Improved fast search random tree path planning method and device based on energy consumption constraint

Abstract

The application discloses an improved fast search random tree path planning method based on energy consumption constraint, and the algorithm is as follows: modeling according to actual geographic information, drawing a digital elevation model of a required map, determining a sampling mode of sampling points according to whether an initial feasible path can be found from a starting point to an ending point, determining a node expansion step length according to a dynamic step length strategy, generating a new expansion node, introducing a slope proportion coefficient by using a dimension reduction idea, adding an energy consumption constraint condition to a heuristic function, generating a logical path length as an optimal path judgment standard, reconnecting and rewriting a node parent node, and reaching the target point when a distance from the new node to the target node is less than a threshold value. The application effectively reduces randomness and blindness of the sampling points, reduces a search space and a range of a search tree, considers an energy consumption influence of an elevation slope on a robot driving process, makes a planned path have the characteristics of being relatively flat and low in energy consumption, and improves a working efficiency of the robot.

Description

Technical Field

[0001] This invention belongs to the field of robot path planning, specifically relating to an improved fast search random tree path planning method and device based on energy consumption constraints. Background Technology

[0002] Path planning is a technique for robots to find the optimal route from a starting point to a destination based on specific environmental conditions and task objectives. Over the past few decades, scholars have proposed many path planning methods, which have achieved good results in practical applications, such as the artificial potential field method, the fast search random tree algorithm, genetic algorithms, A* algorithms, and reinforcement learning.

[0003] Sampling-based path planning has garnered significant attention due to its convenience and speed, with the Rapidly-exploring Random Tree (RRT) algorithm being a prime example. These sampling-based algorithms do not model obstacles but instead perform collision detection at each sampled point. They are highly versatile, capable of solving spatial path planning problems for robots across different dimensions, and boast a very high exploration success rate. However, their drawbacks are also apparent: firstly, the random tree expansion guidance is poor, resulting in slow convergence and slower search speeds in complex spaces; secondly, the obtained path is usually not the optimal solution and deviates significantly from the optimal path; and thirdly, they have limitations when used in dynamic environments. The RRT* algorithm improves upon the basic RRT algorithm, exhibiting probabilistic completeness, with the probability of finding the optimal solution approaching 1 as the number of iterations nears infinity, but significantly increasing the search time. The Informed-RRT* algorithm incorporates prior knowledge into the RRT* algorithm. After finding an initial solution, the Informed-RRT* algorithm focuses its search on an ellipsoidal subset of the state space. This algorithm retains the same completeness and asymptotic optimality as RRT*, while improving convergence speed and the quality of the final solution. However, it is prone to degenerate into the RRT* algorithm, the sampling before finding the initial solution is still random, and it does not take into account the impact of road elevation and slope on the energy consumption of robot path planning. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides an improved fast search random tree path planning method based on energy consumption constraints, which solves the problems of blind search direction, reduced search efficiency due to fixed step size, and lack of consideration for energy consumption factors in traditional fast search random tree algorithms.

[0005] To achieve the above objectives, the technical solution adopted by this invention is: an improved fast search random tree path planning method based on energy consumption constraints, comprising the following steps:

[0006] S1. Model based on actual geographic information, draw the digital elevation model of the required map, and determine the starting point, ending point and initial step size.

[0007] S2, determine whether there is an initial feasible path between the starting point and the ending point. If not, use the quadrant constraint method to constrain the sampling method of the sampling points, restricting the sampling position and orientation of the sampling points. If there is an initial feasible path, establish an elliptical sampling range based on the logical path length of the initial feasible path and the Euclidean geometric distance from the starting point to the ending point, and continuously iterate to find a better path and dynamically update the elliptical sampling range to obtain a series of sampling points.

[0008] S3, based on the sampling points generated in S2, determines the node expansion step size according to the dynamic step size strategy and generates new expanded nodes.

[0009] S4. The path length with added slope constraints is named the logical path length. A slope ratio coefficient is introduced. Based on the idea of ​​dimensionality reduction, the influence of three-dimensional elevation information on robot energy consumption is transformed into a two-dimensional logical path length. Within the elliptical sampling range, the path cost is determined by comparing the logical path lengths of the extended nodes, and the path is updated.

[0010] S5, when the distance from the new node to the target node is less than the set threshold, the target point is reached;

[0011] The system executes the preset number of iterations from S2 to S4, continuously updating and determining the final path. The final path is then optimized by removing redundant points, and the final optimal path based on energy consumption constraints is output.

