CN116700249A

CN116700249A - Shunting automatic driving method, system, equipment, storage medium and product thereof

Info

Publication number: CN116700249A
Application number: CN202310630555.3A
Authority: CN
Inventors: 杨华昌; 马滨博; 栾德杰; 冯军; 徐磊; 唐汇东; 任宛星; 王浩然; 余立伟; 徐鹏; 张弘志; 曹子昱; 陈尚彬; 韩波; 赵阳; 尹德胜; 李继东; 刘靖远; 曹天睿; 李堃
Original assignee: China Academy of Railway Sciences Corp Ltd CARS; Guoneng Shuohuang Railway Development Co Ltd
Current assignee: China Academy of Railway Sciences Corp Ltd CARS; Guoneng Shuohuang Railway Development Co Ltd
Priority date: 2023-05-31
Filing date: 2023-05-31
Publication date: 2023-09-05

Abstract

The invention discloses a shunting automatic driving method, a shunting automatic driving system, shunting automatic driving equipment, a shunting automatic driving device, a shunting automatic driving storage medium and a shunting automatic driving product, and relates to the technical field of shunting operation of locomotives in a station yard; the method comprises the steps of obtaining the position of a locomotive in a station, and obtaining a shunting driving license instruction according to a shunting operation plan; according to the shunting driving permission instruction, acquiring a front obstacle of the locomotive and distance information analysis thereof to generate an automatic driving curve; controlling the locomotive to run based on the automatic driving curve according to the automatic driving curve; acquiring the alignment position of the locomotive, and controlling the locomotive to stop according to the alignment mark of the position of the dumper through the shunting operation instruction; acquiring video and radar point cloud information in front of locomotive traction operation, and analyzing the video information to acquire environmental information, personnel and obstacle intrusion, derailment state and vehicle state information in front of locomotive traction operation; the method can effectively reduce the labor intensity of operators and the occurrence probability of personal safety accidents, obviously improve the control efficiency, the control precision and the safety operation level of shunting operation, and compress the transit residence time of an operation locomotive and a train in a station.

Description

Shunting automatic driving method, system, equipment, storage medium and product thereof

Technical Field

The invention discloses a shunting automatic driving method, a shunting automatic driving system, shunting automatic driving equipment, a shunting automatic driving storage medium and a shunting automatic driving storage medium product, and relates to the technical field of shunting operation of locomotives in yards.

Background

At present, the urban rail transit field and the high-speed railway field both realize the large-scale application of the automatic train steering technology (namely an ATO system), and part of freight lines also carry out automatic steering tests when freight trains run on large lines. However, the automatic driving in the urban rail transit field or the high-speed railway and freight train field is mainly based on the automatic driving in the traction state based on the railway signal. Compared with a front line operation locomotive, the in-station shunting operation scene is much more complex, and comprises the steps of grouping, disassembling, unhooking, line transferring, vehicle taking, transferring, locomotive entering and exiting sections, intermediate station line transferring and the like of a train. In addition, in many scenes, the train needs to be pushed to run, and a driver and passengers cannot intuitively grasp the state of the line in front of the running route, so that the shunting machine completely depends on manual driving to carry out shunting operation.

By researching the current situation of the yellow Yes harbor station shunting operation of the yellow railway, the current station shunting operation owners are finished by means of manual operation, and the automation level is relatively low. The shunting machine is driven manually, drivers can observe the forward route and signal conditions by means of visual observation and manual joint control, frequent talking connection with shunting groups and stations is needed, operation execution depends on personal experience, the operation efficiency is low, the labor intensity is high, and safety accidents are easy to occur.

