CN110147109A - A kind of archetype development system of automated driving system - Google Patents
A kind of archetype development system of automated driving system Download PDFInfo
- Publication number
- CN110147109A CN110147109A CN201910512711.XA CN201910512711A CN110147109A CN 110147109 A CN110147109 A CN 110147109A CN 201910512711 A CN201910512711 A CN 201910512711A CN 110147109 A CN110147109 A CN 110147109A
- Authority
- CN
- China
- Prior art keywords
- sensing
- module
- fusion
- lane line
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000004927 fusion Effects 0.000 claims abstract description 235
- 230000001427 coherent effect Effects 0.000 claims description 24
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 8
- 239000000203 mixture Substances 0.000 claims description 7
- 230000008447 perception Effects 0.000 claims description 2
- 230000005641 tunneling Effects 0.000 description 11
- 230000000694 effects Effects 0.000 description 8
- 230000004888 barrier function Effects 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000000465 moulding Methods 0.000 description 3
- 238000002604 ultrasonography Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 239000000155 melt Substances 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000000034 method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/0088—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0255—Control of position or course in two dimensions specially adapted to land vehicles using acoustic signals, e.g. ultra-sonic singals
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Game Theory and Decision Science (AREA)
- Medical Informatics (AREA)
- Electromagnetism (AREA)
- Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of archetype development systems of automated driving system, including rapid prototyping controller, camera, millimetre-wave radar, the first sensing Fusion Module, m sensing module and the n sensing Fusion Module merged respectively to the lane line that n sensing group detects with obstacle information, control algolithm module, signal processing module, the first sensing fusion parsing module and n sensing fusion parsing module, n sensing group is integrated in rapid prototyping controller to be combined and constituted by camera, millimetre-wave radar and 1~m sensing module;N sensing fusion parsing module receives the fuse information that n sensing Fusion Module is sent respectively and is parsed, and exports target level lane line and obstacle signal, signal processing module are selected and be delivered to control algolithm module after handling.Function for Automatic Pilot is able to achieve using the present invention flexibly rapidly to develop.
Description
Technical field
The invention belongs to automatic Pilot fields, and in particular to a kind of archetype development system of automated driving system.
Background technique
L3 grades of automatic Pilots are during prototyping early period, ADAS (Advanced Driving Assistant
System) control algolithm is usually all by the mode development of modularization, hardware and software platform, and the variation of external signal does not influence as far as possible
Control module.
The archetype development system of existing automated driving system, including rapid prototyping controller, camera, millimeter wave
Radar and first senses Fusion Module, and control algolithm module, signal processing module, the first biography are integrated in rapid prototyping controller
Sense fusion parsing module and vehicle CAN parsing module, signal processing module, the first sensing fusion parsing module and vehicle CAN solution
Analyse module composition control interface module, the lane line and barrier that the first sensing Fusion Module detects camera, millimetre-wave radar
Object information is hindered to be merged, the first sensing fusion parsing module receives the fuse information and progress that the first sensing Fusion Module is sent
Parsing, target level lane line and obstacle signal are to signal processing module all the way for output, and signal processing module is to target level lane
Line and obstacle signal are delivered to control algolithm module after being handled, vehicle CAN parsing module receives vehicle CAN signal and goes forward side by side
Row parsing exports vehicle coherent signal to signal processing module, and signal processing module is defeated after handling vehicle coherent signal
It send to control algolithm module;Control algolithm module is carried out according to target level lane line and obstacle signal and vehicle coherent signal
Automatic Pilot decision rule and control.
Because there is very big uncertainty in the type framework of the sensing modules such as sensor, high-precision map on vehicle,
It is put down in the switching of each aware platform, such as from the perception being made of camera, millimetre-wave radar and the first sensing Fusion Module
Platform is switched to by camera, millimetre-wave radar, laser radar and the second sensing Fusion Module (for camera, millimeter wave thunder
Reach, the lane line that laser radar separately detects is merged with obstacle information) composition aware platform when, need to control
The first sensing fusion parsing module in interface module is changed to the second sensing fusion solution corresponding with the second sensing Fusion Module
When analysing module, or switching to other aware platforms again, it is still desirable to be changed to the first sensing fusion parsing module corresponding
Other sensing fusion parsing modules, it is troublesome in poeration, especially when comparing the control effect of different aware platform inputs, need frequent
Increase sensing Fusion Module, frequently change sensing fusion parsing module, frequent way of compiling control model is time-consuming very long.
Summary of the invention
The object of the present invention is to provide a kind of archetype development systems of automated driving system, in automated driving system
The archetype development phase realize Function for Automatic Pilot flexibly rapidly develop.
