Robot motion path planning method and system in complex scene

Abstract

The invention discloses a robot motion path planning method and system in a complex scene, and the method comprises the steps: taking a motion region where a robot is located as a target region, and discretizing the target region into a series of grids; screening passable grids from the series of grids, and determining avoidance weight values of the passable grids; taking the passable grids of which the avoidance weight values meet a set condition as potential passing areas; calculating the shortest path from the starting point to the target point from the potential traffic area; and carrying out smoothing processing on the shortest path to obtain a processed operation path. According to the invention, a local optimal or impassable area can be avoided, the path planning complexity is reduced, and the path planning efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of path planning, and particularly to a method and system for robot motion path planning in complex scenarios. Background Art

[0002] With the continuous development of robot technology, mobile robots have been widely used in many fields such as industry, medical treatment, agriculture, warehousing and logistics. In order to enable robots to efficiently and safely complete tasks in complex environments, their motion path planning technology has become one of the key research directions. The main goal of path planning is to ensure that the robot can avoid obstacles while moving from the starting point to the target point, and can reach the target point in the shortest time or the optimal path, and ensure the smoothness of operation.

[0003] Currently, the technical solutions for robot path planning are mainly divided into global path planning and local path planning. Global path planning usually relies on known environmental information to determine the best path. For example, Dijkstra's algorithm, D algorithm, etc. are common global path planning methods. Although these methods are widely used, they have problems such as high computational complexity and low efficiency of the planned path. Local path planning focuses on the avoidance and adjustment of unknown or obstacles during the movement of the robot. Common methods include the dynamic window approach (DWA), artificial potential field method (APF), rapidly-exploring random tree (RRT), etc. Although these methods can enable the robot to avoid obstacles autonomously in an unknown environment, they are prone to falling into local optima.

[0004] Therefore, to solve the above problems, a method and system for robot motion path planning in complex scenarios are needed, which can avoid falling into local optima or impassable areas, reduce the complexity of path planning, and improve the efficiency of path planning. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to overcome the defects in the prior art, and provide a method and system for robot motion path planning in complex scenarios, which can avoid falling into local optima or impassable areas, reduce the complexity of path planning, and improve the efficiency of path planning.

[0006] The method for robot motion path planning in complex scenarios of the present invention includes:

[0007] Taking the motion area where the robot is located as the target area, and discretizing the target area into a series of grids;

[0008] Screening passable grids from a series of grids, and determining the avoidance weight value of the passable grids;

[0009] Taking the passable grids whose avoidance weight values meet the set conditions as the potential passing areas;

[0010] Calculate the shortest path from the starting point to the target point in the potential passage area;

[0011] Smooth the shortest path to obtain the processed running path.

[0012] Furthermore, determine the avoidance weight value of the passable grid, specifically including:

[0013] Starting from the four directions of up, down, left, and right of the passable grid respectively, calculate the distances from the passable grid to the nearest obstacles in the corresponding directions to obtain the distances in the four directions, and take the minimum value among the four-direction distances as the distance from the passable grid to the nearest obstacle;

[0014] According to the distance from the passable grid to the nearest obstacle, calculate the avoidance weight value using the following formula:

[0015]

[0016] Among them, C i is the avoidance weight value of the passable grid i; C max is the maximum avoidance weight value; D i is the distance from the passable grid i to the nearest obstacle; α is the attenuation coefficient;

[0017] Determine the attenuation coefficient α according to the following formula:

[0018] Among them, α min is the minimum attenuation coefficient; a max is the maximum attenuation coefficient; v represents the current running speed of the robot; v max is the maximum running speed allowed for the robot during operation.

