CN115223144A - Unmanned mine car sensor data screening method and device based on cloud data - Google Patents
Unmanned mine car sensor data screening method and device based on cloud data Download PDFInfo
- Publication number
- CN115223144A CN115223144A CN202210837403.6A CN202210837403A CN115223144A CN 115223144 A CN115223144 A CN 115223144A CN 202210837403 A CN202210837403 A CN 202210837403A CN 115223144 A CN115223144 A CN 115223144A
- Authority
- CN
- China
- Prior art keywords
- data
- vehicle
- mine car
- road
- cloud
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P3/00—Measuring linear or angular speed; Measuring differences of linear or angular speeds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Abstract
According to the method, data acquired by a vehicle-mounted camera sensor of the unmanned mine car are uploaded to a cloud server, characteristic identification is carried out on the cloud data through the cloud server, then road curvature and gradient are selected according to the characteristics of a mining area, whether obstacles exist in the front road and whether a turnout exists in the front road are used as characteristic parameters to achieve vehicle-mounted camera data screening, data input to a decision model of the unmanned mine car are reduced, meanwhile, in the vehicle-mounted camera data screening process, speed information of the unmanned mine car is referred, screening weight is dynamically determined, and therefore decision time of the decision model is shortened, most of garbage data causing interference are screened according to the characteristics of the mining area and the screening weight is dynamically determined, and decision precision of a decision layer is improved while the decision time is shortened.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a method and a device for screening unmanned mine car sensor data based on cloud data.
Background
The unmanned technology is characterized in that the surrounding environment is sensed by utilizing multiple technologies such as radar, laser, ultrasonic waves, GPS, odometer and computer vision, obstacles and various identification boards are identified through an advanced computer control system, a proper path is planned to control the running of a mine car, the unmanned car becomes the future development trend of the automobile industry along with the rapid development and wide application of the automobile intelligent technology, and the unmanned car is also a very popular research field at present due to the characteristics of innovation, practicability, complexity, multidisciplinary intersection and the like, and a plurality of international and domestic companies develop researches on the unmanned technology.
With the gradual improvement of communication means, in order to improve the real-time performance of unmanned path and driving parameter prediction, data acquired by a sensor in the unmanned process is uploaded to a cloud server, and then the prediction of the path and driving parameters is realized as a big hotspot of current research, generally, 4-6 vehicle-mounted cameras are integrated on an unmanned vehicle for decision-making assistance, so that a large amount of data can be generated by the vehicle-mounted cameras, but not all the data have an effect on decision-making, and due to the unmanned property of the unmanned vehicle, the subsequent path and driving parameters need to be predicted in a near real-time and accurate manner, if the data generated by the sensor is not screened, the data are input into a prediction model to participate in operation, so that the time delay is greatly increased, and the prediction precision is reduced.
Meanwhile, because of less sudden interference and simple road state in a mine, the method gradually becomes the hot research field of unmanned mine cars, and the prior art does not have a scheme for screening vehicle-mounted camera data of the unmanned mine cars aiming at the characteristics of the mine, thereby reducing time delay and improving the prediction accuracy of the driving parameters of the unmanned mine cars.
Disclosure of Invention
The invention aims to solve the technical problem of providing a method and a device for screening data of a sensor of an unmanned mine car based on cloud data, aiming at the application scene of a mine field, the data collected by the sensor of the unmanned mine car is screened, so that the data input into a prediction model is reduced, and the purposes of reducing time delay and improving accuracy are achieved.
The concept to which the present application relates will be first explained below with reference to the drawings. It should be noted that the following descriptions of the concepts are only for the purpose of facilitating understanding of the contents of the present application, and do not represent limitations on the scope of the present application.
In order to achieve the above object, according to one aspect of the present invention, a method for screening sensor data of unmanned mine car based on cloud data comprises:
step 1: acquiring video data acquired by a plurality of vehicle-mounted camera sensors during the running of the unmanned mine car;
step 2: converting the video data acquired by the plurality of vehicle-mounted cameras into an image format by taking frames as units, and uploading the image format to a cloud server after a timestamp is printed;
and step 3: the characteristic information of each frame of image is obtained through an image recognition technology and is used as cloud data, and the cloud data comprises: the method comprises the following steps of (1) running front road surface gradient information, front road curvature, whether a front road has an obstacle or not, and whether the front is turnout information or not;
and 4, step 4: and (4) screening each frame of image according to the cloud data obtained in the step (3) in combination with the driving speed.
