310] Step S102, the acquisition step: Control the vehicle to drive according to the initial global trajectory to obtain the actual running duration of the vehicle reaching the target path point. The target path point is one of the multiple path points;

[0311] Step S103, the first comparison step: When the difference between the actual running duration and the path point planning duration is less than or equal to a predetermined threshold, execute the acquisition step at least once until the vehicle reaches the destination. The path point planning duration is the running duration of the initially planned target path point;

[0312] Step S104, the second comparison step: When the difference between the actual running duration and the path point planning duration is greater than the predetermined threshold, calculate the remaining total running duration and the remaining global path, and sequentially execute the first planning step and the acquisition step at least once until the vehicle reaches the destination.

[0313] The device in this article can be a server, a PC, a PAD, a mobile phone, etc.

[0314] The present application also provides a computer program product, which, when executed on a data processing device, is adapted to execute a program initialized with at least the following method steps:

[0315] Step S101, the first planning step: Using a speed planning algorithm, perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the received vehicle to obtain an initial global trajectory. The initial global trajectory is used to represent the speed information and time information of each path point, and the path point is a point divided at a predetermined interval on the global driving path;

[0316] Step S102, the acquisition step: Control the vehicle to drive according to the initial global trajectory to obtain the actual running duration of the vehicle reaching the target path point. The target path point is one of the multiple path points;

[0317] Step S103, the first comparison step, when the difference between the actual running duration and the path point planning duration is less than or equal to a predetermined threshold, execute the above-mentioned acquisition step at least once until the vehicle reaches the destination, where the path point planning duration is the running duration of the initially planned target path point.

[0318] Step S104, the second comparison step, when the difference between the actual running duration and the path point planning duration is greater than the predetermined threshold, calculate the remaining total running duration and the remaining global path, and sequentially execute the above-mentioned first planning step and the above-mentioned acquisition step at least once until the vehicle reaches the destination.

[0319] In the above embodiments of the present invention, the descriptions of the respective embodiments have their own emphases. For parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0320] In several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the above-mentioned unit division can be a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of units or modules can be in an electrical or other form.

[0321] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0322] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0323] When the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the above methods in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0324] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0325] 1). In the global speed planning method of the present application, first, based on the speed planning algorithm, speed planning is performed on the total running duration, departure time, global driving path, and section speed limit information of the received vehicle to obtain an initial global trajectory. Then, during the process of the vehicle driving along the initial global trajectory, the actual running duration of the vehicle reaching the target path point is obtained. After that, the difference between the actual running duration of the vehicle at the target path point and the path point planning duration is calculated. If the difference is less than or equal to a predetermined threshold, it indicates that the vehicle can reach the destination on time, so the vehicle can continue to be controlled to run along the initial global trajectory until the vehicle reaches the destination. If the difference is greater than the predetermined threshold, it indicates that the vehicle cannot reach the destination on time. Calculate the remaining total running duration and remaining global path of the vehicle, and then perform re-speed planning according to the remaining total running duration, remaining global path, departure time, and section speed limit information to enable the vehicle to reach the destination on time. Compared with the method of obtaining the global trajectory of the vehicle through dynamic programming or heuristic optimization method in the prior art, the global speed planning method of the present application only needs to use the speed planning algorithm to perform speed planning on the total running duration, departure time, global driving path, and section speed limit information of the vehicle to obtain the initial global trajectory of the vehicle, which ensures that the calculation amount for determining the initial global trajectory is small and does not need to be carried on a high-computing-power platform at the vehicle specification level, ensuring that the cost of the global speed planning method of the present application is low. In addition, the global speed planning method of the present application dynamically adjusts the running strategy of the vehicle under the influence of dynamic traffic flow by considering the requirement of the vehicle to reach on time, ensuring that the vehicle can reach the destination on time, thus solving the problem in the prior art that it is difficult to determine the global trajectory of the vehicle with less calculation amount while considering the requirement of the vehicle to reach on time.

