Water surface unmanned air cushion boat obstacle avoidance method and system based on self-adaptive dynamic window method

Abstract

The invention discloses a water surface unmanned air cushion boat obstacle avoidance method and system based on a self-adaptive dynamic window method, and belongs to the technical field of intelligent ship autonomous path planning. According to the method, trajectory evaluation functions under different obstacle states are constructed, and a self-adaptive adjustment strategy of an azimuth angle evaluation function is designed for a static obstacle in order to avoid an infeasible path when a plurality of obstacles exist between a current position and a target point; for a dynamic obstacle, a collision danger coefficient is introduced to replace an original distance evaluation function, and a more comprehensive trajectory evaluation function is designed. According to the invention, an azimuth angle evaluation function can be adaptively adjusted, so that an infeasible path in a multi-obstacle environment can be effectively avoided; the collision danger coefficient is added, the collision avoidance requirement of the dynamic obstacle is comprehensively considered, the adaptability and robustness of the algorithm to the complex environment are improved, and meanwhile the navigation safety of the unmanned air cushion boat is guaranteed.

Description

Technical Field

[0001] The present invention belongs to the technical field of autonomous path planning for intelligent ships, and particularly relates to an obstacle avoidance method and system for a surface unmanned air-cushion vehicle based on the adaptive dynamic window method. Background Art

[0002] With the development of unmanned driving technology, surface unmanned air-cushion vehicles have been widely used in fields such as ocean exploration, environmental monitoring, and maritime rescue. However, due to the complexity of the water surface environment, especially in an environment with multiple static and dynamic obstacles, ensuring the safe obstacle avoidance of the unmanned air-cushion vehicle and smoothly reaching the target has become a technical difficulty. Existing obstacle avoidance algorithms mostly adopt distance-based evaluation criteria, but they often have limitations in complex environments, may not be able to provide the best navigation path, and may even fall into local optimal solutions.

[0003] To address these problems, the Dynamic Window Algorithm (DWA) has been widely applied to the motion planning of unmanned systems. This method selects the optimal path by evaluating feasible trajectories at different speeds and accelerations. However, the states of different obstacles in the water surface environment pose higher requirements for path planning. The traditional DWA algorithm is prone to falling into locally infeasible paths in the case of multiple obstacles and is not comprehensive enough in dealing with dynamic obstacles. Therefore, how to enhance the adaptability of the algorithm to the complex water surface environment has become the key to improving the obstacle avoidance ability of unmanned air-cushion vehicles.

[0004] Therefore, the present invention proposes an obstacle avoidance method for a surface unmanned air-cushion vehicle based on the adaptive dynamic window method. By introducing an adaptive adjustment strategy for the azimuth evaluation function, it is optimized for static obstacles to avoid local stagnation problems caused by multiple obstacles. For dynamic obstacles, a collision risk coefficient is introduced to replace the traditional distance evaluation function, comprehensively considering the safety and collision avoidance requirements of the unmanned air-cushion vehicle. This method effectively improves the obstacle avoidance ability of the surface unmanned air-cushion vehicle in complex environments, ensures navigation safety, and enhances the robustness and adaptability of the algorithm. Summary of the Invention

[0005] The purpose of the present invention is to provide an obstacle avoidance method and system for a surface unmanned air-cushion vehicle based on the adaptive dynamic window method, which is used to help the unmanned air-cushion vehicle navigate safely in complex environments, can enhance the adaptability of the autonomous obstacle avoidance algorithm of the unmanned air-cushion vehicle under different sea conditions, overcome the defect that the traditional dynamic window method is prone to falling into local optimality, and effectively improve the robustness of the algorithm.

[0006] The purpose of the present invention is achieved through the following technical solutions:

[0007] An obstacle avoidance method for a surface unmanned air-cushion vehicle based on the adaptive dynamic window method, the specific steps are as follows:

[0008] Step 1: Establish a model of an unmanned surface air-cushion vehicle;

[0009] Step 2: Generate a set of velocities according to the current speed and acceleration limits of the model in Step 1, and generate a dynamic window through the intersection of the velocity sets;

[0010] Step 3: Predict the motion trajectory of the unmanned air-cushion vehicle according to each set of velocity sets in the velocity sampling;

[0011] Step 4: Evaluate all candidate motion trajectories obtained in Step 3 through an evaluation function;

[0012] Step 5: Select the trajectory with the highest score in Step 4, and convert it into speed and angular velocity commands to make the unmanned air-cushion vehicle execute the corresponding movement;

[0013] Step 6: After executing the trajectory selected in Step 5, repeat Steps 2 to 5, update the state of the unmanned air-cushion vehicle and the information of surrounding obstacles, and re-plan the path until it reaches within the target point threshold range.

