A hanging basket movement control method and system based on running state recognition

Abstract

The application discloses a hanging basket motion control method and system based on operation state recognition, and particularly relates to the technical field of bridge construction automation, and is used for solving the problem of poor synchronous control of the hanging basket movement; the application accurately identifies the working condition through multi-source sensing data fusion and a finite state machine, and realizes adaptive setting of a servo system parameter based on working condition parameter mapping; a model prediction control algorithm fusing pose feedback and wind load feedforward is used to generate a disturbance-resistant motion trajectory; walking synchronization and platform stability are realized through cross-coupling control and attitude closed-loop fine adjustment; an independent safety guardian thread provides emergency response, and a double closed loop of control and protection is formed, so that the motion precision, disturbance resistance and intrinsic safety level of the hanging basket under variable load and strong disturbance working conditions are effectively improved, and the energy consumption and mechanical vibration frequency in the motion process of the hanging basket are reduced.

Description

Technical Field

[0001] This invention relates to the field of bridge construction automation technology, and more specifically, to a method and system for controlling the motion of a hanging basket based on operational status recognition. Background Technology

[0002] In the cantilever construction of large bridges, the walking positioning and posture stability of the formwork are core control aspects. Existing technical solutions focus on optimizing the synchronization accuracy and position servo control of the traveling motor. Through sensor feedback and PID control algorithms, hydraulic or mechanical actuators are driven to achieve the preset travel distance of the formwork on the track. Such systems generally adopt a combination of programmed segmented control and manual experience intervention to ensure walking safety. The technical implementation involves the electric drive and automatic control of engineering machinery.

[0003] However, existing control methods have limitations in dynamic adaptability. There is an inherent contradiction between the preset nature of control commands and the time-varying nature of construction site conditions. Commands issued by the controller based on fixed parameters are difficult to respond in real time to complex dynamic effects caused by sudden load changes, wind disturbances, and track surface changes. This leads to asynchronous, swaying, and even out-of-tolerance operation of the hanging basket. This rigidity of the control strategy makes it impossible for the system to maintain a high-precision and high-stability motion posture in a changing construction environment, thus creating potential risks to structural safety and construction quality. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a hanging basket motion control method and system based on operation status recognition, so as to solve the problem of poor synchronous control of hanging basket movement in the above-mentioned background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A hanging basket motion control method based on operational status recognition includes the following steps:

[0007] Based on the multi-source sensor terminal deployed on the hanging basket body, the structural stress, spatial pose, platform tilt angle and environmental wind load data are collected simultaneously, and timestamp alignment and data encapsulation are performed to construct a unified state vector.

[0008] Based on preset stress thresholds and tilt angle thresholds, finite state machine reasoning is performed on the state vector to determine the core operating condition of the hanging basket.

[0009] Based on the constructed working condition parameter mapping table, the core parameters of the servo drive and motion controller are adaptively tuned according to the identified core operating conditions.

[0010] Based on the tuned core parameters, the system integrates spatial pose feedback and environmental wind load feedforward, and uses a model predictive control algorithm to execute dynamic trajectory planning with mechanical and attitude constraints. It generates disturbance-resistant motion commands and drives the dual-side walking motors to execute them in coordination through cross-coupling control and attitude closed-loop fine-tuning to control walking synchronization and platform stability.

[0011] An independent safety protection thread runs, which monitors and judges critical stress and abnormal tilt angle in real time based on a pre-set rule base, and triggers the response with the highest priority, forming a safety protection closed loop in parallel with the main control loop.

[0012] In a preferred embodiment, based on a multi-source sensor terminal deployed on the hanging basket itself, structural stress, spatial pose, platform tilt angle, and environmental wind load data are simultaneously collected. Timestamp alignment and data encapsulation are performed to construct a unified state vector. The specific process is as follows:

[0013] Assign hardware synchronization signals to GNSS receivers, encoders, resistance strain gauges, tilt sensors and ultrasonic anemometers, or add a unified timestamp to all asynchronously acquired data frames at the software level;

[0014] Based on a unified timestamp, data from different sensors are cached and interpolated within a set fusion period to ensure that all data correspond to the same physical time.

[0015] The aligned data is encapsulated according to a predefined order and data structure to form a unified state vector containing time stamps.

[0016] In a preferred embodiment, finite state machine reasoning is performed on the state vector based on preset stress thresholds and tilt angle thresholds to determine the core operating condition of the hanging basket. The specific process is as follows:

[0017] Define the set of states of the finite state machine, which includes at least the no-load steady state, the heavy-load steady state, and the dynamic disturbance state triggered by strong wind or off-center load.

[0018] Based on a unified state vector, the sliding average and instantaneous peak values ​​of structural stress, the absolute value of platform tilt angle, and the sliding average value of ambient wind speed are calculated in each control cycle.

[0019] When the average sliding stress of the structure is lower than the first stress threshold and the absolute value of the platform tilt angle is lower than the first tilt angle threshold, the state is determined to be the no-load stable state.

[0020] When the average sliding stress of the structure is higher than the first stress threshold but lower than the second stress threshold, and the absolute value of the platform tilt angle is lower than the second tilt angle threshold, the state is determined to be a heavy load stable state.

[0021] When the average sliding wind speed exceeds the wind speed threshold, or when the weighted composite index of the instantaneous peak value of structural stress and the absolute value of the platform tilt angle exceeds the dynamic stability threshold, the system is forced to migrate to the dynamic disturbance state.

[0022] In a preferred embodiment, based on the constructed operating condition parameter mapping table, the core parameters of the servo drive and motion controller are adaptively tuned according to the identified core operating conditions. The specific process is as follows:

[0023] The operating condition parameter mapping table is a pre-set lookup table. The row index of the operating condition parameter mapping table is the operating condition inferred from the finite state machine, the column index is the core parameter of the controller to be tuned, and the table entry content is the optimal parameter value obtained through offline optimization or learning from historical data.

[0024] When the operating condition shifts to a heavy-load steady state, the corresponding parameter combination in the mapping table is automatically called. The parameter combination includes increasing the torque limit value of the servo drive, increasing the proportional gain of the position loop, and decreasing the integral time constant of the speed loop.

