CN114872741B

CN114872741B - Locomotive auxiliary automatic driving system and method based on safety guidance

Info

Publication number: CN114872741B
Application number: CN202210602378.3A
Authority: CN
Inventors: 曲长萍; 王瑞; 姜正; 杜海宾; 任育琦
Original assignee: CRRC Dalian Institute Co Ltd
Current assignee: CRRC Dalian Institute Co Ltd
Priority date: 2022-05-30
Filing date: 2022-05-30
Publication date: 2023-10-03
Anticipated expiration: 2042-05-30
Also published as: CN114872741A

Abstract

The invention discloses a locomotive auxiliary automatic driving system and method based on safety guidance, comprising the following steps: the system comprises a vehicle-mounted module, a locomotive interface module, an interaction module and an environment safety module; the locomotive interface module is connected with the vehicle-mounted module; the interaction module is connected with the vehicle-mounted module; the environment safety module is connected with the vehicle-mounted module; the millimeter wave radar of the environment safety module is connected with the vehicle-mounted module; the laser radar is connected with the vehicle-mounted module. According to the invention, the environment safety module is arranged in the locomotive auxiliary automatic driving system, so that locomotive auxiliary automatic driving based on safety guidance is realized, the technical problems that a single sensor has a vision blind area, the measurement accuracy of a video sensor is reduced along with the increase of distance, the detection definition of objects at a longer distance is insufficient, and accurate distance information of the objects cannot be obtained are solved. The accuracy of obstacle detection can be improved. The running safety of the locomotive is greatly improved, and the fatigue degree of a driver in the driving process is reduced.

Description

Locomotive auxiliary automatic driving system and method based on safety guidance

Technical Field

The invention relates to the technical field of locomotive auxiliary automatic driving systems, in particular to a locomotive auxiliary automatic driving system and method based on safe guiding.

Background

The existing railway locomotive mainly uses manual driving because of fixed track lines, has potential safety hazards because of long running lines of the locomotive and easy fatigue driving and the like, and can observe the surrounding environment through human eyes and has certain hysteresis in addition to human reaction during manual driving, so that the perception degree of the surrounding environment is weaker, and the emergency situation can not be avoided in time. The driving intensity of a driver can be reduced by adding the multi-sensor equipment, the tension emotion of the driver is relieved, and the running safety of the locomotive is improved.

At present, the existing sensing equipment mainly comprises a single sensor, the single sensor is easy to detect to cause an information blind area, and information expression has limitation. As with the 6A video capture devices deployed on existing locomotives, the accuracy of video sensor measurements decreases with distance, and the detection resolution for objects at longer distances is inadequate, which is a result of the physical properties of the sensor. In addition, the existing video data cannot be processed in real time, can only be used for post problem analysis, and cannot meet the requirement of intelligent rail transit development.

The application of the sensor fusion in the intelligent traffic field is gradually promoted, but the application in the large railway industry is still to be promoted, and especially the railway application environment of special scenes such as heavy load, long and large ramp and the like is more required to realize the long-term development of intelligent rail traffic by utilizing the information fusion of a plurality of sensors.

Disclosure of Invention

The invention provides a locomotive auxiliary automatic driving system and method based on safety guide so as to overcome the technical problems.

In order to achieve the above object, the technical scheme of the present invention is as follows:

the locomotive auxiliary automatic driving system determines the operation safety level of the locomotive according to a safety state evaluation model built in the vehicle-mounted module so as to realize locomotive auxiliary automatic driving; comprising the following steps: the system comprises a vehicle-mounted module, a locomotive interface module, an interaction module and an environment safety module;

the locomotive interface module is connected with the vehicle-mounted module to realize data exchange between the locomotive interface module and the vehicle-mounted module, so that the vehicle-mounted module can evaluate the safety state of the locomotive according to LKJ data, TCMS data, driver and passenger state data, brake data and train tail data, obtain the operation safety level of the locomotive, and realize auxiliary automatic driving of the locomotive according to the operation safety level of the locomotive;

the interaction module is connected with the vehicle-mounted module to realize data exchange between the interaction module and the vehicle-mounted module, and realize the display of dangerous early warning, track following and locomotive running states, planning curves of locomotive running, automatic driving traction/braking force and the interaction of manual operation and automatic driving instructions through the interaction module;

The environment safety module is connected with the vehicle-mounted module to realize data exchange between the environment safety module and the vehicle-mounted module, and the environment data acquired by the environment safety module are transmitted to the vehicle-mounted module;

the environment safety module comprises a plurality of cameras, millimeter wave radars and laser radars; the camera is connected with the vehicle-mounted module to acquire an obstacle image in front of the running direction of the locomotive and transmit the obstacle image to the vehicle-mounted module;

the millimeter wave radar is connected with the vehicle-mounted module to acquire the relative distance data and azimuth data of the obstacle and the current locomotive, and transmits the relative distance data and the azimuth data to the vehicle-mounted module;

the laser radar is connected with the vehicle-mounted module to acquire coordinates, size and movement direction of an obstacle in front of the running direction of the locomotive and relative speed data of the obstacle and the current locomotive, and the relative speed data is transmitted to the vehicle-mounted module.

Further, the vehicle-mounted module comprises an interface unit, a central control unit and a data storage unit;

the interface unit is connected with the locomotive interface module so as to realize data exchange between the vehicle-mounted module and the locomotive interface module through the interface unit; LKJ data, TCMS data, driver and passenger state data, brake data and train tail data of a locomotive interface module are obtained through the locomotive interface unit;

The interface unit is connected with the interaction module; the vehicle-mounted module is used for realizing data exchange between the vehicle-mounted module and the interaction module through the interface unit; the method comprises the steps of realizing the visualization of dangerous early warning, track following, locomotive running state and planning curve of locomotive running through the interaction module, and realizing the interaction of manual operation and automatic cab instructions;

the interface unit is connected with the environment safety module; the vehicle-mounted module is used for realizing data exchange between the vehicle-mounted module and the environment safety module through the interface unit;

the data storage module is connected with the interface unit to realize data exchange between the data storage module and the interface unit, and pre-stores the information of the current running line of the locomotive, including but not limited to the ground gradient and the locomotive speed limit curve; the data storage module establishes a route characteristic database according to the automatic driving operation data of the high-incidence area of the safety accident or the ground slope change area;

furthermore, the storage module can also establish a database for the state data of drivers and passengers, and statistics is carried out on the operation data in actual operation, so as to extract the driving characteristics of the drivers and passengers; taking the driving characteristics of the drivers and passengers as parameters for establishing a driver and passenger state model, and further planning an operation speed curve;

The central control unit is connected with the interface unit to realize data exchange between the central control unit and the interface unit, a route characteristic database is established according to a high-incidence area of a safety accident or a ground gradient change area, characteristics in the route characteristic database are used as parameters in the safety state assessment model to assess the safety state of the locomotive, the operation safety level of the locomotive is obtained, and auxiliary automatic driving of the locomotive is realized according to the operation safety level of the locomotive.

Further, the method for evaluating the safety state of the locomotive comprises the following steps:

determining the running safety level ABCD of the locomotive through a safety state evaluation model; starting corresponding safety early warning according to the operation safety level of the locomotive, and guiding the locomotive to re-plan an operation speed curve;

the safety state evaluation model comprises a locomotive speed guiding model, a locomotive fault guiding model, an environment safety guiding model and a driver and passenger state guiding model;

the locomotive obtains the locomotive speed safety level through the locomotive speed guiding model;

the locomotive obtains the locomotive fault safety level through the locomotive fault guiding model;

the locomotive obtains the locomotive environmental security level through the environmental security guiding model;

The locomotive obtains the safety level of the locomotive driver and passenger through the driver and passenger state guide model;

the safety state evaluation model takes minimum values of a locomotive speed safety level, a locomotive fault safety level, a locomotive environment safety level and a locomotive driver and passenger safety level.

Further, the method for acquiring the locomotive speed safety grade by the locomotive through the locomotive speed guiding model comprises the following steps:

the locomotive speed guiding model monitors the running speed of the locomotive in real time to obtain the locomotive speed safety grade A, A epsilon (0,1,3,5): namely, the locomotive speed operation safety level comprises a safety level 5, a safety level 3, a safety level 1 and a safety level 0;

specifically, when the running speed of the locomotive does not exceed a speed safety threshold, the load of the locomotive does not exceed a load threshold, and the gradient value of the current running line is not greater than a gradient threshold, the running speed safety grade A=5 of the locomotive is not required to be planned again at the moment;

when the speed of the locomotive does not exceed the speed safety threshold, the load of the locomotive exceeds the load threshold and the gradient value of the current running line is greater than the gradient threshold, the safety grade A=3 of the speed of the locomotive, and the locomotive sends a danger early warning prompt to drivers and passengers at the moment, so that the running speed curve does not need to be planned again;

When the speed of the locomotive does not exceed the speed safety threshold and the difference between the running speed of the locomotive and the speed of the current position point on the running curve of the locomotive is smaller than the speed difference threshold, the safety grade A=1 of the speed of the locomotive, and the locomotive adjusts the real-time running speed at the moment without re-planning the running speed curve;

when the speed of the locomotive exceeds a speed safety threshold, or the difference between the speed of the locomotive running speed and the speed of the current position point on the locomotive running curve is smaller than a speed difference threshold, the locomotive speed safety level A=0, the locomotive is converted from an automatic driving mode to an artificial driving mode within a countdown time threshold, and the running speed curve is re-planned.

Further, the method for acquiring the locomotive fault safety level by the locomotive through the locomotive fault guiding model comprises the following steps: the locomotive fault guiding model analyzes whether the current fault affects the automatic driving of the locomotive according to the vehicle-mounted network fault model so as to acquire locomotive fault safety grades B and B epsilon (0,1,3,5); determining whether a re-planning of the running speed curve is required; namely, the locomotive fault safety level comprises a safety level 5, a safety level 3, a safety level 1 and a safety level 0;

specifically, when the locomotive has no fault, the locomotive fault safety level b=5, and the locomotive does not need to reprogram the running speed curve;

When the number of locomotive isolating shafts is smaller than the first isolating shaft number threshold value, the locomotive fault safety level B=3, and the locomotive sends a danger early warning prompt to drivers and passengers at the moment, so that the running speed curve does not need to be planned again;

when the number of the locomotive isolating shafts is larger than the first isolating shaft number threshold and smaller than the second isolating shaft number threshold, the locomotive fault safety level B=1, and the locomotive adjusts the real-time running speed at the moment without re-planning the running speed curve;

when the number of the locomotive isolating shafts is larger than the threshold value of the number of the second isolating shafts, the locomotive fault safety level B=0, the locomotive is switched from the automatic driving mode to the manual driving mode, and the running speed curve is planned again.

Further, the method for acquiring the locomotive environmental safety grade by the locomotive through the environmental safety guide model comprises the following steps:

the environmental safety guide model acquires locomotive environmental safety grades C, C epsilon (0,1,3,5) according to the acquired obstacle size and the distance between the obstacle and the locomotive; namely, the locomotive environment safety level comprises a safety level 5, a safety level 3, a safety level 1 and a safety level 0;

specifically, when no obstacle exists in front of the running direction of the locomotive, or the size of the obstacle is smaller than a size threshold, the hardness of the obstacle is smaller than a hardness threshold, and the distance between the obstacle and the locomotive is larger than a distance threshold, the environmental safety level c=5 of the locomotive is achieved, and the locomotive does not need to reprogram a running speed curve at the moment;

When an obstacle with the size larger than the size threshold and the hardness smaller than the hardness threshold is arranged in front of the running direction of the locomotive, the locomotive environmental safety level C=3, and at the moment, the locomotive sends a danger early warning prompt to drivers and passengers without re-planning the running speed curve;

when a dynamic obstacle with the size larger than a size threshold is arranged in front of the running direction of the locomotive, the environmental safety level C=1 of the locomotive, and the locomotive adjusts the real-time running speed at the moment without re-planning the running speed curve;

when the distance between the obstacle in front of the running direction of the locomotive and the locomotive is smaller than a distance threshold value, the locomotive environmental safety level C=0, the locomotive is converted from an automatic driving mode to a manual driving mode, and the running speed curve is re-planned.

Further, the method for acquiring the safety level of the driver and the passengers of the locomotive by the driver and the passengers state model of the locomotive comprises the following steps: the driver and passenger state model acquires the safety grades D, D epsilon (0,1,3,5) of the driver and passenger of the locomotive according to the 6A system data; namely, the safety level of locomotive drivers and passengers comprises a safety level 5, a safety level 3 and a safety level 0;

specifically, when a driver and a passenger determine that the driver has no sleepiness state or the duration of the driver and the passenger continuously watching the mobile phone does not exceed a time threshold according to a signal of the sleepiness early-warning system of the automobile driver, the safety level D=5 of the driver and the passenger is ensured, and the locomotive does not need to reprogram an operation speed curve at the moment;

When a driver and a passenger determine that the driver is in a doze state or the duration of the driver and the passenger continuously watching a mobile phone exceeds a time threshold according to an alarm signal provided by a doze early-warning system of an automobile driver, the safety level D=3 of the driver and the passenger is ensured, and at the moment, a locomotive sends a danger early-warning prompt to the driver and the passenger without re-planning a running speed curve;

when drivers and passengers are in a doze state according to the alarm signals provided by the doze early-warning system of the automobile driver, the drivers and passengers are still in the doze state after being warned for a plurality of times, the safety level D=0 of the drivers and passengers is used for emergency braking operation, and meanwhile, the running speed curve is planned again.

Further, the environment safety guide model comprises an information acquisition function module, an information perception function module, an environment recognition function module and a target following function module,

the information acquisition functional module performs data integration processing on barrier information in front of the running direction of the locomotive, which is acquired by the camera, the millimeter wave radar and the laser radar, so as to acquire barrier data;

the information perception function module obtains the shape of the obstacle through relative distance data, azimuth angle data and relative speed data of the obstacle and the current locomotive;

The environment recognition function module acquires the obstacle state through a convolution algorithm to establish a safe distance model and acquire the locomotive environment safety level; the obstacle states include static obstacle and dynamic obstacle movement states;

the target following function module carries out real-time following on the dynamic obstacle information; tracking an obstacle in front of the running direction of the locomotive;

the method for establishing the safe distance model comprises the following steps:

and obtaining the safety distance of the current vehicle running of the locomotive according to the obstacle size, the running speed of the locomotive, the current line condition, the locomotive load, the dynamic obstacle movement state and the line characteristics of the road section or the line gradient change area of the high-incidence area of the safety accident.

A design method of a locomotive auxiliary automatic driving system based on safety guidance comprises the following steps:

s1: the locomotive auxiliary automatic driving system acquires locomotive line information, locomotive information and front signal state information;

s2: establishing a locomotive running model according to the locomotive line information, locomotive information and front signal state information;

s3: determining a locomotive operation safety level according to the locomotive operation model and combining the safety state evaluation model so as to guide a locomotive planning operation speed curve according to the locomotive operation safety level;

S4: and outputting the locomotive running speed curve and the early warning information through the interaction module.

Further, when the locomotive is heavily loaded, the step S3 further includes: and establishing a multi-particle model and a coupler buffer model of the heavy-duty locomotive according to the parameters of the heavy-duty locomotive, the parameters of the vehicle and the parameters of the buffer system so as to reduce the longitudinal impact force of the combined locomotive.

The beneficial effects are that: according to the locomotive auxiliary automatic driving system and method based on the safety guide, the environment safety module is arranged in the locomotive auxiliary automatic driving system, so that the locomotive auxiliary automatic driving based on the safety guide is realized, the technical problems that a visual blind area exists in a single sensor, the measurement accuracy of a video sensor is reduced along with the increase of the distance, the detection definition of objects at a longer distance is insufficient, and the accurate distance information of the objects cannot be obtained are solved. The accuracy of obstacle detection can be improved. The running safety of the locomotive is greatly improved, and the fatigue degree of a driver in the driving process is reduced.

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 diagram of a locomotive auxiliary autopilot system equipment connection based on safety guidance;

FIG. 2 is a block diagram of a safety guidance based locomotive auxiliary autopilot system design;

FIG. 3 is a block diagram of an environmentally safe guided locomotive auxiliary autopilot system;

FIG. 4 is a schematic diagram of a locomotive auxiliary autopilot system.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The present embodiment provides a locomotive auxiliary automatic driving system based on safety guidance, as described in fig. 1-4;

on the premise of ensuring that the original locomotive device can normally operate, when the automatic driving condition is met, the man-machine interaction display unit enters an auxiliary automatic driving state of the locomotive through the driving state switching interface. Comprising the following steps: LKJ equipment, a TCMS system, a 6A system, a braking unit, a tail arrangement device, a vehicle-mounted module, a locomotive interface module, an interaction module and an environment safety module;

Specifically, the LKJ equipment is a locomotive operation monitoring device and is a common component of a China railway locomotive operation control system; the TCMS system is a general locomotive control and management system; the 6A system is a locomotive-mounted safety protection system (called as 6A system for short) and is a system which is used for carrying out real-time detection, monitoring and alarming and has the functions of network transmission, unified solid state storage, intelligent man-machine interface and the like aiming at important matters related to safety such as high-voltage insulation, fire prevention, video, locomotive power supply, a braking system, a running part and the like of a locomotive. The LKJ equipment is connected with the locomotive interface module to realize data exchange between the LKJ equipment and the locomotive interface module and obtain LKJ data of locomotive operation; the LKJ data includes: grouping information, total locomotive weight, number of vehicles, ramp condition, locomotive speed limit value, kilometer post, front annunciator state; the TCMS system is connected with the locomotive interface module to realize data exchange between the TCMS system and the locomotive interface module and obtain TCMS data; the TCMS data comprise locomotive running speed, traction/braking force, working condition data and fault information data; the 6A system is connected with the locomotive interface module to realize data exchange between the 6A system and the locomotive interface module and obtain driver and passenger state data; the braking unit is connected with the locomotive interface module to realize data exchange between the braking unit and the locomotive interface module and obtain the brake data in the locomotive running process; the train tail device is connected with the locomotive interface module to realize data exchange between the train tail device and the locomotive interface module and obtain train tail data;

Specifically, the LKJ device provides LKJ data, i.e., locomotive operation basic line data, including, but not limited to, consist information, total locomotive weight, number of vehicles, ramp conditions, locomotive speed limit curves, kilometers scale, front traffic light status, etc.; the TCMS system provides TCMS data, including locomotive running basic data such as locomotive running speed, traction/braking force, working condition, fault information and the like; the 6A system provides 6A system data, including operation videos in the driving process of drivers and passengers; the braking unit provides brake data, namely, executes an air decompression control instruction sent by automatic driving; the train tail device provides train tail data, namely train tail wind pressure data, and feeds back real-time wind pressure information of the tail for the automatic driving system.

Specifically, the vehicle-mounted module is the core of the whole locomotive auxiliary automatic driving system, is the brain of locomotive operation, is a processing center of all data, and combines the data of the environment safety module and the locomotive interface module to realize the functions of planning curve design, danger early warning, track following and the like of locomotive operation based on safety guidance;

preferably, the vehicle-mounted module comprises an interface unit, a central control unit and a data storage unit;

the interface unit is connected with the locomotive interface module so as to realize data exchange between the vehicle-mounted module and the locomotive interface module through the interface unit; LKJ data, TCMS data, driver and passenger state data, brake data and train tail data of a locomotive interface module are obtained through the locomotive interface unit, and interaction is carried out with a brake through a vehicle-mounted module, so that emergency braking operation in an emergency state is realized;

specifically, the interface unit of the vehicle-mounted module interacts with the data of the locomotive interface module to acquire the vehicle-mounted data acquired by the locomotive interface module, and the emergency braking treatment in an emergency state is realized by interacting with the brake; and interacting with the data of the environment safety module to acquire the environment information acquired by the environment safety module through the camera and the millimeter wave radar. The interaction module is used for interacting with the data of the interaction module to realize the interaction information of instructions such as switching between automatic driving and manual driving and the interaction information such as running state display, danger early warning and track following of the locomotive.

specifically, the interaction module in the embodiment can visualize the following information including dangerous early warning, track following, locomotive running state, environmental line, automatic driving curve dynamic planning display, automatic driving traction/braking force, real-time running state, manual operation, automatic driving room instruction interaction and other functions. The interaction module provides voice early warning information when the danger signal is received, and the interaction module can be manually operated to switch to manual driving or can control emergency stopping by sending a braking instruction to the braking unit in an automatic driving mode due to emergency when the countdown time is set.

The central control unit is connected with the interface unit to realize data exchange between the central control unit and the interface unit, and establishes a route characteristic model according to a high-incidence area of a safety accident or a ground gradient change area so as to evaluate the safety state of the locomotive, acquire the operation safety level of the locomotive and realize auxiliary automatic driving of the locomotive according to the operation safety level of the locomotive;

specifically, the central control unit of the vehicle-mounted module is a mature control unit which is universal on a rail locomotive and has the functions of optimizing calculation and self-adaptive control, the vehicle-mounted data acquired by the locomotive interface unit and the data of the environment safety module are received, the vehicle-mounted data and the data of the environment safety module specifically comprise information such as locomotive speed, front signal state, speed limit value and ground gradient, driver driving state, current fault state of the locomotive and the like, the safety guidance of the operation of the locomotive is evaluated by combining with the environment safety module, and the functions of planning curve design, danger early warning, track following and the like of the operation of the locomotive can be realized.

The environment safety module is connected with the vehicle-mounted module to realize data exchange between the environment safety module and the vehicle-mounted module, and the environment data acquired by the environment safety module are transmitted to the vehicle-mounted module; the environmental data includes coordinates, dimensions, and direction of movement of an obstacle forward of the direction of travel of the locomotive, relative distance data and azimuth data of the obstacle from the current locomotive, and relative speed data.

The data storage module is connected with the interface unit to realize data exchange between the data storage module and the interface unit, and pre-stores the information of the current running line of the locomotive, including but not limited to the ground gradient and the locomotive speed limit curve; and a database for storing the automatic driving operation data of the accident high incidence area or the ground gradient change area. Furthermore, the storage module can also establish a database for the state data of drivers and passengers, and statistics is carried out on the operation data in actual operation, so as to extract the driving characteristics of the drivers and passengers; taking the driving characteristics of the drivers and passengers as parameters for establishing a driver and passenger state model, and further planning an operation speed curve; specifically, the method for establishing the driver state model according to the driver driving characteristic as a parameter in this embodiment is a conventional technology in the field, and will not be described in detail herein.

Specifically, the data storage module pre-stores basic information of the current running line of the locomotive, including but not limited to ramp conditions, locomotive speed limit curves and the like, interacts with the vehicle-mounted LKJ during each power-on running, confirms whether the data version information is consistent or not, and if not, re-acquires and stores new LKJ version data through the interface unit, and uploads the data to the display unit of the interaction module through the interface unit.

Preferably, the environment safety module comprises a plurality of cameras, millimeter wave radars and laser radars;

the environment safety module senses the running environment condition of the locomotive, and the sensing and real-time processing capacity of the locomotive to the external environment is increased through sensors such as cameras and radars. Preferably, the camera is connected with the vehicle-mounted module to acquire an obstacle image in front of the running direction of the locomotive and transmit the obstacle image to the vehicle-mounted module;

the laser radar is connected with the vehicle-mounted module to acquire coordinates, size and movement direction of an obstacle in front of the running direction of the locomotive and relative speed data of the obstacle and the current locomotive, and the relative speed data is transmitted to the vehicle-mounted module. In the embodiment of the invention, the camera, the millimeter wave radar and the laser radar are all direct applications to the existing products, and the patent only uses the camera, the millimeter wave radar and the laser radar to collect the information of the locomotive running environment, so that the details are not repeated here.

Specifically, in one embodiment of the present invention, the environmental security module is composed of a camera and a millimeter wave radar, where the camera and the millimeter wave radar collect data for an observation target respectively, then a central control unit of the vehicle-mounted module performs feature extraction and pattern recognition processing on environmental data transmitted by the camera and the millimeter wave radar, and accurately associates the targets according to categories, and finally utilizes a fusion algorithm to integrate all sensor data of the same target, accurately detect positions, types, speeds, and the like of obstacles, and realize a locomotive surrounding environment sensing function. And the environment safety module avoids dead angles in driving through cooperation operation among various sensors according to the difference of the accuracy and the length of data acquired by the camera and the millimeter wave radar. The environment safety module is used for detecting the running road conditions of the vehicle in real time, so that the working intensity of drivers and passengers can be reduced, and the running safety is improved.

In another embodiment of the invention, the environment security module comprises a camera, a millimeter wave radar and a laser radar, and a laser radar sensor is added on the basis of the integration of the camera and the millimeter wave radar to integrate the acquired data of the camera, the millimeter wave radar and the laser radar; the laser radar can divide single line and multiple lines according to the number of scanning lines, and can divide one dimension, two dimensions and three dimensions according to an imaging method, wherein the one dimension is used for measuring the distance of a target, the information characteristics of the target are planar, the two dimensions accurately identify and track the outline of the target object such as a tree, and the three dimensions accurately identify and track the target. The laser radar can not only help the unmanned vehicle control system to acquire the accurate position of the target information, but also acquire other information characteristics such as the speed, the size and the direction of the target information. The environment safety module can accurately identify and track the target in the low-speed running process of the locomotive, especially in the condition that the locomotive runs on a road section with limited speed requirements, the locomotive is prevented from slipping, the locomotive is put in and put out, and the turnout is connected with a residential living area, so that the running safety of the locomotive is improved. The method comprises the steps of performing feature extraction and pattern recognition processing on environment data, accurately associating targets according to categories, and finally integrating all sensor data of the same target by using a fusion algorithm, which are all the prior art, and are not described in detail herein.

Preferably, the method for evaluating the safety state of the locomotive comprises the following steps:

determining the running safety level ABCD of the locomotive through a safety state evaluation model; starting corresponding safety early warning according to the operation safety level of the locomotive, and guiding the locomotive to re-plan an operation speed curve; the safety state evaluation model is a software program built in the vehicle-mounted module;

the safety state evaluation model comprises a locomotive speed guiding model, a locomotive fault guiding model, an environment safety guiding model and a driver and passenger state guiding model; the locomotive obtains the locomotive speed safety level through the locomotive speed guiding model; the locomotive obtains the locomotive fault safety level through the locomotive fault guiding model; the locomotive obtains the locomotive environmental security level through the environmental security guiding model; the locomotive obtains the safety level of the locomotive driver and passenger through the driver and passenger state guide model;

The locomotive operation safety level comprises a safety level 5, a safety level 3, a safety level 1 and a safety level 0; specifically, the design of the security level is set by those skilled in the art according to long-term working experience, and is built in a software program of the vehicle-mounted module for application.

The locomotive speed guiding model monitors the running speed of the locomotive in real time through the LKJ equipment to obtain the locomotive speed safety grade A, A epsilon (0,1,3,5): namely, the locomotive speed operation safety level comprises a safety level 5, a safety level 3, a safety level 1 and a safety level 0; specifically, according to the locomotive speed safety grade obtained by the locomotive speed guiding model, planning a locomotive running curve and controlling locomotive running in real time within a safety threshold range according to the locomotive limiting speed curve requirement, if necessary, combining the current locomotive running road condition, whether the current locomotive running road condition is a slope road, whether the gradient influences locomotive control, locomotive carrying capacity and the like, if the current speed has a wire collision danger, when the current speed needs to be converted into manual driving, further, establishing a database for driver and passenger state data according to a storage module, counting the state data of the driver and passenger in actual running, and extracting the driving characteristics of the driver and passenger; and (3) re-planning a speed curve according to the driving characteristics of different drivers and passengers, wherein if the drivers and passengers like to quickly change gear, the drivers and passengers like to slowly regulate the speed, and the specific planning process needs to be fitted for a plurality of times by combining actual data. Specifically, for the heavy-duty locomotive or long and large downhill scenes, the characteristics of the high-incidence area or the line gradient change area of the safety accident are combined, or the speed limit regulation threshold values are different when other safety guiding states are combined, the LKJ device can be used, and the functions can be realized by independent equipment.

When the running speed of the locomotive does not exceed the speed safety threshold, the load of the locomotive does not exceed the load threshold and the gradient value of the current running line is not greater than the gradient threshold, the safety grade A=5 of the locomotive speed, and the locomotive does not need to re-plan the running speed curve at the moment; specifically, when the load of the locomotive exceeds the load threshold, the locomotive is overloaded, the specific load threshold is different according to locomotive parameters such as the locomotive model number, the locomotive number and the like, and the load threshold of each locomotive is already set before running and is directly defined in the vehicle-mounted system. The gradient threshold value is also preset and completed according to the current running route of the locomotive before the locomotive runs.

when the speed of the locomotive does not exceed the speed safety threshold value and the difference between the speed of the locomotive running speed and the speed of the current position point on the locomotive running curve is smaller than the speed difference threshold value, the locomotive speed safety grade A=1, and the locomotive running speed is close to the running speed curve and has the risk of wire collision but does not influence the normal running of the locomotive, so that the locomotive adjusts the real-time running speed without re-planning the running speed curve;

When the speed of the locomotive exceeds a speed safety threshold value or the difference between the speed of the locomotive running speed and the speed of the current position point on the locomotive running curve is smaller than a speed difference value threshold value, the locomotive speed safety grade A=0, and the locomotive is converted into an automatic driving mode to be in an artificial driving mode, and the running speed curve is re-planned; in another embodiment of the present invention, the locomotive may control emergency stopping by the auxiliary autopilot system sending a braking order to the braking interface module within the countdown time threshold.

The locomotive fault guiding model analyzes whether the current fault affects the automatic driving of the locomotive according to the vehicle-mounted network fault model so as to acquire locomotive fault safety grades B and B epsilon (0,1,3,5); namely, the locomotive fault safety level comprises a safety level 5, a safety level 3, a safety level 1 and a safety level 0; determining whether a re-planning of the running speed profile is required or not and whether adjustment of the traction/braking force is required or not;

when the locomotive is fault-free, the locomotive fault safety level B=5, and the locomotive does not need to reprogram the running speed curve;

when the number of the locomotive isolating shafts is smaller than the first isolating shaft number threshold, and normal operation of the locomotive is not affected at this time, the locomotive fault safety level B=3, and at this time, the locomotive sends a danger early warning prompt to drivers and passengers, and the operation speed curve does not need to be planned again;

When the number of the locomotive isolating shafts is larger than the first isolating shaft number threshold and smaller than the second isolating shaft number threshold, and normal operation of the locomotive is not affected at this time, the locomotive fault safety level B=1, and the locomotive adjusts the real-time operation speed at this time without re-planning the operation speed curve;

when the number of the locomotive isolating shafts is larger than the threshold value of the number of the second isolating shafts, the locomotive cannot normally run at the moment, the locomotive fault safety level B=0, the locomotive is converted into an automatic driving mode into a manual driving mode, and the running speed curve is re-planned; the first isolation shaft number threshold value in the embodiment is smaller than the second isolation shaft number threshold value, and the value ranges of the first isolation shaft number threshold value and the second isolation shaft number threshold value are different according to different locomotive models.

According to the embodiment, the number of the isolated axles of the locomotive is used as the dividing basis of the fault level of the locomotive, specifically, the number of the isolated axles of the 6-axle locomotive and the 8-axle locomotive are different, the number of the isolated axles of the first isolated axle is related to the type of the locomotive, and the threshold value of the number of the isolated axles of the second isolated axle is half of the sum of the number of the axles of the locomotive.

When no obstacle exists in front of the running direction of the locomotive, or the size of the obstacle is smaller than a size threshold value, the hardness of the obstacle is smaller than a hardness threshold value, and the distance between the obstacle and the locomotive is larger than a distance threshold value, the environmental safety level C=5 of the locomotive is achieved, and the locomotive does not need to reprogram a running speed curve;

when the distance between the obstacle in front of the running direction of the locomotive and the locomotive is smaller than a distance threshold value, the environmental safety level C=0 of the locomotive, the locomotive is converted from an automatic driving mode to a manual driving mode, and the running speed curve is re-planned;

specifically, in this embodiment, according to the locomotive operation safety standard and the conventional known technology of those skilled in the art, for the locomotive under different operation lines and different real-time operation speeds, for the size standard of the small-size obstacle, the definition standard of the soft material, the size standard of the oversized static obstacle, the definition standard of the hard material obstacle, and the like, there are all relevant regulations, and the corresponding parameters are preset in the central control unit of the locomotive, so that the locomotive operation safety level is determined by the safety state evaluation model in this embodiment.

Specifically, the embodiment calculates the safe distance of the vehicle running in the current state from the obstacle size, the current running speed of the locomotive, the current line condition, the locomotive load, the dynamic obstacle following state and the like, and only considers the obstacle affecting the locomotive running in the locomotive running range by taking the rail as a reference. Small-size soft material barriers, the size range is within 0.5 m, common such as trackside paper shells, plastics and the like, accord with safety distance standards, and if the operation of a locomotive is not influenced, the safety distance is evaluated and verified to be negligible and belongs to the safety class 5; the oversized static obstacle or hard material obstacle accords with the safety distance standard, does not influence the normal operation of a locomotive planning curve, needs to have danger early warning prompt, needs to remind drivers and passengers of attention, judges whether to slow down running according to the safety distance model, respectively belongs to the safety 3 level or the safety 1 level and the safety 0 level, and can be converted into an artificial driving state if the obstacles such as a garage door and the like are in a garage; the dynamic obstacle is required to follow the movement state of the obstacle in real time even if the dynamic obstacle meets the safety distance standard, the dynamic obstacle comprises people, animals, vehicles and the like which are mistakenly put into a line, the dynamic obstacle is judged according to the safety distance model, emergency braking and whistle warning are adopted when necessary, and the dynamic obstacle belongs to the safety level 1 or the safety level 0 respectively.

The driver and passenger state model acquires the safety grades D, D epsilon (0,1,3,5) of the driver and passenger of the locomotive according to the 6A system data; namely, the safety level of locomotive drivers and passengers comprises a safety level 5, a safety level 3 and a safety level 0;

when a driver and a passenger determine that the driver has no sleepiness state or the duration of the driver and the passenger continuously watching the mobile phone does not exceed a time threshold according to the signal of the sleepiness early-warning system of the automobile driver, the safety level D=5 of the driver and the passenger is ensured, and the locomotive does not need to reprogram the running speed curve at the moment; specifically, the vehicle driver drowsiness warning system for determining the drowsiness state of the driver in the present embodiment is a prior art, and will not be described in detail here.

Specifically, the result of the safety state evaluation model is a combination of a locomotive speed guiding model, a locomotive fault guiding model, an environment safety guiding model and a driver and passenger state guiding model, and the combination is composed of 4 digits and expressed as ABCD, wherein A represents a locomotive speed guiding safety grade value, B represents a locomotive fault guiding safety grade value, C represents an environment safety guiding safety grade value, and D represents a driver and passenger state guiding safety grade value. Namely, when the safety state evaluation function module value is 5555, no potential safety hazard exists, and the locomotive can normally run; when any digit value is less than 5, a danger early warning prompt is sent; when the digital value of any bit is 0, the manual driving mode is immediately switched into. If necessary, emergency braking is initiated.

Preferably, the method for establishing the safe distance model comprises the following steps:

Specifically, the embodiment also extracts the characteristics of the characteristic road section by recording the stored automatic driving operation data of the high-incidence area or the line gradient change area of the safety accident, and the characteristics can be used for establishing the model modeling of the locomotive safety distance.

Specifically, the safety state evaluation model of the embodiment can also be used for assisting a manual driving process in an automatic driving process, the driving characteristics of drivers and passengers are extracted by recording the stored running characteristics of each line of the drivers and passengers, whether the drivers and passengers have a sleepy state or not is analyzed, abnormal driving states such as mobile phone operation and the like are played, and the characteristics of the drivers and passengers can be used for establishing a locomotive safety distance model.

Preferably, the environment safety guide model comprises an information acquisition function module, an information perception function module, an environment recognition function module and a target following function module,

the environment recognition function module acquires the obstacle state through a convolution algorithm to establish a safe distance model and acquire the locomotive environment safety level; the obstacle states include static obstacle and dynamic obstacle movement states; the static barrier is the surrounding environment of the line, the track line is used as a demarcation point, a model is built for the size and the relative distance of the barrier, and the safety level is evaluated by the safety state evaluation module in combination with locomotive data. After the dynamic obstacle needs to be followed by the object following function module in real time, a model is built according to the size and the relative distance of the obstacle, and the safety level is evaluated by combining locomotive operation data. Specifically, in this embodiment, the obstacle is identified by using data filtering, cluster analysis, convolution algorithm, and the like. When the target of interest is too small, the accuracy is affected by only extracting details according to the multi-layer cascade structure; instead, a proper amount of global features of the object of interest can be increased to achieve a better effect. Two different network structures are designed aiming at the problems, the traditional convolutional neural network is modified by adding global features, and then the detection and identification of dynamic obstacles in the unmanned automobile process are realized by connecting the two convolutional neural network structures in parallel.

Preferably, the convolution network of the convolution algorithm comprises a first convolution neural network model and a second convolution neural network model; the first neural network model and the second neural network model are connected in parallel and are connected with the full-connection layer to serve as an output layer;

the first neural network model is used for sampling the detailed characteristics of the environment image, the second neural network model is used for correcting the sampled structure of the first neural network model and sampling the global characteristics of the environment image.

Specifically, the method for identifying the obstacle in front of the running direction of the locomotive comprises the following steps: processing images of obstacles in front of the running direction of the locomotive through a convolutional neural network model;

the convolution network of the convolution algorithm comprises a first convolution neural network model and a second convolution neural network model; the first neural network model and the second neural network model are connected in parallel and are connected with the full-connection layer to serve as an output layer; the first neural network model is used for sampling the detailed characteristics of the environment image, the second neural network model is used for correcting the sampled structure of the first neural network model and sampling the global characteristics of the environment image.

Specifically, an image of an obstacle in front of the locomotive running direction is input in the input layer; sampling, in the first neural network model, a detail feature of an image of an obstacle forward of the locomotive running direction; and correcting the structure sampled by the first neural network model in the second neural network model, and sampling the global features of the images of the obstacles in front of the running direction of the locomotive.

One embodiment of the invention is as follows:

such as a first neural network model consisting of 4 convolutional layers and 3 pooling layers. The first neural network model is composed of 4 convolution layers and 3 pooling layers, in order to ensure enough sampling of details, when the step length of the convolution layers is 1, the convolution kernel of the first convolution layer is 8X8, the number of feature images extracted by the pooling operation of the first layer is 16, the convolution kernels of the second and third convolution layers are 3X3, the number of feature images extracted by the second and third pooling layers is 32, and the convolution kernel of the fourth convolution layer is 2X2.

The second neural network model is composed of a 3-layer convolution layer and a 1-layer pooling layer, and is used for sampling global features of an image. When the step length of the convolution layers is 1, the convolution kernel of the first convolution layer is 5x5, the number of feature images extracted by the first layer pooling operation is 16, the convolution kernels of the second convolution layer and the third convolution layer are 3x3, but only the pooling operation after primary convolution is reserved, so that the global features of the image can be reserved maximally, redundant data can be removed, the displacement of the image is unchanged, and the dimension of the image is reduced.

The total connection layers are 3 in the network structure, 2 total connection layers with 400 units are respectively arranged in the first neural network model and the second neural network model, the image features extracted by the first neural network model and the second neural network model are respectively fused, then the two parts are fused together by 1 total connection layer with 100 units, the first neural network model extracts local image features through the total connection layer 1, the second neural network model extracts all image features through the total connection layer 2, and the improved algorithm instantly fuses the local features and the global features together to obtain more accurate image recognition, so that a layer of 100 units of total connection is adopted, and finally the classification recognition is carried out by using a Softmax classification layer, namely an output layer. An accurate image of the obstacle is output.

Specifically, the network structure in this embodiment is formed by two layers of convolutional neural network models in parallel, and finally, the two layers of recognition results are integrated by the same full-connection layer. The first neural network model is mainly used for sampling the detail characteristics of the image, so that the network structure is not changed, and the second neural network model is mainly used for correcting the structure of the upper layer sampling and sampling the global characteristics of the image.

The target following function module carries out real-time following on the dynamic obstacle information; and tracking the obstacle in front of the running direction of the locomotive. The method mainly comprises the steps of selecting a maneuvering target model, determining following parameters and the like. The maneuvering target model is used for comparing the system principles and application occasions of a constant speed model, a constant acceleration model and a current statistical model, tracking the target in front of the vehicle by adopting a self-adaptive Kalman filtering algorithm based on the current maneuvering target model, and analyzing the influence of tracking system parameters on the algorithm.

Specifically, in the embodiment of the present invention, the information collecting function module, the information sensing function module, the environment identifying function module and the target following function module all use existing modules to implement the functions in the embodiment, so that detailed description is omitted.

The invention also discloses a locomotive auxiliary automatic driving system design method based on the safety guide, which comprises the following steps:

specifically, the locomotive auxiliary automatic driving system in the embodiment combines different application scenes to clip an automatic driving model on the premise of ensuring the operation safety of the locomotive, and realizes the locomotive auxiliary automatic driving model under different scenes by a method of adjusting related parameters. And according to different safety guide models, the safety grade is evaluated in combination with the current locomotive data to set a safety early warning grade, and the locomotive running curve is re-planned according to the safety early warning grade.

specifically, the line information in this embodiment includes line speed limit, station, ramp, curve, position, and score equality; the locomotive information comprises total locomotive weight, locomotive length, number of trucks, locomotive traction/braking characteristics, air braking characteristics, locomotive consist, wheel diameter and real-time locomotive status; the signal information comprises a signal lamp in front of the vehicle, the position of the signal lamp and the like.

S2: according to the locomotive route information, locomotive information and front signal state information; establishing a locomotive model;

specifically, according to relevant regulations such as traction rules, a locomotive model is built in combination with locomotive information, and relevant resistance is calculated, wherein the building of the locomotive model is a common technology in the field of rail locomotives, and the embodiment is not specifically described.

S3: identifying an obstacle in front of the running direction of the locomotive according to the safety guide model, and acquiring a running speed curve of the locomotive by combining a safety state evaluation functional module;

the safety guidance model includes: a locomotive speed safety guiding model, a driver and passenger state guiding model, a locomotive fault safety guiding model and an environment safety guiding model;

Specifically, the locomotive auxiliary automatic driving system is guided to the model according to the speed safety; the running speed of the locomotive is monitored in real time, once the running speed of the locomotive exceeds the specified speed, LKJ can immediately generate an emergency braking instruction, so that the aim of safety protection is achieved, and the device has the functions of outputting four control locomotive instructions, namely voice prompt, releasing locomotive traction, service braking and emergency braking;

establishing a driver and passenger state guide model according to the 6A system data; in this embodiment, the state of drivers and passengers is monitored by real-time analysis of the 6A system video data, so as to distinguish whether drowsiness or other illegal operation behaviors exist, such as mobile phone playing, illegal operation, etc. The anti-drowsy intercom function can continuously remind drivers and passengers with voice for 3 times (or associate no vigilance alarm), if the detection state is not improved after 3 minutes, the danger level is improved, and if necessary, emergency braking is started. Meanwhile, the related operation information of operators to the operation panel can be stored and recorded, and the later fault tracing is convenient.

And establishing a locomotive fault safety guide model according to the vehicle-mounted network fault data so as to analyze whether the current fault affects the automatic driving of the locomotive, and determining whether the operation curve needs to be re-planned, whether the traction/braking force needs to be re-designed and adjusted or not according to the fault influence degree. If the number of the isolated shafts is different, the calculated value of the traction/braking force is also different, and if the number of the isolated shafts is larger than certain data (different according to the vehicle type), an automatic driving function may not be realized, and a driver and passengers are required to be reminded to switch the locomotive model to a manual driving state. Specifically, in another embodiment of the present invention, the fault level of the locomotive is described by taking the HXD2 locomotive as an example, so as to realize that whether the auxiliary automatic driving system needs to be re-programmed according to the fault level is determined by the present invention, the fault code consists of 6 digits, wherein the first two digits represent the fault level, the middle two digits represent the part where the fault occurs, and the last two digits are specific fault codes.

And establishing an environment safety guiding model according to the environment data, wherein the environment safety guiding model is divided into two different states of dynamic operation and static parking. When in dynamic operation, mainly considering the environmental safety problem, checking whether an obstacle exists around a line, wherein the obstacle is static or dynamic, and whether the operation of a locomotive is influenced; whether there is a temporary intrusion into a dynamic obstacle (human, animal, vehicle) or the like. The static parking state mainly considers whether the vehicle has a sliding phenomenon or not. Dynamic obstacle detection requires tracking of laser radar data, including the position polar coordinates of the obstacle points, as well as the echo pulse width of the laser points, which is related to the color, texture and surface roughness of the obstacle. The echo pulse width is generally in the range of 0-500, and the pulse width values corresponding to different obstacles are greatly different, so that the method can be used for matching and correlation of the obstacles.

Specifically, in this embodiment, the locomotive speed safety guiding model, the driver and passenger status guiding model, and the locomotive fault safety guiding model are all established by using existing technologies, and only the established model is used to implement the establishment of the locomotive auxiliary automatic driving system in this embodiment, so a detailed description will not be given.

Preferably, when the locomotive is heavy-loaded, the method further comprises the step of establishing a heavy-loaded locomotive multi-particle model and a coupler buffer model according to the heavy-loaded locomotive parameters, the parameters of the vehicle and the buffer system before the step S3 so as to reduce longitudinal impulse of the combined locomotive and ensure safe and stable operation of the locomotive.

Specifically, when the locomotive is a heavy-duty locomotive, the heavy-duty locomotive has high control difficulty due to the characteristics of long marshalling and large load, a driver needs to pay close attention to the running state and the line condition of the marshalling locomotive during running, the traction force and the braking force are adjusted in time, a braking system of the heavy-duty locomotive mainly uses air braking, a braking signal is transmitted from the locomotive to the tail of the locomotive by using air pressure waves as media, and the transmission limit speed of the air pressure waves is 300m/s. With the traditional air braking, the time for transmitting the braking decompression signal from the head part of the marshalling locomotive to the tail part of the marshalling locomotive is at least about 8s, so that the braking time of the marshalling locomotive is inconsistent, and larger coupler force and longitudinal impulse are generated, which is the main reason for causing the break-off and derailment accidents of the heavy-duty locomotive. Therefore, a coupler force estimation module is added in the auxiliary driving system so as to reduce the longitudinal impact force of the combined locomotive and ensure the safe and stable operation of the locomotive.

The coupler force estimation module is mainly used for establishing a multi-particle model and a coupler buffer model of a heavy-load locomotive according to locomotive traction calculation regulations and combining parameters of a locomotive, a vehicle and a buffer system of the heavy-load combined locomotive, linearizing a coupler buffer model characteristic curve and iterating the coupler force of the computer locomotive during running by utilizing a Newmark algorithm.

Preferably, when the locomotive is on a downhill grade, emergency stops cannot be made in a short time and distance if an emergency is encountered, so the locomotive needs to be brought down to a lower speed before entering a large grade. When the locomotive needs to be decelerated, the electric braking force of the electric brake alone is insufficient to overcome the ramp resistance of the whole locomotive, and the air brake alone is limited by the performance of the brake and the thermal load of a brake shoe, so that the electric and pneumatic combined brake is generally adopted.

The air brake and the electric brake of the long downslope are reasonably matched, so that the impulse of the locomotive is reduced, the air charging time requirement of a locomotive pipe during re-braking is met, and the following points are needed to be noted

(1) Firstly, putting in electric braking force to be more than 400kN, and then adopting air braking;

(2) The inflation time after each air braking is larger than 169s, the interval between two air braking is more than 3km, and the air braking cannot be used for more than 6 times continuously;

(3) The air brake release speed cannot be lower than 35km/h;

(4) After relieving the air brake, the electric brake should remain for at least 10 seconds.

In order to ensure the safety of the locomotive running on a long downhill slope, strict requirements are imposed on the operation of the locomotive in the long downhill slope section, and circulating braking is generally adopted. The cyclic braking is air brake electric braking with a small amount of depressurization.

The driver status in the driver status guidance model includes a cautious driving type, a normal driving type, an expert driving type, and a reckless driving type.

Specifically, driving characteristics of different drivers and passengers are extracted through 6A system data, driving characteristic data of different drivers and passengers are classified into a cautious driving type, a common driving type, an expert driving type and a reckless driving type, and the data can be used for perfecting locomotive safe distance model modeling.

And the automatic driving system realizes the automatic driving of the locomotive according to the locomotive running speed curve.

According to the locomotive auxiliary automatic driving system software function based on the safety guide, on the premise of guaranteeing safety, the locomotive auxiliary automatic driving function under different scenes is achieved through a mode of establishing a model by combining different application scenes and through a mode of adjusting parameters.

Specifically, the working principle of the embodiment is that according to a safety state evaluation model, the safety state evaluation model comprises a locomotive speed guiding model, a locomotive fault guiding model, an environment safety guiding model and a driver and passenger state guiding model; and setting different safety early warning levels by combining the current locomotive operation safety level. And a certain decision is taken according to different safety early warning levels, and an operation curve is required to be re-planned if the curve is required to be re-planned. The safety state evaluates the front line running environment state detected by the multi-sensor module to give different danger levels; according to the 6A system, the driving state of drivers and passengers is obtained, whether the drivers and passengers have sleepiness or not is analyzed, and abnormal driving operation states such as mobile phones are played; and further, the state of the locomotive running environment is evaluated, whether the locomotive running behavior is influenced is judged, whether the locomotive running curve needs to be re-planned or not is determined, and whether an alarm strategy needs to be adopted or not is judged. Under the condition that the normal operation state of the locomotive is not affected, the operation curve of the locomotive is planned according to different operation scenes and a certain rule algorithm, so that the safe and stable operation of the locomotive is realized. . When the normal running state of the locomotive is affected, whether the running curve of the locomotive needs to be re-planned is determined according to the safety state evaluation value, and if necessary, a dangerous early warning is provided.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.

Claims

1. The locomotive auxiliary automatic driving system based on the safety guide is characterized in that the locomotive auxiliary automatic driving system determines the running safety level of the locomotive according to a safety state evaluation model built in a vehicle-mounted module so as to realize locomotive auxiliary automatic driving; comprising the following steps: the system comprises a vehicle-mounted module, a locomotive interface module, an interaction module and an environment safety module;

the vehicle-mounted module comprises an interface unit, a central control unit and a data storage unit;

the interface unit is connected with the locomotive interface module so as to realize data exchange between the vehicle-mounted module and the locomotive interface module through the interface unit; the method comprises the steps of obtaining LKJ data, TCMS data, driver and passenger state data, brake data and train tail data of a locomotive interface module through a locomotive interface unit;

the data storage module is connected with the interface unit to realize data exchange between the data storage module and the interface unit, and pre-stores the current running line information of the locomotive, including ground gradient and locomotive speed limit curve; the data storage module establishes a route characteristic database according to the automatic driving operation data of the high-incidence area of the safety accident or the ground slope change area;

The central control unit is connected with the interface unit to realize data exchange between the central control unit and the interface unit, a route characteristic database is established according to a high-incidence area of a safety accident or a ground gradient change area, characteristics in the route characteristic database are used as parameters in the safety state evaluation model to evaluate the safety state of the locomotive, the operation safety level of the locomotive is obtained, and the auxiliary automatic driving of the locomotive is realized according to the operation safety level of the locomotive;

the method for evaluating the safety state of the locomotive comprises the following steps:

The safety state evaluation model comprises a locomotive speed guiding model, a locomotive fault guiding model, an environment safety guiding model and a driver and passenger state guiding model; the driver states in the driver state guide model comprise a cautious driving type, a common driving type, an expert driving type and a reckless driving type;

the safety state evaluation model takes the minimum values of the locomotive speed safety level, the locomotive fault safety level, the locomotive environment safety level and the locomotive driver and passenger safety level;

the environment safety guide model comprises an information acquisition function module, an information perception function module, an environment recognition function module and a target following function module,

acquiring the safety distance of the current vehicle running of the locomotive according to the obstacle size, the running speed of the locomotive, the current line condition, the locomotive load, the dynamic obstacle movement state and the line characteristics of the road section or the line gradient change area of the high-incidence area of the safety accident;

the environment safety module comprises a plurality of cameras, millimeter wave radars and laser radars;

the camera is connected with the vehicle-mounted module to acquire an obstacle image in front of the running direction of the locomotive and transmit the obstacle image to the vehicle-mounted module;

2. A safety guidance-based locomotive auxiliary autopilot system in accordance with claim 1 wherein,

the method for acquiring the locomotive speed safety grade by the locomotive through the locomotive speed guiding model comprises the following steps:

the locomotive speed guiding model monitors the running speed of the locomotive in real time to obtain the safety level of the locomotive speed A，Aϵ (0,1,3,5): namely, the locomotive speed operation safety level comprises a safety level 5, a safety level 3, a safety level 1 and a safety level 0;

specifically, when the running speed of the locomotive does not exceed the speed safety threshold, the load of the locomotive does not exceed the load threshold, and the gradient value of the current running line is not greater than the gradient threshold, the speed safety level of the locomotiveA=5, at which time the locomotive does not need to reprogram the running speed curve;

when the speed of the locomotive does not exceed the speed safety threshold value, the load of the locomotive exceeds the load threshold value and the gradient value of the current running line is greater than the gradient threshold value, the speed safety level of the locomotiveA=3, at this time, the locomotive sends a dangerous early warning prompt to the driver and passengers, and the running speed curve does not need to be planned again;

when the speed of the locomotive does not exceed the speed safety threshold value and the difference between the speed of the locomotive running speed and the speed of the current position point on the locomotive running curve is less than the speed difference threshold value, the speed safety level of the locomotiveA=1, at this time, the locomotive adjusts the real-time running speed without re-planning the running speed curve;

the locomotive speed safety level when the locomotive speed exceeds a speed safety threshold or the difference between the locomotive operating speed and the speed of the current position point on the locomotive operating curve is less than a speed difference threshold A=0, the locomotive changes from automatic driving mode to manual driving mode within the countdown time threshold, and the running speed curve is re-planned.

3. The locomotive auxiliary automatic driving system based on safe guidance according to claim 2, wherein the method for acquiring the locomotive failure safety level by the locomotive failure guidance model is as follows: the locomotive fault guiding model analyzes whether the current fault affects the automatic driving of the locomotive according to the vehicle-mounted network fault model so as to acquire the locomotive fault safety levelB，Bϵ (0,1,3,5); determining whether a re-planning of the running speed curve is required; i.e. locomotive fail safe etcThe levels include level 5, level 3, level 1 and level 0;

in particular, when a locomotive is fault-free, the locomotive is at a fault-safe levelB=5, the locomotive does not need to re-plan the running speed curve at the moment;

the locomotive failsafe level when the locomotive isolation axis number is less than a first isolation axis number thresholdB=3, the locomotive sends a dangerous early warning prompt to drivers and passengers at the moment, and the running speed curve does not need to be planned again;

the locomotive failsafe level when the locomotive isolation shaft number is greater than the first isolation shaft number threshold and less than the second isolation shaft number threshold B=1, at the moment, the locomotive adjusts the real-time running speed without re-planning the running speed curve;

the locomotive failsafe level when the locomotive isolation axis number is greater than a second isolation axis number thresholdB=0, the locomotive is converted from an automatic driving mode to an artificial driving mode, and the running speed curve is planned again.

4. The locomotive auxiliary automatic driving system based on safety guidance according to claim 2, wherein the method for acquiring the locomotive environmental safety grade by the locomotive through the environmental safety guidance model is as follows:

the environmental safety guide model obtains the environmental safety level of the locomotive according to the size of the obstacle and the distance between the obstacle and the locomotive, which are acquired by the environmental safety moduleC，Cϵ (0,1,3,5); namely, the locomotive environment safety level comprises a safety level 5, a safety level 3, a safety level 1 and a safety level 0;

specifically, when there is no obstacle in front of the running direction of the locomotive, or the size of the obstacle is smaller than a size threshold, the hardness of the obstacle is smaller than a hardness threshold, and the distance between the obstacle and the locomotive is larger than a distance threshold, the locomotive environment safety level is setC=5, the locomotive does not need to re-plan the running speed curve at the moment;

when an obstacle with a size larger than a size threshold and a hardness smaller than a hardness threshold is arranged in front of the running direction of the locomotive, the locomotive environmental safety grade C=3, thisThe timing vehicle sends a danger early warning prompt to drivers and passengers, and the running speed curve does not need to be planned again;

when there is a dynamic obstacle with a size greater than a size threshold in front of the running direction of the locomotive, the locomotive environmental safety levelC=1, at the moment, the locomotive adjusts the real-time running speed without re-planning the running speed curve;

the locomotive environmental safety level when a distance between an obstacle in front of a locomotive running direction and the locomotive is less than a distance thresholdC=0, the locomotive is converted from an automatic driving mode to an artificial driving mode, and the running speed curve is planned again.

5. The locomotive auxiliary automatic driving system based on safety guidance according to claim 2, wherein the method for acquiring the safety level of the locomotive driver through the driver state model comprises the following steps: the driver and passenger state model acquires the safety level of the driver and passenger of the locomotive according to the 6A system dataD，Dϵ (0,1,3,5); namely, the safety level of locomotive drivers and passengers comprises a safety level 5, a safety level 3 and a safety level 0;

specifically, when the driver and the passenger determine that the driver has no sleepiness state or the duration of the driver and the passenger continuously watching the mobile phone does not exceed the time threshold according to the signal of the sleepiness early-warning system of the automobile driver, the safety level of the driver and the passenger is ensured D=5, the locomotive does not need to re-plan the running speed curve at the moment;

when the driver and the passenger determine that the driver is in a doze state or the duration of the driver and the passenger continuously watching the mobile phone exceeds a time threshold according to the alarm signal provided by the doze early-warning system of the automobile driver, the safety level of the driver and the passenger is ensuredD=3, the locomotive sends a dangerous early warning prompt to drivers and passengers at the moment, and the running speed curve does not need to be planned again;

when drivers and passengers determine that the drivers are in a doze state according to the alarm signals provided by the doze early-warning system of the automobile driver, the drivers and passengers are still in the doze state after being warned and reminded for a plurality of times, and the safety level of the drivers and passengers is highD=And 0, the locomotive performs emergency braking operation, and simultaneously, the running speed curve is re-planned.

6. The method for designing a locomotive auxiliary automatic driving system based on safe guidance according to any one of claims 1 to 5, comprising the steps of:

7. The method for designing a safety guidance-based automatic driving system for a locomotive according to claim 6, wherein when the locomotive is heavily loaded, the step S3 further comprises: and establishing a multi-particle model and a coupler buffer model of the heavy-duty locomotive according to the parameters of the heavy-duty locomotive, the parameters of the vehicle and the parameters of the buffer system so as to reduce the longitudinal impact force of the combined locomotive.