An Adaptive Adjustment Method for Harvesting Digging Shovel Posture in Root and Tuber Crops Based on Digging Resistance Feedback
By constructing resistance evolution fingerprints and attitude probing actions, adaptive adjustment of the digging shovel attitude during the harvesting of root and tuber crops was achieved, solving the problem of lack of specificity and feedback correction in the existing technology for digging shovel attitude adjustment, and improving operational stability and efficiency.
Patent Information
- Application Number
- CN202611364077.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-04
- Publication Date
- 2026-11-10
AI Technical Summary
In existing technologies, the methods for adjusting the posture of the digging shovel during the harvesting of root and tuber crops lack specificity and are difficult to distinguish the obstruction mechanisms of different abnormal sources. This results in a lack of specific adjustment strategies and a lack of feedback correction mechanisms based on the adjustment effect, making it impossible to achieve stable and efficient digging operations in complex environments.
By constructing a resistance evolution fingerprint of excavation resistance changes, the source of resistance anomalies is identified, and targeted attitude probing actions are generated. Combined with resistance fallback characteristics, an adjustment path is constructed to achieve adaptive adjustment and feedback optimization of the excavator shovel attitude.
It improves the accuracy of identifying complex soil environments and crop root conditions, enhances the adaptability and operational stability of the excavation process, reduces redundant and ineffective adjustments and mechanical impact, and improves operational efficiency.
Smart Images

Figure CN123067153A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent control technology for agricultural machinery, and in particular to an adaptive adjustment method for the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback. Background Technology
[0002] With the continuous improvement of agricultural mechanization, the harvesting process of root crops (such as potatoes and sweet potatoes) is gradually moving towards automation and intelligence. In harvesting operations, the digging shovel, as a key actuator, directly affects the digging efficiency, damage rate, and overall machine stability. In actual operating environments, the soil structure is complex and variable, often accompanied by various uncertainties such as hard soil compaction, stone distribution, root entanglement, and wet soil adhesion. This results in digging resistance exhibiting significant time-varying and nonlinear characteristics. Therefore, how to intelligently adjust the digging shovel's posture based on dynamic information during the digging process has become an important research direction for improving operational quality.
[0003] However, most existing excavator shovel attitude adjustment methods still employ passive control strategies based on a single resistance threshold or empirical rules. These methods typically adjust based solely on the magnitude of resistance, failing to differentiate the resistance mechanisms corresponding to different sources of anomalies, resulting in a lack of targeted adjustment strategies. Furthermore, existing methods generally lack structured analysis of the resistance change process, failing to effectively utilize the temporal characteristics of resistance changes, and are prone to misjudgment or adjustment lag in complex working conditions. In addition, most existing adjustment strategies are fixed patterns, lacking feedback correction mechanisms based on adjustment effects. This prevents dynamic optimization of control parameters based on actual operational results, leading to unstable adjustment effects in complex environments, and necessitating further improvement in operational efficiency and reliability. Summary of the Invention
[0004] This invention provides an adaptive adjustment method for the posture of a digging shovel for harvesting root and tuber crops based on digging resistance feedback. This method enables accurate identification and differentiated adjustment under different obstruction conditions. Simultaneously, it continuously optimizes subsequent control strategies by updating the adjustment results, thereby improving the stability, adaptability, and overall efficiency of digging operations.
[0005] An adaptive adjustment method for the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback includes the following steps: S1, during the advance of the excavator shovel, continuously collect the excavation resistance change signal, and identify the sequential combination relationship between the resistance surge segment, the resistance stagnation segment and the resistance release segment, and construct the resistance evolution fingerprint of the current excavation object; S2, based on the resistance evolution fingerprint, determine the current resistance anomaly source type, including hard soil compaction, rock collision, root ball entanglement or wet soil adhesion, and generate corresponding attitude probing actions for different source types; S3, control the digging shovel to perform the posture probing action, and determine the target adjustment path based on the resistance fall characteristics after the probing action. The target adjustment path includes at least one of the following: digging angle adjustment, soil penetration depth retraction, shovel tip lifting to avoid, or short-distance vibration to get rid of the center. S4. The digging shovel attitude is adjusted according to the target adjustment path, and the adjusted resistance feedback result is written back to update the resistance evolution fingerprint in order to continuously correct the subsequent digging attitude.
[0006] Optionally, S1 includes: S11. During the excavator shovel advance, the force sensor installed at the excavator shovel connection or transmission force-bearing part continuously collects the original excavation resistance signal in the excavation direction, and records it in time sequence according to the preset sampling period. The collected original excavation resistance signal is subjected to sliding smoothing and change rate calculation to generate a smoothed excavation resistance value and resistance change rate that characterize the current excavation process resistance change state. S12, based on the smoothed excavation resistance value and resistance change rate, the excavation process is divided into resistance segments according to the resistance amplitude level, resistance change rate and duration, including the resistance surge segment, the resistance stagnation segment and the resistance release segment. S13. The identified resistance surge segment, resistance stagnation segment, and resistance release segment are numbered according to time sequence. The sequential connection relationship between adjacent resistance segments, the duration of each segment, and the characteristic intensity of each segment are extracted and encoded into a resistance evolution fingerprint that characterizes the process of the current excavation object being obstructed. This allows the resistance evolution fingerprint to simultaneously characterize the resistance state type, the order of occurrence, and the strength of evolution.
