An improved dwa-based local path planning method for AUV

Abstract

The application relates to the field of path specification, in particular to an AUV local path planning method based on improved DWA, which comprises the following steps: referring to the local path planning of an AUV in a horizontal plane to define a quality matrix; referring to a traditional motion state equation of the AUV to obtain an acceleration function with a control input vector and a speed vector as variables; setting multiple application situations, obtaining maximum acceleration according to specific values of angular velocity, and then evaluating and defining allowable speed; designing adaptive parameters, adopting an adaptive mode to adjust parameters of a speed evaluation function according to obstacle information and the state of the AUV; selecting a positive definite function according to selected optimal expected speed, designing a controller based on a backstepping method, and forming a decision-control closed loop; and the AUV obtains state information and surrounding obstacle information through a perception device carried by the AUV, obtains control expectation and control input through the above algorithm, and realizes flexible and accurate obstacle avoidance.

Description

Technical fields:

[0001] This invention relates to an AUV local path planning method based on improved DWA. Background technology:

[0002] AUV stands for Autonomous Underwater Vehicle, a type of robot commonly used in ocean exploration. It is equipped with a variety of sensors that are not limited by time and space, and has autonomous navigation and obstacle avoidance capabilities. In addition, AUV can autonomously perform specific underwater tasks. The development of AUV involves a wide range of high technologies, such as mechanics, fluid mechanics, underwater acoustics, and optics, as well as many contemporary scientific and technological fields such as electronic communication, navigation, automatic control, computer science, sensor technology, bionics, and artificial intelligence.

[0003] In the field of marine engineering, AUVs are used for dam structure inspection, underwater base installation and dismantling, underwater target observation and search and rescue, and to provide auxiliary work for divers. In the field of marine scientific research, AUVs can be deployed in the marine environment for data collection, underwater shipwreck investigation, seabed geological and geomorphological exploration, and exploration for resources such as oil. In the military field, AUVs are used for mine countermeasures, target detection, intelligence gathering, surveillance and reconnaissance, environmental data collection, and anti-submarine warfare. As human activities gradually expand into the marine domain, AUVs will play an important role in marine exploration.

[0004] For AUVs, path planning is a prerequisite for achieving unmanned operation and the most important manifestation of AUV intelligence. However, in dynamic and uncertain underwater environments, it is difficult to obtain various obstacle information before planning a path. In this case, in addition to a global path planner, AUVs usually still need a local path planner to avoid these unknown and dynamic obstacles. Existing local path planning methods mainly use sonar sensors to obtain information about surrounding obstacles and utilize various local dynamic path planning techniques (such as DWA) to complete obstacle avoidance tasks.

[0005] The Dynamic Window (DWA) algorithm directly searches for the optimal command in the robot's possible command space, taking into account some constraints on the robot. The output command of this algorithm takes into account the robot's dynamic performance and provides good results in many scenarios. However, this algorithm limits the robot's dynamic performance by directly giving the robot's maximum acceleration, which does not reflect the actual situation to some extent. In addition, the optimization evaluation function weights of the traditional DWA algorithm are fixed values, which makes the autonomous underwater vehicle unable to adapt to complex sea conditions. Its evaluation index does not change with the environment, resulting in poor flexibility. Furthermore, it does not form closed-loop control, making it less practical. Summary of the Invention:

[0006] This invention provides an AUV local path planning method based on an improved DWA. The method is rationally designed, limiting the dynamic performance of the AUV from the control input level, and adopting an adaptive parameter adjustment method to improve the flexibility in practical applications. This allows the evaluation indicators to change with environmental changes, making it applicable to complex sea conditions. At the same time, a backstepping algorithm is used to design the speed controller, forming a "decision-control" closed loop. It is applicable to various AUVs according to the actual needs of users, has strong practicality, and solves the problems existing in the prior art.

[0007] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0008] A local path planning method for AUVs based on improved DWA, the local path planning method includes the following steps:

[0009] S1, referring to the local path planning of the AUV on the horizontal plane, define the velocity vector, position vector and control input vector of the AUV, define the mass matrix, which includes mechanically added mass and hydrodynamically added mass, and obtain the conventional motion state equation of the AUV;

[0010] S2, referring to the traditional motion state equation of AUV, uses the control input vector and velocity vector as variables to obtain its acceleration function;

[0011] S3 sets up multiple application scenarios, obtains the maximum acceleration based on the specific value of the angular velocity, and then evaluates and defines the allowable speed to obtain the allowable speed space and speed dynamic window. The feasible speed space is then calculated based on the allowable speed space and speed dynamic window.

