Intelligent ship route planning method based on improved D-star lite algorithm

Abstract

The invention provides an intelligent ship route planning method based on an improved D satellite lite algorithm, and the method comprises the following steps: S1, selecting a diagonal function as a heuristic function of the D satellite lite algorithm for the D satellite lite algorithm, and carrying out the weighting processing of the heuristic function, and obtaining an improved heuristic function of the D satellite lite algorithm; s2, introducing risk factors into the path function of the D-satellite lite algorithm for the D-satellite lite algorithm to obtain the path function, into which the risk factors are introduced, of the D-satellite lite algorithm; s3, aiming at the D satellite lite algorithm, adjusting a search strategy of the D satellite lite algorithm; s31, dynamically adjusting the search step size of a D satellite lite algorithm; S32, increasing the selectable direction of a path in a search strategy; s4, obtaining an improved D-star lite algorithm based on S1, S2 and S3, and performing intelligent ship route planning based on the improved D-star lite algorithm; according to the method, navigation safety is guaranteed, and the planning time, the path length and the path smoothness are optimized, so that the method better fits an actual route.

Description

Technical Field

[0001] The present invention relates to the technical field of ship route planning, and in particular to a ship intelligent route planning method based on an improved D-star lite algorithm. Background Art

[0002] With the rapid development of today's shipping industry and the continuous maturity of intelligent technology, autonomous navigation technology has attracted much attention. The core of intelligent navigation lies in the autonomous navigation capability of ships, that is, the ability of ships to complete navigation tasks safely and efficiently without human intervention according to environmental changes and mission requirements. In this context, intelligent route planning for ships has become one of the important research areas in the shipping industry. Its goal is to enable ships to plan safe and efficient routes based on meteorological information, sea conditions and voyage missions through intelligent means, thereby improving the safety and economy of ship navigation. In addition, the problem of intelligent route planning has become more and more complex, and the requirements for algorithms have become higher and higher.

[0003] Traditional ship route planning methods usually use the A* algorithm and the LPA algorithm. In the A* algorithm, two sets are maintained: the open set is used to save the path points to be examined, and the closed set is used to save the path points that have been examined. The algorithm starts from the starting point and searches for the shortest path through the eight neighborhoods until the target point appears in the closed list, and the search process ends. This method is suitable for path planning in a static environment, but it needs to be called frequently in a dynamic environment, and the amount of calculation is large. The LPA algorithm generates a path from the starting point to the end point by iteratively optimizing the A algorithm. When the map changes, the path is updated through incremental search to improve efficiency. Then the A* algorithm is used again to replan the path from the new starting point to the new end point. This method is suitable for path planning problems with fixed starting and end points, but when the starting point or end point changes, the path needs to be replanned. In addition, when the environment changes dynamically, the current path may no longer maintain optimality. The D star lite algorithm, also known as the D*lite algorithm, uses a reverse search method and can effectively handle dynamic and uncertain path planning problems. The algorithm can quickly adjust the path when the environment changes. Its effect is comparable to that of the D algorithm (Dijkstra algorithm), but it is more efficient. Although the D Lite algorithm is a global path planning algorithm, it can adapt to changes in the local environment due to its incremental nature. When encountering unknown mutation obstacles, the D Lite algorithm can efficiently replan the path. In addition, the algorithm uses heuristic information to guide the search process, so as to find the optimal path faster and improve the efficiency and performance of path planning.

[0004] However, the D-star lite algorithm has the following problems when solving the route planning problem: (1) There are too many extended nodes, especially when facing a complex sea environment, the algorithm efficiency is lower; (2) The planned path is too close to obstacles, and the safety is poor; (3) There are too many turning points in the planned path, which does not conform to the actual navigation. During the actual navigation process, the ship cannot turn continuously and smoothly at the turning points. Summary of the invention

[0005] In view of this, the purpose of the present invention is to propose a ship intelligent route planning method based on an improved D-star lite algorithm to solve the technical problems of low algorithm efficiency, poor security and many turning points in the planned path when the existing D-star lite algorithm is used for route planning.