[0007] Optionally, the resistance segment division includes: When the rate of change of resistance is not less than the threshold for identifying a sudden increase in resistance, it is identified as a sudden increase in resistance. When the smoothed excavation resistance value is not less than the high resistance judgment threshold, the absolute value of the resistance change rate is not greater than the stable change rate threshold, and the duration is not less than the minimum duration, it is identified as a resistance viscous segment. When the rate of change of resistance is not greater than the threshold for identifying the negative resistance release segment, it is identified as a resistance release segment.
[0008] Optionally, S13 includes: S131. According to the order of appearance of each resistance segment during the excavation process, the identified resistance surge segment, resistance stagnation segment and resistance release segment are sequentially numbered, and the segment type, start time, end time and duration corresponding to each resistance segment are extracted to form the basic temporal feature set of each resistance segment. S132, based on the smoothed mining resistance value within the time interval corresponding to each resistance segment, calculate the degree of deviation of each resistance segment from the average resistance level of the current sampling window, and use it as the feature intensity of the corresponding resistance segment. Combine the resistance segment type, start time, end time, duration and feature intensity to construct the feature vector of each resistance segment. S133, arrange the feature vectors of each resistance segment in chronological order to form a serialized feature set that reflects the resistance state type, occurrence order, persistence characteristics and strength of change, and determine this serialized feature set as the resistance evolution fingerprint corresponding to the current mining object.
[0009] Optionally, S2 includes: S21, input the resistance evolution fingerprint into the preset anomaly source template, calculate the matching score between the current resistance evolution fingerprint and each anomaly source template according to the average strength index of the resistance surge segment, the average duration index of the resistance viscous segment, the average strength index of the resistance release segment, and the sequential combination relationship of the resistance segments, and determine the type corresponding to the anomaly source template with the largest matching score as the current resistance anomaly source type. The anomaly source types include hard soil compaction, rock collision, root ball entanglement, and wet soil adhesion. S22, based on the current resistance anomaly source type determined by the judgment, retrieve the target test action parameters corresponding to the anomaly source type from the preset attitude test action library, and generate the attitude test action under the current working condition.
[0010] Optionally, S22 includes: S221. Based on the current resistance anomaly source type determined by the judgment, select the attitude adjustment mode corresponding to the anomaly source type from the preset attitude probing action library, and based on the comprehensive matching score corresponding to the current anomaly source type, adaptively calculate the digging angle adjustment amount, soil penetration depth adjustment amount, shovel tip lifting or sway amplitude adjustment amount to generate attitude probing actions that match the current anomaly level. S222, based on the determined anomaly source type, determine the attitude probing action adjustment mode corresponding to that anomaly source type, and limit the direction of action and dominant adjustment amount of each adjustment parameter in the attitude probing action based on the adjustment mode, so that different anomaly source types correspond to different attitude probing action generation strategies, specifically including: When the abnormality source type is hard soil compaction, the attitude probing action uses the excavation angle adjustment as the main adjustment amount, and the soil penetration depth adjustment amount and the sway adjustment amount as auxiliary adjustment amounts. When the abnormal source type is stone collision, the attitude probing action is mainly adjusted by the shovel tip lifting adjustment amount to achieve rapid avoidance; When the anomaly source type is root tangles, the attitude probing action mainly uses the oscillation adjustment amount as the adjustment amount, and is combined with small angle adjustment to achieve untangling; When the abnormality source type is wet soil adhesion, the attitude probing action uses the soil penetration depth retraction adjustment amount as the main adjustment amount, and combines it with the sway adjustment amount to achieve desorption.
[0011] Optionally, S3 includes: S31, based on the generated attitude test action, control the excavator to perform corresponding digging angle adjustment, soil penetration depth adjustment, shovel tip lifting adjustment or short-distance oscillation adjustment, and continuously collect the digging resistance response signal within the preset observation time window after the attitude test action is performed, forming a resistance response sequence after the test. S32, Based on the resistance response sequence after the trial, extract the resistance fall-off amplitude, resistance fall-off rate and resistance stability recovery degree to form a resistance fall-off feature group; S33, input the resistance drop feature group into the preset adjustment path determination rule set, calculate the path adaptation score of each candidate adjustment path, and determine the candidate adjustment path with the largest path adaptation score as the target adjustment path.
[0012] Optionally, S32 includes: S321, based on the resistance response sequence after the trial action is executed, extract the resistance value at the start of the trial action, and calculate the difference between the minimum resistance value and the initial resistance value within the observation window as the resistance drop amplitude; S322, calculate the resistance retracement rate based on the resistance retracement magnitude and the time interval taken for the resistance to reach its minimum value; S323: After the resistance falls back, the resistance fluctuation is statistically analyzed within the stable observation interval, the degree of resistance stability recovery is calculated, and the resistance fall amplitude, resistance fall rate and resistance stability recovery degree are combined to form a resistance fall characteristic group.
