CN-122015801-A - Strip mine unmanned mine car positioning method and system based on multi-source heterogeneous sensor fusion

CN122015801ACN 122015801 ACN122015801 ACN 122015801ACN-122015801-A

Abstract

The invention discloses a method and a system for positioning an unmanned mine car of a strip mine based on multi-source heterogeneous sensor fusion, which relate to the technical field of unmanned mine car positioning, and are used for determining a plurality of multi-source heterogeneous sensors based on detection of a space region, wherein the multi-source heterogeneous sensors are distributed on the periphery of the unmanned mine car and output corresponding induction data, and fusion data combination is determined based on the plurality of induction data, corresponding data types and the overall morphology of the unmanned mine car, so that the accuracy of fusion data combination is improved. The method comprises the steps of determining a current positioning event of the unmanned mine car based on fusion data combination, a current image of the unmanned mine car and a prediction range of the unmanned mine car relative to the space area, determining a plurality of positioning information based on detection of the prediction positioning event of the unmanned mine car, and determining the current positioning position of the unmanned mine car according to the plurality of positioning information, the corresponding positioning dimension and the overall form of the unmanned mine car, so that the accuracy of the current positioning position of the unmanned mine car is improved.

Inventors

ZHAO HONGZE
Earl de Mutu
LUO XIAO
Huang Shouzhang
SHI YANYAN
Kang Chengxiong
YANG ZIYU

Assignees

中国矿业大学(北京)
准能集团
天地科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20251215

Claims (10)

