CN109683606A - A kind of pilotless automobile automatic obstacle avoiding method - Google Patents
A kind of pilotless automobile automatic obstacle avoiding method Download PDFInfo
- Publication number
- CN109683606A CN109683606A CN201811391198.5A CN201811391198A CN109683606A CN 109683606 A CN109683606 A CN 109683606A CN 201811391198 A CN201811391198 A CN 201811391198A CN 109683606 A CN109683606 A CN 109683606A
- Authority
- CN
- China
- Prior art keywords
- barrier
- vehicle
- obstacle
- ibeo
- data
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
- G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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 or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Computer Networks & Wireless Communication (AREA)
- Traffic Control Systems (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The present invention provides a kind of automatic driving vehicle barrier-avoiding methods, to ensure that vehicle is safely operated in complicated and crowded running environment, it is more accurate to the perception of multidate information, the present invention is based on the vehicle recongnition techniques of visual sensor by having introduced combined monitoring, four line laser radar Ibeo and three-dimensional Velodyne laser radar sensor data input, map Velodyne laser pre-treated, it can be divided into two classes by Environment Obstacles in different times, that is the interim movement of fence that is stored in Ibeo and information movement of the mobile and static partition wall of flow obstacle barrier static state fence.The present invention not only can satisfy the promotion of vehicle monitoring system performance, but also the optimization that vehicle overall performance defect may be implemented improves, reinforces and the system administration operating mechanism of maintenance level intelligence.
Description
Technical field
The present invention relates to pilotless automobile more particularly to a kind of avoidance sides of pilotless automobile automatic obstacle avoiding system
Method.
Background technique
Automobile brings convenience to the mankind, improves the trip and life style of people.However, automobile generates product to people
While pole influences, negative effect is also brought, it is especially very serious to the negative effect of traffic accident.
At present pilotless automobile technology caused society note that more and more scientists concentrate on
In the scientific research to unmanned automobile.As in the world 92% or more traffic accident be all as caused by human factor, such as
It drives when intoxicated, furious driving, fatigue driving and careless drive；In order to avoid even avoiding road traffic accident, most important solution
Certainly method is to make automobile more intelligent.
Automatic driving vehicle is a kind of autonomous vehicle, it not only possesses the normal function of conventional truck, and there are also to environment
Superpower adaptation perception.In identical transportation environment, intelligence system is carried out according to the motion state and traffic environment information of vehicle
Analysis greatly reduces the incidence of road traffic accident with the optimal trip strategy of determination.In the case where no driver,
The computer intelligence identifying system of automobile can observe current situation in the case where 360 degree, it can monitor road and speed
Degree makes more accurate judgement, and predicts and adjust control.In addition, the system is not limited by the body of human manipulation person, such as wine
Essence, fatigue or depression.
In real world, the automobile of driver is not often subject to the influence of random obstacle, because they are each in the world
There is route on ground.In this regard, it is necessary to execute task, in a secure manner to avoid certain obstacles.Complicated and changeable
External environment in, the method that our current scientific and technological levels solve dynamic disorder is still immature, still needs to be solved there are many problem
Certainly.
In short, in today that traffic accident takes place frequently, under the background of very uncertain obstacle, in intelligent vehicle control system
It unites in also very jejune scientific research, studies and the control system of intelligent vehicle is carried out to be in real time very with accurate planing method
Important.
Summary of the invention
Goal of the invention: in order to solve the above technical problems, the present invention provides a kind of pilotless automobile automatic obstacle avoiding method, draw
Enter hybrid control system theory to analyze obstacle avoidance system hybrid characters, that realizes pilotless automobile obstacle avoidance system mixes control
System, can make pilotless automobile that traffic accident and casualties can be effectively reduced.
Technical solution: pilotless automobile automatic obstacle avoiding method of the present invention, comprising the following steps:
Step 1, the data processing of the vehicle and barrier of the barrier of mobile environment；
Step 2, Ibeo exports moving obstacle data processing；
Step 3, Data Fusion of Sensor.
Step 1 the following steps are included:
Step 1.1, withdraw the obstacle of speed disengaging: three-dimensional laser radar Velodyne degree is into and out data format a little,
Method of the Cloud firstly the need of pretreatment cloud data.With advanced grid, project is used, to determine three-dimensional diagonal mesh
Characteristic, according to the grid of each project show.