[0013] Optionally, S33 includes: S331, based on the resistance fallback feature group, the candidate adjustment paths corresponding to different attitude probing actions are identified, and the mapping relationship between each candidate adjustment path and its corresponding resistance fallback amplitude, resistance fallback rate and resistance stability recovery degree is established to form a candidate path feature set for path selection. S332, based on the resistance fallback feature group corresponding to each candidate adjustment path, normalize the resistance fallback amplitude, resistance fallback rate and resistance stability recovery degree, and perform weighted fusion according to preset weights to obtain the path adaptation score of each candidate adjustment path, so as to characterize the degree of adaptation of each adjustment path to the current resistance anomaly. S333: Compare the path fit scores of each candidate adjustment path, select the candidate adjustment path with the largest path fit score as the target adjustment path under the current excavation condition, and output the target adjustment path for the control of the excavator shovel attitude.
[0014] Optionally, S4 includes: S41, according to the determined target adjustment path, control the digging shovel to perform the corresponding attitude adjustment action. When the target adjustment path is digging angle yielding, adjust the digging angle of the digging shovel. When the target adjustment path is soil depth retraction, adjust the soil depth of the digging shovel. When the target adjustment path is shovel tip lifting to avoid, adjust the shovel tip lifting attitude. When the target adjustment path is short-distance vibration to get rid of attachment, control the digging shovel to perform the corresponding number of short-distance vibration actions, thereby completing the attitude adjustment that matches the current resistance anomaly. S42, after completing the adjustment of the excavator shovel posture, continuously collect the adjusted excavation resistance signal within the preset feedback observation window, and calculate the degree of resistance improvement based on the resistance difference before and after the adjustment to form a resistance feedback evaluation quantity that characterizes the execution effect of the target adjustment path. S43, based on the degree of resistance improvement and the adjusted resistance feedback sequence, update the resistance segment feature strength and path adaptation weight corresponding to the current resistance anomaly source type, and write the update result back to the resistance evolution fingerprint to obtain the updated resistance evolution fingerprint. In the next round of mining, the anomaly source type discrimination result and attitude trial action generation result are corrected based on the updated resistance evolution fingerprint to continuously optimize the subsequent mining attitude adjustment strategy.
[0015] The beneficial effects of this invention are: This invention continuously collects excavation resistance change signals during the excavation process and constructs a resistance evolution fingerprint based on the sequential combination relationship between the resistance surge segment, the resistance stagnation segment, and the resistance release segment. This enables a structured expression of the excavation resistance change process, transforming the originally difficult-to-use continuous resistance signal into fingerprint data with temporal characteristics and state semantics. As a result, it can accurately characterize the resistance evolution law under different obstructed working conditions, significantly improving the identification accuracy of complex soil environments and crop root system action states.
[0016] This invention, by matching and identifying resistance evolution fingerprints with preset anomaly source templates, and generating targeted attitude probing actions in conjunction with an attitude probing action library, and then constructing an adjustment path determination rule set based on the resistance fall amplitude, fall rate, and stability recovery degree after the probing actions, realizes the transformation from passive response adjustment to an active probing, feedback evaluation, and path optimization decision-making mechanism. This enables the excavator to adopt differentiated adjustment strategies for different anomaly sources such as hard soil compaction, rock collision, root entanglement, and wet soil adhesion, thereby improving the adaptability and operational stability of the excavation process.
[0017] This invention quantifies and evaluates the resistance feedback results after attitude adjustment and writes back the degree of resistance improvement to the resistance evolution fingerprint and adjustment path weight, forming a feedback-based dynamic update mechanism. This enables the system to continuously correct the anomaly source identification results and attitude trial action generation strategy during subsequent excavation control, achieving continuous adaptive optimization of excavation attitude adjustment. This effectively reduces repetitive and ineffective adjustments, improves work efficiency, and reduces mechanical shock and energy consumption. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 This is a schematic diagram of the adjustment method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the resistance evolution fingerprint construction in an embodiment of the present invention. Detailed Implementation
[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Those skilled in the art may employ other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.
[0021] like Figures 1-2 As shown, a method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback includes the following steps: S1, during the advance of the excavator shovel, continuously collect the excavation resistance change signal, and identify the sequential combination relationship between the resistance surge segment, the resistance stagnation segment and the resistance release segment, and construct the resistance evolution fingerprint of the current excavation object; S2, based on the resistance evolution fingerprint, determines the current resistance anomaly source type, including hard soil compaction, rock collision, root ball entanglement or wet soil adhesion, and generates corresponding attitude probing actions for different source types; S3 controls the digging shovel to perform a posture probing action, and determines the target adjustment path based on the resistance fall characteristics after the probing action. The target adjustment path includes at least one of the following: digging angle adjustment, soil penetration depth retraction, shovel tip lifting to avoid, or short-distance vibration to get rid of the center. S4. The excavator attitude is adjusted according to the target adjustment path, and the adjusted resistance feedback results are written back to update the resistance evolution fingerprint in order to continuously correct the subsequent excavation attitude.
