Multi-fuzzy-control navigation path planning method, storage medium and computer equipment

Abstract

The invention discloses a multi-fuzzy-control navigation path planning method, a storage medium and computer equipment. The multi-fuzzy-control navigation path planning method comprises the following steps: S1, acquiring map environment information, and determining positions of a starting point, a target point and an obstacle; s2, acquiring operation parameters of the vehicle; s3, calculating node collision times and node extension lengths, calculating node priorities by utilizing a fuzzy rule based on the node collision times and the node extension lengths, selecting existing nodes according to the priorities, and generating new nodes; s4, for the generated new node, calculating the distance and the included angle between the new node and the front obstacle; based on the distance and the included angle between the new node and the front obstacle, generating a new extension direction by using a fuzzy rule, and outputting the steering angle of the vehicle; and S5, the step S3 and the step S4 are cycled, and new nodes and new extension directions are continuously generated until the vehicle reaches the target point from the starting point. According to the method, the path planning efficiency and the convergence speed in a complex space environment can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and particularly to a navigation path planning method with multi-fuzzy control, a storage medium, and a computer device. Background Art

[0002] As an important device in the fields of warehousing logistics, medical services, industrial manufacturing, etc., the path planning efficiency and safety of intelligent carts directly affect the operation quality. In complex spaces, intelligent carts often face challenges such as multiple obstacles, narrow passages, and maze-like structures, and their own characteristics of limited steering, such as the minimum turning radius limit caused by mechanical constraints such as differential drive and Ackerman steering, further exacerbate the difficulty of path planning.

[0003] Spaces with complex structures generally have a large number of special environmental constraints, such as numerous and messy obstacles, concave and convex obstacles, narrow passages, and maze-like structures. In such complex space environments, traditional random sampling algorithms have a slow convergence speed and a significant drop in efficiency when performing path planning, and even fall into an infinite loop. Summary of the Invention

[0004] Object of the Invention: In order to overcome the deficiencies in the prior art, the present invention provides a navigation path planning method with multi-fuzzy control, a storage medium, and a computer device, which improves the path planning efficiency and convergence speed in complex space environments by introducing two fuzzy variables, namely steering and node priority.

[0005] Technical Solution: To achieve the above object, the navigation path planning method with multi-fuzzy control of the present invention includes the following steps: Step S1: Obtain map environment information, and determine the positions of the starting point, the target point, and the obstacles; Step S2: Obtain the operating parameters of the vehicle; Step S3: Calculate the number of node collisions and the node extension length, and based on the number of node collisions and the node extension length, use fuzzy rules to calculate the node priority, and select existing nodes according to the priority to generate new nodes; Step S4: For the generated new nodes, calculate the distance and angle between the new nodes and the obstacles ahead; Based on the distance and angle between the new nodes and the obstacles ahead, use fuzzy rules to generate a new expansion direction and output the steering angle of the vehicle; Step S5: Loop Step S3 and Step S4, continuously generate new nodes and new expansion directions until the vehicle reaches the target point from the starting point.

[0006] Further, in Step S3, the calculation formula for the node priority is:

[0007]

[0008] where p is the priority of the node, β is the weight size, n p is the number of times the node hits the wall, L is the node extension length, nq is the number of nodes connected between the node and the initial point, and Δq is the unit step size.

[0009] Further, when calculating the node priority using fuzzy rules, as the extension length of the node increases, the node priority has an increasing trend, and as the number of times the node hits the wall increases, the node priority has a decreasing trend.

[0010] Further, in step S4, the calculation formula for the steering angle is: |θ| = αS + (1 - α)γ; where S is the distance between the node and the obstacle ahead, γ is the angle between the current direction of the node and the obstacle ahead, α is the weight size, and θ is the steering angle of the vehicle for the next step.

[0011] Further, when calculating the expansion direction using fuzzy rules, as the distance between the node and the obstacle increases, the steering angle of the vehicle for the next step has a decreasing trend, and as the angle between the node and the obstacle increases, the steering angle of the vehicle for the next step also has a decreasing trend.

