A method for measuring the inclination posture of a cleaning robot body

Abstract

The present application belongs to the technical field of photovoltaic module cleaning robot posture measurement and track navigation control, and particularly relates to a cleaning robot body inclination posture measurement method, which determines the body position skew angle by establishing a track navigation posture measurement reference system, determines the body position angle fusion value, double-end navigation displacement difference value and position angle navigation observation value within a measurement time window, constructs the position navigation consistent index, double-end position rigid body index and correction pulse controllable index, and outputs the body position posture measurement result and measurement state accordingly. The present application establishes the track navigation posture measurement reference system, correlates the double-end position angle measurement, navigation displacement increment and effective navigation baseline length, constructs the position navigation consistent index, double-end position rigid body index and correction pulse controllable index, and improves the reliability of the body position posture measurement result and the distinguishability of the measurement state determination.

Description

Technical Field

[0001] This invention belongs to the field of photovoltaic module cleaning robot attitude measurement and track navigation control technology, and more specifically, it relates to a method for measuring the body tilt angle attitude of a cleaning robot. Background Technology

[0002] Photovoltaic modules are exposed to the outdoor environment for extended periods, and their surfaces are easily contaminated with dust, sand, bird droppings, fallen leaves, snow, and other obstructions, affecting their ability to receive sunlight and their power generation efficiency. To reduce manual cleaning costs and improve the continuity of cleaning operations, cleaning robots that run along the tracks of photovoltaic modules are increasingly being used in photovoltaic power plant operation and maintenance scenarios. These cleaning robots are typically installed across the photovoltaic module array, with their walking structures at both ends cooperating with the photovoltaic module frames, guide rails, or adjacent support structures to complete the cleaning of the module surfaces during operation.

[0003] In actual cleaning processes, the robot's body has a certain length, and the contact environment of the tracks at both ends of the body is not entirely the same. For example, there are seams between photovoltaic modules, differences in installation height between module frames, and slight deformation, dust accumulation, water film residue, or changes in frictional resistance may occur in local areas of the track. The contact resistance between the cleaning brush and the module surface may also change with the degree of contamination. These factors will cause differences in the displacement increment between the first and second running ends of the robot, resulting in azimuth deviation of the body relative to the track direction. At the same time, local impacts, end jamming, or instantaneous sensor disturbances may also cause asynchronous changes in the azimuth angle measurements at both ends.

[0004] Current methods for measuring the attitude of cleaning robots commonly include determining attitude based on a single-end tilt angle or azimuth angle, estimating the robot's state based on the angle difference between the two ends, or calculating the operational deviation based on the incremental displacement at both ends. These methods can provide effective attitude references under conditions of smooth track, balanced forces at both ends, and stable sensor output. However, when there are local obstructions on the photovoltaic module track, asynchronous end attitudes, or short-term fluctuations in navigation displacement differences, relying solely on azimuth angle measurements can easily misinterpret local anomalies at the ends as overall robot tilt, while relying solely on navigation displacement increments makes it difficult to distinguish between actual robot azimuth tilt and non-navigation displacement disturbances. Especially when the robot's ends maintain an approximately rigid body motion relationship, but the navigation displacement difference and azimuth angle changes are not entirely consistent, the reliability of the robot's azimuth attitude measurement results will be affected without a unified track navigation attitude measurement benchmark and a multi-indicator joint judgment basis.

[0005] Therefore, this application proposes a method for measuring the tilt angle attitude of a cleaning robot. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide a method for measuring the tilt angle and attitude of a cleaning robot.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for measuring the tilt angle attitude of a cleaning robot, applicable to cleaning robots running along the track of photovoltaic modules, includes: Step 1: Establish a track navigation attitude measurement reference system including a first orientation reference direction, a second orientation reference direction, and an attitude measurement reference axis, and determine the body orientation skew angle of the cleaning robot based on the track navigation attitude measurement reference system. Step 2: Within the measurement time window, acquire the first end azimuth angle measurement value, the second end azimuth angle measurement value, the first end navigation displacement increment, the second end navigation displacement increment, and the effective navigation baseline length corresponding to the two ends of the robot body, and determine the body azimuth angle fusion value, the double-end navigation displacement difference value, and the azimuth angle navigation observation value. Step 3: Based on the fuselage azimuth angle fusion value, azimuth angle navigation observation value, dual-end navigation displacement difference value, effective navigation baseline length, and measurement time window, construct the azimuth navigation consistency index; based on the dual-end measurement difference between the first-end azimuth angle measurement value and the second-end azimuth angle measurement value, construct the dual-end azimuth rigid body index. Step 4: Based on the dual-end orientation rigid body index, orientation navigation consistency index and body azimuth angle fusion value, conduct exploratory differential speed measurement excitation evaluation on both ends of the sweeping robot body, and obtain the controllable index of the correction pulse. Step 5: Based on the dual-end orientation rigid body index, orientation navigation consistency index, and correction pulse controllability index, determine the body orientation and attitude measurement results of the cleaning robot, and output the measurement status corresponding to the body orientation and attitude measurement results.

[0009] In one embodiment, the cleaning robot includes a main beam, a first drive component and a first mechanical tilt sensor disposed at a first end of the main beam, a second drive component and a second mechanical tilt sensor disposed at a second end of the main beam, and a navigation displacement acquisition unit. The main beam constitutes the body of the cleaning robot, and the first end and the second end of the main beam are respectively a first running end and a second running end. The first mechanical tilt sensor acquires the first end azimuth angle measurement value corresponding to the first running end, the second mechanical tilt sensor acquires the second end azimuth angle measurement value corresponding to the second running end, and the navigation displacement acquisition unit acquires the first end navigation displacement increment corresponding to the first running end and the second end navigation displacement increment corresponding to the second running end.