[0022] S1 includes: S11, During the advance of the excavator shovel, force sensors installed at the connection or transmission force-bearing parts of the excavator shovel continuously collect the original excavation resistance signal in the excavation direction, and record it sequentially according to a preset sampling period. The collected original excavation resistance signal is then smoothed and its rate of change is calculated to generate a smoothed excavation resistance value and resistance rate of change that characterize the current excavation process resistance change state, expressed as: ; ; in, For the first The original values of excavation resistance were collected at each sampling time. For the first The smoothed excavation resistance value corresponding to each sampling time point To adjust the sliding window length, For the first The rate of change of resistance at each sampling time point The sampling time interval between two adjacent sampling points; S12, based on the smoothed excavation resistance value and resistance change rate, divides the excavation process into resistance segments according to the resistance amplitude level, resistance change rate and duration, including the resistance surge segment, the resistance stagnation segment and the resistance release segment; S13. The identified resistance surge segment, resistance stagnation segment, and resistance release segment are numbered according to time sequence. The sequential connection relationship between adjacent resistance segments, the duration of each segment, and the characteristic intensity of each segment are extracted and encoded into a resistance evolution fingerprint that characterizes the process of the current excavation object being obstructed. This allows the resistance evolution fingerprint to simultaneously characterize the resistance state type, the order of occurrence, and the strength of evolution.
[0023] The resistance section is divided into: When the rate of change of resistance is not less than the threshold for identifying a sudden increase in resistance, it is identified as a sudden increase in resistance. When the smoothed excavation resistance value is not less than the high resistance judgment threshold, the absolute value of the resistance change rate is not greater than the stable change rate threshold, and the duration is not less than the minimum duration, it is identified as a resistance viscous segment. When the rate of change of resistance is not greater than the threshold for identifying the negative resistance release segment, it is identified as a resistance release segment; The threshold for identifying the sudden increase in resistance is expressed as follows: ; in, To identify the threshold for the sudden increase in resistance, , These represent the mean and standard deviation of the positive resistance change rate samples within the current sliding window, respectively. This is a sudden increase in amplification factor; The high resistance threshold is expressed as follows: ; in, The threshold for determining high resistance. , These represent the average and standard deviation of the smoothed excavation resistance values within the current sliding window. It has a high drag amplification factor; The steady-state rate of change threshold is expressed as: ; in, To stabilize the rate of change threshold, , These are the average and standard deviation of the absolute values of the rate of change of resistance within the current sliding window, respectively. The adjustment coefficient is used to stabilize the threshold. The minimum duration is expressed as: ; in, For minimum duration, The reference length for determining the viscous state represents the minimum travel distance required to be considered as continuous adhesion or persistent hindrance in the excavation direction. This represents the current speed at which the excavator shovel advances. The threshold for identifying the resistance release segment is expressed as: ; in, Identify the threshold for the resistance release segment. , These represent the average and standard deviation of the rate of change of negative resistance within the current sliding window, respectively. To release the amplification factor.
[0024] S13 includes: S131, according to the order in which each resistance segment appears during the excavation process, the identified resistance surge segment, resistance stagnation segment, and resistance release segment are sequentially numbered. The segment type, start time, end time, and duration corresponding to each resistance segment are extracted to form the basic temporal feature set of each resistance segment, which characterizes the order of appearance and duration of different resistance states during the current excavation process, as shown below: ; in, For the first Duration of the resistance segment For the first The starting moment of the resistance segment, For the first The end of the resistance phase; S132, based on the smoothed excavation resistance value within the corresponding time interval of each resistance segment, calculate the deviation of each resistance segment from the average resistance level of the current sampling window, which serves as the feature intensity of the corresponding resistance segment. Combine the resistance segment type, start time, end time, duration, and feature intensity to construct the feature vector for each resistance segment, representing its state category, temporal location, and strength of change, as follows: ; ; in, For the first Characteristic strength of the resistance section For the first The number of sampling points contained within the resistance section. This represents the average resistance value within the current sampling window. For the first The characteristic vector of the resistance segment For the first The type of resistance segment is used to characterize whether the resistance segment belongs to a resistance surge segment, a resistance stagnation segment, or a resistance release segment; S133, the feature vectors of each resistance segment are arranged sequentially according to time to form a serialized feature set reflecting the type of resistance state, the order of occurrence, the duration characteristics, and the strength of change. This serialized feature set is determined as the resistance evolution fingerprint corresponding to the current excavation object, to characterize the resistance evolution law of the current excavation object in the processes of obstruction, stagnation, and release, expressed as: ; in, Generate a resistance evolution fingerprint for the current mining object. This represents the characteristic vectors of each resistance segment arranged in chronological order. This represents the total number of resistance segments identified within the current sampling window.