[0012] Further, when calculating the node priority, different fuzzy rules are used for mapping based on the size of the minimum turning radius of the vehicle; under the mapping of different fuzzy rules, for the same distance or angle with the obstacle, when the minimum turning radius of the vehicle increases, the steering angle of the vehicle for the next step also has an increasing trend.

[0013] Further, in step S2, the information of the goods carried on the vehicle is obtained, and the size of the minimum turning radius of the vehicle is judged according to the information of the goods carried on the vehicle.

[0014] Further, in step S2, the weight and slenderness value of the goods carried on the vehicle are obtained, where the slenderness value of the goods carried on the vehicle is the ratio of the square of the height of the goods carried on the vehicle to the cross-sectional area of the goods carried on the vehicle; based on the weight and slenderness value of the goods carried on the vehicle, the size of the minimum turning radius of the vehicle is calculated using fuzzy rules. As the weight of the goods carried on the vehicle increases, the minimum turning radius of the vehicle has an increasing trend, and as the slenderness value of the goods carried on the vehicle increases, the minimum turning radius of the vehicle has an increasing trend.

[0015] Further, a storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the multi-fuzzy control navigation path planning method can be implemented.

[0016] Further, a computer device includes a processor, a storage medium, and a computer program stored in the storage medium and executable by the processor. When the computer program is executed by the processor, the steps of the multi-fuzzy control navigation path planning method can be implemented.

[0017] Beneficial effects: The multi-fuzzy control navigation path planning method, storage medium, and computer device of the present invention have the following beneficial effects:

[0018] 1) Steering fuzzy control optimizes the direction: This algorithm introduces two inputs, the distance and angle to the obstacle ahead, and adjusts the steering angle through fuzzy control. When the vehicle's steering is restricted, it can select an appropriate steering angle according to the position and direction of the obstacle, reducing ineffective expansion and improving the efficiency of path planning;

[0019] 2) Node priority fuzzy control reduces ineffective calculations: By calculating the number of times a node hits the wall and the extension length, the priority of the node is fuzzy-controlled. Nodes with a long extension length and few wall-hitting times are preferentially expanded, avoiding repeated calculations of ineffective nodes, optimizing the search path, and reducing the amount of calculation;

[0020] 3) Improve path smoothness: The path generated by this algorithm is smoother, reducing frequent steering commands, conforming to the vehicle's steering ability, reducing energy consumption and wear during driving, and improving operation efficiency and the service life of the equipment;

[0021] 4) Optimize the rationality of the path: Since the minimum turning radius of the vehicle is affected by the on-vehicle cargo, the minimum turning radius of the vehicle is comprehensively judged based on the weight and slenderness value of the on-vehicle cargo, and then the fuzzy rules for vehicle steering are adjusted based on the minimum turning radius of the vehicle, making the path planned by the algorithm more reasonable. Description of the Drawings

[0022] Attached Figure 1 is a flowchart of the multi-fuzzy control navigation path planning method of the present invention;

[0023] Attached Figure 2 is a node priority fuzzy rule control chart;

[0024] Attached Figure 3 is a three-dimensional mapping diagram of node priority fuzzy control;

[0025] Attached Figure 4 is a steering fuzzy rule control chart when the vehicle's minimum turning radius is small;

[0026] Attached Figure 5 is a three-dimensional mapping diagram of steering fuzzy control when the vehicle's minimum turning radius is small;

[0027] Attached Figure 6 is a steering fuzzy rule control chart when the vehicle's minimum turning radius is large;

[0028] Attached Figure 7 is a three-dimensional mapping diagram of steering fuzzy control when the vehicle's minimum turning radius is large;

[0029] Appendix Figure 8 Fuzzy rule control chart for the minimum turning radius of a vehicle;

[0030] Appendix Figure 9 Three-dimensional mapping diagram of the fuzzy control of the minimum turning radius of a vehicle. Specific implementation manner

[0031] The present invention will be further described below with reference to the accompanying drawings.