Further, the camera sensor is used for collecting environmental information around the mine car.
Further, the environmental information of the periphery of the mine car comprises the running front condition of the mine car and road condition information.
Further, in step 2, data can be uploaded through a 5G communication technology, or data transmission can be realized in a communication mode that an edge network communication node is established in a mining area, an unmanned vehicle-mounted computer is adopted to transmit data streams to the edge node, and the edge network node transmits data to a cloud computer.
Further, the step 4 specifically includes:
step 4.1: generating classification rules through machine learning;
step 4.2: classifying the cloud data in the step 3 by using the classification rule established in the step 4.1, wherein the first class is garbage data; the second type is possible uploading data; the third type is to determine the uploading data;
step 4.3: and determining a data screening and uploading strategy according to the classification rule and the driving speed.
Further, the step 4.1 specifically includes:
step 4.1.1: driving a vehicle to run on the task route by using a manned mine car, recording data information of a running process through a vehicle-mounted camera sensor, and recording running parameters of the vehicle running through a vehicle-mounted computer, wherein the running parameters comprise a vehicle speed, and the following steps are added in step 4.1.2: uploading the information to the cloud server, and performing feature recognition processing on the information; extracting the road curvature and the gradient of each frame of picture, whether an obstacle exists in the front road or not and whether a branch exists in the front road or not as characteristic parameter information for data screening to generate a characteristic information data set, assigning initial weights according to the four characteristic parameters of the road curvature and the gradient, whether an obstacle exists in the front road or not and whether a branch exists in the front road or not, and establishing a classification model as shown in a formula (1):
Ti＝a*Ti_c+b*Ti_s+c*Ti_o+d*Ti_a (1)
the method comprises the following steps that the image characteristic value is Ti, a, b, c and d are weights of each characteristic parameter, ti _ c is the curvature of a road identified by the ith frame of image, ti _ s is the gradient of the road identified by the ith frame of image, ti _ o is the size information of a front road obstacle identified by the ith frame of image, and Ti _ a is whether a road turnout exists or not;
training a classification model by using a characteristic information data set of each frame of picture and a driving parameter at the moment as training data to obtain model parameters a, b, c and d;
step 4.1.3: the method comprises the steps of carrying out feature recognition on collected real-time driving data, determining the curvature of a road, the gradient of the road, the obstacle size information of the road and whether the road has a branch road, inputting the data into a classification model, and then calculating an image feature value Ti, wherein the specific classification rule is as follows according to the image feature value:
if Ti < T1, determining that the data is the third type data; if T1 is not less than Ti and not more than T2, determining the data as second type data; if T2< Ti, it is determined to be the first type of data, where T1< T2.
Further, the classification model is a convolution network model.
Further, the step 4.3 further comprises:
if the driving speed v is less than 35km/h, only uploading the third type of data;
if the running speed 120 >:
k＝(lnV-3.55)/1.24 (2)
wherein V is the running speed of the unmanned mine car, and the unit is km/h; selecting data with the proportion of k in the second class of data and the third class of data as uploading data according to the calculation result;
and if the running speed is more than or equal to 120km/h, selecting all the second type data and the third type data as uploading data.
According to another aspect of the invention, a method for screening unmanned mine car sensor data based on cloud data is provided, which comprises the following steps:
vehicle-mounted camera: the vehicle-mounted cameras are used for acquiring video data collected by the vehicle-mounted cameras when the unmanned mine car runs;
the data conversion and uploading module: the system comprises a plurality of vehicle-mounted cameras, a cloud server and a plurality of video cameras, wherein the vehicle-mounted cameras are used for acquiring video data;
the data screening module: for operating the data classification screening method.
Based on the technical scheme, the unmanned mine car sensor data screening method based on the cloud data has the following technical effects:
according to the method, data acquired by a vehicle-mounted camera sensor of the unmanned mine car are uploaded to a cloud server, characteristic identification is carried out on the cloud data through the cloud server, then road curvature and gradient are selected according to the characteristics of a mining area, whether obstacles exist in the front road and whether a turnout exists in the front road are used as characteristic parameters to achieve vehicle-mounted camera data screening, data input to a decision model of the unmanned mine car are reduced, meanwhile, in the vehicle-mounted camera data screening process, speed information of the unmanned mine car is referred, screening weight is dynamically determined, decision time of a decision layer is shortened, the sensor data is screened according to the characteristics of the mining area and the screening weight is dynamically determined, most of garbage data causing interference are overlooked, and decision precision of the decision layer is improved while the decision time is reduced.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow chart of a method for cloud-based unmanned mine car sensor data screening provided in an embodiment of the present application;
fig. 2 is a comparison graph of the screened data and the original data provided by the embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The concept to which the present application relates will be first explained below with reference to the drawings. It should be noted that the following descriptions of the concepts are only for the purpose of facilitating understanding of the contents of the present application, and do not represent limitations on the scope of the present application.
