Multi-cascade fuzzy controller of unmanned surface vehicle and semi-physical simulation test method
Through the design of a multi-cascade fuzzy controller, the control input variables are processed using the sub-fuzzy controller and the fuzzy rule library, the problem of large calculation volume of the surface unmanned boat controller is solved, and efficient control is achieved in complex marine environments. The unmanned boat was driven and the calculation amount was reduced.
Patent Information
- Application Number
- CN202410915051.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-09
- Publication Date
- 2024-11-26
AI Technical Summary
The existing surface unmanned boat controller has a large design and computing capacity, making it difficult to achieve effective autonomous control in complex and changeable marine environments. In particular, the calculation volume of under-drive systems is too large, which limits its application on light-load platforms.
Using a multi-cascaded fuzzy controller, a multi-cascaded fuzzy control chain is formed through several sub-fuzzy controllers connected in series with each other. Each sub-fuzzy controller is equipped with a fuzzy rule library, and the membership function is calculated based on the fuzzy input amount, reducing the non-linear model. Modeling requirements reduce the design calculation volume and improve control accuracy and adaptability.
It realizes efficient unmanned boat control in complex marine environments, reduces the calculation amount, improves control accuracy and adaptability, is suitable for under-driven unmanned boats, and is easy to deploy, and is not limited to the surface unmanned boat control model.
Smart Images

Figure CN119024714A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of unmanned boat controller testing, and in particular to a multi-cascade fuzzy controller for a surface unmanned boat and a semi-physical simulation testing method. Background Art
[0002] With its small size, flexible movement, and low processing cost, surface unmanned boats have become one of the main unmanned transport platforms that are indispensable in the marine environment. Under normal circumstances, the controllable degree of freedom variables of surface unmanned boats are smaller than their own degrees of freedom of motion, that is, they are equipped with a sway propulsion control system and a bow-swing steering control system. Therefore, most unmanned boat control systems have under-actuated characteristics. The complex motion model of unmanned boats is constrained by the changing marine environment and limited portable power resources. How to achieve autonomous control of under-actuated systems is still a difficult problem that needs to be solved urgently.
[0003] At present, the control methods of underactuated surface unmanned boats are mainly divided into backstepping control, sliding mode control, model predictive control and robust control. Backstepping control is to decompose the entire system into several subsystems and design the controllers of each subsystem based on the Lyapunov function. Although it can ensure the stability of the overall system, this control algorithm is too dependent on the system model, and its controller design is complex and requires a lot of calculations. Sliding mode control is to construct a sliding mode to force the nonlinear system to achieve the expected goal. However, in the actual application of surface unmanned boats, it will appear that the controlled state is difficult to slide along the sliding mode plane to the equilibrium point, and then the unmanned boat will vibrate. Model predictive control is highly practical and can establish the optimal control problem within the effective time according to various models, but its solution speed often depends on the complexity of the control model and the performance of the solver. The resources carried by light-load platforms such as surface unmanned boats do not support large-scale computing, so model predictive control and backstepping control are not suitable.
[0004] Faced with the complex and changeable marine environment and the under-actuated control system, if the controller is designed for each situation, it will inevitably bring huge amount of calculation. Therefore, there is an urgent need for an unmanned boat controller that can reduce the amount of design calculation. Summary of the invention
[0005] In view of this, the present invention provides a multi-cascade fuzzy controller for a surface unmanned boat and a semi-physical simulation test method to solve the technical problem of large amount of calculation in the existing unmanned boat controller design.
[0006] In the first aspect, the present invention provides a multi-cascade fuzzy controller for a surface unmanned boat, comprising: an input unit, used to determine the control input variables according to the target control task and classify the control input variables, and obtain the fuzzy input quantity after fuzzy processing the classified control input variables; a sub-fuzzy controller, wherein a plurality of sub-fuzzy controllers are provided, and the sub-fuzzy controllers are connected in series to form a multi-cascade fuzzy control chain, each of the sub-fuzzy controllers is provided with a corresponding fuzzy rule base, and the fuzzy input quantities of the same category are input into the corresponding sub-fuzzy controller, and the control output variables are obtained after calculation according to the membership function in the fuzzy rule base, and the control output variables of the current sub-fuzzy controller in the multi-cascade fuzzy control chain are used as the fuzzy input quantities of the sub-fuzzy controller of the next level; an output unit, used to output the control output variables of the sub-fuzzy controller of the last level in the multi-cascade fuzzy control chain.
[0007] The multi-cascade fuzzy controller of the surface unmanned boat of the embodiment of the present invention forms a multi-cascade fuzzy control chain by connecting several sub-fuzzy controllers in series. Each sub-fuzzy controller is provided with a corresponding fuzzy rule base. Fuzzy input quantities of the same category are input into the corresponding sub-fuzzy controller. The control output variable is obtained after calculation according to the membership function in the fuzzy rule base. The control output variable of the current sub-fuzzy controller in the multi-cascade fuzzy control chain is used as the fuzzy input quantity of the sub-fuzzy controller of the next level, so as to realize multi-cascade fuzzy control, and the control input variable can be processed more finely, and the control accuracy and adaptability can be improved. Each sub-fuzzy controller corresponds to a fuzzy rule base. The fuzzy logic control system contained in the membership function in the fuzzy rule base can continuously approximate nonlinear functions, and does not need to model nonlinear models one by one, thereby reducing the design calculation amount, and has the advantages of easy deployment and not being limited to the surface unmanned boat control model.
