CN-122007045-A - Foreign matter sorting cooperative control method and system based on hanging rail type double-arm mechanical arm

CN122007045ACN 122007045 ACN122007045 ACN 122007045ACN-122007045-A

Abstract

The invention discloses a foreign matter sorting cooperative control method and system based on a hanging rail type double-arm mechanical arm, and relates to the technical field of industrial control. The method comprises the steps of collecting electromagnetic induction signals of a material flow and surface images of the material flow, fusing, recognizing target foreign matters in the material flow, generating foreign matter perception information, predicting predicted motion tracks of each target foreign matters according to the foreign matter perception information and vibration motion parameters of a vibration distributor, planning non-interference sorting tracks for a first mechanical arm and a second mechanical arm respectively based on attribute types of the target foreign matters and corresponding predicted motion tracks, determining execution priorities, and controlling the first mechanical arm and the second mechanical arm according to the execution priorities and the sorting tracks. The invention realizes high-precision identification, intelligent collaborative sorting and efficient conflict-free operation of foreign matters in a dynamic complex environment.

Inventors

LUO QISHENG
HE LEI
ZHAO YANQIU
WANG WENSHAN
LIU LINGYUN
YANG YUHAO

Assignees

安徽理工大学

Dates

Publication Date: 20260512
Application Date: 20260325

Claims (10)