[0012] S4. Design adaptive parameters and adopt an adaptive approach to adjust the parameters of the speed evaluation function based on obstacle information and its own state. Define a maximization objective function and select the optimal speed within the feasible speed space for flexible obstacle avoidance.

[0013] S5. Based on the selected optimal desired speed, a positive definite function is chosen, and a controller is designed based on the backstepping method to form a decision-control closed loop;

[0014] S6, the AUV obtains its state information z(t) through its onboard sensing devices. k ), x(t) k Based on the information of surrounding obstacles, the desired control value x is obtained by solving the algorithm described above. d (t k+1 ) and control input u(t) k+1 This allows for flexible and precise obstacle avoidance.

[0015] The conventional motion state equation of the AUV is:

[0016]

[0017]

[0018] Among them, x(t)=[u(t),v(t),r(t)] T Let z(t) be the velocity vector of the AUV, and z(t) = [X(t), Y(t), ψ(t)]. T Let u(t) be the position vector of the AUV, and u(t) = [δ r (t), 0, n(t)] T δ is the control input vector for the AUV. r n(t) and n(t) are the AUV rudder angle and AUV propeller speed, respectively; functions f and g are the mappings of AUV motion to forces, including Coriolis force, gravity, and centrifugal force; c is the coefficient of hydrostatic and hydrodynamic forces and torques acting on the AUV in the body coordinate system; function h reflects the kinematic relationship when performing coordinate transformations between the body coordinate system and the global coordinate system.

[0019] The AUV acceleration function is:

[0020]

[0021] It can be simplified to

[0022]

[0023] Let r(t) > 0 and δ r When (t)>0, r(t) < 0 and δ r When (t) < 0, When u(t)>0 and n(t=0,

[0024] When r < 0, take

[0025]

[0026] When r = 0, take

[0027]

[0028] at this time,

[0029] When r>0, take

[0030]

[0031] The and These represent the maximum linear acceleration and maximum angular acceleration of the AUV, respectively.

[0032] The permissible velocity space is:

[0033]

[0034] Where dist1(u,r) and dist2(u,r) represent the arc distance to the nearest obstacle intersecting the trajectory and the size of its corresponding central angle, respectively.

[0035] The speed dynamic window is:

[0036]

[0037] The feasible velocity space is defined as: v r ′=v a ′∩v d ′.

[0038] An adaptive approach is adopted, adjusting the parameters of the speed evaluation function based on obstacle information and the user's own state, defining a maximization objective function, and selecting the optimal speed within the feasible speed space for flexible obstacle avoidance, including the following steps:

[0039] S4.1, the distance threshold is defined as D based on the AUV's dynamic performance. s ;

[0040] S4.2, with D min This represents the distance from the AUV to the nearest obstacle, where α0, β0, and γ0 are initial parameters, and α max γ min Let μ and ρ be the upper and lower bounds of the initial parameters, respectively, and μ and ρ be constants. Then the adaptive parameter design is as follows:

[0041]

[0042] β a =β0

[0043]

[0044] S4.3, Adjust the parameters of the velocity evaluation function and define the maximization objective function as follows:

[0045] G a (u, r) = α a ·heading(u,r)+β a ·dist(u,r)+γ a ·velocity(u, r)

[0046] Here, heading(u,r) is the evaluation component of the AUV's movement toward the target, defined as 180-θ, where θ is the angle between the end heading of the AUV's predicted trajectory and the expected heading at that point; dist(u,r) is the distance between the end of the AUV's predicted trajectory and the nearest obstacle; and velocity(u,r) is the magnitude of the linear velocity of the AUV moving along the arc trajectory.

[0047] The optimal expected speed x of the AUV d (t)=[u d (t),v(t),r d [t], where the positive definite function is a Lyapunov function.

[0048] Where e(t) = x(t) - x d (t), to obtain

[0049]

[0050] The control law of the controller is designed as follows:

[0051]

[0052] Where K is a positive definite constant matrix, from which we can obtain

[0053]

[0054] According to Lyapunov stability theory, the equilibrium state of the system at the origin is uniformly asymptotically stable.