[0008] Optionally, the categories of the control input variables include environmental parameters, azimuth differences and azimuth angle differences, and the sub-fuzzy controller includes a first sub-fuzzy controller, a second sub-fuzzy controller and a third sub-fuzzy controller. After the fuzzy input quantity corresponding to the environmental parameter is input into the third sub-fuzzy controller, the third sub-fuzzy controller calculates the environmental factor as the control output variable according to the membership function, the environmental factor and the azimuth angle difference after fuzzy processing are input into the first sub-fuzzy controller to obtain a first control quantity, the environmental factor and the azimuth angle difference after fuzzy processing are input into the second sub-fuzzy controller to obtain a second control quantity, and a fuzzy control quantity is generated according to the first control quantity and the second control quantity as the control output variable of the sub-fuzzy controller of the last level in the multi-cascade fuzzy control chain.
[0009] In this method, by taking environmental parameters, azimuth difference and azimuth angle difference as control input variables, different types of inputs can be specially processed through cascade processing of multiple sub-fuzzy controllers, thereby improving the pertinence and effectiveness of control, realizing the control of the controlled unmanned boat in the target tracking task, and reducing the computational complexity of each sub-fuzzy controller.
[0010] In a second aspect, the present invention provides a semi-physical simulation test method for a multi-cascade fuzzy controller of a surface unmanned boat, which is used to perform a semi-physical simulation test on the multi-cascade fuzzy controller of a surface unmanned boat described in any one of the first aspects. The semi-physical simulation test method includes: receiving a target control task issued by a shore-based server, and establishing a three-dimensional simulation scene based on dynamics according to the target control task; obtaining control input variables according to the three-dimensional simulation scene, and sending the control input variables to the multi-cascade fuzzy controller of the surface unmanned boat through a communication link; receiving control output variables generated based on the control input variables and returned by the multi-cascade fuzzy controller of the surface unmanned boat; updating the three-dimensional simulation scene according to the control output variables and re-acquiring the control input variables, and performing the next round of simulation control according to the re-acquired control input variables until the target control task is completed and the test results are obtained; and sending the test results to the shore-based server.
[0011] The semi-physical simulation test method of the multi-cascade fuzzy controller of the surface unmanned boat of the embodiment of the present invention receives the target control task issued by the shore-based server, establishes a three-dimensional simulation scene based on dynamics according to the target control task; obtains the control input variable according to the three-dimensional simulation scene, and sends the control input variable to the multi-cascade fuzzy controller of the surface unmanned boat through the communication link; receives the control output variable generated based on the control input variable returned by the multi-cascade fuzzy controller of the surface unmanned boat; updates the three-dimensional simulation scene according to the control output variable and re-acquires the control input variable, performs the next round of simulation control according to the re-acquired control input variable until the target control task is completed and the test result is obtained; sends the test result to the shore-based server, and performs a real communication connection with the shore-based server and the multi-cascade fuzzy controller of the surface unmanned boat during the simulation process, so as to simulate the communication delay, data packet loss and the like in the actual working condition in real time, and the environmental elements of the scene can be supplemented by a series of methods through the three-dimensional simulation scene based on dynamics, which is closer to various weather / hydrological environments that may be encountered in the actual simulation process, so that the simulation test environment is closer to the actual working condition and the test result is more accurate.
[0012] Optionally, the three-dimensional simulation scene includes an unmanned boat dynamics model, a hydrodynamics model, a sensor module, a communication module and a signal conversion module.
[0013] Dynamic simulation is achieved through the unmanned boat dynamics model, hydrodynamics model, sensor module, communication module and signal conversion module. The real environment can be simulated through the sensor module, communication module and signal conversion module, making the simulation closer to the actual situation and improving the reliability and practicability of the simulation.
[0014] Optionally, the unmanned boat dynamics model includes the kinematic equation of the controlled unmanned boat, the kinematic equation of the target unmanned boat and the dynamic equation of the controlled unmanned boat.
[0015] Optionally, the kinematic equation of the controlled unmanned boat is:
[0016]
[0017] In the formula, x s and s Respectively represent the coordinates of the controlled unmanned boat in the x direction and the y direction, and Respectively represent x s and s The linear velocity obtained by derivation, u s 、v s and r s They represent the angular velocities of the controlled unmanned boat, ψ s represents the heading angle of the controlled unmanned boat, Represents ψ s The angular velocity obtained by taking the derivative;
[0018] The kinematic equation of the target unmanned boat is:
[0019]
[0020] In the formula, x t and t Respectively represent the coordinates of the target unmanned boat in the x direction and the y direction, and Respectively represent x t and t The linear velocity obtained by derivation, u t 、v t and r t They represent the angular velocities of the target unmanned boat, ψ t represents the heading angle of the target unmanned boat, Represents ψ t The angular velocity obtained by taking the derivative;
[0021] The dynamic equation of the controlled unmanned boat is:
[0022]
[0023] In the formula, F u (u s , v s , r s ), F v (u s , v s , r s ) and F r (u s , v s , r s ) represent the axial forces in three directions in space, m u 、m v and m r represents the mass and moment of inertia coefficient of the controlled unmanned boat, τ u and τ r They respectively represent the input force and moment in the surge and pitch control directions of the controlled unmanned boat, and Respectively represent u t 、v t and r t Take the derivative and get the angular velocity.
[0024] This method provides the kinematic equations of the unmanned boat and the target unmanned boat and the dynamic equations of the controlled unmanned boat, which provides a detailed physical basis for the simulation and ensures the accuracy of the simulation results.
[0025] Optionally, the equation of the hydrodynamic model is described as:
[0026]
[0027] In the formula, u w 、v w and r w Represents the water speed in three directions in space, and Respectively represent u w and v w The acceleration obtained by derivation, P represents pressure, ρ represents density, f c represents the Coriolis force, f wx and f wy They represent the wind resistance in the x and y directions respectively, g represents the acceleration of gravity, T represents the temperature, and p w represents rainfall, E represents evaporation, S represents salinity, and Q represents heat transfer rate.