[0326] 2) In the global speed planning device of the present application, the planning unit is used to perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the received vehicle based on the speed planning algorithm to obtain an initial global trajectory; the acquisition unit is used to obtain the actual running duration of the vehicle reaching the target path point during the vehicle's driving along the initial global trajectory; the first comparison unit is used to calculate the difference between the actual running duration of the vehicle at the target path point and the path point planning duration. If the difference is less than or equal to the predetermined threshold, it indicates that the vehicle can reach the destination on time, so the vehicle can continue to be controlled to run along the initial global trajectory until the vehicle reaches the destination; the second comparison unit is used to, when the difference is greater than the predetermined threshold, which indicates that the vehicle cannot reach the destination on time, calculate the remaining total running duration and the remaining global path of the vehicle, and then perform re-speed planning based on the remaining total running duration, the remaining global path, the departure time, and the road section speed limit information so that the vehicle can reach the destination on time. Compared with the existing technology of obtaining the global trajectory of the vehicle through dynamic programming or heuristic optimization methods, the present application only needs to use the speed planning algorithm to perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the vehicle to obtain the initial global trajectory of the vehicle. This ensures that the computational amount for determining the initial global trajectory is small and does not require being carried on a high-computing-power platform at the vehicle specification level, ensuring that the cost of the global speed planning method of the present application is low. In addition, by considering the requirement of the vehicle to reach on time, the global speed planning device of the present application dynamically adjusts the running strategy of the vehicle under the influence of dynamic traffic flow, ensuring that the vehicle can reach the destination on time, thus solving the problem that it is difficult for the existing technology to determine the global trajectory of the vehicle with less computational amount while considering the requirement of the vehicle to reach on time.

[0327] 3) The planning system of this application includes a global speed planning system, and the global speed planning system includes a global speed planning device. The above-mentioned global speed planning device is used to execute any one of the above-mentioned global speed planning methods. In the above-mentioned global speed planning method, first, based on the speed planning algorithm, speed planning is performed on the total running duration, departure time, global driving path, and road section speed limit information of the received vehicle to obtain an initial global trajectory. Then, during the process of the vehicle driving along the initial global trajectory, the actual running duration of the vehicle reaching the target path point is obtained. After that, the difference between the actual running duration of the vehicle at the target path point and the path point planning duration is calculated. If the difference is less than or equal to the predetermined threshold, it indicates that the vehicle can reach the destination on time, so the vehicle can continue to be controlled to run along the initial global trajectory until it reaches the destination. If the difference is greater than the predetermined threshold, it indicates that the vehicle cannot reach the destination on time. Calculate the remaining total running duration and the remaining global path of the vehicle, and then perform re-speed planning according to the remaining total running duration, the remaining global path, the departure time, and the road section speed limit information, so that the vehicle can reach the destination on time. Compared with the method of obtaining the global trajectory of the vehicle by dynamic programming or heuristic optimization method in the prior art, the global speed planning method of this application only needs to use the speed planning algorithm to perform speed planning on the total running duration, departure time, global driving path, and road section speed limit information of the vehicle to obtain the initial global trajectory of the vehicle. This ensures that the computational amount for determining the initial global trajectory is small and does not require a high-computing power platform at the vehicle specification level, ensuring that the cost of the global speed planning method of this application is low. In addition, the global speed planning method of this application dynamically adjusts the running strategy of the vehicle under the influence of dynamic traffic flow by considering the requirement of the vehicle to reach on time, ensuring that the vehicle can reach the destination on time, thus solving the problem that it is difficult to determine the global trajectory of the vehicle with less computational amount while considering the requirement of the vehicle to reach on time in the prior art.

[0328] The above are only the preferred embodiments of this application and are not used to limit this application. For those skilled in the art, this application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included in the protection scope of this application.