[0055] This invention employs the aforementioned structure. By referencing the local path planning of an AUV on the horizontal plane, relevant vectors are defined to obtain the traditional motion state equation of the AUV. Based on the traditional motion state equation of the AUV, the acceleration function is obtained using the control input vector and velocity vector as variables. By setting various application scenarios, the maximum acceleration is obtained according to the specific value of the angular velocity, and the allowable speed is evaluated and defined. At the same time, the feasible speed space is obtained by combining the speed window within time Δt. By designing adaptive parameters, the parameters of the speed evaluation function are adjusted in an adaptive manner according to obstacle information and the self-state, defining a maximization objective function, and selecting the optimal speed within the feasible speed space for flexible obstacle avoidance. The controller is designed using the backstepping method, making the system more stable and possessing the advantages of precision, flexibility, stability, and practicality. Attached image description:

[0056] Figure 1 This is a schematic diagram of the AUV model structure of the present invention.

[0057] Figure 2This is a schematic diagram of the feasible velocity space of the present invention.

[0058] Figure 3 This is a schematic diagram illustrating the implementation process of the present invention. Detailed implementation method:

[0059] To clearly illustrate the technical features of this solution, the invention will be described in detail below through specific implementation methods and in conjunction with the accompanying drawings.

[0060] like Figure 1-3 As shown, an AUV local path planning method based on improved DWA is described, the local path planning method comprising the following steps:

[0061] S1, referring to the local path planning of the AUV on the horizontal plane, define the velocity vector, position vector and control input vector of the AUV, define the mass matrix, which includes mechanically added mass and hydrodynamically added mass, and obtain the conventional motion state equation of the AUV;

[0062] S2, referring to the traditional motion state equation of AUV, uses the control input vector and velocity vector as variables to obtain its acceleration function;

[0063] S3 sets up multiple application scenarios, obtains the maximum acceleration based on the specific value of the angular velocity, and then evaluates and defines the allowable speed to obtain the allowable speed space and speed dynamic window. The feasible speed space is then calculated based on the allowable speed space and speed dynamic window.

[0064] S4. Design adaptive parameters and adopt an adaptive approach to adjust the parameters of the speed evaluation function based on obstacle information and its own state. Define a maximization objective function and select the optimal speed within the feasible speed space for flexible obstacle avoidance.

[0065] S5. Based on the selected optimal desired speed, a positive definite function is chosen, and a controller is designed based on the backstepping method to form a decision-control closed loop;

[0066] S6, the AUV obtains its state information z(t) through its onboard sensing devices. k ), x(t) k Based on the information of surrounding obstacles, the desired control value x is obtained by solving the algorithm described above. d (t k+1 ) and control input u(t) k+1 This allows for flexible and precise obstacle avoidance.

[0067] The conventional motion state equation of the AUV is:

[0068]

[0069]

[0070] Among them, x(t)=[u(t),v(t),r(t)] T Let z(t) be the velocity vector of the AUV, and z(t) = [X(t), Y(t), ψ(t)]. T Let u(t) be the position vector of the AUV, and u(t) = [δ r (t), 0, n(t)] T δ is the control input vector for the AUV. r n(t) and n(t) are the AUV rudder angle and AUV propeller speed, respectively; functions f and g are the mappings of AUV motion to forces, including Coriolis force, gravity, and centrifugal force; c is the coefficient of hydrostatic and hydrodynamic forces and torques acting on the AUV in the body coordinate system; function h reflects the kinematic relationship when performing coordinate transformations between the body coordinate system and the global coordinate system.

[0071] The AUV acceleration function is:

[0072]

[0073] It can be simplified to

[0074]

[0075] Let r(t) > 0 and δ r When (t)>0, r(t) < 0 and δ r When (t) < , When u(t)>0 and n(t=0,

[0076] When r < 0, take

[0077]

[0078] When r = 0, take

[0079]

[0080] at this time,

[0081] When r>0, take

[0082]

[0083] The and These represent the maximum linear acceleration and maximum angular acceleration of the AUV, respectively.

[0084] The permissible velocity space is:

[0085]

[0086] Where dist1(u,r) and dist2(u,r) represent the arc distance to the nearest obstacle intersecting the trajectory and the size of its corresponding central angle, respectively.

[0087] The speed dynamic window is:

[0088]

[0089] The feasible velocity space is defined as: V r ′-V a ′∩V d ′.