[0019] Furthermore, calculate the shortest path from the starting point to the target point, specifically including:

[0020] Taking the running ground where the robot is located as a plane and the starting point as the origin, construct a plane rectangular coordinate system; among them, take the center point where the grid in the potential passage area is located as the point in the coordinates;

[0021] Set the path sequence and add the starting point to the path sequence; in the coordinate system, determine several adjacent points adjacent to the starting point, calculate the path costs corresponding to the several adjacent points respectively, and add the adjacent point corresponding to the minimum path cost calculated to the path sequence;

[0022] Taking the adjacent point added to the path sequence above as the current point, in the coordinate system, determine several adjacent points adjacent to the current point, calculate the path costs corresponding to the several adjacent points, and add the adjacent point corresponding to the minimum path cost to the path sequence;

[0023] By analogy with the above steps, several points are added to the path sequence until the target point is added to the path sequence, and a complete path sequence is obtained;

[0024] From the complete path sequence, the points in the sequence are taken out in the order of addition, and the path formed by connecting several points from the starting point to the target point in sequence is used as the shortest path.

[0025] Furthermore, the path cost f is determined according to the following formula:

[0026] f = g + h;

[0027] where g is the path cost from the starting point to the current point, and h is the path cost from the current point to the target point;

[0028] g = g pre + ω·(1 + C / C max ); where ω is a set coefficient; C is the avoidance weight value of the grid to which the current point belongs; C max is the maximum avoidance weight value; g pre is the path cost corresponding to the previous point of the current point;

[0029] h = D1·(dx + dy) + (D2 - 2D1)·min(dx, dy);

[0030] where D1 represents the unit movement cost in the horizontal or vertical direction; D2 is the unit movement cost in the diagonal direction; dx is the coordinate difference between the current point and the target point in the x-axis direction; dy is the coordinate difference between the current point and the target point in the y-axis direction.

[0031] Furthermore, the shortest path is smoothed, which specifically includes:

[0032] Performing interval sampling on the path points in the shortest path to obtain key path points;

[0033] Using a B-spline curve to smoothly approximate a given series of key path points, determining a series of smooth points on the B-spline curve, and generating a continuous and smooth curve;

[0034] where a series of smooth points on the B-spline curve are determined according to the following formula:

[0035]

[0036] where P(t) is the smooth point corresponding to the parameter t on the curve; P i is the key path point; n is the number of key path points; N i,k (t) is the B-spline basis function; k is the order of the spline;

[0037] Limit the parameter t to the interval [t min , t max , and make the parameter t uniformly take values within this interval.

[0038] A robot motion path planning system under complex scenarios, including: a region division module, a region recognition module, a path calculation module, and a path optimization module;

[0039] The region division module is used to take the motion region where the robot is located as the target region and discretize the target region into a series of grids;

[0040] The region recognition module is used to screen the passable grids from a series of grids and determine the avoidance weight value of the passable grids; The passable grids whose avoidance weight values meet the set conditions are used as potential passage regions;

[0041] The path calculation module is used to calculate the shortest path from the starting point to the target point from the potential passage regions;

[0042] The path optimization module is used to smooth the shortest path to obtain the processed running path.

[0043] Furthermore, determining the avoidance weight value of the passable grid specifically includes:

[0044] Starting from the four directions of up, down, left, and right of the passable grid respectively, calculate the distance between the passable grid and the nearest obstacle in the corresponding direction to obtain the distances in the four directions, and take the minimum value of the four direction distances as the distance from the passable grid to the nearest obstacle;

[0045] According to the distance from the passable grid to the nearest obstacle, calculate the avoidance weight value using the following formula:

[0046]

[0047] where C i is the avoidance weight value of the passable grid i; C max is the maximum avoidance weight value; D i is the distance from the passable grid i to the nearest obstacle; α is the attenuation coefficient;

[0048] Determine the attenuation coefficient α according to the following formula:

[0049] where α min is the minimum attenuation coefficient; a max is the maximum attenuation coefficient; v represents the current running speed of the robot; v max is the maximum running speed allowed for the robot during operation.

