A method for path planning and obstacle avoidance of unmanned boat formation

Abstract

The present invention discloses a method for path planning and obstacle avoidance of an unmanned boat formation, and relates to the technical field of formation path planning, including system initialization and basic settings; using a quantum genetic algorithm, decoding quantum chromosomes into actual paths, screening out the optimal path, and discretizing the optimal path to obtain a global navigation path; constructing a follower navigation path according to the formation characteristics and the reference formation, adjusting the initialization formation when the follower starts, and establishing a communication link with the navigator at the same time, and the follower using the improved artificial potential field method to avoid obstacles locally and adjust the local formation; detecting a dynamic obstacle, recalculating the force using the improved potential field method, and optimizing the local path of the replanned section; judging whether the target point has been reached, if it has been reached, then ending, if not, then executing in a loop until the target point has been reached. The present invention enables the formation system to maintain a stable formation in a dynamic unknown environment, and achieve the goals of safe obstacle avoidance and optimal path.

Description

Technical Field

[0001] The invention relates to the technical field of formation path planning, and in particular to a path planning and obstacle avoidance method for an unmanned boat formation. Background Art

[0002] As a key component in the field of marine vehicles, unmanned boats are a type of small intelligent platform that can replace humans to perform complex tasks in the marine environment. Unmanned boats play an irreplaceable role in many fields such as marine development, environmental monitoring, and security patrols. However, with the increasing complexity of marine tasks and the continuous and in-depth development of unmanned boat control theory, the limitations of a single unmanned boat in navigation control have become increasingly prominent. Not only does a single unmanned boat have obvious shortcomings in its ability and efficiency when performing tasks, but in the face of increasingly complex mission environments, a single system has significant defects such as limited carrying capacity, narrow coverage, and insufficient environmental adaptability, making it difficult to independently meet diverse mission requirements. Based on this, the study of unmanned boat formation control technology has become an urgent task in the current field of marine vehicles. Among them, the path planning and obstacle avoidance control of the leader-follower collaborative formation is one of the core links to achieve efficient collaborative operations of unmanned systems at sea.

[0003] In the field of path planning and obstacle avoidance control technology, existing research has achieved certain results. Early solutions mostly used traditional optimization algorithms, such as bee swarm algorithm, particle swarm algorithm, genetic algorithm, artificial potential field method, etc. Algorithms such as bee swarm and particle swarm are very likely to produce local optimal results, which will cause path oscillation problems. Although genetic algorithms have global search capabilities, they have problems such as high computational complexity and poor real-time performance, and it is difficult to meet the needs of rapid decision-making in dynamic ocean environments; artificial potential field laws are prone to fall into local minima due to the limitations of the design of gravitational repulsive field functions, resulting in path planning failure or formation obstacle avoidance failure. For example, in narrow sea areas with dense obstacles, the traditional artificial potential field method will cause excessive superposition of repulsive forces, making it impossible for unmanned boats to approach the target point, and even oscillation will occur. Based on this, researchers began to combine traditional algorithms to reduce the impact of each algorithm, such as the combination of genetic algorithm (GA) and artificial potential field method (APF), but the existing technology did not consider the negative impact on overall work efficiency caused by the collaborative operation of multiple algorithms, which led to a significant increase in the amount of calculation. Summary of the invention

[0004] In order to overcome the above problems existing in the prior art, the present invention proposes a path planning and obstacle avoidance method for an unmanned boat formation.

[0005] The technical solution adopted by the present invention to solve the technical problem is: a method for path planning and obstacle avoidance of an unmanned boat formation, comprising the following steps:

[0006] Step 1: rasterize the complex marine environment into a two-dimensional map, set the reference formation and the corresponding formation threshold, clarify the initial information of the unmanned boat, initialize the parameters of the quantum genetic algorithm and the gravitational and repulsive parameters of the improved artificial potential field method;

[0007] Step 2: Use quantum genetic algorithm to decode quantum chromosomes into actual paths, weigh the fitness of each path, path length, resources consumed or deviations generated by avoiding obstacles, and smoothness, screen out the optimal path, and discretize the optimal path to obtain the global navigation path;

[0008] Step 3: Construct the follower navigation path based on the formation characteristics and the reference formation set in step 1. When the follower starts, adjust the initial formation to match the overall formation layout. At the same time, establish a communication link with the leader. The follower uses the improved artificial potential field method to avoid local obstacles and adjust the local formation.