[0010] In one embodiment, the direction of the guide line of the photovoltaic module track on which the cleaning robot runs is determined as the first azimuth reference direction, the extension direction of the cleaning robot body is determined as the second azimuth reference direction, and the in-plane angle generated by the cleaning robot body around the attitude measurement reference axis relative to the first azimuth reference direction is determined as the body azimuth skew angle.

[0011] In one embodiment, the fuselage azimuth fusion value is determined based on the first end azimuth measurement value and the second end azimuth measurement value, and the difference between the two ends of the navigation displacement is determined based on the difference between the first end navigation displacement increment and the second end navigation displacement increment; the effective navigation baseline length is the baseline length formed between the two ends of the cleaning robot body along the second azimuth reference direction, and a navigation baseline azimuth observation model is established based on the difference between the two ends of the navigation displacement and the effective navigation baseline length, and the azimuth navigation observation value is determined through the navigation baseline azimuth observation model.

[0012] In one embodiment, the azimuth navigation consistency index is constructed by generating the azimuth navigation consistency index based on the angular deviation between the fuselage azimuth angle fusion value and the azimuth angle navigation observation value, the baseline projection relationship between the difference in the two-end navigation displacement and the effective navigation baseline length, and the synchronicity of changes within the measurement time window.

[0013] In one embodiment, the process of evaluating the experimental differential speed measurement excitation at both ends of the cleaning robot body and obtaining the controllable index of the correction pulse includes: determining whether the preset pulse measurement triggering relationship is met based on the dual-end orientation rigid body index, the orientation navigation consistency index, and the fusion value of the body azimuth angle; and determining whether to apply experimental differential speed measurement excitation to both ends of the cleaning robot body based on the judgment result; if experimental differential speed measurement excitation is applied, constructing the controllable index of the correction pulse; if experimental differential speed measurement excitation is not applied, determining the unused evaluation state of the controllable index of the correction pulse, which is a state that does not meet the preset controllability conditions.

[0014] In one embodiment, the construction method of the controllable index of the correction pulse includes: obtaining the fuselage azimuth angle fusion value, the first end azimuth angle measurement value, the second end azimuth angle measurement value and the difference in navigation displacement at both ends before and after the trial differential speed measurement excitation, forming excitation response data containing the changes before and after excitation; and determining the actual azimuth angle response, the first end azimuth angle response, the second end azimuth angle response and the theoretical azimuth angle response based on the excitation response data, the effective navigation baseline length and the orbital navigation attitude measurement reference system, thereby constructing the controllable index of the correction pulse.

[0015] In one embodiment, the tentative differential speed measurement excitation includes at least two correction pulses, each correction pulse having a preset amplitude, a preset duration, and a preset direction.

[0016] In one embodiment, when the dual-end azimuth rigid body index does not meet the preset rigid body consistency condition, the fuselage azimuth angle fusion value is determined as a non-rigid body attitude measurement value, and an attitude safety lock flag is output; when the dual-end azimuth rigid body index meets the preset rigid body consistency condition, if the azimuth navigation consistency index meets the preset azimuth skew condition or the correction pulse controllable index meets the preset controllability condition, the fuselage azimuth angle fusion value is determined as the fuselage azimuth skew angle, and a valid fuselage azimuth attitude measurement result is output; if the azimuth navigation consistency index does not meet the preset azimuth skew condition and the correction pulse controllable index does not meet the preset controllability condition, the fuselage azimuth angle fusion value is determined as a non-navigation displacement difference disturbance angle, and an invalid fuselage azimuth attitude measurement result is output.

[0017] In one embodiment, the abnormal state corresponding to the attitude safety lock flag includes at least one of the following: local structural disturbance of the robot body, end attitude asynchrony, and end orientation measurement abnormality.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention acquires the first end azimuth angle measurement value, the second end azimuth angle measurement value, the first end navigation displacement increment, the second end navigation displacement increment, and the effective navigation baseline length corresponding to the two ends of the robot body within a measurement time window. Based on this, it determines the body azimuth angle fusion value, the difference in navigation displacement between the two ends, and the azimuth angle navigation observation value. Furthermore, it constructs an azimuth navigation consistency index and a two-end azimuth rigid body index, so that the body azimuth angle measurement, the change in navigation displacement between the two ends, and the difference in azimuth measurement between the two ends of the body form a correlation judgment relationship. This is helpful in distinguishing between the actual azimuth deviation of the body and the abnormal attitude caused by the difference in the two-end measurement, and improves the reliability of the body azimuth attitude measurement results under complex track operation conditions. Based on the dual-end orientation rigid body index, orientation navigation consistency index, and body azimuth angle fusion value, a trial differential speed measurement excitation evaluation is performed on both ends of the sweeping robot body to obtain the controllable index of the correction pulse. Further, by combining the dual-end orientation rigid body index, orientation navigation consistency index, and controllable index of the correction pulse, the body orientation attitude measurement results and corresponding measurement states are determined. This enables the orientation attitude measurement to simultaneously determine the dual-end rigid body relationship, the orientation and navigation relationship, and the differential speed excitation response evaluation, which helps to improve the discriminability of the body orientation attitude measurement results and provides a basis for the output of the measurement states corresponding to different body orientation attitudes. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the overall process of measuring the orientation and attitude of the cleaning robot in this embodiment of the application. Figure 2 This is a flowchart illustrating the acquisition of measurements and determination of related values ​​within a measurement time window in this application embodiment; Figure 3This is a flowchart illustrating the construction of controllable indexes for exploratory differential speed measurement excitation evaluation and correction pulse in the embodiments of this application. Figure 4 This is a flowchart illustrating the determination of fuselage orientation and attitude measurement results and corresponding measurement states in an embodiment of this application. Detailed Implementation

[0020] Reference Figure 1 A method for measuring the tilt angle attitude of a cleaning robot, applicable to cleaning robots running along the track of photovoltaic modules, comprising: Step 1: Establish a track navigation attitude measurement reference system including a first orientation reference direction, a second orientation reference direction, and an attitude measurement reference axis. Based on the track navigation attitude measurement reference system, determine the body orientation skew angle of the cleaning robot and unify the running direction of the cleaning robot on the photovoltaic module track, the extension direction of the body itself, and the reference axis on which the attitude measurement depends into the same measurement reference. The first orientation reference direction reflects the guiding relationship of the photovoltaic module track, the second orientation reference direction reflects the arrangement relationship of the cleaning robot body relative to the track, and the attitude measurement reference axis defines the measurement axis of the body orientation deviation angle. By determining the body orientation deviation angle through the track navigation attitude measurement reference system, the inconsistency of the orientation reference caused by simply relying on the local angle of the body end can be avoided, so that the body orientation attitude measurement has a clear track reference relationship.