[0050] Further, calculate the shortest path from the starting point to the target point, specifically including:

[0051] Taking the operating ground where the robot is located as a plane and the starting point as the origin, construct a plane rectangular coordinate system; wherein, take the center point of the grid in the potential passage area as the point in the coordinates;

[0052] Set a path sequence, and add the starting point to the path sequence; in the coordinate system, determine several adjacent points adjacent to the starting point, calculate the path costs corresponding to the several adjacent points respectively, and add the adjacent point corresponding to the minimum path cost among the calculated several path costs to the path sequence;

[0053] Taking the adjacent point added to the path sequence above as the current point, in the coordinate system, determine several adjacent points adjacent to the current point, and calculate the path costs corresponding to the several adjacent points, and add the adjacent point corresponding to the minimum path cost to the path sequence;

[0054] By analogy with the above steps, add several points to the path sequence until the target point is added to the path sequence to obtain a complete path sequence;

[0055] From the complete path sequence, take out the points in the sequence in the order of addition, and take the path formed by connecting several points from the starting point to the target point as the shortest path.

[0056] Further, determine the path cost f according to the following formula:

[0057] f = g + h;

[0058] Wherein, g is the path cost from the starting point to the current point, and h is the path cost from the current point to the target point;

[0059] g = g pre + ω·(1 + C / C max ); wherein, ω is a set coefficient; C is the avoidance weight value of the grid to which the current point belongs; C max is the maximum avoidance weight value; g pre is the path cost corresponding to the previous point of the current point;

[0060] h = D1·(dx + dy) + (D2 - 2D1)·min(dx, dy);

[0061] Wherein, D1 represents the unit movement cost in the horizontal or vertical direction; D2 is the unit movement cost in the diagonal direction; dx is the coordinate difference between the current point and the target point in the x-axis direction; dy is the coordinate difference between the current point and the target point in the y-axis direction.

[0062] Further, smooth the shortest path, specifically including:

[0063] Sample the path points in the shortest path at intervals to obtain key path points;

[0064] Use a B-spline curve to smoothly approximate a given series of key path points, determine a series of smooth points on the B-spline curve, and generate a continuous and smooth curve;

[0065] Among them, a series of smooth points on the B-spline curve are determined according to the following formula:

[0066]

[0067] Among them, P(t) is the smooth point corresponding to the parameter t on the curve; P i is the key path point; n is the number of key path points; N i,k (t) is the B-spline basis function; k is the order of the spline;

[0068] Limit the parameter t to the interval [t min , t max , and make the parameter t uniformly take values within this interval.

[0069] The beneficial effects of the present invention are as follows: A method and system for robot motion path planning in a complex scenario disclosed by the present invention calculate the potential passing area by setting the avoidance weight value, thereby ensuring that the generated path has high feasibility and avoiding the robot from falling into a locally optimal or impassable area; by considering the path cost for path planning, the calculated shortest path is more reasonable and efficient; by smoothing the shortest path, the instability caused by sudden changes in the path curvature during the execution of the robot is reduced, thereby improving the executability of the motion and ensuring the stable operation of the robot. Brief Description of the Drawings

[0070] The present invention will be further described below in conjunction with the drawings and embodiments:

[0071] Figure 1 is the flowchart of the motion path planning method of the present invention;

[0072] Figure 2 is the schematic diagram of the motion path planning system of the present invention. Detailed Embodiments

[0073] The following further describes the present invention in conjunction with the drawings of the specification, as shown in the figure:

[0074] This embodiment discloses a method for robot motion path planning in a complex scenario, including the following steps:

[0075] Take the motion area where the robot is located as the target area, and discretize the target area into a series of grids;

[0076] Screen the passable grids from a series of grids and determine the avoidance weight values of the passable grids;

[0077] Take the passable grids whose avoidance weight values meet the set conditions as potential passage areas;

[0078] Calculate the shortest path from the starting point to the target point in the potential passage area;

[0079] Smooth the shortest path to obtain the processed running path.