[0009] Step 4: When a dynamic obstacle is detected, local path replanning is triggered, the force is recalculated using the improved potential field method, and step 2 is repeated to optimize the local path of the replanned section;

[0010] Step 5, determine whether the target point has been reached. If it has been reached, the process ends. If it has not been reached, steps 1-4 are executed repeatedly until the target point is reached.

[0011] In the above-mentioned method for path planning and obstacle avoidance of an unmanned boat formation, step 2 specifically includes:

[0012] Step 2.1, generate the quantum initial population, and convert the chromosome into a path node sequence by decoding;

[0013] Step 2.2, calculate the path fitness value F:

[0014] ;

[0015] in, is the weight coefficient; L is the comprehensive path length; C is the obstacle avoidance cost; S is the smoothness;

[0016] Select the path with high fitness value;

[0017] Step 2.3, according to the fitness, apply the rotating gate operation to the quantum chromosome to adjust the quantum bit phase; perform crossover and mutation operations to generate a new population;

[0018] Step 2.4, determine whether the population has converged. If so, output the global optimal path, discretize it into a navigation point sequence and load it. If not, continue iterating.

[0019] In the above-mentioned method for path planning and obstacle avoidance of an unmanned boat formation, step 3 specifically includes:

[0020] Step 3.1, based on the formation characteristics and the reference formation, construct the follower's initial navigation path;

[0021] Step 3.2, update the position in real time, calculate the repulsion of obstacles and the gravitational force of the target, add virtual obstacles, optimize the repulsion function, calculate the direction of the resultant force, and adjust the local path; optimize the path based on the direction of the resultant force, and the follower moves closer to the leader while avoiding obstacles;

[0022] Step 3.3, monitor the formation distance. If it is within the threshold, the leader moves at the speed and the followers make local adjustments. If it is not within the threshold, the speed of the followers is corrected, the leader's speed remains unchanged, and local adjustments are made after returning to the threshold.

[0023] The above-mentioned method for path planning and obstacle avoidance of an unmanned boat formation, wherein the repulsion function is optimized in step 3.2 for:

[0024] ;

[0025] The repulsive force is obtained by optimizing the repulsive force function. :

[0026] ;

[0027] Among them, q represents the current position, is the location of the obstacle, is the repulsion coefficient, It is to improve the repulsive force range of the artificial potential field method. = Indicates the current distance q and The distance between.

[0028] In the above-mentioned method for path planning and obstacle avoidance of an unmanned boat formation, the step 3.2 of adding a virtual obstacle is specifically as follows: when the unmanned boat is detected to be at a local minimum, the obstacle distribution is evaluated and a virtual obstacle is added. The repulsive force generated by the virtual obstacle The calculation formula is:

[0029] ;

[0030] in, is the virtual repulsion coefficient; The vector pointing from the current position to the virtual obstacle position; for Length of mold Represented as the repulsion range of the virtual obstacle.

[0031] The above-mentioned method for path planning and obstacle avoidance of an unmanned boat formation introduces the same repulsion and mutual repulsion of the unmanned boat formation in step 3.3. The formula is:

[0032] ;

[0033] ;

[0034] in, represents the same type repulsion coefficient; is represented as a smoothing kernel function, Expressed as a distance parameter; Indicates the safety distance set for similar boats; represents the layered safety distance; n represents the radial unit vector; Represents the heading weight coefficient; Indicates the angle between the headings of the unmanned boats; represents the tangent unit vector.

