Intelligent collaborative control method and system for sweeping and patrolling garbage collection and environmental sanitation operation multi-actuator

Abstract

The present application relates to the field of intelligent control technology, and is an intelligent collaborative control method and system for sweeping and collecting environmental sanitation operation multi-actuator, comprising: collecting sensing data reflecting the interaction state of each module and the road facility; calculating the pneumatic efficiency decay representation of the blowing and collecting module in the current operation cycle; inputting the pneumatic efficiency decay representation, the contact mechanics response parameter and the environmental feedback parameter into a preset compensatory overload coupling model to generate joint micro-motion wear driving energy of the collecting module; calculating the damage accumulation energy of the road facility according to the joint micro-motion wear driving energy and the contact mechanics response parameter; calculating the system entropy increase rate of the environmental sanitation vehicle; and dynamically generating and executing a multi-modal collaborative intervention strategy based on the comparison result between the system entropy increase rate and the preset degradation stage threshold. The present application solves the problem that the hidden failure characteristics of multi-actuator collaboration in long-term operation are difficult to be effectively dealt with in the prior art.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and is a method and system for intelligent collaborative control of multiple actuators in sanitation and cleaning operations. Background Technology

[0002] In current multi-functional sanitation multi-actuator task planning and scheduling systems, although physical integration and task-level collaborative scheduling of multiple actuators such as blowing, suction, and picking have been achieved, their control concepts still generally rely on a single-point trigger-independent response mode based on target recognition. That is, after the inspection module identifies a specific type of garbage, it directly starts the corresponding actuator to perform the operation, without considering the compensatory behavior of the actuator in pursuing operational efficiency during the task completion process. In actual operation, when the pneumatic efficiency of the blowing module naturally declines due to long-term operation, in order to maintain the subjective quality goal of road cleanliness, the system will unintentionally trigger suction. The increased frequency of module inching, extended contact time, or enhanced ground contact path are compensatory behaviors not stemming from sudden load changes or command errors, but rather from implicit adaptive strategies driven by cleanliness quality feedback. However, existing control architectures are ill-suited to effectively address equipment degradation caused by such erroneous commands, forming a degradation chain of internal module degradation, enhanced compensatory behavior, external facility damage, and feedback reinforcing internal degradation. Simultaneously, existing control systems lack dynamic adjustment criteria that incorporate the status of external facilities into actuator operating strategies, resulting in hidden failure characteristics where multiple actuators appear to function normally on the surface but are actually experiencing accelerated systemic deterioration during long-term operation. Summary of the Invention

[0003] The technical problem to be solved by this invention is that the hidden failure characteristics of multiple actuators in long-term operation are difficult to deal with effectively in the prior art. The invention proposes an intelligent collaborative control method and system for multiple actuators in sweeping and patrolling sanitation operations.

[0004] To achieve the above objectives, the technical solution of the intelligent collaborative control method for multiple actuators in purging and collection sanitation operations of the present invention includes the following steps: S1: Collect sensor data reflecting the interaction status between the blowing module, suction module, picking module and road facilities. The sensor data shall include at least boundary layer disturbance characteristic parameters, contact mechanical response parameters, load dynamic characteristic parameters and environmental feedback parameters. S2: Based on the waste aggregation state in the boundary layer disturbance characteristic parameters and environmental feedback parameters, calculate the aerodynamic efficiency decay characterization of the blowing module in the current operation cycle; S3: Based on the aerodynamic efficiency attenuation characterization quantity, contact mechanical response parameters, and road surface smoothness information in the environmental feedback parameters, the joint micro-motion wear driving energy of the absorption module is generated using the compensating overload coupling model. S4: Calculate the cumulative damage energy of road infrastructure based on the contact mechanics response parameters and aerodynamic efficiency attenuation characteristics; S5: Based on the aerodynamic efficiency decay characterization quantity, joint fretting wear driving energy, road facility damage accumulation energy, and load dynamic characteristic parameters, the system entropy increase rate of the sanitation vehicle is calculated using the collaborative degradation entropy increase model. S6: Based on the comparison between the system entropy increase rate and the preset degradation stage threshold, dynamically generate and execute a multimodal collaborative intervention strategy.

[0005] Specifically, in S1: Boundary layer disturbance characteristic parameters include local velocity distribution information and near-wall flow disturbance information in the fan outlet region; The contact mechanics response parameters include vertical contact information, tangential contact information, normal approach motion information, and driving state information of each joint of the suction module during the contact process between the end of the suction module and the road surface. The driving state information includes joint torque and joint angular velocity. The load dynamic characteristic parameters include the dynamic change characteristics of the working current of the gripper drive motor of the pickup module during the process of grasping a standard object; Environmental feedback parameters include the road surface height dispersion characteristic, which characterizes road surface smoothness, and the aggregation degree characteristic, which characterizes the aggregation state of waste.

[0006] Specifically, in S2, the steps for calculating the aerodynamic efficiency attenuation characterization quantity include: S21: Based on the boundary layer disturbance characteristic parameters, the uniformity deviation of the flow field at the fan outlet and the degree of high-frequency instability of the boundary layer are jointly evaluated to obtain the intrinsic degradation characterization quantity that characterizes the mechanical degradation and aerodynamic degradation of the fan itself. S22: Based on the aggregation characteristics of the current operation cycle and the joint micro-motion wear driving energy of the previous control cycle, determine the waste distribution deterioration modulation term. The waste distribution deterioration modulation term is used to characterize the amplification effect of the collection module's compensation behavior on the risk of secondary waste dispersion. S23: Generate the aerodynamic efficiency decay characterization quantity for the current operating cycle based on the intrinsic degradation characterization quantity and the waste distribution degradation modulation term; Among them, the waste distribution degradation modulation term is also corrected based on the road segment degradation risk information recorded during historical operations. The corrected waste distribution degradation modulation term participates in the generation of the aerodynamic efficiency decay characterization quantity for the current operation cycle.

[0007] Specifically, in S3, the compensatory overload coupling model includes: S31: Calculate the joint friction dissipation energy corresponding to the cumulative mechanical work done by each joint based on the joint torque and joint angular velocity of each joint in the current working cycle of the collection module. S32: Calculate the compensating contact dissipation energy caused by the decrease in blowing efficiency based on the aerodynamic efficiency attenuation characterization quantity, the vertical contact information and normal approach motion information at the end of the collection module. S33: Calculate the impact input energy based on the impact pulse events in the current work cycle, the vertical contact impulse corresponding to each impact pulse event, and the equivalent mass at the end of the collection module; S34: Based on the joint friction dissipation energy, compensated contact dissipation energy, impact input energy, and road height discrete characteristics, the joint fretting wear driving energy is generated using a compensated overload coupling model; Among them, the compensatory overload coupling model weights and couples the joint friction dissipation energy, the compensatory contact dissipation energy, and the impact input energy, and uses the discrete characteristics of road height to amplify and correct the compensatory contact dissipation energy, so as to characterize the promoting effect of road surface roughness changes on the joint fretting wear of the absorption module.

[0008] Specifically, S4 includes: S41: Obtain the vertical contact information, tangential contact information, and movement speed information along the tangential direction of the road surface at the end of the suction module during the current working cycle; S42: For each impact pulse event, extract the corresponding maximum vertical contact amount, and combine it with the road equivalent support stiffness obtained in the low-speed test contact stage to determine the local deformation energy corresponding to each impact pulse event. S43: Accumulate the local deformation energy corresponding to each impact pulse event within the current working cycle to obtain the cumulative impact deformation energy absorbed by the road surface; S44: Based on the aerodynamic efficiency attenuation characterization quantity, tangential contact information, and motion velocity information along the tangential direction of the road surface, determine the tangential shear work done on the road surface during the tangential sliding process of the end of the suction module along the road surface; S45: Based on the cumulative energy of impact deformation and the work done by tangential shear, the cumulative energy of damage to pavement facilities is generated using the pavement facility damage evolution model. Among them, the pavement facility damage evolution model uses a weighted coupling of impact deformation cumulative energy and tangential shear work to characterize the combined effects of vertical impact and tangential slip on pavement facility damage.