[0025] When the operating condition shifts to dynamic disturbance, the corresponding parameter combination in the mapping table is automatically invoked. The parameter combination includes enabling the strong smoothing mode of the trajectory filter, reducing the prediction time domain and control time domain of the model predictive controller, and shortening the control cycle of attitude closed-loop fine-tuning.

[0026] In a preferred embodiment, a model predictive control algorithm is used to perform dynamic trajectory planning with mechanical and attitude constraints, and the specific process is as follows:

[0027] A discrete state-space model containing the mass, damping, and stiffness of the hanging basket is constructed as the prediction model, with spatial pose as the state variable and motor driving force as the control variable.

[0028] In each control cycle, based on the current state vector, an optimal control problem in a finite time domain is solved by a predictive model in a rolling manner.

[0029] Mechanical and attitude constraints are applied as hard constraints for the optimal control problem, including that the structural stress calculated based on the prediction model must not exceed the allowable stress of the material in the prediction time domain, and the absolute value of the predicted platform lateral tilt angle must not exceed the anti-overturning safety threshold.

[0030] Solve the constrained quadratic programming problem and output the first control command sequence as the disturbance-resistant motion command.

[0031] In a preferred embodiment, the environmental wind load data is incorporated as a feedforward compensation quantity into the model predictive control algorithm in the following specific manner:

[0032] Based on real-time wind speed and direction, the equivalent wind disturbance force is calculated by the wind load coefficient and the windward area of ​​the hanging basket. The disturbance force is then directly introduced into the objective function of the prediction model and the optimal control problem as a known external disturbance input.

[0033] In a preferred embodiment, the dual-side walking motors are executed in coordination through cross-coupling control and attitude closed-loop fine-tuning, as follows:

[0034] The difference between the left and right walking displacements is calculated in real time and defined as the synchronization error.

[0035] The synchronization error is input into a synchronization compensator, which at least includes a proportional control element.

[0036] The outputs of the synchronous compensator are superimposed on the outputs of the position loop controllers of the left and right motors respectively with opposite polarities to form cross feedback;

[0037] The lateral tilt angle signal measured by the high-precision dual-axis tilt sensor is input into an independent proportional-derivative controller, and the output of the proportional-derivative controller is used as a speed correction amount, which is superimposed on the speed commands of the left and right travel motors with opposite signs.

[0038] In a preferred embodiment, a separate safety daemon thread is run to monitor and determine critical stress and abnormal tilt angle in real time based on a pre-set rule base. The specific process is as follows:

[0039] The security daemon thread runs independently of the main control loop and with higher priority. The security daemon thread's condition rule base consists of a series of production rules in the form of IF-THEN.

[0040] The rule conditions section makes logical judgments on the instantaneous values ​​of structural stress, platform tilt angle, and their short-term trends;

[0041] The rule-based action section includes generating safety protection commands at different levels, which at least include deceleration, stopping, and emergency stop.

[0042] When the instantaneous value of structural stress exceeds the stress safety threshold, or the instantaneous value of platform tilt angle exceeds the tilt angle safety threshold, the highest priority rule is triggered, an emergency stop command is immediately generated and executed, and an interrupt signal is sent to the main controller.

[0043] When the standard deviation of structural stress or platform tilt angle continues to increase over several consecutive cycles and exceeds the dynamic stability threshold, a secondary rule is triggered to generate a deceleration or shutdown command and record an early warning log.

[0044] In a preferred embodiment, in areas where positioning signals are limited, path calibration and position alignment are performed, specifically as follows:

[0045] In areas where GNSS signals are blocked, a path calibration mechanism is activated to construct a displacement fingerprint sequence based on mileage reference markers deployed beside the travel track and encoder readings on the basket itself.

[0046] The collected inertial measurement unit data and the odometer marker sequence observed along the route are used to perform path matching and error minimization fitting to complete the loop correction and positioning trajectory correction of the hanging basket's running path.

[0047] A hanging basket motion control system based on operational status recognition, used to implement the aforementioned hanging basket motion control method based on operational status recognition, includes:

[0048] The state perception fusion module is used to simultaneously collect structural stress, spatial pose, platform tilt angle and environmental wind load data based on the multi-source sensor terminal deployed on the hanging basket body, perform timestamp alignment and data encapsulation, and construct a unified state vector.

[0049] The operating condition identification module is used to perform finite state machine reasoning on the state vector based on preset stress thresholds and tilt angle thresholds to determine the core operating condition of the hanging basket.

[0050] The parameter adaptive module is used to adaptively tune the core parameters of the servo drive and motion controller based on the constructed working condition parameter mapping table and the identified core operating conditions.

[0051] The disturbance rejection motion control module is used to perform dynamic trajectory planning with mechanical and attitude constraints based on the core parameters after tuning, integrating spatial pose feedback and environmental wind load feedforward, and using model predictive control algorithm to generate disturbance rejection motion commands. It also drives the dual-side walking motors to perform the commands in coordination through cross-coupling control and attitude closed-loop fine-tuning to achieve walking synchronization and platform stability.

[0052] The closed-loop adjustment module is used to run an independent safety guardian thread, which monitors and judges critical stress and abnormal tilt angle in real time based on a preset rule base, and triggers the response with the highest priority, forming a safety protection closed loop in parallel with the main control loop.

[0053] The technical effects and advantages of this invention are as follows:

[0054] This invention significantly improves the intelligence and safety performance of the hanging basket in complex construction environments by constructing an integrated closed loop of operational status perception, parameter adaptive tuning, and disturbance-resistant motion control. Firstly, based on multi-source sensor data fusion and finite state machine reasoning, it accurately identifies core operating conditions such as no-load stability, heavy-load stability, and dynamic disturbances, effectively overcoming the control mismatch problem caused by misjudgment of operating conditions in traditional control. Secondly, through a pre-set operating condition parameter mapping table and an online self-tuning mechanism, it dynamically adjusts the servo system control parameters, enabling the system to maintain optimal dynamic characteristics under different load and disturbance conditions. Furthermore, it combines... The model predictive control algorithm, which integrates spatial pose feedback and environmental wind load feedforward, generates a smooth, disturbance-resistant trajectory while strictly meeting mechanical and attitude constraints. Then, through cross-coupled control and attitude closed-loop fine-tuning, it achieves high-precision synchronization of the dual-side walking motors and platform attitude stability, effectively suppressing structural fluctuations caused by off-center loading and wind load. Finally, through an independently running safety guardian thread, it provides the highest level of safety assurance through multiple redundant monitoring and millisecond-level response mechanisms. This ensures accurate positioning and stable operation of the hanging basket while significantly reducing energy consumption and mechanical vibration, achieving a simultaneous improvement in construction efficiency and safety reliability. Attached Figure Description

[0055] Figure 1 This is a flowchart of a hanging basket motion control method based on running state recognition according to the present invention.