Disclosure of Invention

Aiming at the problems in the prior art, the invention provides a shunting automatic driving method, a shunting automatic driving system, shunting automatic driving equipment, a shunting automatic driving storage medium and a shunting automatic driving product, and the technical scheme adopted is as follows:

in a first aspect, a shunting autopilot method, the method comprising:

acquiring the position of a locomotive in a station, and acquiring a shunting driving license instruction according to a shunting operation plan;

according to the shunting driving permission instruction, acquiring a front obstacle of the locomotive and distance information analysis thereof to generate an automatic driving curve;

controlling the locomotive to run based on the automatic driving curve according to the automatic driving curve;

acquiring the alignment position of the locomotive, and controlling the locomotive to stop according to the alignment mark of the position of the dumper through the shunting operation instruction;

and acquiring video and radar point cloud information in front of locomotive traction operation, and analyzing the video information to acquire environmental information, personnel and obstacle intrusion, derailment state and vehicle state information in front of locomotive traction operation.

In some implementations, the obtaining the locomotive position in the station, according to a shunting operation plan, obtains a shunting driving license instruction, includes:

carrying out data fusion positioning through Beidou satellite positioning, wheel track ranging and station track states to obtain the real-time position in the station of the locomotive;

And acquiring signal equipment state information of the computer interlocking system according to the shunting operation plan, and analyzing to generate a shunting driving permission instruction.

In some implementations, according to the autopilot curve, controlling the locomotive to travel based on the autopilot curve includes:

acquiring an automatic driving state of a locomotive, and analyzing the driving state through a PID locomotive control algorithm to generate a speed regulation instruction;

and controlling the locomotive to execute the speed regulation command based on the automatic driving curve.

In some implementations, the obtaining the locomotive alignment position, that is, obtaining video and radar point cloud information of the locomotive propulsion alignment position, analyzing the video information to obtain environmental information, personnel and obstacle intrusion limit and vehicle state information of the front end of the propulsion train includes:

collecting image information of the traction front and propulsion alignment position of a locomotive, preprocessing the image information, executing a predefined detection task based on a deep learning network, and generating detection information;

executing a lane line detection task according to the detection information, and determining the current lane line coordinates;

and matching the automatic driving curve according to the current lane line coordinates, and controlling the locomotive to stop according to the turning locomotive position.

In a second aspect, an embodiment of the present invention provides a shunting autopilot control system, the system including:

the ground main control module is used for acquiring the position of the locomotive in the station and acquiring a shunting driving permission instruction according to a shunting operation plan;

the vehicle-mounted shunting safety protection module is used for acquiring a front obstacle of the locomotive and distance information thereof according to the shunting driving permission instruction, and analyzing and generating an automatic driving curve;

the vehicle-mounted automatic driving module is used for controlling the locomotive to run based on the automatic driving curve according to the automatic driving curve;

the track side intelligent sensing module is used for acquiring the alignment position of the locomotive and controlling the locomotive to stop according to the alignment mark of the position of the dumper through the shunting operation instruction;

the vehicle-mounted intelligent sensing module is used for acquiring video and radar point cloud information in front of locomotive traction operation, analyzing the video information and acquiring environmental information, personnel and obstacle intrusion, derailment state and vehicle state information in front of locomotive traction operation.

In some implementations, the ground master control module includes:

the ground shunting safety protection unit is used for carrying out data fusion positioning through Beidou satellite positioning, wheel track ranging and station track states to obtain the real-time position in the station of the locomotive;

And the ground shunting automatic driving control unit is used for acquiring signal equipment state information of the computer interlocking system and analyzing and generating shunting driving permission instructions according to the shunting operation plan.

In some implementations, the in-vehicle autopilot module includes:

the driving safety protection unit is used for acquiring the automatic driving state of the locomotive, analyzing the driving state through a PID locomotive control algorithm and generating a speed regulation instruction;

and the automatic driving control unit is used for controlling the locomotive to execute the speed regulation instruction based on the automatic driving curve.

In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the method according to the first aspect when executing the computer program.

In a fourth aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, where the memory is configured to store one or more computer instructions, and where the one or more computer instructions, when executed by the processor, implement a method as described in the first aspect above.