1. The method for positioning the unmanned mine car of the strip mine based on the multi-source heterogeneous sensor fusion is characterized by comprising the following steps of: In an internal conveying channel of a mine, acquiring a moving route of the unmanned mine car, determining a plurality of mobile nodes based on the identification of the moving route of the unmanned mine car, determining a plurality of dynamic images of the unmanned mine car according to the tracing of the plurality of mobile nodes, and determining a current image of the unmanned mine car based on the plurality of dynamic images and corresponding shooting time; Determining a spatial region of the unmanned mine car relative to the mine according to the identification of the current image of the unmanned mine car, and determining a plurality of multi-source heterogeneous sensors based on the detection of the spatial region; The system comprises a plurality of heterogeneous sensors, a fusion data combination, a fusion data acquisition unit and a fusion data acquisition unit, wherein the heterogeneous sensors are distributed on the periphery of the unmanned mine car and output corresponding induction data; Acquiring a prediction range of the unmanned mine car relative to the space region, and determining a prediction positioning event of the unmanned mine car based on the fusion data combination, the current image of the unmanned mine car and the prediction range of the unmanned mine car relative to the space region; And determining a plurality of positioning information based on the detection of the predicted positioning event of the unmanned mine car, and determining the current positioning position of the unmanned mine car according to the plurality of positioning information, the corresponding positioning dimension and the overall shape of the unmanned mine car.
2. The method for positioning an unmanned mine car of a strip mine based on multi-source heterogeneous sensor fusion according to claim 1, wherein in the internal transportation path of the mine, a moving route of the unmanned mine car is collected, a plurality of mobile nodes are determined based on the identification of the moving route of the unmanned mine car, a plurality of dynamic images of the unmanned mine car are determined according to the tracing of the plurality of mobile nodes, and a current image of the unmanned mine car is determined based on the plurality of dynamic images and corresponding photographing time, comprising: Marking an internal conveying channel of the mine based on the detection of the mine, determining a conveying distribution diagram of the mine according to the detection of the internal conveying channel of the mine, and determining a moving route of the unmanned mine car based on the conveying distribution diagram of the mine, a database of the unmanned mine car and a corresponding transportation task so as to acquire the moving route of the unmanned mine car; Determining a plurality of sub-moving areas according to the detection of the moving route of the unmanned mine car, determining corresponding mobile nodes based on the area positions of the sub-moving areas, the corresponding area forms and the stay positions of the unmanned mine car, determining a plurality of dynamic images of the unmanned mine car based on the tracing of the mobile nodes, marking the shooting events of each dynamic image, and determining the current image of the unmanned mine car based on the plurality of dynamic images and the corresponding shooting time.
3. The method for positioning an unmanned mine car of a strip mine based on multi-source heterogeneous sensor fusion according to claim 1, wherein the determining a spatial region of the unmanned mine car relative to the mine based on the identification of the current image of the unmanned mine car, determining a plurality of multi-source heterogeneous sensors based on the detection of the spatial region, comprises: Collecting a current image of the unmanned mine car, determining a plurality of sub-environment areas according to the detection of the current image of the unmanned mine car, determining corresponding environment characteristics based on the detection of each sub-environment area, and determining a space area of the unmanned mine car relative to a mine according to the overall form of the unmanned mine car, the characteristic forms of the plurality of environment characteristics and the conveying distribution diagram of the mine; The method comprises the steps of determining a plurality of electronic devices based on detection of the space region, determining an induction combination according to the devices of the plurality of electronic devices, and determining a plurality of multi-source heterogeneous sensors based on the induction combination, the space position of the unmanned mine car relative to the space region and response data of the unmanned mine car, wherein the plurality of multi-source heterogeneous sensors are respectively arranged on the periphery of the unmanned mine car and synchronously induce the unmanned mine car.
4. The method for locating the unmanned mine car of the strip mine based on the multi-source heterogeneous sensor fusion of claim 1, wherein the multi-source heterogeneous sensors are distributed on the periphery of the unmanned mine car and output corresponding sensing data, and the method for determining the fusion data combination based on the sensing data, the corresponding data types and the overall morphology of the unmanned mine car comprises the following steps: the method comprises the steps of marking a plurality of multi-source heterogeneous sensors, determining a multi-dimensional detection space of the unmanned mine car according to the multi-source heterogeneous sensors and the unmanned mine car, detecting the unmanned mine car by the multi-source heterogeneous sensors along different directions in the multi-dimensional detection space of the unmanned mine car, outputting corresponding induction data, and matching the induction data in a time dimension.
5. The method for locating a mine car of a strip mine based on multi-source heterogeneous sensor fusion of claim 4, wherein the multi-source heterogeneous sensors are distributed on the periphery of the mine car and output corresponding sensing data, wherein the method for locating the mine car of the strip mine based on multi-source heterogeneous sensors determines a fusion data combination based on the sensing data, the corresponding data type and the overall shape of the mine car, and further comprises: determining a corresponding data type in the plurality of sensing data based on the identification of each sensing data, and determining a first data combination according to the plurality of sensing data and the overall form of the unmanned mining vehicle; And determining a second data combination according to the data types of the plurality of sensing data and the overall shape of the unmanned mining vehicle, determining a corresponding fusion data combination based on the fusion of the first data combination and the first data combination, marking the data priority of each sensing data in the fusion data combination, and carrying out anomaly screening on the plurality of sensing data.
6. The method for positioning unmanned mine vehicles in strip mines based on multi-source heterogeneous sensor fusion according to claim 1, wherein the step of acquiring the predicted range of the unmanned mine vehicle relative to the space region, determining the predicted positioning event of the unmanned mine vehicle based on the fusion data combination, the current image of the unmanned mine vehicle and the predicted range of the unmanned mine vehicle relative to the space region comprises the steps of: The method comprises the steps of collecting a space area, determining a prediction range of an unmanned mine car relative to the space area based on matching of the unmanned mine car and the space area, determining a plurality of sub-prediction areas according to identification of the prediction range, collecting a current image of the unmanned mine car, and determining a current position event of the unmanned mine car according to the current image of the unmanned mine car and the plurality of sub-prediction areas.
7. The method for locating a mine car in a strip mine based on multi-source heterogeneous sensor fusion of claim 6, wherein the acquiring a predicted range of the mine car relative to the spatial region determines a predicted locating event of the mine car based on the fusion data combination, the current image of the mine car and the predicted range of the mine car relative to the spatial region, further comprising: And acquiring a fusion data combination, determining a current form event of the unmanned mine car according to the fusion data combination and a current image of the unmanned mine car, and determining the current position event of the unmanned mine car based on the current position event of the unmanned mine car and the corresponding current form event.
8. The method for locating a mine car of a strip mine based on multi-source heterogeneous sensor fusion according to claim 1, wherein the determining a plurality of locating information based on the detection of the predicted locating event of the mine car, determining the current locating position of the mine car according to the plurality of locating information, the corresponding locating dimension and the overall shape of the mine car comprises: The method comprises the steps of collecting predicted positioning events of the unmanned mine car, determining a plurality of positioning items based on detection of the predicted positioning events of the unmanned mine car, determining a plurality of positioning information according to item contents of the plurality of positioning items, corresponding item priorities and a predicted range of the unmanned mine car relative to the space region, and marking corresponding positioning dimensions.
9. The method for locating a mine car of a strip mine based on multi-source heterogeneous sensor fusion of claim 8, wherein the determining the plurality of locating information based on the detection of the predicted locating event of the mine car, determining the current locating position of the mine car based on the plurality of locating information, the corresponding locating dimension and the overall shape of the mine car, further comprises: Acquiring the overall form of the unmanned mine car, and determining first positioning content according to the overall form of the unmanned mine car and a plurality of positioning information; And determining second positioning content according to the overall form of the unmanned mine car and the positioning dimensions of the positioning information, and determining the current positioning position of the unmanned mine car based on the first positioning content, the second positioning content and the conveying distribution diagram of the mine.
10. The strip mine unmanned mine car positioning system based on the multi-source heterogeneous sensor fusion is characterized in that the strip mine unmanned mine car positioning system based on the multi-source heterogeneous sensor fusion is applied to the strip mine unmanned mine car positioning method based on the multi-source heterogeneous sensor fusion according to any one of claims 1 to 9, and the strip mine unmanned mine car positioning system based on the multi-source heterogeneous sensor fusion comprises: The current image module is used for collecting the moving route of the unmanned mine car in the internal conveying channel of the mine, determining a plurality of mobile nodes based on the identification of the moving route of the unmanned mine car, determining a plurality of dynamic images of the unmanned mine car according to the tracing of the plurality of mobile nodes, and determining the current image of the unmanned mine car based on the plurality of dynamic images and corresponding shooting time; the detection module is used for determining a space area of the unmanned mine car relative to the mine according to the identification of the current image of the unmanned mine car, and determining a plurality of multi-source heterogeneous sensors based on the detection of the space area; The fusion data combination module is used for distributing a plurality of multi-source heterogeneous sensors on the periphery of the unmanned mine car and outputting corresponding induction data; The prediction positioning event module is used for acquiring the prediction range of the unmanned mine car relative to the space area and determining the prediction positioning event of the unmanned mine car based on the fusion data combination, the current image of the unmanned mine car and the prediction range of the unmanned mine car relative to the space area; And the current positioning position module is used for determining a plurality of positioning information based on detection of the predicted positioning event of the unmanned mine car, and determining the current positioning position of the unmanned mine car according to the plurality of positioning information, the corresponding positioning dimension and the overall form of the unmanned mine car.