[0032] As in the Figures 1 to 5 multi-fuzzy control-based navigation path planning method, which includes the following steps: Step S1: Obtain map environment information, and determine the positions of the starting point, the target point, and the obstacles;

[0033] Step S2: Obtain the operating parameters of the vehicle, including basic operating parameters of the vehicle such as vehicle size and motion control thresholds, and also including relevant information about the on-vehicle cargo;

[0034] Step S3: Calculate the number of node collisions and the node extension length. Based on the number of node collisions and the node extension length, use fuzzy rules to calculate the node priority, and select existing nodes according to the priority to generate new nodes;

[0035] Step S4: For the generated new nodes, calculate the distance and angle between the new nodes and the obstacles ahead; Based on the distance and angle between the new nodes and the obstacles ahead, use fuzzy rules to generate new expansion directions and output the steering angle of the vehicle;

[0036] Step S5: Loop through Step S3 and Step S4, continuously generate new nodes and new expansion directions until the vehicle reaches the target point from the starting point.

[0037] In practical application scenarios such as warehousing logistics, vehicles often face complex environments such as dense distribution of multiple obstacles, narrow channels, and maze-like spaces. Traditional path planning methods are difficult to efficiently process multi-dimensional environmental information in complex environments, such as obstacle positions, spatial layout constraints, etc., resulting in either the planned path being unable to avoid obstacles or being infeasible due to not considering kinematic constraints such as limited steering of intelligent trolleys.

[0038] The multi-fuzzy control-based navigation path planning method proposed by the present invention, through a multi-fuzzy control mechanism, on the one hand, uses fuzzy rules to deeply analyze environmental parameters such as the distance and angle between nodes and obstacles ahead, and dynamically generates steering angles that meet the obstacle avoidance requirements and vehicle motion characteristics; on the other hand, combines fuzzy evaluation of node priorities to specifically optimize the node expansion strategy, enabling the algorithm to accurately adapt to spatial constraints and obstacle distributions in complex environments, effectively solving the problems of feasibility and accuracy of path planning in complex environments, and improving the operation ability of vehicles in complex scenarios.

[0039] In step 4, after the steering angle of the output vehicle is obtained, the required angular velocity and linear velocity for driving are calculated in combination with the dynamic model, and the differential steering system controls the motion execution mechanism in sequence to achieve differential steering motion. At the same time, the positioning module real-time feedbacks the vehicle position to form a closed-loop control, enabling the vehicle to continuously and cyclically execute path planning and motion adjustment based on fuzzy control in a complex environment, achieving efficient obstacle avoidance and precise path driving control.

[0040] In step S3, the present invention designs a fuzzy controller, which consists of two fuzzy input variables and one output variable. The first input variable is the number of times n that the node hits the wall p , by calculating the number of times each node hits the wall, the probability of calculating the node with more wall-hitting times is reduced. The second input variable is the length L of the node extension, calculating the length between each node and the initial node, and increasing the probability of calculating the node with a longer extension distance. The output variable is the priority p of the calculated node. The higher the priority of the node, the greater the probability of calculating this node. The calculation formula for the node priority is as follows:

[0041]

[0042] L=(n q +1)Δq;

[0043] where p is the priority of the node, β is the weight size, the value of β ranges from 0 to 1, n p is the number of times the node hits the wall, L is the length of the node extension, n q is the number of nodes connected between the node and the initial point, and Δq is the unit step size.

[0044] When calculating the node priority using fuzzy rules, as the length of the node extension increases, the node priority has an increasing trend, and as the number of times the node hits the wall increases, the node priority has a decreasing trend.

[0045] In a specific embodiment, in step S3, the fuzzy language of the number of times the node hits the wall is set as {zero, small, large}, represented by {ZO, S, B}, and the corresponding change range is [0, 6], and the universe of discourse is {0, 1, 3, 6}.

[0046] The maximum value of the node extension length is set as the distance from the starting point to the target point. Assuming this distance is 450m, the fuzzy language of the node extension length is set as {small, medium, large}, represented by {S, M, B}, and the corresponding change range is [0, 450], and the universe of discourse is {0, 150, 300, 450}, with the unit being m.

[0047] The fuzzy language of the node priority is set as {small, medium, large}, represented by {S, M, B}, and the corresponding change range is [0.1, 1.0], and the universe of discourse is {0.1, 0.4, 0.7, 1.0}.