[0021] When the cleaning robot runs along the photovoltaic module track, the main beam serves as the main support for the robot's posture. A first drive component and a first mechanical tilt sensor are arranged at the first end of the main beam, and a second drive component and a second mechanical tilt sensor are arranged at the second end of the main beam. The navigation displacement acquisition unit collects the displacement changes at both ends of the main beam as it runs along the track. The first end of the main beam serves as the first running end, and the second end serves as the second running end. The first mechanical tilt sensor is installed at the first running end near the end of the main beam, and the second mechanical tilt sensor is installed at the second running end near the end of the main beam. The installation directions of both sensors maintain a defined assembly relationship with the extension direction of the main beam. The first mechanical tilt sensor obtains the first end azimuth angle measurement value corresponding to the first running end, and the second mechanical tilt sensor obtains the second end azimuth angle measurement value corresponding to the second running end. The navigation displacement acquisition unit obtains the first end navigation displacement increment corresponding to the first running end and the second end navigation displacement increment corresponding to the second running end, respectively. For example, when the cleaning robot straddles the surface of a row of inclined photovoltaic modules and moves laterally along the photovoltaic module track, the first running end and the second running end are located on both sides of the main beam along its length. The first driving component and the second driving component may have a slight difference in travel due to track friction, the height difference of the module frame, or local dust accumulation. The first mechanical tilt sensor and the second mechanical tilt sensor record the orientation changes at both ends, and the navigation displacement acquisition unit records the travel increment at both ends, forming the original measurement quantity related to the robot's body posture. When establishing the orbital navigation attitude measurement reference system, the direction of the guide line of the photovoltaic module track on which the cleaning robot runs is determined as the first azimuth reference direction, the extension direction of the cleaning robot body is determined as the second azimuth reference direction, and the attitude measurement reference axis is used as the angle measurement axis for the azimuth deviation angle of the cleaning robot body. In this orbital navigation attitude measurement reference system, the first azimuth reference direction reflects the track direction that the cleaning robot should follow when walking along the photovoltaic module track, the second azimuth reference direction reflects the actual spatial arrangement direction of the main beam on the cleaning robot body, the attitude measurement reference axis, the first azimuth reference direction and the second azimuth reference direction together define the measurement plane of the body azimuth deviation angle, and the in-plane angle generated by the cleaning robot body around the attitude measurement reference axis relative to the first azimuth reference direction is determined as the body azimuth deviation angle. For example, when the first running end lags behind the second running end due to local obstruction, the second orientation reference direction of the main beam deflects relative to the guide line direction of the photovoltaic module track. This deflection forms a body orientation skew angle in the plane defined by the attitude measurement reference axis. The first end azimuth angle measurement value, the second end azimuth angle measurement value, the first end navigation displacement increment, and the second end navigation displacement increment are all interpreted around the track navigation attitude measurement reference system, avoiding the introduction of end local errors by judging the body attitude of the cleaning robot solely based on single-end tilt angle data.

[0022] In this embodiment, all angles are measured in radians, and the angle normalization function is denoted as . Its output range is When the main beam is in a preset reference attitude and the fuselage azimuth angle is zero, the first and second mechanical tilt sensors are calibrated to zero position. End at the The relative angle of the end, obtained by converting the sensor sampling output at each sampling time using the calibration curve, is denoted as . , No. The zero-position offset of the terminal is denoted as , No. The measured value of the end azimuth angle is denoted as ,in, When the fuselage rotates in the positive direction relative to the first azimuth reference direction along the attitude measurement reference axis, the azimuth angle takes a positive value; when rotating in the reverse direction, the azimuth angle takes a negative value, satisfying the following:

[0023] Reference Figure 2 Step 2: Within the measurement time window, acquire the first end azimuth angle measurement value, the second end azimuth angle measurement value, the first end navigation displacement increment, the second end navigation displacement increment, and the effective navigation baseline length corresponding to the two ends of the robot body. Also, determine the body azimuth angle fusion value, the difference in navigation displacement between the two ends, and the azimuth angle navigation observation value. Simultaneously acquire the azimuth angle information and navigation displacement information of the two ends of the robot body within the same measurement time window to form a set of measurement data reflecting the overall posture of the body and the difference in operation between the two ends. The first and second end azimuth measurements reflect the azimuth status at both ends of the fuselage. The first and second end navigation displacement increments reflect the displacement changes at both ends of the fuselage during orbital operation. The effective navigation baseline length reflects the effective scale for establishing azimuth observation relationships between the two ends. The fuselage azimuth fusion value reflects the comprehensive result of the dual-end azimuth measurements. The dual-end navigation displacement difference reflects the degree of inconsistency between the two ends' navigation displacements. The azimuth navigation observation value reflects the azimuth observation result derived from the navigation displacement difference and the effective navigation baseline length.