The archetype development system of automated driving system of the present invention, including rapid prototyping controller, camera,
Millimetre-wave radar and the first sensing Fusion Module, be integrated in rapid prototyping controller control algolithm module, signal processing module,
First sensing fusion parsing module and vehicle CAN parsing module, the first sensing Fusion Module detect camera, millimetre-wave radar
To lane line merged with obstacle information, first sensing fusion parsing module receive first sensing Fusion Module send
Fuse information is simultaneously parsed, and to signal processing module, vehicle CAN is parsed for target level lane line and obstacle signal all the way for output
Module receives vehicle CAN signal and is parsed, and exports vehicle coherent signal to signal processing module, signal processing module is to whole
Vehicle coherent signal is delivered to control algolithm module after being handled;The archetype development system further includes for detecting lane
M sensing module of line and obstacle information and the lane line that n sensing group detects is carried out respectively with obstacle information
N sensing Fusion Module of fusion, the n sensing group is by camera, millimetre-wave radar and 1~m the sensing module group
It closes and constitutes;It is also integrated in the rapid prototyping controller and is merged with the one-to-one n sensing of the n sensing Fusion Module
Parsing module, n sensing fusion parsing module receive the fuse information that n sensing Fusion Module is sent respectively and are parsed,
The road n target level lane line and obstacle signal are exported to signal processing module, signal processing module merges parsing module to sensing
The timestamp in corresponding channel carries out overtime detection, selects at the target level lane line and obstacle signal in some channel
Reason, is delivered to control algolithm module, control algolithm module is according to target level lane line and obstacle signal and vehicle after processing
Coherent signal carries out automatic Pilot decision rule and control;Wherein, m is integer, and m >=1, n are integer, and It indicates to take out the number of combinations that i sensing module is combined from m sensing module.
Preferably, the value that the value of the m is 1, n is 1;1 sensing module is laser radar or ultrasound
Wave radar or high-precision map, 1 sensing group (can be with by camera, millimetre-wave radar and 1 sensing module
It is laser radar, is also possible to ultrasonic radar, can also be high-precision map) it constitutes, 1 sensing Fusion Module is
Second sensing Fusion Module, 1 sensing fusion parsing module correspond to the second sensing fusion parsing module.Wherein, believe
Number processing module carries out overtime detection to the timestamp in the corresponding channel of sensing fusion parsing module, selects the target in some channel
The concrete mode that grade lane line and obstacle signal are handled are as follows: signal processing module merges parsing module according to the second sensing
The timestamp in corresponding channel judges whether the break period in the channel is less than the time threshold a of setting, if it is, selecting the
The target level lane line and obstacle signal in the corresponding channel of two sensing fusion parsing modules are handled and (select the second sensing
The lane line of target level all the way and obstacle signal of fusion parsing module output are handled), otherwise select the first sensing fusion
The target level lane line and obstacle signal in the corresponding channel of parsing module, which are handled, (selects the first sensing fusion parsing mould
The lane line of target level all the way and obstacle signal of block output are handled).
Preferably, the value that the value of the m is 2, n is 3;2 sensing modules are respectively laser radar and surpass
Sound radar perhaps laser radar and high-precision map or ultrasonic radar and high-precision map;3 sensing groups
In first sensing group is by camera, millimetre-wave radar and wherein 1 sensing module is constituted, second sensing group is by imaging
Head, millimetre-wave radar and another 1 sensing module are constituted, and third sensing group is by camera, millimetre-wave radar and 2 sensing modules
It constitutes, 3 sensing Fusion Modules are respectively the lane line detected to first sensing group and obstacle information progress
The third that the second of fusion senses Fusion Module, merged to the lane line that second sensing group detects with obstacle information
Sensing Fusion Module, the 4th sensing fusion mould that the lane line that third sensing group detects is merged with obstacle information
Block, 3 sensings fusion parsing module respectively correspond as the second sensing fusion parsing module, third sensing fusion parsing mould
Block, the 4th sensing fusion parsing module.Wherein, timestamp of the signal processing module to the corresponding channel of sensing fusion parsing module
Carry out overtime detection, the concrete mode for selecting the target level lane line in some channel and obstacle signal to be handled are as follows: signal
Processing module judges whether the break period in the channel is small according to the timestamp in the corresponding channel of the 4th sensing fusion parsing module
In the time threshold a of setting, if it is, select the corresponding channel of the 4th sensing fusion parsing module target level lane line and
Obstacle signal, which is handled, (selects the lane line of target level all the way and barrier letter of the 4th sensing fusion parsing module output
Number handled), when otherwise judging the interruption in the channel according to the timestamp that third senses the corresponding channel of fusion parsing module
Between whether be less than the time threshold a of setting, if it is, selecting the target level in the corresponding channel of third sensing fusion parsing module
Lane line and obstacle signal handled (i.e. selection third sensing fusion parsing module output the lane line of target level all the way and
Obstacle signal is handled), the channel is otherwise judged according to the timestamp in the corresponding channel of the second sensing fusion parsing module
Break period whether be less than the time threshold a of setting, if it is, selecting the corresponding channel of the second sensing fusion parsing module
Target level lane line and obstacle signal handled (i.e. select second sensing fusion parsing module output target level all the way
Lane line and obstacle signal are handled), otherwise select the target level vehicle in the corresponding channel of the first sensing fusion parsing module
Diatom and obstacle signal, which are handled, (selects the lane line of target level all the way and barrier of the first sensing fusion parsing module output
Object signal is hindered to be handled).
