A method and system for local path planning for a commercial vehicle

Abstract

The application discloses a kind of commercial vehicle local path planning method and system, after obtaining the current data of vehicle, vehicle state information, vehicle position distance barrier data, vehicle position and distance global path data are substituted into path planning algorithm to obtain original solution, then the original solution is extended to obtain convex space, in convex space, the best fuel consumption strategy, driving safety strategy, driving efficiency strategy are used as revenue function to modify original solution to obtain modified solution, and the modified solution and original solution are adjusted to obtain the optimal path, which can reduce vehicle fuel consumption, ensure vehicle driving safety and improve vehicle driving efficiency on the basis of ensuring the operation speed of the method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automobiles, in particular to a local path planning method and system for commercial vehicles. BACKGROUND

[0002] Economical driving is an important direction for energy saving and emission reduction of self-driving cars, especially in the field of self-driving commercial vehicles. However, most of the current research on local path planning algorithms focuses on small vehicles such as passenger cars and how to improve the speed of algorithm solving, ignoring some practical applications of the algorithm, such as fuel consumption, safety, efficiency, etc. For the field of global path planning research, most of them optimize the global path distance, reduce the time of red light, etc. to improve the driving efficiency of the vehicle, thereby reducing the fuel consumption rate. There is little research on the best fuel consumption planning method in the local path.

[0003] The path planning algorithm in the prior art needs to establish a path planning model in advance, but the scene of path planning is variable, so this method is not universal and has low reliability. The method of obtaining model parameters of path planning by using historical data has certain limitations and low efficiency because it is difficult to obtain the latest data due to the variability of road conditions. The model needed to be trained in advance can solve some specific problems, but it is difficult to cover all scenarios in reality, and when a new scenario is encountered, a path planning model needs to be re-established, which obviously increases the cost and is inefficient. In addition, only the driving distance and driving time during the driving process are considered, and the actual problems such as fuel consumption reduction and driving safety are not considered, which cannot fully meet the actual demand. SUMMARY

[0004] The present application provides a local path planning method and system for commercial vehicles to solve the technical problem of reducing fuel consumption while meeting the running efficiency of the path planning algorithm in the path planning of the vehicle.

[0005] To solve the above technical problems, the present application provides a local path planning method for commercial vehicles, comprising:

[0006] Obtain vehicle current state data, vehicle position distance obstacle data, and vehicle position and distance global path data;

[0007] According to the preset path planning algorithm, the vehicle current state data, the vehicle position distance obstacle data, and the vehicle position and distance global path data are calculated to obtain an original solution. Then, the convex space is expanded according to the original solution, the original solution is modified using a preset benefit function to obtain a modified solution, and the modified solution and the original solution are adjusted to obtain an optimal path. The path planning algorithm is provided with a cost function, and the benefit function is provided with a best fuel consumption strategy, a driving safety strategy, and a driving efficiency strategy.

[0008] According to the embodiment of the present application, after the current data of the vehicle is acquired, the vehicle state information, the data of the distance between the vehicle position and the obstacle, and the data of the distance between the vehicle position and the global path are substituted into the path planning algorithm to obtain an original solution, the original solution is then extended to obtain a convex space, the original solution is modified by using the best fuel consumption strategy, the driving safety strategy and the driving efficiency strategy as the benefit function in the convex space to obtain a modified solution, and the modified solution and the original solution are adjusted to obtain an optimal path, which can reduce the fuel consumption of the vehicle, ensure the driving safety of the vehicle and improve the driving efficiency of the vehicle on the basis of ensuring the operation speed of the method.

[0009] As a preferred solution, the vehicle current state data is solved according to the preset path planning algorithm to obtain an original solution, and specifically, the original solution is obtained by:

[0010] The cost function is calculated according to the continuity of the path point, the distance between the path point and the obstacle and the distance between the path point and the global path, and specifically, the cost function is calculated by the following formula:

[0011] The continuity of the path point is calculated, and the formula is as follows:

[0012] u smooth =u(s i )

[0013] u(s i )=∑f′(s i ) 2

[0014] Wherein, c smooth represents the continuity of the path point, s i represents the i-th path, and u(s i ) represents the derivative of the curve of the i-th path.

[0015] The distance between the path point and the obstacle is calculated, and the formula is as follows:

[0016] u barrier_distance =h(d)

[0017]

[0018]

[0019] Wherein, u barrier-distance represents the distance between the path point and the obstacle, (s, l) and (s i , l i ) represent the coordinates of the obstacle and the coordinates of a path point respectively, and d represents the distance between the path point and the obstacle.