[0024] The measurement time window is set to match the continuous operation cycle of the cleaning robot along the photovoltaic module track. The sampling time covers the synchronous change process of the two ends of the body in the same posture evolution state. Within this time window, the azimuth angle measurement values ​​of the first end and the second end are collected around the first and second running ends of the main beam, respectively. At the same time, the navigation displacement increment of the first end and the navigation displacement increment of the second end are obtained simultaneously, and the effective navigation baseline length formed between the two ends of the main beam along the second azimuth reference direction is measured. For example, when a cleaning robot traverses the joint area of ​​photovoltaic modules, a short-term speed difference may occur between the first and second operating ends, resulting in non-consistent changes in azimuth and displacement increments. These differences are fully recorded within a unified time window to avoid attitude misjudgment due to sampling misalignment. Based on this data, a fuselage azimuth fusion process is performed, fusing the azimuth measurements from the first and second ends according to a weighted consistency rule to form a fuselage azimuth fusion value. This fusion value reflects the overall azimuth trend of the fuselage and weakens the impact of momentary disturbances at the ends. The weight consistency rule is to fuse angles with equal weights at both ends. The second end azimuth measurement is expanded based on the first end azimuth measurement, and the expanded second end azimuth measurement is denoted as... The fuselage azimuth fusion value is denoted as ,satisfy:

[0025]

[0026] Differential calculations are performed on the first-end navigation displacement increment and the second-end navigation displacement increment to obtain the double-end navigation displacement difference. This difference reflects the degree of inconsistency in the operation of the two ends of the main beam in the track direction, which is more obvious when there are changes in local track resistance or differences in the adhesion of components. The effective navigation baseline length serves as a geometric constraint to limit the spatial distribution relationship between the two ends of the fuselage. This length is extended along the second azimuth reference direction and stably corresponds to the main beam structural dimensions. The measurement time window is denoted as The measurement time window includes Each synchronous sampling time ,in, . No. At the sampling time The cumulative navigation trip is recorded as The first end navigation displacement increment is denoted as The second-end navigation displacement increment is denoted as The difference in displacement between the two navigation ends is denoted as The displacement difference between the two navigation ends is positive when the first operating end leads the second operating end in travel. This positive direction is consistent with the positive direction of the fuselage azimuth angle during zero-position calibration. The effective navigation baseline length is denoted as... , The fixed distance formed between the corresponding azimuth measurement positions at both ends of the main beam along the second azimuth reference direction, and ,satisfy:

[0027]

[0028]

[0029] Furthermore, a navigation baseline azimuth observation model is constructed based on the difference between the two-end navigation displacements and the effective navigation baseline length. This model maps the displacement difference to the azimuth change and outputs the azimuth angle navigation observation value by combining the direction constraint relationship in the orbital navigation attitude measurement reference system. When the change in fuselage orientation within the measurement time window is within the preset small angle measurement range, the starting sampling time of the measurement time window... The corresponding fuselage azimuth fusion value is used as the initial navigation azimuth value, and the azimuth navigation observation value is denoted as... ,satisfy:

[0030] The estimated value of the dual-end navigation displacement difference corresponding to the fuselage azimuth fusion change is denoted as: The baseline projection residual is denoted as ,satisfy:

[0031]

[0032] For example, when the cleaning robot is running in a slightly tilted area of ​​the photovoltaic module, the difference in navigation displacement between the two ends increases while the effective navigation baseline length remains stable. At this time, the azimuth angle navigation observation value shows a trend consistent with the azimuth deflection angle of the body, thus providing an independent observation basis for subsequent azimuth consistency determination. In another operating scenario, when the ground friction conditions are uniform, the difference in navigation displacement between the two ends tends to stabilize, and the azimuth angle navigation observation value and the fused value of the azimuth angle of the body maintain a high degree of consistency.

[0033] Step 3: Based on the fuselage azimuth angle fusion value, azimuth navigation observation value, dual-end navigation displacement difference value, effective navigation baseline length, and measurement time window, construct an azimuth navigation consistency index; based on the dual-end measurement difference between the first-end azimuth angle measurement value and the second-end azimuth angle measurement value, construct a dual-end azimuth rigidity index, evaluate the consistency of the correspondence between the fuselage azimuth angle fusion value and the azimuth navigation observation value, and evaluate the rigidity of the azimuth measurement difference at both ends of the robot body; The azimuth navigation consistency index reflects whether there is a consistent azimuth relationship between the fuselage azimuth fusion value and the navigation displacement change. It can reflect the degree of matching between the azimuth change, the difference in navigation displacement between the two ends, and the effective navigation baseline length within the measurement time window. The dual-end azimuth rigidity index reflects whether the first-end azimuth measurement value and the second-end azimuth measurement value maintain measurement consistency with the fuselage rigid structure. It can distinguish the difference between the overall fuselage azimuth deviation and the end measurement anomalies and local structural disturbances.

[0034] Within the same measurement time window, the fuselage azimuth fusion value, azimuth navigation observation value, dual-end navigation displacement difference value, and effective navigation baseline length are time-aligned, and multiple sets of azimuth navigation observation data are generated according to the corresponding sampling time. For each set of azimuth navigation observation data, the angle deviation between the fuselage azimuth fusion value and the azimuth navigation observation value is calculated, and the difference exceeding the angle boundary is normalized in combination with the angle periodicity to avoid abrupt errors when the azimuth crosses the angle boundary. Based on the baseline projection relationship between the dual-end navigation displacement difference value and the effective navigation baseline length, the geometric azimuth change corresponding to the dual-end operational difference is determined, and the direction, magnitude, and timing of the geometric azimuth change are compared with the fuselage azimuth fusion value within the measurement time window. When the direction of change is consistent, the magnitude of change is within the set tolerance range, and the timing of change remains synchronized, the azimuth navigation consistency index is improved; when the angle deviation increases, the baseline projection relationship deviates, or the synchronization of change decreases, the azimuth navigation consistency index is reduced. In this embodiment, the angular deviation between the fuselage azimuth angle fusion value and the azimuth angle navigation observation value is denoted as . ,satisfy:

[0035] Within the measurement time window, the average absolute value of the angle deviation is obtained by averaging the absolute values ​​of the angle deviations at each sampling time, and the average absolute value of the baseline projection residual is obtained by averaging the absolute values ​​of the baseline projection residuals at each sampling time. The fuselage azimuth fusion mean is obtained by averaging the values ​​after expansion based on the fuselage azimuth fusion value at the initial sampling time within the measurement time window. The change synchronization ratio is obtained by the proportion of the number of fuselage azimuth fusion changes that are consistent with the azimuth navigation observation changes in adjacent sampling intervals to the total number of sampling intervals. When the average absolute value of the angle deviation is not greater than the preset angle consistency tolerance, the average absolute value of the baseline projection residual is not greater than the preset baseline projection residual tolerance, the change synchronization ratio is not less than the preset synchronization ratio threshold, and the absolute value of the fuselage azimuth fusion mean is not less than the minimum verifiable skew angle, the azimuth navigation consistency index takes the evaluation result that meets the preset azimuth skew condition; in other cases, the azimuth navigation consistency index takes the evaluation result that does not meet the preset azimuth skew condition. For example, when the cleaning robot passes through the frame seam of the photovoltaic module, the first running end is briefly hindered, and the difference in navigation displacement between the two ends increases accordingly. The fused value of the body azimuth angle and the observed value of the azimuth angle navigation both change in the same direction at adjacent sampling times, and the amount of change conforms to the geometric relationship limited by the effective navigation baseline length. At this time, the azimuth navigation consistency index remains at a high level. If the observed value of the azimuth angle navigation changes significantly while the fused value of the body azimuth angle remains basically unchanged, it is determined that there is inconsistency in the angle deviation, baseline projection relationship, or change synchronicity within this time period. A dual-end azimuth rigid body index is constructed based on the difference between the first and second end azimuth angle measurements. Specifically, the difference between the first and second end azimuth angle measurements at each sampling moment within the measurement time window is calculated, and the average level, fluctuation amplitude, duration, and continuity of the dual-end measurement difference are statistically analyzed. When the dual-end measurement difference remains within the set rigid body tolerance range for a long period, and the first and second end azimuth angle measurements show consistent direction and similar amplitude, it is determined that the first and second running ends of the main beam maintain an attitude response that conforms to the rigid body motion relationship of the cleaning robot body, and the dual-end azimuth rigid body index takes a higher evaluation result. When the dual-end measurement difference exceeds the set rigid body tolerance range for multiple consecutive sampling moments, or when the first and second end azimuth angle measurements show reverse changes, single-end abrupt changes, or continuous asynchrony, the dual-end azimuth rigid body index takes a lower evaluation result. The residual assembly offset between the first and second end azimuth measurements is denoted as . The difference in orientation measurements at both ends is recorded as ,satisfy:

[0036] Within the measurement time window, the average absolute value of the difference between the two azimuth measurements is obtained by averaging the absolute values ​​of the differences between the two azimuth measurements at each sampling time. The fluctuation range of the difference between the two azimuth measurements is obtained by the root mean square deviation of the difference between the two azimuth measurements at each sampling time relative to its average value. The synchronization ratio of the two-end changes is obtained by the proportion of the number of sampling intervals in which the first and second azimuth changes are in the same direction to the total number of sampling intervals. The number of consecutively exceeding sampling points is obtained by the number of sampling points in which the absolute value of the difference between the two azimuth measurements continuously exceeds the maximum number of sampling points with a preset tolerance for the difference between the two azimuth measurements. When the average absolute value of the difference between the two azimuth measurements is not greater than the preset tolerance for the difference between the two azimuth measurements, the fluctuation range of the difference between the two azimuth measurements is not greater than the preset tolerance for the fluctuation between the two azimuth measurements, the number of consecutively exceeding sampling points is not greater than the preset number of consecutively exceeding sampling points, and the synchronization ratio of the two-end changes is not less than the preset threshold for the synchronization ratio between the two azimuth measurements, the evaluation result of the rigid body index of the two azimuth measurements is taken as meeting the preset rigid body consistency condition; otherwise, the evaluation result of the rigid body index of the two azimuth measurements is taken as not meeting the preset rigid body consistency condition. For example, when the cleaning robot deflects in the plane relative to the photovoltaic module track, the azimuth angle measurement values ​​at the first end and the second end usually change in approximately synchronously. This situation is classified as an overall change in the robot's orientation. However, when the first mechanical tilt sensor is subjected to local vibration, or when a short-term structural disturbance occurs at one end of the main beam, only the corresponding end's azimuth angle measurement value fluctuates abnormally, and the difference between the two ends is significantly amplified. Based on this, the local change at the end can be distinguished from the overall orientation deviation of the cleaning robot's body.

[0037] Reference Figure 3 Step four: Based on the dual-end orientation rigid body index, orientation navigation consistency index and body azimuth angle fusion value, conduct trial differential speed measurement excitation evaluation on both ends of the sweeping robot body and obtain the correction pulse controllability index. On the basis of the existing orientation measurement results and navigation consistency evaluation, introduce trial differential speed measurement excitation evaluation to further determine whether the response of both ends of the sweeping robot body to the correction pulse meets the controllability requirements. The dual-end orientation rigid body index reflects whether the two ends of the robot body have the conditions for overall attitude measurement. The orientation navigation consistency index reflects the degree of agreement between the orientation measurement results and the navigation displacement relationship. The fusion value of the robot body azimuth angle reflects the current orientation state of the robot body. By conducting exploratory differential speed measurement excitation evaluation on the two ends of the robot body, the controllable correction pulse index can reflect the response relationship between the change of the robot body azimuth angle and the differential speed excitation, serving as an important criterion for distinguishing between the actual robot body orientation deviation and non-navigation displacement difference disturbances.