[0025] S2 includes: S21, input the resistance evolution fingerprint into the preset anomaly source template. Based on the average strength index of the resistance surge segment, the average duration index of the resistance viscous segment, the average strength index of the resistance release segment, and the sequential combination relationship of the resistance segments, calculate the matching score between the current resistance evolution fingerprint and each anomaly source template. The anomaly source template with the highest matching score is determined as the current resistance anomaly source type. Anomaly source types include hard soil compaction, rock collision, root ball entanglement, and wet soil adhesion, represented as: ; ; ; ; ; ; ; ; in, For the current resistance evolution fingerprint and the first The overall matching score between the templates of the exception source class , , These are the weight coefficients for sequential matching, intensity matching, and persistence matching, respectively. For the current resistance evolution fingerprint and the first Order matching items between exception source templates. For the current resistance evolution fingerprint and the first Strength matching items between templates representing different exception sources For the current resistance evolution fingerprint and the first Continuous matching items between exception source templates. The current resistance segment sequence and the first The length of the sequence of successful matches in the exception source template. For the first The standard ordinal length corresponding to the exception source template. For the first Reference strength index for the resistance surge segment corresponding to the anomaly source template. This is the average strength index for the period of sudden increase in resistance. This is the average strength index of the resistance release section. For the first Reference strength index for the resistance release segment corresponding to the anomaly source template. To prevent extremely small positive numbers with a denominator of zero, The number of segments with a sudden increase in resistance. This is the set of numbers corresponding to all segments of sudden increase in resistance. The number of resistance release segments, This is the set of numbers corresponding to all resistance release segments. This is the average duration index for the resistance viscous segment. For the first The reference persistence index for the resistance viscous segment corresponding to the anomaly source template. The number of viscous segments that resist resistance. This is the set of numbers corresponding to all resistance viscous segments. The template number representing the anomaly source with the highest overall matching score. These correspond to hard soil compaction, stone collision, root ball entanglement, and wet soil adhesion, respectively. S22, based on the current resistance anomaly source type determined by the judgment, retrieve the target test action parameters corresponding to the anomaly source type from the preset attitude test action library, and generate the attitude test action under the current working condition.
[0026] The anomaly source templates are constructed based on the resistance evolution patterns under different types of obstructed working conditions, including the resistance segment sequence pattern and the corresponding reference intensity index for the sudden increase segment, the reference duration index for the stagnant segment, and the reference intensity index for the release segment. Each anomaly source template is obtained by statistical analysis of historical mining operation data. Specifically, it is obtained by feature extraction and mean statistics of multiple sets of resistance evolution fingerprints under the same anomaly source type, thus forming a standardized reference template for anomaly source identification.
[0027] The attitude probing action library is constructed by statistically analyzing the experience and historical operation data of excavation attitude adjustment under different anomaly source types. It combines the excavation angle adjustment, soil penetration depth adjustment, shovel tip lifting or oscillation amplitude and oscillation execution number to form parameterized attitude probing action templates corresponding to different anomaly source types.
[0028] S22 includes: S221, based on the determined type of current resistance anomaly, select the attitude adjustment mode corresponding to the anomaly type from the preset attitude probing action library, and adaptively calculate the adjustment amount of digging angle, soil penetration depth, and shovel tip lifting or sway amplitude based on the comprehensive matching score corresponding to the current anomaly type, generating an attitude probing action that matches the current anomaly level, represented as: ; ; ; ; ; in, This is a probing action to test the posture. The current anomaly source type number, To adjust the excavation angle, This is the adjustment amount for the soil penetration depth. This refers to the lifting or oscillation amplitude of the shovel tip. The number of times the oscillation is executed. , , These are the corresponding adjustment coefficients. For the first The number of basic oscillation executions corresponding to the type of exception source. For the first The matching score adjustment coefficient corresponding to the anomaly source. For the first The hysteresis persistence adjustment coefficient corresponding to the anomaly source; S222, based on the determined anomaly source type, determine the attitude probing action adjustment mode corresponding to that anomaly source type, and limit the direction of action and dominant adjustment amount of each adjustment parameter in the attitude probing action based on the adjustment mode, so that different anomaly source types correspond to different attitude probing action generation strategies, specifically including: When the anomaly source type is hard soil compaction, the attitude probing action takes the excavation angle adjustment as the main adjustment, and the soil penetration depth adjustment and vibration adjustment as auxiliary adjustment. When the abnormal source type is rock collision, the attitude probing action mainly adjusts the shovel tip lift to achieve rapid avoidance; When the anomaly source type is root tangles, the attitude probing action mainly uses the oscillation adjustment amount, combined with small angle adjustment to achieve untangling; When the source of the anomaly is wet soil adhesion, the attitude probing action is mainly adjusted by the soil penetration depth retraction adjustment, combined with the oscillation adjustment to achieve desorption.
[0029] S3 includes: S31, based on the generated attitude test action, control the excavator to perform corresponding digging angle adjustment, soil penetration depth adjustment, shovel tip lifting adjustment, or short-distance oscillation adjustment, and continuously collect the digging resistance response signal within the preset observation window after the attitude test action, forming a resistance response sequence after the test, represented as: ; in, The resistance response sequence after the initial test. For the first The smoothed excavation resistance value corresponding to each sampling time point This is the first observation sampling moment after the attitude probing action begins. This represents the total number of sampling points within the observation window after the initial trial. S32, based on the resistance response sequence after the trial, extract the resistance pullback amplitude, resistance pullback rate and resistance stability recovery degree to form a resistance pullback feature group; S33, input the resistance pullback feature group into the preset adjustment path judgment rule set, calculate the path adaptation score of each candidate adjustment path, and determine the candidate adjustment path with the largest path adaptation score as the target adjustment path.
[0030] The rule set for determining the adjustment path is constructed by normalizing and weighting the magnitude of the resistance decline, the rate of the resistance decline, and the degree of resistance stability recovery. The weight of each indicator is determined by statistical analysis or online updates of historical adjustment effect data, thereby enabling the assessment of the suitability of different candidate adjustment paths and the selection of the optimal path.