[0014] Furthermore, the velocity set in Step 2 is:

[0015] The limit of the maximum and minimum speeds of the unmanned air-cushion vehicle itself: V s ={(u,r)|u∈[u min ,u max ,r∈[r min ,r max}

[0016] where Vs is the set of all vector velocities that the unmanned air-cushion vehicle can reach; u min and u max are the minimum linear velocity and the maximum linear velocity; r min and r max are the minimum angular velocity and the maximum angular velocity;

[0017] Influence of the unmanned air-cushion vehicle affected by the motor performance:

[0018] where u c and r c are the velocity and angular velocity at the current moment; and are the maximum deceleration of the velocity and the maximum acceleration of the velocity; and are the maximum deceleration of the angular velocity and the maximum acceleration of the angular velocity;

[0019] Influence of the path obstacle avoidance requirement:

[0020] Among them, dist(u,r) is the distance closest to the obstacle on the trajectory corresponding to (u,r);

[0021] The dynamic window is the intersection of the above three velocity sets, that is, V r = V s ∩V d ∩V a .

[0022] Furthermore, the predicted motion trajectory of the unmanned hovercraft in step 3 is:

[0023]

[0024] Among them, x and y are positions, Ψ is the heading, u is the forward speed, v is the cross drift speed, and r is the yaw angular velocity.

[0025] Furthermore, the evaluation function in step 4 is:

[0026] For static obstacles, the evaluation function is:

[0027] G(u,r) = σ[α 1 ·Head(u,r) + β 1 ·Dist(u,r) + γ 1 ·Vel(u,r)], (u,r) ∈ V r

[0028] For the encounter of two hovercrafts, the evaluation function is:

[0029]

[0030] Among them, α, β, and γ are the coefficients of each evaluation sub-function, and σ is the smoothing factor; if CRI ≥ 0.5, it means there is a collision risk, and it is more accurate to use the collision risk assessment function CRI(u,r) to replace the obstacle distance evaluation function Dist(u,r); if there is no collision risk, at this time, it is expected to reach the target point faster, so only the azimuth evaluation function Head(u,r) and the speed evaluation function Vel(u,r) are retained.

[0031] Furthermore, the azimuth evaluation function:

[0032]

[0033] Among them, θ is the pose angle of the line connecting the current position and the target point in the inertial coordinate system; θ s is the direction with the fewest obstacles within ±50 degrees of the current heading; n(obstacles) is the number of obstacles within 15m of the current position;

[0034] The obstacle distance evaluation function:

[0035]

[0036] The speed evaluation function:

[0037] Vel(u,r) = u c

[0038] The collision risk evaluation function:

[0039] CRI(u,r) = 1 - CRI;

[0040] To balance the dimensions and value ranges of different evaluation indicators and ensure the reasonable contribution of each indicator to the total score, the sub-evaluation function is normalized:

[0041]

[0042] where σ is the smoothing factor, n is the total number of predicted trajectories within a simulation period, and i is the currently evaluated feasible predicted trajectory.

[0043] Furthermore, in the collision risk evaluation function, CRI is used to describe the degree of collision risk. The higher its value, the greater the degree of collision risk, and its value range is [0,1]; considering the influence of five factors: the distance S between the two boats, the azimuth angle θ T , the speed ratio K, the closest distance of approach DCPA, and the time to closest point of approach TCPA, the fuzzy mathematics method is used to calculate CRI:

[0044]

[0045] where γ is the membership function of each factor and ω is the weight factor.