As shown in fig. 1, a method for screening unmanned mine car sensor data based on cloud data comprises:
step 1: acquiring video data acquired by a plurality of vehicle-mounted camera sensors during the running of the unmanned mine car;
the camera sensor is mainly used for collecting environmental information around the mine car, including but not limited to information such as the running front condition of the mine car and the road condition;
step 2: converting the video data acquired by the plurality of vehicle-mounted cameras into an image format by taking frames as units, and uploading the image format to a cloud server after a timestamp is printed;
specifically, data uploading can be realized through a 5G communication technology, or data transmission can be realized in a communication mode that an edge network communication node is established in a mining area, an unmanned vehicle-mounted computer is adopted to transmit data streams to the edge node, and the edge network node transmits data to a cloud computer;
and 3, step 3: the characteristic information of each frame of image is obtained through an image recognition technology and is used as cloud data, and the cloud data comprises: information such as gradient information of a running front road surface, curvature of a front road, whether an obstacle exists on the front road, whether the front road is a fork and the like;
the cloud server has better performance relative to the vehicle-mounted computer, so that a complex model can be operated, and the characteristic information of each frame of image acquired by adopting an image recognition technology in the cloud server is used as cloud data to participate in the operation of the decision layer
And 4, step 4: screening each frame of image according to the cloud data obtained in the step 3 in combination with the driving speed;
specifically, the step 4 specifically includes:
step 4.1: generating classification rules through machine learning;
due to the existence of the plurality of vehicle-mounted cameras of the unmanned mine car, the unmanned mine car generates a large amount of data information in the running process, and the large data with huge body facing volume is obtained efficiently through machine learning, so that the unmanned mine car has gradually become a main driving force for the development of the current machine learning technology, and specifically, the step 4.1 comprises the following steps: step 4.1.1: driving a vehicle to run on the task route by using a manned mine car, recording data information of a running process through a vehicle-mounted camera sensor, and recording running parameters of the vehicle running through a vehicle-mounted computer, wherein the running parameters comprise vehicle speed, acceleration and steering or not; recording data information and driving parameters of the driving process through a vehicle-mounted camera sensor;
step 4.1.2: step 4.1.2: uploading the information to the cloud server, and performing feature recognition processing on the information; extracting the road curvature and the gradient of each frame of picture, whether an obstacle exists in the front road or not and whether a branch exists in the front road or not as characteristic parameter information for data screening to generate a characteristic information data set, assigning initial weights according to the four characteristic parameters of the road curvature and the gradient, whether an obstacle exists in the front road or not and whether a branch exists in the front road or not, and establishing a classification model as shown in a formula (1):
Ti＝a*Ti_c+b*Ti_s+c*Ti_o+d*Ti_a (1)
the method comprises the following steps that the image characteristic value is Ti, a, b, c and d are weights of each characteristic parameter, ti _ c is the curvature of a road identified by the ith frame of image, ti _ s is the gradient of the road identified by the ith frame of image, ti _ o is the size information of a front road obstacle identified by the ith frame of image, and Ti _ a is whether a road turnout exists or not; specifically, a convolutional network model is adopted as a classification model; training the convolution network model through training data so as to establish a classification rule; during training, an Adam optimizer is adopted, the initial learning rate is 0.01, the initial learning rate is influenced by software and hardware environments, the size of a characteristic information data set input into the learning network model is 4x4xC, wherein C is the number of parameters in each characteristic parameter data set input into the network model, and it is worth mentioning that the road curvature, the gradient, whether obstacles exist in a front road or not and whether a fork exists in the front road or not are adopted as characteristic parameters in the method, so that the value of C is 4; illustratively, taking a certain operation route as an example, specifically, initial weights are assigned to four characteristic parameters, such as curvature and gradient of a road, whether an obstacle exists on a front road and whether a branch exists on the front road, and then a characteristic value of each frame of image is calculated according to a formula (1), a classification model is trained through training data, and the weights of the characteristic parameters are mainly trained, so that optimal parameters are obtained.