Step 1.2, the cluster and extraction of the unknown obstacle of environment: the extraction of barrier.Firstly, we must be prevented from blocking
Obstacle.The obstacle of grid is as made of regional cluster algorithm combination.An open list is created, for storing all grids
Grid, then in a random list randomly choose a grid, adjacent grid is then checked after inspection, until
It is closed and the list opened is empty.
Step 1.3, obstacle block tracks: when tracking obstacle block, it would be desirable to by obstacle block and be stored in current time collection
Block in the obstacle list of group associates.The present invention is associated using maximum correlation value method.Vehicle is moved to suitably
Then destination channel position relocates the position of vehicle when vehicle starts.
Working principle: the present invention is based on the vehicle identification of visual sensor, combined monitoring, four line laser radar Ibeo and
The technologies such as three-dimensional laser radar Velodyne sensor data acquisition improve the avoidance respond of pilotless automobile, together
When shorten the avoidance time.
The motor-driven vehicle system hidden obstacle and protect system current, the purpose is to meet investigation, prediction and prevention
Current obstacle meets timeliness and accuracy that daily life generally requires, and main method is: radar laser and radar millimeter
The information of their sendings of wave will read information format control and analyze these data by calculating come data fusion, and analysis obtains
The information of barrier in the information and environment of the position of Intelligent unattended vehicle in the environment.Mistake and the surrounding obstacles letter of information
The inaccuracy of breath may will affect the behavior decision of intelligent vehicle.Grid wall is built, including about accurate real-time, decision position one
Bodyization is based on four line laser radar of information wire and three-dimensional check moves to grid wall problem, this method detects laser three-D laser
Sensor Velodyne periphery monitoring, filter analysis, monitoring and prediction regional movement, it is desirable that establish in use
It the conceptual data of Velodyne line and integrates Ibeo and moves knot in grid wall, laser scanning on the position of moving obstacle
Fruit, to handle data.
The utility model has the advantages that for the flowing of the obstacle motor vehicles encountered under complex road condition environment and multiple when present invention research
Miscellaneous road conditions detect and hinder before and after vehicle in accurate area using the obstacle motor vehicles that external environment laser acquisition is current
Hinder motion detection.Based on the fusion of four line laser radars and three-dimensional laser radar, to receive the obstacle of signal laser return, detection
Signal is detected by positioning in laser reflection position.The method use a three-dimensional laser sensor Velodyne and four lines
Laser radar Ibeo for detecting a wide range of barrier of vehicle periphery, and predicts using Kalman's filter the fortune of barrier
Dynamic state.According to a series of convergence as a result, being corrected to the extension of sensing data processing time.This method is not only
The accuracy of front detection is improved, and keeps it more effective.
Therefore, compared with prior art, the present invention has real-time and accuracy, both can guarantee the real-time performance of system,
To adapt to the variation of environment, and more true environment description can be provided for unmanned automobile collision prevention process.
Detailed description of the invention
Fig. 1 is to carry out rasterizing result figure to the raster pattern of generation；
Fig. 2 is that barrier occupies grid cluster result；
Fig. 3 is barrier block parametrization figure；
Fig. 4 is BOX data synchronization result figure；
Fig. 5 is moving obstacle position correction result figure；
Fig. 6 is velocity magnitude comparative result figure；
Fig. 7 is directional velocity comparative result figure；
Fig. 8 is that unmanned vehicle moves obstacle avoidance system function implementation flow chart.
Specific embodiment
The present invention the following steps are included:
Step 1, the data processing of the vehicle and barrier of the barrier of mobile environment；
The pretreatment of step 1.1Velodyne output data；
Velodyne output data is point cloud format, is freely accessible to intelligent speed in pattern framework and limits；Road vehicle is logical
Often modeling, analysis environment intelligence during escape, required minimum safe distance；The laser radar sensor recalled, the whole world are fixed
Position system and communication protocol data, intelligent external environment are established.Firstly, it is necessary to pre-process to these cloud data, nothing is studied
Man-machine preprocess method.The grid of projection is grid power according to obtained from the information of each point projection in a network.
The flowing of the main control export intelligent vehicle running determined, control strategy and longitudinally controlled strategy, on foot
Lateral supervision strategic angle changes steering wheel, and in order to change the direction of vehicle, it can be used for roading.Control vehicle
Traveling is in longitudinal framing, and in general, the vehicle and pedestrian of the barrier of mobile environment is mainly higher obstacle, so this
Invention is more directly minimized, the method for net center of a lattice the center of the grid map heart within a grid each point, three
Tie up the value between maximum height minimum height values, on the basis of the demand for the intelligent vehicle that do not equip, independent suspension
It is installed on bridge at rear and later: before steering mechanism after having used mechanical gear converter, the rear rotation of rear-wheel
Dynamic to reduce the radius of gyration, motivation installs the brake of wrong brake sign, and propulsion system drives automobile charging.In the control of automobile
In system processed, the spool of front results in the gearbox not upgraded automatically.