[0031] S32 includes: S321, based on the resistance response sequence after the trial action, extract the resistance value at the start of the trial action, and calculate the difference between the minimum resistance value within the observation window and the initial resistance value, as the resistance drop amplitude, to characterize the degree of reduction of abnormal resistance by the current attitude trial action, expressed as: ; in, The magnitude of the pullback due to resistance. To test the smoothed excavation resistance value at the start of the action, For the first Smoothed excavation resistance value at each sampling time. This is the set of sampling points within the observation window after the initial probe. S322, based on the magnitude of the resistance drop and the time interval taken for the resistance to reach its minimum value, calculate the resistance drop rate to characterize the efficiency of the attitude probing action in resolving the resistance anomaly, expressed as: ; in, The rate of decline due to resistance. This refers to the sampling time when the resistance reaches its minimum value after the initial test. To test the timing of the start of the action, The sampling time interval; S323, after the resistance pullback is completed, the resistance fluctuation is statistically analyzed within the stable observation interval, and the degree of resistance stability recovery is calculated to characterize whether the system has entered a stable state after the resistance pullback. The resistance pullback amplitude, resistance pullback rate, and resistance stability recovery degree are combined to form a resistance pullback characteristic group, expressed as: ; ; in, To determine the degree of resistance to recovery. This is the set of sampling points corresponding to the stable observation interval after the initial trial. To stabilize the number of sampling points within the observation interval, To stabilize the average resistance value within the observation interval, To prevent extremely small positive numbers with a denominator of zero, This is a characteristic group of resistance and pullback.
[0032] S33 includes: S331, based on the resistance fallback feature set, candidate adjustment paths corresponding to different attitude probing actions are identified, and a mapping relationship is established between each candidate adjustment path and its corresponding resistance fallback amplitude, resistance fallback rate, and resistance stability recovery degree, to form a candidate path feature set for path selection, represented as: ; in, For the first The resistance pullback characteristic group corresponding to each candidate adjustment path For the first The magnitude of the resistance pullback corresponding to each candidate adjustment path For the first The rate of resistance decline corresponding to each candidate adjustment path For the first The degree of resistance stability recovery corresponding to each candidate adjustment path Number the candidate regulation path. These correspond to excavation angle clearance, soil penetration depth retraction, shovel tip lifting to avoid, and short-distance vibration to get rid of attachment, respectively. S332, based on the resistance fallback characteristic group corresponding to each candidate adjustment path, normalize the resistance fallback amplitude, resistance fallback rate, and resistance stability recovery degree, and perform weighted fusion according to preset weights to obtain the path adaptation score of each candidate adjustment path, which characterizes the degree of adaptation of each adjustment path to the current resistance anomaly, expressed as: ; in, For the first The path fit score of each candidate regulation path This represents the maximum value among the resistance pullback amplitudes corresponding to the current candidate paths. This represents the maximum resistance retracement rate among all candidate paths. , , These are the weighting coefficients for the magnitude of the resistance pullback, the rate of pullback, and the degree of stable recovery, respectively. S333: Compare the path fit scores of each candidate adjustment path, select the candidate adjustment path with the highest path fit score as the target adjustment path under the current excavation condition, and output the target adjustment path for excavator shovel attitude adjustment control, expressed as: ; in, Adjust the path number for the target. For the first The path fit score of each candidate regulation path.
[0033] S4 includes: S41, based on the determined target adjustment path, control the digging shovel to perform corresponding attitude adjustment actions. When the target adjustment path is digging angle yielding, adjust the digging angle of the digging shovel; when the target adjustment path is soil depth retraction, adjust the soil depth of the digging shovel; when the target adjustment path is shovel tip lifting to avoid obstruction, adjust the shovel tip lifting attitude; when the target adjustment path is short-distance vibration to break free from attachment, control the digging shovel to perform a corresponding number of short-distance vibration actions, thereby completing the attitude adjustment that matches the current resistance anomaly, represented as: ; ; in, Adjust the attitude amount corresponding to the target path. Adjust the excavation angle corresponding to the target path by adjusting the amount of adjustment. Adjust the depth of soil penetration corresponding to the target adjustment path. Adjust the amount of shovel tip lifting or oscillation amplitude corresponding to the target adjustment path. Adjust the number of oscillations corresponding to the target path. To adjust the attitude state vector of the excavator shovel. To adjust the attitude state vector of the front excavator shovel; S42, after completing the digging shovel attitude adjustment, continuously collect the adjusted digging resistance signal within the preset feedback observation window, and calculate the degree of resistance improvement based on the resistance difference before and after the adjustment, forming a resistance feedback evaluation quantity characterizing the execution effect of this target adjustment path, which is used for subsequent resistance evolution fingerprint updates, expressed as: ; ; ; in, For the adjusted resistance feedback sequence, For the first The smoothed mining resistance value corresponding to each feedback sampling time. To provide feedback on the starting sampling time of the observation window, To provide feedback on the total number of sampling points within the observation window, To provide feedback on the average resistance value within the observation window, The reference average resistance value before adjusting the path to achieve the target. To the degree of improvement in resistance, when A larger value indicates that the attitude adjustment has a better effect on reducing drag. A smaller value indicates that the effect of this attitude adjustment is limited; S43, based on the degree of resistance improvement and the adjusted resistance feedback sequence, update the resistance segment feature strength and path adaptation weight corresponding to the current resistance anomaly source type, and write the update result back into the resistance evolution fingerprint to obtain the updated resistance evolution fingerprint. In the next round of mining, the anomaly source type discrimination result and attitude probing action generation result are corrected based on the updated resistance evolution fingerprint to continuously optimize the subsequent mining attitude adjustment strategy, expressed as: ; ; ; ; in, For the first time before the update The characteristic strength of each resistance segment For the updated version The characteristic strength of each resistance segment This is the update coefficient for the characteristic strength of the resistance segment based on the resistance feedback results. Adjust the adaptation weights of the path to the target before updating. Adjust the adaptation weights of the path to the updated target. Update the path weight coefficients. For the updated resistance evolution fingerprint, For the updated number Each resistance segment feature vector For the first Type marking of each resistance segment, For the first The starting time of each resistance segment For the first The end of each resistance segment For the first The duration of each resistance segment This represents the total number of resistance segments in the current resistance evolution fingerprint.
