CN-121994095-A - Intelligent charging system for tunneling

CN121994095ACN 121994095 ACN121994095 ACN 121994095ACN-121994095-A

Abstract

The invention discloses an intelligent charging system and method for tunneling, and belongs to the field of manipulators and control systems thereof. The system comprises an environment sensing and data acquisition module, a space topology planning module, a master-slave vision collaborative guiding module, a pose compensation and alignment execution module, an intelligent push control module, a safety interlocking module, a safety installation module and a file recording module. According to the invention, a blasthole topological relation network is constructed through global scanning, an optimal sequence is planned, the mechanical arm is moved and pre-aiming is processed in parallel by combining master-slave vision collaborative guidance, tolerance alignment is realized by using a flexible guide sleeve in an execution stage, and a vibration gap-finding self-adaptive pushing mechanism based on real-time force feedback is introduced. The invention can effectively solve the problems of insufficient illumination, dust interference and the like in the severe tunnel environment, realize rapid, accurate and safe automatic charging, and improve the full-section charging operation efficiency, the single success rate and the autonomy of equipment operation.

Inventors

WANG ZHIGUO
Xing Wanghui
HOU LIPING
MA TIANLIANG
LI JING
HAN ZAO

Assignees

山西壶化集团爆破有限公司
山西壶化集团股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260408

Claims (10)