[0056] Figure 2 This is a schematic diagram of the structure of a hanging basket motion control system based on operation status recognition according to the present invention. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] Example 1: As Figure 1 As shown, a hanging basket motion control method based on running state recognition includes the following steps:

[0059] Based on the multi-source sensor terminal deployed on the hanging basket body, the structural stress, spatial pose, platform tilt angle and environmental wind load data are collected simultaneously, and timestamp alignment and data encapsulation are performed to construct a unified state vector.

[0060] Based on preset stress thresholds and tilt angle thresholds, finite state machine reasoning is performed on the state vector to determine the core operating condition of the hanging basket.

[0061] Based on the constructed working condition parameter mapping table, the core parameters of the servo drive and motion controller are adaptively tuned according to the identified core operating conditions.

[0062] Based on the tuned core parameters, the system integrates spatial pose feedback and environmental wind load feedforward, and uses a model predictive control algorithm to execute dynamic trajectory planning with mechanical and attitude constraints. It generates disturbance-resistant motion commands and drives the dual-side walking motors to execute them in coordination through cross-coupling control and attitude closed-loop fine-tuning to control walking synchronization and platform stability.

[0063] An independent safety protection thread runs, which monitors and judges critical stress and abnormal tilt angle in real time based on a pre-set rule base, and triggers the response with the highest priority, forming a safety protection closed loop in parallel with the main control loop.

[0064] Step 1: Based on the multi-source sensor terminals deployed on the hanging basket itself, simultaneously collect structural stress, spatial pose, platform tilt angle, and environmental wind load data, perform timestamp alignment and data encapsulation, and construct a unified state vector. The specific implementation is as follows:

[0065] When the hanging basket is ready to be moved in the bridge construction section, the multi-source sensing terminal deployed on its body starts to work. The structural stress data is measured by resistance strain gauges welded to key stress points of the main beam of the hanging basket, such as the root of the main truss and the anchor point of the front sling. These strain gauges form a Wheatstone bridge circuit, and the weak voltage signal output by the strain gauge is amplified by the signal conditioning amplifier to reflect the real-time stress of the hanging basket structure.

[0066] Spatial pose data is provided by a GNSS receiver (such as a GPS / BeiDou dual-mode receiver that supports RTK positioning) and an inertial measurement unit installed at the center of the top of the hanging basket.

[0067] The GNSS receiver provides absolute position information in latitude, longitude and elevation, while the inertial measurement unit measures angular velocity and linear acceleration through its internal gyroscope and accelerometer. The two are fused using a Kalman filter algorithm to finally output complete pose information with high precision and high update rate, including three-dimensional coordinates, heading angle, pitch angle and roll angle.

[0068] The platform tilt data is directly measured by a high-precision dual-axis tilt sensor installed on the front working platform of the hanging basket. The sensor outputs the longitudinal tilt angle (forward and backward pitch) and lateral tilt angle (left and right tilt) of the platform relative to the horizontal plane. The environmental wind load data is continuously measured by an ultrasonic anemometer installed at the highest point of the hanging basket. This device directly outputs the instantaneous wind speed and wind direction angle by calculating the propagation time difference of ultrasonic waves on the downwind and upwind paths.

[0069] To overcome the timing chaos caused by different sampling rates and communication delays of different sensors, timestamp alignment is performed. First, hardware synchronization is achieved by the system's main controller generating a precise, periodic synchronization pulse signal (e.g., a square wave pulse with a frequency of 10Hz), which is then distributed in parallel to devices that support external synchronization functions, such as walking motor encoders, resistance strain gauge data acquisition modules, tilt sensors, and ultrasonic anemometers, through a dedicated synchronization cable. This forces all devices to use this pulse as a reference for synchronous sampling.

[0070] Then, software synchronization is performed. In scenarios where wiring is difficult or the equipment does not support hardware synchronization, the system runs a high-precision clock service in the main control computer. This service is calibrated by receiving the PPS (pulses per second) signal from GNSS or the network NTP protocol. Based on this, every sensor data frame that arrives asynchronously via Ethernet or CAN bus, regardless of whether it comes from a strain gauge, inclinometer, or anemometer, is given a unified timestamp accurate to the millisecond level.

[0071] After data collection and timestamping are completed, the data fusion and vector construction stage begins, which involves setting a fixed data fusion period, such as 100 milliseconds.

[0072] At the beginning of each fusion cycle, all timestamped sensor data arriving in the buffer within the past 100 milliseconds are checked. For some slower-updating data that has not yet arrived by the end of the cycle, such as GNSS data with an update rate of only 20Hz, a linear interpolation algorithm is used for prediction and completion.

[0073] For example, if the previous GNSS location point is at time T0 and the next is at time T1, and the current data is needed at time T0+10ms, then an interpolated coordinate at time T0+10ms is calculated linearly according to the coordinate values ​​at time T0 and T1. Through this caching and interpolation mechanism, it is ensured that all the data used to construct the state vector at the end of each fusion cycle are strictly aligned in physical time.

[0074] Step 2: Based on preset stress and tilt angle thresholds, perform finite state machine reasoning on the state vector to determine the core operating condition of the hanging basket. Specifically, the implementation is as follows:

[0075] After the unified state vector is constructed, the operation condition reasoning based on the finite state machine is performed to determine the core working state of the hanging basket. First, in the system initialization phase, the set of states contained in the finite state machine needs to be clearly defined.

[0076] Define three core states to cover the main operating modes of the hanging basket:

[0077] The no-load stable state indicates that the concrete inside the hanging basket has been poured and reached the required strength. After the formwork is removed, the hanging basket is in a state of no construction load and stable operation.

[0078] Heavy load stable state means that the hanging formwork is carrying concrete pouring or bearing a large amount of steel bars and other materials, and is subjected to a large construction load, but the structural response and attitude are still within a safe and stable range.