[0048] When calculating the node priority, the following fuzzy rules are used for mapping. For details, please refer to the appendix Figure 2 as shown below:

[0049] If the number of times a node hits the wall is zero and the node extension length is small, the priority of the node is medium;

[0050] If the number of times a node hits the wall is zero and the node extension length is medium, the priority of the node is large;

[0051] If the number of times a node hits the wall is zero and the node extension length is large, the priority of the node is large;

[0052] If the number of times a node hits the wall is small and the node extension length is small, the priority of the node is small;

[0053] If the number of times a node hits the wall is small and the node extension length is medium, the priority of the node is small;

[0054] If the number of times a node hits the wall is small and the node extension length is large, the priority of the node is medium;

[0055] If the number of times a node hits the wall is large and the node extension length is small, the priority of the node is small;

[0056] If the number of times a node hits the wall is large and the node extension length is medium, the priority of the node is small;

[0057] If the number of times a node hits the wall is large and the node extension length is large, the priority of the node is small.

[0058] In step S4, the present invention also designs a fuzzy controller, taking the distance S between the vehicle and the obstacle ahead and the angle γ between the current direction of the node and the obstacle ahead as the inputs of the fuzzy controller, and taking the next steering angle θ of the vehicle as the output of the fuzzy controller. Herein, the current direction of the node refers to the current moving direction of the vehicle when the vehicle moves to this node. The specific calculation formula for the steering angle is:

[0059] |θ| = αS + (1 - α)γ;

[0060] where S is the distance between the node and the obstacle ahead, γ is the angle between the current direction of the node and the obstacle ahead, α is the weight value, with a value between 0 and 1, and θ is the next steering angle of the vehicle.

[0061] When calculating the extension direction using fuzzy rules, as the distance between the node and the obstacle increases, the next steering angle of the vehicle has a tendency to decrease, and as the angle between the node and the obstacle increases, the next steering angle of the vehicle also has a tendency to decrease.

[0062] In a specific embodiment, the fuzzy language for setting the distance between the node and the obstacle is {small, medium, large, extremely large}, represented by {S, M, B, VB}, with the corresponding variation range being [10, 50], the universe of discourse being {10, 20, 30, 40, 50}, and the unit being m.

[0063] The fuzzy language for setting the absolute value of the angle between the node and the obstacle is {small, medium, large}, represented by {S, M, B}, with the corresponding variation range being [0, 90], and the universe of discourse being {0, 30, 60, 90}.

[0064] The fuzzy language for setting the steering angle of the vehicle is {small, medium, large}, represented by {S, M, B}, with the corresponding variation range being [0, 180], and the universe of discourse being {0, 45, 90, 180}.

[0065] In warehousing logistics, the load situation on the cargo vehicle will change. The vehicle may be unloaded or fully loaded. Even if it is fully loaded, the total weight of the goods on the vehicle may also be different. In addition, for goods of the same weight, the shape of the goods may also vary. Some goods are slender and some are short and fat. When the goods carried on the vehicle are different, the inertia and the center of gravity of the vehicle will change, thus affecting the actual minimum turning radius of the vehicle. Suppose the vehicle can turn smoothly at a certain turning radius when it is unloaded. However, when the vehicle is fully loaded and the goods on the vehicle are very high, when the vehicle turns at the said turning radius, the vehicle may overturn. Therefore, the minimum turning radius of the vehicle is affected by the on-vehicle goods.

[0066] When the distance between the vehicle and the obstacle is relatively small, if the vehicle wants to avoid the obstacle, the vehicle usually needs to make a large-amplitude turn. When the vehicle makes a large-amplitude turn, if the minimum turning radius of the vehicle is small, then the turning space required by the vehicle is also small, and the vehicle is not likely to hit the obstacle; while if the minimum turning radius of the vehicle is large, then the turning space required by the vehicle is also large. Then, on the premise that the vehicle itself is already close to the obstacle, the vehicle is very likely to hit the obstacle during the turning process, resulting in the vehicle being unable to complete the turn smoothly.