[0028] This method describes the hydrodynamic model and takes into account the impact of factors such as water flow and wind on the unmanned boat, making the control strategy more comprehensive and able to adapt to complex water environments.
[0029] Optionally, obtaining control input variables according to the three-dimensional simulation scene includes: collecting corresponding environmental parameters through the sensor module in the three-dimensional simulation scene; judging whether the collected environmental parameters are complete; if the collected environmental parameters are incomplete, judging whether all sensors in the sensor module are associated; if they are all associated, using interpolation to supplement the missing environmental parameters; and using the environmental parameters and the task initialization parameters in the target control task as control input variables.
[0030] In this manner, the missing environmental parameters are supplemented by interpolation to ensure the integrity and accuracy of the control input, thereby preventing the simulation test from being unable to run due to missing environmental parameters.
[0031] In a third aspect, the present invention provides a real-time simulation machine comprising a processor and a communication unit, wherein the processor is used to execute a semi-physical simulation test method for a multi-cascade fuzzy controller of a surface unmanned boat as described in any one of the second aspects of the present invention, and the communication unit is used to realize a communication connection between the processor and a shore-based server, and to realize a communication connection between the processor and the multi-cascade fuzzy controller of a surface unmanned boat.
[0032] In a fourth aspect, the present invention provides a semi-physical simulation test system for a multi-cascade fuzzy controller of a surface unmanned boat, characterized in that it includes a shore-based server, a multi-cascade fuzzy controller for a surface unmanned boat as described in any one of the first aspects of the present invention, and a real-time simulator as described in the third aspect of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0034] Figure 1 is a schematic diagram of the structure of a multi-cascade fuzzy controller for an unmanned surface boat according to an embodiment of the present invention;
[0035] Figure 2 Schematic diagram of the working principle of the multi-cascade fuzzy controller of the unmanned surface boat according to an embodiment of the present invention;
[0036] Figure 3 It is a flow chart of a semi-physical simulation test method of a multi-cascade fuzzy controller of an unmanned surface boat according to an embodiment of the present invention;
[0037] Figure 4 is a schematic diagram of a flow chart of obtaining control input variables according to an embodiment of the present invention;
[0038] Figure 5 It is a module block diagram of a three-dimensional simulation scene according to an embodiment of the present invention;
[0039] Figure 6 It is a structural schematic diagram of a semi-physical simulation test system of a multi-cascade fuzzy controller of an unmanned surface boat according to an embodiment of the present invention. DETAILED DESCRIPTION
[0040] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0041] Faced with the complex and changeable marine environment and under-actuated control systems, if the controller is designed accordingly for each situation, it will inevitably bring a huge amount of calculation. The embodiment of the present invention adopts a multi-cascade fuzzy control surface unmanned boat controller, which has the advantages of small amount of calculation, easy deployment, and not limited to the surface unmanned boat control model. It can also be simulated in real time in a real-time simulator according to the actual variable sea conditions to test the performance of the controller. The fuzzy logic control system contained in it can continuously approximate nonlinear functions, thereby eliminating the disadvantages of modeling nonlinear models one by one. The multi-cascade fuzzy controller of the surface unmanned boat is highly applicable and can be used for various types of under-actuated unmanned boats to perform single target tracking tasks under various sea conditions.
[0042] In addition, the embodiment of the present invention adopts semi-physical simulation technology, which effectively solves the problem that the digital simulation environment is not real and the physical environment is not easy to test. With the help of a real-time simulator, efficient semi-physical simulation testing of multi-cascade fuzzy controllers of surface unmanned boats can be achieved. By simulating actual control tasks and environments, the performance of the controller can be tested without actually launching the boat, saving costs, reducing risks, and enabling rapid iteration and optimization of control strategies.
[0043] In this embodiment, a multi-cascade fuzzy controller for an unmanned surface boat is provided. Figure 1 and Figure 2As shown, it includes: an input unit, which is used to determine the control input variables according to the target control task and classify the control input variables, and obtain the fuzzy input quantity after fuzzy processing on the classified control input variables; a sub-fuzzy controller, wherein the sub-fuzzy controllers are provided with a plurality of sub-fuzzy controllers, and the sub-fuzzy controllers are connected in series to form a multi-cascade fuzzy control chain, each of the sub-fuzzy controllers is provided with a corresponding fuzzy rule base, and the fuzzy input quantities of the same category are input into the corresponding sub-fuzzy controller, and the control output variables are obtained after calculation according to the membership function in the fuzzy rule base, and the control output variables of the current sub-fuzzy controller in the multi-cascade fuzzy control chain are used as the fuzzy input quantities of the sub-fuzzy controller of the next level; an output unit, which is used to output the control output variables of the sub-fuzzy controller of the last level in the multi-cascade fuzzy control chain.
[0044] Specifically, there can be multiple multi-cascade fuzzy control chains, and each multi-cascade fuzzy control chain is mainly composed of multiple sub-fuzzy controllers connected in series. Each sub-fuzzy controller has a fuzzy language input, output and fuzzy function library. In a specific task, the sub-fuzzy controller first needs to clarify the controlled input and output quantities one by one, and then cascade to form a multi-cascade fuzzy control chain.
[0045] In one example, the target control task is a surface unmanned boat target tracking task. Correspondingly, the categories of the control input variables include environmental parameters, azimuth difference and azimuth angle difference, the sub-fuzzy controller includes a first sub-fuzzy controller, a second sub-fuzzy controller and a third sub-fuzzy controller, after the fuzzy input corresponding to the environmental parameter is input into the third sub-fuzzy controller, the third sub-fuzzy controller calculates the environmental factor as the control output variable according to the membership function, the environmental factor and the azimuth difference after fuzzy processing are input into the first sub-fuzzy controller to obtain the first control quantity, the environmental factor and the azimuth angle difference after fuzzy processing are input into the second sub-fuzzy controller to obtain the second control quantity, and the fuzzy control quantity is generated according to the first control quantity and the second control quantity as the control output variable of the sub-fuzzy controller at the last level in the multi-cascade fuzzy control chain.
[0046] In the surface unmanned boat target tracking task, the actual control output variable is the bow azimuth angle ψ′ of the controlled unmanned boat. s and the propulsion force τ′ u According to the actual control output variables, it can be known that the target / controlled unmanned boat azimuth difference Δ(X, Y), the target / controlled unmanned boat azimuth angle difference Δψ, environmental parameters and other variables are used as fuzzy control input variables of the multi-cascade fuzzy controller of the surface unmanned boat.
[0047] The azimuth difference Δ(X, Y) and azimuth angle difference Δψ between the controlled unmanned boat and the target unmanned boat are used as the fuzzy control input variables of the first sub-fuzzy controller F1 and the second sub-fuzzy controller F2 respectively. The environmental parameters can be divided into two categories, namely direct environmental effect and indirect environmental effect, according to the description in the hydrodynamic model. The direct environmental effect category includes water speed [u w , v w , r w ], pressure P(x, y), density ρ, Coriolis force f c , wind resistance [f wx , f wy ] variables, which are divided into three categories according to water velocity, density pressure and external force as the fuzzy control input variables of the third sub-fuzzy controller F3. The control output variables of the third sub-fuzzy controller F3 are respectively combined with the azimuth difference Δ(X, Y) and the azimuth angle difference Δψ to form the control input variables of the first sub-fuzzy controller F1 and the second sub-fuzzy controller F2. The first sub-fuzzy controller F1 and the second sub-fuzzy controller F2 are respectively connected in series with the third sub-fuzzy controller F3 and output (ψ′ s , τ′ u ).
[0048] For each sub-fuzzy controller, the fuzzy function library establishment process is as follows:
[0049] Each type of control input variable or control output variable will form an odd number of fuzzy sets after fuzzy processing, such as {negative large (NB), negative small (NS), zero (O), positive small (PS), positive large (PB)}. At the same time, if the control input variable is continuous in its own domain, such as water speed, external force, azimuth difference, etc., it is also necessary to normalize it to its own fuzzy domain [-1, 1] according to the upper and lower limits of its own domain, so that a continuous fuzzy input quantity matching the discrete input quantity can be obtained.
[0050] Taking five fuzzy sets I as an example, I = {negative large (NB), negative small (NS), zero (O), positive small (PS), positive large (PB)} correspond to [-1, -0.67), [-0.67, -0.33), [-0.33, 0.33], (0.33, 0.67], (0.67, 1). After normalization according to the predetermined upper and lower limits of each input quantity, we get: water speed [u w , v w , r w ]=[0.8,-0.2,0]=[PB,O,O], pressure density [P(x,y),ρ]=[0.43,-0.2]=[PS,O], external force [f c , f wx , F wy]=[0.4,-0.8,0.6]=[PS,NB,PS]. The three types of environmental fuzzy control input variables will output an environmental factor according to the membership function through the third sub-fuzzy controller F3. The membership function in each sub-fuzzy controller is segmented using a trapezoidal function, and an "if-then" rule is added to ensure that all fuzzy input-output pairs are covered. For example, if the water speed is [u w , V w , r w ]=[PB,O,O],pressure density[P(x,y),ρ]=[PS,O],external force[f c , f wx , f wy ]=[PS,NB,PS], then The other input control variables are fuzzy processed in the same way.
[0051] Getting environmental factors back, The fuzzy inputs of the first sub-fuzzy controller F1 and the second sub-fuzzy controller F2 are formed with the azimuth difference Δ(X, Y) and the azimuth angle difference Δψ, and the fuzzy control quantity (ψ′) is output according to the membership function library rules set in the first sub-fuzzy controller F1 and the second sub-fuzzy controller F2. s , τ′ u ) is transmitted to the semi-physical simulation machine to complete a control cycle.
[0052] By taking environmental parameters, azimuth difference and azimuth angle difference as control input variables and cascading multiple sub-fuzzy controllers, special processing can be performed on different types of inputs, thereby improving the pertinence and effectiveness of control, realizing the control of the controlled unmanned boat in the target tracking task, and reducing the computational complexity of each sub-fuzzy controller.
[0053] The multi-cascade fuzzy controller of the surface unmanned boat of the embodiment of the present invention forms a multi-cascade fuzzy control chain by connecting several sub-fuzzy controllers in series. Each sub-fuzzy controller is provided with a corresponding fuzzy rule base. Fuzzy input quantities of the same category are input into the corresponding sub-fuzzy controller. The control output variable is obtained after calculation according to the membership function in the fuzzy rule base. The control output variable of the current sub-fuzzy controller in the multi-cascade fuzzy control chain is used as the fuzzy input quantity of the sub-fuzzy controller of the next level, so as to realize multi-cascade fuzzy control, and the control input variable can be processed more finely, and the control accuracy and adaptability can be improved. Each sub-fuzzy controller corresponds to a fuzzy rule base. The fuzzy logic control system contained in the membership function in the fuzzy rule base can continuously approximate nonlinear functions, and does not need to model nonlinear models one by one, thereby reducing the design calculation amount, and has the advantages of easy deployment and not being limited to the surface unmanned boat control model.