[0035] The beneficial effect of the present invention is that in terms of global path planning, a quantum genetic algorithm is used to decode quantum chromosomes, calculate path fitness, use quantum crossover, mutation and revolving door update and other operations. Compared with traditional genetic algorithms, it not only solves the problems of large computational complexity and easy to fall into local optimality, but also greatly improves the global path search efficiency and can quickly output the optimal path.

[0036] In terms of local obstacle avoidance, the potential field function of the artificial potential field method is improved, its range and intensity curve are optimized, and virtual obstacles are added to solve the problem that the traditional artificial potential field method is prone to falling into local minima. At the same time, the problem of path oscillation is dealt with, making local obstacle avoidance more reliable and effectively reducing the failure of path planning.

[0037] In terms of collaborative control, the formation is monitored in real time to maintain the optimal formation as much as possible, solving the problem of easy formation dispersion in existing technologies. At the same time, the repulsion of the same type of formation is increased, effectively avoiding collisions and sudden shocks between members when avoiding obstacles or adjusting the formation.

[0038] In addition, when the system detects a dynamic obstacle, it can combine the improved artificial potential field method to recalculate the force and adjust the follower's movement direction. At the same time, it can update the global path through incremental optimization of the quantum genetic algorithm, thereby enhancing the ability to adapt to complex environments and ensuring the safe operation of the maritime unmanned system formation in complex environments, taking into account both path optimality and obstacle avoidance stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the process of the present invention;

[0040] Figure 2 It is the principle diagram of the artificial potential field method of the present invention;

[0041] Figure 3 It is a schematic diagram of the situation where the artificial potential field method of the present invention is reasonably zero;

[0042] Figure 4 is a schematic diagram of the situation in which a virtual obstacle is added to make the resultant force non-zero in the present invention;

[0043] Figure 5 It is a basic principle diagram of the same type of repulsion of the formation of the present invention. DETAILED DESCRIPTION

[0044] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0045] This embodiment discloses a path planning and obstacle avoidance method for a leader-follower collaborative formation, such as Figure 1 As shown, specifically including:

[0046] 1. Perform environmental preprocessing and parameter initialization. It is necessary to rasterize the complex marine environment into a two-dimensional map, and transform obstacles of various shapes into regular shapes to facilitate the subsequent algorithm processing related calculations. At the same time, set the reference formation, that is, the shape that the formation is expected to maintain during the march, and the corresponding formation threshold to define the acceptable formation deviation range in actual operation. Clarify the initial information of the unmanned boat, covering the basic data such as the initial position and heading of the unmanned boat. On this basis, initialize the key parameters of the quantum genetic algorithm, among which the population size determines the range of the number of paths searched by the algorithm, the number of iterations limits the maximum number of loop optimizations, and the quantum revolving door parameters affect the update direction and amplitude of the quantum bit state. These parameters jointly determine the search characteristics of the algorithm. For the improved artificial potential field method, initialize its gravitational and repulsive coefficients. The gravitational coefficient determines the attraction strength of the unmanned boat to the target point, and the repulsive coefficient determines the repulsion strength of the obstacle to the unmanned boat. Reasonable setting of the coefficient is the basis for accurate obstacle avoidance and efficient path planning.

[0047] 1. Environmental Modeling

[0048] The environment is rasterized into a two-dimensional map, obstacles are converted into regular grid cells, the grid occupancy state (obstacle occupation or feasible area) is defined, and a standardized environment model is constructed.

[0049] 2. Parameter configuration

[0050] Formation parameters: Set parameters such as the number of unmanned boats, reference formation and formation threshold.

[0051] Algorithm parameters: Determine the population size, number of iterations, and quantum revolving door parameters of the quantum genetic algorithm; initialize the gravitational coefficient and repulsive coefficient of the artificial potential field method. At the same time, obtain the initial position, speed, and other kinematic parameters of the unmanned boat.