[0038] In this embodiment, the azimuth navigation consistency index being in a suspected skew state means that the azimuth navigation consistency index has not reached the evaluation result that satisfies the preset azimuth skew condition, and the absolute value of the fuselage azimuth angle fusion mean within the measurement time window is not less than the preset pulse measurement trigger angle. When the dual-end azimuth rigid body index satisfies the preset rigid body consistency condition, the azimuth navigation consistency index is in a suspected skew state, and the absolute value of the fuselage azimuth angle fusion mean is not greater than the preset pulse safety angle, it is determined that the preset pulse measurement trigger relationship is satisfied, and a trial differential speed measurement excitation is applied to the first drive component and the second drive component. If the aforementioned conditions are not simultaneously satisfied, no trial differential speed measurement excitation is applied, the correction pulse controllability index is in an inactive evaluation state, and the inactive evaluation state is treated as not satisfying the preset controllability condition, but this does not indicate that the first drive component or the second drive component has malfunctioned.

[0039] For example, when the cleaning robot passes through the junction of the photovoltaic module frame, the azimuth angle measurement values ​​of the first and second running ends still change synchronously, but there is a short-term inconsistency between the difference in navigation displacement between the two ends and the fused value of the body azimuth angle. At this time, the robot selects to enter the trial differential speed measurement excitation evaluation through the preset pulse measurement trigger relationship. If the azimuth angle measurement value of the first end jumps abruptly and the azimuth angle measurement value of the second end does not change accordingly, the trial differential speed measurement excitation is not applied to avoid mistaking the abnormal end measurement as a correctable body posture. When an exploratory differential speed measurement excitation is applied, the exploratory differential speed measurement excitation includes... One correction pulse, . No. Each correction pulse has a preset amplitude, preset duration, and preset direction, wherein, The preset amplitude is the absolute value of the difference in commanded speed between the first and second driving components; the preset direction includes the positive and negative directions of the double-ended navigation displacement difference. A preset interval is maintained between adjacent correction pulses, and at least two correction pulses have opposite preset directions.

[0040] For example, a set of correction pulses causes a short-term differential speed between the first operating end and the second operating end, while another set of correction pulses is applied in the opposite preset direction to check whether the fuselage azimuth fusion value shows a identifiable change with the differential speed direction. Before and after the exploratory differential speed measurement excitation, the fuselage azimuth fusion value, the first end azimuth measurement value, the second end azimuth measurement value, and the difference in navigation displacement between the two ends are acquired respectively, and excitation response data containing the changes before and after excitation are formed. Combining the effective navigation baseline length and the orbital navigation attitude measurement reference system, the actual azimuth response, the first end azimuth response, the second end azimuth response, and the theoretical azimuth response are calculated, and the deviation between the actual azimuth response and the theoretical azimuth response, the degree of synchronization between the first end azimuth response and the second end azimuth response, and the direction of change of the difference in navigation displacement between the two ends under the action of the correction pulses are compared. No. The stable sampling interval before the application of each correction pulse is denoted as . The stable sampling interval after the correction pulse ends and a preset stabilization interval is denoted as . . The number of internal sampling points is denoted as , The number of internal sampling points is denoted as ,in, , . and None of the calculations include changes from correction pulse commands. The azimuth angles used in the summation are all expanded based on the first sample value of the corresponding stable sampling interval. The azimuth angle values ​​in the following formulas refer to the values ​​after this azimuth expansion. The actual azimuth angle response is denoted as... The first azimuth response is denoted as The second azimuth response is denoted as The change in displacement difference between the two navigation ends is denoted as The theoretical azimuth response is denoted as ,satisfy:

[0041]

[0042]

[0043]

[0044]

[0045] For each correction pulse, evaluation is performed based on the direction of change of the dual-end navigation displacement difference, the actual azimuth response, the theoretical azimuth response, the first-end azimuth response, and the second-end azimuth response. The correction pulse satisfies the preset controllability condition when: the direction of change of the dual-end navigation displacement difference is consistent with the preset direction of the correction pulse; the actual azimuth response is consistent with the theoretical azimuth response; the absolute values ​​of both the actual and theoretical azimuth responses are not lower than the minimum effective azimuth response; the absolute deviation between the actual and theoretical azimuth responses is not greater than the preset azimuth response tolerance; and the absolute deviation between the first and second-end azimuth responses is not greater than the preset dual-end response synchronization tolerance. When all correction pulses satisfy the preset controllability condition, the controllability index of the correction pulse is taken as the evaluation result of satisfying the preset controllability condition; when any correction pulse does not satisfy the preset controllability condition, the controllability index of the correction pulse is taken as the evaluation result of not satisfying the preset controllability condition.

[0046] Reference Figure 4 Step 5: Based on the dual-end orientation rigid body index, orientation navigation consistency index, and correction pulse controllable index, determine the body orientation attitude measurement results of the cleaning robot, and output the measurement status corresponding to the body orientation attitude measurement results. The dual-end orientation rigid body index, orientation navigation consistency index, and correction pulse controllable index are comprehensively judged to give the body orientation attitude measurement results and corresponding measurement status of the cleaning robot. The dual-end orientation rigid body index serves as the fundamental condition for the reliability of orientation measurements at both ends of the robot body. The orientation navigation consistency index serves as the criterion for judging whether the relationship between orientation measurement and navigation displacement is consistent. The correction pulse controllability index serves as the criterion for judging whether the robot body produces a controllable response to the exploratory differential speed measurement excitation. By combining the above indices, the robot body orientation and attitude measurement results can be distinguished into valid robot body orientation and attitude measurement results, invalid robot body orientation and attitude measurement results, or measurement states with attitude safety lock indicators, ensuring that the robot body orientation and attitude measurement results have clear state outputs during the operation of the cleaning robot on the photovoltaic module track.