[0006] The technical means adopted by the present invention are as follows:

[0007] A method for intelligent ship route planning based on an improved D-star lite algorithm comprises the following steps:

[0008] S1. For the D-star lite algorithm, the diagonal function is selected as the heuristic function of the D-star lite algorithm, and the heuristic function is weighted to obtain the improved heuristic function of the D-star lite algorithm;

[0009] S2. For the D-star lite algorithm, a risk factor is introduced into the path function of the D-star lite algorithm to obtain the path function of the D-star lite algorithm with the risk factor introduced;

[0010] S3. Adjust the search strategy of the D-star lite algorithm for the D-star lite algorithm;

[0011] S31, Dynamically adjust the search step size of the D star lite algorithm

[0012] S32, increasing the selectable direction of the path in the search strategy;

[0013] S4, based on S1, S2 and S3, an improved D-star lite algorithm is obtained, and intelligent ship route planning is performed based on the improved D-star lite algorithm;

[0014] S41. For the problem of ship route planning, the D-star lite algorithm has many extended nodes, low efficiency in complex environments, many planning path turning points, and is too close to obstacles. Through the improved D-star lite algorithm, a route planning model is constructed. Based on the grid environment map, the starting point and the end point are set, and the route planning model is used to perform the first route planning to obtain the pre-planned route;

[0015] S42, the ship is caused to sail along the pre-planned route. When the ship detects that the navigation environment has changed, the current starting point is used as a new starting point, and a new planned route is dynamically replanned according to the path field information obtained from the first route planning, and the ship continues to sail along the new planned route;

[0016] S43, the ship is caused to sail along the newly planned route. When the ship detects that the navigation environment has changed, the current starting point is used as the new starting point, and the new planned route is dynamically replanned according to the path field information obtained by the new planned route planning, and the ship continues to sail along the new planned route;

[0017] S44, repeat S43 until the voyage mission is completed.

[0018] Furthermore, in S1, the calculation formula of the improved heuristic function is as follows:

[0019] h(s,s start )=0.9(x s -x start |+|y s -y start |+(d'-d)min(x s -x start |,|y s -y start |))

[0020] In the formula, h(s,s start ) is the heuristic function, which represents the cost of the starting node and the current node; (x s ,y s ) represents the coordinates of the current node, (x start ,y start ) represents the coordinates of the starting point, d represents the moving distance of the horizontally or vertically adjacent grids, and d' represents the moving distance of the diagonally adjacent grids.

[0021] Furthermore, in S2, the calculation formula of the path function that introduces the risk factor is as follows:

[0022]

[0023] In the formula, (x s ,y s ) is the coordinate of the current node s, (x s' ,y s' ) is the coordinate of the successor node s' of node s, and the risk factor Danger selects an appropriate safety distance value according to actual needs.

[0024] Furthermore, S31 specifically includes the following steps:

[0025] A mixed search strategy with different step sizes is adopted. The search step size is initially set to 3 for path search. If there are obstacles within the step size range, the search step size is dynamically adjusted to 2 to continue searching. If there are still obstacles within the step size range, the search step size is further adjusted to 1, thereby achieving a smooth path while avoiding oblique crossing of obstacles.

[0026] Furthermore, S41 specifically includes the following steps: based on the improved D-star lite algorithm, a ship route planning model is constructed, and a path is searched from the end point until the starting point is reached, and the current shortest route, i.e., the pre-planned route, is found.

[0027] Furthermore, S42 specifically includes the following steps: the ship sails according to the pre-planned route and obtains environmental information at regular intervals. If the environmental information changes, that is, the state of some grid nodes changes from 0 to 1 or from 1 to 0, the current node is used as the starting point, and the state information of the node and the affected nodes is updated according to the previous path field information, and a path from the end point to the current node is replanned, and the ship continues to sail along the new route.

[0028] The present invention also provides a storage medium, which includes a stored program, wherein when the program is run, any of the above-mentioned ship intelligent route planning methods based on the improved D-star lite algorithm is executed.

[0029] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes any one of the above-mentioned ship intelligent route planning methods based on the improved D-star lite algorithm through the operation of the computer program.