[0009] Specifically, in S5, the calculation of the system entropy increase rate includes: S51: Based on the aerodynamic efficiency attenuation characterization quantity, joint fretting wear driving energy, road facility damage accumulation energy and gripper current dynamic change characteristics in load dynamic characteristic parameters, construct the system degradation state vector, where the gripper current dynamic change characteristics are characterized by the rate of change of the current working cycle relative to the factory reference state. S52: Normalize each state component in the system's degraded state vector according to its corresponding preset reference value to obtain a normalized state vector; S53: Calculate the covariance matrix representing the fluctuation and correlation relationships of each degradation dimension based on multiple normalized state vector samples within the sliding time window; S54: Determine the system entropy based on the covariance matrix, and calculate the system entropy increase rate based on the system entropy change between adjacent work cycles; The system entropy increase rate is used to characterize the degree of collaborative degradation among the blowing module, the suction module, the road infrastructure, and the pickup module.

[0010] Specifically, in S6, multimodal collaborative intervention strategies are dynamically generated and executed, including: When the system entropy increase rate is lower than the threshold of the first degradation stage, the system is determined to be in steady-state operation and the current cooperative operation mode is maintained. When the system entropy increase rate is not lower than the threshold of the first degradation stage and is lower than the threshold of the second degradation stage, the system is determined to have entered the implicit compensation period, and the first type of intervention strategy is executed. When the system entropy increase rate is not lower than the threshold of the second degradation stage, the system is determined to have entered the facility backlash period, and the second type of intervention strategy is implemented. The threshold for the first degradation stage is lower than the threshold for the second degradation stage.

[0011] Specifically, S22 also includes: An environmental memory field corresponding to the operation location is established. The environmental memory field is used to record the temporal cumulative impact of historical high-risk operation events on the corresponding road segment. The size of the environmental memory field is determined by the occurrence time of the historical high-risk operation event, the memory decay coefficient, the road segment topology weight, and the number of historical events. When executing S2 in the current job cycle, query the environment memory field corresponding to the current job position; The waste distribution degradation modulation term is corrected based on the environmental memory field obtained from the query, so that the historical high-risk road sections can affect the calculation of the aerodynamic efficiency attenuation characterization quantity of the current operation cycle. The modified waste distribution degradation modulation term is used to generate the aerodynamic efficiency decay characterization quantity for the current operating cycle.

[0012] In addition, the intelligent collaborative control system for multiple actuators in sanitation sweeping and collection operations of the present invention includes the following modules: The module includes a multi-source collaborative sensing module, a trigger judgment module, an endogenous degradation prediction module, an exogenous damage assessment module, an entropy increase assessment module, and a collaborative decision-making module. The multi-source collaborative sensing module is used to collect sensor data reflecting the interaction status between the blowing module, the suction module, the picking module and the road facilities. The triggering judgment module generates a measure of the aerodynamic efficiency decay of the blowing module in the current working cycle based on the convergence characteristics in the boundary layer disturbance characteristic parameters and the environmental feedback parameters. The intrinsic degradation prediction module generates the joint micro-motion wear driving energy of the absorption module using a compensating overload coupling model based on the aerodynamic efficiency decay characterization quantity, contact mechanical response parameters, and road surface smoothness information in the environmental feedback parameters. The exogenous damage assessment module generates the cumulative damage energy of the road infrastructure based on the contact mechanics response parameters and the aerodynamic efficiency attenuation characterization quantity. The entropy increase assessment module generates the system entropy increase rate based on the aerodynamic efficiency decay characterization quantity, joint micro-motion wear driving energy, road facility damage accumulation energy, and load dynamic characteristic parameters using a collaborative degradation entropy increase model. The collaborative decision-making module dynamically generates and executes a multimodal collaborative intervention strategy based on the comparison between the system entropy increase rate and the preset degradation stage threshold.

[0013] Compared with the prior art, the technical effects of the present invention are as follows: 1. This invention identifies the compensatory operating behavior implied by the performance degradation of the blower module, enabling the control system to actively capture the high-frequency micro-motion wear trend of the suction module under the absence of obvious overload signals. Unlike the traditional health management method based on threshold alarms, this invention achieves early warning and proactive intervention for the inherent degradation process, delays the hidden failure of key actuators, and improves the long-term operational reliability of the entire machine. 2. This invention constructs an energy coupling feedback channel between the external road surface facility status and the internal actuator degradation, incorporating environmental factors such as paving loosening and joint deformation into the input variables of collaborative decision-making. This allows the system to not only consider the cleaning effect when generating operation strategies, but also simultaneously assess the cumulative damage risk to infrastructure, avoiding the vicious cycle of the cleaner the better. It achieves synergistic optimization of the equipment's own lifespan and the lifespan of urban sanitation infrastructure, solving the technical defect of existing sanitation robots that are highly efficient in high-frequency and refined operations but unsustainable. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart illustrating the intelligent collaborative control method for multiple actuators in sanitation and cleaning operations according to the present invention. Figure 2 This is a schematic diagram of the intelligent collaborative control system for multiple actuators in sanitation and cleaning operations according to the present invention. Detailed Implementation

[0015] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0016] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0017] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0018] Example 1: like Figure 1 As shown, the intelligent collaborative control method for multiple actuators in sanitation operations according to the present invention constructs a four-body dynamic degradation model including a blowing module, a suction module, a pickup module, and external road facilities based on the implicit compensation behavior between modules of sanitation equipment during operation to maintain the cleaning effect. The method includes: S1: Collect sensor data reflecting the interaction status between the blowing module, suction module, picking module and road facilities. The sensor data shall include at least boundary layer disturbance characteristic parameters, contact mechanical response parameters, load dynamic characteristic parameters and environmental feedback parameters. The environmental feedback parameters are used to characterize road surface smoothness and garbage aggregation status. In S1: The boundary layer disturbance characteristic parameters include: a local velocity profile acquired by a sensor array deployed at the wind turbine outlet section. With wall shear stress ; In this embodiment, it should be noted that, regarding the blower module, considering that long-term operation of the fan can lead to distortion of the outlet airflow velocity profile and high-frequency instability of the boundary layer due to bearing wear and impeller dynamic imbalance, which directly affects the blower efficiency and airflow direction stability, this embodiment collects the local velocity profile of the fan outlet section. Quantify the velocity field inhomogeneity and simultaneously collect wall shear stress. To capture boundary layer separation and turbulent fluctuation characteristics, which together constitute the boundary layer disturbance characteristic parameters in this embodiment; It should be noted that, for the suction module, considering the contact mechanics between the end effector and the road surface during ground-hugging operations—that is, the vertical contact force reflects the downward pressure exerted by the suction nozzle on the road surface, the tangential contact force reflects the sliding friction state, the normal approach velocity reflects the impact intensity at the moment of contact, and the torque and angular velocity of each joint reflect the driving cost borne by the robotic arm to maintain its ground-hugging posture—in this embodiment, the contact mechanics response parameters include: the vertical contact force collected by a multi-dimensional force sensor deployed at the end effector of the suction robotic arm. Tangential contact force ,in and normal approach velocity ; It should be noted that the normal approach velocity The acquisition strategy is as follows: the joint angular velocity is converted into the normal approach velocity using the Jacobian matrix of the robotic arm. ; The contact mechanics response parameters also include: joint torque vectors acquired by torque sensors deployed at each joint. and joint angular velocity ,in , For joint degrees of freedom; The load dynamic characteristic parameters include: the slope of the operating current of the gripper drive motor of the pickup module as a function of time during the process of gripping a standard object. ; It should be noted that, for the pickup module, the current change slope... It is a sensitive indicator characterizing the motor output rate and transmission efficiency. Considering that the increase in backlash and deterioration of lubrication in the gripper transmission chain will be directly reflected in the response sensitivity of the gripping action, in this embodiment, the slope of the current change over time is recorded during each gripping of a standard object. This is done in order to achieve online monitoring of the degradation state of the picking module in a low-cost manner.