1. The foreign matter sorting cooperative control method based on the suspended track type double-arm mechanical arm is characterized by comprising the following steps of: Synchronously acquiring electromagnetic induction signals and surface images of the material flow through an electromagnetic sensor and a high-speed vision sensor which are arranged above the vibration distributing device; Carrying out fusion processing on the electromagnetic induction signals and the surface images of the material flow, identifying target foreign matters in the material flow, and generating foreign matter perception information containing attribute categories, spatial positions and time stamps of each target foreign matters, wherein the attribute categories comprise metal foreign matters or nonmetal foreign matters; Predicting a predicted motion track of each target foreign object in a future period according to the foreign object perception information and the vibration motion parameters of the vibration distributor, and planning a non-interference sorting track for the first mechanical arm and the second mechanical arm and determining an execution priority based on the attribute category of the target foreign object and the corresponding predicted motion track; according to the execution priority and the sorting track, the first mechanical arm is controlled to adsorb and remove the metal foreign matters through the magnetic actuator at the tail end, and the second mechanical arm is controlled to grasp and remove the nonmetal foreign matters through the clamping jaw actuator at the tail end.
2. The cooperative control method for sorting foreign matters based on a hanger rail type double-arm mechanical arm according to claim 1, wherein the electromagnetic induction signal and the surface image of the material flow are collected synchronously by an electromagnetic sensor and a high-speed vision sensor which are arranged above the vibration distributor, comprising: according to the environment illumination condition above the vibration distributor, dynamically adjusting the gain parameter, the exposure time parameter and the light supplementing intensity of the light supplementing lamp of the high-speed vision sensor to obtain an image acquisition parameter which is adapted to the current environment illumination condition; Under the image acquisition parameters, controlling the high-speed vision sensor to acquire a material flow surface image of the material flow; And synchronously controlling the electromagnetic sensor to acquire electromagnetic induction signals of the material flow, and aligning the electromagnetic induction signals with an acquisition time stamp of the surface image of the material flow.
3. The cooperative control method for foreign matter sorting based on a suspended track type double-arm mechanical arm according to claim 2, wherein dynamically adjusting gain parameters, exposure time parameters and light supplementing intensity of a light supplementing lamp of the high-speed vision sensor according to an ambient lighting condition above the vibration distributor to obtain an image acquisition parameter adapted to a current ambient lighting condition comprises the following steps: Acquiring an average brightness value and an average contrast value of a monitoring area above the vibration distributing device; Calculating the brightness deviation between the average brightness value and a preset standard brightness value, and calculating the contrast deviation between the average contrast value and the preset standard contrast value; Adjusting a gain parameter and an exposure time parameter of the high-speed vision sensor based on the brightness deviation, wherein the adjustment directions of the gain parameter and the exposure time parameter are opposite to the brightness deviation; And adjusting the light supplementing intensity of the light supplementing lamp based on the contrast deviation, wherein the adjusting direction of the light supplementing intensity is opposite to the direction of the contrast deviation.
4. The cooperative control method for foreign matter sorting based on a suspended track type double-arm mechanical arm according to claim 1, wherein the steps of performing fusion processing on the electromagnetic induction signals and the surface images of the material flow, identifying target foreign matters in the material flow, and generating foreign matter perception information including attribute types, spatial positions and time stamps of each target foreign matter, comprise: Based on the material flow surface image, acquiring a three-dimensional volume estimated value and surface visual texture characteristics of each material to be identified in the material flow through visual analysis; synchronously analyzing the electromagnetic induction signals to obtain electromagnetic response intensity of the spatial position of each material to be identified; The three-dimensional volume estimated value, the surface visual texture characteristics and the electromagnetic response intensity of each material to be identified are input into a pre-trained multi-stage classification model together, wherein the multi-stage classification model comprises a first-stage classification module and a second-stage classification module which are connected in series; the first-stage classification module divides each material to be identified into normal materials or target foreign matters according to the three-dimensional volume estimation value and the surface visual texture characteristics; the second-stage classification module combines the electromagnetic response intensity of the target foreign matters, classifies the target foreign matters into metal foreign matters or nonmetal foreign matters, and determines corresponding space position coordinates; and generating foreign matter perception information comprising attribute categories, spatial positions and acquisition time stamps of each target foreign matter according to the classification result and the spatial position coordinates output by the second-stage classification module.
5. The cooperative control method for foreign matter sorting based on a suspended track type double-arm mechanical arm according to claim 4, wherein the step of obtaining the three-dimensional volume estimated value and the surface visual texture feature of each material to be identified in the material flow through visual analysis based on the material flow surface image comprises the following steps: Reconstructing three-dimensional point cloud data of the material flow surface by adopting a grating deformation analysis method based on the material flow surface image; Dividing the three-dimensional point cloud data, and separating out a point cloud subset of each material to be identified in the material flow according to the space aggregation characteristics of the three-dimensional point cloud data; calculating a three-dimensional volume estimated value of each material to be identified based on the partitioned point cloud subsets; extracting a two-dimensional image area matched with the projection area of the material to be identified from the material flow surface image aiming at each material to be identified; and extracting the surface visual texture characteristics of each two-dimensional image area by using a convolutional neural network, wherein the surface visual texture characteristics comprise roughness, contrast and directivity values.