Claims

1. A global speed planning method, characterized in that, Including: A first planning step, according to a predetermined rule, dividing the global driving path of the received vehicle to obtain a plurality of path segments, where the predetermined rule is a rule determined according to the forward and backward movement of the vehicle; obtaining at least two key points on each of the path segments, where the key points are one of a plurality of path points; for consecutive first key point and second key point on a target path segment, determining a first speed and a second speed of the first key point, and a third speed of the second key point, where the target path segment is one of the plurality of path segments, and both the first key point and the second key point are the key points; in the case where the first speed is greater than the second speed and the second speed is greater than the third speed, based on the maximum deceleration, calculating a first distance for decelerating from the second speed to the third speed, and determining the target key point on the target path segment at least according to the first distance; in the case where the second speed is greater than the first speed and the first speed is greater than the third speed, based on the maximum deceleration, calculating a second distance for decelerating from the first speed to the third speed, and determining the target key point on the target path segment at least according to the second distance; performing trapezoidal speed planning according to the target key points on each of the path segments, the total running duration, the departure time, the global driving path, and the road section speed limit information to obtain an initial global trajectory of the vehicle, where the initial global trajectory is used to represent the speed information and time information of each of the path points, and the path points are points divided at a predetermined interval on the global driving path; An obtaining step, controlling the vehicle to travel according to the initial global trajectory to obtain the actual running duration for the vehicle to reach a target path point, where the target path point is one of the plurality of path points; A first comparison step, in the case where the difference between the actual running duration and the path point planning duration is less than or equal to a predetermined threshold, performing the obtaining step at least once until the vehicle reaches the destination, where the path point planning duration is the running duration of the initially planned target path point; A second comparison step, in the case where the difference between the actual running duration and the path point planning duration is greater than the predetermined threshold, calculating the remaining total running duration and the remaining global path, and sequentially performing the first planning step and the obtaining step at least once until the vehicle reaches the destination.

2. The global speed planning method according to claim 1, wherein Determining the target key point on the target path segment at least according to the first distance includes: Calculating the distance between the first key point and the second key point to obtain a target key distance; In the case where the first distance is greater than the target key distance, determining the second key point as the target key point, determining the fourth speed of the second key point as the first speed, and deleting the first key point; In the case where the first distance is less than or equal to the target key distance, determining both the first key point and the second key point as the target key points.

3. The global speed planning method according to claim 1, wherein Determining the target key point on the target path segment according to at least the second distance includes: Calculating the distance between the first key point and the second key point to obtain a target key distance; When the second distance is greater than the target key distance, determining the second key point as the target key point, determining the fourth speed of the second key point as the first speed, and deleting the first key point; When the second distance is less than or equal to the target key distance, determining both the first key point and the second key point as the target key points.

4. The global speed planning method according to claim 1, wherein Performing trapezoidal speed planning according to the target key points on each path segment, the total running duration, the departure time, the global driving path, and the road section speed limit information to obtain the initial global trajectory of the vehicle, including: A second planning step of performing trapezoidal speed planning on the target key points on each path segment according to at least the road section speed limit information to obtain a global speed curve of the global driving path; A first determination step of determining a global trajectory duration according to the global speed curve and the global driving path; A second determination step of determining the initial global trajectory according to the global trajectory duration, the total running duration, and the departure time.

5. The global speed planning method according to claim 4, wherein Determining the initial global trajectory according to the global trajectory duration, the total running duration, and the departure time includes: When the global trajectory duration is greater than the total running duration, generating a first planning failure message and sending the first planning failure message to the dispatching system; When the global trajectory duration is less than or equal to the total running duration, calculating the difference between the global trajectory duration and the total running duration to obtain a target duration difference, and determining the initial global trajectory of the vehicle according to at least the target duration difference and the departure time.

6. The global speed planning method according to claim 5, wherein Determining the initial global trajectory of the vehicle according to at least the target duration difference and the departure time includes: When the target duration difference is less than or equal to a duration difference threshold, adding speed information and time information to each path point according to the global speed curve and the departure time to obtain the initial global trajectory of the vehicle; When the target duration difference is greater than the duration difference threshold, correcting the road section speed limit information according to the target duration difference, and performing the second planning step, the first determination step, and the second determination step at least once until the target duration difference is less than or equal to the duration difference threshold.