[0080] In this embodiment, the motion area where the robot is located is used as the target area, and the target area is discretized into a series of grids, and each grid represents a sub-area in the target area; determine the size of a single grid, for example, 0.05m×0.05m or 0.1m×0.1m, and the grid size can be adjusted according to the actual working conditions.

[0081] Use a lidar or a depth camera to collect the environmental information of the target area to identify obstacles; if a certain grid is occupied by an obstacle, mark this grid as an impassable area; if the grid has no obstacles, mark this grid as a passable area. Calculate the distance from each passable grid to the nearest obstacle, and set the avoidance weight value of the grid based on this distance, specifically including:

[0082] Starting from the four directions of up, down, left, and right of the passable grid respectively, calculate the distance from the passable grid to the nearest obstacle in the corresponding direction to obtain the distances in the four directions, and take the minimum value of the four-direction distances as the distance from the passable grid to the nearest obstacle;

[0083] According to the distance from the passable grid to the nearest obstacle, use the following formula to calculate the avoidance weight value:

[0084]

[0085] where C i is the avoidance weight value of the passable grid i; D i is the distance from the passable grid i to the nearest obstacle; C max is the maximum avoidance weight value, usually taken as 255; α is the attenuation coefficient. As the value of α increases, the obtained avoidance weight value becomes smaller, which is suitable for refined obstacle avoidance, but once it is greater than a certain value, the collision risk will increase; on the contrary, if the value of α is smaller, it will make the robot too conservative during the traveling process, and the distance from the obstacle will be farther, which will lead to a too far detour of the running path; so the appropriate value can be selected according to the specific robot characteristics and environmental requirements, such as taking the value of 3 - 5.

[0086] In practical applications, adjust the size of α dynamically according to the following formula:

[0087] Among them, α min is the minimum attenuation coefficient, with a value of 2; a max is the maximum attenuation coefficient, with a value of 8; v represents the current running speed of the robot; v max is the maximum running speed of the robot, that is, during the running process, the maximum running speed allowed for the robot.

[0088] On the basis of setting the maximum avoidance weight value, the avoidance weight threshold can be set according to the actual working conditions, so as to screen out the potential passing areas of the robot before passing. For example, the avoidance weight threshold is 180, that is, the passable grids with an avoidance weight value less than or equal to 180 are used as potential passing areas.

[0089] In this embodiment, to obtain the shortest path from the starting point to the target point from the potential passing areas, it specifically includes:

[0090] Taking the running ground where the robot is located as a plane and the starting point as the origin, a plane rectangular coordinate system is constructed; among them, the center points of the grids in the potential passing areas are used as the points in the coordinates;

[0091] Set the path sequence, and add the starting point to the path sequence; in the coordinate system, determine several adjacent points adjacent to the starting point, calculate the path costs corresponding to the several adjacent points respectively, and add the adjacent point corresponding to the minimum path cost among the calculated several path costs to the path sequence;

[0092] Taking the adjacent point added to the path sequence above as the current point, in the coordinate system, determine several adjacent points adjacent to the current point, and calculate the path costs corresponding to the several adjacent points, and add the adjacent point corresponding to the minimum path cost to the path sequence;

[0093] By analogy with the above steps, add several points to the path sequence until the target point is added to the path sequence, and a complete path sequence is obtained;

[0094] From the complete path sequence, take out the points in the sequence in the order of addition, and the path formed by connecting the several points from the starting point to the target point in sequence is used as the shortest path.

[0095] In this embodiment, the path cost f is determined according to the following formula:

[0096] f = g + h;

[0097] Among them, g is the path cost from the starting point to the current point, and h is the path cost from the current point to the target point.

[0098] g = g pre + ω·(1 + C / Cmax )); where ω is a coefficient with a value ranging from 1 to 2; C is the avoidance weight value of the grid to which the current point belongs; C max is the maximum avoidance weight value; g pre is the path cost corresponding to the previous point of the current point. If the previous point of the current point is the starting point, then g pre has a value of 0.

