A-star algorithm robot path planning method and system with improved search freedom

By improving the A-star algorithm's search freedom, and combining local path merging and an improved heuristic A-star algorithm, a smoother and shorter robot path is generated. This solves the problems of non-smooth paths and high energy consumption in the traditional A-star algorithm, and improves the efficiency of robot path planning.

CN117232541BActive Publication Date: 2024-11-15HARBIN INST OF TECH AT WEIHAI
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310661492.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-02
Publication Date
2024-11-15
Estimated Expiration
2043-06-02

AI Technical Summary

Technical Problem

The traditional A-star algorithm fails to comprehensively consider path length and speed factors in robot path planning, resulting in feasible paths that are not theoretically optimal and have many path inflection points and large turning angles.

Method used

By improving the A-star algorithm's search freedom, an improved heuristic A-star algorithm is used to increase the number of directions the current point can search in the surrounding area. Furthermore, by processing the intersections of line segments and obstacles and combining local path merging, a smoother global path is generated.

Benefits of technology

It generates smoother and shorter paths, reducing the time and power consumption during robot movement and improving work efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117232541B_ABST
    Figure CN117232541B_ABST
Patent Text Reader

Abstract

The invention discloses an A-star algorithm robot path planning method and system for improving the search freedom, and relates to the field of robot path planning technology. In order to solve the problem that the existing robot path planning does not consider the time and power required for the mobile robot to complete the task, nor does it comprehensively consider factors such as the path length and speed, and the conventional A-star algorithm is used, because the search strategy has many path turning points and large turning angles, resulting in the resulting feasible path not being the theoretical optimal path. Technical points: The conventional A-star algorithm introduces the number of directions that describe the expansion of the current point to the surrounding area, and obtains a straight line segment by connecting the starting point and the end point to find the intersection point. When intersecting with an obstacle, an intersection point will be obtained, and then the point is set as the center of the circle to draw a circle with R as the radius, and the obstacle boundary point is obtained. Next, the first intersection point is used as the local starting point, and the last obstacle boundary point is used as the local end point, and is brought into the improved A-star algorithm to obtain a local path, and other local paths are connected by straight lines. Finally, these local paths are combined into a global path.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of robot path planning, and in particular to an A-star algorithm robot path planning method and system for improving search freedom. Background Art

[0002] Mobile robots play an important role in the country's industrialization and informatization process. They have been widely used in agricultural production, industrial manufacturing, marine development, transportation, aerospace and national defense. In recent years, autonomous driving car technology is also gradually being realized, which has promoted the development of mobile robot technology. However, enabling robots to complete tasks autonomously in complex environments is still one of the key and difficult points of research, because planning an effective and feasible path and moving from the starting point to the end point without collision can complete navigation. In order to consider the time and power required for the mobile robot to complete the task, the path obtained by path planning needs to comprehensively consider factors such as path length and speed. The traditional A-star algorithm has the disadvantages of many path turning points and large turning angles in the search strategy, and the feasible path generated is not the theoretically optimal path. Summary of the invention

[0003] The technical problem to be solved by the present invention is: in order to solve the problems that the existing robot path planning does not take into account the time and power consumption required for the mobile robot to complete the task, nor does it comprehensively consider factors such as path length and speed, and the traditional A-star algorithm is used because the search strategy has the disadvantages of many path turning points and large turning angles, resulting in the resulting feasible path not being the theoretically optimal path. Thus, an A-star algorithm robot path planning method and system with improved search freedom are provided.