In a fifth aspect, embodiments of the present invention provide a computer storage medium having a computer program stored therein, which when executed by a processor, is adapted to carry out the method according to the first aspect.

One or more embodiments of the present invention can provide at least the following advantages:

the beneficial effects of the invention are as follows: the method of the invention is used for shunting and automatically driving, and the operation before turning over the station changes the existing front-port traction, front-port pushing and front-port pushing into the operation before turning over, thereby reducing the operation links of one-time parking, single machine turning back and the like, and greatly reducing the operation scene operators such as unmanned vehicle taking, vehicle state inspection after turning over and the like;

the heavy-load train can stably run to the accurate position of the check point before and after turning the car dumper according to the planned speed curve to stop, so that the actions of rapid acceleration and rapid deceleration in the running process are reduced, the complicated train-ground joint control link is avoided, and the control efficiency of the shunting locomotive is effectively improved;

the driver and passengers are greatly reduced, the operation of manual operation of the locomotive by the double drivers in the past is changed into automatic driving under the supervision of a single person, even the operation is completely unmanned, the traction and pushing operation is unmanned, and the occurrence probability of personal injury accidents is greatly reduced. The intelligent sensing system is developed to sense the line obstacle in front of the train operation in advance, so that accidents such as shunting off-line and collision are effectively avoided, and the safety risk is effectively reduced.

The method can effectively reduce the labor intensity of operators and the occurrence probability of personal safety accidents, remarkably improve the control efficiency, the control precision and the safety operation level of shunting operation, compress the transit stay time of the operation locomotive and train in the station, and further improve the automation, informatization and intelligence levels of the station.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it will be obvious that the drawings in the following description are some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort to a person skilled in the art.

FIG. 1 is a flow chart of a method for automatic driving of a shunting device according to an embodiment of the present invention;

FIG. 2 is a flow chart of a vehicle-mounted autopilot module control locomotive operation provided by an embodiment of the present invention;

FIG. 3 is a target ranging flow chart of a shunting autopilot system provided by an embodiment of the present invention;

fig. 4 is a vehicle-mounted intelligent sensing flow chart of a shunting autopilot system provided by an embodiment of the invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, are intended to be within the scope of the present invention.

Embodiment one:

fig. 1 shows a flowchart of a shunting autopilot method, and as shown in fig. 1, the shunting autopilot method provided in this embodiment includes:

s10, acquiring the position of a locomotive in a station, and acquiring a shunting driving license instruction according to a shunting operation plan;

s20, acquiring a front obstacle and distance information of the locomotive according to the shunting driving permission instruction, and analyzing and generating an automatic driving curve;

S30, controlling the locomotive to run based on the automatic driving curve according to the automatic driving curve;

s40, acquiring the alignment position of the locomotive, and controlling the locomotive to stop according to the alignment mark of the position of the dumper through the shunting operation instruction;

s50, acquiring video and radar point cloud information in front of locomotive traction operation, and analyzing the video information to acquire environmental information, personnel and obstacle intrusion, derailment state and vehicle state information in front of locomotive traction operation.

In some implementations, the S10 includes:

s11, carrying out data fusion positioning through Beidou satellite positioning, wheel track ranging and station track states to obtain the real-time position in the station of the locomotive;

s12, acquiring signal equipment state information of the computer interlocking system, analyzing and generating a shunting driving permission instruction according to a shunting operation plan.

According to the method, a driving permission instruction is generated according to S11 and S12, and real-time positions of a shunting locomotive and an operation train in a station are automatically tracked according to a multi-data fusion positioning mode of Beidou satellite differential positioning, wheel track ranging and station track state checking. The real-time positioning method specifically comprises the steps of real-time positioning of the locomotive itself and the positions of two ends of the locomotive traction or propulsion train, so that real-time position tracking of the train during traction and propulsion running in a station is realized.