[0067] Therefore, it can be concluded that when the minimum turning radius of the vehicle is different, the optimal path of the vehicle from the starting point to the target point will be different. Therefore, when calculating the node priority, different fuzzy rules are used for mapping based on the size of the minimum turning radius of the vehicle. Under the mapping of different fuzzy rules, for the same distance or angle with the obstacle, when the minimum turning radius of the vehicle increases, the next steering angle of the vehicle also has an increasing trend. The following provides a specific embodiment:

[0068] In the first case, when the minimum turning radius of the vehicle is small, the following fuzzy rules are used for mapping, which can be referred to in the appendix Figure 4 as shown below:

[0069] If the distance between the node and the obstacle is small and the angle between the node and the obstacle is small, the steering angle of the vehicle is medium;

[0070] If the distance between the node and the obstacle is small and the angle between the node and the obstacle is medium, the steering angle of the vehicle is large;

[0071] If the distance between the node and the obstacle is small and the angle between the node and the obstacle is large, the steering angle of the vehicle is large;

[0072] If the distance between the node and the obstacle is medium and the angle between the node and the obstacle is small, the steering angle of the vehicle is small;

[0073] If the distance between the node and the obstacle is medium and the angle between the node and the obstacle is medium, the steering angle of the vehicle is medium;

[0074] If the distance between the node and the obstacle is medium and the angle between the node and the obstacle is large, the steering angle of the vehicle is large;

[0075] If the distance between the node and the obstacle is large and the angle between the node and the obstacle is small, the steering angle of the vehicle is small;

[0076] If the distance between the node and the obstacle is large and the angle between the node and the obstacle is medium, the steering angle of the vehicle is small;

[0077] If the distance between the node and the obstacle is large and the angle between the node and the obstacle is large, the steering angle of the vehicle is medium;

[0078] If the distance between the node and the obstacle is extremely large and the angle between the node and the obstacle is small, the steering angle of the vehicle is small;

[0079] If the distance between the node and the obstacle is extremely large and the angle between the node and the obstacle is medium, the steering angle of the vehicle is small;

[0080] If the distance between the node and the obstacle is extremely large and the angle between the node and the obstacle is large, the steering angle of the vehicle is small.

[0081] In the second case, when the minimum turning radius of the vehicle is large, the following fuzzy rules are used for mapping, which can be referred to in the appendix Figure 6 as shown below:

[0082] If the distance between the node and the obstacle is small and the angle between the node and the obstacle is small, the steering angle of the vehicle is medium;

[0083] If the distance between the node and the obstacle is small and the angle between the node and the obstacle is medium, the steering angle of the vehicle is large;

[0084] If the distance between the node and the obstacle is small and the angle between the node and the obstacle is large, the steering angle of the vehicle is large;

[0085] If the distance between the node and the obstacle is medium and the angle between the node and the obstacle is small, the steering angle of the vehicle is medium;

[0086] If the distance between the node and the obstacle is medium and the angle between the node and the obstacle is medium, the steering angle of the vehicle is medium;

[0087] If the distance between the node and the obstacle is medium and the angle between the node and the obstacle is large, the steering angle of the vehicle is large;

[0088] If the distance between the node and the obstacle is large and the angle between the node and the obstacle is small, the steering angle of the vehicle is small;

[0089] If the distance between the node and the obstacle is large and the angle between the node and the obstacle is medium, the steering angle of the vehicle is medium;

[0090] If the distance between the node and the obstacle is large and the angle between the node and the obstacle is large, the steering angle of the vehicle is medium;

[0091] If the distance between the node and the obstacle is extremely large and the angle between the node and the obstacle is small, the steering angle of the vehicle is small;

[0092] If the distance between the node and the obstacle is extremely large and the angle between the node and the obstacle is medium, the steering angle of the vehicle is small;

[0093] If the distance between the node and the obstacle is extremely large and the angle between the node and the obstacle is large, the steering angle of the vehicle is small.

[0094] Compare Figure 4 and Figure 6 It can be seen that when the minimum turning radius of the vehicle is small, in some mapping relationships, the next steering angle of the vehicle will tend to be smaller. When the minimum turning radius of the vehicle is large, in some mapping relationships, the next steering angle of the vehicle will tend to be larger. For example, when the distance between the node and the obstacle is medium and the angle between the node and the obstacle is small, when the minimum turning radius of the vehicle is small, the steering angle of the vehicle is small, while when the minimum turning radius of the vehicle is large, the steering angle of the vehicle is medium.