[0079] Dynamic disturbance refers to a dynamic instability state in which the hanging basket experiences severe eccentric loading due to strong winds, asymmetrical box girder sections, uneven concrete pouring, or other reasons, resulting in drastic fluctuations in structural stress or significant tilting of the platform.

[0080] A series of preset judgment thresholds are used to achieve automatic transition between states. These judgment thresholds are preset and stored in the controller based on the hanging basket design specifications, historical safe operation data and engineering experience.

[0081] Key thresholds include: a first stress threshold (e.g., 50 MPa), used to distinguish between no-load and low-load conditions; a second stress threshold (e.g., 120 MPa), serving as a safe proportion of the material's allowable stress, used to distinguish between heavy load and overload risk; a first tilt angle threshold (e.g., 0.5 degrees), used for stability assessment under no-load conditions; a second tilt angle threshold (e.g., 1.5 degrees), used to allow for greater attitude changes under heavy load conditions; a wind speed threshold (e.g., 15 m / s), exceeding which indicates that the wind load effect is not negligible; and a dynamic stability threshold, an indicator used to comprehensively evaluate instantaneous impact and attitude stability.

[0082] Within each control cycle, e.g., 100 milliseconds, key performance indicators are calculated based on the latest unified state vector. Specifically, the moving average of structural stress is the arithmetic mean of stress values ​​at all key measuring points over the past 2 seconds, reflecting the sustained load level. The instantaneous peak value of structural stress is the maximum value selected from all measuring points within the same time window, used to capture sudden stress concentrations. The absolute value of the platform tilt angle is directly taken from the larger of the lateral and longitudinal tilt angles in the state vector, used to assess the overall tilt of the platform. The moving average of ambient wind speed is the average of wind speed data over the past 10 seconds, used to smooth out gust disturbances and reflect sustained wind intensity.

[0083] The logical decision for state transition is executed cyclically in each control cycle. When the calculated average sliding stress of the structure is lower than the preset first stress threshold and the absolute value of the platform tilt angle is lower than the preset first tilt angle threshold, the finite state machine will determine that the hanging basket is currently in an unloaded stable state. When the average sliding stress of the structure is higher than the first stress threshold but lower than the more stringent second stress threshold and the absolute value of the platform tilt angle is lower than the second tilt angle threshold, the hanging basket will be determined to enter a heavy-load stable state.

[0084] In addition, there are two conditions for forcibly transitioning a state to a dynamic perturbation state; the transition is triggered as long as either condition is met.

[0085] The first condition is: the calculated sliding average of the ambient wind speed exceeds the preset wind speed threshold.

[0086] The second condition is: the value of a weighted composite index exceeds the preset dynamic stability threshold.

[0087] The calculation process for the weighted composite index is as follows:

[0088] First, divide the instantaneous peak value of the structural stress obtained in the current cycle by a stress normalization coefficient, which can be taken as the second stress threshold, to obtain a dimensionless stress ratio. At the same time, divide the absolute value of the platform tilt angle by a tilt angle normalization coefficient, which can be taken as the second tilt angle threshold, to obtain a dimensionless tilt angle ratio.

[0089] Then, the stress ratio is multiplied by a stress weighting coefficient (e.g., 0.6), and the tilt ratio is multiplied by a tilt weighting coefficient (e.g., 0.4). Finally, the two weighted results are added together to obtain the weighted composite index. When the composite index is greater than 1, it indicates that the instantaneous anomaly has exceeded the stable range. Once the system determines that it has entered a dynamic disturbance state, it will continue to maintain this state until the calculated environmental wind speed sliding average value is lower than the wind speed threshold and the weighted composite index value is lower than the dynamic stability threshold. Only after a preset stability delay is confirmed will it migrate back to other stable states.

[0090] Step 3: Based on the constructed operating condition parameter mapping table, adaptively tune the core parameters of the servo drive and motion controller according to the identified core operating conditions. Specifically, this is implemented as follows:

[0091] After identifying the core operating conditions, the system enters the adaptive tuning stage of servo drive and motion controller parameters. This step is based on a pre-built operating condition parameter mapping table stored in the system memory. This operating condition parameter mapping table defines the optimal set of servo drive and motion controller parameters corresponding to each core operating condition. These parameters are determined by a combination of optimization through a large amount of field trial operation data, theoretical calculations and engineering experience, aiming to enable the hanging basket to obtain the best motion performance and stability under different operating conditions.

[0092] Specifically, the operating condition parameter mapping table should include at least the following key parameter settings: For servo drives, the main settings are the speed loop proportional gain and integral time constant; for motion controllers, the main settings are the position loop proportional gain, integral time constant, and derivative time constant. For example, the parameters configured for the no-load steady state in the mapping table are:

[0093] The servo driver speed loop uses a high proportional gain and a short integral time constant, while the motion controller position loop uses a high proportional gain and a moderate derivative time constant, in order to pursue fast and accurate positioning response.

[0094] The parameters configured for heavy-load steady state are:

[0095] The proportional gain and integral time constant of the servo driver's speed loop are appropriately reduced, and the proportional gain of the motion controller's position loop is also reduced accordingly. Simultaneously, the integral time constant is increased. The aim is to reduce system stiffness, avoid overshoot and oscillation under heavy inertial loads, and ensure smooth start-up and shutdown processes.

[0096] The parameters configured for dynamic disturbance are: the gain of all loops of the servo drive and motion controller are set to a low level, especially the integral action is significantly weakened, thereby sacrificing some response speed in exchange for extremely strong anti-interference robustness, effectively suppressing continuous fluctuations caused by wind load or off-center load.

[0097] The parameter tuning execution process is as follows: In each control cycle, after the finite state machine infers a new core operating condition, it will immediately query the parameter mapping table of that operating condition, read the corresponding set of preset parameter values, and then send these parameter values ​​to the specific servo drive and motion controller through a dedicated communication protocol, such as CANopen or EtherCAT, so that they can immediately run according to the new parameters.