And S2, calculating a running speed curve according to the shunting running permission instruction sent by S12, the running obstacle identification information, the parking spot distance information, the train length information and the empty/heavy vehicle information in the job ticket, which are given by the intelligent sensing system beside the track.

In some implementations, according to S30, the method includes:

s31, acquiring an automatic driving state of the locomotive, and analyzing the driving state through a PID locomotive control algorithm (PID: proportion integration differentiation; proportional integral derivative locomotive control algorithm) to generate a speed regulation instruction;

s32, controlling the locomotive to execute the speed regulation command based on the automatic driving curve.

And according to the steps S31 and S32, acquiring the state of the locomotive automatic home ten, and transmitting corresponding running speed, direction, braking and other operation information by utilizing an interface of a locomotive control module. The speed is slowly increased, decreased and regulated by a PID locomotive control algorithm. And stopping the vehicle according to the distance between the vehicle and the opposite site or the distance between the vehicle and the obstacle to finish the parking alignment operation.

In some implementations, the S40 includes:

s41, acquiring image information of the traction front and propulsion para position of the locomotive, preprocessing the image information, executing a predefined detection task based on a deep learning network, and generating detection information;

S42, executing a lane line detection task according to the detection information, and determining current lane line coordinates;

s43, matching the automatic driving curve according to the current lane line coordinates, and controlling the locomotive to stop according to the turning locomotive position.

In accordance with steps S41, S42 and S43, vehicles, other locomotives, personnel, etc. on the line limits in front of the locomotive operation are detected. And acquiring the detected obstacle information, and controlling the locomotive to stop at a reduced speed according to the obstacle distance.

The method is based on advanced technologies such as video image recognition, intelligent combination, 5G communication and edge calculation, and realizes automatic driving of shunting operation of the shunting locomotive. The system adopts a multi-data fusion positioning mode of Beidou satellite positioning, wheel track ranging and station track state checking, and automatically tracks the position of a shunting locomotive in a station. According to a shunting operation plan, shunting driving permission instructions are calculated through collecting signal equipment state information of a computer interlocking system, and an automatic driving curve is calculated in real time by combining intelligent sensing systems at two ends of a locomotive to obtain a forward obstacle and distance information thereof, so that shunting operation such as traction, propulsion, connection and hanging, alignment and the like is automatically controlled by the shunting locomotive. The method of the invention sets the intelligent sensing equipment beside the rail at the front turning parking position, detects the vehicle alignment position in real time, and controls the locomotive to finish the accurate position alignment and parking before the car dumper. After the vehicle is turned over, the running part and the vehicle state of the vehicle are checked by intelligent video checking equipment. Meanwhile, in the operation process, the independent shunting safety protection system monitors operation by calculating a shunting operation protection curve, so that operation safety is ensured.

Embodiment two:

in a second aspect, the embodiment of the invention provides a shunting autopilot control system, which comprises a ground main control module, a vehicle-mounted shunting safety protection module, a vehicle-mounted autopilot module, a trackside intelligent sensing module and a vehicle-mounted intelligent sensing module,

In some implementations, the ground master control module includes:

The ground shunting safety protection unit in the ground main control module realizes the safety protection of the shunting locomotive when shunting operation is carried out, generates driving permission information aiming at the controlled locomotive according to the information such as track occupation, route state, speed limit information, perception obstacle, line parameters and the like in the control range of the controlled locomotive, and transmits the driving permission information to the vehicle-mounted shunting safety protection module through a wireless communication system such as a 5G or high-speed data transmission radio station and the like, thereby ensuring the shunting operation safety of the locomotive in the jurisdiction range of the shunting locomotive.