[0030] Compared with the prior art, the present invention has the following advantages:

[0031] The method of the present invention improves the heuristic function, effectively solves the problem of excessive node expansion and low efficiency of the D-star lite algorithm in a complex sea environment, and is more in line with the algorithm principle.

[0032] The present invention introduces a risk factor to ensure that a certain safe distance is maintained between the planned route and obstacles, thereby solving the problem of the ship being too close to obstacles when the D-star lite algorithm plans the route.

[0033] According to the actual navigation situation, the method of the present invention realizes path smoothing by dynamically adjusting the search step size, solves the problem of discontinuous navigation of the D-star lite algorithm at the ship turning point, and is more in line with the D-star lite algorithm principle and the actual route planning requirements.

[0034] The method of the present invention is used in the process of ship navigation, and when a change in the navigation environment is detected, the route can be dynamically re-planned in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0036] Figure 1 The figure is a flow chart of the route planning method of the present invention.

[0037] Figure 2 This is a comparison chart of the search effects of different heuristic functions of the present invention.

[0038] Figure 3 This is a comparison diagram of different step lengths and selectable directions of the present invention.

[0039] Figure 4 Schematic diagram of the search strategy of the present invention.

[0040] Figure 5 Schematic diagram of the route planning model of the present invention.

[0041] Figure 6 This is the flow chart of the improved D-star lite algorithm of the present invention. DETAILED DESCRIPTION

[0042] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0043] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0044] like Figure 1 As shown, a method for intelligent ship route planning based on an improved D-star lite algorithm according to an embodiment of the present invention comprises the following steps:

[0045] S1. Aiming at the planning efficiency problem of intelligent route planning method, the heuristic function of D-star lite algorithm is improved.

[0046] The choice of heuristic function will directly affect the search effect and speed of the algorithm, such as Figure 2 As shown in the figure, the Euclidean distance is used to calculate the shortest distance between two points. It is suitable for distance calculation in any angle movement direction, but the calculation is complex and there are many extended nodes. The diagonal distance is suitable for the distance of diagonal movement, and the calculation is simple, the number of node extensions is small, and it can effectively improve the operation speed while more accurately describing the path cost of the grid graph. However, considering the incremental characteristics of the D-star lite algorithm, the path field information searched during the first planning will be used, so the number of extended nodes cannot be too small.

[0047] Therefore, the heuristic function of the improved D-star lite algorithm application is:

[0048] h(s,s start )=0.9(|x s -x start |+|y s -y start |+(d'-d)min(|x s -x start |,|y s -y start |))

[0049] Among them, h(s,s start ) is the heuristic function, which represents the cost of the starting node and the current node; (x s ,y s ) represents the coordinates of the current node, (x start ,y start ) represents the coordinates of the starting point, d represents the moving distance of the horizontally or vertically adjacent grids, and d' represents the moving distance of the diagonally adjacent grids.

[0050] S2. Aiming at the problem of route safety in intelligent route planning method, the path function of D-star lite algorithm is improved.

[0051] Since the method of the present invention adds the path search step length and direction in the subsequent part, the diagonal distance cannot accurately express the actual cost of the path, so the Euclidean distance is selected as the path cost evaluation function between nodes.

[0052] The risk factor introduced into the cost function is to give a certain penalty to the grid near the obstacle, so that the planned path maintains a certain safe distance from the obstacle. The calculation formula is as follows:

[0053]

[0054] Among them, (x s ,y s ) is the coordinate of the current node s, (x s' ,y s' ) is the coordinate of the successor node s' of node s. The risk factor Danger can select an appropriate safety distance value according to actual needs.

[0055] S3. To address the route smoothness problem of the intelligent route planning method, adjust the search strategy of the D-star lite algorithm.

[0056] According to the different optimization timing and methods, curve optimization can be divided into post-processing optimization and in-process optimization. The most widely used method is post-processing optimization, that is, adjustment and optimization are performed after the curve is generated or calculated, such as B-spline curve, Bezier curve, linear interpolation, spline interpolation and least squares curve fitting, etc., to post-process the planned path to make the path smoother. Since the post-processing optimization method will modify the path points of the pre-planned path, causing the original path information to become invalid, thereby affecting the accuracy of the re-planning result, the path pre-planned by the D-star lite algorithm is not suitable for optimization through post-processing.