[0012] S2 specifically includes the following steps:

[0013] S2.1: Before finding the initial feasible path, the quadrant constraint method is used. A sampling quadrant constraint mark d is set to determine the coordinates of the starting point and the ending point. A relative coordinate system is established based on the positions of the starting point and the ending point to determine the optimal search quadrant where the ending point is located. The sampling range of the sampling points is preferentially controlled in one quadrant, which is the optimal weight sampling quadrant.

[0014] S2.2: Find the initial feasible path, denoted as c. best Starting from x start , endpoint x goal As the focus of the ellipse, let the major axis 'a' of the ellipse be equal to half the initial path length, that is... Obtain all parameters of the ellipse equation and the eccentricity of the ellipse.

[0015] Where c min Starting from x start To x goal The straight-line distance between them is the theoretical minimum cost between them, cbest To find the actual minimum distance from the starting point to the ending point, the feasible paths are iteratively updated. Each time a path with a smaller cost is found, its cost is used as the new value for c. best Update the elliptic sampling space.

[0016] In S2.1, the specific rules for determining the optimal weight sampling quadrant are as follows:

[0017] And if x2-x1>0, then d=1, meaning the optimal weight quadrant is the first quadrant;

[0018] And if x2-x1<0, then d=2, meaning the optimal weight quadrant is the second quadrant;

[0019] And if x2-x1<0, then d=3, meaning the optimal weight quadrant is the third quadrant;

[0020] And if x2-x1>0, then d=4, meaning the optimal weight quadrant is the fourth quadrant;

[0021] During execution, sampling points are randomly generated. If the generated sampling point is not in the current optimal weight quadrant, it is discarded and a new sampling point is generated. If the generated sampling point is in the current optimal weight quadrant, it is retained.

[0022] S3 utilizes a dynamic step-size strategy to determine the expansion step size when expanding nodes. This strategy is mainly divided into two different expansion strategies: one when feasible solutions for the start and end points are not found, and the other after they are found.

[0023] S3.1: When no initial feasible solution is found at the starting and ending points, random sampling is performed within the feasible region, starting from the starting point with an initial step size γ. init The random tree is expanded. Collision detection is performed each time a new node is generated during the expansion. If no obstacle is detected, the next expansion step size is γ. next =γ now +0.5*γ init This process continues in subsequent searches until a collision is detected during a node expansion. The next expansion step size is γ. next =γno w -0.5γ init The minimum expansion step size is γ init ;

[0024] S3.2: After finding the initial feasible solutions for the start and end points, the search domain becomes an elliptical subset of the state space, using an initial step size γ. nit Expand nodes; if no collision is detected in this expansion, the next expansion step size is γ. next =γnow +0.1*γ init This process continues in subsequent searches until a collision is detected during a node expansion, at which point the next expansion step size becomes γ. next =γ now -0.1*γ init If no collision is detected, the minimum sampling step size within the ellipse is γ. init .

[0025] In S4, based on the concept of dimensionality reduction, the three-dimensional geographic information is reduced to two dimensions. A slope ratio coefficient is introduced to transform the impact of three-dimensional elevation information on robot energy consumption into a two-dimensional logical path length, including:

[0026] S4.1 measures the time required for the robot to traverse a fixed distance with constant power across different slopes, using the energy consumption factor β.

[0027] S4.2, perform function fitting on the measured values ​​of energy consumption and slope to establish a slope energy consumption fitting function;

[0028] S4.3, using the established fitting function, calculate the energy consumption factor β corresponding to different slopes;

[0029] S4.4, normalize the slope ratio coefficient k, and determine the logical path length of the slope constraint E(x) based on the slope ratio coefficient k and the Euclidean geometric distance between two points in the two-dimensional map.

[0030] In S4.1, the slope energy consumption fitting function is:

[0031] f(x) = p1*x 2 +p2*x+p3

[0032] Where p1, p2, and p3 are coefficients.

[0033] In S4.4, the formula for calculating the logical path length with the added slope constraint E(x) is as follows:

[0034] Path E(x) =path true (1+k*10)

[0035]

[0036] k = 1 - f(x)

[0037] Where, Path E(x) The path represents the logical path length after adding the slope constraint E(x). true Let f(x) represent the Euclidean geometric distance between two points in a two-dimensional map, and f(x) be the energy consumption factor value of the fitting function.