[0034] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.
[0035] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
Claims
1. A method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback, characterized in that, Includes the following steps: S1, during the advance of the excavator shovel, continuously collect the excavation resistance change signal, and identify the sequential combination relationship between the resistance surge segment, the resistance stagnation segment and the resistance release segment, and construct the resistance evolution fingerprint of the current excavation object; S2, based on the resistance evolution fingerprint, determine the current resistance anomaly source type, including hard soil compaction, rock collision, root ball entanglement or wet soil adhesion, and generate corresponding attitude probing actions for different source types; S3, control the digging shovel to perform the posture probing action, and determine the target adjustment path based on the resistance fall characteristics after the probing action. The target adjustment path includes at least one of the following: digging angle adjustment, soil penetration depth reduction, shovel tip lifting to avoid, or short-distance vibration to get rid of the center. S4. The digging shovel attitude is adjusted according to the target adjustment path, and the adjusted resistance feedback result is written back to update the resistance evolution fingerprint in order to continuously correct the subsequent digging attitude.
2. The method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback according to claim 1, characterized in that, S1 includes: S11. During the excavator shovel advance, the force sensor installed at the excavator shovel connection or transmission force-bearing part continuously collects the original excavation resistance signal in the excavation direction, and records it in time sequence according to the preset sampling period. The collected original excavation resistance signal is subjected to sliding smoothing and change rate calculation to generate a smoothed excavation resistance value and resistance change rate that characterize the current excavation process resistance change state. S12, based on the smoothed excavation resistance value and resistance change rate, the excavation process is divided into resistance segments according to the resistance amplitude level, resistance change rate and duration, including the resistance surge segment, the resistance stagnation segment and the resistance release segment. S13. The identified resistance surge segment, resistance stagnation segment, and resistance release segment are numbered according to time sequence. The sequential connection relationship between adjacent resistance segments, the duration of each segment, and the characteristic intensity of each segment are extracted and encoded into a resistance evolution fingerprint that characterizes the process of the current excavation object being obstructed. This allows the resistance evolution fingerprint to simultaneously characterize the resistance state type, the order of occurrence, and the strength of evolution.
3. The method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback according to claim 2, characterized in that, The resistance section division includes: When the rate of change of resistance is not less than the threshold for identifying a sudden increase in resistance, it is identified as a sudden increase in resistance. When the smoothed excavation resistance value is not less than the high resistance judgment threshold, the absolute value of the resistance change rate is not greater than the stable change rate threshold, and the duration is not less than the minimum duration, it is identified as a resistance viscous segment. When the rate of change of resistance is not greater than the threshold for identifying the negative resistance release segment, it is identified as a resistance release segment.
4. The method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback according to claim 3, characterized in that, S13 includes: S131. According to the order of appearance of each resistance segment during the excavation process, the identified resistance surge segment, resistance stagnation segment and resistance release segment are sequentially numbered, and the segment type, start time, end time and duration corresponding to each resistance segment are extracted to form the basic temporal feature set of each resistance segment. S132, based on the smoothed mining resistance value within the time interval corresponding to each resistance segment, calculate the degree of deviation of each resistance segment from the average resistance level of the current sampling window, and use it as the feature intensity of the corresponding resistance segment. Combine the resistance segment type, start time, end time, duration and feature intensity to construct the feature vector of each resistance segment. S133, arrange the feature vectors of each resistance segment in chronological order to form a serialized feature set that reflects the resistance state type, occurrence order, persistence characteristics and strength of change, and determine this serialized feature set as the resistance evolution fingerprint corresponding to the current mining object.
5. The method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback according to claim 1, characterized in that, S2 includes: S21, input the resistance evolution fingerprint into the preset anomaly source template, calculate the matching score between the current resistance evolution fingerprint and each anomaly source template according to the average strength index of the resistance surge segment, the average duration index of the resistance viscous segment, the average strength index of the resistance release segment, and the sequential combination relationship of the resistance segments, and determine the type corresponding to the anomaly source template with the largest matching score as the current resistance anomaly source type. The anomaly source types include hard soil compaction, rock collision, root ball entanglement, and wet soil adhesion. S22, based on the determined type of current resistance anomaly, retrieve the target test action parameters corresponding to the anomaly source type from the preset attitude test action library, and generate the attitude test action under the current working condition.