[0046] Furthermore, the membership function of DCPA is:

[0047]

[0048] where d 1 is the minimum safe distance of approach, and d 2 is the minimum safe passing distance;

[0049] The membership function of TCPA is:

[0050]

[0051] where t 1 represents the time required for the unmanned hovercraft to travel from the latest safe avoidance position to the closest point of approach, and t 2 represents the time required for the speed ratio to travel from the current position to the closest point of approach;

[0052] The membership function of the distance S between the two boats is:

[0053]

[0054] Among them, D 1 , D 2 is the nearest and farthest avoidance distance;

[0055] The membership function of the azimuth angle θ T is:

[0056]

[0057] The membership function of the ship speed ratio K is:

[0058]

[0059] A computer device / system, including a memory, a processor, and a computer program stored on the memory, where the processor executes the computer program to implement the steps of an obstacle avoidance method for an unmanned surface air-cushion vehicle based on the adaptive dynamic window method.

[0060] A computer-readable storage medium, on which a computer program / instructions are stored, and when the computer program / instructions are executed by a processor, the steps of an obstacle avoidance method for an unmanned surface air-cushion vehicle based on the adaptive dynamic window method are implemented.

[0061] The beneficial effects of the present invention are as follows:

[0062] The present invention can effectively avoid getting into an infeasible path in a multi-obstacle environment by adaptively adjusting the azimuth evaluation function; adding a collision risk coefficient, comprehensively considering the collision avoidance requirements of dynamic obstacles, improving the adaptability and robustness of the algorithm to complex environments, and at the same time ensuring the safety of the unmanned air-cushion vehicle navigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a coordinate reference system diagram of the ship three-degree-of-freedom motion model in the present invention;

[0064] Figure 2 It is a flowchart of an obstacle avoidance method for an unmanned surface air-cushion vehicle based on the adaptive dynamic window method;

[0065] Figure 3 It is a path effect diagram of autonomous obstacle avoidance in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0066] The following further describes the present invention with reference to the drawings.

[0067] Embodiment 1:

[0068] According to Figures 1 to 2, a method for obstacle avoidance of an unmanned surface air-cushion vehicle based on the adaptive dynamic window method, the specific steps are as follows:

[0069] Step 1, establish a mathematical model of the unmanned air-cushion vehicle;

[0070] Considering only the motion in the horizontal plane, the motion of the unmanned air-cushion vehicle can be represented by a three-degree-of-freedom ship model, including forward motion, lateral drift, and yaw. To accurately describe the motion of the unmanned air-cushion vehicle, two coordinate systems are established. One is a coordinate system (X E , O E , Y E ) fixed to the earth's sea surface, also known as the inertial coordinate system, and its origin is usually the center-of-gravity position of the ship at time t = 0. The other is a coordinate system (X B , O B , Y B ) fixed to the ship, also known as the appended-body coordinate system, and the origin of the coordinate system is usually selected at the center of gravity of the ship. In the appended-body coordinate system, the forward speed is u, the lateral drift speed is v, and the yaw angular velocity is r. The velocity vector is v = [u vr] T . In the inertial coordinate system, the course is Ψ, and the position vector is η = [x yψ] T . The transformation between the two coordinates can be carried out through the transformation matrix T(ψ):

[0071]

[0072] Step 2, generate a dynamic window;

[0073] To better predict the trajectory, it is necessary to consider the kinematic and environmental constraints on the unmanned air-cushion vehicle during navigation. According to the current speed and acceleration limits, a set of speeds is generated, including the linear speed range and the angular speed range.

[0074] 1) Limits of its own maximum and minimum speeds:

[0075] V s = {(u, r)|u ∈ [u min , u max , r ∈ [r min , r max}

[0076] Among them, Vs is the set of all vector speeds that the unmanned air-cushion vehicle can reach; u min and u max are the minimum and maximum linear speeds; r min and r max are the minimum and maximum angular speeds.

[0077] 2) Influence of motor performance:

[0078] Due to the limited torque of the motor, there is an acceleration limit. Only the speed that can be reached within a certain time of maximum acceleration or maximum deceleration will be retained.

[0079]

[0080] Among them, u c and r c are the speed and angular velocity at the current moment; and are the maximum deceleration of speed and the maximum acceleration of speed; and are the maximum deceleration of angular velocity and the maximum acceleration of angular velocity.

[0081] 3) Limitation of obstacle avoidance effect:

[0082] When the unmanned hovercraft encounters an obstacle, a braking distance needs to be reserved for deceleration. Therefore, the speed at which it can stop safely needs to be considered, and they should meet the following conditions:

[0083]

[0084] Among them, dist(u, r) is the closest distance to the obstacle on the trajectory corresponding to (u, r).

[0085] The dynamic window is the intersection of the above three speed sets, that is, V r = V s ∩V d ∩V a .

[0086] Step 3: Predict the trajectory;

[0087] According to each speed combination in the speed sampling space, simulate the movement trajectory of the unmanned hovercraft in the future for a period of time. Through the predicted trajectory, the algorithm can judge the position that the unmanned hovercraft will reach under different speed combinations. According to the motion model and the sampled speed, the trajectory prediction is as follows:

[0088]

[0089] Among them, x and y are positions, Ψ is the course, u is the forward speed, v is the cross drift speed, and r is the yaw angular velocity.

[0090] Step 4: Evaluate the candidate trajectories;

[0091] For each predicted trajectory, it needs to be evaluated through an evaluation function to determine its quality. For static obstacles, the evaluation function is:

[0092] G(u, r) = σ[α 1 ·Head(u, r)+β1 ·Dist(u, r) + γ 1 ·Vel(u, r)], (u, r) ∈ V r

[0093] For the encounter of two boats, the evaluation function is as follows:

[0094]

[0095] Among them, α, β, and γ are the coefficients of each evaluation sub-function, and σ is the smoothing factor. If CRI ≥ 0.5, it indicates a collision risk, and it is more accurate to use the collision risk assessment function CRI(u, r) instead of the obstacle distance evaluation function Dist(u, r); if there is no collision risk, it is expected to reach the target point faster at this time, so only the azimuth angle Head(u, r) and the speed evaluation function Vel(u, r) are retained.

[0096] 1) Speed evaluation function:

[0097] The speed evaluation is used to measure the efficiency of traveling under the speed combination. A higher linear speed will get a higher score:

[0098] Vel(u, r) = u c

[0099] 2) Obstacle distance evaluation function:

[0100] The obstacle distance evaluation is used to measure the safety of the predicted trajectory. The farther the trajectory is from the obstacle, the higher the score:

[0101]

[0102] 3) Azimuth angle evaluation function:

[0103] The evaluation of the direction angle was originally used to evaluate the angular difference between the predicted trajectory and the target point. However, when there are too many obstacles between the current position and the target point, it will fall into an infeasible path and spin in place. Therefore, considering the situation of nearby obstacles, the direction that can avoid these obstacles can be selected as the orientation.

[0104]

[0105] where θ is the pose angle of the line connecting the current position and the target point in the inertial coordinate system; θ s is the direction with the fewest obstacles within ±50 degrees of the current heading; n(obstacles) is the number of obstacles within 15 m of the current position.

[0106] 4) Collision risk evaluation function:

[0107] CRI(u, r) = 1 - CRI

[0108] The CRI is used to describe the collision risk degree. The higher its value, the greater the collision risk. Its value range is [0, 1]. When ships encounter each other, various motion parameters need to be calculated to determine whether there is a collision risk.

[0109]

[0110] Among them, v o , v t are the speeds of the unmanned hovercraft and the approaching ship; are the headings of the unmanned hovercraft and the approaching ship; [x o , y o is the position of the unmanned hovercraft, and [x t , y t is the position of the approaching ship; v r is the relative speed of the two ships. The relative bearing of the two ships is:

[0111]

[0112] The relative heading of the two ships is:

[0113]

[0114] The closest distance of approach DCPA and the time to closest point of approach TCPA are:

[0115]

[0116] Considering the influence of the distance S between the two ships, the azimuth angle θ T , the ship speed ratio K, DCPA, and TCPA, a total of five factors, the CRI is calculated using the fuzzy mathematics method:

[0117]

[0118] Among them, γ represents the membership function of each factor, and ω represents the weight factor. The membership function of DCPA is:

[0119]

[0120] Among them, d 1 represents the minimum safe distance of approach, and d 2 represents the minimum safe passing distance. The membership function of TCPA is:

[0121]

[0122] Among them, t 1 represents the time required for the unmanned hovercraft to travel from the latest safe avoidance position to the closest point of approach, and t 2It represents the time required for the unmanned hovercraft to travel from its current position to the closest point of encounter. The membership function of the distance S between the two vessels is as follows:

[0123]

[0124] where D 1 , D 2 represent the closest and farthest avoidance distances. The membership function of the azimuth angle θ T is as follows:

[0125]

[0126] The membership function of the ship speed ratio K is as follows:

[0127]

[0128] In order to balance the dimensions and value ranges of different evaluation indicators and ensure the reasonable contribution of each indicator to the total score, it is necessary to normalize the sub-evaluation function:

[0129]

[0130] where n is the total number of predicted trajectories within a simulation period, and i is the currently evaluated feasible predicted trajectory.

[0131] Step 5: Execute the movement strategy;

[0132] Compare the total evaluation functions of the candidate trajectories, select the one with the highest score as the optimal trajectory, and convert it into speed and angular velocity commands to make the unmanned hovercraft execute the corresponding movement.

[0133] Step 6: Repeat the iteration.

[0134] After executing the movement, repeat the above steps, continuously update the state of the unmanned hovercraft and the information of surrounding obstacles, and re-plan the path until it reaches within the target point threshold range.

[0135] Embodiment 2:

[0136] An obstacle avoidance system for a surface unmanned hovercraft based on the adaptive dynamic window method of the present invention includes: sensors and an unmanned hovercraft; the sensors obtain the motion parameters of the real-time navigation of the unmanned hovercraft, detect the obstacle situation, and transmit it to the planning module. The path planning algorithm adopts corresponding evaluation functions for different obstacles and selects an optimal path. By dynamically adjusting the speed and steering of the unmanned hovercraft, the movement efficiency is maximized, enabling the unmanned hovercraft to reach the target point as soon as possible while ensuring safety.

[0137] The present invention relates to a method for an obstacle avoidance system of an unmanned surface air-cushion vehicle based on an adaptive dynamic window method, which adopts an improved DWA path planning algorithm; the improved DWA path planning algorithm includes: a dynamic window, trajectory prediction, and an evaluation function.

[0138] The dynamic window is used to describe the range of travelable speed and angular velocity. According to the current state and dynamic model of the unmanned air-cushion vehicle, the speed range is calculated, and the dynamic window is composed of four parameters: the minimum speed, the maximum speed, the minimum angular velocity, and the maximum angular velocity.

[0139] The trajectory prediction is used to describe the trajectory of the unmanned air-cushion vehicle after executing the current speed and angular velocity for a period of time. Based on the dynamic model, the dynamic window is traversed to generate candidate trajectories for each combination of speed and angular velocity.

[0140] The evaluation function is used to describe the score of each predicted trajectory. The scoring criteria include speed evaluation, navigation evaluation, obstacle distance evaluation, and collision risk evaluation. Each criterion assigns a score to each trajectory, and finally these scores are combined to select the best trajectory.

[0141] Embodiment 3:

[0142] In the embodiment, the weight coefficients in the evaluation function can be selected with reference to the following table:

[0143] Weight coefficient 1 2 3 α 0.05 0.1 0.05 β 0.3 \ 0.5 γ 0.05 0.2 0.1

[0144] The numbers in the first row of the table represent the subscripts of each coefficient.

[0145] According to Figure 3 As shown, it is the implementation of a method for an obstacle avoidance of an unmanned surface air-cushion vehicle based on an adaptive dynamic window method and a common dynamic window method in an obstacle environment. The yellow dots in the figure are the starting points, the green dots are the path end points, the red figures represent obstacles, the blue lines represent the obstacle avoidance paths under the method of the present invention, and the black lines are the obstacle avoidance paths under the common dynamic window method.

[0146] In the embodiments of the present invention, when implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0147] The foregoing are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for avoiding obstacles on a water surface unmanned hovercraft based on an adaptive dynamic window method, characterized in that: The specific steps are as follows: Step 1: Build a model of the unmanned hovercraft on the water surface; Step 2: Generate a velocity set based on the current velocity and acceleration limit of the model in step 1, and generate a dynamic window through the intersection of the velocity sets; Step 3: Predict the motion trajectory of the unmanned hovercraft based on each set of speeds in the speed sampling; Step 4: Evaluate all candidate motion trajectories obtained in step 3 using the evaluation function; Step 5: Select the trajectory with the highest score in step 4 and convert it into velocity and angular velocity commands to make the unmanned hovercraft perform corresponding movements; Step 6: After executing the trajectory selected in step 5, repeat steps 2 to 5 to update the status of the unmanned hovercraft and the surrounding obstacle information and replan the path until it reaches the target point threshold range.

2. The method for avoiding obstacles on a water surface unmanned hovercraft based on an adaptive dynamic window method according to claim 1, characterized in that: The speed set in step 2 is: The maximum and minimum speed limits of the unmanned hovercraft: V s ={(u,r)|u∈[u min ,u max ],r∈[r min ,r max ]} Where Vs is the set of all vector velocities that the unmanned hovercraft can reach; u min and u max is the minimum and maximum linear speed; r min and r max are the minimum and maximum angular velocities; Unmanned hovercraft is affected by motor performance: Among them, u c and r c is the velocity and angular velocity at the current moment; and is the maximum deceleration of speed and the maximum acceleration of speed; and is the maximum deceleration of angular velocity and the maximum acceleration of angular velocity; Impact of path obstacle avoidance requirements: Among them, dist(u,r) is the distance to the obstacle on the trajectory corresponding to (u,r); The dynamic window is the intersection of the above three speed sets, namely V r =V s ∩V d ∩V a .

3. The method for avoiding obstacles on a water surface unmanned air-cushion craft based on an adaptive dynamic window method according to claim 1, characterized in that: The motion trajectory of the unmanned hovercraft predicted in step 3 is: Where x and y are positions, Ψ is heading, u is forward speed, v is roll speed, and r is yaw angular velocity.

4. The method for avoiding obstacles on a water surface unmanned air-cushion craft based on an adaptive dynamic window method according to claim 1, characterized in that: The evaluation function in step 4 is: For static obstacles, the evaluation function is: G(u,r)=σ[α1·Head(u,r)+β1·Dist(u,r)+γ1·Vel(u,r)],(u,r)∈V r For the encounter between two boats, the evaluation function is: Among them, α, β, γ are the coefficients of each evaluation sub-function, and σ is the smoothing factor; if CRI ≥ 0.5, it means there is a collision risk, and it is more accurate to use the collision risk assessment function CRI (u, r) instead of the obstacle distance evaluation function Dist (u, r); if there is no collision risk, it is expected to reach the target point faster, so only the azimuth evaluation function Head (u, r) and the speed evaluation function Vel (u, r) are retained.

5. The method for avoiding obstacles on a water surface unmanned air-cushion craft based on an adaptive dynamic window method according to claim 4, characterized in that: The azimuth evaluation function: Among them, θ is the pose angle of the line connecting the current position and the target point in the inertial coordinate system; θ s is the direction with the fewest obstacles within ±50 degrees of the current heading; n(obstacles) is the number of obstacles within 15m of the current position; The obstacle distance evaluation function: The speed evaluation function: Vel(u,r)=u c The collision risk evaluation function: CRI(u,r)=1-CRI; In order to balance the dimensions and value ranges of different evaluation indicators and ensure that each indicator contributes reasonably to the total score, the sub-evaluation function is normalized: Where σ is the smoothing factor, n is the total number of forecast trajectories within a simulation period, and i is the feasible forecast trajectory currently evaluated.

6. The method for avoiding obstacles on a water surface unmanned air-cushion craft based on an adaptive dynamic window method according to claim 5, characterized in that: In the collision risk evaluation function, CRI is used to describe the collision risk. The higher the value, the greater the collision risk. The value range is [0,1]. Considering the distance S between the two ships, the azimuth θ T , ship speed ratio K, shortest encounter distance DCPA and shortest encounter time TCPA, and fuzzy mathematics method is used to calculate CRI: Among them, γ is the membership function of each factor, and ω is the weight factor.

7. The method for avoiding obstacles on a water surface unmanned air-cushion craft based on an adaptive dynamic window method according to claim 6, characterized in that: The membership function of DCPA is: Among them, d1 is the minimum safe encounter distance, and d2 is the minimum safe passing distance; The membership function of TCPA is: Among them, t1 represents the time required for the unmanned hovercraft to travel from the latest safe avoidance position to the nearest encounter point, and t2 represents the time required for the unmanned hovercraft to travel from the current position to the nearest encounter point; The membership function of the distance S between the two boats is: Among them, D1 and D2 are the closest and farthest avoidance distances; Azimuth θ T The membership function of is: The membership function of the ship speed ratio K is:

8. A computer device / equipment / system comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images