[0020] The distance between the path point and the global path is calculated, and the formula is as follows:

[0021] u global_distance =u(l)

[0022]

[0023] wherein u global-distance represents the distance between the path point and the global path, l i represents the vertical distance between the i-th path point and the reference line, and u(l) represents the fitting degree of the distance between the path point and the global path;

[0024] The cost function is calculated, specifically by the following formula:

[0025]

[0026] wherein ω1, ω2, ω3 represent the weight values, and the value is a number in (0, 1], and ∑f′(s i ) 2 represents the continuity of the path point, represents the profit function of the distance between the path point and the obstacle, represents the fitting degree of the distance between the path point and the global path;

[0027] The improved path planning algorithm is obtained according to the cost function, specifically by the following formula:

[0028]

[0029] wherein (s i , l i ) is the coordinate of the path intermediate state n i , (s, l) is the coordinate of the obstacle, s i represents the i-th path, and l i represents the vertical distance between the i-th path point and the reference line;

[0030] According to the improved path planning algorithm, the current state data of the vehicle, the distance data between the vehicle position and the obstacle, and the distance data between the vehicle position and the global path are calculated to obtain the original solution.

[0031] By designing the cost constraint function and substituting the cost constraint function into the path planning algorithm to obtain the improved path planning algorithm, the original solution is optimized on the basis of ensuring the operation speed.

[0032] As a preferred solution, the convex space is expanded according to the original solution, specifically:

[0033] The original solution and the obstacle are input into the coordinate system;

[0034] The original solution is projected on the road boundary and the obstacle one by one to obtain the boundary of the convex space;

[0035] The coordinates of the convex space boundary are obtained by projecting the global path, and thus the coordinates of all points in the convex space are obtained.

[0036] As a preferred solution, the original solution is modified by using a preset benefit function to obtain a modified solution, specifically as follows:

[0037] The original solution is modified in the convex space by using the best fuel consumption strategy, the driving safety strategy and the driving efficiency strategy as the benefit function to obtain a modified solution, specifically as follows:

[0038] The best fuel consumption strategy is as follows:

[0039] u consumption =u(s i )≤α

[0040] u(s i )=ω1·∑f′ 2 (s i )+ω2.∑f″ 2 (s i )+ω3·∑f″′ 2 (s i )

[0041] Wherein, s i represents the i-th path, u(s i ) represents the smoothness of the i-th path, the closer the path is to a straight line, the smoother it is, and α is the smoothness threshold;

[0042] The driving safety strategy is as follows:

[0043] u safety =u(d)

[0044] d1=|l i -l maxi |>γ

[0045] d2=|l i -l mini |>γ

[0046] Wherein, l i represents the longitudinal coordinate of the i-th path point, (l mini , l maxi ) represents the projection of the i-th point on the road edge line, γ is the minimum safety distance, d is the distance between the path point and the road edge, and u(d) represents the normalized constraint d;

[0047] The driving efficiency strategy is as follows:

[0048] uefficiency = u(t)

[0049] t = ∫f(s i )ds i ≤ β

[0050] wherein s i represents the i-th path, β is an integral threshold, and u(t) represents normalizing the constraint t;

[0051] The embodiments of the present application adopt the best fuel consumption strategy, the driving safety strategy and the driving efficiency strategy as the benefit function, correct the original solution in the convex space to obtain a corrected solution, so that the vehicle can reduce fuel consumption, improve driving safety and driving efficiency when selecting the optimal path.

[0052] As a preferred solution, the optimal path is obtained by adjusting the corrected solution and the original solution, specifically as follows:

[0053] A strategy set is determined, so that the original solution and the corrected solution select different strategies in the strategy set to obtain benefits;

[0054] The benefits between the original solution and the corrected solution are compared to determine whether to correct, and then the optimal path is obtained.

[0055] As a preferred solution, a strategy set is determined, so that the original solution and the corrected solution select different strategies in the strategy set to obtain benefits, specifically as follows:

[0056] A common mixed strategy set S i = (s′1, s″1, s′2, s″2, s′3, s″3) is determined;

[0057] wherein s′1, s″1, s′2, s″2, s′3, s″3 represent the best fuel consumption strategy before and after correction, the driving safety strategy before and after correction, and the driving efficiency strategy before and after correction, respectively;

[0058] When the original solution and the corrected solution select different strategies in the strategy set, a benefit function is used to obtain benefits.