[0038] The present invention may also provide a computer device, including a processor and a memory, wherein the memory is used to store a computer executable program, the processor reads part or all of the computer executable program from the memory and executes it, and when the processor executes part or all of the compute executable program, it can implement some or all of the steps of the improved fast search random tree path planning method based on energy consumption constraints.

[0039] Simultaneously, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, it can implement the improved fast search random tree path planning method based on energy consumption constraints described in this invention.

[0040] Compared with the prior art, the present invention has at least the following beneficial effects:

[0041] This invention employs a quadrant constraint method for the random search points before obtaining the initial feasible path at the start and end points, reducing the randomness of sampling point generation and accelerating the search efficiency for generating the initial feasible path. Different step size strategies are selected before and after generating the initial feasible path, improving search efficiency in open spaces and shortening the time to find the initial feasible path. A slope constraint condition is added to the heuristic function to consider the impact of ground elevation slope on the robot's energy consumption during movement. A dimensionality reduction approach is adopted, proposing a slope ratio coefficient to reduce the three-dimensional search space to a two-dimensional search space, significantly shortening the algorithm's search time and improving search efficiency. The results are normalized to logical path lengths for easy comparison and calculation. The improved algorithm can prioritize planning paths with small slopes, flat surfaces, and low energy consumption, improving robot efficiency, reducing operating costs, and saving energy. The improved algorithm is applicable to the field of robot path planning. Attached Figure Description

[0042] Figure 1 A flowchart for improving path planning methods.

[0043] Figure 2 This is a schematic diagram of an elliptical sampling space. Detailed Implementation

[0044] In the following description, many technical details are presented to help the reader better understand this application. However, those skilled in the art will understand that the technical solutions claimed in this application can be implemented even without these technical details and various variations and modifications based on the following embodiments.

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be further described in detail below with reference to the accompanying drawings, including:

[0046] Step 1: Model based on actual map information, determine the required digital elevation model of the map, and determine the starting point p. start (x1, y1), endpoint p goal (x2, y2) and initial step size γ;

[0047] Step 2: When no initial feasible path from the starting point to the ending point is found, the quadrant constraint method is used to constrain the sampling points, limiting their sampling position and orientation. Once an initial feasible path is found, dynamic elliptic sampling is used. The specific method is as follows:

[0048] Step 2.1: Before finding the initial feasible path, the quadrant constraint method is used. The quadrant constraint method requires setting a sampling quadrant constraint marker d (direction), determining the coordinates of the start and end points, establishing a relative coordinate system based on the positions of the start and end points, determining the optimal search quadrant where the end point is located, and prioritizing the sampling range of sampling points within one quadrant. Specifically, a new coordinate system is established with the start point as the origin, and the relative quadrant of the coordinate system established with respect to the start point is determined; this quadrant is the optimal weight sampling quadrant. When expanding nodes, random sampling is preferentially performed within the optimal weight sampling quadrant. The specific judgment rules are as follows:

[0049] And if x2-x1>0, then d=1, meaning the optimal weight quadrant is the first quadrant;

[0050] And if x2-x1<0, then d=2, meaning the optimal weight quadrant is the second quadrant;

[0051] And if x2-x1<0, then d=3, meaning the optimal weight quadrant is the third quadrant;

[0052] And if x2-x1>0, then d=4, meaning the optimal weight quadrant is the fourth quadrant.

[0053] During algorithm execution, sampling points are randomly generated. It is determined whether they are located in the current optimal weight quadrant. If not, they are discarded and new sampling points are generated. If they are located in the current optimal weight quadrant, they are retained.

[0054] Step 2.2: Find the initial feasible path, denoted as c. best The standard equation of an ellipse is: The focal coordinates are The major axis has a length of 2a, and the minor axis has a length of 2b. Starting from point x... start , endpoint x goal As the focus of the ellipse, let the major axis 'a' of the ellipse be equal to half the initial path length, that is... Derive all the parameters of the ellipse equation. The eccentricity of the ellipse. Appendix Figure 2 This is a schematic diagram of an elliptical sampling space.

