Intelligent construction-oriented tower crane obstacle avoidance trajectory planning and self-adaptive tracking control method

Abstract

The invention discloses a tower crane obstacle avoidance trajectory planning and self-adaptive tracking control method for intelligent construction, and relates to the field of mechanical system dynamics and control, an optimization problem is established based on a constraint equation by taking minimization of transportation time as a target, and the optimization problem is solved to obtain optimal transportation time and obstacle occurrence time; constructing a data set based on the target position of the tower crane, the obstacle position, the optimal transportation time and the obstacle appearance time; obtaining a trained fuzzy neural network based on the data set; and utilizing the trained fuzzy neural network to predict the optimal transportation time and the obstacle appearance time, and obtaining the reference trajectory of the tower crane based on the optimal transportation time and the obstacle appearance time. And a self-adaptive tracking controller with error constraints is designed based on the reference trajectory. Obstacle avoidance and tracking error constraints are considered, a dichotomy and a fuzzy neural network are combined, and an optimal time trajectory considering the position and the size of an obstacle in tower crane transportation is planned.

Description

Technical Field

[0001] The present invention relates to the field of mechanical system dynamics and control, and in particular to a tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art.

[0003] With the popularity of smart building scenarios, tower cranes have become one of the most important construction machinery. Since tower cranes usually work in complex outdoor environments, obstacles on the transportation path are inevitable, and it is crucial to achieve collision-free and efficient operation of tower cranes. On the other hand, as a typical under-actuated system, tower cranes exhibit distinct dynamic characteristics such as nonlinearity, strong coupling, and under-actuation. Therefore, in order to achieve safe and efficient transportation, it is challenging and necessary to propose an efficient tracking controller.

[0004] In the past few decades or so, many methods have been proposed to achieve the goals of precise positioning and anti-sway, including adaptive methods, sliding mode control, fuzzy control, backstepping control, etc. Because neural networks have powerful uncertainty processing capabilities, they have attracted great attention from scholars. For example, an adaptive controller based on radial basis function neural network (RBFNN) is proposed to handle uncertain parameters, external disturbances and nonlinearity. Some authors have also introduced a Bayesian regularized artificial neural network (BRANN) that dynamically updates the shaper parameters based on the current system parameters. In some studies, it is shown that control methods that consider constraints can usually obtain better dynamic performance. For example, some authors designed auxiliary terms to strengthen the coupling between the underdriven state and the driven state. On this basis, a controller with underdriven state constraints was further proposed to solve the positioning and anti-sway control problems. In other literatures, tracking control with state constraints was designed to solve the motion control problem of tower cranes under actual physical constraints. Although the above methods can bring good control performance to tower crane systems, there are few studies on obstacle avoidance. Some authors pre-planned the load path by considering the translation speed of the trolley and the rotation speed of the cantilever, and then designed an obstacle avoidance controller. Some authors have also proposed optimal time trajectory planning methods for double-swing bridge cranes and slewing crane systems. In the above works, obstacles are usually approximated as single points, and their volume is ignored. In the real world, the volume of obstacles is difficult to ignore, especially in complex external construction scenes. Moreover, the volume of obstacles will greatly affect the trajectory planning results and tracking control performance. Summary of the invention

[0005] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction. In order to effectively deal with the obstacle avoidance problem, new auxiliary terms are introduced to deal with the drive and under-actuated state variables. On this basis, considering the obstacle avoidance and tracking error constraints, an improved state equation is established, and the dichotomy method is combined with a fuzzy neural network (FNN) to plan the optimal time trajectory considering the position and volume of obstacles in tower crane transportation.