[0059] As a preferred solution, the benefits between the original solution and the corrected solution are compared to determine whether to correct, and then the optimal path is obtained, specifically as follows:

[0060] The benefits between the original solution and the corrected solution are compared to determine whether to correct, and the formula is as follows:

[0061] u i (s′ i , s -i ) > u i (s″ i , s -i )

[0062] wherein s' i denotes the strategy selected by the modified solution, s" i denotes the strategy selected by the original solution, S -i denotes the income of other participants;

[0063] If the income of the original solution after modification is higher than that before modification, the solution is modified into the modified solution, and if the income of the original solution after modification is lower than that before modification, the original solution is kept unchanged;

[0064] If the income of the modified solution after modification is higher than that before modification, the modified solution is kept unchanged, and if the income of the modified solution after modification is lower than that before modification, the modified solution is changed into the original solution.

[0065] If the strategy set satisfies the following formula:

[0066]

[0067] wherein, denotes the optimal strategy set;

[0068] denotes the optimal strategy set except

[0069] denotes that in the total strategy set S i , there is the optimal strategy set S i * , so as to obtain the optimal path.

[0070] By comparing the income of the original solution with the income of the modified solution, if the income of the original solution after modification is higher than that before modification, the solution is modified into the modified solution, otherwise, the original solution is kept unchanged; if the income of the modified solution after modification is higher than that before modification, the modified solution is kept unchanged, otherwise, the modified solution is changed into the original solution, and the original solution and the modified solution are adjusted, so that the reliability of the method is improved.

[0071] As a preferred solution, to solve the same technical problem, the embodiment of the present application also provides a commercial vehicle local path planning system, comprising: a vehicle state data acquisition module and an optimal path calculation module;

[0072] The vehicle state data acquisition module is used to acquire vehicle current state data, vehicle position distance obstacle data and vehicle position and distance global path data.

[0073] ​The optimal path calculation module is configured to solve the current state data of the vehicle according to a preset path planning algorithm to obtain an original solution, extend a convex space according to the original solution, correct the original solution by using a preset benefit function to obtain a corrected solution, and adjust the corrected solution and the original solution to obtain the optimal path; wherein the path planning algorithm is provided with a cost function, and the benefit function is provided with a best fuel consumption strategy, a driving safety strategy and a driving efficiency strategy.

[0074] As a preferred solution, the optimal path calculation module comprises an original solution calculation unit, an original solution correction unit and an optimal path adjustment unit.

[0075] The original solution calculation unit is configured to solve the current state data of the vehicle according to a preset path planning algorithm to obtain an original solution, specifically, a cost function is calculated according to the continuity of path points, the distance between path points and obstacles and the distance between path points and a global path, specifically, the cost function is calculated by the following formula:

[0076] The continuity of path points is calculated by the following formula:

[0077] u smooth =u(s i )

[0078] u(s i )=∑f′(s i ) 2

[0079] wherein u smooth represents the continuity of path points, s i represents the i-th path, and u(s i ) represents the derivative of the curve of the i-th path.

[0080] The distance between path points and obstacles is calculated by the following formula:

[0081] u barrier_distance =h(d)

[0082]

[0083]

[0084] wherein u barrier-distance represents the distance between path points and obstacles, (s, l) and (s i , l i ) represent the coordinates of obstacles and a path point, respectively, and d represents the distance between the path point and the obstacles. The distance between path points and a global path is calculated by the following formula:

[0085] u global_distance =u(l)

[0086]

[0087] wherein, l i a vertical distance between the i-th path point and the reference line, u(l) represents a fitting degree of the path point to the global path;

[0088] The cost function is calculated, specifically, by the following formula:

[0089]

[0090] wherein, ω1, ω2, ω3 represent weight values, and the value is a number (0, 1], ∑f′(s i ) 2 representing the continuity of the path point, representing the profit function of the path point and the obstacle distance, representing the fitting degree of the path point to the global path;

[0091] The improved path planning algorithm is obtained according to the cost function, specifically, by the following formula:

[0092]

[0093] wherein, (s i , l i ) is the coordinate of the path intermediate state n i , (s, l) is the coordinate of the obstacle, s i represents the i-th path, l i a vertical distance between the i-th path point and the reference line;

[0094] According to the improved path planning algorithm, the current state data of the vehicle is solved to obtain the original solution;

[0095] The original solution modification unit is used to expand the convex space according to the original solution, and the original solution is modified by using a preset profit function to obtain a modified solution;

[0096] The optimal path adjustment unit is used to adjust the modified solution and the original solution to obtain the optimal path. BRIEF DESCRIPTION OF DRAWINGS

[0097] Figure 1 : a flowchart of an embodiment of the commercial vehicle local path planning method provided by the present application;

[0098] Figure 2 : a process diagram of convex space expansion and original solution modification of an embodiment of the commercial vehicle local path planning method provided by the present application;

[0099] Figure 3 Fig. 1 is a schematic diagram of an original solution solving process based on an improved A* algorithm of an embodiment of the commercial vehicle local path planning method provided by the present application;

[0100] Figure 4 Fig. 2 is a structural schematic diagram of another embodiment of the commercial vehicle local path planning method provided by the present application. DETAILED DESCRIPTION

[0101] The technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the present application.