[0052] Carry out global path planning for the navigator. Use quantum genetic algorithms to decode quantum chromosomes into actual paths. In order to select the best path, it is necessary to consider many factors: calculate the fitness of each path, path length, resources consumed or deviations generated by avoiding obstacles, smoothness, etc. By weighing various factors, the optimal individual is selected. During this period, the algorithm convergence is constantly judged. If it does not converge, the quantum crossover operation is used to exchange some gene fragments between different quantum chromosomes to increase the diversity of the population; the state of some quantum bits is changed through quantum mutation to introduce new search directions; the quantum revolving gate is used to update the state of the quantum bits to promote the evolution of the population in the direction of a better solution. This cycle is repeated until the algorithm converges and the optimal path of the navigator is output. The path is then discretized so that it can be recognized by the system and loaded as a global navigation path, providing macro guidance for the actions of subsequent followers.

[0053] 1. Quantum population initialization and path decoding

[0054] Generate the quantum initial population, and the quantum chromosome is composed of quantum bit coding. The chromosome is converted into a path node sequence by decoding. For example, if binary coding is used, the gene bit corresponds to the grid map node, and the path from the starting point to the target point is formed after decoding.

[0055] Path fitness evaluation

[0056] Calculate the fitness value, comprehensive path length L, obstacle avoidance cost C, smoothness S, the formula is:

[0057] ;

[0058] in, is the weight coefficient, and the path with high fitness value is preferred.

[0059] Quantum operations and population updates

[0060] Quantum revolving gate update: According to the fitness, the revolving gate operation is applied to the quantum chromosome to adjust the quantum bit phase. The formula is:

[0061] ;

[0062] in, is the rotation angle, which is dynamically adjusted according to the difference between the current solution and the optimal solution; and is the probability amplitude of the quantum bit.

[0063] A qubit is the basic unit of information in quantum computing. and They represent the probability amplitudes of the quantum bit being in different states, and they satisfy ,in represents the probability of a quantum bit being in a certain state, Represents the probability of being in another state.

[0064] Quantum crossover and mutation: perform crossover (exchange genes) and mutation (randomly change quantum bits) operations to generate new populations and avoid premature algorithm maturation.

[0065] Convergence judgment and optimal path output

[0066] Determine whether the population converges. If so, output the global optimal path, discretize it into a sequence of navigation points and load it; if not, continue iterating.

[0067] 3. Enter the follower path construction and obstacle avoidance phase. The follower navigation path is constructed based on the characteristics of the formation and the reference formation set in the early stage. When the follower starts, it quickly adjusts the initialization formation to match the overall layout of the formation. At the same time, a stable communication link with the navigator is established to receive the navigator's position, heading and other information in real time. During operation, the follower uses the improved artificial potential field method to perform local obstacle avoidance. By calculating the repulsion of the obstacle on itself, the resultant force and direction of the resultant force in the x / y direction are obtained. This solution optimizes the repulsive field function, adjusts its range and intensity curve, adds virtual obstacles, and solves the problems of local minima and path oscillation that are prone to occur in the traditional artificial potential field method, thereby ensuring that the follower can safely avoid obstacles. During the obstacle avoidance process, the follower synchronously adjusts the formation formation locally to maintain the integrity of the formation, and updates its own position information in real time to ensure that the entire formation is always in an orderly and collaborative state.

[0068] 1. Initial formation of followers

[0069] Based on the formation characteristics and reference formation, the initial navigation path of the followers is constructed.

[0070] Improved artificial potential field method for local obstacle avoidance

[0071] The artificial potential field method is a classic robot path planning and obstacle avoidance method. It constructs the robot's workspace as a virtual "potential field" and uses the combined force of the target's gravity and the obstacle's repulsion to guide the robot to move and avoid obstacles. The principle is similar to the movement of electric charges in an electric field. For the specific principle, see Figure 2 The target point attracts the unmanned boat through the gravitational potential field to generate gravitational force F2, and the obstacle repels the unmanned boat through the repulsive potential field to generate repulsive force F1. The unmanned boat uses the combined force F of gravity and repulsion to guide the movement and obstacle avoidance.