[0006] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: In a first aspect, the present invention provides a tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction, comprising: Obtain the relevant variables of the tower crane's trolley, cantilever and load, and establish a dynamic model of the tower crane; Establishing a state constraint taking into account obstacle avoidance based on the state variables of the dynamic model to obtain a state constraint equation; An optimization problem is established based on the constraint equation with the goal of minimizing the transportation time, and the optimization problem is solved to obtain the optimal transportation time and the obstacle appearance time; a data set is constructed based on the target position of the tower crane, the obstacle position, the optimal transportation time and the obstacle appearance time; Establishing a fuzzy neural network and training it based on the data set to obtain a trained fuzzy neural network; The target position and obstacle position of the tower crane to be planned are input into the trained fuzzy neural network to predict the optimal transportation time and obstacle appearance time, and the reference trajectory of the tower crane is obtained based on the optimal transportation time and obstacle appearance time.

[0007] In a further technical solution, the kinetic model is expressed as:

[0008] In the formula, represents the state variable, represents the control input variable, represents the friction or disturbance variable; represents the cantilever rotation angle, represents the displacement of the trolley, Indicates the radial swivel angle of the load, represents the load tangential swing angle, represents the cantilever rotation inertia, represents the trolley's kinetic force, represents the friction torque, represents the friction force, represents the air resistance parameter, Indicates the length of the rope.

[0009] In a further technical solution, the constraint equation is expressed as: ,

[0010] ,

[0011] ,

[0012] In the formula, represents the cantilever rotation angle at time t, , They represent the maximum velocity and acceleration of the cantilever rotation angle respectively; represents the displacement of the trolley at time t, , They represent the maximum velocity and acceleration of the car’s displacement respectively; , They represent the radial swing angle and tangential swing angle of the load at time t, respectively. , They represent the maximum amplitude of the load radial swing angle and the maximum amplitude of the load tangential swing angle respectively.

[0013] According to a further technical solution, the optimization problem is to minimize the transportation time based on the constraint equation.

[0014] A further technical solution is to use a dichotomy method to solve the optimization problem.

[0015] According to a further technical solution, the fuzzy neural network includes an input layer, a fuzzification layer, an inference layer, a normalization layer and an output layer which are connected in sequence.

[0016] A further technical solution is to design an adaptive tracking controller with error constraints based on the reference trajectory, which is expressed as:

[0017]

[0018] In the formula, represents the cantilever rotation moment; , , and indicates positive gain; represents the tracking error of the cantilever; Indicates the cantilever tracking error range; and Represents state variables separated from parameters; Indicates the trolley's kinetic force; Indicates the tracking error of the car; Indicates the tracking error range of the car; and is a positive parameter; and Denote the uncertain parameter vectors and An estimate of ; T represents the vector transpose.

[0019] A further technical solution is to use an adaptive law to estimate the uncertain parameter vector.

[0020] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction as described in the first aspect.

[0021] In a third aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction as described in the first aspect are implemented.

[0022] One or more of the above technical solutions have the following beneficial effects: This paper constructs an improved trajectory planning method for collision avoidance for the first time. In order to effectively deal with the obstacle avoidance problem, new auxiliary terms are introduced to deal with the driving and underactuated state variables. On this basis, the improved state equation is established by considering the obstacle avoidance and tracking error constraints. Then the dichotomy method is combined with the fuzzy neural network (FNN) to plan the optimal time trajectory considering the position and volume of obstacles in tower crane transportation. Not only accurate positioning and swing angle elimination are achieved, but also collision with obstacles in the motion path is avoided.

[0023] The present invention designs an adaptive tracking controller that considers obstacle avoidance trajectory planning, takes into account obstacle avoidance trajectory planning and tracking error constraints, and ensures that the obstacle avoidance trajectory reaches the target position. In order to ensure the tracking performance, a pre-set safety margin is given, and an adaptive law is used to estimate uncertain / unknown parameters.

[0024] The invention theoretically proves the stability of the system through Lyapunov technology and LaSalle invariant set principle. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0026] Figure 1 is a tower crane model with obstacles in an embodiment of the present invention; Figure 2 is a top view of the obstacle avoidance trajectory designed in an embodiment of the present invention; Figure 3 It is a structural diagram of a fuzzy neural network FNN in an embodiment of the present invention; Figure 4 It is a flowchart of a controller in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0028] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0029] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0030] Embodiment 1 like Figure 1 As shown, this embodiment discloses a tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction, which includes the following steps: S1: Obtain the relevant variables of the tower crane's trolley, cantilever and load, and establish a dynamic model of the tower crane; In this embodiment, a method for obstacle avoidance trajectory planning is proposed for a four-degree-of-freedom tower crane.