[0095] Therefore, when the minimum turning radius of the vehicle is relatively large, during the process of the vehicle gradually approaching an obstacle, a larger steering angle will be adopted in advance, which is equivalent to the vehicle steering in advance to avoid the obstacle. Therefore, when the distance between the vehicle and the obstacle is very close, the angle between the vehicle and the obstacle will be relatively small, enabling the vehicle to avoid the obstacle with a larger turning radius and not easily collide with the obstacle.

[0096] Conversely, if the minimum turning radius of the vehicle is relatively small, then during the process of the vehicle gradually approaching an obstacle, a smaller steering angle will be selected. Then, when the distance between the vehicle and the obstacle is very close, the vehicle can avoid the obstacle with a smaller turning radius and will not collide with the obstacle.

[0097] In summary, in the present invention, since the fuzzy rules of vehicle steering can be adjusted according to the size of the minimum turning radius of the vehicle, the path of the vehicle can be optimized specifically.

[0098] Since the on-vehicle cargo will affect the minimum turning radius of the vehicle, in step S2, it is necessary to obtain the operating parameters of the vehicle, including the weight and slenderness value of the on-vehicle cargo. When the weight of the cargo is relatively large, the inertia of the vehicle will change, making the vehicle more prone to tipping over, so the minimum turning radius of the vehicle will decrease. The weight of the cargo can be detected by a weight sensor. The slenderness value of the on-vehicle cargo is the ratio of the square of the height of the on-vehicle cargo to the cross-sectional area of the on-vehicle cargo. For example, if the on-vehicle cargo is 1 m long, 1 m wide, and 1 m high, then its slenderness value is 1; if the on-vehicle cargo is 2 m long, 1 m wide, and 1 m high, then its slenderness value is 4; if the on-vehicle cargo is 2 m long, 2 m wide, and 1 m high, then its slenderness value is 2. When the on-vehicle cargo tends to be slender, the center of gravity of the vehicle will move upward, making the vehicle prone to tipping over, so the minimum turning radius of the vehicle will decrease. The slenderness value of the cargo can be detected by a vision sensor. In addition, when the cargo is placed on the vehicle, it is defaulted that the cargo will not extend outside the vehicle and affect the size of the vehicle.

[0099] Obtain the weight and slenderness value of the on-vehicle cargo, where the slenderness value of the on-vehicle cargo is the ratio of the square of the height of the on-vehicle cargo to the cross-sectional area of the on-vehicle cargo; based on the weight and slenderness value of the on-vehicle cargo, use fuzzy rules to calculate the size of the minimum turning radius of the vehicle. As the weight of the on-vehicle cargo increases, the minimum turning radius of the vehicle has an increasing trend. As the slenderness value of the on-vehicle cargo increases, the minimum turning radius of the vehicle has an increasing trend.

[0100] Based on the weight and slenderness value of the on-vehicle cargo, use fuzzy rules to calculate the size of the minimum turning radius of the vehicle. Then, when calculating the node priority in step S4, different fuzzy rules are adopted for mapping based on the size of the minimum turning radius of the vehicle.

[0101] In a specific embodiment, in step S2, the fuzzy language for setting the weight of the on-vehicle goods is {small, medium, large}, represented by {S, M, B}, with the corresponding change range being [0, 12], the universe of discourse being {0, 4, 8, 12}, and the unit being kg.

[0102] The fuzzy language for setting the slenderness value of the on-vehicle goods is {small, medium, large}, represented by {S, M, B}; the corresponding change range is [0, 9], and the universe of discourse is {0, 1, 4, 9}.

[0103] The fuzzy language for setting the minimum turning radius of the vehicle is {small, large}, represented by {S, B}; the corresponding change range is [1, 2], the universe of discourse is {1, 1.5, 2}, and the unit is m.