[0099] h = D1·(dx + dy)+(D2 - 2D1)·min(dx,dy);

[0100] where D1 represents the unit movement cost in the horizontal or vertical direction and can take a value of 1; D2 is the unit movement cost in the diagonal direction and can take a value of dx is the coordinate difference between the current point and the target point in the x-axis direction; dy is the coordinate difference between the current point and the target point in the y-axis direction; D1·(dx + dy) takes into account the total cost in the horizontal and vertical directions; and (D2 - 2D1)·min(dx,dy) takes into account the impact on the path cost when moving in the diagonal direction. Since the diagonal step length is shorter than two horizontal movements (in the x-axis direction) or vertical movements (in the y-axis direction) (D2 < 2D1), the extra unit movement cost needs to be calculated. Further considering that diagonal movement is less than horizontal or vertical movement, the smaller number of movement steps in the x-axis and y-axis directions is taken as a reference, that is, min(dx,dy) is taken as the reference path distance, so as to obtain the path cost correction value (D2 - 2D1)·min(dx,dy).

[0101] In this embodiment, through the above processing, the obtained shortest path is often composed of discrete points, and the path is in a zigzag shape and not smooth enough, which will affect the actual operation of the robot. In order to improve the executability of the trajectory, the above shortest path can be smoothed, so that the path is more continuous, smooth, and the sharp turns are reduced.

[0102] Smoothing the above shortest path specifically includes:

[0103] Since the generated shortest path may contain many small inflection points, some key path points can be sampled at intervals, such as sampling once every 2 - 3 points, to reduce the density of path points while retaining the main inflection points to prevent loss of path information;

[0104] Using a B-spline curve to smoothly approximate a given series of key path points, determining a series of smooth points on the B-spline curve, and generating a continuous and smooth curve;

[0105] Among them, a series of smooth points on the B-spline curve are determined according to the following formula:

[0106]

[0107] Among them, P(t) is the smooth point corresponding to the parameter t on the curve; P i is the critical path point; n is the number of critical path points; N i,k (t) is the B-spline basis function, which determines the shape of the curve. k is the order of the spline, and k takes the value of 3. The parameter t is limited to the interval [0, 1], and the parameter t is uniformly taken within this interval. For example, t can take values such as 0, 0.01, 0.02, …, 1.

[0108] Through the above smoothing process, the inflection points are reduced and the robot motion becomes smoother. Of course, in order to further optimize the above smoothing process, the path can also be adapted to the maximum speed and acceleration of the robot by adjusting the B-spline parameters, which will not be elaborated here.

[0109] The present invention also relates to a robot motion path planning system in a complex scenario. The path planning system corresponds to the path planning method of the above embodiment and can be understood as a path planning system for implementing the above path planning method. The path planning system includes: a region division module, a region recognition module, a path calculation module, and a path optimization module;

[0110] The region division module is used to take the motion region where the robot is located as the target region and discretize the target region into a series of grids;

[0111] The region recognition module is used to screen the passable grids from a series of grids and determine the avoidance weight values of the passable grids; the passable grids whose avoidance weight values meet the set conditions are used as potential passage regions;

[0112] The path calculation module is used to calculate the shortest path from the starting point to the target point from the potential passage regions;

[0113] The path optimization module is used to smooth the shortest path to obtain the processed running path.

[0114] The present invention provides an objective and scientific robot motion path planning method and system, which comprehensively considers the obstacle avoidance ability, feasibility, and executability of the path, thereby obtaining a reliable and optimized running path, and can better adapt to the robot operation and navigation requirements in a complex environment, providing technical support for improving the intelligent level of robot navigation.