[0054] According to an embodiment of the present invention, an embodiment of a semi-physical simulation test method for a multi-cascade fuzzy controller of a surface unmanned boat is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0055] In this embodiment, a semi-physical simulation test method for a multi-cascade fuzzy controller of a surface unmanned boat is provided, which is used to perform a semi-physical simulation test on the multi-cascade fuzzy controller of a surface unmanned boat in the above embodiment. The semi-physical simulation test method for the multi-cascade fuzzy controller of a surface unmanned boat can be executed in a real-time simulator, which has a multi-core processor and an I / O board card such as cPCI / PCIe / PXIe / PXI. Figure 3 As shown, the process includes the following steps:
[0056] Step S101, receiving a target control task sent by a shore-based server, and establishing a dynamics-based three-dimensional simulation scene according to the target control task.
[0057] Specifically, the shore-based server sends the target control task to the real-time simulator for testing. The real-time simulator restores the task scene through the target control task requirements, including establishing a seawater model based on hydrodynamics, an unmanned boat model based on dynamics, etc., and establishing corresponding sensor modules, thereby establishing a three-dimensional simulation scene based on dynamics. The real-time simulator introduces the Unreal Engine to render the three-dimensional water scene.
[0058] In one example, the target control task is a target tracking test task, the unmanned boat model of the three-dimensional simulation scene includes a target unmanned boat and a controlled unmanned boat, and the output of the multi-cascade fuzzy controller of the surface unmanned boat is used to control the movement of the controlled unmanned boat and track the target unmanned boat.
[0059] The shore-based server issues target tracking test tasks, including target position and dynamics, surface unmanned boat position and dynamics, environmental conditions, tracking forms, etc. The semi-physical simulation test content is closely centered around the real surface unmanned boat conditions.
[0060] Step S102, obtaining control input variables according to the three-dimensional simulation scene, and sending the control input variables to the multi-cascade fuzzy controller of the surface unmanned vehicle through a communication link.
[0061] Specifically, in the three-dimensional simulation scene, a model is established through sensors to collect corresponding environmental parameters as one of the control input variables. In addition, initial position parameters such as the target / controlled unmanned boat azimuth difference and the target / controlled unmanned boat azimuth angle difference can be obtained as control input variables based on the positions of the current controlled unmanned boat model and the target unmanned boat in the three-dimensional simulation scene.
[0062] Step S103, receiving the control output variable generated based on the control input variable and returned by the multi-cascade fuzzy controller of the unmanned surface vehicle.
[0063] Specifically, in the multi-cascade fuzzy controller of the surface unmanned boat, after each type of input is classified, the same type of control input variables will be attributed to a sub-fuzzy controller. Each sub-fuzzy controller designs a fuzzy rule base and independently outputs control output variables according to actual needs. The sub-fuzzy controllers do not interfere with each other, and uniformly output the control input variables to the real-time simulator after passing through the last layer of sub-fuzzy controllers for real-time interaction. The real-time simulator thus receives the control output variables generated based on the control input variables returned by the multi-cascade fuzzy controller of the surface unmanned boat.
[0064] In one example, the control input variables include environmental parameters, azimuth difference and azimuth angle difference, and the control output variables include the bow azimuth angle ψ′ of the controlled unmanned boat. s and the propulsion force τ′ u .
[0065] Step S104, updating the three-dimensional simulation scene according to the control output variable and reacquiring the control input variable, and performing the next round of simulation control according to the reacquired control input variable until the target control task is completed and the test result is obtained.
[0066] Specifically, after receiving the control output variables, the real-time simulation machine updates the position information of the unmanned boat model in the three-dimensional simulation scene based on the principle of dynamics, and reacquires the control input variables, and performs the next round of simulation control according to the reacquired control input variables until the target control task is completed and the test results are obtained.
[0067] Step S105, sending the test result to the shore-based server.
[0068] Specifically, in combination with the specific mission scenario, the real-time simulator and the multi-cascade fuzzy controller of the surface unmanned boat are connected through a communication link to transmit control input variables and control output variables to each other. When the entire test task is completed, the real-time simulator returns the test results to the shore-based server.
[0069] The semi-physical simulation test method of the multi-cascade fuzzy controller of the surface unmanned boat of the embodiment of the present invention receives the target control task issued by the shore-based server, establishes a three-dimensional simulation scene based on dynamics according to the target control task; obtains the control input variable according to the three-dimensional simulation scene, and sends the control input variable to the multi-cascade fuzzy controller of the surface unmanned boat through the communication link; receives the control output variable generated based on the control input variable returned by the multi-cascade fuzzy controller of the surface unmanned boat; updates the three-dimensional simulation scene according to the control output variable and re-acquires the control input variable, performs the next round of simulation control according to the re-acquired control input variable until the target control task is completed and the test result is obtained; sends the test result to the shore-based server, and performs a real communication connection with the shore-based server and the multi-cascade fuzzy controller of the surface unmanned boat during the simulation process, so as to simulate the communication delay, data packet loss and the like in the actual working condition in real time, and the environmental elements of the scene can be supplemented by a series of methods through the three-dimensional simulation scene based on dynamics, which is closer to various weather / hydrological environments that may be encountered in the actual simulation process, so that the simulation test environment is closer to the actual working condition and the test result is more accurate.
[0070] In one embodiment, if Figure 5 As shown, the three-dimensional simulation scene includes an unmanned boat dynamics model, a hydrodynamics model, a sensor module, a communication module and a signal conversion module.