[0072] The commonly used expression of gravity is:

[0073] ;

[0074] in is the gravitational function.

[0075] The commonly used expression of repulsive force is:

[0076] ;

[0077] in is the repulsion function.

[0078] In the traditional artificial potential field method, in complex environments, the superposition of gravity and repulsion may form a local minimum point, causing the robot to be trapped in a certain position and unable to reach the target. For example, there may be an area between two obstacles that are close to each other, where the resultant force on the unmanned boat is zero, such as Figure 3 As shown, the obstacle repulsion F3 exerted on the unmanned boat is equal to the target point attraction F4, and the resultant force F5 is zero, so the unmanned boat cannot continue to avoid obstacles and move forward.

[0079] What's worse, in some cases, the unmanned boat may produce path oscillations under the action of gravity and repulsion, especially at the edge of obstacles or in areas where gravity and repulsion change greatly. This will lead to unstable movement, increase energy consumption and movement time.

[0080] Therefore, we improve some of the effects brought by the traditional artificial potential field method and optimize the repulsion function. Here we perform the following steps:

[0081] Force field calculation and path optimization: Update position in real time, calculate obstacle repulsion and target gravity. Improve potential field function, calculate the direction of resultant force, and adjust local path. Optimize path based on the direction of resultant force, so followers will move closer to the leader while avoiding obstacles, reducing formation disorder caused by obstacle avoidance, which leads to a large distance from the leader.

[0082] The optimized repulsion function is as follows:

[0083] ;

[0084] The repulsive force is obtained by optimizing the repulsive force function. :

[0085] ;

[0086] Among them, q represents the current position, is the location of the obstacle, is the repulsion coefficient, It is to improve the repulsive force range of the artificial potential field method. = Indicates the current distance q and The distance between.

[0087] Optimize the repulsion function and introduce the exponential saturation kernel , which greatly reduces the singularity problem of the original function where the repulsion becomes very large when the distance between the unmanned boat and the obstacle is small. When the unmanned boat is very close to the obstacle, the repulsion growth rate slows down and tends to a finite value, simulating the flexible contact of the spring buffer to avoid rigid collision shock (avoiding the damage of the rigid body such as the unmanned boat due to excessive impact force). At the same time, through the gradient continuous design, the mutation point of the original function at the boundary of the repulsion is eliminated, making the unmanned boat move more smoothly when approaching or leaving the obstacle, effectively solving the local extremely small traps and the path oscillation that may occur at the edge of the obstacle or in the area where the gravitational repulsion changes greatly, taking into account its safety and path continuity.

[0088] Although optimizing the potential field parameters can effectively reduce the probability of falling into the local optimum, it cannot completely eliminate it. Therefore, this embodiment adds another way to improve the artificial potential field method, adding virtual obstacles. The following is the formula for the force generated by the virtual obstacle:

[0089] ;

[0090] in, The repulsive force generated by the virtual obstacle; is the virtual repulsion coefficient; The vector pointing from the current position to the virtual obstacle position; for Length of mold Represented as the repulsion range of the virtual obstacle.

[0091] When it is detected that the unmanned boat is at a local minimum point, virtual obstacles can be added by evaluating the distribution of obstacles to help the unmanned boat escape from the local minimum point. In this case, by adding virtual obstacles, the resultant force acting on the ship will change, and the virtual obstacles will provide additional force for the unmanned boat to help it avoid the local minimum point. In addition, the existence of repulsive force can prevent the unmanned boat from entering the lowest point area again. Generally, the repulsive force generated by the virtual obstacle does not need to be very strong. Its main function is to break the deadlock of the ship in the local minimum and help it jump out of the current predicament without causing too much interference to the overall path planning, ensuring that the planning results can still meet the global optimal requirements. Demonstration details are as follows Figure 4 As shown, the obstacle repulsion F6 on the unmanned boat is equal to the target point attraction F7. At this time, a virtual obstacle is introduced, and the unmanned boat is subjected to the virtual obstacle repulsion F8, so that the final resultant force F9 of the unmanned boat is not equal to 0.