[0019] The environmental feedback parameters include: the standard deviation of local road surface height obtained through the real-time positioning and mapping system. And the degree of leaf aggregation obtained through a visual perception system. ; It should also be noted that, regarding environmental feedback, firstly, during the actual operation of sanitation vehicles, the smoothness of the road surface usually directly affects the contact stability between the suction nozzle and the ground. For example, loose paving stones and height differences at the joints can increase the frequency of abnormal impacts. Therefore, in this embodiment, the standard deviation of local road surface height is obtained through a SLAM system. This method aims to quantify the degree of road surface loosening. Simultaneously, considering that the degree of leaf aggregation directly reflects the actual operational effectiveness of the collection module, and more specifically, if the aggregation decreases (i.e., the garbage is dispersed), the collection module will typically be forced to expand its search range, thus increasing the frequency of ineffective actions and contact with the road surface. Therefore, this embodiment also introduces the leaf aggregation degree. To quantify the gathering and agglomeration effect; Preferably, in this embodiment, the aforementioned local road surface height standard deviation and the degree of aggregation of fallen leaves Together they constitute the environmental feedback parameters; In another embodiment, the preferred ratio of the projected area of ​​the collected waste within the work area to the total projected area of ​​waste within the work area is used as the degree of leaf collection.

[0020] The normal approach velocity The acquisition method is as follows: joint angular velocity is collected in real time through the encoders of each joint. Calculate the Jacobian matrix based on the current configuration of the robotic arm. From the terminal velocity relationship Extracting the velocity component perpendicular to the ground to obtain ; The slope of the gripper current change The acquisition method is as follows: during each grasping of a standard object (such as a 500mL plastic bottle), the current of the drive motor is recorded at a sampling rate of not less than 1kHz. For the initial moment of closure Linear fitting was performed on nearby current data, and the slope of the fitted line was taken as... ; S2: Based on the waste aggregation state in the boundary layer disturbance characteristic parameters and environmental feedback parameters, calculate the aerodynamic efficiency decay characterization of the blowing module in the current operation cycle; The aerodynamic efficiency attenuation characterization quantity is generated by jointly evaluating the degradation characteristics of the fan itself and the deterioration characteristics of waste distribution caused by the compensation behavior of the collection module. In S2, the aerodynamic efficiency attenuation characterization quantity is calculated. The steps include: S21: Based on the boundary layer disturbance characteristic parameters, quantify the degree of mechanical and aerodynamic degradation of the wind turbine itself, and calculate the intrinsic degradation characterization quantity. : ; in, The rated outlet wind speed of the fan. For export area, This represents the Hilbert transform used to extract the high-frequency envelope. To characterize the preset characteristic frequency bands of high-frequency instability in the boundary layer, This indicates the average amplitude within that frequency band. These are the wall shear stress calibration values ​​under factory reference conditions. The preset vibration-aerodynamic coupling weighting coefficient; It should be noted that during long-term operation, bearing wear and impeller dynamic balance deterioration of the blower in the sanitation vehicle's blower module will simultaneously lead to the following phenomena: the velocity profile at the outlet section will no longer be uniform, and the degree of deviation of the local flow velocity from the rated value will be aggravated; at the same time, the near-wall boundary layer of the sanitation vehicle will experience high-frequency instability due to the increase in blade surface roughness and changes in blade tip clearance. More specifically, this will manifest as a significant increase in the pulsation amplitude of the wall shear stress in the characteristic frequency band.

[0021] Therefore, in this embodiment, the spatial dispersion of the velocity profile is weighted and superimposed with the high-frequency envelope amplitude of the shear stress, wherein the first term The quantization of velocity field inhomogeneity is achieved by extracting the mean value of the shear stress envelope in the 9-13kHz frequency band using Hilbert transform and comparing it with the factory reference value. The comparison aims to quantify the degree of boundary layer instability.

[0022] In one embodiment, the wall shear stress The data is directly collected by a thin-film shear stress sensor installed on the inner wall of the air outlet duct of the blower module. S22: Based on the aforementioned leaf aggregation degree The joint fretting wear driving energy calculated in the previous control cycle Calculate the garbage distribution degradation modulation term caused by the compensation behavior of the collection module. : ; in, This is the baseline convergence level under new machine conditions. The preset reference wear driving energy, , The preset positive modulation coefficient; The degree of leaf aggregation The value is calculated by segmenting the garbage distribution image within the target area using the visual perception module. It is preferably defined as the ratio of the garbage pixel area within the core area to the total garbage pixel area within the target operation area, or as the inverse mapping value of the average normalized distance from the garbage centroid to the target collection center.

[0023] In another embodiment, a method is provided. The optimal acquisition strategy is as follows: record the aggregation degree under a standard fallen leaf scene using a vision system, and take the average value of 10 repeated trials as the optimal acquisition strategy. ; It should also be noted that the actual operating effect of the blowing module depends not only on the condition of the blower itself, but also on the reverse influence of the compensation behavior of the suction module. This is because when the suction module becomes sluggish or its positioning accuracy decreases due to overload wear, its efficiency in collecting ground debris decreases. This can cause the debris that has been gathered by the blowing airflow to disperse again, i.e., the degree of leaf aggregation. In actual operation, this deterioration in waste distribution caused by the collection module will further increase the workload of the blowing module, creating a vicious cycle. Therefore, in this embodiment, two product factors are constructed, specifically: pass It reflects the rate of degradation of the current clustering relative to the state of a new machine; the lower the clustering, the larger this factor. Considering that joint fretting wear of the suction module reduces the suction port's posture retention capability and increases the probability of residual debris redispersal, the joint fretting wear driving energy from the previous control cycle is introduced as a historical compensation factor for the debris distribution degradation modulation term, i.e., through... Introducing the joint fretting wear driving energy from the previous cycle This characterizes the contribution of the current wear level of the collection module to waste dispersion; that is, the more severe the wear, the slower the compensation action, and the more significant the secondary dispersion of waste.

[0024] S23: Based on S21 and S22, the aerodynamic efficiency attenuation characterization quantity is obtained by combining them. : ; Among them, the waste distribution degradation modulation term The correction was made based on road degradation risk information recorded during historical operations. Then with the modified modulation terms participate The calculation.

[0025] S23 multiplies the intrinsic degradation characterization term by the degradation modulation term to obtain the final aerodynamic efficiency attenuation characterization term. In this embodiment, a multiplication structure is adopted because it is considered that the deterioration of waste distribution caused by the collection module is not a separate source of degradation from the degradation of the fan, but rather an amplification and modulation of the intrinsic degradation effect of the fan. Based on the multiplication model, when When close to zero (good aggregation, slight adsorption wear), ;when When it increases, A sharp rise.

[0026] It should also be noted that, for step S2, the characteristic frequency band The preferred frequency band is the sideband near the wind turbine blade passing frequency and its second harmonic. In this embodiment, the preferred range is 9kHz to 13kHz. Several experiments have verified that this frequency band is most sensitive to boundary layer separation and changes in turbulence intensity.