[0071] Specifically, the unmanned boat dynamics model includes a three-dimensional model of the controlled unmanned boat and a three-dimensional model of the target unmanned boat, wherein the motion process of the unmanned boat is simulated by combining the corresponding kinematic equations. The kinematic equations in the unmanned boat dynamics model include the kinematic equations of the controlled unmanned boat, the kinematic equations of the target unmanned boat and the dynamic equations of the controlled unmanned boat.
[0072] Among them, since the research object is the surface unmanned boat, the unmanned boat in the earth coordinate system [X E , Y E ] plane motion. The controlled surface unmanned boat is at [x s ,y s ], heading angle ψ s , the mass and moment of inertia coefficient of the unmanned boat are [m u , m v , m r ]. Assume that the input force and moment of the controlled unmanned boat in the surge and bow roll control direction are [τ u , τ r ], the angular velocities of surge, drift and pitch are [u s , v s , r s ]. The kinematic equation of the controlled unmanned boat is:
[0073]
[0074] In the formula, x s and s Respectively represent the coordinates of the controlled unmanned boat in the x direction and the y direction, and Respectively represent x s and s The linear velocity obtained by derivation, u s 、v s and r s They represent the angular velocities of the controlled unmanned boat, ψ s represents the heading angle of the controlled unmanned boat, Represents ψ s Take the derivative and get the angular velocity.
[0075] The target unmanned boat is in the same coordinate system [x t ,y t ] at. Heading angle ψ t , the angular velocities of surge, drift and pitch are [u t , v t , r t ]. The kinematic equation of the target surface unmanned boat is described as follows:
[0076]
[0077] In the formula, x t and t Respectively represent the coordinates of the target unmanned boat in the x direction and the y direction, and Respectively represent x t and t The linear velocity obtained by derivation, u t 、v t and r t They represent the angular velocities of the target unmanned boat, ψ t represents the heading angle of the target unmanned boat, Represents ψ t Take the derivative and get the angular velocity.
[0078] Based on the physical quantities and hydrodynamic resistance of the controlled unmanned boat [F u (u s , v s , r s ), F v (u s , v s , r s ), Fr (u s , v s , r s )] and fluid dynamics, the dynamic equation of the controlled unmanned boat can be obtained as:
[0079]
[0080] In the formula, F u (u s , v s , r s ), F v (u s , v s , r s ) and F r (u s , v s , r s ) represent the axial forces in three directions in space, m u 、m v and m r represents the mass and moment of inertia coefficient of the controlled unmanned boat, τ u and τ r They respectively represent the input force and moment in the surge and pitch control directions of the controlled unmanned boat, and Respectively represent u t 、v t and r t Take the derivative and get the angular velocity.
[0081] This method provides the kinematic equations of the unmanned boat and the target unmanned boat and the dynamic equations of the controlled unmanned boat. When the real-time simulator receives the control output variables returned by the multi-cascade fuzzy controller of the surface unmanned boat, it can update the position information of the unmanned boat and the target unmanned boat in the three-dimensional simulation scene based on the control output variables, combined with the kinematic equations of the unmanned boat and the target unmanned boat and the dynamic equations of the controlled unmanned boat, providing a detailed physical basis for the simulation and ensuring the accuracy of the simulation results.
[0082] Furthermore, the motion equations describing the conservation of momentum of viscous incompressible fluids are the Navier-Stokes equations. In combination with the actual working environment of surface unmanned boats, the water velocity in space [u w , V w , r w ], pressure P(x, y), Coriolis force f c , wind resistance [f wx , f wy ], temperature T, salinity S(x t ,y t , z t), density ρ, evaporation E, rainfall P w A series of environmental factors that may affect the motion control of the surface unmanned boat are considered, and a hydrodynamic model is established as a random ocean dynamics model. The dynamic equation of the hydrodynamic model is described as:
[0083]
[0084] In the formula, u w 、v w and r w Indicates the water speed in three directions in space, and Respectively represent u w and v w The acceleration obtained by derivation, P represents pressure, ρ represents density, f c represents the Coriolis force, f wx and f wy They represent the wind resistance in the x and y directions, g represents the acceleration of gravity, T represents the temperature, and P w represents rainfall, E represents evaporation, S represents salinity, and Q represents heat transfer rate.
[0085] By describing the hydrodynamic model, the influence of factors such as water flow and wind force on the unmanned boat is taken into account, making the control strategy more comprehensive and able to adapt to complex water environments.
[0086] After the unmanned boat dynamics model and hydrodynamic model are established, in order to restore the real unmanned boat mission scene, it is also necessary to establish models for a series of sensors configured on it, such as inertial measurement unit, differential GPS, compass, speed sensor, visual sensor, temperature and humidity salt fog sensor, etc. These sensor modules are simulated and integrated into the corresponding positions on the unmanned boat to receive or collect environmental signals in the three-dimensional simulation scene. In order to restore the working conditions of the sensors in the real physical environment, Gaussian noise will also be randomly added to the sensors. Under actual working conditions, if the collected signal is missing or delayed in part during the transmission process, the missing sensor signal will be supplemented by predicting it.
[0087] Furthermore, the established three-dimensional simulation scene also includes a communication module, which can be divided into two parts, the in-boat communication module and the boat-shore communication module. The in-boat communication module is mainly reflected in the wired or wireless communication between the sensors in the boat and the controller. The boat-shore communication module not only needs to monitor the signals from the sensors in the boat in real time during the test, but also needs to locate the working position of the unmanned boat in real time through wireless communication. In the semi-physical simulation test, the communication module also needs to introduce actual conditions such as time delay and signal loss to simulate the actual working conditions.