[0031] Based on the Lagrangian modeling technology, the dynamic equation of the four-degree-of-freedom tower crane is expressed in matrix-vector form as follows: (1) In the formula, represents the state variable, represents the control input variable, represents the friction / disturbance variable.

[0032] like Figure 1As shown in the figure, the 4-DOF tower crane model includes a cantilever, on which a moving trolley is arranged, on which a load is fixed by ropes, and in the transportation path of the load there are obstacles of a certain volume. The dynamic model of the 4-DOF tower crane can be further expressed as: (2) (3) (4) (5) Friction torque and friction The expression is defined as: (6) (7) In the formula, , , , , , is a parameter related to friction.

[0033] Table 1 Parameters and variables

[0034] S2: establishing a state constraint taking into account obstacle avoidance based on the state variables of the dynamic model to obtain a state constraint equation; In this embodiment, in order to facilitate the trajectory planning process, some new auxiliary terms are designed to represent the state variables of the tower crane system. Figure 1 The spatial geometric relationship in the coordinate system can be expressed as: (8) (9) Furthermore, the load coordinates can be calculated as: (10) (11) Considering the constraints on the swing angle within a small range, equations (10)-(11) can be further rewritten as (when the swing angle is very small, equations (10)-(11) satisfy the approximate relationship: , , , ): (12) (13) The second-order derivative of equations (12)-(13) can be expressed as: (14) (15) Further processing of equations (14)-(15) yields: (16) (17) According to equations (4), (5), (16), and (17), we can obtain the following relationship: (18) (19) On the other hand, from equations (12)-(13), we can calculate the variable , The expression is: (20) (twenty one) Then, multiply equations (18)-(19) by , , and calculate the elimination with equations (9), (20)-(21), The expression is: (twenty two) In the formula, .

[0035] Substitute equation (22) into equation (9): Expression, we get: (twenty three) Then, substituting equation (22) into equations (20)-(21), we obtain: (twenty four) (25) Formulas (22)-(25) use all state variables , , These auxiliary terms are used for the expressions of state variable constraints and reference trajectories, which are crucial for solving the optimization problem.

[0036] The following is a detailed description of the tower crane's obstacle avoidance trajectory planning: In order to achieve precise positioning and sway elimination, the constraint equation of the actual motion physical signal is set, which is expressed as follows: , , , (26) , , , (27) , , , (28) In the formula, The higher-order derivatives representing the position of the load, =1,2,3,4,5; represents the high-order derivative of the position of the car, =1,2,3; Load starting position (x-axis position), (y-axis position); target position , ; The y-axis starting and target positions of the car , The calculation expression is: , , , , , .

[0037] To ensure safety and traceability, state variables should be restricted to appropriate scopes, namely: ,

[0038] ,

[0039] , (29) In the formula, represents the cantilever rotation angle at time t; represents the first derivative of the cantilever rotation angle at time t; represents the second derivative of the cantilever rotation angle at time t; represents the displacement of the trolley at time t; , They represent the radial swing angle of the load and the tangential swing angle of the load at time t respectively; , They represent the maximum velocity and acceleration of the cantilever rotation angle respectively; , They represent the maximum velocity and acceleration of the car’s displacement respectively; , Represents the load swing angle , The maximum upper limit of the change, which respectively represents the maximum amplitude of the load radial swing angle and the maximum amplitude of the load tangential swing angle.

[0040] To satisfy equation (29), we further design an optimal trajectory with the minimum transportation time. Next, the auxiliary signal , , , will be parameterized by a suitable function. , The reference trajectory uses an 11th-order polynomial curve, and the 7th-order polynomial curve is selected as the expected y-axis trajectory of the car. The expression is as follows: (30) In the formula, , , ={1,…,11}, , ={1,…,9} are parameters to be determined. represents the time-dependent normalized parameter, Indicates the shipping time.