[0047] Priority is given to the dual-end orientation rigid body index, which serves as a prerequisite for whether the body orientation attitude measurement results can be included in the orientation confirmation. When the dual-end orientation rigid body index does not meet the preset rigid body consistency condition, it indicates that the orientation response between the first and second running ends has deviated from the rigid body motion relationship of the main beam. At this time, the body orientation angle fusion value is no longer directly identified as the actual orientation deviation generated by the cleaning robot running along the photovoltaic module track. Instead, the body orientation angle fusion value is identified as a non-rigid body attitude measurement value, and an attitude safety lock mark is output. This attitude safety lock mark corresponds to at least one of the following: local structural disturbance of the cleaning robot body, end attitude asynchrony, and end orientation measurement abnormality. For example, the first mechanical tilt sensor may experience a single-end jump after being impacted by the module frame, or the first end of the main beam may experience a short-term twist due to local obstruction, while the second running end does not show a matching orientation change. In such cases, although the body orientation angle fusion value has a numerical output, the value has been mixed with the end abnormal component. The attitude safety lock mark can distinguish this type of non-rigid body measurement state from the normal body orientation deviation state. If the rigid body index of the two-end orientation meets the preset rigid body consistency condition, the orientation navigation consistency index and the controllable index of the correction pulse are included in the judgment. If the orientation navigation consistency index meets the preset orientation deviation condition, or the controllable index of the correction pulse meets the preset controllability condition, it indicates that there is a verifiable orientation correspondence between the fuselage value of the body orientation angle and the displacement relationship of the two-end navigation, or the two ends of the cleaning robot body generate a controllable response in accordance with the preset direction under the excitation of the trial differential speed measurement. At this time, the fuselage value of the body orientation angle is determined as the body orientation deviation angle, and the effective body orientation attitude measurement result is output. For example, if both the first and second operating ends maintain synchronous rigid body changes, and the azimuth change formed by the difference in navigation displacement between the two ends and the effective navigation baseline length is consistent with the fused value of the body azimuth angle, it can be confirmed that the cleaning robot body has a real deviation relative to the photovoltaic module track guide line. If the azimuth navigation consistency index does not meet the preset azimuth deviation condition and the correction pulse controllability index does not meet the preset controllability condition, it means that although the fused value of the body azimuth angle does not violate the consistency requirement of the rigid body at both ends, its change cannot be explained by the difference in navigation displacement at both ends, nor does it show a controllable response under the action of the correction pulse. At this time, the fused value of the body azimuth angle is determined as the non-navigation displacement difference disturbance angle, and an invalid body azimuth attitude measurement result is output. For example, when a cleaning robot passes through a localized dusty area, the orientation measurements at both ends remain similar, but the navigation displacement difference and the correction pulse response do not correspond, and this angle is not used as the body orientation skew angle to participate in attitude confirmation.

[0048] Zero offset Residual assembly bias Effective navigation baseline length Measurement time window Number of simultaneous samples Number of correction pulses The preset amplitude, duration, interval, stabilization interval, small angle measurement range, angle consistency tolerance, baseline projection residual tolerance, synchronization ratio threshold, minimum verifiable skew angle, double-end azimuth difference tolerance, double-end fluctuation tolerance, number of consecutive exceedances, double-end synchronization ratio threshold, pulse measurement trigger angle, pulse safety angle, minimum effective azimuth response, azimuth response tolerance, and double-end response synchronization tolerance of the cleaning robot are all determined based on the calibration data after the robot is installed. The calibration data includes zero-position data of the main beam in a preset reference posture, double-end consistency data during synchronous operation of the first and second running ends, azimuth and navigation displacement data under a known body azimuth skew angle, and pulse response data under preset differential excitation. The sensor sampling outputs, cumulative navigation travel, correction pulse commands, and response data within the stable sampling interval of the first and second ends are recorded according to the same time reference. The effective navigation baseline length is positive, the number of synchronous samples within the measurement time window is no less than two, and the number of correction pulses is no less than two. All tolerance and response thresholds are positive, and all proportional thresholds are proportional values ​​greater than zero and no greater than one. The preset small angle measurement range is determined based on the linearization error of the ratio of the difference in dual-end navigation displacement to the effective navigation baseline length relative to the actual change in the body azimuth angle. This linearization error is no greater than the preset angle consistency tolerance. The preset pulse measurement trigger angle is no greater than the preset pulse safety angle. The preset amplitude, preset duration, preset interval, and preset stabilization interval of the correction pulse are determined based on the pulse response data under preset differential excitation and the safe operating boundary of the cleaning robot. The preset amplitude and preset duration of the correction pulse do not exceed the safe operating boundary of the cleaning robot, and the preset stabilization interval is no shorter than the time required for the body azimuth angle response to enter a stable sampling state after the correction pulse ends. The preset interval between adjacent correction pulses should ensure the stable sampling interval corresponding to the previous correction pulse. Sampling complete. The azimuth response generated by the correction pulse is not less than the effective resolution of the sensor after angle conversion.