Description

Strip mine unmanned mine car positioning method and system based on multi-source heterogeneous sensor fusion Technical Field The invention relates to the technical field of unmanned mine car positioning, in particular to a strip mine unmanned mine car positioning method and system based on multi-source heterogeneous sensor fusion. Background Along with development of science and technology, unmanned mine cars are gradually applied to mines and move in internal conveying channels of the mines, and at the moment, the unmanned mine cars can automatically complete intelligent vehicles for transportation operation in mine environments. In the prior art, the form of the internal conveying channel of the mine is collected, the form of the internal conveying channel of the mine is complex, the unmanned mine car is unmanned in the internal conveying channel of the mine, a plurality of images are collected, the positions of the unmanned mine car are determined based on the images, sensing data corresponding to a plurality of multi-source heterogeneous sensors are ignored, the combination accuracy of fusion data is affected, and the accuracy of the current positioning position of the unmanned mine car is reduced. Disclosure of Invention The invention aims to overcome the defects of the prior art, and provides a positioning method and a positioning system for an unmanned mine car of a strip mine based on multi-source heterogeneous sensor fusion. The embodiment of the invention provides a positioning method of an unmanned mine car of a strip mine based on multi-source heterogeneous sensor fusion, which comprises the following steps: In an internal conveying channel of a mine, acquiring a moving route of the unmanned mine car, determining a plurality of mobile nodes based on the identification of the moving route of the unmanned mine car, determining a plurality of dynamic images of the unmanned mine car according to the tracing of the plurality of mobile nodes, and determining a current image of the unmanned mine car based on the plurality of dynamic images and corresponding shooting time; Determining a spatial region of the unmanned mine car relative to the mine according to the identification of the current image of the unmanned mine car, and determining a plurality of multi-source heterogeneous sensors based on the detection of the spatial region; The system comprises a plurality of heterogeneous sensors, a fusion data combination, a fusion data acquisition unit and a fusion data acquisition unit, wherein the heterogeneous sensors are distributed on the periphery of the unmanned mine car and output corresponding induction data; Acquiring a prediction range of the unmanned mine car relative to the space region, and determining a prediction positioning event of the unmanned mine car based on the fusion data combination, the current image of the unmanned mine car and the prediction range of the unmanned mine car relative to the space region; And determining a plurality of positioning information based on the detection of the predicted positioning event of the unmanned mine car, and determining the current positioning position of the unmanned mine car according to the plurality of positioning information, the corresponding positioning dimension and the overall shape of the unmanned mine car. The embodiment of the invention provides a strip mine unmanned mine car positioning system based on multi-source heterogeneous sensor fusion, which is applied to the strip mine unmanned mine car positioning method based on multi-source heterogeneous sensor fusion, and comprises the following steps: The current image module is used for collecting the moving route of the unmanned mine car in the internal conveying channel of the mine, determining a plurality of mobile nodes based on the identification of the moving route of the unmanned mine car, determining a plurality of dynamic images of the unmanned mine car according to the tracing of the plurality of mobile nodes, and determining the current image of the unmanned mine car based on the plurality of dynamic images and corresponding shooting time; the detection module is used for determining a space area of the unmanned mine car relative to the mine according to the identification of the current image of the unmanned mine car, and determining a plurality of multi-source heterogeneous sensors based on the detection of the space area; The fusion data combination module is used for distributing a plurality of multi-source heterogeneous sensors on the periphery of the unmanned mine car and outputting corresponding induction data; The prediction positioning event module is used for acquiring the prediction range of the unmanned mine car relative to the space area and determining the prediction positioning event of the unmanned mine car based on the fusion data combination, the current image of the unmanned mine car and the prediction range of the unmanned mine car relative to the space a