[0098] Furthermore, for dynamic disturbances, the system introduces a more refined adaptive mechanism. Since the disturbance intensity under dynamic disturbance conditions is continuously changing, instead of simply using a set of fixed parameters in the mapping table, the system dynamically fine-tunes the parameters based on a baseline parameter defined by the mapping table, using the real-time calculated sliding average of the ambient wind speed and the weighted comprehensive index. For example, using the servo speed loop proportional gain preset for dynamic disturbances in the mapping table as the baseline value, the system automatically reduces the proportional gain by 5% for every five meters per second increase in the average wind speed. This continuous fine-tuning enables the control system to adapt to changes in the intensity of external disturbances more smoothly and accurately, achieving integrated control from macroscopic condition identification to microscopic parameter adaptation.

[0099] It should be noted that the construction of this operating condition parameter mapping table is not a simple assignment of empirical values, but rather based on a systematic parameter optimization process. Specifically, in the initial stage of basket debugging, for each defined core operating condition, the system will automatically execute a series of test trajectories containing different accelerations and velocities within a safe test range, such as under no-load conditions. During this process, senior engineers will observe key performance indicators such as overshoot, settling time, and number of oscillations based on the step response curve. They will manually or through the built-in self-tuning algorithm find a set of servo and controller parameters that make the system response both fast and stable, and permanently bind them to the operating condition, storing them in the mapping table. This process ensures that each set of parameters in the mapping table has been verified in practice and is the optimal performance solution for a specific operating condition.

[0100] In addition to manual tuning by senior engineers, the system can also incorporate built-in automated parameter tuning algorithms. For example, a self-tuning method based on step response can be used: a small step speed command is applied within the safe test range. Based on the measured response curve, characteristic parameters such as delay time and rise time are calculated. Using classic formulas such as Ziegler-Nichols or the built-in expert system, the initial recommended values ​​of the proportional gain and integral time constant of the speed loop are automatically calculated. Subsequently, a small step response test can be performed based on these recommended values. Based on the measured overshoot and settling time, further fine-tuning is performed using the gradient descent method until the step response meets the preset performance indicators (such as overshoot <5%, settling time <2 seconds). Finally, the optimized parameter set is stored in the operating condition parameter mapping table.

[0101] Through the above methods, the motion control system of the hanging basket is no longer fixed, but becomes an organic whole that can sense its own state and environmental changes and make intelligent adjustments accordingly, thus ensuring the safety and stability of the movement process under various complex construction conditions.

[0102] Step 4: Based on the tuned core parameters, spatial pose feedback and environmental wind load feedforward are integrated, and a model predictive control algorithm is used to execute dynamic trajectory planning with mechanical and attitude constraints. This generates disturbance-resistant motion commands, which are then executed collaboratively by the dual-side walking motors through cross-coupling control and attitude closed-loop fine-tuning to control walking synchronization and platform stability. The specific implementation is as follows:

[0103] After completing the adaptive tuning of the core parameters, the high-precision motion control execution stage is entered. This step uses a model predictive control algorithm to generate smooth motion commands that can actively resist disturbances. In specific implementation, an internal predictive model describing the dynamics and kinematics of the hanging basket walking is first established.

[0104] The core of a Model Predictive Controller (MPC) is an internal predictive model capable of predicting the future dynamics of a system. This internal predictive model takes the form of a discrete-time state-space equation, and its standard expression is: Where k represents the current sampling time, Let the system's state vector be the state variable. In the example of hanging basket longitudinal motion control, the state variables are defined as follows: ,in: Longitudinal displacement , Longitudinal velocity , Platform pitch angle , Platform pitch angular velocity ; The control input vector represents the driving quantity of the actuator. In this application, it is mainly the combined torque command of the two walking motors. The measurable disturbance vector mainly refers to the environmental wind load. The system uses real-time wind speed and direction sensors to calculate the equivalent wind pressure and wind overturning moment acting on the hanging basket based on aerodynamic parameters such as the windward area of ​​the hanging basket, and uses them as feedforward disturbance input. The system matrix, determined by the inherent physical parameters of the hanging basket, mainly includes the total mass. Moment of inertia about the horizontal axis and the system damping coefficient; The input matrix describes the control input. How to influence changes in state; The disturbance matrix describes the wind load disturbance. How it affects the system state, n represents the number of system states, m represents the number of control inputs, and p represents the number of disturbances.

[0105] The internal prediction model has parameter adaptation capabilities. Key parameters in the internal prediction model, such as total mass... and moment of inertia It is not fixed. After each concrete pour, the operator inputs the volume of concrete to be poured, and the system automatically calculates the mass increment based on the concrete density. The moment of inertia is updated according to the following relationship: ,in, and These are the moments of inertia before and after the update, respectively. The equivalent arm length from the center of mass to the axis of rotation.

[0106] Subsequently, the system matrix and input matrix All of the above and All relevant elements are recalculated and updated in real time. This mechanism ensures that the prediction model always remains consistent with the current actual load state of the hanging basket, thereby maintaining high-precision prediction performance under various operating conditions. In other words, the MPC controller obtains the optimal control sequence by solving the optimization problem of minimizing J under all preset constraints.

[0107] Workflow example: When the total mass of the hanging basket is updated from 10,000 kg to 12,000 kg, the MPC controller immediately adopts the new system matrix. and input matrix Within each control cycle, the controller solves an optimal control problem in a finite time domain, constrained by the aforementioned state-space equations. The goal is to find a set of control sequences. This allows the basket to move from its current position to the target position, and its predicted trajectory (especially the pitch angle) is determined accordingly. It is always stably constrained within a safe boundary (e.g., ±1.0 degrees) while maintaining optimal dynamic performance.

[0108] The internal prediction model is a simplified state-space model of the hanging basket's mechanics. Its state variables include at least the system's position, velocity, and equivalent disturbance torque caused by wind load and eccentric load. Model parameters, such as the hanging basket's equivalent mass and moment of inertia, are loaded according to the hanging basket's preset design drawings during system initialization and are manually updated by the operator after each concrete pour based on the pour volume to ensure that the prediction model can reflect the hanging basket's current true load characteristics.

[0109] This internal prediction model, within each control cycle, uses the current spatial pose feedback as the initial state, the tuned servo parameters as the model's intrinsic characteristics, and incorporates feedforward compensation for future environmental wind loads. Within a finite time domain, such as the next five seconds, it predicts multiple potential motion trajectories at fixed time intervals. The planning of these trajectories strictly follows preset mechanical and attitude constraints. The mechanical constraints include ensuring that the predicted structural stress values ​​at all key points do not exceed the second stress threshold, while the attitude constraints require that the predicted absolute value of the platform tilt angle is always lower than the second tilt angle threshold.