Preferably, the value that the value of the m is 3, n is 2, and 3 sensing modules are respectively laser radar, ultrasound
Wave radar and high-precision map, first sensing group in 2 sensing groups is by camera, millimetre-wave radar and wherein 1
A sensing module is constituted, and second sensing group is made of camera, millimetre-wave radar and 3 sensing modules, 2 sensings
Fusion Module is respectively that the second sensing merged to the lane line that first sensing group detects with obstacle information merges
Module, the third merged to the lane line that second sensing group detects with obstacle information sense Fusion Module, described
2 sensings fusion parsing module respectively correspond for second sensing fusion parsing module, third sensing fusion parsing module.Its
In, signal processing module carries out overtime detection to the timestamp in the corresponding channel of sensing fusion parsing module, selects some channel
Target level lane line and the concrete mode that is handled of obstacle signal are as follows: signal processing module senses fusion solution according to third
The timestamp in the corresponding channel of analysis module judges whether the break period in the channel is less than the time threshold a of setting, if it is,
It selects the target level lane line in the corresponding channel of third sensing fusion parsing module and obstacle signal to be handled and (selects the
The lane line of target level all the way and obstacle signal of three sensing fusion parsing module outputs are handled), otherwise passed according to second
The timestamp in the corresponding channel of sense fusion parsing module judges whether the break period in the channel is less than the time threshold a of setting,
If it is, the target level lane line in the corresponding channel of the second sensing fusion parsing module and obstacle signal is selected to be handled
(lane line of target level all the way of the second sensing fusion parsing module output and obstacle signal is selected to be handled), otherwise selects
It is handled with the target level lane line in the corresponding channel of the first sensing fusion parsing module and obstacle signal and (is selected first
The lane line of target level all the way and obstacle signal of sensing fusion parsing module output are handled).
There are the present invention multiple sensings to merge parsing module, can configure and develop demand certainly according to the sensing module of vehicle
It is dynamic to switch to required sensing fusion parsing module, the automatic target for selecting the required corresponding channel of sensing fusion parsing module
Grade lane line and obstacle signal are handled, and realize automatic Pilot function in the archetype development phase of automated driving system
It flexibly can rapidly develop, improve development efficiency.
Detailed description of the invention
Fig. 1 is the functional block diagram of embodiment 1.
Fig. 2 is that signal processing module selects the target level lane line in some channel and obstacle signal to carry out in embodiment 1
The logical flow chart of processing.
Fig. 3 is the functional block diagram of embodiment 2.
Fig. 4 is that signal processing module selects the target level lane line in some channel and obstacle signal to carry out in embodiment 2
The logical flow chart of processing.
Fig. 5 is the functional block diagram of embodiment 3.
Fig. 6 is that signal processing module selects the target level lane line in some channel and obstacle signal to carry out in embodiment 3
The logical flow chart of processing.
Specific embodiment
It elaborates with reference to the accompanying drawing to the present invention.
Embodiment 1: the archetype development system of automated driving system as shown in Figure 1, including rapid prototyping controller
1, camera 2, millimetre-wave radar 3, laser radar 5, first sense Fusion Module 4 and the second sensing Fusion Module 6, rapid prototyping
Control algolithm module 10 is integrated in controller 1, the sensing fusion parsing module 12, second of signal processing module 11, first senses
Merge parsing module 14 and vehicle CAN parsing module 13, the sensing fusion parsing module 12, second of signal processing module 11, first
Sensing fusion parsing module 14 and 13 composition control interface module of vehicle CAN parsing module;First 4 pairs of Fusion Module of sensing camera shooting
First 2, the lane line that millimetre-wave radar 3 separately detects is merged with obstacle information, the first sensing fusion parsing module 12
It receives the fuse information that the first sensing Fusion Module 4 is sent and is parsed, target level lane line and barrier are believed all the way for output
Number to signal processing module 11, the second sensing Fusion Module 6 separately detects camera 2, millimetre-wave radar 3 and laser radar 5
To lane line merged with obstacle information, second sensing fusion parsing module 14 receive second sensing Fusion Module 6 send out
The fuse information that send simultaneously is parsed, output all the way target level lane line and obstacle signal to signal processing module 11.
As shown in Fig. 2, timestamp of the signal processing module 11 according to the second sensing fusion corresponding channel of parsing module 14
Judge whether the break period in the channel is less than the time threshold a of setting, if it is, indicating that the tunneling traffic is normal, selects
The lane line of target level all the way and obstacle signal that second sensing fusion parsing module 14 exports are handled, and otherwise indicate second
The sensing fusion corresponding tunneling traffic of parsing module 14 is interrupted, and then selects the first sensing fusion parsing module 12 output all the way
Target level lane line and obstacle signal are handled;Vehicle CAN parsing module 13 receives vehicle CAN signal and is parsed,
Vehicle coherent signal is exported to signal processing module 11, signal processing module 11 is delivered to after handling vehicle coherent signal
Control algolithm module 10;Control algolithm module 10 is carried out according to target level lane line and obstacle signal and vehicle coherent signal
Automatic Pilot decision rule and control.
If vehicle is only configured with camera 2, millimetre-wave radar 3, when the first sensing Fusion Module 4 works normally, close
Closing the second sensing Fusion Module 6 can be realized the output of the selection of signal processing module 11 first sensing fusion parsing module 12 all the way
Target level lane line and obstacle signal handled (because second sensing Fusion Module 6 be closed after, corresponding channel
Break period can be greater than setting time threshold a), control algolithm module 10 according to the target level lane line and obstacle signal with
And vehicle coherent signal carries out automatic Pilot decision rule and control;If vehicle configuration camera 2,3 and of millimetre-wave radar
Laser radar 5, then starting the second sensing Fusion Module 6 can be realized the selection of signal processing module 11 second sensing fusion parsing mould
The lane line of target level all the way and obstacle signal that block 14 exports are handled (because the second sensing Fusion Module 6 works normally
When, the break period in corresponding channel is less than the time threshold a) of setting, and control algolithm module 10 is according to the target level lane
Line and obstacle signal and vehicle coherent signal carry out automatic Pilot decision rule and control.In this embodiment 1, if wanted
By camera 2, millimetre-wave radar 3 as a sensing group and camera 2, millimetre-wave radar 3 and laser radar 5 as a biography
Sense group, when comparing the control effect of different sensing group inputs, it is only necessary to make the first sensing Fusion Module 4 keep working condition, lead to
Crossing the sensing Fusion Module 6 of on/off second can be realized the control effect comparison of two sensing groups inputs.This archetype
Development system can promote development efficiency.