[0057] The method of the present invention adopts the process optimization method and realizes curve optimization by adjusting the search strategy of the D-star lite algorithm, which is as follows: first, by adjusting the search step size of the D-star lite algorithm, the algorithm expansion is no longer limited to adjacent grids. Secondly, the optional direction of the path is increased so that the algorithm can select other better grids.

[0058] However, by Figure 3 It can be seen intuitively that as the search step size and the selectable direction increase, although the path length and the number of turning points decrease, the increased algorithm complexity leads to an increase in path planning time and a decrease in efficiency. To address this problem, the present invention proposes a fan-shaped search strategy to solve the problem of increased algorithm complexity caused by increasing the search step size and the selectable direction. Figure 4 shown.

[0059] At the same time, from Figure 4 It can be seen that increasing the search step size may cause the path to cross the non-navigable grid obliquely, thereby reducing the safety of ship navigation. Therefore, the present invention solves the problem of crossing the non-navigable grid obliquely caused by increasing the search step size by dynamically adjusting the search step size, that is, adopting a mixed search with different step sizes, and improves the path safety.

[0060] Dynamically adjust the search step size, that is, adopt a mixed search strategy with different step sizes. Since the planning time of the search strategy with 4 steps (56 directions) is too long and the calculation is complicated, the search step size is initially set to 3 for path search. If there are obstacles within the step size range, the search step size is dynamically adjusted to 2 to continue searching. If there are still obstacles within the step size range, the search step size is further adjusted to 1, so as to achieve a smooth path while avoiding oblique obstacles.

[0061] Increase the selectable directions of the path in the search strategy, that is, in each step, not only 8 directions are considered, but more directions can also be considered, such as 16 selectable directions for 2 steps or 32 selectable directions for 3 steps.

[0062] S4. Regarding the ship route planning problem, the D*lite algorithm has many expansion nodes, low efficiency in complex environments, many turning points in the planned path, and is too close to obstacles. Based on these problems, the D*lite algorithm is improved through the above S1, S2 and S3, and a route planning model is constructed. Based on the grid environment map, the starting point and the end point are set, and the first route planning is performed to obtain the pre-planned route. The ship sails along the pre-planned route. When the ship detects that the navigation environment has changed, it takes the current starting point as the new starting point and dynamically replans a route based on the path field information obtained from the first route planning. The ship continues to sail along the new route until the voyage mission is completed.

[0063] For the problem of intelligent route planning, the ship route planning model is constructed as follows: Figure 5 The improved D-star lite algorithm flow chart is as follows Figure 6 shown.

[0064] The D-star lite algorithm is a graph search algorithm that can be implemented and run on a rasterized map. It simplifies the complex environment by dividing it into small cells, each of which can be passable or impassable, thereby taking advantage of this map representation method to improve the efficiency and adaptability of path planning. In order to apply the graph search algorithm and plan ship routes using the D-star lite algorithm, it is first necessary to construct a sea area grid map. Based on the acquired hydrological and meteorological information, the navigation environment is processed using the grid method and converted into a simple geometric relationship environment model, thereby constructing a sea area environment model.

[0065] Based on the constructed marine environment model and ship route planning model, the first route is carried out from the end point until it reaches the starting point, and the current shortest route is found, that is, the pre-planned route. The ship sails according to the pre-planned route, and obtains environmental information at a certain interval, compares the current environmental information with the last environmental information obtained, and if the environmental information changes, the current node is used as the starting point, and the status information of the node and the affected nodes is updated according to the previous path field information, and a path from the end point to the current node is re-planned. The ship continues to sail along the new route, and repeats the above steps until the voyage mission is completed.