Step 4.1.3: the method comprises the steps of carrying out feature recognition on collected real-time driving data, determining the curvature of a road, the gradient of the road, the size information of obstacles on the road and whether the road has a turnout, inputting the turnout into a classification model, calculating an image feature value Ti, and according to the image feature value, carrying out the following specific classification rules:
if Ti < T1, determining that the data is the third type data; if T1 is not less than Ti and not more than T2, determining the data as second type data; if T2< Ti, it is determined to be the first type of data, where T1< T2.
Step 4.2: calculating the characteristic value of each frame of picture by adopting the characteristic value calculation method in the step 4.1, and then classifying the cloud data in the step 3 by utilizing the classification rule established in the step 4.1, wherein the first type is junk data; the second type is possible uploading data; and the third type is to determine uploading data, wherein the garbage data is not input into a decision model to participate in decision due to small decision influence on the unmanned mine car, while the uploading proportion is judged by the following steps for possible uploading data, and the garbage data is determined to participate in decision due to the fact that the garbage data is essential data for participating in decision.
Step 4.3: determining a data screening and uploading strategy according to the classification rule and the driving speed; the speed of the unmanned vehicle plays a crucial role in making a correct decision of the vehicle, when the speed is high, more information needs to be uploaded to improve decision accuracy so as to reduce accident risk, and when the speed is low, less information is uploaded to reduce decision time; therefore, the influence of the running speed of the unmanned mine car on the mine car decision making is mainly considered, so that the decision making is carried out on the sensor data screening, specifically, the running speed of the unmanned mine car is obtained in real time, and if the running speed is less than 35km/h, only the third type of data is uploaded; if the running speed 120 >:
k＝(lnV-3.55)/1.24 (2)
v is the running speed of the unmanned mine car, the unit is km/h, and data with the proportion of k in the second type of data and the third type of data are selected as uploading data according to the calculation result; and if the running speed is more than or equal to 120km/h, selecting all the second type data and the third type data as uploading data.
For example, when the speed is 45km/h, and k is 0.21, randomly selecting 21% of data in the second type of information to upload;
so far, accomplish the screening of on-vehicle camera sensor data, upload useful information, make the data of input to decision-making model reduce, thereby the decision-making time of decision-making layer operation has been reduced, and, this application is according to the characteristics in unmanned mining area, select road curvature, slope, whether have the barrier in the place ahead road and realize on-vehicle camera data screening as characteristic parameter, make the useful information that contains in the data that upload increase, what have considered most rubbish data that cause the interference more, decision-making precision on decision-making layer has been improved.
According to another aspect of the invention, a method for screening unmanned mine car sensor data based on cloud data is provided, which comprises the following steps:
vehicle-mounted camera: the vehicle-mounted cameras are used for acquiring video data collected by the vehicle-mounted cameras when the unmanned mine car runs;
the data conversion and uploading module: the system comprises a plurality of vehicle-mounted cameras, a cloud server and a plurality of video cameras, wherein the vehicle-mounted cameras are used for acquiring video data;
the data screening module: for running the above-described data classification screening method.
According to the method, a certain operation route in a certain mining area is used as a specific task, the scheme is used for testing, a driver with many years of driving experience is selected to drive the route, and during driving, the driver adopts accurate driving parameters to realize driving according to road condition information; the method comprises the steps of obtaining 10580 frames of pictures and driving data corresponding to each frame of picture, uploading the information to a cloud server, and classifying 36580 frames of pictures by using the cloud server to operate the sensor data screening method, wherein the first type of data are 17683 frames, the second type of data are 8519 frames, and the third type of data are 10378 frames, so that a considerable proportion of garbage data can be screened by the screening method, then unmanned driving parameter prediction time delay of the operation route is counted according to the driving speed of the unmanned mine car, as shown in fig. 2, wherein the vertical axis is the decision time when the screening method is adopted to screen the data and then input into a decision model, the solid line is the decision time when data screening is not carried out, the dotted line is the decision time when the screening method is adopted as input data, the decision time is reduced by 21% on average, and meanwhile, the decision accuracy is obviously improved by checking.