[0102] Embodiment one

[0103] Please refer to Figure 1 The commercial vehicle local path planning method provided by the embodiment of the present application is suitable for an unmanned vehicle. In the embodiment, an optimal path is calculated by improving a path planning algorithm and ensuring the operation speed. The commercial vehicle local path planning method comprises steps 101 to 102, and each step is specifically as follows:

[0104] Step 101: obtaining vehicle current state data, vehicle position distance obstacle data and vehicle position and distance global path data.

[0105] In the embodiment, in the path planning, the vehicle current state data, the distance between a certain path point in the path and the obstacle, and the vertical distance between a certain path point and the reference line are obtained.

[0106] Step 102: calculating the vehicle current state data, the vehicle position distance obstacle data and the vehicle position and distance global path data according to a preset path planning algorithm to obtain an original solution, then expanding a convex space according to the original solution, using a preset benefit function to modify the original solution to obtain a modified solution, and adjusting the modified solution and the original solution to obtain an optimal path; wherein the path planning algorithm is provided with a cost function, and the benefit function is provided with a best fuel consumption strategy, a driving safety strategy and a driving efficiency strategy.

[0107] Optionally, the vehicle current state data, the vehicle position distance obstacle data and the vehicle position and distance global path data are calculated according to the preset path planning algorithm to obtain the original solution, and the calculation is specifically as follows:

[0108] The cost function is calculated according to the continuity of the path point, the distance between the path point and the obstacle, and the distance between the path point and the global path. Specifically, the cost function is calculated by the following formula:

[0109] The continuity of the path point is calculated by the following formula:

[0110] u smooth =u(s i )

[0111] u(s i )=∑f′(s i ) 2

[0112] Wherein, u smooth represents the continuity of the path point, s i represents the i-th path, and u(s i ) represents the derivative of the curve of the i-th path.

[0113] The distance between the path point and the obstacle is calculated by the following formula:

[0114] u barrier_distance =h(d)

[0115]

[0116]

[0117] Wherein, u barrier-distance represents the distance between the path point and the obstacle, (s, l) and (s i , l i ) represent the coordinates of the obstacle and the coordinates of a path point respectively, and d represents the distance between a path point and the obstacle.

[0118] The distance between the path point and the global path is calculated by the following formula:

[0119] u global_distance =u(l)

[0120]

[0121] Wherein, u global-distance represents the distance between the path point and the global path, l i represents the vertical distance between the i-th path point and the reference line, and u(l) represents the fitting degree of the distance between the path point and the global path.

[0122] The cost function is calculated by the following formula:

[0123]

[0124] Wherein, ω1, ω2, ω3 represent weight values respectively, and the value is a number (0, 1]; ∑f'(s i ) 2 Indicates the continuity of the path point, Indicates the reward function of the distance between the path point and the obstacle, Indicates the fitting degree representing the distance between the path point and the global path;

[0125] According to the improved path planning algorithm obtained by the cost function, specifically, the following formula is used for calculation:

[0126]

[0127] Wherein, (s i , l i ) is the coordinate of the intermediate state n i of the path, (s, l) is the coordinate of the obstacle, s i Indicates the i-th path, l i Indicates the vertical distance between the i-th path point and the reference line;

[0128] According to the improved path planning algorithm, the current state data of the vehicle, the distance data between the vehicle position and the obstacle, and the distance data between the vehicle position and the global path are calculated to obtain the original solution.

[0129] In this embodiment, the overall cost function N COST is determined by the smoothness, the distance from the obstacle and the distance from the global path, and then the A* algorithm is improved by using the overall cost function. The A* algorithm is commonly used to solve the shortest distance of the path, and the formula is as follows:

[0130] F(n i )=G(n i )+H(n i )

[0131] Wherein, G(n i ) represents the minimum cost consumed from the starting point (the initial initial state n0) to the intermediate state n i ; H(n i ) represents the Manhattan distance between the intermediate state n i and the terminal point; the distance between each state point is a Manhattan unit length M (as shown in Figure 1 ); after finding the intermediate state n1, the initial initial state n0 is replaced by the intermediate state n1, and so on until the intermediate state n i is equal to the terminal point, then the solution is completed;

[0132] The formula G(n i ) is replaced by the cost function N COST :

[0133]