[0066] The planned route obtained by adopting the present application is more suitable for ship route planning, whether in the pre-planning stage or the re-planning stage. While ensuring navigation safety, it can optimize the planning time, path length and path smoothness, and is more in line with the actual navigation path.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A ship intelligent route planning method based on improved D-star lite algorithm, characterized in that: The steps include: S1. For the D-star lite algorithm, the diagonal function is selected as the heuristic function of the D-star lite algorithm, and the heuristic function is weighted to obtain the improved heuristic function of the D-star lite algorithm; S2. For the D-star lite algorithm, a risk factor is introduced into the path function of the D-star lite algorithm to obtain the path function of the D-star lite algorithm with the risk factor introduced; S3. Adjust the search strategy of the D-star lite algorithm for the D-star lite algorithm; S31, dynamically adjust the search step size of the D star lite algorithm; S32, increasing the selectable direction of the path in the search strategy; S4, based on S1, S2 and S3, an improved D-star lite algorithm is obtained, and ship intelligent route planning is performed based on the improved D-star lite algorithm; S41. For the problem of ship route planning, the D-star lite algorithm has many extended nodes, low efficiency in complex environments, many planning path turning points, and is too close to obstacles. Through the improved D-star lite algorithm, a route planning model is constructed. Based on the grid environment map, the starting point and the end point are set, and the route planning model is used to perform the first route planning to obtain the pre-planned route; S42, the ship is caused to sail along the pre-planned route. When the ship detects that the navigation environment has changed, the current starting point is used as a new starting point, and a new planned route is dynamically replanned based on the path field information obtained in the first route planning, that is, the node status information searched in the first route planning, and the ship continues to sail along the new planned route; S43, the ship is caused to sail along the newly planned route. When the ship detects that the navigation environment has changed, the current starting point is used as the new starting point, and the new planned route is dynamically replanned according to the path field information obtained by the new planned route planning, and the ship continues to sail along the new planned route; S44, repeat S43 until the voyage mission is completed.

2. The ship intelligent route planning method based on the improved D-star lite algorithm according to claim 1 is characterized in that: In S1, the calculation formula of the improved heuristic function is as follows: h(s,s start )=0.9(|x s -x start |+|y s -y start| +(d'-d)min(|x s -x start |,|y s -y start |)) In the formula, h(s,s start ) is the heuristic function, which represents the cost of the starting node and the current node; (x s ,y s ) represents the coordinates of the current node, (x start ,y start ) represents the coordinates of the starting point, d represents the moving distance of the horizontally or vertically adjacent grids, and d' represents the moving distance of the diagonally adjacent grids.

3. The ship intelligent route planning method based on the improved D-star lite algorithm according to claim 1 is characterized in that: In S2, the calculation formula of the path function introducing the risk factor is as follows: In the formula, (x s ,y s ) is the coordinate of the current node s, (x s' ,y s' ) is the coordinate of the successor node s' of node s, and the risk factor Danger selects an appropriate safety distance value according to actual needs.

4. The method for ship intelligent route planning based on improved D-star lite algorithm according to claim 1 is characterized in that: S31 specifically includes the following steps: A mixed search strategy with different step sizes is adopted. The search step size is initially set to 3 for path search. If there are obstacles within the step size range, the search step size is dynamically adjusted to 2 to continue searching. If there are still obstacles within the step size range, the search step size is further adjusted to 1, thereby achieving a smooth path while avoiding oblique crossing of obstacles.

5. The method for ship intelligent route planning based on improved D-star lite algorithm according to claim 1 is characterized in that: S41 specifically includes the following steps: based on the improved D-star lite algorithm, a ship route planning model is constructed, and a path is searched from the end point until the starting point is reached, and the current shortest route, i.e., the pre-planned route, is found.

6. The method for ship intelligent route planning based on improved D-star lite algorithm according to claim 1 is characterized in that: S42 specifically includes the following steps: the ship sails according to the pre-planned route and obtains environmental information at regular intervals. If the environmental information changes, that is, the state of some grid nodes changes from 0 to 1 or from 1 to 0, the current node is used as the starting point, and the state information of the node and the affected nodes is updated according to the previous path field information, and a path from the end point to the current node is replanned, and the ship continues to sail along the new route.

7. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is run, the ship intelligent route planning method based on the improved D-star lite algorithm described in any one of claims 1 to 6 is executed.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The processor executes the ship intelligent route planning method based on the improved D-star lite algorithm as described in any one of claims 1 to 6 through the computer program.

Images