[0092] Adding virtual obstacles can help ships get rid of the local optimal dilemma, change the force field distribution, break the local balance to explore the global optimal path. The virtual force generated can increase the safe distance between the ship and the dangerous area, improve navigation safety, and provide more control parameters for path planning, making planning more flexible and more in line with diverse actual needs.

[0093] Dynamic maintenance of formation

[0094] Monitor the formation distance (too close or too far): If it is within the threshold, the leader moves at the speed and the follower makes local adjustments; if it is not within the threshold, correct the follower speed (the leader remains unchanged), and make local adjustments after returning to the threshold.

[0095] In order to prevent the formation from colliding within the formation due to the failure to adjust the speed in time during local obstacle avoidance or formation adjustment, this embodiment adds the same kind of repulsion force to the unmanned boat formation. The following is the formula for the same kind of repulsion force:

[0096] ;

[0097] in represents the same type repulsion coefficient; is represented as a smooth kernel function, where Expressed as a distance parameter; represents the layered safety distance; n represents the radial unit vector; Represents the heading weight coefficient; Indicates the angle between the headings of the unmanned boats; represents the tangent unit vector.

[0098] In the mutual repulsion formula of multiple unmanned boats, As the repulsion coefficient of the same type, it distinguishes the repulsion strength between the unmanned boat and the obstacles, making the repulsion between the same type of boats softer and conducive to formation. This smooth kernel function makes the repulsive force change smoothly at close distances through a specific expression to avoid oscillation caused by sudden changes. Reflecting the layered safety distance, based on the actual distance A safe distance set for similar boats To determine whether the repulsive force is effective and its magnitude. n is the radial unit vector, which specifies the direction of the traditional repulsive force; As the heading weight coefficient, combined with the heading angle between the two boats and the tangent unit vector , so that the tangential force is adjusted according to the heading, and collision avoidance is achieved in accordance with navigation rules. The basic principle diagram is as follows Figure 5 As shown in the figure, the internal repulsion of the formation refers to the repulsive force generated between the boats in the unmanned boat formation to avoid mutual collision. Figure 5The follower boat 01 and the navigator enter each other's repulsive potential field. There are repulsive forces F01 and F10 between the navigator and the follower boat 01, while the follower 02 does not enter the repulsive range, so no repulsive force is generated, so the distance and formation between them are maintained stable.

[0099] Fourth, the system has the ability to continuously detect dynamic obstacles. When a dynamic obstacle is detected, local path replanning is triggered immediately. At this time, the force situation is recalculated in combination with the improved artificial potential field method, and the follower's movement direction is accurately adjusted according to the new force analysis results, so that it can avoid obstacles in time. At the same time, the incremental optimization characteristics of the quantum genetic algorithm are used to update the global path to adapt to changes in the environment caused by the appearance of dynamic obstacles, ensuring the rationality and safety of the entire formation's navigation path.

[0100] 5. Process judgment and loop stage. The system continuously judges whether the target point has been reached. If it has been reached, the path planning and obstacle avoidance control process ends; if it has not been reached, the following operations are performed in a loop: update the environment and obtain the latest environmental information in real time, including obstacle position changes, sea conditions and other information; re-plan the path and optimize the leader and follower paths based on the updated environmental information; perform obstacle avoidance control to deal with possible obstacles; adjust the formation to maintain a reasonable formation. Through repeated cycles, the unmanned system formation at sea can be ensured to safely and efficiently drive to the target point in a complex environment.

[0101] Through the above-mentioned implementation methods, this patent realizes efficient path planning and safe obstacle avoidance of the leader-follower collaborative formation in complex marine environments, solves the problems of large computational complexity, easy to fall into local optimality, path oscillation and collision within the team in traditional algorithms, and improves the reliability and efficiency of collaborative operations of unmanned systems at sea.

[0102] The above embodiments are only exemplary embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art may make various modifications or equivalent substitutions to the present invention within the essence and protection scope of the present invention, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present invention.

Claims

1. A method for path planning and obstacle avoidance of an unmanned boat formation, characterized in that: The steps include: Step 1: rasterize the complex marine environment into a two-dimensional map, set the reference formation and the corresponding formation threshold, clarify the initial information of the unmanned boat, initialize the parameters of the quantum genetic algorithm and the gravitational and repulsive parameters of the improved artificial potential field method; Step 2: Use quantum genetic algorithm to decode quantum chromosomes into actual paths, weigh the fitness of each path, path length, resources consumed or deviations generated by avoiding obstacles, and smoothness, screen out the optimal path, and discretize the optimal path to obtain the global navigation path; Step 3: Construct the follower navigation path based on the formation characteristics and the reference formation set in step 1. When the follower starts, adjust the initial formation to match the overall formation layout. At the same time, establish a communication link with the leader. The follower uses the improved artificial potential field method to avoid local obstacles and adjust the local formation. Step 4: When a dynamic obstacle is detected, local path replanning is triggered, the force is recalculated using the improved potential field method, and step 2 is repeated to optimize the local path of the replanned section; Step 5, determine whether the target point has been reached, if reached, then end, if not reached, then loop through steps 1-4 until the target point is reached; The step 3 specifically includes: Step 3.1, based on the formation characteristics and the reference formation, construct the follower's initial navigation path; Step 3.2, update the position in real time, calculate the repulsion of obstacles and the gravitational force of the target, add virtual obstacles, optimize the repulsion function, calculate the direction of the resultant force, and adjust the local path; optimize the path based on the direction of the resultant force, and the follower moves closer to the leader while avoiding obstacles; Step 3.3, monitor the formation distance. If it is within the threshold, the leader moves at the speed and the followers make local adjustments. If it is not within the threshold, the followers' speed is corrected, the leader's speed remains unchanged, and local adjustments are made after returning to the threshold. In step 3.3, the same repulsion and mutual repulsion of the unmanned boat formation are introduced. The formula is: ; ; in, represents the same type repulsion coefficient; Expressed as a smoothing kernel function, Expressed as a distance parameter; Indicates the safety distance set for similar boats; represents the layered safety distance; n represents the radial unit vector; Represents the heading weight coefficient; Indicates the angle between the headings of the unmanned boats; represents the tangent unit vector.

2. The method for path planning and obstacle avoidance of an unmanned boat formation according to claim 1, characterized in that: The step 2 specifically includes: Step 2.1, generate the quantum initial population, and convert the chromosome into a path node sequence by decoding; Step 2.2, calculate the path fitness value F: ; in, is the weight coefficient; L is the comprehensive path length; C is the obstacle avoidance cost; S is the smoothness; Select the path with high fitness value; Step 2.3, according to the fitness, apply the rotating gate operation to the quantum chromosome to adjust the quantum bit phase; perform crossover and mutation operations to generate a new population; Step 2.4, determine whether the population has converged. If so, output the global optimal path, discretize it into a navigation point sequence and load it. If not, continue iterating.

3. The method for path planning and obstacle avoidance of an unmanned boat formation according to claim 1, characterized in that: Optimize the repulsion function in step 3.2 for: ； The repulsive force is obtained by optimizing the repulsive force function. : ； Among them, q represents the current position, is the location of the obstacle, is the repulsion coefficient, It is to improve the repulsive force range of the artificial potential field method. = Indicates the current distance q and The distance between.

4. The method for path planning and obstacle avoidance of an unmanned boat formation according to claim 1, characterized in that: The specific method of adding virtual obstacles in step 3.2 is as follows: when the unmanned boat is detected to be at a local minimum, the obstacle distribution is evaluated and a virtual obstacle is added. The repulsive force generated by the virtual obstacle The calculation formula is: ; in, is the virtual repulsion coefficient; The vector pointing from the current position to the virtual obstacle position; for Length of mold Represented as the repulsion range of the virtual obstacle.

Images