1. An intelligent charging system for tunnelling, the system comprising: the environment sensing and data acquisition module is used for acquiring the global depth image of the tunnel blasting working face acquired by the global vision device and acquiring blast hole design data of the working face; The space topology planning module is used for generating a blasthole space topology relation network comprising all blasthole nodes and space relations among the nodes based on the global depth image and blasthole design data, and generating a serialization operation instruction for guiding the action of the mechanical arm according to the blasthole space topology relation network planning; The master-slave vision collaborative guiding module is used for driving the mechanical arm to move towards the target blast hole according to the serialization operation instruction and synchronously starting a master-slave vision collaborative guiding process, wherein the master-slave vision collaborative guiding process is used for fusing first positioning data acquired by the global vision device and second positioning data acquired by an end vision device mounted at the tail end of the mechanical arm to generate an end pose dynamic adjustment instruction; The pose compensation and alignment execution module is used for responding to the terminal pose dynamic adjustment instruction, controlling an actuator at the tail end of the mechanical arm to move towards the blast hole, and completing tolerance alignment of the tube by contacting the flexible guide sleeve at the front end of the actuator with the blast hole; The intelligent pushing control module is used for sending a vibration gap searching pushing instruction after the tolerance alignment is finished, receiving the thrust and speed feedback data which are monitored and fed back in real time by an internal sensor of the pipe feeder, generating a self-adaptive pushing control instruction or a constant-speed pushing control instruction, and controlling the pipe feeder to finish pushing operation; The safety interlocking and mounting module is used for acquiring an in-situ confirmation signal of the explosive tube reflecting the in-situ state of the explosive tube and an in-situ confirmation signal of the initiating explosive reflecting the state of the material supplied by the initiating explosive, and controlling the mechanical arm to execute the mounting operation of the initiating explosive based on the safety interlocking logic; and the file recording module is used for recording and uploading the charging process file containing the thrust and speed feedback data.
2. An intelligent charging system for tunnelling as claimed in claim 1, wherein the spatial topology planning module comprises: The initial pose extraction unit is used for processing the global depth image by utilizing a target detection algorithm and identifying initial pose sets of all visible blastholes in the visual field range; The space matching and complementing unit is used for carrying out space registration on the initial pose set and the blasthole design data through a registration algorithm, correcting the blasthole design data by applying a transformation matrix, and constructing a blasthole space topological relation network containing all target blasthole real poses; The moving cost calculation unit is used for calculating the moving cost between any two blasthole nodes in the blasthole space topological relation network, and the moving cost is formed by weighting the space distance between the two nodes and the expected recognition confidence of the terminal vision device on the target blasthole node; And the operation instruction planning unit is used for applying a path searching algorithm, solving and traversing paths with minimum total movement cost of all blasthole nodes according to the movement cost, and generating a serialized operation instruction.
3. An intelligent charging system for tunnelling as claimed in claim 1, wherein the master-slave visual co-pilot module comprises: The macroscopic offset tracking unit is used for continuously tracking the spatial position of the tail end of the mechanical arm relative to the current target blast hole group region through the global vision device so as to acquire a rough spatial offset as first positioning data; The local fine scanning unit is used for carrying out three-dimensional scanning on the current target blast hole through the tail end vision device in the moving process of the mechanical arm so as to acquire fine point cloud data of the blast hole as second positioning data; The data fusion estimation unit is used for carrying out fusion processing on the rough spatial offset and the fine point cloud data by applying a Kalman filtering algorithm so as to estimate the real-time relative pose of the actuator relative to a target gun orifice; And the track dynamic correction unit is used for comparing the real-time relative pose with a preset optimal observation pose to calculate pose errors, and generating a speed instruction for driving each joint of the mechanical arm to move according to the pose errors and taking the speed instruction as a terminal pose dynamic adjustment instruction.
4. An intelligent charging system for tunnelling as claimed in claim 1, wherein the pose compensation and alignment execution module comprises: the pose error compensation unit is used for driving the tail end of the mechanical arm to perform translation and rotation compensation according to the tail end pose dynamic adjustment instruction so as to align the central axis of the actuator with the orifice normal vector of the target blast hole; The flexible contact feedback unit is used for judging whether the actuator and the gun hole are in reliable flexible sealing connection or not by monitoring the compression deformation state of the flexible guide sleeve in the process that the actuator approaches the gun hole; And the tolerance alignment checking unit is used for confirming that the passage of the medicine tube entering the blast hole is communicated within a preset allowable radial deviation range based on the physical limit space of the flexible guide sleeve.
5. An intelligent charging system for tunnelling as claimed in claim 1, wherein the intelligent push control module comprises: the vibration excitation control unit is used for controlling the pipe feeder to perform high-frequency micro-vibration with set vibration frequency and vibration amplitude; the initial low-speed pushing unit is used for driving the tube feeder to push the medicine tube forwards at a set initial low speed in the high-frequency micro-vibration state; And the antifriction and clearance-seeking guiding unit is used for guiding the medicine tube to advance along a path with minimum resistance in the initial section of the blast hole through antifriction effect generated by the coupling effect of the high-frequency micro-vibration and the initial low speed.
6. An intelligent charging system for tunnelling as claimed in claim 1, wherein the safety interlock and mounting module comprises: The pre-condition checking unit is used for checking whether an in-place confirmation signal of the explosive tube reflecting the in-place of the explosive tube in the actuator and an in-place confirmation signal of the initiating explosive charge reflecting the sufficient material of a target station of the initiating explosive charge supply device are received simultaneously before a control instruction for grabbing the initiating explosive charge is generated; The instruction authorization control unit is used for authorizing the execution of the control instruction for grabbing the initiating explosive only when the verification results of the pre-condition verification unit are all valid; And the grabbing state confirming unit is used for checking grabbing success feedback signals from the actuator clamping jaw after executing the control instruction for grabbing the primary explosive, and judging whether the subsequent primary explosive installation operation is continuously executed or not.
7. An intelligent charge system for tunnelling as claimed in claim 5 wherein the intelligent push control module further comprises: The blocking state judging unit is used for judging that blocking occurs and triggering a blocking response strategy when the thrust and speed feedback data meet preset blocking conditions of abnormal increase of the thrust and abnormal decrease of the speed; the adaptive instruction generation unit is used for responding to the resistance response strategy and generating an adaptive pushing control instruction comprising a back pumping instruction, a vibration mode adjustment instruction and a secondary pushing instruction; And the strategy iteration execution unit is used for controlling the pipe feeder to execute the back-pumping instruction to relieve stress and execute a secondary propulsion instruction to try to surmount the obstacle after adjusting the vibration frequency or the vibration amplitude.
8. An intelligent charging system for tunnelling as claimed in claim 1, wherein the system further comprises a visual self-maintenance module: The confidence real-time computing unit is used for computing the continuous recognition confidence of the target blast hole output by the tail end vision device in real time; the pollution event judging unit is used for judging that a lens pollution triggering event occurs when the continuous recognition confidence is lower than a set threshold value and meets a preset degradation characteristic on the premise that the ambient light condition is optimized; And the self-cleaning execution unit is used for responding to the lens pollution triggering event and automatically controlling a cleaning mechanism arranged on the end vision device protective cover to execute cleaning action.
9. An intelligent charging system for tunnelling as claimed in claim 8, wherein the visual self-maintenance module further comprises: the gray feature analysis unit is used for acquiring gray histogram distribution features of the current image acquired by the terminal vision device; the light supplementing signal generating unit is used for generating a dynamic light supplementing control signal according to the difference between the gray level histogram distribution characteristics and a preset ideal contrast interval; And the brightness closed-loop adjusting unit is used for adjusting the brightness of the light supplementing lamp according to the dynamic light supplementing control signal so that the contrast of the current image approaches to the target interval, and improving the continuous recognition confidence by improving illumination.
10. An intelligent charging method for tunnelling, the method comprising: acquiring a global depth image of a tunnel blasting working surface acquired by a global vision device, and acquiring blast hole design data of the working surface; Based on the global depth image and the blasthole design data, generating a blasthole space topological relation network comprising all blasthole nodes and space relations among the nodes, and generating a serialization operation instruction for guiding the action of the mechanical arm according to the blasthole space topological relation network plan; Driving a mechanical arm to move towards a target blast hole according to the serialization operation instruction, and synchronously starting a master-slave vision collaborative guiding process, wherein the master-slave vision collaborative guiding process is used for fusing first positioning data acquired by the global vision device and second positioning data acquired by an end vision device mounted at the tail end of the mechanical arm to generate an end pose dynamic adjustment instruction; responding to the terminal pose dynamic adjustment instruction, controlling an actuator at the terminal of the mechanical arm to move towards the muzzle, and completing tolerance alignment of the tube by contacting the flexible guide sleeve at the front end of the actuator with the muzzle; after the tolerance alignment is finished, a vibration gap searching pushing instruction is sent, thrust and speed feedback data which are fed back by an internal sensor of the pipe feeder in real time are received, and an adaptive pushing control instruction or a constant-speed pushing control instruction is generated to control the pipe feeder to finish pushing operation; Acquiring an in-situ confirmation signal of the explosive tube reflecting the in-situ state of the explosive tube and an in-situ confirmation signal of the initiating explosive reflecting the state of the material supplied by the initiating explosive, and controlling the mechanical arm to execute the installation operation of the initiating explosive based on the safety interlocking logic; and recording and uploading a charging process file containing the thrust and speed feedback data.