[0027] In another embodiment, a Hilbert transform in step S2 is also provided. The preferred implementation method is: for the wall shear stress signal Perform bandpass filtering, calculate the magnitude of its analytic signal as the envelope, and then obtain the envelope in... The average value within; More specifically, benchmark value , , The machine is calibrated in a standard test environment under factory conditions, i.e., the fan is running at rated speed, and the velocity profile and shear stress are measured at the outlet section. The average value of multiple measurements is taken as the reference value. In this embodiment, the weighting coefficient Typical values ​​range from 0.3 to 0.8, modulation coefficient , The typical value range is 0.2 to 0.7. The specific value is determined by fitting accelerated degradation experimental data. For example, under different wind turbine wear levels (simulated by adjusting the blade wear amount) and different suction overload levels (simulated by changing the road surface convergence), the cumulative increment of the suction robot arm joint torque is measured as a reference target, and the coefficient value is optimized and determined by the least squares method.

[0028] S3: Based on the aerodynamic efficiency attenuation characterization quantity, contact mechanical response parameters, and road surface smoothness information in the environmental feedback parameters, the joint micro-motion wear driving energy of the absorption module is generated using the compensating overload coupling model. In S3, the compensatory overload coupling model specifically includes: S31: Within the set operation cycle, obtain the torque-time history of each joint of the suction module. Time history of angular velocity ,in, , For the joint degrees of freedom; based on the torque and angular velocity of each joint, calculate the cumulative mechanical work done by the joints of the absorbing module within a set working cycle, and obtain the joint friction dissipation energy. Specifically: ;in, This represents the duration of the current work cycle. It should be noted that during the cleaning operation, the suction robot arm continuously outputs torque and generates angular displacement to maintain the ground-hugging posture and path tracking of the end-effector nozzle. During torque transmission, the joint reducers and harmonic gears continuously dissipate mechanical energy due to factors such as tooth surface friction and lubrication shearing. This frictional dissipation is the fundamental energy source for joint fretting wear. Furthermore, the amount of fretting wear has an approximately linear relationship with the accumulated frictional dissipation in materials science. Therefore, this step involves adjusting the torque of each joint. With angular velocity Summing the absolute values ​​of the products and integrating over time yields the joint friction dissipation energy. ; It should be noted that the absolute value is used because regardless of whether the joint rotates forward or backward, the frictional force always does negative work and consumes energy, so its contribution to wear is in the same direction.

[0029] It should also be noted that, for step S31, the joint friction dissipation energy The preferred implementation method is to synchronously acquire the torque of each joint at a sampling rate of not less than 100Hz. With angular velocity Numerical integration (e.g., trapezoidal integral) is performed on the absolute value of the product of each sampling point within one operation cycle, with the cycle duration being... Determined by the start and end timestamps of the task.

[0030] S32: Obtain the aerodynamic efficiency attenuation characteristic quantity corresponding to the blower module. and the contact force at the end of the suction module in the vertical direction. Approaching speed in the normal direction ; Based on the aforementioned aerodynamic efficiency decay characteristic, the compensating contact work caused by the decrease in blowing efficiency is calculated, and the compensating contact dissipation energy is obtained. Specifically: ; in, In this step, it participates in the weighting as a degradation amplification factor to characterize the effect of blow-collecting degradation in compensating for contact enhancement in the absorber module. This does not change the fundamental physical meaning of the vertical contact mechanical work and is relevant for compensating for contact dissipation energy. It should be noted that when the aerodynamic efficiency of the blower module decreases ( When the airflow at the fan outlet decreases (as it rises), the suction module typically increases its contact frequency and adhesion to the ground at its end in order to search for the blown-away debris. Therefore, this embodiment uses the normal approach speed... Vertical contact force Quantitative compensation for contact dissipation energy.

[0031] The normal approach velocity It is obtained by projecting the velocity vector at the end of the absorption module onto the current road surface normal direction; Furthermore, in this embodiment, an aerodynamic efficiency attenuation characterization quantity is introduced during integration. The aim is to make the energy term reflect the mechanical work of the contact process while quantifying the impact of the degree of blow-collecting degradation. Specifically, the more severe the blow-collecting degradation, the higher the equivalent dissipation weight is assigned to the same contact behavior.

[0032] S33: Obtain all impact pulse events within the set operation cycle, and calculate the impulse of each impact pulse event: ; in, , The vertical contact force during the current work cycle The total number of impact events is calculated after comparing with the preset impact threshold. and The first The start and end times of the next impact pulse; Simultaneously, combining the equivalent mass at the end of the absorption module Calculate the impact input energy from the impact pulse input. Specifically: ; The equivalent mass at the end of the absorption module The equivalent mass lookup table for different typical postures can also be determined by converting the inertia matrix of the robotic arm's operating space under the current posture; in another simplified embodiment, it can also be determined by offline collision calibration experiments.

[0033] For impact input energy The specific quantitative logic is as follows: Considering that during the contact process between the suction nozzle and the ground, due to the height difference between the tile joints or the unevenness of the road surface, short-lived impact pulse events will frequently occur, therefore, according to the mechanical relationship... The impulse input to the end effector of each impact pulse. After end equivalent quality Converted into kinetic energy; It should also be noted that the detection method for the impact pulse event is: based on the vertical contact force. Real-time monitoring and preset impact thresholds are implemented. The typical value is 5N, when If the event exceeds the threshold and lasts for less than 2 ms, it is considered an impact event, and its start and end times are recorded. and .

[0034] In another preferred embodiment, the equivalent mass at the end of the suction module The method of obtaining the parameters is as follows: In the state of a new machine leaving the factory, the mass and inertia parameters of each link are obtained through a dynamic parameter identification experiment of the robotic arm (such as the excitation trajectory method based on the least squares method), and then the equivalent mass of the end effector is calculated, which is 0.5kg to 2.0kg. It should be noted that the specific value depends on the configuration of the robotic arm and the mass of the end effector nozzle.

[0035] S34: Obtain the standard deviation of road surface height in the current work area And calculate the corresponding road surface height variance. ; Based on the pre-set benchmark road surface height variance Surface roughness factor: ; Then the joint friction dissipates energy Compensation for contact dissipation energy Impact input energy The road surface height variance is imported into the compensated overload coupling model to obtain the joint fretting wear driving energy of the absorbing module. Specifically: ; in, , , The preset weighting coefficients, This is the amplification factor for road surface loosening.

[0036] It should also be noted that the compensatory overload coupling model is an equivalent energy model for assessing the trend of joint fretting wear. Considering that the three energy sources in S31-S33 have different contribution weights to joint fretting wear, specifically, joint friction dissipation is the basic source of degradation, compensatory contact dissipation is an additional source of blow-collector degradation, and impact input is a random source caused by uneven road surface. At the same time, a loose road surface will also cause the nozzle's compensatory action to produce more ineffective slippage and repeated contact, that is, the loose state of the external road surface will also amplify the actual impact of compensatory contact dissipation on wear.

[0037] Therefore, in this embodiment, a weighted superposition method is adopted, wherein the road surface height variance... As an amplification factor, it is multiplied into the compensation contact dissipation term to make the same A higher wear-driving weight is assigned to loose surfaces; In another embodiment, the road surface height variance The acquisition method is as follows: local point cloud data output by the SLAM system is divided into grids within the work area, the variance of the height values ​​within each grid is calculated, and the mean of the variances of each grid along the work path is taken as the current period's data. .