6. The cooperative control method for foreign matter sorting based on a suspended track type double-arm mechanical arm according to claim 1, wherein predicting a predicted motion trajectory of each target foreign matter in a future period according to the foreign matter sensing information and a vibration motion parameter of the vibration distributor comprises: Acquiring real-time vibration phase signals of a vibration distributor driving motor, and inclination angle parameters, vibration frequency parameters and amplitude parameters of a vibration distributor bed surface in real time; calculating a throwing index under the current vibration condition based on the vibration frequency parameter and the amplitude parameter; determining a critical throwing phase angle at which the material starts to perform throwing motion according to the throwing index and the friction coefficient between the material and the bed surface of the vibration distributor; Comparing the current phase angle of the real-time vibration phase signal with the critical throwing phase angle, if the current phase angle is smaller than the critical throwing phase angle, judging that the target foreign matter is in a sliding state contacting with the bed surface, and if the current phase angle is larger than or equal to the critical throwing phase angle, judging that the target foreign matter is in a throwing state thrown by the bed surface; If the target foreign matter is judged to be in a sliding state, calculating a predicted motion track of the target foreign matter in real time based on a coulomb friction dynamics equation according to the spatial position coordinates, the real-time vibration phase signals, the vibration frequency parameters, the amplitude parameters and the attribute types of the target foreign matter in the foreign matter perception information; If the target foreign matter is in the throwing state, the spatial position coordinates, the real-time vibration phase signals, the inclination angle parameters, the vibration frequency parameters, the amplitude parameters and the attribute types of the target foreign matter in the foreign matter sensing information are input into the pre-trained throwing track prediction intelligent agent together, and the predicted motion track of the target foreign matter in a future period is obtained through output.
7. The cooperative control method for foreign matter sorting based on a suspended track type double-arm mechanical arm according to claim 6, wherein the training process of the projection track prediction intelligent agent comprises the following steps: Constructing a projection track prediction intelligent agent based on machine learning; based on the physical parameters of the vibration distributor and the typical foreign matters, constructing a parameterized digital twin simulation system; in the digital twin simulation system, a sample data set is formed by randomly setting sample space position coordinates, sample vibration motion parameters, foreign matter attributes and initial states, running high-fidelity simulation and generating corresponding real casting tracks in batches; And training the cast track prediction intelligent body by using the sample data set to obtain the trained cast track prediction intelligent body.
8. The cooperative control method for foreign matter sorting based on suspended track type double-arm mechanical arm according to claim 1, wherein planning non-interference sorting tracks for the first mechanical arm and the second mechanical arm and determining execution priority based on the attribute type of the target foreign matter and the corresponding predicted motion track respectively comprises: Calculating the departure time of each target foreign object from the effective sorting working area of the mechanical arm based on the space position of each target foreign object on the vibration distributing device, the motion direction of the vibration distributing device and the corresponding predicted motion track; Determining execution priority for the first mechanical arm and the second mechanical arm according to the attribute category of each target foreign object and the corresponding leaving time, wherein the execution priority is used for designating the execution sequence of the first mechanical arm and the second mechanical arm, so that only one mechanical arm is allowed to execute sorting action at any moment; And sequentially driving the first mechanical arm or the second mechanical arm according to the determined execution priority, and planning a sorting track from the current pose to the corresponding target foreign object grabbing point for the currently driven mechanical arm based on a space-time planning algorithm in a lifting track space shared by the first mechanical arm and the second mechanical arm.
9. The cooperative control method for foreign matter sorting based on suspended track type double-arm mechanical arms according to claim 8, wherein determining the execution priority for the first mechanical arm and the second mechanical arm according to the attribute category and the corresponding departure time of each target foreign matter comprises: counting the first quantity of metal foreign matters and the second quantity of non-metal foreign matters in all the current identified but unfinished sorted target foreign matters; Normalizing the first quantity and the second quantity to obtain a first quantity normalized value and a second quantity normalized value respectively, wherein the first quantity normalized value is equal to the second quantity divided by the sum of the first quantity and the second quantity, and the second quantity normalized value is equal to the first quantity divided by the sum of the first quantity and the second quantity; taking the first quantity normalized value as a basic priority value of a metal foreign matter sorting task, and taking the second quantity normalized value as a basic priority value of a nonmetal foreign matter sorting task; According to the attribute category of the target foreign matters, respectively calculating a first average value of the departure time of all the metal foreign matters and a second average value of the departure time of all the nonmetal foreign matters; Normalizing the first average value and the second average value to respectively obtain a first emergency degree parameter corresponding to a metal foreign matter sorting task and a second emergency degree parameter corresponding to a nonmetallic foreign matter sorting task; the method comprises the steps of carrying out weighted fusion on a basic priority value of a metal foreign matter sorting task and a first emergency degree parameter to obtain a comprehensive priority score of the metal foreign matter sorting task, and carrying out weighted fusion on the basic priority value of a non-metal foreign matter sorting task and a second emergency degree parameter to obtain the comprehensive priority score of the non-metal foreign matter sorting task, wherein in the weighted fusion process, the weight of the emergency degree parameter is larger than that of the basic priority value; And comparing the comprehensive priority scores of the metal foreign matter sorting tasks with the comprehensive priority scores of the non-metal foreign matter sorting tasks, wherein the mechanical arm corresponding to the sorting task with the higher comprehensive priority score has the priority execution weight.
10. A co-control system for sorting foreign matter based on a suspended track type double arm mechanical arm, characterized by being used for executing the co-control method for sorting foreign matter based on a suspended track type double arm mechanical arm according to any one of claims 1 to 9, comprising: the sensor acquisition module is used for synchronously acquiring electromagnetic induction signals of the material flow and surface images of the material flow through an electromagnetic sensor and a high-speed vision sensor which are arranged above the vibration distributor; The information fusion and identification module is used for carrying out fusion processing on the electromagnetic induction signals and the surface images of the material flow, identifying target foreign matters in the material flow, and generating foreign matter perception information containing attribute types, spatial positions and time stamps of each target foreign matter, wherein the attribute types comprise metal foreign matters or nonmetal foreign matters; The track prediction and planning module is used for predicting the predicted motion track of each target foreign object in a future period according to the foreign object sensing information and the vibration motion parameters of the vibration distributor, planning a non-interference sorting track for the first mechanical arm and the second mechanical arm respectively based on the attribute type of the target foreign object and the corresponding predicted motion track, and determining the execution priority; And the sorting execution control module is used for controlling the first mechanical arm to adsorb and remove the metal foreign matters through the magnetic actuator at the tail end according to the execution priority and the sorting track, and controlling the second mechanical arm to grasp and remove the nonmetal foreign matters through the clamping jaw actuator at the tail end.