[0049] 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 method for measuring the tilt angle and attitude of a cleaning robot, applicable to cleaning robots running along the track of photovoltaic modules, characterized in that, include: Step 1: Establish a track navigation attitude measurement reference system including a first orientation reference direction, a second orientation reference direction, and an attitude measurement reference axis, and determine the body orientation skew angle of the cleaning robot based on the track navigation attitude measurement reference system. The direction of the guide line of the photovoltaic module track on which the cleaning robot runs is determined as the first azimuth reference direction, the extension direction of the cleaning robot body is determined as the second azimuth reference direction, and the in-plane angle generated by the cleaning robot body around the attitude measurement reference axis relative to the first azimuth reference direction is determined as the body azimuth skew angle. Step 2: Within the measurement time window, acquire the first end azimuth angle measurement value, the second end azimuth angle measurement value, the first end navigation displacement increment, the second end navigation displacement increment, and the effective navigation baseline length corresponding to the two ends of the robot body. Determine the fused azimuth angle value, the difference in navigation displacement between the two ends, and the azimuth angle navigation observation value. The effective navigation baseline length is the baseline length formed between the two ends of the robot body along the second azimuth reference direction. Establish a navigation baseline azimuth observation model based on the difference in navigation displacement between the two ends and the effective navigation baseline length, and determine the azimuth angle navigation observation value through the navigation baseline azimuth observation model. Step 3: Based on the fuselage azimuth angle fusion value, azimuth angle navigation observation value, dual-end navigation displacement difference value, effective navigation baseline length, and measurement time window, construct the azimuth navigation consistency index; based on the dual-end measurement difference between the first-end azimuth angle measurement value and the second-end azimuth angle measurement value, construct the dual-end azimuth rigid body index. The azimuth navigation consistency index reflects whether there is a consistent azimuth relationship between the fuselage azimuth fusion value and the navigation displacement change; The dual-end azimuth rigid body index reflects whether the measured values ​​of the first and second end azimuth angles maintain measurement consistency with the rigid body structure of the fuselage. Step 4: Based on the dual-end orientation rigid body index, orientation navigation consistency index and body azimuth angle fusion value, conduct exploratory differential speed measurement excitation evaluation on both ends of the sweeping robot body, and obtain the controllable index of the correction pulse. Step 5: Based on the dual-end orientation rigid body index, orientation navigation consistency index, and correction pulse controllability index, determine the body orientation and attitude measurement results of the cleaning robot, and output the measurement status corresponding to the body orientation and attitude measurement results.

2. The method for measuring the tilt angle attitude of a cleaning robot according to claim 1, characterized in that, The cleaning robot includes a main beam, a first drive component and a first mechanical tilt sensor disposed at the first end of the main beam, a second drive component and a second mechanical tilt sensor disposed at the second end of the main beam, and a navigation displacement acquisition unit. The main beam constitutes the body of the cleaning robot, and the first end and the second end of the main beam are respectively the first running end and the second running end. The first mechanical tilt sensor acquires the first end azimuth angle measurement value corresponding to the first running end, the second mechanical tilt sensor acquires the second end azimuth angle measurement value corresponding to the second running end, and the navigation displacement acquisition unit acquires the first end navigation displacement increment corresponding to the first running end and the second end navigation displacement increment corresponding to the second running end.

3. The method for measuring the tilt angle attitude of a cleaning robot according to claim 2, characterized in that, The fuselage azimuth fusion value is determined based on the first end azimuth measurement value and the second end azimuth measurement value, and the dual-end navigation displacement difference is determined based on the difference between the first end navigation displacement increment and the second end navigation displacement increment.

4. The method for measuring the tilt angle attitude of a cleaning robot according to claim 3, characterized in that, The method for constructing the azimuth navigation consistency index includes: generating the azimuth navigation consistency index based on the angular deviation between the fuselage azimuth angle fusion value and the azimuth angle navigation observation value, the baseline projection relationship between the difference in the two-end navigation displacement and the effective navigation baseline length, and the synchronicity of changes within the measurement time window.

5. The method for measuring the tilt angle attitude of a cleaning robot according to claim 1, characterized in that, The evaluation of the experimental differential speed measurement excitation at both ends of the cleaning robot body and the resulting controllable index of the correction pulse include: determining whether the preset pulse measurement trigger relationship is met based on the dual-end orientation rigid body index, orientation navigation consistency index, and body azimuth angle fusion value, and determining whether to apply experimental differential speed measurement excitation to both ends of the cleaning robot body based on the judgment result; constructing the controllable index of the correction pulse when the experimental differential speed measurement excitation is applied; and determining the unused evaluation state of the controllable index of the correction pulse when the experimental differential speed measurement excitation is not applied, which is a state that does not meet the preset controllability conditions.

6. The method for measuring the tilt angle attitude of a cleaning robot according to claim 5, characterized in that, The construction method of the controllable index of the correction pulse includes: obtaining the fuselage azimuth angle fusion value, the first end azimuth angle measurement value, the second end azimuth angle measurement value and the difference in navigation displacement at both ends before and after the trial differential speed measurement excitation, forming excitation response data containing the changes before and after excitation; based on the excitation response data, the effective navigation baseline length and the orbital navigation attitude measurement reference system, determining the actual azimuth angle response, the first end azimuth angle response, the second end azimuth angle response and the theoretical azimuth angle response, and then constructing the controllable index of the correction pulse.

7. The method for measuring the tilt angle attitude of a cleaning robot according to claim 6, characterized in that, The exploratory differential speed measurement excitation includes at least two correction pulses, each with a preset amplitude, preset duration, and preset direction.

8. A method for measuring the tilt angle attitude of a cleaning robot according to any one of claims 1-7, characterized in that, When the dual-end azimuth rigid body index does not meet the preset rigid body consistency condition, the fuselage azimuth angle fusion value is determined as a non-rigid body attitude measurement value, and an attitude safety lock flag is output. When the dual-end azimuth rigid body index meets the preset rigid body consistency condition, if the azimuth navigation consistency index meets the preset azimuth deviation condition or the correction pulse controllability index meets the preset controllability condition, then the fuselage azimuth angle fusion value is determined as the fuselage azimuth deviation angle, and the effective fuselage azimuth attitude measurement result is output. If the azimuth navigation consistency index does not meet the preset azimuth deviation condition and the correction pulse controllability index does not meet the preset controllability condition, then the fuselage azimuth angle fusion value will be determined as the non-navigation displacement difference disturbance angle, and invalid fuselage azimuth attitude measurement results will be output.

9. A method for measuring the tilt angle and attitude of a cleaning robot according to claim 8, characterized in that, The abnormal states corresponding to the attitude safety lock flag include at least one of the following: local structural disturbance of the robot body, end-effector attitude asynchrony, and end-effector orientation measurement abnormality.

Images