[0004] The technical solution adopted by the present invention to solve the above technical problems is:

[0005] An A-star algorithm robot path planning method for improving search freedom, the implementation process of the method is:

[0006] Step 1: The robot obtains the environmental information around the current map, which includes its own position, the robot's position in the environment, the starting point, the end point, and the obstacle information. The starting point in the environmental information is A(x start ,y start ), target point B(x goal ,y goal ), connect the starting point and the target point to get a straight line Define that there are at least two obstacles between the starting point A and the end point B;

[0007] Step 2: Use analytic geometry to find the straight line segment based on the coordinates of the starting point and the end point The equation is:

[0008] (yy start )(x goal -x start )=(xx start )(y goal -y start ) (1)

[0009] Step 3: Detect the straight line segment connecting the start and end points Whether to pass through the obstacle. If not, the optimal path between the starting point and the end point is a straight line segment. Otherwise, jump to step 4;

[0010] Step 4: Find the obstacles and straight line segments first The intersection point C 1i (x obs ,y obs )(i=1,2,3., take the first intersection point C 11 (x obs ,y obs ), with C 11 (x obs ,y obs ) is the center of the circle, and r is the radius. The equation of the circle is:

[0011] (xx obs ) 2 +(yy obs ) 2 =r 2 (2) Calculate the circle and line segments based on the equations of the line and circle The intersection of

[0012] Step 5: Find the edge points that pass through the obstacle (points E1(x1,y1), E2(x2,y2)), and calculate the edge points E1, E2 and the intersection point C 11 (x obs ,y obs ) distance:

[0013]

[0014] If d1>d2, keep point E2, otherwise keep point E1, assuming that E1 is at a distance of 100 m from the intersection point C. 11 (x obs ,y obs ) are closer, so E1 is retained and E2 is discarded;

[0015] Step 6: Determine the number of intersections between line AB and the obstacle; if the number of intersections is equal to 1, execute step 7; if the number of intersections is greater than 1, execute step 8;

[0016] Step 7: Determine the number of intersections between line AB and the obstacle. If the number of intersections is 1, use point As the starting point, point E1(x1,y1) is the end point, and the A-star algorithm is used to find the shortest path as local path 1, that is, point The curve segment to the point E1(x1,y1);

[0017] Starting point A(x start ,y st )arrive The straight line connects the local path 2 from point E1(x1,y1) to the end point B(x goal ,y go ) The local path 3 is obtained by connecting the two paths in a straight line. Then, the local paths 1-3 are merged, i.e., A→D1→E1→B, to obtain the optimal path.

[0018] Step 8: If the number of intersection points is greater than 1, use point E1(x1,y1) as the new starting point and find the distance from point E1(x1,y1) to point B(x goal ,y goal ) and repeat steps 2 to 4 to get point C 21 (x obs2 ,y obs2 ),

[0019] Step 9: Repeat step 5 to get edge points E3(x3,y3) and E4(x4,y4); determine the distance from point E1(x1,y1) to point B(x goal ,y goal ) The number of intersections of the straight line segments connected by the two obstacles. If the number of intersections is greater than 1, execute step eight.

[0020] If the number of intersection points is equal to 1, then the point As the starting point, point E3 (x3, y3) is the end point, and the A-star algorithm is used to solve the local path 1. The starting point A (x start ,y start )arrive The straight line connects the local path 2 from point E3(x3,y3) to the end point B(x goal ,y goal ) connect the local path 3 by straight lines, and merge the local paths 1-3 to get the optimal path, namely A→D1→E3→B.

[0021] Furthermore, it is defined that there are at least two long obstacles between the starting point A and the end point B.

[0022] Furthermore, the A-star algorithm is an improved heuristic A-star algorithm, which improves the robot's search freedom by increasing the number of directions searched around the current point n, thereby making the planned path smoother.

[0023] Furthermore, the improved heuristic A-star algorithm is specifically:

[0024] Taking the current position of the robot (current point) as the benchmark, the cost function is used to score the surrounding points, and the point with the smallest score is selected as the next point. The next point is the position that the robot will reach at the next moment.