[0134] wherein (s i , l i ) represents the coordinates of the intermediate state n i of the path, and (s, l) represents the coordinates of the obstacle;

[0135] H(n i ) is the Manhattan distance, as shown in the following formula:

[0136] H(n i ) = |s i -s i+a | + |l i -l i+a |

[0137] wherein (s i+a , l i+a ) represents the coordinates of the end point, but in one planning, the end point does not mean the point at which the planning ends, and can be a certain stage point, so the size of a can be set to determine the number of points to be calculated in one planning cycle according to the needs;

[0138] Therefore, the improved A* algorithm is as follows:

[0139]

[0140] wherein (s i , l i ) represents the coordinates of the intermediate state n i of the path, and (s, l) represents the coordinates of the obstacle;

[0141] As shown in the following formula, the original solution is obtained by using the improved A* algorithm for solving. Figure 2

[0142] Optionally, the convex space is expanded according to the original solution, and the convex space is expanded according to the original solution, and the convex space is expanded according to the original solution.

[0143] The original solution and the obstacle information are input into the coordinate system;

[0144] The original solution is projected on the road boundary and the obstacle one by one to obtain the boundary of the convex space;

[0145] The coordinates of the boundary of the convex space are obtained by projecting the global path, so that the coordinates of all points in the convex space are obtained.

[0146] In this embodiment, the boundary (l maxi , l mini ) of the convex space is obtained by projecting the original solution composed of a set of discrete points on the road boundary one by one, and the coordinates s i ​, thus obtaining the SL coordinates of all points in the convex space, at this time: maxi ≤f(s i )≤l mini , thus obtaining the extended convex space.

[0147] Optionally, the original solution is modified using a preset benefit function to obtain a modified solution, specifically:

[0148] The original solution is modified in the convex space using the best fuel consumption strategy, the driving safety strategy and the driving efficiency strategy as the benefit function to obtain a modified solution, specifically, the following formula is used for calculation:

[0149] The best fuel consumption strategy formula is as follows:

[0150] u consumption =u(s i )≤α

[0151] u(s i )=ω1·∑f′ 2 (s i )+ω2·∑f″ 2 (s i )+ω3·∑f″′ 2 (s i )

[0152] Wherein, s i represents the i-th path, u(s i ) represents the smoothness of the i-th path, the closer the path is to a straight line, the smoother it is, and α is the smoothness threshold;

[0153] The driving safety strategy formula is as follows:

[0154] u safety =u(d)

[0155] d1=|l i -l maxi |>γ

[0156] d2=|l i -l mini |>γ

[0157] Wherein, l i represents the longitudinal coordinate of the i-th path point, (l mini , l maxi ) represents the projection of the i-th point on the road edge line, γ is the minimum safety distance, d is the distance between the path point and the road edge, and u(d) represents the normalized constraint d;

[0158] The driving efficiency strategy formula is as follows:

[0159] u efficiency =u(t)

[0160] t=∫f(s i )ds i ≤β

[0161] Among them, s i Let represent the i-th path segment, β be the integration threshold, and u(t) represent normalizing constraint t.

[0162] In this embodiment, a strategy set is constructed based on the optimal fuel consumption strategy, the safety constraint strategy, and the efficiency strategy. In the convex space, the original solution is modified, and the modification process is as follows: Figure 3 As shown, the original solution and obstacle coordinates are input into the reference frame. The convex space is obtained by projecting the original solution onto the road boundary and obstacles. The original solution is then corrected within the convex space.

[0163] As an example of this embodiment, the original solution is corrected within the convex space. First, the coordinates of the i-th point of the original solution are taken, and it is calculated whether the optimal fuel consumption meets the constraint conditions. If not, it is first moved by one Manhattan unit length M towards the S-axis (global path), and the condition is calculated again. If it meets the condition, the calculation continues downward; if not, the movement continues until the condition is met, and the calculation continues downward. If the condition is met, the calculation continues downward. Then, it is calculated whether the driving safety strategy is met. When d1 > γ and d2 < γ, it means that the i-th point of the original solution is close to the lower boundary of the convex space, so it is moved upward by one Manhattan unit distance M. When d1 > γ and d2 > γ, it means that the i-th point of the original solution is neither close to the upper boundary nor the lower boundary, so no movement is needed. When d1 < γ and d2 > γ, it means that the i-th point of the original solution is close to the upper boundary of the convex space, so it is moved downward by one Manhattan unit distance M. If the driving safety strategy is satisfied, calculate whether the driving efficiency strategy is satisfied. If the driving efficiency benefit is <β and >0, it means that the original solution is above the global path, so move it down by one Manhattan unit distance. If the driving efficiency benefit is <0, it means that the original solution is below the global path, so move it down by one Manhattan unit distance. Finally, the correction is completed.

[0164] Optionally, the corrected solution and the original solution are adjusted to obtain the optimal path, specifically as follows:

[0165] Determine a set of strategies so that the original solution and the modified solution can each obtain a benefit by choosing different strategies from the set.

[0166] By comparing the gains between the original solution and the modified solution, we determine whether to make a correction and thus obtain the optimal path.

[0167] Optionally, a set of strategies is determined such that the original solution and the modified solution each select different strategies from the set to obtain benefits, specifically:

[0168] Determine the common mixing strategy set S i= (s'1, s"1, s'2, s"2, s'3, s"3);

[0169] wherein s'1, s"1, s'2, s"2, s'3, s"3 represent the modified and unmodified optimal fuel consumption strategy, the modified and unmodified driving safety strategy, and the modified and unmodified driving efficiency strategy, respectively;

[0170] The original solution and the modified solution select different strategies in the strategy set, and the revenue function is used to obtain the revenue.

[0171] Optionally, the size of the revenue between the original solution and the modified solution is compared to determine whether to modify, and then the optimal path is obtained, specifically as follows:

[0172] The size of the revenue between the original solution and the modified solution is compared to determine whether to modify, and the formula is as follows:

[0173] u i (s' i ,s -i )>u i (s" i ,s -i )

[0174] wherein s′ i represents the strategy selected by the modified solution, s″ i represents the strategy selected by the original solution, and s -i represents the revenue of other participants;

[0175] If the revenue of the original solution after modification is higher than that before modification, the solution is modified as the modified solution; if the revenue of the original solution after modification is lower than that before modification, the original solution remains unchanged.

[0176] If the revenue of the modified solution after modification is higher than that before modification, the modified solution remains unchanged; if the revenue of the modified solution after modification is lower than that before modification, the modified solution is changed to the original solution.

[0177] If the strategy set satisfies the following formula:

[0178]

[0179] wherein represents the optimal strategy set;

[0180] represents the optimal strategy set except ; and

[0181] In the embodiment, when the original solution and the modified solution select different strategies in the strategy set, different revenues u i (i=1, 2) are obtained, the revenues are compared to perform modification, and the modification rules are shown in the following table:

[0182] Table 1 correction rule table

[0183] Profit formula Original solution T Corrected solution F u i (S i ,S -i )>u i (S i ,S -i )]]> T→F F u i (S i ,S -i )<u i (S i ,S -i )]]> T F→T

[0184] By using the Nash equilibrium solving method, decisions are made on the original solution and the corrected solution, and finally the Nash equilibrium is reached, the path planning solving strategy and method based on the best fuel consumption are realized, and the decision reliability is improved.

[0185] In the embodiment, the A* algorithm is improved by designing a cost function for solving the optimal path, the improved A* algorithm is used to calculate the current state data, the vehicle position distance obstacle data and the vehicle position and distance global path data, the optimal path original solution is obtained, the convex space of the region is expanded according to the original solution, and finally the accurate solution, i.e., the optimal path for the best fuel consumption, is obtained by correcting the original solution based on the Nash equilibrium with the best fuel consumption strategy, the driving safety strategy and the driving efficiency strategy as the revenue function.

[0186] Embodiment two

[0187] Correspondingly, referring to Figure 4 , Figure 4 is a structure schematic diagram of a local path planning system of a commercial vehicle provided by the application. As shown in the figure, the local path planning system of the commercial vehicle comprises a vehicle state data acquisition module 401 and an optimal path calculation module 402, wherein the specific units of each module are as follows:

[0188] The vehicle state data acquisition module 401 is configured to acquire the vehicle current state data, the vehicle position distance obstacle data and the vehicle position and distance global path data.

[0189] The optimal path calculation module 402 is configured to calculate the vehicle current state data, the vehicle position distance obstacle data and the vehicle position and distance global path data according to a preset path planning algorithm, obtain an original solution, expand a convex space according to the original solution, correct the original solution using a preset revenue function to obtain a corrected solution, and adjust the corrected solution and the original solution to obtain an optimal path.

[0190] In the embodiment, the optimal path calculation module comprises an original solution calculation unit 4021, an original solution correction unit 4022 and an optimal path adjustment unit 4023.

[0191] The original solution calculation unit 4021 is configured to calculate the vehicle current state data according to a preset path planning algorithm to obtain an original solution.

[0192] The original solution correction unit 4022 is configured to correct the original solution to obtain a corrected solution according to the original solution expansion convex space and using a preset benefit function.

[0193] The optimal path adjustment unit 4023 is configured to adjust the corrected solution and the original solution to obtain an optimal path.