[0090] An adaptive approach is adopted, adjusting the parameters of the speed evaluation function based on obstacle information and the user's own state, defining a maximization objective function, and selecting the optimal speed within the feasible speed space for flexible obstacle avoidance, including the following steps:

[0091] S4.1, Define the distance threshold as D based on the AUV's dynamic performance (such as braking distance). s ;

[0092] S4.2, with D min This represents the distance from the AUV to the nearest obstacle, where α0, β0, and γ0 are initial parameters, and α max γ min Let μ and ρ be the upper and lower bounds of the initial parameters, respectively, and μ and ρ be constants. Then the adaptive parameter design is as follows:

[0093]

[0094] β a -β0

[0095]

[0096] S4.3, Adjust the parameters of the velocity evaluation function and define the maximization objective function as follows:

[0097] G a (u, r) = α a ·heading(u,r)+β a ·dist(u,r)+γ a ·velocity(u, r)

[0098] Here, heading(u,r) is the evaluation component of the AUV's movement toward the target, defined as 180-θ, where θ is the angle between the end heading of the AUV's predicted trajectory and the expected heading at that point; dist(u,r) is the distance between the end of the AUV's predicted trajectory and the nearest obstacle; and velocity(u,r) is the magnitude of the linear velocity of the AUV moving along the arc trajectory.

[0099] The optimal expected speed x of the AUV d (t)=[u d (t),v(t),r d [t], where the positive definite function is a Lyapunov function.

[0100] Where e(t) = x(t) - x d (t), to obtain

[0101]

[0102] The control law of the controller is designed as follows:

[0103]

[0104] Where K is a positive definite constant matrix, from which we can obtain

[0105]

[0106] According to Lyapunov stability theory, the equilibrium state of the system at the origin is uniformly asymptotically stable.

[0107] The working principle of an AUV local path planning method based on improved DWA in this embodiment of the invention is as follows: The dynamic performance of the AUV is limited at the control input level, and adaptive parameter adjustment is used to improve flexibility in practical applications, allowing evaluation indicators to change with environmental variations, thus making it applicable to complex sea conditions. Simultaneously, a backstepping algorithm is used to design a speed controller, forming a "decision-control" closed loop. This method is suitable for various AUVs according to actual user needs, demonstrating strong practicality. Compared to existing conventional local path planning algorithms, it better meets the application requirements of actual sea conditions, ensuring that the AUV can automatically and flexibly avoid obstacles.

[0108] An AUV is a type of underwater vehicle that integrates artificial intelligence and other advanced computing technologies. It incorporates deep-sea submersibles, sensors, environmental effects, computer software, energy storage, conversion and propulsion, new materials and processes, and underwater intelligent weapons. Therefore, underwater path planning and obstacle avoidance are essential functions for AUVs.

[0109] In recent years, many path planning methods have emerged to study the avoidance of unknown and dynamic obstacles, such as rapid exploration random trees, artificial potential fields, fuzzy logic algorithms, neural networks, reinforcement learning, and even deep reinforcement learning. Unlike other robotic platforms or aircraft on land, these methods mainly use sonar sensors to obtain the real-time distance and angle of surrounding obstacles and use local dynamic path planning technology to complete the obstacle avoidance task, which greatly improves the autonomy of AUVs.

[0110] In the overall scheme of this invention, the local path planning method mainly includes the following steps: referring to the local path planning of an AUV on a horizontal plane, defining the velocity vector, position vector, and control input vector of the AUV, defining a mass matrix, the mass matrix including mechanically added mass and hydrodynamically added mass, and obtaining the traditional motion state equation of the AUV; referring to the AUV state equation, using the control input vector and velocity vector as variables, obtaining its acceleration function; setting multiple application scenarios, obtaining the maximum acceleration based on the specific value of the angular velocity, and then evaluating and defining the permissible speed, obtaining the permissible speed space and speed dynamic window, and calculating the feasible speed space based on the permissible speed space and speed dynamic window; designing adaptive parameters, adopting an adaptive approach, adjusting the parameters of the speed evaluation function according to obstacle information and its own state, defining a maximization objective function, selecting the optimal speed within the feasible speed space for flexible obstacle avoidance; selecting a positive definite function based on the selected optimal expected speed, designing a controller based on the backstepping method, forming a decision-control closed loop; the AUV obtains its state information z(t) through its onboard sensing devices. k ), x(t) k Based on the information of surrounding obstacles, the desired control value x is obtained by solving the algorithm described above. d (t k+1 ) and control input u(t) k+1 This allows for flexible and precise obstacle avoidance.