[0025] The cost function of the A-star algorithm is defined as:

[0026] f(n)=h(n)+g(n) (4)

[0027] Where n is the current inspection point; f(n) is the sum of the evaluation functions; g(n) is the actual path cost from the starting point to the current inspection point; h(n) is the estimated cost from the current inspection point to the target point; h(n) is calculated using the Euclidean distance between the inspection point and the target point:

[0028]

[0029] Where (x n ,y n ) is the coordinate of the current observation point; (x goal ,y goal ) are the coordinates of the target point;

[0030] Increase the number of directions searched around the current point n. When the parameter Connecting_distance is defined as 1, 2, 3, and 4, 8, 16, 32, and 48 directions are generated respectively. That is, when the connecting distance Connecting_distance = 1, 2, 3, and 4, the A-star algorithm will expand in 8, 16, 32, and 48 directions.

[0031] The improved heuristic static path planning system for mobile robots is characterized in that the system has a program module corresponding to the steps of the above technical solution, and executes the steps of the above-mentioned A-star algorithm robot path planning method with improved search freedom during operation.

[0032] A computer-readable storage medium stores a computer program, wherein the computer program is configured to implement the steps of the above-mentioned A-star algorithm robot path planning method for improving search freedom when called by a processor.

[0033] The present invention has the following beneficial technical effects:

[0034] The A-star algorithm robot path planning method with improved search freedom proposed in the present invention can evaluate its performance by designing different map environments. At the same time, the evaluation index is defined to compare with the existing algorithm through simulation experiments to evaluate the effectiveness and feasibility of the path planning algorithm based on the heuristic method. The innovation of the present invention is as follows: the number of directions describing the expansion of the current point to the surrounding is introduced into the traditional A-star algorithm, and a straight line segment is obtained by connecting the starting point and the end point to find the intersection point. When intersecting with an obstacle, an intersection point will be obtained, and then the point is set as the center of the circle to draw a circle with R as the radius to find the obstacle boundary point. Next, the first intersection point is used as the local starting point and the last obstacle boundary point is used as the local end point, and brought into the improved A-star algorithm to obtain a local path 1, and other local paths are connected by straight lines. Finally, these local paths are combined into a global path. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A flowchart of the method of the present invention;

[0036] Figure 2 It is a detailed schematic diagram of the specific steps of the implementation process of the present invention;

[0037] Figure 3 The following are the azimuth expansion diagrams using different Connecting_distance parameters: (a) is the azimuth expansion diagram when the parameter Connecting_distance=1, (b) is the azimuth expansion diagram when the parameter Connecting_distance=2, (c) is the azimuth expansion diagram when the parameter Connecting_distance=3, and (d) is the azimuth expansion diagram when the parameter Connecting_distance=4;

[0038] Figure 4 This is a comparison chart of the traditional A-star and the improved algorithm when there is no obstacle.

[0039] Figure 5 The results of the traditional A-star algorithm and the improved algorithm using different Connecting_distance parameters are shown in the figure: (a) is the result comparison diagram when the parameter Connecting_distance = 1, (b) is the result comparison diagram when the parameter Connecting_distance = 2, (c) is the result comparison diagram when the parameter Connecting_distance = 3, and (d) is the result comparison diagram when the parameter Connecting_distance = 4;

[0040] Figure 6 The results of the traditional A-star algorithm and the improved algorithm using different Connecting_distance parameters are shown in the figure: (a) is the result comparison diagram when the parameter Connecting_distance = 1, (b) is the result comparison diagram when the parameter Connecting_distance = 2, (c) is the result comparison diagram when the parameter Connecting_distance = 3, and (d) is the result comparison diagram when the parameter Connecting_distance = 4. DETAILED DESCRIPTION

[0041] Combined with Figures 1 to 6 The implementation process of the A-star algorithm robot path planning method for improving the search freedom described in the present invention is described as follows:

[0042] Step 1: The robot obtains the environmental information around the current map, such as its own position, the robot's position in the environment, the starting point, the end point, and obstacle information. Figure 1 The starting point in the environmental information is A(x start ,y start ), target point B(x goal ,y goal ), connect the starting point and the target point. Get the straight line

[0043] Step 2: Use analytic geometry to find the straight line segment based on the coordinates of the starting point and the end point The equation is:

[0044] (yy start )(x goal -x start )=(xx start )(y goal -y start ) (1)

[0045] Step 3: Detect the straight line segment connecting the start and end points Whether to pass through the obstacle. If not, the optimal path between the starting point and the end point is a straight line segment. Otherwise, jump to step 4;

[0046] Step 4: Find the obstacles and straight line segments first The intersection point C 1i (x obs ,y obs )(i=1,2,3., take the first intersection point C 11 (x obs ,y obs ), with C 11 (x obs,y obs ) is the center of the circle, and r is the radius. The equation of the circle is:

[0047] (xx obs ) 2 +(yy obs ) 2 =r 2 (2)

[0048] Calculate the circle and line segments based on the equations of the line and circle The intersection of

[0049] Step 5: Find the edge points that pass through the obstacle (points E1(x1,y1), E2(x2,y2)), and calculate the edge points E1, E2 and the intersection point C 11 (x obs ,y obs ) distance:

[0050]

[0051] If d1>d2, keep point E2, otherwise keep point E1. Figure 2 The obvious point E1 is close to the intersection point, so keep E1 and discard E2;

[0052] Step 6: Determine the number of intersections between the straight line and the obstacle. If the number of intersections is equal to 1, use the point As the starting point, point E1(x1,y1) is the end point, and the A-star algorithm is used to find the shortest path as local path 1, such as Figure 2 (a) Point The blue line to point E1(x1,y1);

[0053] The A-star algorithm is a typical heuristic algorithm. Given a global (static) map and a starting point and an end point, the A-star algorithm can generate a more reasonable path. Its core principle is to use the current point as a benchmark, use the cost function to score the surrounding points, and select the point with the smallest score as the next point.

[0054] The cost function of the A-star algorithm is defined as:

[0055] f(n)=h(n)+g(n) (4)

[0056] Where n is the current inspection point; f(n) is the sum of the evaluation functions; g(n) is the actual path cost from the starting point to the current inspection point; h(n) is the estimated cost from the current inspection point to the target point. The calculation method of h(n) uses the Euclidean distance between the inspection point and the target point:

[0057]

[0058] Where (x n ,y n ) is the coordinate of the current observation point; (x goal ,y goal ) are the coordinates of the target point.

[0059] In the standard A-star algorithm, only 8 surrounding points are searched, which may lead to a suboptimal path. In order to make the robot path smoother, the number of directions from the current point to the surrounding points is increased on this basis. When the parameter Connecting_distance is defined as 1, 2, 3 and 4, 8, 16, 32 and 48 possible directions are generated respectively. In other words, when the connecting distance Connecting_distance = 1, 2, 3 and 4, the algorithm will expand in 8, 16, 32 and 48 directions. The effect diagram is as follows Figure 3 As shown in Table 1, the mapping relationship between the number of directions and the parameter Connecting_distance is shown in Table 1.

[0060] Table 1 Relationship between Connecting_distance and the number of extended directions

[0061]

[0062] Starting point A(x start ,y st )arrive The straight line connects the local path 2 from point E1(x1,y1) to the end point B(x goal ,y go ) is connected by straight lines to obtain local path 3, then the local paths are merged: A→D1→E1→B, such as Figure 2 (a)

[0063] Step 7: If the number of intersection points is greater than 1, use point E1(x1,y1) as the new starting point and find the distance from point E1(x1,y1) to point B(x goal ,y goal ), repeat steps 2 to 4 to get the point

[0064] Step 8: Repeat steps 6 to 7 to get points E3(x3,y3) and E4(x4,y4). Determine the straight line from point E1(x1,y1) to point B(x goal ,y goal ) passes through the number of intersection points of obstacles. Figure 2 (b) Map situation, when the number of intersection points is equal to 1, then the point As the starting point, point E3 (x3, y3) is the end point, and the A-star algorithm is used to solve the local path 1. The starting point A (x start ,y start )arrive The straight line connects the local path 2 from point E3(x3,y3) to the end point B(x goal ,y goal ) The local path 3 is obtained by connecting the straight lines. The final modification method merges the local paths into: A→D1→E3→B. Figure 2 (b) as shown.