The above-described embodiments and/or implementations are only for illustrating the preferred embodiments and/or implementations of the present technology, and are not intended to limit the implementations of the present technology in any way, and those skilled in the art can make many modifications or changes without departing from the scope of the technology disclosed in the present disclosure, but should be construed as technology or implementations that are substantially the same as the present technology.
Claims (9)
1. A method for screening unmanned mine car sensor data based on cloud data is characterized by comprising the following steps:
step 1: acquiring video data acquired by a plurality of vehicle-mounted camera sensors during the running of the unmanned mine car;
and 2, step: converting the video data collected by the plurality of vehicle-mounted cameras into an image format by taking a frame as a unit, and uploading the image format to a cloud server after a timestamp is printed;
and 3, step 3: the characteristic information of each frame of image is obtained through an image recognition technology and is used as cloud data, and the cloud data comprises: the method comprises the following steps of (1) running front road surface gradient information, front road curvature, whether a front road has an obstacle or not, and whether the front is turnout information or not;
and 4, step 4: and (4) screening each frame of image according to the cloud data obtained in the step (3) in combination with the driving speed.
2. The unmanned mine car sensor data screening method based on cloud data of claim 1, wherein the camera sensor is used to collect environmental information around the mine car.
3. The unmanned mine car sensor data screening method based on cloud data as claimed in claim 2, wherein the environmental information of the mine car periphery comprises mine car driving front condition and road condition information.
4. The unmanned mine vehicle sensor data screening method based on cloud data as claimed in claim 1, wherein in step 2, data uploading can be achieved through a 5G communication technology, or data transmission can be achieved through a communication mode that an unmanned vehicle-mounted computer transmits data streams to edge nodes and the edge nodes transmit data to a cloud computer by establishing edge network communication nodes in a mine area.
5. The cloud-data-based unmanned mine car sensor data screening method of claim 1, wherein step 4 specifically comprises:
step 4.1: generating classification rules through machine learning;
and 4.2: classifying the cloud data in the step 3 by using the classification rule established in the step 4.1, wherein the first type of data is garbage data; the second type of data is possible uploading data; the third kind of data is determined uploading data;
step 4.3: and determining a data screening and uploading strategy according to the classification rule and the driving speed.
6. The unmanned mine car sensor data screening method based on cloud data of claim 5, wherein the step 4.1 specifically comprises:
step 4.1.1: driving a vehicle to run on the task route by using a manned mine car, recording data information of a running process through a vehicle-mounted camera sensor, and recording running parameters of the vehicle running through a vehicle-mounted computer, wherein the running parameters comprise vehicle speed, acceleration and steering or not; recording data information and driving parameters of the driving process through a vehicle-mounted camera sensor;
step 4.1.2: uploading the information to the cloud server, and performing feature recognition processing on the information; extracting the road curvature and the gradient of each frame of picture, whether an obstacle exists in the front road or not and whether a branch exists in the front road or not as characteristic parameter information for data screening to generate a characteristic information data set, assigning initial weights according to the four characteristic parameters of the road curvature and the gradient, whether an obstacle exists in the front road or not and whether a branch exists in the front road or not, and establishing a classification model as shown in a formula (1):
Ti＝a*Ti_c+b*Ti_s+c*Ti_o+d*Ti_a (1)
the method comprises the following steps of obtaining an image characteristic value, wherein Ti is the image characteristic value, a, b, c and d are the weight of each characteristic parameter, ti _ c is the curvature of a road identified by the ith frame of image, ti _ s is the gradient of the road identified by the ith frame of image, ti _ o is the size information of a front road obstacle identified by the ith frame of image, and Ti _ a is whether a road turnout exists:
training a classification model by using a characteristic information data set of each frame of picture and a driving parameter at the moment as training data to obtain model parameters a, b, c and d;
step 4.1.3: the method comprises the steps of carrying out feature recognition on collected real-time driving data, determining the curvature of a road, the gradient of the road, the size information of obstacles on the road and whether the road has a turnout, inputting the turnout into a classification model, calculating an image feature value Ti, and according to the image feature value, carrying out the following specific classification rules:
if Ti < T1, determining that the data is the third type data; if T1 is not less than Ti and not more than T2, determining the data as second type data; if T2< Ti, it is determined to be the first type of data, where T1< T2.
7. The method for unmanned mine car sensor data screening based on cloud data as claimed in claim 6, wherein the classification model is a convolutional network model.