[0111] The traditional motion state equation for an AUV is as follows:

[0112]

[0113]

[0114] Among them, x(t)=[u(t),v(t),r(t)] T The velocity vector of the AUV.

[0115] z(t)=[X(t),Y(t),ψ(t)] T Let u(t) be the position vector of the AUV, and u(t) = [δ r (t), 0, n(t)] T δ is the control input vector for the AUV. rn(t) and n(t) are the AUV rudder angle and AUV propeller speed, respectively; functions f and g are the mappings of AUV motion to forces, including Coriolis force, gravity, and centrifugal force; c is the coefficient of hydrostatic and hydrodynamic forces and torques acting on the AUV in the body coordinate system; function h reflects the kinematic relationship when performing coordinate transformations between the body coordinate system and the global coordinate system.

[0116] The dynamic window method takes into account the dynamic performance of the AUV and presents a local collision avoidance method that only calculates the linear velocity and angular velocity that the robot can reach within a given time.

[0117] In this application, based on the traditional motion state equations of an AUV and combined with external environmental factors, the acceleration function is obtained using the control input vector and velocity vector as variables:

[0118]

[0119] It can be simplified to

[0120]

[0121] To approximate real-world application scenarios, three application scenarios were used to evaluate whether the speed was permissible, resulting in the permissible speed space and the speed dynamic window.

[0122] Let r(t) > 0 and δ r When (t)>0, r(t) < 0 and δ r When (t) < 0, When u(t)>0 and n(t=0,

[0123] When r < 0, take

[0124]

[0125] When r = 0, take

[0126]

[0127] at this time,

[0128] When r>0, take

[0129]

[0130] The and These represent the maximum linear acceleration and maximum angular acceleration of the AUV, respectively.

[0131] The permissible velocity space is:

[0132]

[0133] Where dist1(u,r) and dist2(u,r) represent the arc distance to the nearest obstacle intersecting the trajectory and the size of its corresponding central angle, respectively.

[0134] The speed dynamic window is:

[0135]

[0136] Then the feasible velocity space V is calculated. r ′=V a ′∩V d ′.

[0137] To select the optimal desired speed, unlike the traditional DWA algorithm, this application adopts an adaptive parameter design approach. The parameters of the speed evaluation function are adjusted based on obstacle information and the user's own state to achieve automatic, flexible, and accurate obstacle avoidance.

[0138] Specifically, the adaptive approach using adaptive parameters mainly includes the following steps: defining a distance threshold, designing the corresponding adaptive parameters, and defining the objective function to maximize the target.

[0139] After obtaining the optimal desired speed, a transformation operation is performed using a positive definite Lyapunov function to design and obtain...

[0140]

[0141] Where K is a positive definite constant matrix, from which we can obtain

[0142]

[0143] According to Lyapunov stability theory, the equilibrium state of the system at the origin is uniformly asymptotically stable.

[0144] Following the steps described above, the AUV acquires its state information z(t) through its onboard sensing devices. k ), x(t) k The system obtains information about the surrounding obstacles and corresponding control expectations and inputs, enabling the AUV to achieve automatic, sensitive, and precise obstacle avoidance when performing tasks.

[0145] It should be noted that the sensing devices (mainly sonar, inertial navigation, etc.) mounted on the AUV should be designed to avoid detection errors as much as possible in order to obtain the corresponding obstacle information and self-state information in real time and accurately.

[0146] In summary, the AUV local path planning method based on improved DWA in this embodiment of the invention limits the dynamic performance of the AUV at the control input level and improves the flexibility in practical applications by adopting adaptive parameter adjustment, so that the evaluation index changes with the environment, thus making it applicable to complex sea conditions. At the same time, a backstepping algorithm is used to design the speed controller, forming a "decision-control" closed loop. According to the actual needs of users, it is applicable to a variety of different AUVs, has strong practicality, and is more in line with the application requirements of actual sea conditions compared with existing conventional local path planning algorithms, ensuring that the AUV can automatically and flexibly avoid obstacles.

[0147] The above specific embodiments should not be construed as limiting the scope of protection of the present invention. For those skilled in the art, any alternative improvements or modifications made to the embodiments of the present invention shall fall within the scope of protection of the present invention.

[0148] Any aspects of this invention not described in detail are well-known to those skilled in the art.