Embodiment 2: the archetype development system of automated driving system as shown in Figure 3, including rapid prototyping controller
1, camera 2, millimetre-wave radar 3, laser radar 5, ultrasonic radar 7, first sense the sensing fusion mould of Fusion Module 4, second
Block 6, third sensing Fusion Module the 8, the 4th sense Fusion Module 9, are integrated with control algolithm module in rapid prototyping controller 1
10, the sensing of signal processing module 11, first fusion parsing module 12, second sensing fusion parsing module 14, third sensing fusion
The sensing of parsing module the 15, the 4th fusion parsing module 16 and vehicle CAN parsing module 13, signal processing module 11, first sense
Merge the sensing fusion of parsing module 12, second parsing module 14, the third sensing fusion sensing fusion parsing of parsing module the 15, the 4th
13 composition control interface module of module 16 and vehicle CAN parsing module;First sensing Fusion Module 4 is to camera 2, millimeter wave thunder
It is merged up to the lane line that 3 separately detect with obstacle information, the first sensing fusion parsing module 12 receives the first sensing
The fuse information of the transmission of Fusion Module 4 is simultaneously parsed, and exports target level lane line all the way and obstacle signal to signal processing
Module 11, the lane line that the second sensing Fusion Module 6 separately detects camera 2, millimetre-wave radar 3 and laser radar 5 with
Obstacle information is merged, and the second sensing fusion parsing module 14 receives the fuse information that the second sensing Fusion Module 6 is sent
And parsed, to signal processing module 11, third senses Fusion Module 8 for target level lane line and obstacle signal all the way for output
The lane line separately detected to camera 2, millimetre-wave radar 3 and ultrasonic radar 7 is merged with obstacle information, third
It senses the fuse information that the fusion reception third sensing Fusion Module 8 of parsing module 15 is sent simultaneously to be parsed, exports target all the way
Grade lane line and obstacle signal to signal processing module 11, the 4th sensing Fusion Module 9 to camera 2, millimetre-wave radar 3,
The lane line that laser radar 5 and ultrasonic radar 7 separately detect is merged with obstacle information, the 4th sensing fusion parsing
Module 16 receives the fuse information that the 4th sensing Fusion Module 9 is sent and is parsed, and exports target level lane line and barrier all the way
Hinder object signal to signal processing module 11.
As shown in figure 4, timestamp of the signal processing module 11 according to the 4th sensing fusion corresponding channel of parsing module 16
Judge whether the break period in the channel is less than the time threshold a of setting, if it is, indicating that the tunneling traffic is normal, selects
The lane line of target level all the way and obstacle signal that 4th sensing fusion parsing module 16 exports are handled, and otherwise indicate the 4th
The sensing fusion corresponding tunneling traffic of parsing module 16 is interrupted, and senses the fusion corresponding channel of parsing module 15 further according to third
Timestamp judges whether the break period in the channel is less than the time threshold a of setting, if it is, indicating the tunneling traffic just
Often, the lane line of target level all the way and obstacle signal for selecting third sensing fusion parsing module 15 to export are handled, otherwise
Indicate that the third sensing fusion corresponding tunneling traffic of parsing module 15 is interrupted, it is corresponding further according to the second sensing fusion parsing module 14
The timestamp in channel judge whether the break period in the channel is less than the time threshold a of setting, if it is, indicating the channel
Communication is normal, at the lane line of target level all the way and obstacle signal for selecting the second sensing fusion parsing module 14 to export
Otherwise reason indicates that the second sensing fusion corresponding tunneling traffic of parsing module 14 is interrupted, then selects the first sensing fusion parsing
The lane line of target level all the way and obstacle signal that module 12 exports are handled;Vehicle CAN parsing module 13 receives vehicle CAN
Signal is simultaneously parsed, and exports vehicle coherent signal to signal processing module 11, signal processing module 11 is to vehicle coherent signal
Control algolithm module 10 is delivered to after being handled;Control algolithm module 10 according to target level lane line and obstacle signal and
Vehicle coherent signal carries out automatic Pilot decision rule and control.
If vehicle is only configured with camera 2, millimetre-wave radar 3, when the first sensing Fusion Module 4 works normally, close
Close the second sensing Fusion Module 6, third sensing Fusion Module the 8, the 4th senses Fusion Module 9 and signal processing module 11 can be realized
The lane line of target level all the way and obstacle signal for selecting the first sensing fusion parsing module 12 to export are handled (because the
Two, after third, the 4th sensing Fusion Module are closed, the break period in corresponding channel can be greater than the time threshold of setting
A), control algolithm module 10 is determined according to the target level lane line and obstacle signal and vehicle coherent signal progress automatic Pilot
Plan planning and control;If vehicle configuration camera 2, millimetre-wave radar 3 and laser radar 5, Fusion Module 6 is sensed second
When normal work, closing the third sensing sensing Fusion Module 9 of Fusion Module the 8, the 4th can be realized the selection of signal processing module 11
The lane line of target level all the way and obstacle signal that second sensing fusion parsing module 14 exports are handled (because of third, the
After four sensing Fusion Modules are closed, the break period in corresponding channel can be greater than the time threshold a) of setting, control algolithm
Module 10 carries out automatic Pilot decision rule and control according to the target level lane line and obstacle signal and vehicle coherent signal
System;If vehicle configuration camera 2, millimetre-wave radar 3 and ultrasonic radar 7, worked normally in third sensing Fusion Module 8
When, the 4th sensing Fusion Module 9 of closing can be realized signal processing module 11 and select the third sensing fusion output of parsing module 15
The lane line of target level all the way and obstacle signal handled (because the 4th sensing Fusion Module be closed after, it is corresponding
The break period in channel can be greater than the time threshold a) of setting, and control algolithm module 10 is according to the target level lane line and barrier
Signal and vehicle coherent signal carry out automatic Pilot decision rule and control;If vehicle configuration camera 2, millimeter wave thunder
Up to 3, laser radar 5 and ultrasonic radar 7, then starting the 4th sensing Fusion Module 9 can be realized the selection of signal processing module 11
The lane line of target level all the way and obstacle signal that 4th sensing fusion parsing module 16 exports are handled (because of the 4th sensing
When Fusion Module 9 works normally, the break period in corresponding channel is less than the time threshold a) of setting, control algolithm module 10
Automatic Pilot decision rule and control are carried out according to the target level lane line and obstacle signal and vehicle coherent signal.Herein
In embodiment 1, if regarding camera 2, millimetre-wave radar 3 as a sensing group, by camera 2, millimetre-wave radar 3 and swash
Optical radar 5 is used as a sensing group, regard camera 2, millimetre-wave radar 3 and ultrasonic radar 7 as a sensing group, will image
First 2, millimetre-wave radar 3, laser radar 5 and ultrasonic radar 7 are used as a sensing group, compare the control of different sensing group inputs
When effect, it is only necessary to make the first sensing Fusion Module 4 keep working condition, by/off second, third, the 4th sensing
The control effect comparison of four sensing group inputs can be realized in Fusion Module.This archetype development system can promote exploitation effect
Rate.
Embodiment 3: the archetype development system of automated driving system as shown in Figure 5, including rapid prototyping controller
1, camera 2, millimetre-wave radar 3, laser radar 5, ultrasonic radar 7, accurately Figure 20, the first sensing Fusion Module 4, the
Two sensing Fusion Modules 6 and third sense Fusion Module 8, are integrated with control algolithm module 10, signal in rapid prototyping controller 1
The sensing fusion sensing fusion of parsing module 12, second of processing module 11, first parsing module 14, third sensing fusion parsing module
15 and vehicle CAN parsing module 13, the sensing fusion sensing fusion parsing of parsing module 12, second of signal processing module 11, first
Module 14, third sensing fusion parsing module 15 and 13 composition control interface module of vehicle CAN parsing module;First sensing fusion
The lane line that module 4 separately detects camera 2, millimetre-wave radar 3 is merged with obstacle information, the first sensing fusion
Parsing module 12 receives the fuse information that the first sensing Fusion Module 4 is sent and is parsed, and exports target level lane line all the way
And obstacle signal, to signal processing module 11, the second sensing Fusion Module 6 is to camera 2, millimetre-wave radar 3 and laser radar
5 lane lines separately detected are merged with obstacle information, and the second sensing fusion parsing module 14 receives the second sensing and melts
The fuse information of the molding transmission of block 6 is simultaneously parsed, and exports target level lane line all the way and obstacle signal to signal processing mould
Block 11, third sense Fusion Module 8 to camera 2, millimetre-wave radar 3, laser radar 5, ultrasonic radar 7 and high-precision map
20 lane lines separately detected are merged with obstacle information, and third sensing fusion parsing module 15 receives third sensing and melts
The fuse information of the molding transmission of block 8 is simultaneously parsed, and exports target level lane line all the way and obstacle signal to signal processing mould
Block 11.
As shown in fig. 6, signal processing module 11 senses the timestamp in the fusion corresponding channel of parsing module 15 according to third
Judge whether the break period in the channel is less than the time threshold a of setting, if it is, indicating that the tunneling traffic is normal, selects
The lane line of target level all the way and obstacle signal that third sensing fusion parsing module 15 exports are handled, and otherwise indicate third
The sensing fusion corresponding tunneling traffic of parsing module 15 is interrupted, further according to the second sensing fusion corresponding channel of parsing module 14
Timestamp judges whether the break period in the channel is less than the time threshold a of setting, if it is, indicating the tunneling traffic just
Often, the lane line of target level all the way and obstacle signal for selecting the second sensing fusion parsing module 14 to export are handled, otherwise
It indicates that the second sensing fusion corresponding tunneling traffic of parsing module 14 is interrupted, then selects the first sensing fusion parsing module 12 defeated
The lane line of target level all the way and obstacle signal out is handled;Vehicle CAN parsing module 13 receives vehicle CAN signal and goes forward side by side
Row parsing exports vehicle coherent signal to signal processing module 11, and signal processing module 11 handles vehicle coherent signal
After be delivered to control algolithm module 10;Control algolithm module 10 is related according to target level lane line and obstacle signal and vehicle
Signal carries out automatic Pilot decision rule and control.
If vehicle is only configured with camera 2, millimetre-wave radar 3, when the first sensing Fusion Module 4 works normally, close
Close the second sensing Fusion Module 6, third sensing Fusion Module 8 can be realized signal processing module 11 and select the first sensing fusion solution
The lane line of target level all the way and obstacle signal that analysis module 12 exports are handled (because of second, third sensing Fusion Module
After being closed, the break period in corresponding channel can be greater than the time threshold a) of setting, and control algolithm module 10 is according to the mesh
It marks grade lane line and obstacle signal and vehicle coherent signal carries out automatic Pilot decision rule and control;If vehicle configuration
Camera 2, millimetre-wave radar 3 and laser radar 5 are closed third sensing and are melted when the second sensing Fusion Module 6 works normally
The target level lane line all the way that signal processing module 11 selects the second sensing fusion parsing module 14 to export can be realized in molding block 8
And obstacle signal is handled that (because after third sensing Fusion Module is closed, break period in corresponding channel can be big
In the time threshold a) of setting, control algolithm module 10 is believed according to the target level lane line and obstacle signal and vehicle correlation
Number carry out automatic Pilot decision rule and control;If vehicle configuration camera 2, millimetre-wave radar 3, laser radar 5, ultrasound
Wave radar 7 and accurately Figure 20, then starting third sensing Fusion Module 8 can be realized the selection third biography of signal processing module 11
The lane line of target level all the way and obstacle signal that sense fusion parsing module 15 exports are handled (because of third sensing fusion mould
When block 8 works normally, the break period in corresponding channel is less than the time threshold a) of setting, and control algolithm module 10 is according to this
Target level lane line and obstacle signal and vehicle coherent signal carry out automatic Pilot decision rule and control.In this embodiment
In 1, if camera 2, millimetre-wave radar 3 are regard as a sensing group, by camera 2, millimetre-wave radar 3 and laser radar
5 be used as a sensing groups, using camera 2, millimetre-wave radar 3, laser radar 5, ultrasonic radar 7 and accurately Figure 20 as
One sensing group, when comparing the control effect of different sensing group inputs, it is only necessary to the first sensing Fusion Module 4 be made to keep work shape
State, by/off, second, third, which senses Fusion Module, can be realized the control effect comparison of three sensing groups inputs.This
Kind archetype development system can promote development efficiency.
Claims (7)
1. a kind of archetype development system of automated driving system, including rapid prototyping controller (1), camera (2), millimeter
Wave radar (3) and the first sensing Fusion Module (4), rapid prototyping controller (1) is interior to be integrated with control algolithm module (10), signal
Processing module (11), the first sensing fusion parsing module (12) and vehicle CAN parsing module (13), the first sensing Fusion Module
(4) lane line detected to camera (2), millimetre-wave radar (3) is merged with obstacle information, the first sensing fusion solution
Analysis module (12) receives the fuse information that the first sensing Fusion Module (4) is sent and is parsed, and exports target level lane all the way
To signal processing module (11), vehicle CAN parsing module (13) receives vehicle CAN signal and is solved for line and obstacle signal
Analysis exports vehicle coherent signal to signal processing module (11), and signal processing module (11) handles vehicle coherent signal
After be delivered to control algolithm module (10);It is characterized by also including the m perception for detecting lane line and obstacle information
Module and the n sensing Fusion Module that the lane line that n sensing group detects is merged respectively with obstacle information, institute
It states n sensing group and composition is combined by camera (2), millimetre-wave radar (3) and 1~m sensing module;The rapid prototyping
It is also integrated in controller (1) and merges parsing module, n sensing with the one-to-one n sensing of the n sensing Fusion Module
Fusion parsing module receives the fuse information that n sensing Fusion Module is sent respectively and is parsed, and exports the road n target level lane
Line and obstacle signal are to signal processing module (11), and signal processing module (11) is to the corresponding channel of sensing fusion parsing module
Timestamp carry out overtime detection, select the target level lane line in some channel and obstacle signal to be handled, it is defeated after processing
It send to control algolithm module (10), control algolithm module (10) is related according to target level lane line and obstacle signal and vehicle
Signal carries out automatic Pilot decision rule and control;Wherein, m is integer, and m >=1, n are integer, and
2. the development system of automated driving system according to claim 1, it is characterised in that: the value of the m is 1, n's
Value is 1;1 sensing module is laser radar or ultrasonic radar or high-precision map, 1 sensing
Group is made of camera (2), millimetre-wave radar (3) and 1 sensing module, and 1 sensing Fusion Module is second
Fusion Module is sensed, 1 sensing fusion parsing module corresponds to the second sensing fusion parsing module.
3. the archetype development system of automated driving system according to claim 1, it is characterised in that: the m's takes
The value that value is 2, n is 3;2 sensing modules are respectively laser radar and ultrasonic radar or laser radar and height
Precision map or ultrasonic radar and high-precision map;First sensing group in 3 sensing groups is by camera
(2), millimetre-wave radar (3) and wherein 1 sensing module composition, second sensing group is by camera (2), millimetre-wave radar (3)
It is constituted with another 1 sensing module, third sensing group is made of camera (2), millimetre-wave radar (3) and 2 sensing modules, institute
State 3 sensing Fusion Modules are respectively the merged to the lane line that first sensing group detects with obstacle information
Two sensing Fusion Modules, the third sensing fusion that the lane line that second sensing group detects is merged with obstacle information
Module, the 4th sensing Fusion Module that the lane line that third sensing group detects is merged with obstacle information, it is described
3 sensings fusion parsing module respectively correspond for second sensing fusion parsing module, third sensing fusion parsing module, the 4th
Sensing fusion parsing module.
4. the archetype development system of automated driving system according to claim 1, it is characterised in that: the m's takes
The value that value is 3, n is 2, and 3 sensing modules are respectively laser radar, ultrasonic radar and high-precision map, described
2 sensing groups in first sensing group is by camera (2), millimetre-wave radar (3) and wherein 1 sensing module is constituted,
Two sensing groups are made of camera (2), millimetre-wave radar (3) and 3 sensing modules, 2 sensing Fusion Modules point
The lane line that first sensing group Wei not be detected merged with obstacle information second sensing Fusion Module, to second
The third that the lane line that a sensing group detects is merged with obstacle information senses Fusion Module, and 2 sensings are melted
Parsing module is closed to respectively correspond as the second sensing fusion parsing module, third sensing fusion parsing module.
5. the archetype development system of automated driving system according to claim 2, it is characterised in that: at the signal
It manages module (11) and overtime detection is carried out to the timestamp in the corresponding channel of sensing fusion parsing module, select the target in some channel
The concrete mode that grade lane line and obstacle signal are handled are as follows: according to the corresponding channel of the second sensing fusion parsing module
Timestamp judges whether the break period in the channel is less than the time threshold a of setting, if it is, selecting the second sensing fusion solution
The target level lane line and obstacle signal in the corresponding channel of analysis module are handled, and the first sensing fusion parsing mould is otherwise selected
The target level lane line and obstacle signal in the corresponding channel of block are handled.
6. the archetype development system of automated driving system according to claim 3, it is characterised in that: at the signal
It manages module (11) and overtime detection is carried out to the timestamp in the corresponding channel of sensing fusion parsing module, select the target in some channel
The concrete mode that grade lane line and obstacle signal are handled are as follows: according to the corresponding channel of the 4th sensing fusion parsing module
Timestamp judges whether the break period in the channel is less than the time threshold a of setting, if it is, selecting the 4th sensing fusion solution
The target level lane line and obstacle signal in the corresponding channel of analysis module are handled, and otherwise sense fusion parsing mould according to third
The timestamp in the corresponding channel of block judges whether the break period in the channel is less than the time threshold a of setting, if it is, selecting
The target level lane line and obstacle signal in the corresponding channel of third sensing fusion parsing module are handled, otherwise according to second
The timestamp in the corresponding channel of sensing fusion parsing module judges whether the break period in the channel is less than the time threshold of setting
A, if it is, selecting at the target level lane line and obstacle signal in the corresponding channel of the second sensing fusion parsing module
Otherwise reason selects the target level lane line in the corresponding channel of the first sensing fusion parsing module and obstacle signal to be handled.
7. the archetype development system of automated driving system according to claim 4, it is characterised in that: at the signal
It manages module (11) and overtime detection is carried out to the timestamp in the corresponding channel of sensing fusion parsing module, select the target in some channel
The concrete mode that grade lane line and obstacle signal are handled are as follows: the corresponding channel of fusion parsing module is sensed according to third
Timestamp judges whether the break period in the channel is less than the time threshold a of setting, if it is, selecting third sensing fusion solution
The target level lane line and obstacle signal in the corresponding channel of analysis module are handled, otherwise according to the second sensing fusion parsing mould
The timestamp in the corresponding channel of block judges whether the break period in the channel is less than the time threshold a of setting, if it is, selecting
The target level lane line and obstacle signal in the corresponding channel of the second sensing fusion parsing module are handled, and otherwise select first
The target level lane line and obstacle signal in the corresponding channel of sensing fusion parsing module are handled.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910421454 | 2019-05-21 | ||
CN2019104214549 | 2019-05-21 |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110147109A true CN110147109A (en) | 2019-08-20 |
Family
ID=67591293
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910512711.XA Pending CN110147109A (en) | 2019-05-21 | 2019-06-13 | A kind of archetype development system of automated driving system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110147109A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110794406A (en) * | 2019-11-12 | 2020-02-14 | 北京经纬恒润科技有限公司 | Multi-source sensor data fusion system and method |
CN112776797A (en) * | 2021-02-27 | 2021-05-11 | 重庆长安汽车股份有限公司 | Original parking space parking establishment method and system, vehicle and storage medium |
CN113479191A (en) * | 2021-06-30 | 2021-10-08 | 重庆长安汽车股份有限公司 | Lane-line-free lane boundary detection system and method for parking and vehicle |
CN113734166A (en) * | 2021-09-30 | 2021-12-03 | 重庆长安汽车股份有限公司 | Automobile automatic driving control system and method based on perception fusion SWC |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105857310A (en) * | 2016-03-29 | 2016-08-17 | 谢奇 | Autonomous vehicle following method and system |
WO2018012165A1 (en) * | 2016-07-15 | 2018-01-18 | 日立オートモティブシステムズ株式会社 | Sensor device and autonomous vehicle |
DE102015100316B4 (en) * | 2014-01-16 | 2018-05-17 | Toyota Motor Engineering & Manufacturing North America Inc. | Planning device for a sideways maneuver for an automated driving system |
CN108556767A (en) * | 2018-03-01 | 2018-09-21 | 李洪运 | A kind of expansible intelligent driving auxiliary system |
CN108776472A (en) * | 2018-05-17 | 2018-11-09 | 驭势(上海)汽车科技有限公司 | Intelligent driving control method and system, onboard control device and intelligent driving vehicle |
CN108958284A (en) * | 2018-06-29 | 2018-12-07 | 湖北电鹰科技有限公司 | A kind of unmanned plane obstacle avoidance system and method |
CN109756429A (en) * | 2017-11-06 | 2019-05-14 | 阿里巴巴集团控股有限公司 | Bandwidth allocation methods and equipment |
-
2019
- 2019-06-13 CN CN201910512711.XA patent/CN110147109A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102015100316B4 (en) * | 2014-01-16 | 2018-05-17 | Toyota Motor Engineering & Manufacturing North America Inc. | Planning device for a sideways maneuver for an automated driving system |
CN105857310A (en) * | 2016-03-29 | 2016-08-17 | 谢奇 | Autonomous vehicle following method and system |
WO2018012165A1 (en) * | 2016-07-15 | 2018-01-18 | 日立オートモティブシステムズ株式会社 | Sensor device and autonomous vehicle |
CN109756429A (en) * | 2017-11-06 | 2019-05-14 | 阿里巴巴集团控股有限公司 | Bandwidth allocation methods and equipment |
CN108556767A (en) * | 2018-03-01 | 2018-09-21 | 李洪运 | A kind of expansible intelligent driving auxiliary system |
CN108776472A (en) * | 2018-05-17 | 2018-11-09 | 驭势(上海)汽车科技有限公司 | Intelligent driving control method and system, onboard control device and intelligent driving vehicle |
CN108958284A (en) * | 2018-06-29 | 2018-12-07 | 湖北电鹰科技有限公司 | A kind of unmanned plane obstacle avoidance system and method |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110794406A (en) * | 2019-11-12 | 2020-02-14 | 北京经纬恒润科技有限公司 | Multi-source sensor data fusion system and method |
CN112776797A (en) * | 2021-02-27 | 2021-05-11 | 重庆长安汽车股份有限公司 | Original parking space parking establishment method and system, vehicle and storage medium |
CN113479191A (en) * | 2021-06-30 | 2021-10-08 | 重庆长安汽车股份有限公司 | Lane-line-free lane boundary detection system and method for parking and vehicle |
CN113479191B (en) * | 2021-06-30 | 2023-04-07 | 重庆长安汽车股份有限公司 | Lane-line-free lane boundary detection system and method for parking and vehicle |
CN113734166A (en) * | 2021-09-30 | 2021-12-03 | 重庆长安汽车股份有限公司 | Automobile automatic driving control system and method based on perception fusion SWC |
CN113734166B (en) * | 2021-09-30 | 2023-09-22 | 重庆长安汽车股份有限公司 | Automatic automobile driving control system and method based on sensing fusion SWC |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110147109A (en) | A kind of archetype development system of automated driving system | |
EP3549119B1 (en) | Method, apparatus and computer program product for estimation of road traffic condition using traffic signal data | |
CN105929823A (en) | Automatic driving system and driving method based on existing map | |
US8625850B2 (en) | Environment recognition device and environment recognition method | |
CN102428505B (en) | Vehicular Environment Estimation Device | |
KR20200101307A (en) | Interaction method and apparatus between vehicles | |
CN107894767A (en) | A kind of automatic driving vehicle transverse movement control object system of selection | |
EP2784762A1 (en) | Vehicle identification device | |
CN110691990B (en) | Obstacle detection method and equipment | |
CN101241188A (en) | Collision avoidance system and method of detecting overpass locations using data fusion | |
CN109922439A (en) | The fusion method of multi-sensor data, the object detection method based on V2X and system | |
CN110356339B (en) | Lane change blind area monitoring method and system and vehicle | |
EP3418686A1 (en) | Feature data structure, control device, storage device, control method, program, and storage medium | |
CN109270524A (en) | Based on unpiloted multi-data fusion obstacle detector and its detection method | |
CN113060131B (en) | Obstacle map generation method and device, vehicle and storage medium | |
KR20190084614A (en) | Traffic signal control system and method for cross-road | |
CN110275167A (en) | A kind of control method of radar detection, controller and terminal | |
CN111505623A (en) | Method and system for detecting obstacle in driving process of unmanned vehicle and vehicle | |
CN202130447U (en) | Novel lane line deviation detection device | |
CN112105539A (en) | Autonomous control method for rear vehicles in fleet | |
CN109164805A (en) | The dispatching method and device of robot driving path | |
CN110395255A (en) | Judgment method, system and the computer readable storage medium of adjacent lane lateral direction of car motion state | |
US11885640B2 (en) | Map generation device and map generation method | |
CN109932721A (en) | Error and detecting probability analysis method applied to the fusion of more sensors | |
KR101446545B1 (en) | Display system of vehicle information based on location in cross-road |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190820 |