[0038] In another embodiment, the weighting coefficient and road surface loosening amplification factor The calibration method is as follows: on the joint fretting wear accelerated test bench, at different road surface smoothness levels (based on... Quantification) and different blow-collection degradation states (in terms of Under quantified conditions, run the standard cleaning cycle and record each energy component. After the experiment, the joint was disassembled to measure the actual fretting wear (e.g., recording the radial clearance increment of the joint bearing or the backlash increment of the transmission chain). Using actual wear as the target variable and each energy component as the independent variable, multiple linear regression was used to determine the appropriate values. and In this embodiment, the value of is , , In this embodiment, The typical range of values ​​is Its value was determined by fitting accelerated wear experiments under different road surface roughness levels.

[0039] S4: Calculate the cumulative damage energy of road infrastructure based on the contact mechanics response parameters and aerodynamic efficiency attenuation characteristics; S4 specifically includes: S41: Obtain the vertical contact force at the end of the suction module. and tangential contact force ,in, ; Simultaneously, the velocity of the end effector along the tangential direction of the road surface is obtained by combining the kinematic information of the robotic arm's end effector. The tangential velocity The projection component of the end velocity vector, obtained by converting the joint angular velocity through the Jacobian matrix of the robotic arm, is determined in the tangential plane of the road surface. For example, the tangential velocity The engineering implementation method is based on joint angular velocity. Jacobian matrix of robotic arm kinematics Calculate the terminal velocity vector By combining the road surface normal vector estimation provided by the SLAM system, Projected onto the road surface tangent plane, the magnitude of the projection vector is taken as... .

[0040] S42: For all impact pulse events, extract the maximum vertical contact force for each impact event. Based on the force-displacement slope during the low-speed trial contact phase, the equivalent support stiffness of the road surface at the current working position is estimated. ; In another embodiment, the equivalent road surface support stiffness is first provided. The preferred estimation method is as follows: during the trial contact phase when the end of the collection module is pressed vertically against the road surface at a low speed (e.g., 1 mm / s), the vertical force is recorded simultaneously. Vertical displacement at the end The slope of the initial linear segment of the force-displacement curve is taken as... Furthermore, it also provides a maximum vertical contact force for an impact event. The extraction method is specifically as follows: within the start and end intervals of each impact event determined in S33. Inside, search for vertical contact force The peak value.

[0041] Calculate the local deformation energy of the road surface input for each impact event. Specifically: ;in, ; The local deformation energy of a single impact depends not only on the magnitude of the impact force, but also on the current support conditions of the road surface. Specifically, for example, the same impact force acting on loose paving stones will produce greater local deformation and joint displacement, and loose road surfaces are often more susceptible to damage.

[0042] S43: Within a set operating cycle, the local deformation energy of all impact events is accumulated to obtain the cumulative impact deformation energy absorbed by the road surface. Specifically: ; S44: A measure of aerodynamic efficiency decay in conjunction with the blower module. During the tangential sliding of the end of the collection module along the road surface, the tangential shear work it performs on the road surface is calculated. Specifically: ; in, In this step, it is used as a damage risk amplification factor in the weighting to characterize the effect of blow-collector degradation on the increased frequency of attitude adjustment and the exacerbation of tangential slip, without changing the physical meaning of tangential mechanical work itself.

[0043] It should be noted that the damage between the end of the suction module and the road surface mainly manifests in two mechanical forms: vertical impact and tangential slippage. Specifically, the degree of indentation and impact fatigue experienced by the paving tile joints is determined by the vertical contact force. Characterization, representing tangential contact force With tangential velocity Multiplying these together gives the instantaneous power of the mechanical work done by the end of the device on the road surface through scraping and friction.

[0044] It should also be noted that when the aerodynamic efficiency of the blower module decreases ( When the airflow direction at the fan outlet becomes turbulent and its focusing ability decreases, the suction module is forced to increase the frequency of attitude adjustments and the sliding amplitude to maintain the cleaning effect, resulting in a significant increase in tangential friction work under the same working path. For the above process, this embodiment multiplies the integral by a quantity representing aerodynamic efficiency attenuation. As an amplification factor, it aims to reflect the indirect contribution of blow-collection degradation to damage to external facilities. Blow-collection degradation will induce the collection module to increase the frequency of compensating actions and the amplitude of sliding in order to maintain the cleaning effect, thereby increasing the tangential contact work and its corresponding equivalent damage risk.

[0045] S45: Accumulates impact deformation energy Work done by tangential shear Importing the pavement infrastructure damage evolution model, the cumulative damage energy of the pavement infrastructure is obtained. Specifically: The road infrastructure damage evolution model is an empirically weighted model for assessing the damage trends of different road materials. and These are the preset weighting coefficients.

[0046] In another embodiment, preferably, the weighting coefficient and The calibration method is as follows: standard damage acceleration test is carried out on pavement of different materials (such as permeable bricks, asphalt, granite curb stones), with different combinations of impact number and different tangential slip distance. After the test, the mass of pavement sealant falling off or the surface wear depth is measured as the damage quantification index. Record the experiment and Multiple linear regression was performed with the damage index to obtain the fit. and ,For example , .

[0047] It should be noted that during the actual operation of sanitation vehicles, vertical impact damage and tangential shear damage are the two main energy sources of road infrastructure degradation. Furthermore, their contribution weights differ depending on the type of infrastructure (e.g., permeable brick joints are mainly affected by vertical impact, while granite curb stones are mainly affected by tangential scraping). Therefore, the cumulative energy from impact deformation should be considered... Work done by tangential shear Weighted summation yields the comprehensive damage accumulation energy. Weighting coefficient , It can be marked separately according to the type of road surface material.

[0048] Example 2: The intelligent collaborative control method for multiple actuators in purging and collection sanitation operations according to embodiments of the present invention further includes: S5: Input the aerodynamic efficiency attenuation characterization quantity, the joint micro-motion wear driving energy, the damage accumulation energy, and the load dynamic characteristic parameters into the collaborative degradation entropy increase model to calculate the system entropy increase rate of the sanitation vehicle; In S5, the entropy increase rate of the system is calculated. include: S51: Constructing the system's degenerate state vector: ,in, The gripper current slope is one of the dynamic characteristic parameters of the load. The rate of change relative to the factory benchmark value in the current period is specifically defined as... , The slope of the reference current when gripping a standard object in its factory condition; The aerodynamic efficiency attenuation parameters have been calculated for S2 to S4 above. (Characterizing the degradation of the blower module), joint fretting wear driving energy (Characterizing the degradation of the absorption module), cumulative energy of road infrastructure damage (Characterizing damage to external facilities), corresponding to three main dimensions in the four-body dynamic degradation model; while the picking module is the fourth main dimension. In this embodiment, its degradation state is measured by the rate of change of the gripper current slope. This is used to characterize the decrease in response sensitivity caused by increased backlash in the gripper drive chain or deterioration in lubrication.

[0049] S52: Normalize the state vector to obtain a normalized state vector. ,in , , , These are the preset baseline values ​​for the corresponding variables; For example, the normalized reference value , , , All values ​​are positive. The setting method is as follows: Under factory conditions, the machine is run for one complete operating cycle in a standard test environment (flat road surface, rated operating conditions). The maximum value or steady-state mean of each variable during this cycle is recorded as the baseline value. Take 1.0, and Typical values ​​are the measured values ​​of joint friction dissipation energy and pavement damage accumulation energy in a single cycle under new machine conditions. Take 0.1.

[0050] S53: Calculate the covariance matrix of the normalized state vector within the sliding time window. ; The sliding time window contains no less than a preset number of continuous operation cycle samples, preferably no less than 3 times the dimension of the state vector.

[0051] It should also be noted that, considering the degradation of sanitation vehicles is a slow, stochastic process, the state value at a single moment is greatly affected by sensor noise and instantaneous operating condition disturbances, and lacks statistical stability. Therefore, in step S53, the most recent value is taken. The normalized state vectors of each work cycle (10-20 cycles) form a sample set, and their covariance matrix is ​​calculated. .

[0052] For example, the covariance matrix The calculation uses the sample covariance formula: ,in For the first in the window One sample, This represents the mean of the samples within the window.

[0053] The diagonal elements of the covariance matrix reflect the fluctuation variance of each degradation dimension, while the off-diagonal elements reflect the correlation strength between any two degradation dimensions. Therefore, when blow-gathering degradation causes absorption overload and absorption overload aggravates road damage, the covariance of the corresponding dimension will increase significantly, and the off-diagonal structure of the covariance matrix will change accordingly.

[0054] S54: Calculate the system entropy increase rate Specifically: First, the covariance matrix obtained in step S53... After performing regularization, we get: ;in, Preset small positive numbers, It is an identity matrix.

[0055] Then, trace normalization is performed on the regularized covariance matrix to obtain: ;in, Represents the trace operation of a matrix.

[0056] Based on normalized matrix The system entropy is defined as: ; Furthermore, under discrete operation cycle conditions, the entropy increase rate of the system is calculated using backward difference: ;in, This is the time interval between two adjacent work cycles.

[0057] It should be noted that the entropy increase rate of the system is used to quantify the degree of synergistic deterioration among the four degradation dimensions of the blowing module, the suction module, the road facilities and the picking module. When each module degrades independently, the entropy increase rate is low. When synergistic deterioration occurs (such as blowing degradation causing suction overload, and suction overload aggravating road damage), the off-diagonal elements of the covariance matrix increase, the entropy value rises accordingly, and the entropy increase rate jumps.

[0058] It should be noted that in a healthy state, the degradation of each dimension of the sanitation vehicle's four-body degradation system is relatively independent and the correlation is weak. At this time, the covariance matrix is ​​approximately a diagonal matrix, and the system entropy is low. However, when implicit compensation behavior drives cross-module coupled degradation, the correlation between each dimension increases, the off-diagonal elements of the covariance matrix increase, and the information entropy rises accordingly.

[0059] For example, matrix entropy can comprehensively reflect the total uncertainty carried by the covariance matrix. Furthermore, the entropy increase rate can be obtained by taking its time derivative. Instead of directly using the entropy value itself, this is because different devices have individual differences in their steady-state entropy values ​​due to differences in operating conditions. However, the rate of entropy increase has stronger universality and comparability. Specifically, when the system is in a steady state, the degradation of each dimension is slow and independent, and the entropy increase rate remains at a low level. When entering the implicit compensation period, the blow-collecting degradation begins to drive the absorption overload, the covariance of the two dimensions increases, and the entropy increase rate jumps to the first threshold range. When entering the facility backlash period, the absorption overload further drives the road surface damage to intensify, the three dimensions form a positive feedback, the off-diagonal structure of the covariance matrix changes drastically, and the entropy increase rate breaks through the second threshold.

[0060] In another embodiment, the system entropy increase rate In discrete-time systems, the calculation is performed using the backward difference method, specifically as follows: ;in, This refers to the duration of the work cycle.

[0061] In another embodiment, the matrix logarithm The calculation is achieved through eigenvalue decomposition, specifically: [the normalized matrix is ​​then processed]. Perform eigenvalue decomposition: Then there is ; in, Represents the diagonal matrix with respect to eigenvalues Take each diagonal element as Logarithm with base 0.

[0062] S6: Based on the comparison between the system entropy increase rate and the preset degradation stage threshold, dynamically generate and execute a multimodal collaborative intervention strategy.

[0063] In S6, multimodal collaborative intervention strategies are dynamically generated and executed, specifically including: when When the system is in a steady-state operation period, the current collaborative operation mode of each module is maintained. when When the system is deemed to have entered a period of implicit compensation, the first type of intervention strategy is implemented. In another embodiment, the first type of intervention strategy includes: The fan speed of the control blower module is reduced to its rated speed. Within the range, adjust the pitch angle of the nozzle to optimize the airflow direction; Adjust the working path planning of the suction module to shorten the distance between adjacent purging paths by 5%; The slope of the gripper current Continuous monitoring is performed, and when an abnormal change in its value is detected that exceeds the preset drift threshold, the pre-calibration process of the picking module is triggered. In another embodiment, the preset drift threshold is set by: statistically analyzing data in the factory-shipped state of the new machine. Standard deviation ,by As a threshold for judging abnormal drift; during continuous monitoring, if 5 consecutive captures... If all values ​​exceed this range, the pre-calibration process is triggered, reminding the operator to check the backlash of the gripper drive chain.

[0064] when When the system is deemed to have entered a period of facility backlash, the second type of intervention strategy is implemented; In another embodiment, the second type of intervention strategy includes: The upper limit of the outlet airflow velocity of the blow-collecting module is forcibly constrained, and the airflow deflection angle is limited to a preset safe range; For example, the safe airflow deflection range is set as follows: based on the relative position of the fan outlet geometry and the ground, a range of angles that do not directly impact the edge joints of the floor tiles is preset within the range of the nozzle pitch angle adjustment. For example, the yaw angle is limited to within ±8°.

[0065] The control module uses variable stiffness impedance control during the contact phase with the ground in order to reduce the impact stiffness of the robotic arm end on the road surface. For example, the variable stiffness impedance control is implemented as follows: when the end of the absorption module contacts the ground, the controller dynamically adjusts the target impedance parameter in force control mode. In the initial contact phase (less than 200ms), the stiffness parameter is set to a low value (e.g., 80N / m) to absorb impact energy. After the contact stabilizes, it is switched to normal stiffness (e.g., 150N / m) to ensure the ground adhesion effect.

[0066] Based on the real-time positioning and mapping system, the current work section is marked as a high-risk area for damage, and facility maintenance prompts containing the coordinates and risk index of the section are generated. in, and The preset entropy increase rate threshold is, and .

[0067] For example, the entropy increase rate threshold and The optimal method for determination is as follows: First, run multiple standard operation cycles in the factory-shipped state of the new machine and record the system entropy increase rate. The steady-state distribution is taken as 1.5 to 2 times the steady-state mean. (e.g., 0.01 bit / min); then, through accelerated degradation experiments (e.g., artificially increasing joint gaps, loosening pavement), the system is simulated to enter a state of implicit compensation and facility backlash, and the distribution range of entropy increase rate under the corresponding state is recorded. The lower limit of the implicit compensation period is taken as... (e.g., 0.03 bits / min).

[0068] Example 3: In another preferred embodiment, step S22 further includes: First, construct an environmental memory field. ,as follows: ; in, To be at the current location The timing of historical high-entropy events The preset memory decay coefficient, Weighting factors based on road segment topology. This is an indicator function that takes the value 1 when the condition is true and 0 otherwise. It is used to filter events that only reach the facility backlash level and include them in the memory field. The total number of recorded historical events; The initial state of the environmental memory field is all zeros, that is... Whenever the current location is detected within a work cycle Then, the time and weight corresponding to the event are appended to the record of the raster, and the raster is updated. Field value; It should be noted that the system can maintain a hash table or raster map indexed by location to store the memory field value of each location, and query it in real time according to the current location each time S2 is calculated.

[0069] In another embodiment, the memory decay coefficient The settings are based on a preset memory half-life: for example, if it is expected that the impact of historical events will decay to half after 7 days, then... .

[0070] In another embodiment, the road segment topology weighting factor The setting method is as follows: the functional level of the current road segment in the SLAM map is determined, for example, high-frequency operation areas such as main road intersections and bus stop areas are given higher weights. Ordinary road sections are assigned a baseline weight ( ); It should be noted that the environmental memory field is used to record and quantify the potential degradation risk accumulated by a specific road segment due to historical high-entropy events; Then, during the execution of S2 in the current job cycle, the environment memory field is queried in real time, and based on the results obtained from the query... Correct the aforementioned waste distribution degradation modulation term: ; in, The preset memory strength coefficient typically ranges from 0.1 to 0.5. Those skilled in the art can calibrate it according to actual engineering needs based on prior knowledge of road segment vulnerability or cloud-shared data. This will not be elaborated here. This coefficient controls the conservative margin of historical risk information in assessing current periodic degradation. The larger the value, the more obvious the amplification effect of historical high-entropy events.

[0071] Finally, the revised version Alternative Used in S23 The calculation.

[0072] It should be noted that, considering that sanitation equipment will repeatedly pass through the same road sections during long-term operations, and some road sections may be in a sub-healthy state due to historical reasons (such as loose paving stones, aging grout, roadbed settlement, etc.), even if the current real-time sensing data does not show obvious abnormalities, the road surface facilities of these road sections are more likely to suffer further damage when facing the same blowing and suction operation load, and the compensation action of the suction module on such road sections is often more violent, and the risk of secondary dispersion of garbage is higher.

[0073] If we rely solely on real-time sensing of leaf aggregation and road surface height variance to assess the degree of degradation modulation, we cannot capture the potential risks brought about by historical cumulative effects, because the road surface may temporarily appear stable in the current cycle, but historical high-entropy events have already revealed the structural vulnerability of the road section.

[0074] To address the aforementioned issues, this embodiment constructs an environmental memory field. It is defined as a historical high-entropy event (i.e., the system entropy increase rate exceeds the second threshold). The weighted time-decayed cumulative sum at the current position (at the moment): .

[0075] In this formula, the indicator function Only high-risk events that truly enter the facility's backlash phase are selected; exponential decay term Weighting recent events higher than long-term events helps prevent the system from becoming overly conservative due to excessive accumulation of historical data; road segment topology weighting. The values ​​are then assigned differently based on the road grade (such as high-frequency operation areas like main road intersections and bus stop areas).

[0076] During the execution of S2 in the current work cycle, the system queries the memory field value of the corresponding grid in real time based on the current location coordinates. If a facility backlash event has occurred in this road segment, then... This is achieved by amplifying the modulatory term for waste distribution degradation through modification.

[0077] After the above processing, in historically high-risk road sections, even if the current aggregation and road surface smoothness are acceptable, the sanitation vehicle system will proactively consider the actual impact of the deteriorated garbage distribution at that location to be more severe than the real-time sensing data, thus incorporating this into the calculation of aerodynamic efficiency degradation metrics. By adding a conservative margin proportional to the level of historical risk, the following technical effects can be achieved: when sanitation equipment passes through known vulnerable road sections, it will reduce the blowing power, shorten the blowing interval, or use a flexible contact strategy in advance to avoid repeated aggravation of road surface damage. At the same time, the data of the environmental memory field can be shared among the equipment clusters through the cloud, ultimately forming a city-level degradation path memory network.

[0078] Example 4: like Figure 2 As shown in the embodiment of the present invention, the intelligent collaborative control system for multiple actuators in sanitation sweeping and collection operations is as follows: Figure 2 As shown, it includes the following modules: The module includes a multi-source collaborative sensing module, a trigger judgment module, an endogenous degradation prediction module, an exogenous damage assessment module, an entropy increase assessment module, and a collaborative decision-making module. The multi-source collaborative sensing module is used to collect sensor data reflecting the interaction status between the blowing module, the suction module, the picking module and the road facilities. The triggering judgment module generates a measure of the aerodynamic efficiency decay of the blowing module in the current working cycle based on the convergence characteristics in the boundary layer disturbance characteristic parameters and the environmental feedback parameters. The intrinsic degradation prediction module generates the joint micro-motion wear driving energy of the absorption module using a compensating overload coupling model based on the aerodynamic efficiency decay characterization quantity, contact mechanical response parameters, and road surface smoothness information in the environmental feedback parameters. The exogenous damage assessment module generates the cumulative damage energy of the road infrastructure based on the contact mechanics response parameters and the aerodynamic efficiency attenuation characterization quantity. The entropy increase assessment module generates the system entropy increase rate based on the aerodynamic efficiency decay characterization quantity, joint micro-motion wear driving energy, road facility damage accumulation energy, and load dynamic characteristic parameters using a collaborative degradation entropy increase model. The collaborative decision-making module dynamically generates and executes a multimodal collaborative intervention strategy based on the comparison between the system entropy increase rate and the preset degradation stage threshold.

[0079] Example 5: This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the aforementioned intelligent collaborative control method for multiple actuators in sanitation and sweeping operations by calling computer programs stored in memory.

[0080] The electronic device can vary considerably depending on its configuration and performance. It may include one or more processors (Central Processing Units, CPUs) and one or more memories, wherein the memory stores at least one computer program. This computer program is loaded and executed by the processor to implement the intelligent collaborative control method for multiple actuators in sanitation operations provided in the above-described method embodiment. The electronic device may also include other components for implementing its functions. For example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Details will not be elaborated upon in this embodiment.

[0081] Example 6: This embodiment proposes a computer-readable storage medium on which an erasable and rewritable computer program is stored. When the computer program runs on the computer device, it enables the computer device to execute the intelligent collaborative control method of the multi-actuator for sanitation and cleaning operations described above.

[0082] For example, computer-readable storage media can be read-only memory (ROM), random access memory (RAM), compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage devices.

[0083] 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.

[0084] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent collaborative control method for multiple actuators in sanitation sweeping and collection operations, characterized in that, The method includes: S1: Collect sensor data reflecting the interaction status between the blowing module, suction module, picking module and road facilities. The sensor data shall include at least boundary layer disturbance characteristic parameters, contact mechanical response parameters, load dynamic characteristic parameters and environmental feedback parameters. S2: Based on the waste aggregation state in the boundary layer disturbance characteristic parameters and environmental feedback parameters, calculate the aerodynamic efficiency decay characterization of the blowing module in the current operation cycle; S3: Based on the aerodynamic efficiency attenuation characterization quantity, contact mechanical response parameters, and road surface smoothness information in the environmental feedback parameters, the joint micro-motion wear driving energy of the absorption module is generated using the compensating overload coupling model. In S3, the compensatory overload coupling model specifically includes: S31: Calculate the joint friction dissipation energy corresponding to the cumulative mechanical work done by each joint based on the joint torque and joint angular velocity of each joint in the current working cycle of the collection module. S32: Calculate the compensating contact dissipation energy caused by the decrease in blowing efficiency based on the aerodynamic efficiency attenuation characterization quantity, the vertical contact information and normal approach motion information at the end of the collection module. S33: Calculate the impact input energy based on the impact pulse events in the current work cycle, the vertical contact impulse corresponding to each impact pulse event, and the equivalent mass at the end of the collection module; S34: Based on the joint friction dissipation energy, compensated contact dissipation energy, impact input energy, and road height discrete characteristics, the joint fretting wear driving energy is generated using a compensated overload coupling model; Among them, the compensatory overload coupling model weights and couples the joint friction dissipation energy, the compensatory contact dissipation energy and the impact input energy, and uses the road height discrete characteristics to amplify and correct the compensatory contact dissipation energy, so as to characterize the promoting effect of road surface roughness change on the joint fretting wear of the absorption module. S4: Calculate the cumulative damage energy of road infrastructure based on the contact mechanics response parameters and aerodynamic efficiency attenuation characteristics; S4 specifically includes: S41: Obtain the vertical contact information, tangential contact information, and movement speed information along the tangential direction of the road surface at the end of the suction module during the current working cycle; S42: For each impact pulse event, extract the corresponding maximum vertical contact amount, and combine it with the road equivalent support stiffness obtained in the low-speed test contact stage to determine the local deformation energy corresponding to each impact pulse event. S43: Accumulate the local deformation energy corresponding to each impact pulse event within the current working cycle to obtain the cumulative impact deformation energy absorbed by the road surface; S44: Based on the aerodynamic efficiency attenuation characterization quantity, tangential contact information, and motion velocity information along the tangential direction of the road surface, determine the tangential shear work done on the road surface during the tangential sliding process of the end of the suction module along the road surface; S45: Based on the cumulative energy of impact deformation and the work done by tangential shear, the cumulative energy of damage to pavement facilities is generated using the pavement facility damage evolution model. Among them, the road infrastructure damage evolution model uses a weighted coupling of impact deformation cumulative energy and tangential shear work to characterize the combined effects of vertical impact and tangential slip on road infrastructure damage. S5: Based on the aerodynamic efficiency decay characterization quantity, joint fretting wear driving energy, road facility damage accumulation energy, and load dynamic characteristic parameters, the system entropy increase rate of the sanitation vehicle is calculated using the collaborative degradation entropy increase model. S6: Based on the comparison between the system entropy increase rate and the preset degradation stage threshold, dynamically generate and execute a multimodal collaborative intervention strategy.