The vehicle-mounted shunting safety protection module transmits information such as a controlled request message, locomotive running data, train real-time position, manual operation confirmation information and the like to the ground main control module through a wireless communication system such as a 5G or high-speed data transmission radio station and the like, and generates a shunting safety protection curve according to shunting driving permission instruction information sent by the ground main control module and driving obstacle identification information which is intelligently perceived and output. And sending the station yard representation information, the front position of the train and the route information to an LKJ system, displaying a station yard graph, a front signal and a route on an LKJ display, and ensuring the safe running of the locomotive through safety protection and speed limiting. Meanwhile, the shunting safety protection curve is sent to the vehicle-mounted automatic driving module for speed control of the automatic driving module. When the speed of the locomotive approaches to the protection curve, the system sends out an alarm prompt tone and sends a deceleration instruction to the automatic driving module; when the speed of the locomotive reaches or exceeds the protection curve, the system directly controls the locomotive brake to brake and stop.

The difference between the intelligent sensing system and the existing wireless shunting locomotive signal and monitoring System (STP) is that the intelligent sensing system is added with safety protection of personnel, obstacles, derailers, reserved vehicles and the like in the range of the forward running path of the train detected by the intelligent sensing system besides covering all functions of the STP. The vehicle-ground communication technology is upgraded, and the 5G or high-speed data transmission radio station and the like are adopted to realize the bidirectional safety communication between vehicles and ground.

In some implementations, the in-vehicle autopilot module includes:

The vehicle-mounted automatic driving module and the ground shunting automatic driving control unit form automatic driving control system ground equipment, and are mainly used for generating driving permissions, and automatically tracking real-time positions of shunting locomotives and operation trains in a station according to multiple data fusion positioning modes of Beidou satellite differential positioning, wheel track ranging and station track state checking. The real-time positioning method specifically comprises the steps of real-time positioning of the locomotive itself and the positions of two ends of the locomotive traction or propulsion train, so that real-time position tracking of the train during traction and propulsion running in a station is realized.

The ground equipment of the automatic driving control system comprises a vehicle-mounted host, an interface with a locomotive traction circuit, an interface with a locomotive brake and the like. The automatic driving control system calculates a running speed curve according to a shunting running permission instruction sent by the ground main control module and combining running obstacle identification information given by the vehicle-mounted intelligent sensing module, parking spot distance information given by the trackside intelligent sensing module, train length information and empty/heavy vehicle information in an operation sheet. And transmitting corresponding running speed, direction, braking and other operation information by using an interface with a locomotive control module. The speed is slowly increased, decreased and regulated by a PID locomotive control algorithm. And stopping the vehicle according to the distance between the vehicle and the opposite site or the distance between the vehicle and the obstacle to finish the parking alignment operation.

The automatic driving control system has a mode switching function such as manual driving and automatic driving. When the mode is switched, self-checking is firstly carried out on the permission conditions of automatic driving, including the equipment state of an automatic driving control system, the state of a locomotive control system and the state of a brake system, so that the normal availability of the automatic driving function is ensured when the automatic driving mode is entered, and potential safety hazards caused by abnormal situations, incapacitation and the like are avoided; secondly, the locomotive driver operating part is placed at a preset position, the direction handle is in a neutral position, the self-valve and independent brake valve handles are in a release position, the traction handle is in idle rotation, and then an automatic driving input button of the man-machine interaction unit is clicked and confirmed according to a prompt; in the automatic driving mode, a locomotive driver or a remote operator has the authority of emergency intervention, an automatic driving control system responds in time after emergency braking and air braking are performed, and when the safety level of manual operation is higher than the operation level of the automatic driving control system, the automatic driving control system exits from the automatic driving mode, the locomotive control authority is manually controlled, and meanwhile, voice and text prompt is carried out in a man-machine interaction unit.

Furthermore, the vehicle-mounted intelligent sensing module is responsible for detecting vehicles, other locomotives, personnel and the like on the line limit in front of the locomotive operation. The system sends the detected obstacle information to the vehicle-mounted shunting safety protection module and the vehicle-mounted automatic driving module, and the vehicle-mounted automatic driving controls the locomotive to stop at a reduced speed according to the obstacle distance and forwards the information to the ground main control module.