Description

Foreign matter sorting cooperative control method and system based on hanging rail type double-arm mechanical arm Technical Field The invention relates to the technical field of industrial control, in particular to a cooperative foreign matter sorting control method and system based on a hanging rail type double-arm mechanical arm. Background In the industrial automatic production process, aiming at the scenes of bulk material treatment such as coal dressing, ore sorting and the like, metal or nonmetal foreign matters are mixed in the material flow frequently, the purity and quality of the final product can be reduced due to the existence of various foreign matters, and key equipment such as downstream crushing, conveying and the like can be worn or even seriously damaged, so that continuous and stable operation of the whole production line is affected. At present, the prior art scheme for sorting foreign matters in material flows is mostly dependent on a single type of sensor, for example, a vision sensor or a metal detector is singly used, and obvious limitations exist in a complex dynamic environment, namely, the vision scheme is difficult to reliably distinguish nonmetallic foreign matters with similar colors and textures from normal materials and is easy to be interfered by illumination change and dust, and the metal detection scheme cannot identify nonmetallic foreign matters. In addition, the existing mechanical arm sorting system usually adopts fixed time sequence or simple rules to grasp, and lacks accurate prediction of material flow dynamic motion, particularly complex motion such as throwing, sliding and the like caused by a vibration distributor, so that the grasping success rate is low. When a plurality of mechanical arms are deployed for cooperation, the existing method often lacks an effective task allocation and anti-collision track planning mechanism, so that the problems of mutual interference of the mechanical arms, waiting for idle consumption and the like are easily generated, and dynamic optimization and balance among sorting efficiency, operation safety and system energy consumption cannot be realized. Disclosure of Invention The invention provides a cooperative control method and a cooperative control system for foreign matter sorting based on a suspended track type double-arm mechanical arm, aiming at the technical problems of incomplete foreign matter identification category and insufficient reliability caused by single sensor dependence, inaccurate prediction of dynamic material flow motion, low sorting efficiency, easy collision and high energy consumption in the prior art. The technical scheme for solving the technical problems is as follows: in a first aspect, the present invention provides a cooperative control method for foreign matter sorting based on a suspended rail type double-arm mechanical arm, comprising: Synchronously acquiring electromagnetic induction signals and surface images of the material flow through an electromagnetic sensor and a high-speed vision sensor which are arranged above the vibration distributing device; Carrying out fusion processing on the electromagnetic induction signals and the surface images of the material flow, identifying target foreign matters in the material flow, and generating foreign matter perception information containing attribute categories, spatial positions and time stamps of each target foreign matters, wherein the attribute categories comprise metal foreign matters or nonmetal foreign matters; Predicting a predicted motion track of each target foreign object in a future period according to the foreign object perception information and the vibration motion parameters of the vibration distributor, and planning a non-interference sorting track for the first mechanical arm and the second mechanical arm and determining an execution priority based on the attribute category of the target foreign object and the corresponding predicted motion track; according to the execution priority and the sorting track, the first mechanical arm is controlled to adsorb and remove the metal foreign matters through the magnetic actuator at the tail end, and the second mechanical arm is controlled to grasp and remove the nonmetal foreign matters through the clamping jaw actuator at the tail end. In a second aspect, the present invention provides a cooperative control system for foreign matter sorting based on a suspended rail type double-arm mechanical arm, comprising: the sensor acquisition module is used for synchronously acquiring electromagnetic induction signals of the material flow and surface images of the material flow through an electromagnetic sensor and a high-speed vision sensor which are arranged above the vibration distributor; The information fusion and identification module is used for carrying out fusion processing on the electromagnetic induction signals and the surface images of the material flow, identifying target foreign matters i