2. The intelligent collaborative control method for multiple actuators in sweeping and collection sanitation operations according to claim 1, characterized in that, In S1: Boundary layer disturbance characteristic parameters include local velocity distribution information and near-wall flow disturbance information in the fan outlet region; The contact mechanics response parameters include vertical contact information, tangential contact information, normal approach motion information, and driving state information of each joint of the suction module during the contact process between the end of the suction module and the road surface. The driving state information includes joint torque and joint angular velocity. The load dynamic characteristic parameters include the dynamic change characteristics of the working current of the gripper drive motor of the pickup module during the process of grasping a standard object; Environmental feedback parameters include the road surface height dispersion characteristic, which characterizes road surface smoothness, and the aggregation degree characteristic, which characterizes the aggregation state of waste.

3. The intelligent collaborative control method for multiple actuators in sweeping and collection sanitation operations according to claim 2, characterized in that, In S2, the steps for calculating the aerodynamic efficiency attenuation characterization quantity include: S21: Based on the boundary layer disturbance characteristic parameters, the uniformity deviation of the flow field at the fan outlet and the degree of high-frequency instability of the boundary layer are jointly evaluated to obtain the intrinsic degradation characterization quantity that characterizes the mechanical degradation and aerodynamic degradation of the fan itself. S22: Based on the aggregation characteristics of the current operation cycle and the joint micro-motion wear driving energy of the previous control cycle, determine the waste distribution deterioration modulation term. The waste distribution deterioration modulation term is used to characterize the amplification effect of the collection module's compensation behavior on the risk of secondary waste dispersion. S23: Generate the aerodynamic efficiency decay characterization quantity for the current operating cycle based on the intrinsic degradation characterization quantity and the waste distribution degradation modulation term; Among them, the waste distribution degradation modulation term is also corrected based on the road segment degradation risk information recorded during historical operations. The corrected waste distribution degradation modulation term participates in the generation of the aerodynamic efficiency decay characterization quantity for the current operation cycle.

4. The intelligent collaborative control method for multiple actuators in sweeping and collection sanitation operations according to claim 3, characterized in that, In S5, calculating the system entropy increase rate includes: S51: Based on the aerodynamic efficiency attenuation characterization quantity, joint fretting wear driving energy, road facility damage accumulation energy and gripper current dynamic change characteristics in load dynamic characteristic parameters, construct the system degradation state vector, where the gripper current dynamic change characteristics are characterized by the rate of change of the current working cycle relative to the factory reference state. S52: Normalize each state component in the system's degraded state vector according to its corresponding preset reference value to obtain a normalized state vector; S53: Calculate the covariance matrix representing the fluctuation and correlation relationships of each degradation dimension based on multiple normalized state vector samples within the sliding time window; S54: Determine the system entropy based on the covariance matrix, and calculate the system entropy increase rate based on the system entropy change between adjacent work cycles; The system entropy increase rate is used to characterize the degree of collaborative degradation among the blowing module, the suction module, the road infrastructure, and the pickup module.

5. The intelligent collaborative control method for multiple actuators in sweeping and collection sanitation operations according to claim 4, characterized in that, In S6, multimodal collaborative intervention strategies are dynamically generated and executed, specifically including: When the system entropy increase rate is lower than the threshold of the first degradation stage, the system is determined to be in steady-state operation and the current cooperative operation mode is maintained. When the system entropy increase rate is not lower than the threshold of the first degradation stage and is lower than the threshold of the second degradation stage, the system is determined to have entered the implicit compensation period, and the first type of intervention strategy is executed. When the system entropy increase rate is not lower than the threshold of the second degradation stage, the system is determined to have entered the facility backlash period, and the second type of intervention strategy is implemented. The threshold for the first degradation stage is lower than the threshold for the second degradation stage.

6. The intelligent collaborative control method for multiple actuators in purging and sweeping sanitation operations according to claim 3, characterized in that, S22 also includes: An environmental memory field corresponding to the operation location is established. The environmental memory field is used to record the temporal cumulative impact of historical high-risk operation events on the corresponding road segment. The size of the environmental memory field is determined by the occurrence time of the historical high-risk operation event, the memory decay coefficient, the road segment topology weight, and the number of historical events. When executing S2 in the current job cycle, query the environment memory field corresponding to the current job position; The waste distribution degradation modulation term is corrected based on the environmental memory field obtained from the query, so that the historical high-risk road sections can affect the calculation of the aerodynamic efficiency attenuation characterization quantity of the current operation cycle. The modified waste distribution degradation modulation term is used to generate the aerodynamic efficiency decay characterization quantity for the current operating cycle.

7. An intelligent collaborative control system for multiple actuators in purging and sweeping sanitation operations, used to implement the intelligent collaborative control method for multiple actuators in purging and sweeping sanitation operations as described in any one of claims 1-6, characterized in that, The system includes: The module includes a multi-source collaborative sensing module, a trigger judgment module, an endogenous degradation prediction module, an exogenous damage assessment module, an entropy increase assessment module, and a collaborative decision-making module. The multi-source collaborative sensing module is used to collect sensor data reflecting the interaction status between the blowing module, the suction module, the picking module and the road facilities. The triggering judgment module generates a measure of the aerodynamic efficiency decay of the blowing module in the current working cycle based on the convergence characteristics in the boundary layer disturbance characteristic parameters and the environmental feedback parameters. The intrinsic degradation prediction module generates the joint micro-motion wear driving energy of the absorption module using a compensating overload coupling model based on the aerodynamic efficiency decay characterization quantity, contact mechanical response parameters, and road surface smoothness information in the environmental feedback parameters. The exogenous damage assessment module generates the cumulative damage energy of the road infrastructure based on the contact mechanics response parameters and the aerodynamic efficiency attenuation characterization quantity. The entropy increase assessment module generates the system entropy increase rate based on the aerodynamic efficiency decay characterization quantity, joint micro-motion wear driving energy, road facility damage accumulation energy, and load dynamic characteristic parameters using a collaborative degradation entropy increase model. The collaborative decision-making module dynamically generates and executes a multimodal collaborative intervention strategy based on the comparison between the system entropy increase rate and the preset degradation stage threshold.

Images