[0115] 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 them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for robot motion path planning in complex scenarios, characterized in that: including: Regarding the movement area where the robot is located as the target area, discretizing the target area into a series of grids; Screening passable grids from the series of grids and determining the avoidance weight values of the passable grids; Regarding the passable grids whose avoidance weight values meet the set conditions as the potential passage areas; Calculating the shortest path from the starting point to the target point from the potential passage areas; Smoothing the shortest path to obtain the processed running path.

2. The method for robot motion path planning in a complex scenario according to claim 1, wherein: Determining the avoidance weight values of the passable grids specifically includes: Starting from the four directions of up, down, left, and right of the passable grid respectively, calculating the distances from the passable grid to the nearest obstacle in the corresponding directions to obtain the distances in the four directions, and taking the minimum value of the four direction distances as the distance from the passable grid to the nearest obstacle; Calculating the avoidance weight value according to the distance from the passable grid to the nearest obstacle by using the following formula: Among them, C i is the avoidance weight value of the passable grid i; C max is the maximum avoidance weight value; D i is the distance from the passable grid i to the nearest obstacle; α is the attenuation coefficient; Determining the attenuation coefficient α according to the following formula: Among them, α min is the minimum attenuation coefficient; a max is the maximum attenuation coefficient; v represents the current running speed of the robot; v max is the maximum running speed allowed for the robot during operation.

3. The method for robot motion path planning in a complex scenario according to claim 1, wherein: Calculating the shortest path from the starting point to the target point specifically includes: Taking the running ground where the robot is located as a plane and the starting point as the origin to construct a plane rectangular coordinate system; among them, taking the center points of the grids in the potential passage areas as the points in the coordinates; Setting a path sequence and adding the starting point to the path sequence; in the coordinate system, determining several adjacent points adjacent to the starting point, calculating the path costs corresponding to the several adjacent points respectively, and adding the adjacent point corresponding to the minimum path cost among the calculated several path costs to the path sequence; Taking the adjacent point added to the path sequence above as the current point, in the coordinate system, determining several adjacent points adjacent to the current point, and calculating the path costs corresponding to the several adjacent points, and adding the adjacent point corresponding to the minimum path cost to the path sequence; By analogy with the above steps, adding several points to the path sequence until the target point is added to the path sequence, obtaining a complete path sequence; Taking out the points in the sequence in the order of addition from the complete path sequence, and taking the path formed by connecting the several points from the starting point to the target point in sequence as the shortest path.

4. The method for robot motion path planning in a complex scenario according to claim 3, wherein: Determining the path cost f according to the following formula: f = g + h; where g is the path cost from the starting point to the current point, and h is the path cost from the current point to the target point; g = g pre + ω·(1 + C / C max ); where ω is a set coefficient; C is the avoidance weight value of the grid to which the current point belongs; C max is the maximum avoidance weight value; g pre is the path cost corresponding to the previous point of the current point; h = D1·(dx + dy) + (D2 - 2D1)·min(dx, dy); where D1 represents the unit movement cost in the horizontal or vertical direction; D2 is the unit movement cost in the diagonal direction; dx is the coordinate difference between the current point and the target point in the x-axis direction; dy is the coordinate difference between the current point and the target point in the y-axis direction.

5. The method for robot motion path planning in a complex scenario according to claim 1, wherein: Smoothing the shortest path specifically includes: Sampling the path points in the shortest path at intervals to obtain key path points; Using a B-spline curve to smoothly approximate a given series of key path points, determining a series of smooth points on the B-spline curve, and generating a continuous and smooth curve; where determining a series of smooth points on the B-spline curve according to the following formula: Among them, P(t) is the smooth point corresponding to the parameter t on the curve; P i is the critical path point; n is the number of critical path points; N i,k (t) is the B-spline basis function; k is the order of the spline; Limit the parameter t to the interval [t min , t max , and make the parameter t uniformly take values within this interval.