[0110] The algorithm iteratively calculates and selects the optimal solution from all feasible trajectories that achieves the highest positioning accuracy, lowest energy consumption, and smoothest motion. Finally, it outputs an anti-disturbance motion command sequence, which precisely specifies the target displacement, target velocity, and target acceleration of the hanging basket in the future.

[0111] Next, the high-level anti-disturbance motion commands generated by the model predictive control algorithm are converted into specific drive signals for the dual-side traveling motors. This conversion process is completed in coordination with cross-coupling control and attitude closed-loop fine-tuning. Cross-coupling control is responsible for maintaining the synchronization between the two traveling motors, collecting encoder feedback from the two motors in real time, and calculating their actual displacement difference. This actual displacement difference is not only multiplied by a synchronization compensation gain and directly fed back to their respective speed loops, but also accumulated through a synchronization error integrator to eliminate steady-state synchronization error. This design ensures that even when the friction coefficients of the two tracks are different, the basket can still maintain straight-line forward movement and avoid jamming.

[0112] Before the final synthesized motor command, the outputs of cross-coupling control and attitude closed-loop fine-tuning pass through an output limiting stage. This stage ensures that the absolute values ​​of the synchronous compensation from cross-coupling control and the speed compensation from attitude fine-tuning do not exceed a preset limit, such as 5% of the rated speed of a single motor. This is to prevent an excessively large command output from a compensation loop under extreme abnormal conditions, which could lead to system instability. Thus, while pursuing high performance, the bottom line of stability is firmly maintained.

[0113] Meanwhile, the attitude closed-loop fine-tuning works in parallel as an independent compensation loop, continuously monitoring the platform roll angle data provided by the inertial measurement unit. Once a slight tilt trend is detected, such as the platform starting to tilt to the right by 0.1 degrees, the loop will immediately take action. Based on the tilt angle and the rate of change of angle, it calculates an additional speed compensation amount. This compensation amount is superimposed on the motor base speed command output by the cross-coupling control, specifically by slightly reducing the target speed of the right motor or slightly increasing the target speed of the left motor, thereby generating a corrective rotational torque to restore the platform to level.

[0114] Finally, after synchronization by cross-coupling control and fine-tuning compensation of attitude closed loop, the two independent motor drive commands are sent to their respective servo drivers via fieldbus. The drivers use the core parameters that have been tuned according to the working conditions to precisely execute speed and torque control. For example, when starting from a heavy load steady state, since the parameters have been tuned to a flexible "S"-shaped speed curve, the two motors will accelerate smoothly. At the same time, the cross-coupling control ensures that they are in sync, and the attitude fine-tuning loop suppresses the head-up or head-down phenomenon that may be caused by uneven load in real time.

[0115] Through a multi-loop fusion control strategy, the hanging basket not only strictly tracks the preset anti-disturbance trajectory on a macroscopic level, but also achieves high-precision walking synchronization and platform stability on a microscopic level, ultimately moving safely and smoothly to the target position.

[0116] Step 5: Run an independent safety guardian thread to monitor and judge critical stress and abnormal tilt angle in real time based on a pre-set rule base, and trigger the response with the highest priority, forming a safety protection closed loop in parallel with the main control loop. The specific implementation is as follows:

[0117] Create and run an independent security daemon thread. This thread is given the highest task priority in the real-time operating system and runs in parallel with the main control loop responsible for regular motion control. It independently occupies a processor core or is implemented using a hardware watchdog circuit. It continuously monitors key safety signals and triggers handling actions with a millisecond-level response speed based on a pre-set rule base, forming a security protection closed loop that is not affected by the main loop being busy or abnormal.

[0118] This security protection mechanism is a rule base pre-installed in the system's non-volatile memory, which defines several hard security thresholds that cannot be modified by regular programs and their corresponding response logic.

[0119] Key thresholds include, but are not limited to: a third stress threshold, which is higher than the second stress threshold used for condition identification and is usually set to 90% of the material's yield strength as the last line of defense against structural plastic failure; a third tilt angle threshold, which is also higher than the second tilt angle threshold, for example, set to 3.5 degrees, to prevent the hanging basket from overturning due to excessive tilting; and an emergency wind speed threshold, which is higher than the wind speed threshold that triggers dynamic disturbances, for example, set to 20 meters per second, indicating that the wind load has exceeded the design resistance capacity.

[0120] The real-time monitoring and decision-making process of the safety daemon thread is as follows: This thread directly reads the raw data from the stress sensor and inertial measurement unit through a dedicated high-speed data channel, completely bypassing the data processing flow of the main control loop. Within each extremely short safety scan cycle, such as ten milliseconds, it performs the following decision:

[0121] First, check if the instantaneous value of structural stress at any key measuring point exceeds the third stress threshold; second, check if the larger of the platform's lateral and longitudinal tilt angles (i.e., the absolute value of the platform tilt angle) exceeds the third tilt angle threshold; finally, check if the instantaneous value of the ambient wind speed exceeds the emergency wind speed threshold.

[0122] The above check uses OR logic; as long as any one condition is met, the thread will immediately set a global safety alert flag.

[0123] Once the judgment conditions are met, the safety guardian thread will not hesitate to execute the highest priority response. First, it will send an emergency stop signal directly to all servo drives through the hardware-level output point. This signal can cut off the motor enable and activate the mechanical brake. At the same time, it will send a command to the hydraulic control system to make all hydraulic cylinders switch to pressure holding or safe retraction mode.

[0124] It should be noted that, in order to ensure the reliability of monitoring data, after reading the raw data from the sensors, the safety protection thread will perform a simple and effective two-out-of-three logic judgment. Specifically, for key stress monitoring points, the system will deploy two or three sensors for redundant measurement. The safety protection thread will compare the readings of multiple sensors at the same monitoring point. Only when two or more of the readings exceed the third stress threshold at the same time will the stress be finally judged as exceeding the limit.

[0125] Specifically, the safety protection thread itself also has a self-diagnostic function, periodically sending test commands to the sensors or checking whether data updates have timed out, to confirm that the entire safety monitoring chain (from sensors to I / O modules to processor) is working properly. When the self-diagnosis finds that two sensors in the three-out-of-two logic are simultaneously faulty, the system will degrade to one-out-of-one logic and immediately alarm, prompting that maintenance is required. After the emergency stop is triggered, the system will enter a safety lockout state. The system can only be restarted when the operator confirms on-site that the safety hazard has been eliminated and manually resets it using a special key or password, thereby preventing secondary accidents caused by accidental reset.