[0055] Where c min Starting from x start To x goal The straight-line distance between them is the theoretical minimum cost between them, c best Let c be the cost of the actual minimum distance from the starting point to the ending point, which is also the cost of the best solution found so far. Sampling is continuously performed in the generated elliptical sampling space. When a sampling point reappears near the ending point, indicating that a less costly path has been found, the cost of this path is used as the new c. best Update the elliptic sampling space, perform sampling in the new elliptic sampling space, and continue the above iterative process until the preset number of sampling points is reached, at which point the program ends.

[0056] According to the properties of an ellipse: the sum of the distances from any point on the ellipse to the two foci is equal to 2a; the sum of the distances from points outside the ellipse to the two foci is greater than 2a; and the sum of the distances from points inside the ellipse to the two foci is less than 2a. Therefore, after obtaining the initial feasible path to the starting and ending points, a path better than the current solution must be within the range of the ellipse drawn by this algorithm rule. Furthermore, as the path length decreases, the size of the ellipse also decreases, and the path gradually approaches the optimal value.

[0057] Step 3: Based on the sampling points generated in Step 2, a dynamic step size strategy is used to determine the expansion step size of the node when expanding the node, and a new expanded node is generated; the dynamic step size strategy is divided into two different methods: before finding the initial path of the start and end point and after finding the initial path of the start and end point.

[0058] Step 3.1: When no initial feasible solution is found at the starting point and the ending point, random sampling is performed within the feasible region, starting from the starting point with an initial step size γ. init The random tree is expanded. Collision detection is performed each time a new node is generated during the expansion. If no obstacle is detected, the next expansion step size is γ. next =γ now +0.5*γ init This process continues in subsequent searches until a collision is detected during a node expansion. The next expansion step size is γ. next =γ now -0.5γ init The minimum expansion step size is γ init .

[0059] Step 3.2: After the algorithm finds the initial feasible solutions for the start and end points, the search domain becomes an elliptical subset of the state space, using an initial step size γ. initExpand nodes; if no collision is detected in this expansion, the next expansion step size is γ. next =γ now +0.1*γ init This process continues in subsequent searches until a collision is detected during a node expansion, at which point the next expansion step size becomes γ. next =γ now -0.1*γ init If no collision is detected, the minimum sampling step size within the ellipse is γ. init .

[0060] Step 4: Add slope constraint E(x). In this invention, the path length with slope constraint is named the logical path length, and this is used as the standard for judging the path cost. The concept of dimensionality reduction is introduced to reduce the three-dimensional geographic information to two dimensions. The slope ratio coefficient k is determined, and the impact of three-dimensional elevation information on robot energy consumption is transformed into a two-dimensional logical path length. The result is then normalized to facilitate the search for the lowest energy consumption path. Detailed steps are as follows:

[0061] Step 4.1: First, measure the time and energy consumption factor β required for the robot to travel a fixed distance with constant power across different slopes. As an example, the fixed distance is 10m. The specific information is shown in the table below.

[0062]

[0063] Step 4.2: Fit the measured values ​​of energy consumption and slope to a function, establish a mathematical model for slope energy consumption, and determine the fitting function information as follows:

[0064] f(x) = p1*x 2 +p2*x+p3

[0065] Coefficients (confidence margin at 95%):

[0066] p1=-57.14(-96.81,-17.48);

[0067] p2=-0.7143(-1.953, 0.5242);

[0068] p3 = 1.002(0.9942, 1.01)

[0069] The present invention determines p1 = -57.14, p2 = -0.7143, and p3 = 1.000.

[0070] Step 4.3: By establishing the fitting function, the energy consumption factor β corresponding to different slopes can be calculated. When the slope x > 40°, the robot cannot pass through, which has no practical significance and will not be discussed.

[0071] The formula for calculating slope is:

[0072] Elevation difference = starting point 高程 -end 高程

[0073] To simplify calculations and ensure consistent results, the calculated slope is divided by 1000 and used as the input to the fitting function.

[0074] Step 4.4: Normalize the slope ratio coefficient k and determine the logical path calculation formula for adding slope constraint E(x).

[0075] Path E(x) =path true (1+k*10)

[0076]

[0077] k = 1 - f(x)

[0078] Where PathE(x) represents the logical path length after adding the slope constraint E(x), pathtrue represents the Euclidean geometric distance between two points in the two-dimensional map, and f(x) is the energy consumption factor value of the fitting function.

[0079] Step 5: When the distance from the new node to the target node is less than the threshold, it is considered that the target point has been reached.