[0104] When calculating the minimum turning radius of the vehicle, the following fuzzy rules are used for mapping, which can be referred to as shown in the appendix Figure 8 as follows:

[0105] If the weight of the on-vehicle goods is small and the slenderness value of the on-vehicle goods is small, then the minimum turning radius of the vehicle is small;

[0106] If the weight of the on-vehicle goods is small and the slenderness value of the on-vehicle goods is medium, then the minimum turning radius of the vehicle is small;

[0107] If the weight of the on-vehicle goods is small and the slenderness value of the on-vehicle goods is large, then the minimum turning radius of the vehicle is small;

[0108] If the weight of the on-vehicle goods is medium and the slenderness value of the on-vehicle goods is small, then the minimum turning radius of the vehicle is small;

[0109] If the weight of the on-vehicle goods is medium and the slenderness value of the on-vehicle goods is medium, then the minimum turning radius of the vehicle is small;

[0110] If the weight of the on-vehicle goods is medium and the slenderness value of the on-vehicle goods is large, then the minimum turning radius of the vehicle is large;

[0111] If the weight of the on-vehicle goods is large and the slenderness value of the on-vehicle goods is small, then the minimum turning radius of the vehicle is large;

[0112] If the weight of the on-vehicle goods is large and the slenderness value of the on-vehicle goods is medium, then the minimum turning radius of the vehicle is large;

[0113] If the weight of the on-vehicle goods is large and the slenderness value of the on-vehicle goods is large, then the minimum turning radius of the vehicle is large.

[0114] Based on the above fuzzy rules, the minimum turning radius of the vehicle can be determined by combining the slenderness value and weight of the on-vehicle goods. When the output is small, it indicates that the minimum turning radius of the vehicle is relatively small; when the output is large, it indicates that the minimum turning radius of the vehicle is relatively large.

[0115] The algorithm flowchart of the present invention is as shown in the appendix Figure 1 . First, the initial point is clearly set to construct the starting node of the random tree, establishing a benchmark starting point for subsequent path planning and completing the preliminary preparation work for the algorithm operation.

[0116] Then, the weight and slenderness ratio information of the on-vehicle goods are obtained, and the minimum turning radius of the vehicle is evaluated according to the fuzzy rules in Figure 8 . If the minimum turning radius of the vehicle is small, then when calculating the steering of the vehicle, the fuzzy rules in Figure 4 are adopted; if the minimum turning radius of the vehicle is large, then when calculating the steering of the vehicle, the fuzzy rules in Figure 6 are adopted instead.

[0117] Next, the nodes in the random tree are evaluated and calculated. According to the two key parameters of "node collision times" and "node extension length", the node priority is calculated and evaluated according to the fuzzy rules in Figure 2 . Through fuzzy logic operations, the node priority determination result is formed, providing a quantitative basis for subsequent node selection and optimizing the node selection strategy.

[0118] After the priority calculation is completed, according to the priority result, the target node is selected from the existing nodes to ensure that the selected node has better expansion value and improve the search efficiency. For the selected node, first calculate its spatial distance from the target point, and then further analyze and calculate the distance and angle between the node and the front obstacle to accurately obtain the environmental constraint information where the current node is located, providing data support for formulating the expansion strategy.

[0119] In addition, differential expansion decisions are implemented based on environmental parameters, and the distance parameter S is determined by the logic judgment module. That is to say, in step S4, a loose distance value, such as 50m, is set, and the distance parameter S between the node and the obstacle is determined. When S is greater than the loose distance value, that is, when the distance between the node and the obstacle is greater than 50m, it indicates that the current environmental constraint is relatively loose, and the new node is controlled to randomly expand towards the target point. When S is less than or equal to the loose distance value, that is, when the distance between the node and the obstacle is less than or equal to 50m, the new node is controlled to generate a new expansion direction using fuzzy rules. Through the fuzzy logic operation that fuses the obstacle distance and the angle, an expansion direction that meets the obstacle avoidance requirements is generated to achieve intelligent expansion decisions in complex environments.

[0120] After completing the expansion of new nodes, the validity of the expanded nodes is verified. If the expanded nodes are determined to be valid, it is further determined whether the termination condition of "reaching the target point" has been achieved. If this condition is met, the process ends, indicating that the path planning has been successfully completed. If the target point has not been reached, the current node is subjected to a counting operation of "collision count + 1", and the process returns to the upstream for continuous iteration. Through continuous iterative optimization, the search process is gradually improved, and ultimately an efficient and accurate path planning is achieved.