[0126] For example, during concrete pouring, if a sensor malfunction causes the main controller to fail to detect off-center loading, resulting in the platform tilt angle continuously increasing to 3.5 degrees, the safety protection thread will detect this situation within ten milliseconds and directly trigger an emergency stop, forcing the entire system to stop and thus preventing a rollover accident. The time, triggering rules, and handling actions of all safety events are recorded in a separate black box memory for post-event analysis. Through this independent and highest-priority protection system, the final safety barrier is built for the automated operation of the hanging basket.

[0127] It should be noted that the thresholds involved in the embodiments can be determined according to specific scenarios and needs.

[0128] This invention significantly improves the intelligence and safety performance of the hanging basket in complex construction environments by constructing an integrated closed loop of operational status perception, parameter adaptive tuning, and disturbance-resistant motion control. Firstly, based on multi-source sensor data fusion and finite state machine reasoning, it accurately identifies core operating conditions such as no-load stability, heavy-load stability, and dynamic disturbances, effectively overcoming the control mismatch problem caused by misjudgment of operating conditions in traditional control. Secondly, through a pre-set operating condition parameter mapping table and an online self-tuning mechanism, it dynamically adjusts the servo system control parameters, enabling the system to maintain optimal dynamic characteristics under different load and disturbance conditions. Furthermore, it combines... The model predictive control algorithm, which integrates spatial pose feedback and environmental wind load feedforward, generates a smooth, disturbance-resistant trajectory while strictly meeting mechanical and attitude constraints. Then, through cross-coupled control and attitude closed-loop fine-tuning, it achieves high-precision synchronization of the dual-side walking motors and platform attitude stability, effectively suppressing structural fluctuations caused by off-center loading and wind load. Finally, through an independently running safety guardian thread, it provides the highest level of safety assurance through multiple redundant monitoring and millisecond-level response mechanisms. This ensures accurate positioning and stable operation of the hanging basket while significantly reducing energy consumption and mechanical vibration, achieving a simultaneous improvement in construction efficiency and safety reliability.

[0129] Example 2: A hanging basket motion control system based on operating status recognition, such as Figure 2 As shown, it specifically includes:

[0130] The state perception fusion module is used to simultaneously collect structural stress, spatial pose, platform tilt angle and environmental wind load data based on the multi-source sensor terminal deployed on the hanging basket body, perform timestamp alignment and data encapsulation, and construct a unified state vector.

[0131] The operating condition identification module is used to perform finite state machine reasoning on the state vector based on preset stress thresholds and tilt angle thresholds to determine the core operating condition of the hanging basket.

[0132] The parameter adaptive module is used to adaptively tune the core parameters of the servo drive and motion controller based on the constructed working condition parameter mapping table and the identified core operating conditions.

[0133] The disturbance rejection motion control module is used to perform dynamic trajectory planning with mechanical and attitude constraints based on the core parameters after tuning, integrating spatial pose feedback and environmental wind load feedforward, and using model predictive control algorithm to generate disturbance rejection motion commands. It also drives the dual-side walking motors to perform the commands in coordination through cross-coupling control and attitude closed-loop fine-tuning to achieve walking synchronization and platform stability.

[0134] The closed-loop adjustment module is used to run an independent safety guardian thread, which monitors and judges critical stress and abnormal tilt angle in real time based on a preset rule base, and triggers the response with the highest priority, forming a safety protection closed loop in parallel with the main control loop.

[0135] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, ATA hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state ATA hard disk.

[0136] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0137] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0138] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0139] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0140] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0141] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A hanging basket movement control method based on running state recognition, characterized by, The method comprises the following steps: Synchronously collecting structural stress, spatial pose, platform inclination and environmental wind load data based on a multi-source sensing terminal arranged on the hanging basket body, performing timestamp alignment and data encapsulation, and constructing a unified state vector; Performing finite state machine reasoning on the state vector based on preset stress threshold and inclination threshold to determine the core operating condition of the hanging basket; Based on the constructed operating condition parameter mapping table, the core parameters of the servo drive and motion controller are adaptively set according to the identified core operating condition; Based on the set core parameters, the spatial pose feedback and environmental wind load feedforward are fused, the model predictive control algorithm is used to perform dynamic trajectory planning with mechanical and attitude constraints, anti-disturbance motion instructions are generated, and the double walking motors are driven through cross-coupling control and attitude closed-loop fine adjustment to execute the instructions, which are used to control walking synchronization and platform stability; An independent safety daemon thread is run to monitor and determine critical stress and abnormal inclination in real time based on a preset rule library, a highest priority response is triggered, and a safety protection closed loop parallel to the main control loop is formed; The specific process of the model predictive control algorithm for dynamic trajectory planning with mechanical and attitude constraints is as follows: A discrete state space model containing hanging basket mass, damping and stiffness is constructed as a prediction model, the spatial pose is taken as a state variable, and the motor driving force is taken as a control variable; In each control cycle, the optimal control problem in a limited time domain is solved by the prediction model based on the current state vector; The mechanical and attitude constraints are applied as hard constraints of the optimal control problem, including that the structural stress calculated based on the prediction model should not exceed the material allowable stress, and the absolute value of the predicted platform transverse inclination should not exceed the anti-overturning safety threshold; The quadratic programming problem with constraints is solved, and the obtained first control instruction sequence is taken as the anti-disturbance motion instruction output; The specific way of the environmental wind load data as the feedforward compensation quantity into the model predictive control algorithm is as follows: Based on real-time wind speed and wind direction, the equivalent wind disturbance force is calculated through the wind load coefficient and the hanging basket windward area, and the equivalent wind disturbance force is taken as a known external disturbance input, which is introduced into the prediction model and the objective function of the optimal control problem; The specific process of the independent safety daemon thread for monitoring and determining critical stress and abnormal inclination in real time based on the preset rule library is as follows: The safety daemon thread is independent of the main control loop and runs at a higher priority, and the safety daemon thread condition rule library is composed of a series of IF-THEN production rules; The rule condition part logically judges the structural stress instantaneous value, the platform inclination instantaneous value and their change trend; The rule action part includes generating different levels of safety protection instructions, and the safety protection instructions at least include deceleration, shutdown and emergency stop; When the structural stress instantaneous value exceeds the stress safety threshold or the platform inclination instantaneous value exceeds the inclination safety threshold, the highest priority rule is triggered, the emergency stop instruction is immediately generated and executed, and an interrupt signal is sent to the main controller. When the standard deviation of the structural stress or the platform inclination angle continues to increase and exceeds the dynamic stability threshold value in consecutive cycles, a secondary rule is triggered, a deceleration or shutdown instruction is generated, and a warning log is recorded.