6. The method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback according to claim 5, characterized in that, S22 includes: S221. Based on the current resistance anomaly source type determined by the judgment, select the attitude adjustment mode corresponding to the anomaly source type from the preset attitude probing action library, and based on the comprehensive matching score corresponding to the current anomaly source type, adaptively calculate the digging angle adjustment amount, soil penetration depth adjustment amount, shovel tip lifting or sway amplitude adjustment amount to generate attitude probing actions that match the current anomaly level. S222, based on the determined anomaly source type, determine the attitude probing action adjustment mode corresponding to that anomaly source type, and limit the direction of action and dominant adjustment amount of each adjustment parameter in the attitude probing action based on the adjustment mode, so that different anomaly source types correspond to different attitude probing action generation strategies, specifically including: When the abnormality source type is hard soil compaction, the attitude probing action uses the excavation angle adjustment as the main adjustment amount, and the soil penetration depth adjustment amount and the sway adjustment amount as auxiliary adjustment amounts. When the abnormal source type is stone collision, the attitude probing action is mainly adjusted by the shovel tip lifting adjustment amount to achieve rapid avoidance; When the anomaly source type is root tangles, the attitude probing action mainly uses the oscillation adjustment amount as the adjustment amount, and is combined with small angle adjustment to achieve untangling; When the abnormality source type is wet soil adhesion, the attitude probing action uses the soil penetration depth retraction adjustment amount as the main adjustment amount, and combines it with the sway adjustment amount to achieve desorption.
7. The method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback according to claim 6, characterized in that, S3 includes: S31, based on the generated attitude test action, control the excavator to perform corresponding digging angle adjustment, soil penetration depth adjustment, shovel tip lifting adjustment or short-distance oscillation adjustment, and continuously collect the digging resistance response signal within the preset observation time window after the attitude test action is performed, forming a resistance response sequence after the test. S32, Based on the resistance response sequence after the trial, extract the resistance fall-off amplitude, resistance fall-off rate and resistance stability recovery degree to form a resistance fall-off feature group; S33, input the resistance drop feature group into the preset adjustment path determination rule set, calculate the path adaptation score of each candidate adjustment path, and determine the candidate adjustment path with the largest path adaptation score as the target adjustment path.
8. The method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback according to claim 7, characterized in that, S32 includes: S321, based on the resistance response sequence after the trial action is executed, extract the resistance value at the start of the trial action, and calculate the difference between the minimum resistance value and the initial resistance value within the observation window as the resistance drop amplitude; S322, calculate the resistance retracement rate based on the resistance retracement magnitude and the time interval taken for the resistance to reach its minimum value; S323: After the resistance falls back, the resistance fluctuation is statistically analyzed within the stable observation interval, the degree of resistance stability recovery is calculated, and the resistance fall amplitude, resistance fall rate and resistance stability recovery degree are combined to form a resistance fall characteristic group.
9. A method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback, as described in claim 8, is characterized in that... S33 includes: S331, based on the resistance fallback feature group, the candidate adjustment paths corresponding to different attitude probing actions are identified, and the mapping relationship between each candidate adjustment path and its corresponding resistance fallback amplitude, resistance fallback rate and resistance stability recovery degree is established to form a candidate path feature set for path selection. S332, based on the resistance fallback feature group corresponding to each candidate adjustment path, normalize the resistance fallback amplitude, resistance fallback rate and resistance stability recovery degree, and perform weighted fusion according to preset weights to obtain the path adaptation score of each candidate adjustment path, so as to characterize the degree of adaptation of each adjustment path to the current resistance anomaly. S333: Compare the path fit scores of each candidate adjustment path, select the candidate adjustment path with the largest path fit score as the target adjustment path under the current excavation condition, and output the target adjustment path for the control of the excavator shovel attitude.
10. The method for adaptive adjustment of the posture of a harvesting shovel for root and tuber crops based on digging resistance feedback according to claim 9, characterized in that, S4 includes: S41, according to the determined target adjustment path, control the digging shovel to perform the corresponding attitude adjustment action. When the target adjustment path is digging angle yielding, adjust the digging angle of the digging shovel. When the target adjustment path is soil depth retraction, adjust the soil depth of the digging shovel. When the target adjustment path is shovel tip lifting to avoid, adjust the shovel tip lifting attitude. When the target adjustment path is short-distance vibration to get rid of attachment, control the digging shovel to perform the corresponding number of short-distance vibration actions, thereby completing the attitude adjustment that matches the current resistance anomaly. S42, after completing the adjustment of the excavator shovel posture, continuously collect the adjusted excavation resistance signal within the preset feedback observation window, and calculate the degree of resistance improvement based on the resistance difference before and after the adjustment to form a resistance feedback evaluation quantity that characterizes the execution effect of the target adjustment path. S43, based on the degree of resistance improvement and the adjusted resistance feedback sequence, update the resistance segment feature strength and path adaptation weight corresponding to the current resistance anomaly source type, and write the update result back to the resistance evolution fingerprint to obtain the updated resistance evolution fingerprint. In the next round of mining, the anomaly source type discrimination result and attitude trial action generation result are corrected based on the updated resistance evolution fingerprint to continuously optimize the subsequent mining attitude adjustment strategy.