8. A cloud-based unmanned mine car sensor data screening method according to claim 5, wherein the step 4.3 further comprises:
if the driving speed v is less than 35km/h, only uploading the third type of data;
if the running speed 120 >:
k＝(lnV-3.55)/1.24 (2)
wherein V is the running speed of the unmanned mine car, and the unit is km/h; according to the calculation result, selecting data with the proportion of k in the second class of data and the third class of data as uploading data;
and if the running speed is more than or equal to 120km/h, selecting all the second type data and the third type data as uploading data.
9. The utility model provides an unmanned mine car sensor data screening system based on high in clouds data, includes:
vehicle-mounted camera: the vehicle-mounted cameras are used for acquiring video data collected by the vehicle-mounted cameras when the unmanned tramcar runs;
the data conversion and uploading module: the system comprises a plurality of vehicle-mounted cameras, a cloud server and a plurality of video cameras, wherein the vehicle-mounted cameras are used for acquiring video data;
the data screening module: for operating the sensor data classification screening method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210837403.6A CN115223144A (en) | 2022-07-15 | 2022-07-15 | Unmanned mine car sensor data screening method and device based on cloud data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210837403.6A CN115223144A (en) | 2022-07-15 | 2022-07-15 | Unmanned mine car sensor data screening method and device based on cloud data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115223144A true CN115223144A (en) | 2022-10-21 |
Family
ID=83612872
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210837403.6A Pending CN115223144A (en) | 2022-07-15 | 2022-07-15 | Unmanned mine car sensor data screening method and device based on cloud data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115223144A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117558147A (en) * | 2024-01-11 | 2024-02-13 | 上海伯镭智能科技有限公司 | Mining area unmanned vehicle road right distribution remote control method |
-
2022
- 2022-07-15 CN CN202210837403.6A patent/CN115223144A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117558147A (en) * | 2024-01-11 | 2024-02-13 | 上海伯镭智能科技有限公司 | Mining area unmanned vehicle road right distribution remote control method |
| CN117558147B (en) * | 2024-01-11 | 2024-03-26 | 上海伯镭智能科技有限公司 | Mining area unmanned vehicle road right distribution remote control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP4152204A1 (en) | Lane line detection method, and related apparatus | |
| CN110850854A (en) | Autonomous driver agent and policy server for providing policies to autonomous driver agents | |
| CN110850861A (en) | Attention-based hierarchical lane change depth reinforcement learning | |
| US10963706B2 (en) | Distributable representation learning for associating observations from multiple vehicles | |
| WO2020098456A1 (en) | Method and system for generating predicted occupancy grid maps | |
| EP3822852B1 (en) | Method, apparatus, computer storage medium and program for training a trajectory planning model | |
| CN112249032B (en) | Automatic driving decision method, system, equipment and computer storage medium | |
| CN110356412B (en) | Method and apparatus for automatic rule learning for autonomous driving | |
| CN109767619B (en) | Intelligent networking pure electric vehicle running condition prediction method | |
| CN112614373B (en) | BiLSTM-based weekly vehicle lane change intention prediction method | |
| CN114915646B (en) | Data grading uploading method and device for unmanned mine car | |
| US20200174474A1 (en) | Method and system for context and content aware sensor in a vehicle | |
| CN113291320A (en) | Vehicle track prediction method, device, equipment and storage medium | |
| CN111971725A (en) | Method for determining lane change instructions for a vehicle, computer-readable storage medium, and vehicle | |
| CN115223144A (en) | Unmanned mine car sensor data screening method and device based on cloud data | |
| CN117693459A (en) | Track consistency measurement for autonomous vehicle operation | |
| CN115359332A (en) | Data fusion method and device based on vehicle-road cooperation, electronic equipment and system | |
| EP4257443A1 (en) | Method and system for automatic driving data collection and closed-loop management | |
| US11544899B2 (en) | System and method for generating terrain maps | |
| CN114863685B (en) | Traffic participant trajectory prediction method and system based on risk acceptance degree | |
| WO2023206236A1 (en) | Method for detecting target and related device | |
| EP4148600A1 (en) | Attentional sampling for long range detection in autonomous vehicles | |
| EP4145358A1 (en) | Systems and methods for onboard enforcement of allowable behavior based on probabilistic model of automated functional components | |
| US20230195977A1 (en) | Method and system for classifying scenarios of a virtual test, and training method | |
| US20230048926A1 (en) | Methods and Systems for Predicting Properties of a Plurality of Objects in a Vicinity of a Vehicle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |