Unmanned ship obstacle avoidance control and wireless transmission method and system based on data fusion

Abstract

The invention discloses an unmanned ship obstacle avoidance control and wireless transmission method and system based on data fusion, and belongs to the field of unmanned ship navigation obstacle avoidance. Comprising the following steps: S1, constructing an unmanned ship initialization dynamic model, S2, constructing a path navigation feature vector, S3, arranging detection sensors around the unmanned ship in a 360-degree full-coverage mode, and arranging the detection sensors in a non-uniform distribution mode by adopting a dimension reduction mode in the preparation process; and S4, making a collision avoidance decision, and making the collision avoidance decision which meets the target of the unmanned ship and is dynamically limited by fusing the obtained various information of the obstacles and combining the condition of the unmanned ship. According to the technical scheme provided by the invention, the problem that unmanned ship navigation obstacle avoidance cannot be accurately and efficiently carried out in the prior art is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned ship navigation and obstacle avoidance, and in particular to an unmanned ship obstacle avoidance control and wireless transmission method and system based on data fusion. Background Art

[0002] Autonomous surface unmanned ships need to consider many factors during navigation to ensure safe and efficient operation. In the marine environment, unmanned ships face various obstacles and need to comply with the International Regulations for Preventing Collisions at Sea. Traditional collision avoidance methods may have certain limitations in complex dynamic environments and may not be able to well consider the properties of obstacles and the actions required by the International Regulations for Preventing Collisions at Sea. It is difficult to achieve the unmanned ship's own goals and adapt to its dynamic limited collision avoidance trajectory. Summary of the invention

[0003] In view of the defects of the related technology, the present invention provides an unmanned ship obstacle avoidance control and wireless transmission method and system based on data fusion, aiming to solve the problem that the existing technology cannot efficiently and accurately achieve unmanned ship collision avoidance.

[0004] The technical solution is as follows:

[0005] An unmanned ship obstacle avoidance control and wireless transmission method based on data fusion includes the following steps:

[0006] Step S1: constructing an initialization dynamic model of the unmanned ship, by defining a coordinate frame, state space constraints, and ship driving environment constraints to construct the initialization dynamic model of the unmanned ship;

[0007] Step S2: construct a path navigation feature vector, define the representation of the parameterized path, and characterize the path navigation feature vector by setting path reference points, Euclidean distances, forward path reference points, path deviations, heading errors, and path angles that are continuously updated based on the current position of the unmanned ship on the path to be navigated;

[0008] Step S3: Setting a data acquisition and processing module, equipping the unmanned ship with detection sensors in a 360-degree full coverage manner, and using a dimension reduction method during the equipping process to set the detection sensors to a non-uniform distribution;

[0009] Step S4: Make a collision avoidance decision. By integrating various information of obstacles acquired and combining the situation of the unmanned ship itself, a collision avoidance decision that meets the unmanned ship's own goals and is dynamically restricted is made;

[0010] Step S5: Through the wireless data transmission module including the DTU unit and the Beidou client, the obstacle avoidance information and control instructions required during the driving of the unmanned ship are wirelessly transmitted to other unmanned ships and the collaborative control center.

[0011] An unmanned ship obstacle avoidance control and wireless transmission system based on data fusion, the system comprising:

[0012] Unmanned ship obstacle avoidance control device and collaborative control center;

[0013] The collaborative control center is set up on the shore, and is used to establish wireless communication connection with the unmanned ship, send control instructions, and wirelessly keep the unmanned ship online in real time; it includes a wireless communication module, which has a DTU unit and a Beidou client, and communicates with the unmanned ship obstacle avoidance control device through radio or satellite signals via the 5G network or Beidou satellite. During the communication process, when the 5G network is interrupted or fails within a certain period of time, it switches to Beidou satellite communication;

[0014] The unmanned boat obstacle avoidance control device includes:

[0015] The dynamic model construction module is used to construct the initialization dynamic model of the unmanned ship. The initialization dynamic model of the unmanned ship is constructed by defining the coordinate frame, state space constraints, and ship driving environment constraints;

[0016] The path navigation feature vector construction module defines the representation of the parameterized path, and characterizes the path navigation feature vector by setting the path reference points, Euclidean distance, forward path reference points, path deviation, heading error and path angle that are continuously updated based on the current position of the unmanned ship on the path to be navigated;

[0017] The data acquisition and processing module equips the unmanned ship with detection sensors in a 360-degree full coverage manner. During the installation process, the detection sensors are arranged in a non-uniform distribution by using a dimension reduction method.

[0018] The collision avoidance decision-making module integrates various obstacle information obtained and combines the unmanned ship's own situation to make collision avoidance decisions that meet the unmanned ship's own goals and are dynamically restricted;

[0019] The wireless data transmission module, including the DTU unit and Beidou client, is used for wireless communication with other unmanned ships and the collaborative control center.

[0020] The present invention has the following beneficial effects:

[0021] The present invention provides an unmanned ship obstacle avoidance control and wireless transmission method and system based on data fusion. Through a unique arrangement of unmanned ship sensors, the present invention solves the problem of high-dimensional complex network data, slow control response and sluggish control in the control process of the unmanned ship. The construction of path navigation feature vectors optimizes the channel optimization problem caused by the minimum obstacle distance and the maximum obstacle distance. Accurate and efficient obstacle avoidance is achieved through collision avoidance control decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of an unmanned ship obstacle avoidance control method based on data fusion provided by an embodiment of the present invention;

[0023] Figure 2 is a flow chart for determining a dynamics model of an unmanned ship provided by an embodiment of the present invention;

[0024] Figure 3 is a flow chart of constructing a path navigation feature vector provided by an embodiment of the present invention;

[0025] Figure 4 It is a flow chart of the configuration of the acquisition module provided by an embodiment of the present invention;

[0026] Figure 5 is a collision avoidance decision-making flow chart provided by an embodiment of the present invention;

[0027] Figure 6 It is a structural diagram of an unmanned ship obstacle avoidance control and wireless transmission system based on data fusion provided by an embodiment of the present invention;

[0028] Figure 7 It is a structural diagram of an unmanned ship obstacle avoidance control device based on data fusion provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0030] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0031] See also Figure 1 The present invention provides an unmanned ship obstacle avoidance control and wireless transmission method based on data fusion, comprising the following steps:

[0032] Step S1: constructing an initialization dynamic model of the unmanned ship, by defining a coordinate frame, state space constraints, and ship driving environment constraints to construct the initialization dynamic model of the unmanned ship;

[0033] Step S2: construct a path navigation feature vector, define the representation of the parameterized path, and characterize the path navigation feature vector by setting path reference points, Euclidean distances, forward path reference points, path deviations, heading errors, and path angles that are continuously updated based on the current position of the unmanned ship on the path to be navigated;

[0034] Step S3: Setting a data acquisition and processing module, equipping the unmanned ship with detection sensors in a 360-degree full coverage manner, and using a dimension reduction method during the equipping process to set the detection sensors to a non-uniform distribution;

[0035] Step S4: Make a collision avoidance decision. By integrating various information of obstacles acquired and combining the situation of the unmanned ship itself, a collision avoidance decision that meets the unmanned ship's own goals and is dynamically restricted is made;

[0036] Step S5: Through the wireless data transmission module including the DTU unit and the Beidou client, the obstacle avoidance information and control instructions required during the driving of the unmanned ship are wirelessly transmitted to other unmanned ships and the collaborative control center.

[0037] See also Figure 2 , Step S1: Construct the unmanned ship initialization dynamics model, and construct the unmanned ship initialization dynamics model by defining the coordinate frame, state space constraints, and ship driving environment constraints, specifically:

[0038] Step S11: Define that the unmanned ship is always located on the surface of the earth without floating up and down, pitching and rolling. The coordinate frame of the environment where the unmanned ship is located is n=(x n ,y n ,z n ), which forms a tangent plane with the Earth's surface, where x n The axis points to the north, y n The axis points east, z n The axis points to the center of the Earth;

[0039] Step S12: define the state vector of the unmanned ship η = [x n ,y n ,ψ] T , where x n and nare the north and east positions in the coordinate frame of the environment, respectively; ψ is the heading angle, i.e. the angle between the longitudinal axis of the unmanned ship and the north axis; the translation and angular velocity vector v = [u, h, r] T It is composed of foreground velocity u, lateral velocity h and yaw rate r;

[0040] Step S13: The unmanned boat is equipped with a rudder and a propeller at the tail, and can move freely along any trajectory in the state space. The control vector for controlling the unmanned boat is f = [T u ,T r ], where T u Represents the torque input in the forward direction, T r Represents the torque input in the yaw direction, then the initialization dynamic model of the unmanned ship is expressed as:

[0041]

[0042] Where M, C(v), and D(v) are all 3×3 state space vectors, that is, M is the mass matrix of the unmanned ship, covering the rigid body effect and the added mass effect; C(v) is the centripetal force parameter matrix; D(v) is the damping matrix; B is the actuator configuration matrix, which belongs to the 3×2 state space vector; R z,ψ (η) represents the direction along z n The axis rotation ψ in radians.

[0043] See also Figure 3 ,Step S2: Construct a path navigation feature vector, define the representation of the parameterized path, and characterize the path navigation feature vector by setting the path reference point, Euclidean distance, forward path reference point, path deviation, heading error and path angle that are continuously updated based on the current position of the unmanned ship on the path to be navigated, specifically:

[0044] Step S21: define the parameterized path as p d (ω)=[x d (ω),y d (ω)] T , where ω is the path variable, p d (ω) represents the path reference point, which is set to be the point on the right side of the unmanned ship at the Euclidean distance from the unmanned ship. d (ω),y d (ω) represents the coordinates of the path reference point. Correspondingly, the forward path reference point needs to be set It is the path reference point p d (ω) Go forward Δ along the path LA Point of distance;

[0045] Step S22: Finding the path reference point p d(ω), it is necessary to calculate the corresponding value of the path variable ω at each time step

[0046]

[0047] Define the path deviation between the expected path and the current trajectory as ε, and define the heading error required for the unmanned ship to sail directly from the current position to the forward path reference point have:

[0048]

[0049] Further define the path angle Taking the inverse of the first-order path as a reference, the smoothness of the unmanned ship trajectory when it relies on the forward path reference point is:

[0050]

[0051] Step S23: By converting the path variable ω calculated in the above step into the corresponding value of each time step Path deviation ε, heading error Path Angle Together they form the path navigation feature vector s (t) n .

[0052] See also Figure 4 , Step S3: Set up a data acquisition and processing module, equip the unmanned ship with detection sensors in a 360-degree full coverage manner, and use a dimension reduction method during the equipment process to set the detection sensors to a non-uniform distribution, specifically:

[0053] Step S31: Equip a distance measuring sensor as a detection sensor, the number of distance measuring sensors is N, and the maximum detection range is S r , equipped around the unmanned ship with 360-degree full coverage;

[0054] In an embodiment provided by the present invention, the ranging sensor may be a laser radar, a sonar, a depth camera, etc.;

[0055] Although the area behind the unmanned boat is not very important for path planning and obstacle avoidance, and there is no need to consider obstacles behind it in pure static terrain navigation, in order to consider a variety of scenarios, such as the possibility that a moving obstacle approaches from the rear, and the unmanned boat determines the distance to the rear obstacle when overtaking the obstacle and changing the course, it is necessary to equip the range-finding sensors around the unmanned boat in a 360-degree full coverage manner;

[0056] During the navigation and obstacle avoidance experiment, when fusing the path information feature vector with the output data of the ranging sensor, it was found that the control information input to the unmanned boat was large, which generated high-dimensional complex network data, resulting in control sluggishness and slow response. Even if the number of ranging sensors is reduced, the same problem still exists. If the number of ranging sensors is too small, the accuracy of obstacle detection is reduced. Therefore, the data of the ranging sensor needs to be processed as follows;

[0057] Step S32: Divide the 360-degree full coverage ranging sensors into D groups, each group generates a scalar detection value, so that it is included in the final observation vector;

[0058] Since we want to minimize the dimension of the data input to the unmanned ship, evenly dividing the distribution of the ranging sensors into groups of equal size cannot meet the optimal solution. The obstacles in front of the unmanned ship are more critical than those behind the ship in the actual navigation and obstacle avoidance process, and require higher perception accuracy. Therefore, we need to perform non-uniform division processing according to the following steps;

[0059] Step S33: Define the following formula to sort the distance measuring sensors and the divided groups counterclockwise and clockwise starting from the rear of the unmanned ship:

[0060]

[0061] in, is the nonlinear mapping part, is the constant offset part; maps the given sensor index i∈{1,…,N} to the group index κ∈{1,…,D}, where tanh is the logic function γC is a scaling parameter used to control the density of the group distribution. The group near the front of the unmanned ship is the narrowest. At the same time, the distance measurement vector measured by the distance measurement sensor of the κth group is defined as ω κ ;

[0062] For the selection of the measurement standard value of a group, the minimum obstacle distance is usually selected as a representative, which is a conservative and safe value. However, this selection may be too restrictive in some scenarios, resulting in the neglect of the passable channels between obstacles, or ignoring the existence of small obstacles and selecting the maximum obstacle distance as a representative. This selection may cause danger in some scenarios. The present invention considers the position of the ranging sensor reading and the width of the unmanned ship, calculates the maximum distance that can be reached by the sensor in each group, iterates the ranging sensor data in ascending order corresponding to the ranging sensor, and checks whether the ship can pass for each distance level obtained, thereby optimizing the problems caused by the minimum obstacle distance and the maximum obstacle distance. The specific process is as follows:

[0063] Step S34: define the width of the unmanned ship as W, the angle between adjacent ranging sensors of the 360-degree full coverage ranging sensor as θ, and the measurement value of a certain group of ranging sensors as x = {x1, ..., x n}, arrange the measured values ​​x in ascending order to form an index collection I, then calculate the arc length between adjacent ranging sensors for each index, set the initial opening width that can pass between obstacles based on half of the arc length, perform a loop from 0 to n, traverse the readings of all ranging sensors, and when it is found that the width of a certain opening j is greater than the current opening width, the opening width is updated by adding the arc corresponding to the opening width of the opening j to the value of the current opening. If the updated opening width is greater than the width W of the unmanned boat, a true mark is attached to the opening and the loop is exited. If the updated opening width is not greater than the width W of the unmanned boat, the opening width is updated by adding half of the arc corresponding to the opening width of the opening j to the value of the current opening width, and the above process is repeated for loop traversal; select all openings with true marks as potential passage options, and if there is no opening with a true mark, it means that it is impassable;

[0064] See also Figure 5 ,Step S4: Make a collision avoidance decision. By integrating various information of obstacles obtained and combining the situation of the unmanned ship itself, a collision avoidance decision that meets the unmanned ship's own goals and is dynamically restricted is made. Specifically:

[0065] Step S41: Simplify the unmanned ship dynamics model in step S1 and define the dynamics model as Vector w represents the input caused by environmental disturbance; vector u is the control input β is a constant parameter, using the estimated state and interference To track the reference speed v ref , then the dynamic model can be expressed as By using a short time frame t Hd The integration is performed using the discretization interval t s The determined N d discrete sampling time, and obtain the predicted position that can be achieved based on the dynamic model The predicted position is used to calculate the distance between the unmanned ship and each obstacle. j The predicted distance between Respectively represent the position of the unmanned ship and the position of the obstacle;

[0066] Step S42: Control to reduce the deviation from the preferred speed and define the candidate speed set as Ω c ,Ω c =Ω grid ∪Ω int , where Ω gridis a set of velocity grid points, according to the preferred velocity v of the unmanned ship pref The fixed offset selected; speed set Ω int is to satisfy min||vv pref ||Reference speed set for optimization conditions;

[0067] Step S43: Reduce the instructions with large speed changes among the control instructions, and define an objective function to discard those instructions that cause large speed changes. The objective function is:

[0068]

[0069] Among them, q v ,q χ , Δq v , Δq χ and q τ is the corresponding empirical coefficient, v pref is the optimal speed of the unmanned ship, χ pref is the preferred heading of the unmanned ship, v last is the speed of the unmanned ship last time, χ last is the last heading of the unmanned ship, and τ is a time-dependent variable, which evaluates the time-varying responsibility of the unmanned ship in the collision process.

[0070] See also Figure 6 The present invention provides an unmanned ship obstacle avoidance control and wireless transmission system based on data fusion, which is used to implement the unmanned ship obstacle avoidance control and wireless transmission method based on data fusion. The system includes:

[0071] Unmanned boat obstacle avoidance control device 100 and collaborative control center 200;

[0072] The collaborative control center 200 is set up on the shore, and is used to establish wireless communication connection with the unmanned ship, send control instructions, and wirelessly keep the unmanned ship online in real time; it includes a wireless communication module 201, and the wireless communication module 201 has a DTU unit and a Beidou client, and communicates with the unmanned ship obstacle avoidance control device 100 by radio or satellite signals through the 5G network or Beidou satellite. During the communication process, when the 5G network is interrupted or fails within a certain period of time, it switches to Beidou satellite communication.

[0073] See also Figure 7 , the unmanned ship obstacle avoidance control device 100 comprises:

[0074] The dynamic model construction module 101 is used to construct the initialization dynamic model of the unmanned ship, and the initialization dynamic model of the unmanned ship is constructed by defining the coordinate frame, state space constraints, and ship driving environment constraints;

[0075] A path navigation feature vector construction module 102 defines a representation of a parameterized path, and characterizes a path navigation feature vector by setting a path reference point, a Euclidean distance, a forward path reference point, a path deviation, a heading error, and a path angle that are continuously updated based on the current position of the unmanned ship on the path to be navigated;

[0076] The data acquisition and processing module 103 is equipped with detection sensors around the unmanned ship in a 360-degree full coverage manner, and in the process of equipping, a dimension reduction method is adopted to set the detection sensors to a non-uniform distribution;

[0077] The collision avoidance decision-making module 104 makes a collision avoidance decision that satisfies the unmanned ship's own goals and is dynamically restricted by integrating various information about obstacles obtained and combining the unmanned ship's own conditions;

[0078] The wireless data transmission module 105 includes a DTU unit and a Beidou client, and is used for wireless communication with other unmanned ships and a collaborative control center.

[0079] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An unmanned ship obstacle avoidance control and wireless transmission method based on data fusion, characterized in that: The following steps are involved: Step S1: constructing an initialization dynamic model of the unmanned ship, by defining a coordinate frame, state space constraints, and ship driving environment constraints to construct the initialization dynamic model of the unmanned ship; Step S2: construct a path navigation feature vector, define the representation of the parameterized path, and characterize the path navigation feature vector by setting path reference points, Euclidean distances, forward path reference points, path deviations, heading errors, and path angles that are continuously updated based on the current position of the unmanned ship on the path to be navigated; Step S3: Setting a data acquisition and processing module, equipping the unmanned ship with detection sensors in a 360-degree full coverage manner, and using a dimension reduction method during the equipping process to set the detection sensors to a non-uniform distribution; Step S4: Make a collision avoidance decision. By integrating various information of obstacles acquired and combining the situation of the unmanned ship itself, a collision avoidance decision that meets the unmanned ship's own goals and is dynamically restricted is made; Step S5: Through the wireless data transmission module including the DTU unit and the Beidou client, the obstacle avoidance information and control instructions required during the driving of the unmanned ship are wirelessly transmitted to other unmanned ships and the collaborative control center.

2. The unmanned ship obstacle avoidance control and wireless transmission method based on data fusion as claimed in claim 1, characterized in that: The step S1 specifically includes: Step S11: Define that the unmanned ship is always located on the surface of the earth without floating up and down, pitching and rolling. The coordinate frame of the environment where the unmanned ship is located is n=(x n ,y n ,z n ), which forms a tangent plane with the Earth's surface, where x n The axis points to the north, y n The axis points east, z n The axis points to the center of the Earth; Step S12: define the state vector of the unmanned ship η = [x n ,y n ,ψ] T , where x n and n are the north and east positions in the coordinate frame of the environment, respectively; ψ is the heading angle, i.e. the angle between the longitudinal axis of the unmanned ship and the north axis; the translation and angular velocity vector v = [u, h, r] T It is composed of foreground velocity u, lateral velocity h and yaw rate r; Step S13: The unmanned boat is equipped with a rudder and a propeller at the tail, and can move freely along any trajectory in the state space. The control vector for controlling the unmanned boat is f = [T u ,T r ], where T u Represents the torque input in the forward direction, T r Represents the torque input in the yaw direction, then the initialization dynamic model of the unmanned ship is expressed as: Where M, C(v), and D(v) are all 3×3 state space vectors, that is, M is the mass matrix of the unmanned ship, covering the rigid body effect and the added mass effect; C(v) is the centripetal force parameter matrix; D(v) is the damping matrix; B is the actuator configuration matrix, which belongs to the 3×2 state space vector; R z,ψ (η) represents the direction along z n The axis rotation ψ in radians.

3. The unmanned ship obstacle avoidance control and wireless transmission method based on data fusion as claimed in claim 1, characterized in that: The step S2 specifically includes: Step S21: define the parameterized path as p d (ω)=[x d (ω),y d (ω)] T , where ω is the path variable, p d (ω) represents the path reference point, which is set to be the point on the right side of the unmanned ship at the Euclidean distance from the unmanned ship. d (ω),y d (ω) represents the coordinates of the path reference point. Correspondingly, the forward path reference point needs to be set It is the path reference point p d (ω) Go forward Δ along the path LA Distance point; Step S22: Finding the path reference point p d (ω), it is necessary to calculate the corresponding value of the path variable ω at each time step The path deviation between the expected path and the current trajectory is defined as ε, and the heading error ψ required for the unmanned ship to sail directly from the current position to the forward path reference point is defined as: Further define the path angle Taking the inverse of the first-order path as a reference, the smoothness of the unmanned ship trajectory when it relies on the forward path reference point is: Step S23: By converting the path variable ω calculated in the above step into the corresponding value of each time step Path deviation ε, heading error Path Angle Together they form the path navigation feature vector s (t) n .

4. The unmanned ship obstacle avoidance control and wireless transmission method based on data fusion as claimed in claim 1, characterized in that: The step S3 specifically includes: Step S31: Equip a distance measuring sensor as a detection sensor, the number of distance measuring sensors is N, and the maximum detection range is S r , equipped around the unmanned ship with 360-degree full coverage; Step S32: Divide the 360-degree full coverage ranging sensors into D groups, each group generates a scalar detection value, so that it is included in the final observation vector; Step S33: Define the following formula to sort the distance measuring sensors and the divided groups counterclockwise and clockwise starting from the rear of the unmanned ship: in, is the nonlinear mapping part, is the constant offset part; maps the given sensor index i∈{1,…,N} to the group index κ∈{1,…,D}, where tanh is the logic function γC is a scaling parameter used to control the density of the group distribution. The group near the front of the unmanned ship is the narrowest. At the same time, the distance measurement vector measured by the distance measurement sensor of the κth group is defined as ω κ ; Step S34: define the width of the unmanned ship as W, the angle between adjacent ranging sensors of the 360-degree full coverage ranging sensor as θ, and the measurement value of a certain group of ranging sensors as x = {x1, ..., x n }, arrange the measured values ​​x in ascending order to form an index collection I, then calculate the arc length between adjacent ranging sensors for each index, set the initial opening width that can pass between obstacles based on half of the arc length, perform a loop from 0 to n, traverse the readings of all ranging sensors, and when it is found that the width of a certain opening j is greater than the current opening width, the opening width is updated by adding the arc corresponding to the opening width of the opening j to the value of the current opening. If the updated opening width is greater than the width W of the unmanned boat, a true flag is attached to the opening and the loop is exited. If the updated opening width is not greater than the width W of the unmanned boat, the opening width is updated by adding half of the arc corresponding to the opening width of the opening j to the value of the current opening width, and the above process is repeated for loop traversal; select all openings marked as true as potential passage options, and if no opening is marked as true, it means that it is impassable.

5. The unmanned ship obstacle avoidance control and wireless transmission method based on data fusion as claimed in claim 1, characterized in that: The step S4 specifically includes: Step S41: Simplify the unmanned ship dynamics model in step S1 and define the dynamics model as Vector w represents the input caused by environmental disturbance; vector u is the control input β is a constant parameter, using the estimated state and interference To track the reference speed v ref , then the dynamic model can be expressed as By using a short time frame t Hd The integration is performed using the discretization interval t s The determined N d discrete sampling time, and obtain the predicted position that can be achieved based on the dynamic model The predicted position is used to calculate the distance between the unmanned ship and each obstacle. j The predicted distance between Respectively represent the position of the unmanned ship and the position of the obstacle; Step S42: Control to reduce the deviation from the preferred speed and define the candidate speed set as Ω c ,Ω c =Ω grid ∪Ω int , where Ω grid is a set of velocity grid points, according to the preferred velocity v of the unmanned ship pref The fixed offset selected; speed set Ω int is to satisfy min||vv pref ||Reference speed set for optimization conditions; Step S43: Reduce the instructions with large speed changes among the control instructions, and define an objective function to discard those instructions that cause large speed changes. The objective function is: Among them, q v ,q χ , Δq v , Δq χ and q τ is the corresponding empirical coefficient, v pref is the optimal speed of the unmanned ship, χ pref is the preferred heading of the unmanned ship, v last is the speed of the unmanned ship last time, χ last is the last heading of the unmanned ship, and τ is a time-dependent variable, which evaluates the time-varying responsibility of the unmanned ship in the collision process.

6. An unmanned ship obstacle avoidance control and wireless transmission system based on data fusion, used to implement an unmanned ship obstacle avoidance control and wireless transmission method based on data fusion as described in any one of claims 1 to 5, characterized in that: The system includes: Unmanned boat obstacle avoidance control device 100 and collaborative control center 200; The collaborative control center 200 is set up on the shore, and is used to establish wireless communication connection with the unmanned ship, send control instructions, and wirelessly keep the unmanned ship online in real time; it includes a wireless communication module 201, and the wireless communication module 201 has a DTU unit and a Beidou client, and communicates with the unmanned ship obstacle avoidance control device 100 by radio or satellite signals through the 5G network or Beidou satellite. During the communication process, when the 5G network is interrupted or fails within a certain period of time, it switches to Beidou satellite communication.

7. The unmanned ship obstacle avoidance control and wireless transmission system based on data fusion as claimed in claim 6, characterized in that: The unmanned ship obstacle avoidance control device 100 includes: The dynamic model construction module 101 is used to construct the initialization dynamic model of the unmanned ship, and the initialization dynamic model of the unmanned ship is constructed by defining the coordinate frame, state space constraints, and ship driving environment constraints; A path navigation feature vector construction module 102 defines a representation of a parameterized path, and characterizes a path navigation feature vector by setting a path reference point, a Euclidean distance, a forward path reference point, a path deviation, a heading error, and a path angle that are continuously updated based on the current position of the unmanned ship on the path to be navigated; The data acquisition and processing module 103 is equipped with detection sensors around the unmanned ship in a 360-degree full coverage manner, and in the process of equipping, a dimension reduction method is adopted to set the detection sensors to a non-uniform distribution; The collision avoidance decision-making module 104 makes a collision avoidance decision that satisfies the unmanned ship's own goals and is dynamically restricted by integrating various information about obstacles obtained and combining the unmanned ship's own conditions; The wireless data transmission module 105 includes a DTU unit and a Beidou client, and is used for wireless communication with other unmanned ships and a collaborative control center.

Images