[0065] The simulation analysis for the present invention is as follows:

[0066] Mobile robot map environment establishment: a grid map with a length of 140 and a width of 120. When the index value corresponding to the position in the grid map is 1, it is an obstacle, and when the index value corresponding to the position in the grid map is 0, it means that there is no obstacle here.

[0067] The configuration of the evaluation platform is as follows: CPU: Intel(R) Core(TM) i5-7200U CPU@2.50GHz 2.70GHz, 8G RAM, Windows 10 64bit, evaluation environment: Matlab R2016a.

[0068] When there are no obstacles between the starting point and the end point, the corresponding simulation results are as follows: Figure 4 The specific data are shown in Table 2.

[0069] Table 2 Comparison data without obstacles

[0070]

[0071] For the simulation results of two obstacles, the starting point is set to (20, 10) and the end point is set to (70, 110). The comparison results of the improved algorithm and the traditional algorithm corresponding to different directional diagrams are as follows Figure 5 The specific data are shown in Table 3.

[0072] Table 3 Results of traditional A-star algorithm and improved algorithm using different Connecting_distance parameters

[0073]

[0074]

[0075] For multiple obstacle simulation results, the starting point (10, 10) and the end point (70, 110) are set. The comparison results of the improved algorithm and the traditional algorithm corresponding to different directional diagrams are as follows Figure 6The specific data are shown in Table 4.

[0076] Table 4 Results of traditional A-star algorithm and improved algorithm using different Connecting_distance parameters

[0077]

[0078] According to the above results, it can be inferred that the improved A-star algorithm performs better than the traditional A-star algorithm in calculating this problem and can obtain a shorter path length. At the same time, by comparing the performance at each connection distance, we can find that the improved A-star algorithm performs better than the traditional A-star algorithm at all connection distances, which further proves that the improved algorithm has better performance.

[0079] The above study made a detailed comparison of the performance of the traditional A-star algorithm and the improved algorithm in different static environments of mobile robots, and reached the following conclusions: the traditional A-star algorithm has defects such as too large turning angles, uneven paths, and non-optimal paths; the improved algorithm is superior to the traditional algorithm, not only planning a better path, but also with the expansion of the orientation, the obtained path is smoother, and the path length is shortened, thereby reducing the time and power energy consumed by the robot during movement. This also ensures that the robot has enough power when performing tasks, improving the robot's work efficiency. Therefore, the present invention verifies the feasibility, effectiveness and superiority of the improved algorithm in the path planning problem of mobile robots, which has important practical significance.