Claims

1. A local path planning method for AUVs based on improved DWA, characterized in that, The local path planning method Includes the following steps: S1, referring to the local path planning of the AUV on the horizontal plane, define the velocity vector, position vector and control input vector of the AUV, define the mass matrix, which includes mechanically added mass and hydrodynamically added mass, and obtain the conventional motion state equation of the AUV; S2, referring to the traditional motion state equation of AUV, uses the control input vector and velocity vector as variables to obtain its acceleration function; S3 sets up multiple application scenarios, obtains the maximum acceleration based on the specific value of the angular velocity, and then evaluates and defines the allowable speed to obtain the allowable speed space and speed dynamic window. The feasible speed space is then calculated based on the allowable speed space and speed dynamic window. S4. Design adaptive parameters and adopt an adaptive approach to adjust the parameters of the speed evaluation function based on obstacle information and its own state. Define a maximization objective function and select the optimal speed within the feasible speed space for flexible obstacle avoidance. S5. Based on the selected optimal desired speed, a positive definite function is chosen, and a controller is designed based on the backstepping method to form a decision-control closed loop; S6, the AUV obtains its state information z(t) through its onboard sensing devices. k ), x(t) k Based on the information of surrounding obstacles, the desired control value x is obtained by solving the algorithm. d (t k+1 ) and control input u(t) k+1 To achieve flexible and precise obstacle avoidance; The AUV acceleration function is: ； It can be simplified to ； set up hour, ; hour, ; hour, ; When r < 0, take ； When r=0, take ； at this time, ; When r>0, take ； The (u,r,t) and (u, r, t) represent the maximum linear acceleration and maximum angular acceleration of the AUV, respectively; The permissible velocity space is: ； ； Where dist1(u,r) and dist2(u,r) represent the arc distance to the nearest obstacle intersecting the trajectory and the size of its corresponding central angle, respectively.

2. The AUV local path planning method based on improved DWA according to claim 1, characterized in that, The conventional motion state equation of the AUV is: ； Among them, x(t)=[u(t),v(t),r(t)] T Let z(t) be the velocity vector of the AUV, and z(t) = [X(t), Y(t), ψ(t)]. T Let u(t) be the position vector of the AUV. r (t), 0, n(t)] T δ is the control input vector for the AUV. r n(t) and n(t) are the AUV rudder angle and AUV propeller speed, respectively; functions f and g are the mappings of AUV motion to forces, including Coriolis force, gravity, and centrifugal force; c is the coefficient of hydrostatic and hydrodynamic forces and torques acting on the AUV in the body coordinate system; function h reflects the kinematic relationship when performing coordinate transformations between the body coordinate system and the global coordinate system.

3. The AUV local path planning method based on improved DWA according to claim 1, characterized in that, The speed dynamic window is: ； ； The feasible velocity space is defined as follows: .

4. The AUV local path planning method based on improved DWA according to claim 3, characterized in that, An adaptive approach is adopted, adjusting the parameters of the speed evaluation function based on obstacle information and the user's own state, defining a maximization objective function, and selecting the optimal speed within the feasible speed space for flexible obstacle avoidance, including the following steps: S4.1, the distance threshold is defined as D based on the AUV's dynamic performance. s ; S4.2, with D min This represents the distance from the AUV to the nearest obstacle, where α0, β0, and γ0 are initial parameters, and α max γ min Let μ and ρ be the upper and lower bounds of the initial parameters, respectively, and μ and ρ be constants. Then the adaptive parameter design is as follows: ； ； S4.3, Adjust the parameters of the velocity evaluation function and define the maximization objective function as follows: ； Here, heading(u,r) is the evaluation component of the AUV's movement toward the target, defined as 180-θ, where θ is the angle between the end heading of the AUV's predicted trajectory and the expected heading at that point; dist(u,r) is the distance between the end of the AUV's predicted trajectory and the nearest obstacle; and velocity(u,r) is the magnitude of the linear velocity of the AUV moving along the arc trajectory.

5. The AUV local path planning method based on improved DWA according to claim 1, characterized in that, The optimal expected speed x of the AUV d (t)=[u d (t), v(t),r d [t], where the positive definite function is a Lyapunov function. ; Where e(t) = x(t) - x d (t), to obtain ； The control law of the controller is designed as follows: ； Where K is a positive definite constant matrix, from which we can obtain ； According to Lyapunov stability theory, the equilibrium state of the system at the origin is uniformly asymptotically stable.

Images