[0041] The n-order derivative of formula (30) is as follows: (31) In the formula, represents the higher-order derivative of the load, =1,2,3,4,5; represents the higher-order derivative of the car, =1,2,3.

[0042] Substituting equations (26)-(28) into equations (30)-(31), we can obtain the parameters: , ,

[0043] , , ,

[0044] , , , , , (32) At this point, the expanded expressions of all constraints in equation (29) can be expressed by equations (22)-(25) and (30)-(32). Substituting equations (30)-(32) into equations (22)-(25), the constraint equations can be expressed by auxiliary terms , , express.

[0045] S3: establishing an optimization problem based on the constraint equation with the goal of minimizing the transportation time, solving the optimization problem to obtain the optimal transportation time and the obstacle appearance time; constructing a data set based on the target position of the tower crane, the obstacle position, the optimal transportation time and the obstacle appearance time; In this embodiment, the optimization problem is established based on the state constraint equation as follows: a series of constraint problems are transformed into solutions The minimum value of and Respectively The upper and lower limits of Represents the maximum error. The optimization problem is: based on the state constraint equation, minimize the transportation time.

[0046] Then, the optimization problem is solved by using the bisection method to obtain the minimum value of the transportation time. The specific optimization problem is: Based on formula (29), minimize the transportation time That is to say, solving equation (29) yields the minimum value of the transportation time that satisfies all inequality conditions.

[0047] Next, assume that the obstacle is Always appear on the motion path. To avoid collision, The load should reach the lower right corner of the safety range at all times. , , The obstacle avoidance trajectory is designed as: (33) (34) (35) In the formula, , , is the length of the obstacle, is the width of the obstacle. is the solution of: (36) Substituting equations (33)-(35) into equations (22)-(23), we can obtain , Final trajectory.

[0048] The optimization problem is solved by using the dichotomy method to obtain the solution result, that is, the minimum value of the transportation time (optimal transportation time), and the appearance time of the obstacle on the load transportation path is obtained based on the obstacle position. The dichotomy solution results (optimal transportation time, obstacle appearance time), the target position of the tower crane (trolley target position, cantilever target position), and the obstacle position (x-axis and y-axis position of the obstacle) are collected to construct a data set as the training data of the fuzzy neural network.

[0049] S4: establishing a fuzzy neural network and training it based on the data set to obtain a trained fuzzy neural network; The traditional trajectory planning method needs to calculate the transportation time first, which is cumbersome and time-consuming. Therefore, in this embodiment, the fuzzy neural network FNN is incorporated into the design of the obstacle avoidance trajectory to improve the traditional planning method. After training based on the collected data set, the time parameters (the minimum value of the transportation time, the time when the obstacle appears) can be predicted by FNN, and the reference trajectory can be calculated based on the time parameters, which only requires the position parameters and no calculation is required.

[0050] , The datasets are computed and collected in S3. Figure 3 As shown, by using the training data, FNN , , , and output variables , FNN combines the advantages of fuzzy logic control reasoning and neural network fault tolerance. It uses the nonlinear mapping and self-learning ability of neural networks to adapt to fuzzy logic. FNN uses optimization algorithms to adjust the weights and membership functions of fuzzy systems. Figure 3 As shown in Figure 1, the structure of FNN consists of five layers, including the input layer, fuzzification layer, inference layer, normalization layer and output layer connected in sequence. The composition and function of each layer are described as follows: Layer 1: Input layer of the fuzzy neural network.

[0051] In the input layer, there are four nodes: the target location of the load , , the center point of the obstacle , In this embodiment, the input ranges of the four nodes are , , , .

[0052] Layer 2: Fuzzification layer.

[0053] Gaussian function is used as membership function. Each input variable has 4 membership functions, i.e. 4 fuzzy linguistic sets, and the 4 membership functions are PL (positive large), PS (positive small), NS (negative small) and NL (negative large). This layer consists of 16 nodes.

[0054] , = =1,2,3,4 (37) In the formula, , Represents output nodes and input nodes, , Represent the centroid and width of the membership function respectively.

[0055] Layer 3: Fuzzy reasoning layer.

[0056] This layer consists of 256 nodes, each node represents a fuzzy rule.

[0057] (38) In the formula, Represents the output of the third layer.

[0058] Layer 4: Normalization layer of the fuzzy neural network.

[0059] The nodes of the fuzzy inference layer are normalized in this layer, and the number of nodes in this layer is the same as that in the third layer.

[0060] (39) Layer 5: Output layer of the fuzzy neural network (defuzzification layer).

[0061] Defuzzification using weighted average method .

[0062] (40) In the formula, Represents the connection weight parameter between the fourth and fifth layers.

[0063] After optimization, update the centroid of the iterative membership function and width , and the connection weight coefficient .

[0064] As shown in Table 2, eight groups of experiments were selected by the present invention for testing, and the transportation time predicted by FNN in the eight groups of experiments was given. and obstacle appearance time The present invention uses four evaluation indicators (i.e., mean absolute error MAE, root mean square error RMSE, mean absolute percentage error MAPE and symmetric mean absolute percentage error SMAPE) to evaluate the prediction performance of FNN; Table 3 shows the prediction results of the present invention and the real data (transportation time , the actual obstacle appearance time ) The MAE, RMSE, MAPE and SMAPE error values. By using the results of the four error evaluation methods, MAE, RMSE, MAPE and SMAPE, it can be concluded that the error of FNN prediction is very small, the MAPE and SMAPE indicators are less than 2%, the MAE and RMSE indicators are less than 0.1, and the prediction accuracy is high.

[0065] Table 2 FNN prediction results

[0066] Table 3 Model evaluation indicators

[0067] S5: Input the target position and obstacle position of the tower crane to be planned into the trained fuzzy neural network, predict the optimal transportation time and obstacle appearance time, and obtain the reference trajectory of the tower crane based on the optimal transportation time and obstacle appearance time.

[0068] In this embodiment, the position parameters are input into the trained fuzzy neural network to predict the corresponding time parameters. Specifically, the target position and obstacle position of the tower crane to be planned are input into the trained fuzzy neural network to predict the time parameters, namely the optimal transportation time and the obstacle appearance time.

[0069] Based on the optimal transportation time and obstacle appearance time , Substituting equations (33)-(35) into equations (22)-(23), we can obtain the reference trajectory of the tower crane, that is, the reference trajectory of the cantilever rotation angle and the reference trajectory of the car displacement .

[0070] In this embodiment, if Figure 4 As shown in Figure 2, in order to accurately track the reference trajectory obtained above, an adaptive tracking controller with error constraints is also designed for the four-DOF tower crane.

[0071] For system (1) (i.e., formula (1), formula (1) represents a system), design a suitable controller to achieve the following goals: (41) (42) In the formula, Indicates the expected transportation time, which can be calculated by S5. and The reference trajectory is obtained based on equations (22)-(23) and (33)-(35).

[0072] During the entire control process, the tracking errors of the cantilever and the trolley are within a small range: , (43) In the formula, and respectively Represents the cantilever and trolley tracking error: (44) (45) In the formula, and Respectively represent and the border.

[0073] In order to ensure the tracking performance, the following conditions need to be met: the reference trajectory of the cantilever and the car should be When the initial position to In addition, all velocity and acceleration signals of the boom and the trolley are and Therefore, the following conditions hold: , , , , ,

[0074] , , ,

[0075] , , , , =1,2,3, when (46) The trajectory-tracking controller design steps are as follows: Similar to the structure of the system energy function, the error-based scalar function is defined as follows: (47) In the formula, yes The error vector.

[0076] The time derivative of can be written as: (48) The last two items are and After elimination, formula (48) can be expressed as follows: (49) In the formula, yes The expected vector of .

[0077] By substituting formula (1) into formula (49) and calculating, the following results are obtained: (50) Where:

[0078]

[0079]

[0080]

[0081] In the formula, , , Defined as: ,

[0082]

[0083] and The expression is as follows:

[0084]

[0085] Therefore, the controller of this embodiment is designed as follows: (51) (52) In the formula, , , and indicates positive gain; and is a positive parameter; and Denote the uncertain parameter vectors and The estimated value of ; T represents the vector transpose; and Represents state variables separated from parameters. Use the following explicit expression:

[0086]

[0087] , The adaptive law is as follows: , (53) In the formula, and represents a nonnegative definite diagonal matrix.

[0088] The following is a stability analysis of the controller designed above: Theorem 1: The controller in this embodiment can meet the following control objectives: (54) Proof: In order to prove the stability of the tower crane system, the Lyapunov method is used to construct the Lyapunov function, which is defined as follows: (55) In the formula, and represents a nonnegative definite diagonal matrix.

[0089] Estimation Error and Defined as: , (56) The time derivative of formula (55) is as follows: (57) Substituting formula (48) into formula (57), we get: (58) , ,when (59) Based on this, when , formula (58) becomes: (60) Therefore, formula (60) can derive the following results: (61) Then, the following results can be obtained: (62) Using formula (47) and , We can further get: (63) Then, according to equations (51)-(53) and (2)-(7), the following results can be obtained: (64) According to formula (46), . Assume the following situation: the car tracking error Towards the border ,Right now ,but This is consistent with the convergence of equations (55) and (61). Therefore, this situation does not exist. Then we have: , (65) right Carry out corresponding analysis, namely: , (66) In the formula, Indicates a tendency to Indicates not tending to.

[0090] Subsequently, the LaSalle invariant set principle will be applied to summarize the stability analysis. To further complete the analysis of asymptotic stability, the following invariant sets will be introduced: (67) Then, The maximum invariant set in is defined as . Combining formula (60), we find that when When is 0, each part of the expression is 0, resulting in the following result: (68) Calculate the first-order derivative expression and the second-order derivative expression of equation (44) and equation (45) respectively, and combine the conclusion of equation (68) and equation (46) , , , Substitute the obtained first-order derivative expression and second-order derivative expression into the obtained first-order derivative expression and draw the conclusion: , , , (69) Substituting equations (6)-(7) and (68)-(69) into equations (2)-(3) and performing elimination calculations, we obtain the following results: , (70) Considering , In the explicit expressions in equations (51)-(52), substituting equations (46) and (68)-(70) into equations (51)-(52), and performing elimination, we obtain: , (71) Substituting equations (68)-(69) into equations (4)-(5), we can deduce the following results: , (72) Then, it can be concluded that: , (73) It is obvious from equations (71) and (73) that all equilibrium points of the system are located in the maximum invariant set Therefore, by adopting the LaSalle invariant set theorem, it can be concluded that the error signal and the swing angle converge to zero asymptotically, that is: , , , . (74) In summary, for the obstacle avoidance problem, the present invention designs an optimal trajectory and adaptive controller for a 4-DOF tower crane. Specifically, a set of auxiliary signals are carefully designed to represent the state variables. Then a series of physical constraints are considered and the optimal problem is established. Based on the dichotomy method, the optimal problem can be solved and the optimal trajectory is designed by FNN. In particular, the tracking error is limited to a set of a priori boundaries, and the control method is theoretically analyzed using the Lyapunov method and the LaSalle invariant set theorem. In general, the innovative method of the present invention not only achieves precise positioning and swing angle elimination, but also avoids collisions with obstacles in the motion path.

[0091] Embodiment 2 This embodiment provides a tower crane obstacle avoidance trajectory planning and adaptive tracking control method system for intelligent construction, including: A model building module is configured to: obtain relevant variables of a trolley, a cantilever and a load of the tower crane, and establish a dynamic model of the tower crane; A data set construction module is configured to: establish a state constraint taking obstacle avoidance into consideration based on the state variables of the dynamic model to obtain a state constraint equation; establish an optimization problem based on the constraint equation with the goal of minimizing the transportation time, solve the optimization problem to obtain an optimal transportation time and an obstacle appearance time; construct a data set based on a target position of a tower crane, an obstacle position, the optimal transportation time and an obstacle appearance time; A network building module is configured to: establish a fuzzy neural network and train it based on the data set to obtain a trained fuzzy neural network; The network prediction module is configured to: input the target position and obstacle position of the tower crane to be planned into the trained fuzzy neural network, predict its optimal transportation time and obstacle appearance time, and obtain the reference trajectory of the tower crane based on the optimal transportation time and obstacle appearance time.

[0092] In this embodiment, the system further includes a tracking controller design module, which is configured to: design an adaptive tracking controller with error constraints based on the reference trajectory.

[0093] Embodiment 3 The purpose of this embodiment is to provide a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method of embodiment 1 when executing the program.

[0094] Embodiment 4 The purpose of this embodiment is to provide a computer-readable storage medium, a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of the method of embodiment 1 are performed.

[0095] The steps involved in the apparatus of the above embodiments 3 and 4 correspond to the method embodiment 1, and the specific implementation method can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0096] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0097] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0098] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction, characterized in that: include: Obtain the relevant variables of the tower crane's trolley, cantilever and load, and establish a dynamic model of the tower crane; Establishing a state constraint taking into account obstacle avoidance based on the state variables of the dynamic model to obtain a state constraint equation; Based on the constraint equation, an optimization problem is established with the goal of minimizing the transportation time, and the optimization problem is solved to obtain the optimal transportation time and obstacle appearance time; constructing a data set based on a target position of the tower crane, an obstacle position, the optimal transportation time, and an obstacle appearance time; Establishing a fuzzy neural network and training it based on the data set to obtain a trained fuzzy neural network; The target position and obstacle position of the tower crane to be planned are input into the trained fuzzy neural network to predict the optimal transportation time and obstacle appearance time, and the reference trajectory of the tower crane is obtained based on the optimal transportation time and obstacle appearance time.

2. The tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction according to claim 1, characterized in that: The kinetic model is expressed as: In the formula, represents the state variable, represents the control input variable, represents the friction or disturbance variable; represents the cantilever rotation angle, represents the displacement of the trolley, Indicates the radial swivel angle of the load, represents the load tangential swing angle, represents the cantilever rotation inertia, represents the trolley's kinetic force, represents the friction torque, represents the friction force, represents the air resistance parameter, Indicates the length of the rope.

3. The tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction according to claim 1, characterized in that: The constraint equation is expressed as: ， ， ， In the formula, represents the cantilever rotation angle at time t, , They represent the maximum velocity and acceleration of the cantilever rotation angle respectively; represents the displacement of the trolley at time t, , They represent the maximum velocity and acceleration of the car’s displacement respectively; , They represent the radial swing angle and tangential swing angle of the load at time t, respectively. , They represent the maximum amplitude of the load radial swing angle and the maximum amplitude of the load tangential swing angle respectively.

4. The tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction according to claim 1, characterized in that: The optimization problem is to minimize the transportation time based on the constraint equation.

5. The tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction according to claim 4, characterized in that: The optimization problem is solved by using a binary search method.

6. The tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction according to claim 1, characterized in that: The fuzzy neural network includes an input layer, a fuzzification layer, an inference layer, a normalization layer and an output layer which are connected in sequence.

7. The tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction according to claim 1, characterized in that: Based on the reference trajectory, an adaptive tracking controller with error constraints is also designed, which is expressed as: In the formula, represents the cantilever rotation moment; , , and indicates positive gain; represents the tracking error of the cantilever; Indicates the cantilever tracking error range; and Represents state variables separated from parameters; Indicates the trolley's kinetic force; Indicates the tracking error of the car; Indicates the tracking error range of the car; and is a positive parameter; and Denote the uncertain parameter vectors and An estimate of ; T represents the vector transpose.

8. The tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction according to claim 7, characterized in that: An adaptive law is used to estimate the uncertain parameter vector.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction as described in any one of claims 1-8 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the tower crane obstacle avoidance trajectory planning and adaptive tracking control method for intelligent construction as described in any one of claims 1-8 are implemented.

Images