Claims

1. A robot path planning method using the A-star algorithm to improve search freedom, characterized in that: The implementation process of the method is: Step 1: The robot obtains the environmental information around the current map, which includes its own position, the robot's position in the environment, the starting point, the end point, and the obstacle information. The starting point in the environmental information is A(x start ,y start ), target point B(x goal ,y goal ), connect the starting point and the target point to get a straight line Define that there are at least two obstacles between the starting point A and the end point B; Step 2: Use analytic geometry to find the straight line segment based on the coordinates of the starting point and the end point The equation is: (yy start )(x goal -x start )=(xx start )(and goal -and start ) (1) Step 3: Detect the straight line segment connecting the start and end points Whether to pass through the obstacle. If not, the optimal path between the starting point and the end point is a straight line segment. Otherwise, jump to step 4; Step 4: Find the obstacles and straight line segments first The intersection point C 1i (x obs ,y obs ), i=1,2,3..., take the first intersection point C 11 (x obs ,y obs ), with C 11 (x obs ,y obs ) is the center of the circle, and r is the radius. The equation of the circle is: (x-x obs ) 2 +(y-y obs ) 2 =r 2 (2) Calculate the circle and line segments based on the equations of the line and circle The intersection of Step 5: Find the edge points E1(x1,y1) and E2(x2,y2) that pass through the obstacle, and calculate the intersection point C with the edge points E1 and E2. 11 (x obs ,y obs ) distance: If d1>d2, keep point E2, otherwise keep point E1, assuming that E1 is at a distance of 100 m from the intersection point C. 11 (x obs ,y obs ) are closer, so E1 is retained and E2 is discarded; Step 6: Determine the number of intersections between straight line AB and obstacles; If the number of intersection points is equal to 1, execute step 7; if the number of intersection points is greater than 1, execute step 8; Step 7: Determine the number of intersections between line AB and the obstacle. If the number of intersections is 1, use point As the starting point, point E1(x1,y1) is the end point, and the A-star algorithm is used to find the shortest path as local path 1, that is, point The curve segment to point E1(x1,y1); Starting point A(x start ,y start )arrive The straight line connects the local path 2 from point E1(x1,y1) to the end point B(x goal ,y goal ) The local path 3 is obtained by connecting the two paths in a straight line. Then, the local paths 1-3 are merged, i.e., A→D1→E1→B, to obtain the optimal path. Step 8: If the number of intersection points is greater than 1, use point E1(x1,y1) as the new starting point and find the distance from point E1(x1,y1) to point B(x goal ,y goal ) and repeat steps 2 to 4 to get point C 21 (x obs2 ,y obs2 ), Step 9: Repeat step 5 to get edge points E3(x3,y3) and E4(x4,y4); determine the distance from point E1(x1,y1) to point B(x goal ,y goal ) The number of intersections of the straight line segments connected by the two obstacles. If the number of intersections is greater than 1, execute step eight. If the number of intersection points is equal to 1, then the point As the starting point, point E3 (x3, y3) is the end point, and the A-star algorithm is used to solve the local path 1. The starting point A (x start ,y start )arrive The straight line connection obtains the local path 2, from point E3(x3,y3) to the end point B(x goal ,y goal ) connect the local path 3 by straight lines, and merge the local paths 1-3 to get the optimal path, namely A→D1→E3→B.

2. The A-star algorithm robot path planning method for improving search freedom according to claim 1, characterized in that: Define that there are at least two long obstacles between the starting point A and the end point B.

3. The A-star algorithm robot path planning method for improving search freedom according to claim 1 or 2, characterized in that: The A-star algorithm is an improved heuristic A-star algorithm, which improves the robot's search freedom by increasing the number of directions searched around the current point n, thereby making the planned path smoother.

4. The A-star algorithm robot path planning method for improving search freedom according to claim 3 is characterized in that: The improved heuristic A-star algorithm is specifically: Taking the current position of the robot as the benchmark, the cost function is used to score the surrounding points, and the point with the smallest score is selected as the next point. The next point is the position that the robot will reach at the next moment. The cost function of the A-star algorithm is defined as: f(n)=h(n)+g(n) (4) Where n is the current inspection point; f(n) is the sum of the evaluation functions; g(n) is the actual path cost from the starting point to the current inspection point; h(n) is the estimated cost from the current inspection point to the target point; h(n) is calculated using the Euclidean distance between the inspection point and the target point: Where (x n ,y n ) is the coordinate of the current observation point; (x goal ,y goal ) are the coordinates of the target point; Increase the number of directions searched around the current point n. When the parameter Connecting_distance is defined as 1, 2, 3, and 4, 8, 16, 32, and 48 directions are generated respectively. That is, when the connecting distance Connecting_distance = 1, 2, 3, and 4, the A-star algorithm will expand in 8, 16, 32, and 48 directions.

5. Applied to improve the heuristic static path planning system of mobile robots, characterized by: The system has a program module corresponding to the steps of any one of claims 1 to 4 above, and executes the steps of the A-star algorithm robot path planning method for improving search freedom when running.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is configured to implement the steps of an A-star algorithm robot path planning method for improving search freedom according to any one of claims 1 to 4 when called by a processor.