Description

Intelligent charging system for tunneling Technical Field The invention relates to the field of manipulators and control systems thereof, in particular to an intelligent charging system for tunneling. Background Tunneling is a key link in infrastructure construction, wherein the drilling and blasting method is a common construction method. In the tunneling blasting process, the charge quality directly influences the blasting effect and the construction safety. In order to improve the working efficiency, the use of industrial robots or mechanical arms to replace manual work for blast hole positioning, cartridge conveying and primary explosive charge installation has become an important development direction in the field of tunnel engineering automation. In the related art, chinese patent publication No. CN111664762A discloses an automatic charging control system and method of a mixed loading explosive truck based on drilling positioning parameters, wherein the system comprises a data acquisition module, a calculation module and a charging positioning module. The technology comprises the steps of acquiring three-dimensional coordinates and hole depth of a blast hole when a drilling machine drills, calculating the explosive loading amount corresponding to each blast hole, positioning each time a loading person moves a loading hose to one blast hole, and matching the positioning with the acquired coordinates, so that an explosive truck is controlled to load the explosive with the corresponding explosive loading amount into the blast hole. In view of the above related art, the inventor considers that the technical requirement on personnel is reduced, but the technical problem still exists that firstly, the charging positioning depends on the manual operation of a hose to an orifice by a charging personnel, the automation degree is limited, and the efficiency is lower in large-scale full-face operation. Secondly, the technology adopts a rigid pushing and matching mode, has extremely high requirements on the regularity of blast holes, is extremely easy to cause blocking or positioning deviation of a charging hose when the charging hose is interfered by irregular blast holes or rock fragments in a tunnel, leads to operation interruption, lacks effective environment sensing and fault self-recovery capability, and is not beneficial to long-term stable use under complex and changeable tunnel working conditions. Disclosure of Invention In order to solve the problems, the invention provides an intelligent charging system for tunneling, which adopts the technical scheme of combining space topology planning, master-slave vision collaborative guiding, flexible tolerance alignment and intelligent pushing control, and can control a mechanical arm to realize efficient, accurate and safe automatic charging operation. In order to achieve the above purpose, the application adopts the following technical scheme: The first aspect provides an intelligent medicine loading system for tunneling, comprising an environment sensing and data acquisition module, a control module and a control module, wherein the environment sensing and data acquisition module is used for acquiring a global depth image of a tunnel blasting working face acquired by a global vision device and acquiring blast hole design data of the working face; the system comprises a global depth image and a blasthole design data, a spatial topology planning module, a master-slave visual collaborative guiding module, an intelligent push control module, a constant speed control and self-adaptive push control module, a position and orientation compensation and alignment execution module, a speed control module and a self-adaptive push control module, wherein the global depth image and the blasthole design data are used for generating a blasthole spatial topological relation network comprising all blasthole nodes and spatial relations among the nodes, the blasthole spatial topological relation network plans to generate a sequencing operation instruction for guiding the action of a mechanical arm according to the blasthole spatial topological relation network, the master-slave visual collaborative guiding module is used for driving the mechanical arm to move towards a target blasthole according to the sequencing operation instruction, and synchronously starting a master-slave visual collaborative guiding process, the master-slave visual collaborative guiding process is used for fusing first positioning data acquired by the global visual device and second positioning data acquired by an end visual device mounted at the tail end of the mechanical arm to generate a tail end pose dynamic adjustment instruction, the pose compensation and alignment execution module is used for responding to the tail end pose dynamic adjustment instruction, an executor at the tail end of the mechanical arm is controlled to move towards a blasthole opening, and the front end of the mechanical arm is con