2. The hanging basket movement control method based on running state recognition according to claim 1, characterized in that: Based on the multi-source sensing terminal deployed on the hanging basket body, the structural stress, spatial pose, platform inclination angle and environmental wind load data are synchronously collected, time stamp alignment and data packaging are performed, a unified state vector is constructed, and the specific process is as follows: The GNSS receiver, encoder, resistance strain gauge, inclination sensor and ultrasonic anemometer are assigned with hardware synchronization signals, or all asynchronously collected data frames are given a unified time stamp at the software layer; Based on the unified time stamp, the data from different sensors are buffered and interpolated to align within the set fusion period, ensuring that all data correspond to the same physical time; The aligned data is packaged according to the predefined order and data structure to form a unified state vector containing a time marker.

3. The hanging basket movement control method based on running state recognition according to claim 2, characterized in that: Based on the preset stress threshold and inclination threshold, the state vector is subjected to finite state machine reasoning to determine the core operating condition of the hanging basket, and the specific process is as follows: Define the state set of the finite state machine, which at least includes the empty load stable state, the heavy load stable state, and the dynamic disturbance state triggered by strong wind or unbalanced load; Based on the unified state vector, the sliding average value of the structural stress and the instantaneous peak value, the absolute value of the platform inclination angle, and the sliding average value of the environmental wind speed are calculated within each control period; When the sliding average value of the structural stress is lower than the first stress threshold, and the absolute value of the platform inclination angle is lower than the first inclination threshold, the state is determined to be the empty load stable state; When the sliding average value of the structural stress is higher than the first stress threshold but lower than the second stress threshold, and the absolute value of the platform inclination angle is lower than the second inclination threshold, the state is determined to be the heavy load stable state; When the sliding average value of the environmental wind speed exceeds the wind speed threshold, or the weighted comprehensive index of the instantaneous peak value of the structural stress and the absolute value of the platform inclination angle exceeds the dynamic stability threshold, it is forcibly migrated to the dynamic disturbance state.

4. The hanging basket movement control method based on running state recognition according to claim 3, characterized in that: Based on the constructed operating condition parameter mapping table, the core parameters of the servo drive and motion controller are adaptively set according to the identified core operating condition, and the specific process is as follows: The operating condition parameter mapping table is a pre-set lookup table, the row index of the operating condition parameter mapping table is the operating condition inferred by the finite state machine, the column index is the controller core parameter to be set, and the table content is the optimal parameter value obtained through offline optimization or historical data learning; When the operating condition is migrated to the heavy load stable state, the corresponding parameter combination in the mapping table is automatically called, which includes increasing the torque limit value of the lifting servo drive, increasing the position loop proportional gain, and reducing the speed loop integral time constant; When the operating condition is migrated to the dynamic disturbance state, the corresponding parameter combination in the mapping table is automatically called, which includes enabling the strong smoothing mode of the trajectory filter, reducing the prediction time domain and control time domain of the model predictive controller, and shortening the control period of the attitude closed-loop fine tuning.

5. The hanging basket movement control method based on running state recognition according to claim 4, characterized in that: Through cross-coupling control and attitude closed-loop fine tuning, the dual walking motors are driven cooperatively, and the specific process is as follows: The difference between the left walking displacement and the right walking displacement is calculated in real time and defined as the synchronization error. The synchronization error is input into a synchronization compensator, and the synchronization compensator at least comprises a proportional control link; The output of the synchronization compensator is superimposed on the output of the position loop controller of each of the left and right motors in opposite polarities, respectively, to form cross feedback; The lateral inclination signal measured by the dual-axis inclination sensor is input into an independent proportional-differential controller, and the output of the proportional-differential controller is taken as a speed correction amount, which is superimposed on the speed command of the left and right traveling motors in opposite signs, respectively.

6. The hanging basket movement control method based on running state recognition according to claim 5, characterized in that: In the positioning signal limited area, path calibration and position alignment are performed, and the specific process is as follows: In the GNSS signal shielding area, the path calibration mechanism is started, and the displacement fingerprint sequence is constructed based on the mileage reference marker information deployed beside the walking track and the encoder reading on the hanging basket body; The collected inertial measurement unit data and the observed mileage marker sequence are used to perform path matching and error minimization fitting, complete the loop correction of the hanging basket running path and the correction of the positioning trajectory.

7. A hanging basket movement control system based on running state recognition, for implementing the hanging basket movement control method based on running state recognition according to any one of claims 1-6, characterized in that, Comprise: The state perception fusion module is used for synchronously collecting structural stress, spatial pose, platform inclination and environmental wind load data based on the multi-source sensing terminal deployed on the hanging basket body, performing timestamp alignment and data packaging, and constructing a unified state vector; The running condition recognition module is used for performing finite state machine reasoning on the state vector based on the preset stress threshold and inclination threshold to determine the core running condition of the hanging basket; The parameter adaptive module is used for self-adapting the core parameters of the servo drive and the motion controller according to the identified core running condition based on the constructed condition parameter mapping table; The anti-disturbance motion control module is used for fusing spatial pose feedback and environmental wind load feedforward based on the set core parameters, adopting a model predictive control algorithm to perform dynamic trajectory planning with mechanical and attitude constraints, generating anti-disturbance motion instructions, and driving the double traveling motors through cross-coupling control and attitude closed-loop fine adjustment to realize walking synchronization and platform stability; The closed-loop adjustment module is used for running an independent safety daemon thread, monitoring and judging the critical stress and abnormal inclination in real time based on the preset rule library, triggering the disposal response with the highest priority, and constituting a safety protection closed loop parallel to the main control loop.

Images