The vehicle-mounted intelligent sensing module comprises a vehicle-mounted host, a laser radar, a millimeter wave radar, a camera, a temperature and humidity sensor, a sound sensor and the like, wherein the camera is responsible for providing image sensing target identification, the laser radar and the millimeter wave radar are responsible for distance sensing, and sensing results are output after host operation. The vehicle-mounted operation host adopts a CPU and GPU processing system, and provides high-performance computing power for a system image recognition and point cloud data processing algorithm. The camera adopts a low-delay, low-illumination and wide-view star light level camera to provide image data for the identification of the station information. The laser radar adopts a 3D high-resolution, long-distance and high-precision detection radar to provide accurate point cloud data for distance measurement and foreign matter detection. The millimeter wave radar adopts a multi-point, high-precision, long-distance and industrial-grade radar, and the distance measurement redundancy check data is improved.

The camera is adopted to identify the target, and the laser radar and the millimeter wave radar are adopted to conduct redundant ranging on the target, so that the accuracy of identifying the obstacle can be improved.

Further, the intelligent sensing module beside the shunting rail comprises:

the line obstacle recognition unit is used for collecting image information of the traction front and the propulsion counterpoint position of the locomotive, preprocessing the image information, executing a predefined detection task based on a deep learning network and generating detection information;

the alignment operation unit is used for executing a lane line detection task according to the detection information and determining the current lane line coordinates;

The intelligent sensing module beside the shunting rail provides ground sensing information on a shunting access, realizes V2X road cooperation in the shunting operation and driving process of the locomotive, provides barrier information such as locomotive vehicles, personnel and the like for locomotive automatic driving, and realizes shunting automation, safety and less humanization.

By constructing a low-point camera of a key area along the shunting operation line, by using a 5G network or a high-speed data transmission radio station as a communication carrier, when the shunting machine enters different operation areas, vehicle-mounted equipment automatically displays corresponding camera images along the line, and the front operation condition is assisted to be confirmed. And identifying personnel, vehicles and other larger obstacle targets entering the line limit, and providing the targets for a ground main control module to realize shunting driving permission fusion operation. In addition, at the operation place of propulsion linkage and propulsion alignment (for example, front turning, pushing and parking), an independent camera and a laser radar are arranged, so that accurate parking distance measurement is realized.

The vehicle-mounted intelligent sensing module is combined with the trackside intelligent sensing module to identify vehicles, personnel and obstacles on the shunting operation line, calculate the positions of the obstacles on the shunting access in advance, correspondingly adjust the driving speed and avoid collision accidents.

In the operation process of shunting propulsion operation, the intelligent sensing module beside the shunting rail is connected with the station video monitoring system to acquire video monitoring information in a station yard, identify obstacles on a line in the station yard through AI video analysis, send identification results to the ground main control module, realize shunting driving permission fusion operation by the ground main control module and instruct the vehicle-mounted automatic driving module to control the operation of the locomotive. The flow chart is shown in fig. 4.

The alignment operation unit needs to accurately align in the forward pushing operation, a locomotive is adjusted to push vehicles to an appointed parking point at the entrance of the dumper, the distance between indication boards of the parking point is 1m, an independent camera and a laser radar are arranged at the operation site aiming at the operation characteristics, the vehicles close to the pushing operation unit are subjected to three-dimensional scanning, the accurate operation on the target distance is realized through the fusion of laser radar point cloud data and a video AI recognition algorithm, and the recognition result is transmitted to a ground main control module and is commanded to control parking by a vehicle-mounted automatic driving module. Meanwhile, the train approaching image obtained by the camera can be transmitted to the locomotive for display, so that a driver can see an intuitive visual effect. The target ranging flowchart is shown in fig. 3.

Embodiment III:

on the basis of the second embodiment, the main functions of the automatic driving control system comprise automatic speed regulation of shunting, including functions of right-of-way, traction, braking, parking, reversing, whistling and the like. Taking the parking alignment operation before turning over as an example, the method specifically comprises the following steps:

automatic start of shunting

When the shunting has operation conditions, if a driving license instruction sent by the ground main control module is received, the intelligent perception system checks the state of no obstacle in front, and the operation list plans to receive normal system self-checking. The automatic driving control system is used for finishing locomotive starting operations such as setting a steering handle, releasing braking, whistling and the like.

Automatic operation of the section

In the alignment process of the machine, the automatic driving control system selects a proper driving strategy according to the distance from the alignment point. The acceleration, the deceleration and the idle running of the shunting are controlled, so that the automatic control of the shunting running speed is realized.

Automatic target parking

The shunting approaches to the opposite point, and the automatic driving control system automatically controls the train to stop at the opposite point according to the distance between the opposite point and the conditions (such as the length, the weight and the like) of the train.

Automatic stopping of obstacle

In the running process of the automatic driving control system, the sensing information of the intelligent sensing system is judged in the whole process, once an obstacle is encountered, the safety speed limit and the running speed are recalculated according to the distance of the obstacle sensed by the intelligent sensing system, and the vehicle is controlled to stop 20 meters in front of the obstacle.

Embodiment four:

on the basis of the third embodiment, as shown in fig. 4, the vehicle-mounted intelligent sensing module mainly includes:

(1) Target positioning: the RGB camera collects color images and sends the color images to the vehicle-mounted intelligent perception host through the communication interface. After receiving the RGB image, the vehicle-mounted intelligent perception host machine preprocesses the image, performs detection tasks based on a deep learning network, predefines detection objects including vehicles, personnel, locomotives and the like, and obtains the target category in the image and the position coordinates thereof in the RGB image; executing a lane line detection task, and determining the current lane line coordinates by means of slope constraint, coordinate constraint and the like; performing line matching, and determining whether a predefined target exists in the current lane line;

(2) The radar sensor collects information of vertical angle, horizontal angle, echo ranging and echo reflection intensity, the information is transmitted to the vehicle-mounted operation host through a network, and the vehicle-mounted operation host calculates XYZ coordinate values according to the ranging and angle information to generate PCL point cloud data;

(3) Carrying out point cloud image data fusion, and carrying out data fusion according to the internal and external parameters of the camera, which are obtained after the joint calibration of the camera, the laser radar and the millimeter wave radar: and the corresponding relation between the three-dimensional space point of the point cloud and the pixel point of the two-dimensional plane of the image is determined by projecting the point cloud onto the two-dimensional plane of the image, the time stamp verification is required in the fusion process, the time consistency of the image and the point cloud is ensured, and finally, the distance from the current train to the target in the front operation track is obtained according to the target position found in the target positioning task.

The vehicle-mounted intelligent sensing host transmits the intelligent sensing result and the obstacle distance to the vehicle-mounted shunting safety protection module and the vehicle-mounted automatic driving control system in real time, the vehicle-mounted shunting safety protection module and the vehicle-mounted automatic driving control system update driving permission in real time, the sensing distance and the safety side (namely smaller value) protection distance of the front part of the train from the parking signal distance are taken for controlling the train, and the train is reasonably controlled to park.

Fifth embodiment:

the present embodiment also provides a computer device, including a memory and a processor, where the memory is configured to store one or more computer instructions, and where the one or more computer instructions, when executed by the processor, implement the method of the first embodiment;

in practical applications, the processor may be an application specific integrated circuit (Application Specific Integrated Circuit, abbreviated as ASIC), a digital signal processor (Digital Signal Processor, abbreviated as DSP), a digital signal processing device (Digital Signal Processing Device, abbreviated as DSPD), a programmable logic device (Programmable Logic Device, abbreviated as PLD), a field programmable gate array (Field Programmable Gate Array, abbreviated as FPGA), a controller, a microcontroller (Microcontroller Unit, MCU), a microprocessor or other electronic component implementation for executing the method in the above embodiment.

The method implemented by the embodiment comprises the following steps:

fig. 1 shows a flow chart of a shunting automatic driving method, and as shown in fig. 1, the shunting automatic driving method provided in this embodiment includes:

In some implementations, the S10 includes:

In some implementations, according to S30, the method includes:

In some implementations, the S40 includes:

Example six:

the present embodiment also provides a computer storage medium having a computer program stored therein, which when executed by one or more processors, implements the method of the first embodiment;

the computer readable storage medium may be implemented by any type or combination of volatile or nonvolatile Memory devices, such as static random access Memory (Static Random Access Memory, SRAM for short), electrically erasable programmable Read-Only Memory (Electrically Erasable Programmable Read-Only Memory, EPROM for short), programmable Read-Only Memory (Programmable Read-Only Memory, PROM for short), read-Only Memory (ROM for short), magnetic Memory, flash Memory, magnetic disk, or optical disk.

The method implemented by the embodiment comprises the following steps:

In some implementations, the S10 includes:

In some implementations, according to S30, the method includes:

In some implementations, the S40 includes:

In the several embodiments provided in the embodiments of the present application, it should be understood that the disclosed system and method may be implemented in other manners. The system and method embodiments described above are merely illustrative.

It should be noted that, in this document, the terms "first," "second," and the like in the description and the claims of the present application and the above drawings are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

Although the embodiments of the present application are described above, the embodiments are only used for facilitating understanding of the present application, and are not intended to limit the present application. Any person skilled in the art can make any modification and variation in form and detail without departing from the spirit and scope of the present disclosure, but the scope of the present disclosure is still subject to the scope of the appended claims.

Claims

1. A shunting autopilot method, the method comprising:

2. The method of claim 1, wherein the obtaining the in-station locomotive position, obtaining a shunting service approval command according to a shunting operation plan, comprises:

and according to the shunting operation plan, acquiring signal equipment state information analysis of the computer interlocking system and generating a shunting driving permission instruction based on the current position of the locomotive and the target operation area.

3. The method of claim 2, wherein controlling the locomotive to travel based on the autopilot profile according to the autopilot profile comprises:

4. The method of claim 3, wherein the obtaining the locomotive alignment position, that is, obtaining video and radar point cloud information of the locomotive propulsion alignment position, parsing the video information to obtain environmental information, personnel and obstacle intrusion and vehicle status information of a front end of a propulsion train, comprises:

5. A shunting autopilot control system, the system comprising:

the vehicle-mounted intelligent sensing module is used for acquiring video images and radar point cloud information of the position in front of the locomotive running, and analyzing and acquiring environmental information, personnel, barriers and vehicle state information in a limiting range of the locomotive after multi-sensor fusion is carried out on the video and radar point cloud information.

6. The system of claim 5, wherein the ground master control module comprises:

7. The system of claim 6, wherein the on-board autopilot module comprises:

the shunting safety protection unit is used for safety protection in the shunting operation process, and prevents the shunting train from impoverishment of a front closing signal, extrusion of a turnout, collision of a dead end line earth shield, wrong crossing of a station landmark, overspeed running, overspeed connection hanging and the like through real-time tracking of the shunting train position;

and the automatic driving control unit is used for acquiring the automatic driving state of the locomotive, analyzing the driving state through a PID locomotive control algorithm to generate a speed regulation instruction, and controlling the locomotive to execute the speed regulation instruction based on the automatic driving curve.

8. A computer device comprising a memory and a processor, the memory for storing one or more computer instructions, wherein the one or more computer instructions when executed by the processor implement the method of any of claims 1-4.

9. An electronic device comprising a memory and a processor, the memory configured to store one or more computer instructions, wherein the one or more computer instructions when executed by the processor implement the in-station job automatic notification method of any of claims 1-4.

10. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, which when executed by a processor, is configured to implement the in-station job automatic notification method according to any one of the preceding claims 1-4.