[0194] The optimal path adjustment unit 4023 includes a benefit acquisition subunit 40231 and a benefit comparison subunit 40232.

[0195] The benefit acquisition subunit 40231 is configured to determine a strategy set, and make the original solution and the corrected solution select different strategies in the strategy set to obtain benefits.

[0196] The benefit comparison subunit 40232 is configured to compare the benefits of the original solution and the corrected solution to determine whether to correct, and then obtain the optimal path.

[0197] The local path planning method of the vehicle can be more detailed, but is not limited to the content described above.

[0198] Compared with the prior art, the vehicle state information, the vehicle position distance obstacle data, and the vehicle position and distance global path data are substituted into the path planning algorithm to obtain an original solution, the original solution is expanded to obtain a convex space, the original solution is corrected in the convex space using a best fuel consumption strategy, a driving safety strategy, and a driving efficiency strategy as a benefit function to obtain a corrected solution, and the corrected solution and the original solution are adjusted to obtain an optimal path, which can reduce vehicle fuel consumption, ensure vehicle driving safety, and improve vehicle driving efficiency on the basis of ensuring the operation speed of the method.

[0199] The above embodiments further illustrate the purpose, technical solutions, and advantages of the present application. It should be understood that the above embodiments are only specific embodiments of the present application and are not intended to limit the protection scope of the present application. It should be particularly pointed out that any modifications, equivalent replacements, improvements, etc. made by those skilled in the art within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for local path planning for a commercial vehicle, characterized in that The application relates to a vehicle path planning method and device. The application comprises the following steps: According to a preset path planning algorithm, vehicle current state data, vehicle position distance obstacle data and vehicle position distance global path data are calculated to obtain an original solution, and then a convex space is expanded according to the original solution, a modified solution is obtained by modifying the original solution by using a preset benefit function, and finally the modified solution and the original solution are adjusted to obtain an optimal path; wherein the path planning algorithm is provided with a cost function, and the benefit function is provided with an optimal fuel consumption strategy, a driving safety strategy and a driving efficiency strategy. The original solution is obtained by calculating the vehicle current state data, the vehicle position distance obstacle data and the vehicle position distance global path data according to the preset path planning algorithm, and the calculation specifically comprises the following steps: The cost function is calculated according to the continuity of the path point, the distance between the path point and the obstacle and the distance between the path point and the global path, and the calculation specifically comprises the following steps: The continuity of the path point is calculated, and the calculation formula is as follows: wherein, represents the continuity of the path point, represents the first segment path, represents the derivative of the curve of the first segment. The distance between the path point and the obstacle is calculated, and the calculation formula is as follows: wherein, represents the distance of a path point to an obstacle, , represent obstacle coordinates, respectively, a path point coordinate, represents the distance between a path point and an obstacle; The distance between the path point and the global path is calculated, and the calculation formula is as follows: wherein, represents the distance of the path point from the global path, the vertical distance between the first path point and the reference line, represents the degree of fit of the path point's distance from the global path; The cost function is calculated, and the calculation specifically comprises the following steps: wherein, , , respectively represent the weight values, the values are numbers in (0, 1], represents the continuity of the path points, represents the distance between the path points and the obstacles, represents the fitting degree representing the distance between the path points and the global path; The improved path planning algorithm is obtained according to the cost function, and the calculation specifically comprises the following steps: wherein, is the coordinate of the intermediate state of the path , is the coordinate of the obstacle, denotes the th segment of the path, the vertical distance between the th path point and the reference line; The original solution is obtained by calculating the vehicle current state data, the vehicle position distance obstacle data and the vehicle position distance global path data according to the improved path planning algorithm.

2. The commercial vehicle local path planning method of claim 1, wherein, The original solution and obstacle information are input into a coordinate system, and the calculation specifically comprises the following steps: The boundary of the convex space is obtained by projecting the original solution on the road boundary and the obstacle one by one. The coordinates of the boundary of the convex space are obtained by projecting the global path, so that the coordinates of all points in the convex space are obtained. The modified solution is obtained by modifying the original solution by using the preset benefit function, and the calculation specifically comprises the following steps:

3. The commercial vehicle local path planning method of claim 1, wherein, The modified solution is obtained by modifying the original solution in the convex space by taking the optimal fuel consumption strategy, the driving safety strategy and the driving efficiency strategy as the benefit function, and the calculation specifically comprises the following steps: The optimal fuel consumption strategy formula is as follows: The driving safety strategy formula is as follows: wherein, represents the first segment path, represents the degree of smoothness of the first segment path, the path being smoother the closer it is to a straight line, is a threshold value for the degree of smoothness; The driving efficiency strategy formula is as follows: wherein represents the longitudinal coordinate of the th path point, represents the projection of the th point on the curb line, is the minimum safety distance, is the distance between the path point and the curb represents normalizing the constraint d; The optimal path is obtained by adjusting the modified solution and the original solution, and the calculation specifically comprises the following steps: wherein, represents the first segment path, is an integral threshold, represents normalizing the constraint t.

4. The commercial vehicle local path planning method of claim 1, wherein, A strategy set is determined, and the original solution and the modified solution select different strategies in the strategy set to obtain benefits. Whether the original solution and the modified solution are modified is determined by comparing the benefits of the original solution and the modified solution, and then the optimal path is obtained. The benefits of the original solution and the modified solution are obtained when the original solution and the modified solution select different strategies in the strategy set.

5. The commercial vehicle local path planning method of claim 4, wherein, Whether the original solution and the modified solution are modified is determined by comparing the benefits of the original solution and the modified solution, and then the optimal path is obtained. Determining a common mixing policy set wherein, respectively represent the modified and unmodified optimal fuel consumption strategy, the modified and unmodified travel safety strategy, and the modified and unmodified travel efficiency strategy. Whether the original solution and the modified solution are modified is determined by comparing the benefits of the original solution and the modified solution, and the calculation formula is as follows:

6. The commercial vehicle local path planning method of claim 4, wherein, ​ ​ wherein, represents the strategy chosen by the revised solution, represents the strategy chosen by the original solution, represents the payoff of the other player; If the yield of the original solution after modification is higher than that before modification, the solution is modified as the modified solution, and if the yield of the original solution after modification is lower than that before modification, the original solution is kept unchanged; If the yield of the modified solution after modification is higher than that before modification, the modified solution is kept unchanged, and if the yield of the modified solution after modification is lower than that before modification, the modified solution is changed as the original solution; If the strategy set satisfies the following formula: wherein, represents the optimal strategy set represents the optimal strategy set except represents the optimal strategy set except denotes the optimal strategy set in the total strategy set , there exists an optimal strategy set , thereby obtaining an optimal path.

7. A commercial vehicle local path planning system characterized by, The application relates to a vehicle optimal path calculation method and device, and belongs to the field of intelligent transportation. The vehicle state data acquisition module is used for acquiring vehicle current state data, vehicle position distance obstacle data and vehicle position and distance global path data. The optimal path calculation module is used for calculating the vehicle current state data, the vehicle position distance obstacle data and the vehicle position and distance global path data according to a preset path planning algorithm, obtaining an original solution, then extending a convex space according to the original solution, modifying the original solution by using a preset yield function to obtain a modified solution, and adjusting the modified solution and the original solution to obtain an optimal path; wherein the path planning algorithm is provided with a cost function, and the yield function is provided with a best fuel consumption strategy, a driving safety strategy and a driving efficiency strategy. The optimal path calculation module comprises an original solution calculation unit, an original solution modification unit and an optimal path adjustment unit. The original solution calculation unit is used for solving the vehicle current state data according to a preset path planning algorithm to obtain an original solution, specifically, a cost function is calculated according to the continuity of path points, the distance between path points and obstacles and the distance between path points and a global path, specifically, the continuity of path points is calculated by the following formula: The distance between path points and obstacles is calculated by the following formula: The cost function is calculated by the following formula: wherein, denotes the continuity of the path points, denotes the first segment path, denotes the first derivative of the curve of the segment. An improved path planning algorithm is obtained according to the cost function, specifically, the improved path planning algorithm is calculated by the following formula: wherein, represents the distance of a path point to an obstacle, , respectively represent the obstacle coordinates, the coordinates of a path point, represents the distance between a path point and an obstacle; the distance of a path point to the global path is calculated as follows: wherein, represents the distance of the path point from the global path, the vertical distance between the first path point and the reference line, represents the degree of fit of the path point's distance from the global path; The vehicle current state data is solved according to the improved path planning algorithm to obtain an original solution; wherein, , , respectively represent the weight values, the value is a number in (0, 1], represents the continuity of the path points, represents the distance between the path points and the obstacles, represents the fitting degree representing the distance between the path points and the global path; The original solution modification unit is used for extending a convex space according to the original solution, and modifying the original solution by using a preset yield function to obtain a modified solution; wherein, is the coordinate of the intermediate state of the path is the coordinate of the obstacle, is the coordinate of the obstacle, denotes the segment of the path, the vertical distance between the path point and the reference line; The optimal path adjustment unit is used for adjusting the modified solution and the original solution to obtain an optimal path. ​ ​

Images