[0080] The system performs a preset number of iterations to determine the final path. It then optimizes the final path by removing redundant points and outputs the node matrix of the optimal path based on energy consumption constraints.

[0081] The pseudocode for the method described in this invention is as follows:

[0082]

[0083] The pseudocode for the sampling function is as follows:

[0084]

[0085] The pseudocode for the function to calculate the logical path length is as follows:

[0086]

[0087]

[0088] In summary, this invention discloses an improved fast search random tree path planning method based on energy consumption constraints. The algorithm is as follows: Modeling is performed based on actual geographic information, drawing a digital elevation model of the required map, and determining the starting point, ending point, and initial step size; it is determined whether an initial feasible path has been found at the starting and ending points. If not, a quadrant constraint method is used to constrain the sampling method of the sampling points, limiting the sampling domain; if an initial feasible path exists, an elliptical sampling range is established based on the logical path length of the initial feasible path and the Euclidean geometric distance from the starting point to the ending point, and the elliptical sampling range is dynamically updated as a better path is found through continuous iteration; the node expansion step size is determined according to a dynamic step size strategy, generating new expansion nodes; using dimensionality reduction, a slope ratio coefficient is introduced, and energy consumption constraints are added to the heuristic function to generate a logical path length as the optimal path judgment criterion, performing reconnection and rewriting of the parent nodes of nodes; when the distance from the new node to the target node is less than a threshold, the target point is reached. A preset number of iterations are executed to continuously update and determine the final path, and the final path is optimized by deleting redundant points. The algorithm of this invention effectively reduces the randomness and blindness of sampling points, reduces the search space and range of the search tree, improves search efficiency, and takes into account the impact of elevation slope on the energy consumption of the robot's driving process, so that the planned path has the characteristics of being relatively flat and having low energy consumption, thereby improving the robot's working efficiency, reducing working costs, and saving energy.

[0089] The present invention also provides a computer device, including but not limited to one or more processors and a memory, wherein the memory is used to store a computer executable program, the processor reads part or all of the computer executable program from the memory and executes it, and when the processor executes part or all of the computed executable program, it can implement some or all of the steps of the improved fast search random tree path planning method based on energy consumption constraints described in the present invention.

[0090] A computer-readable storage medium storing a computer program, which, when executed by a processor, enables the implementation of the improved fast search random tree path planning method based on energy consumption constraints as described in this invention.

[0091] The computer device may be a laptop, tablet, desktop computer, mobile phone, or workstation. The processor may be a central processing unit (CPU), digital signal processor (DSP), application-specific integrated circuit (ASIC), or off-the-shelf programmable gate array (FPGA).

[0092] The memory described in this invention can be an internal storage unit of a laptop, tablet, desktop computer, mobile phone, or workstation, such as memory or hard disk; or it can be an external storage unit, such as a portable hard disk or flash memory card.

[0093] The specific hardware and software environment used in the implementation of this patent is shown below:

[0094] Table 1. Software and Hardware Environment Parameters

[0095]

[0096] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of the claims of the present invention.

Claims

1. An improved fast search random tree path planning method based on energy consumption constraints, characterized in that, Includes the following steps: S1. Model based on actual geographic information, draw the digital elevation model of the required map, and determine the starting point, ending point and initial step size. S2, determine whether there is an initial feasible path between the starting point and the ending point. If not, use the quadrant constraint method to constrain the sampling method of the sampling points, restricting the sampling position and orientation of the sampling points. If there is an initial feasible path, establish an elliptical sampling range based on the logical path length of the initial feasible path and the Euclidean geometric distance from the starting point to the ending point, and continuously iterate to find a better path and dynamically update the elliptical sampling range to obtain a series of sampling points. Specifically, the following steps are included: S2.1: Before finding the initial feasible path, the quadrant constraint method is used. A sampling quadrant constraint mark d is set to determine the coordinates of the starting point and the ending point. A relative coordinate system is established based on the positions of the starting point and the ending point, that is, with the starting point as the origin. Then, the coordinate difference between the ending point and the starting point is calculated. )as well as The sign of the sample determines the quadrant in which the endpoint is located; the specific rules for determining the optimal weight sampling quadrant are as follows: and If d=1, then the optimal weight quadrant is the first quadrant; and If d=2, then the optimal weight quadrant is the second quadrant; and If d=3, then the optimal weight quadrant is the third quadrant; and If d=4, then the optimal weight quadrant is the fourth quadrant; During execution, sampling points are randomly generated. If the generated sampling point is not in the current optimal weight quadrant, it is discarded and a new sampling point is generated. If the generated sampling point is in the current optimal weight quadrant, it is retained. S2.2: Find the initial feasible path, denoted as... Starting point ,end As the focus of the ellipse, let the major axis of the ellipse be... It is equal to half the initial path length, that is , , This yields all the parameters of the ellipse equation and the eccentricity of the ellipse. ; in Starting point arrive The straight-line distance between them represents the theoretical minimum cost between them. To find the actual minimum distance from the starting point to the ending point, feasible paths are iteratively updated. Each time a path with a smaller cost is found, its cost is used as the new minimum distance. Update the elliptic sampling space; S3, based on the sampling points generated in S2, determines the node expansion step size according to the dynamic step size strategy and generates new expanded nodes; S4, the path length with added slope constraints is named logical path length, and a slope ratio coefficient is introduced. Based on the idea of ​​dimensionality reduction, the influence of three-dimensional elevation information on robot energy consumption is transformed into two-dimensional logical path length. Within the elliptical sampling range, the path cost is determined by comparing the logical path length of the extended node obtained in S3, and the path is updated. S5, when the distance from the new node to the target node is less than the set threshold, the target point is reached; The system executes the preset number of iterations from S2 to S4, continuously updating and determining the final path. The final path is then optimized by removing redundant points, and the final optimal path based on energy consumption constraints is output.

2. The improved fast search random tree path planning method based on energy consumption constraints according to claim 1, characterized in that, S3 utilizes a dynamic step-size strategy to determine the expansion step size when expanding nodes. This strategy is mainly divided into two different expansion strategies: one when feasible solutions for the start and end points are not found, and the other after they are found. S3.1: When no initial feasible solution is found at the starting and ending points, random sampling is performed within the feasible region, starting with an initial step size. The random tree is expanded. Collision detection is performed each time a new node is generated during the expansion. If no obstacle is detected, the next expansion step size is... And so on in subsequent searches, until a collision is detected during a node expansion, at which point the next expansion step size is... The minimum expansion step size is ; S3.2: After finding an initial feasible solution for the start and end points, the search domain becomes an elliptical subset of the state space, using an initial step size. Expand nodes; if no collision is detected in this expansion, the next expansion step size is... This process continues in subsequent searches until a collision is detected during a node expansion, at which point the next expansion step size becomes... If no collision is detected, the minimum sampling step size within the ellipse is... .

3. The improved fast search random tree path planning method based on energy consumption constraints according to claim 1, characterized in that, In S4, based on the concept of dimensionality reduction, the three-dimensional geographic information is reduced to two dimensions. A slope ratio coefficient is introduced to transform the impact of three-dimensional elevation information on robot energy consumption into a two-dimensional logical path length, including: S4.1, the time and energy consumption factor required for the robot to traverse a fixed distance with constant power across different slopes. Take measurements. S4.2, perform function fitting on the measured values ​​of energy consumption and slope to establish a slope energy consumption fitting function; S4.3, using the established fitting function, calculate the energy consumption factor corresponding to different slopes. ; S4.4, Normalized slope ratio coefficient According to the slope ratio coefficient And the Euclidean geometric distance between two points in a 2D map is determined by incorporating slope constraints. The logical path length.

4. The improved fast search random tree path planning method based on energy consumption constraints according to claim 3, characterized in that, In S4.2, the slope energy consumption fitting function is: in, , 、 is a coefficient.

5. The improved fast search random tree path planning method based on energy consumption constraints according to claim 4, characterized in that, In S4.4, slope constraints are added. The formula for calculating the logical path length is as follows: in, Indicates the addition of slope constraints The length of the logical path following the condition. This represents the Euclidean geometric distance between two points on a two-dimensional map.

6. A computer device, characterized in that, It includes a processor and a memory, the memory being used to store a computer-executable program, the processor reading part or all of the computer-executable program from the memory and executing it, and the processor executing part or all of the computed executable program is able to implement the improved fast search random tree path planning method based on energy consumption constraints as described in any one of claims 1-5.

7. A computer-readable storage medium, characterized in that, A computer-readable storage medium stores a computer program that, when executed by a processor, enables the implementation of the improved fast search random tree path planning method based on energy consumption constraints as described in any one of claims 1-5.

Images