[0088] Furthermore, the established three-dimensional simulation scene also includes a signal conversion module. The signal conversion module mainly serves voltage signal conversion and data signal conversion. The voltage required for each sensor and drive system in the unmanned boat comes from the DC power module carried, so a DC voltage conversion module is required to convert it into the working voltage required by other modules. The data signal conversion module includes the mutual conversion between digital signals and analog signals. The analog signal collected by the sensor module in the three-dimensional simulation scene needs to be converted into a digital signal before it can be transmitted to the processor for high-speed calculation.
[0089] Dynamic simulation is achieved through the unmanned boat dynamics model, hydrodynamics model, sensor module, communication module and signal conversion module. The real environment can be simulated through the sensor module, communication module and signal conversion module, making the simulation closer to the actual situation and improving the reliability and practicability of the simulation.
[0090] In one embodiment, if Figure 4 As shown, in step S102, obtaining control input variables according to the three-dimensional simulation scene includes:
[0091] Step S1021, collecting corresponding environmental parameters through the sensor module in the three-dimensional simulation scene;
[0092] Step S1022, determining whether the collected environmental parameters are complete. If the collected environmental parameters are incomplete, determining whether all sensors in the sensor module are associated. If all sensors are associated, interpolation is used to supplement the missing environmental parameters.
[0093] Step S1023: Using the environmental parameters and the task initialization parameters in the target control task as control input variables.
[0094] Specifically, the dynamics of the surface unmanned boat, environmental parameters, and target tasks are obtained from the three-dimensional simulation scene established by the real-time simulator. If the relevant environmental parameters of the perception layer are missing, they are supplemented by interpolation to meet the input of the multi-cascade fuzzy controller of the surface unmanned boat, otherwise the real-time simulator will enter the system self-check and end this test task. When the total input and output conditions of the multi-cascade fuzzy controller of the surface unmanned boat are met, the sub-fuzzy controller is subdivided according to the task and the corresponding rule base is formulated.
[0095] The embodiment of the present invention ensures the integrity and accuracy of the control input by supplementing the missing environmental parameters through the interpolation method, and avoids the failure of the simulation test to run due to the missing environmental parameters.
[0096] The semi-physical simulation test method of a multi-cascade fuzzy controller of an unmanned surface boat according to an embodiment of the present invention has the following advantages:
[0097] 1. The surface unmanned boat itself carries limited resources. If a large amount of resources are spent on its own control, it will be difficult to complete other complex tasks. By simulating the real unmanned boat model and the changes in the ocean environment in the semi-physical simulator, the control performance of the multi-cascade fuzzy controller of the surface unmanned boat can be obtained. The semi-physical simulation technology is used to effectively solve the problems of the unreality of the digital simulation environment and the difficulty of testing the physical environment. With the help of the real-time simulator to build a three-dimensional simulation scene based on dynamics and interact with the multi-cascade fuzzy controller of the surface unmanned boat, efficient semi-physical testing of the surface unmanned boat controller can be achieved, reducing the money and time costs of making real unmanned boats;
[0098] 2. Dynamic simulation is realized through the unmanned boat dynamics model, hydrodynamics model, sensor module, communication module and signal conversion module. The real environment can be simulated through the sensor module, communication module and signal conversion module, making the simulation closer to the actual situation and improving the reliability and practicality of the simulation;
[0099] 3. During the simulation process, the real-time simulator and the shore-based server as well as the multi-cascade fuzzy controller of the surface unmanned boat establish real communication connections, which can simulate communication delays, data packet loss and other situations in actual working conditions in real time.
[0100] An embodiment of the present invention also provides a real-time simulation machine, including a processor and a communication unit, wherein the processor is used to execute a semi-physical simulation test method of a multi-cascade fuzzy controller of a surface unmanned boat as in the above-mentioned embodiment, and the communication unit is used to realize a communication connection between the processor and a shore-based server, and to realize a communication connection between the processor and the multi-cascade fuzzy controller of a surface unmanned boat.
[0101] The processor may be a central processing unit, a network processor or a combination thereof. The processor may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.
[0102] The embodiment of the present invention also provides a semi-physical simulation test system for a multi-cascade fuzzy controller of an unmanned surface boat, such as Figure 6 As shown, it includes a shore-based server, a multi-cascade fuzzy controller for an unmanned surface boat as in the above-mentioned embodiment, and a real-time simulator.
[0103] The working process of the semi-physical simulation test system of the multi-cascade fuzzy controller of the surface unmanned boat is as follows: First, the test task is sent to the real-time simulator through the shore-based server, and the task scene is restored in the real-time simulator through the test task requirements. Secondly, combined with the specific task scene, the real-time simulator and the multi-cascade fuzzy controller of the surface unmanned boat are connected through a communication link to transmit control input and output signals to each other. When the entire test task is completed, the real-time simulator returns the test results to the server.
[0104] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.
Claims
1. A multi-cascade fuzzy controller for an unmanned surface boat, characterized in that: include: An input unit, used for determining control input variables according to a target control task and classifying the control input variables, and performing fuzzy processing on the classified control input variables to obtain fuzzy input quantities; A sub-fuzzy controller, wherein a plurality of sub-fuzzy controllers are provided, and the sub-fuzzy controllers are connected in series to form a multi-cascade fuzzy control chain, each of the sub-fuzzy controllers is provided with a corresponding fuzzy rule base, and the fuzzy input quantities of the same category are input into the corresponding sub-fuzzy controller, and a control output variable is obtained after calculation according to the membership function in the fuzzy rule base, and the control output variable of the current sub-fuzzy controller in the multi-cascade fuzzy control chain is used as the fuzzy input quantity of the sub-fuzzy controller of the next level; The output unit is used to output the control output variable of the sub-fuzzy controller at the last stage in the multi-cascade fuzzy control chain.