6. A robot motion path planning system under complex scenarios, characterized in that: including: a region division module, a region recognition module, a path calculation module, and a path optimization module; The area division module is used to take the movement area where the robot is located as the target area and discretize the target area into a series of grids; The area recognition module is used to screen passable grids from a series of grids and determine the avoidance weight values of the passable grids; the passable grids whose avoidance weight values meet the set conditions are used as potential passage areas; The path calculation module is used to calculate the shortest path from the starting point to the target point in the potential passage area; The path optimization module is used to smooth the shortest path to obtain the processed running path.

7. The robot motion path planning system in a complex scenario according to claim 6, wherein: Determining the avoidance weight value of the passable grid specifically includes: Starting from the four directions of up, down, left, and right of the passable grid respectively, calculate the distance from the passable grid to the nearest obstacle in the corresponding direction to obtain the distances in the four directions, and take the minimum value of the four-direction distances as the distance from the passable grid to the nearest obstacle; According to the distance from the passable grid to the nearest obstacle, calculate the avoidance weight value using the following formula: Among them, C i is the avoidance weight value of the passable grid i; C max is the maximum avoidance weight value; D i is the distance from the passable grid i to the nearest obstacle; α is the attenuation coefficient; Determine the attenuation coefficient α according to the following formula: Among them, α min is the minimum attenuation coefficient; a max is the maximum attenuation coefficient; v represents the current running speed of the robot; v max is the maximum running speed allowed for the robot during operation.

8. The robot motion path planning system in a complex scenario according to claim 6, characterized in that: Calculating the shortest path from the starting point to the target point specifically includes: Taking the running ground where the robot is located as a plane and the starting point as the origin, construct a plane rectangular coordinate system; among them, take the center point where the grid is located in the potential passage area as the point in the coordinate; Set a path sequence and add the starting point to the path sequence; in the coordinate system, determine several adjacent points adjacent to the starting point, calculate the path costs corresponding to the several adjacent points respectively, and add the adjacent point corresponding to the minimum path cost calculated to the path sequence; Taking the adjacent point added to the path sequence above as the current point, in the coordinate system, determine several adjacent points adjacent to the current point, and calculate the path costs corresponding to the several adjacent points, and add the adjacent point corresponding to the minimum path cost to the path sequence; By analogy with the above steps, add several points to the path sequence until the target point is added to the path sequence to obtain a complete path sequence; From the complete path sequence, take out the points in the sequence in the order of addition, and take the path formed by connecting several points from the starting point to the target point in turn as the shortest path.

9. The robot motion path planning system in a complex scenario according to claim 8, wherein: Determine the path cost f according to the following formula: f = g + h; Among them, g is the path cost from the starting point to the current point, and h is the path cost from the current point to the target point; g = g pre + ω·(1 + C / C max ); where ω is a set coefficient; C is the avoidance weight value of the grid to which the current point belongs; C max is the maximum avoidance weight value; g pre is the path cost corresponding to the previous point of the current point; h = D1·(dx + dy) + (D2 - 2D1)·min(dx, dy); Among them, D1 represents the unit movement cost in the horizontal or vertical direction; D2 is the unit movement cost in the diagonal direction; dx is the coordinate difference between the current point and the target point in the x-axis direction; dy is the coordinate difference between the current point and the target point in the y-axis direction.

10. The robot motion path planning system in a complex scenario according to claim 6, wherein: Smoothing the shortest path specifically includes: Performing interval sampling on the path points in the shortest path to obtain key path points; Using a B-spline curve to smoothly approximate a given series of key path points, determine a series of smooth points on the B-spline curve, and generate a continuous and smooth curve; Among them, determine a series of smooth points on the B-spline curve according to the following formula: Among them, P(t) is the smooth point corresponding to the parameter t on the curve; P i is the critical path point; n is the number of critical path points; N i,k (t) is the B-spline basis function; k is the order of the spline; The parameter t is limited to the interval [t min , t max , and the parameter t takes values uniformly within this interval.

Images