[0121] The present invention also provides a storage medium in which a computer program is stored. When the computer program is executed by a processor, the steps of the multi-fuzzy control navigation path planning method can be implemented.

[0122] The present invention also provides a computer device, including a processor, a storage medium, and a computer program stored in the storage medium and executable by the processor. When the computer program is executed by the processor, the steps of the multi-fuzzy control navigation path planning method can be implemented.

[0123] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A navigation path planning method based on multi-fuzzy control, characterized in that: It includes the following steps: Step S1: Obtain the map environment information, and determine the positions of the starting point, the target point, and the obstacles; Step S2: Obtain the running parameters of the vehicle; Step S3: Calculate the number of node collisions and the node extension length. Based on the number of node collisions and the node extension length, use fuzzy rules to calculate the node priority, and select existing nodes according to the priority to generate new nodes; Step S4: For the generated new nodes, calculate the distance and angle between the new node and the obstacle in front; Based on the distance and angle between the new node and the obstacle in front, use fuzzy rules to generate a new expansion direction, and output the steering angle of the vehicle; Step S5: Loop through Step S3 and Step S4, continuously generate new nodes and new expansion directions until the vehicle reaches the target point from the starting point.

2. The multi-fuzzy control-based navigation path planning method according to claim 1, wherein: In Step S3, the calculation formula for the node priority is: Among them, p is the priority of the node, β is the weight size, and n p is the number of times the node hits the wall, L is the extension length of the node, and n q is the number of nodes connected between the node and the initial point, and Δq is the unit step size.

3. The navigation path planning method with multi-fuzzy control according to claim 2, characterized in that: When using fuzzy rules to calculate the node priority, as the node extension length increases, the node priority has an increasing trend, and as the number of node collisions increases, the node priority has a decreasing trend.

4. The multi-fuzzy-control-based navigation path planning method according to claim 1, wherein: In Step S4, the calculation formula for the steering angle is: |θ| = αS + (1 - α)γ; where S is the distance between the node and the obstacle in front, γ is the angle between the current direction of the node and the obstacle in front, α is the weight value, and θ is the steering angle of the vehicle for the next step.

5. The navigation path planning method with multi-fuzzy control according to claim 4, characterized in that: When using fuzzy rules to calculate the expansion direction, as the distance between the node and the obstacle increases, the steering angle of the vehicle for the next step has a decreasing trend, and as the angle between the node and the obstacle increases, the steering angle of the vehicle for the next step also has a decreasing trend.

6. The navigation path planning method with multi-fuzzy control according to claim 5, characterized in that: When calculating the node priority, different fuzzy rules are used for mapping based on the size of the minimum turning radius of the vehicle; under the mapping of different fuzzy rules, for the same size of the distance or angle with the obstacle, when the minimum turning radius of the vehicle increases, the steering angle of the vehicle for the next step also has an increasing trend.

7. The navigation path planning method with multi-fuzzy control according to claim 6, characterized in that: In Step S2, obtain the information of the goods carried on the vehicle, and judge the size of the minimum turning radius of the vehicle according to the information of the on-vehicle goods.

8. The navigation path planning method with multi-fuzzy control according to claim 7, characterized in that: In Step S2, obtain the weight and slenderness value of the on-vehicle goods, where the slenderness value of the on-vehicle goods is the ratio of the square of the height of the on-vehicle goods to the cross-sectional area of the on-vehicle goods; Based on the weight and slenderness value of the on-vehicle goods, use fuzzy rules to calculate the size of the minimum turning radius of the vehicle. As the weight of the on-vehicle goods increases, the minimum turning radius of the vehicle has an increasing trend, and as the slenderness value of the on-vehicle goods increases, the minimum turning radius of the vehicle has an increasing trend.

9. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, it can implement the steps of the multi-fuzzy control navigation path planning method described in any one of claims 1 to 8.

10. A computer device, characterized in that: It includes a processor, a storage medium, and a computer program stored in the storage medium and executable by the processor. When the computer program is executed by the processor, it can implement the steps of the multi-fuzzy control navigation path planning method described in any one of claims 1 to 8.

Images