2. The multi-cascade fuzzy controller for unmanned surface boat according to claim 1, characterized in that: The categories of the control input variables include environmental parameters, azimuth differences and azimuth angle differences. The sub-fuzzy controller includes a first sub-fuzzy controller, a second sub-fuzzy controller and a third sub-fuzzy controller. After the fuzzy input quantity corresponding to the environmental parameter is input into the third sub-fuzzy controller, the third sub-fuzzy controller obtains the environmental factor as the control output variable after calculation according to the membership function. The environmental factor and the azimuth angle difference after fuzzy processing are input into the first sub-fuzzy controller to obtain the first control quantity. The environmental factor and the azimuth angle difference after fuzzy processing are input into the second sub-fuzzy controller to obtain the second control quantity. A fuzzy control quantity is generated according to the first control quantity and the second control quantity as the control output variable of the sub-fuzzy controller at the last level in the multi-cascade fuzzy control chain.
3. A semi-physical simulation test method for a multi-cascade fuzzy controller of an unmanned surface boat, characterized in that: Used to perform a semi-physical simulation test on the multi-cascade fuzzy controller of the unmanned surface boat according to any one of claims 1 to 2, the semi-physical simulation test method comprising: Receiving a target control task sent by a shore-based server, and establishing a dynamics-based three-dimensional simulation scene according to the target control task; Acquire control input variables according to the three-dimensional simulation scene, and send the control input variables to the multi-cascade fuzzy controller of the surface unmanned boat through a communication link; Receiving a control output variable generated based on the control input variable and returned by the multi-cascade fuzzy controller of the unmanned surface boat; updating the three-dimensional simulation scene according to the control output variable and reacquiring the control input variable, and performing the next round of simulation control according to the reacquired control input variable until the target control task is completed and the test result is obtained; The test results are sent to the shore-based server.
4. The semi-physical simulation test method of the multi-cascade fuzzy controller of the unmanned surface boat according to claim 3 is characterized in that: The three-dimensional simulation scene includes an unmanned boat dynamics model, a hydrodynamics model, a sensor module, a communication module and a signal conversion module.
5. The semi-physical simulation test method of the multi-cascade fuzzy controller of the unmanned surface boat according to claim 4 is characterized in that: The unmanned boat dynamics model includes the kinematic equation of the controlled unmanned boat, the kinematic equation of the target unmanned boat and the dynamic equation of the controlled unmanned boat.
6. The semi-physical simulation test method of the multi-cascade fuzzy controller of the unmanned surface boat according to claim 5 is characterized in that: The kinematic equation of the controlled unmanned boat is: In the formula, x s and s Respectively represent the coordinates of the controlled unmanned boat in the x direction and the y direction, and Respectively represent x s and s The linear velocity obtained by derivation, u s 、v s and r s They represent the angular velocities of the controlled unmanned boat, ψ s represents the heading angle of the controlled unmanned boat, Represents ψ s The angular velocity obtained by taking the derivative; The kinematic equation of the target unmanned boat is: In the formula, x t and t Respectively represent the coordinates of the target unmanned boat in the x direction and the y direction, and Respectively represent x t and t The linear velocity obtained by derivation, u t 、v t and r t They represent the angular velocities of the target unmanned boat, ψ t represents the heading angle of the target unmanned boat, Represents ψ t The angular velocity obtained by taking the derivative; The dynamic equation of the controlled unmanned boat is: In the formula, F u (u s , v s , r s ), F v (u s , v s , r s ) and F r (u s , v s , r s ) represent the axial forces in three directions in space, m u 、m v and m r represents the mass and moment of inertia coefficient of the controlled unmanned boat, τ u and τ r They respectively represent the input force and moment in the surge and pitch control directions of the controlled unmanned boat, and Respectively represent u t 、v t and r t Take the derivative and get the angular velocity.
7. The semi-physical simulation test method of the multi-cascade fuzzy controller of the unmanned surface boat according to claim 4 is characterized in that: The equations of the hydrodynamic model are described as: In the formula, u w 、v w and r w Indicates the water speed in three directions in space, and Respectively represent u w and v w The acceleration obtained by derivation, P represents pressure, ρ represents density, f c represents the Coriolis force, f wx and f wy They represent the wind resistance in the x and y directions, g represents the acceleration of gravity, T represents the temperature, and P w represents rainfall, E represents evaporation, S represents salinity, and Q represents heat transfer rate.
8. The semi-physical simulation test method of the multi-cascade fuzzy controller of the unmanned surface boat according to claim 4 is characterized in that: Acquiring control input variables according to the three-dimensional simulation scene includes: Collecting corresponding environmental parameters through the sensor module in the three-dimensional simulation scene; Determine whether the collected environmental parameters are complete. If the collected environmental parameters are incomplete, determine whether all sensors in the sensor module are associated. If all sensors are associated, use interpolation to supplement the missing environmental parameters. The environmental parameters and the task initialization parameters in the target control task are used as control input variables.
9. A real-time simulation machine, characterized in that: It includes a processor and a communication unit, the processor is used to execute the semi-physical simulation test method of the multi-cascade fuzzy controller of the surface unmanned boat as described in any one of claims 3 to 8, and the communication unit is used to realize the communication connection between the processor and the shore-based server, and realize the communication connection between the processor and the multi-cascade fuzzy controller of the surface unmanned boat.
10. A semi-physical simulation test system for multi-cascade fuzzy controllers of unmanned surface boats, characterized in that: It comprises a shore-based server, a multi-cascade fuzzy controller for an unmanned surface boat as claimed in any one of claims 1 to 2, and a real-time simulator as claimed in claim 9.