7. The global speed planning method according to claim 4, wherein Performing trapezoidal speed planning on the target key points on each path segment according to at least the road section speed limit information to obtain a global speed curve of the global driving path, including: A third planning step of performing kinematic speed planning based on at least the speed information of the target key points on each path segment to obtain a plurality of initial path speed curves, where one path segment corresponds to one initial path speed curve; Optimization step: Optimize each of the initial path speed curves to obtain an initial global speed curve; Iteration step: Based on the road section speed limit information, perform iterative processing on the initial global speed curve to obtain the global speed curve.

8. The global speed planning method according to claim 7, wherein, Optimizing each of the initial path speed curves to obtain an initial global speed curve includes: Constructing an optimization function based on the speed information of at least two adjacent path points to obtain multiple optimization objective functions; Using the gradient descent method to perform optimization solution on the multiple optimization objective functions to obtain the initial global speed curve.

9. The global speed planning method according to claim 7, wherein Based on the road section speed limit information, performing iterative processing on the initial global speed curve to obtain the global speed curve includes: Determining the actual path running duration of an alternative path point according to the initial global speed curve, where the alternative path point is one of the multiple path points; Calculating the difference between the actual path running duration and the planned path running duration to obtain a path duration difference, where the planned path running duration is the running duration of the planned alternative path point; When the path duration difference is greater than the predetermined threshold, correcting the road section speed limit information of the alternative path point according to the corrected speed limit speed to obtain corrected road section speed limit information, and performing the third planning step, the optimization step, and the iteration step at least once until the difference between the global trajectory duration of the global speed curve and the total running duration meets the preset range, where the corrected speed limit speed is determined based on at least the initial corrected speed limit speed, the actual path running duration, and the planned path running duration.

10. A global speed planning device, characterized in that, including: A planning unit, for a first planning step, divides the received global driving path of a vehicle according to a predetermined rule to obtain a plurality of path segments, where the predetermined rule is a rule determined according to the forward and backward movement of the vehicle; obtains at least two key points on each of the path segments, where the key points are one of a plurality of path points; for a consecutive first key point and second key point on a target path segment, determines a first speed and a second speed of the first key point, and a third speed of the second key point, where the target path segment is one of the plurality of path segments, and both the first key point and the second key point are the key points; in the case where the first speed is greater than the second speed and the second speed is greater than the third speed, calculates a first distance for decelerating from the second speed to the third speed based on the maximum deceleration, and determines the target key point on the target path segment at least according to the first distance; in the case where the second speed is greater than the first speed and the first speed is greater than the third speed, calculates a second distance for decelerating from the first speed to the third speed based on the maximum deceleration, and determines the target key point on the target path segment at least according to the second distance; performs trapezoidal speed planning according to the target key points on each of the path segments, the total running duration, the departure time, the global driving path, and the road section speed limit information to obtain an initial global trajectory of the vehicle, where the initial global trajectory is used to represent the speed information and time information of each of the path points, and the path points are points divided at a predetermined interval on the global driving path; An acquisition unit, for an acquisition step, controls the vehicle to travel according to the initial global trajectory to obtain the actual running duration of the vehicle reaching a target path point, where the target path point is one of the plurality of path points; A first comparison unit, for a first comparison step, in the case where the difference between the actual running duration and the path point planning duration is less than or equal to a predetermined threshold, executes the acquisition step at least once until the vehicle reaches the destination, and the path point planning duration is the running duration of the initially planned target path point; A second comparison unit, for a second comparison step, in the case where the difference between the actual running duration and the path point planning duration is greater than the predetermined threshold, calculates the remaining total running duration and the remaining global path, and sequentially executes the first planning step and the acquisition step at least once until the vehicle reaches the destination.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, where the program executes the global speed planning method according to any one of claims 1 to 9.

12. A processor, characterized in that, The processor is used to run a program, where the program runs to execute the global speed planning method according to any one of claims 1 to 9.

13. A planning system, characterized in that, Comprising: A scheduling system, where the scheduling system is used to send the total running duration and the departure time to the global speed planning system; Global path planning system, which is used to send the global driving path and section speed limit information to the global speed planning system; The global speed planning system includes a global speed planning device, and the global speed planning device is used to execute the global speed planning method described in any one of claims 1 to 9.

Images