As shown in Figure 1, being divided using maximum method and the smallest method the intersection three-dimensional data from laser
Grade, obtains the image of barrier grid.
Step 1.2 barrier block cluster and feature extraction；
For the feature of barrier in extraction environment, it is necessary first to gather a barrier grid in barrier block.
The present invention captures grid to the barrier in grid map using region growing cluster and clusters, as a result grid as shown in Figure 2
Barrier in map in cluster areas occupies grid and is clustered into barrier block one by one.Cluster is that barrier block is collected
Data are also convenient for the obstacle tracking of next step convenient for analyzing and improving steering direction accurately judgement property.
The present invention uses feature of the minimum rectangle parameter that can cover barrier block as barrier block.Such as Fig. 3 institute
Show, parameter includes: long side length L, short side long R, center position O (x, y), occupancy k of the barrier to rectangle.
Step 1.3 obstacle tracking；
Track barrier block, it is necessary to contact current block with the block being stored on domestic barrier inventory
Come.It is considered as the movement for being parallel to road surface that the pause of vehicle, which is in the influence of mobility that the same time carries out,.Current library
It deposits and has been used to connect block and maximally related method, include the following three types situation:
1. it is stored in the list of moving obstacle, but it is limited in the block that one does not meet current class
On block；Other values are constant；
2. there is no hiding obstacles for current Synchrone movement to prevent obstacle, and are added to moving disorder column
In table.The speed and speed of acceleration are zero, and the velocity and acceleration of covariance is preliminary, initial value zero.In database
The paces of the position obstacle about detection of the level of security of the barrier and obstacle block obstacle spread module of flow meter, algorithm are accelerated
Variance and speed, variance and acceleration.The label line detection algorithms on complicated road the problem of, image pixel is applied.Change
Access road image partition method, to describe boundary line profile；Scanning and digital image sampling are carried out, the calibration line the characteristics of
In, extract and selected selected environment road conditions feature；Braking system includes selection and the meter of each element parameter of brake apparatus
It calculates, and calculates the parameter of its composition mechanism；Intelligent unattended vehicle enters the structure of the program analysis of system movement and type exists
Dyskinesia on Ibeo laser sensor directly exported can arrive environmental information, these information are the dark wheels according to scanning
For exterior feature come what is modeled, they use each obstacle, are exported in the form of environmental information.For each barrier and obstacle
Table, for each of list barrier block OBiEach obstacle OM obtained with current timejAll there is a relating value
fij.Modular responsibility method changes firstly the need of center:
tij=tj-ti
Wherein, tiAnd tj, it is OB respectivelyiIt obtains refreshing the time of obstacle cluster and refresh the moment.Represent barrier most
New speed,Represent the newest moment.WithRepresent the position coordinates of different time.
Wherein L is rectangle side length, and R is that rectangle short side is long, center position O (x, y), occupancy k of the barrier to rectangle.
Threshold correlation value f is set0, long side length L, short side long R, center position O (x, y), occupancy of the barrier to rectangle
Rate k.
Maximum value fmnIf they are not less than threshold value f0, then it is identified.About object block, deleted from decision matrix
The positive link of all correlations, and propose a new decision matrix:
Until finding highest threshold value, combined value is less than threshold value or decision matrix is sky.
Preferably, step time as needed for data collection and analysis is different, analysis center position with
Velodyne is identical.The position for the mobile barrier that data processing obtains has a certain difference, and needs the center with these boxes
Position keeps synchronizing.
3. being to establish railroad tracks markup model according to the characteristics of visual field road graticule, dynamically tracked in calibration line
The introducing of Kalman filter, to realize detection and monitoring to calibration line.
X (k)=AX (k-1)+BU (k)+W (k)
Z (k)=HX (k)+C (k)+V (k)
WhereinCoefficient A=1, coefficient
C (k)=s (k-1), W (k) and V (k) respectively indicate measurement process and noise.
The covariance of door is respectively q and r, this allows us to predict current state according to the preceding state of system:
X (k | k-1)=AX (k-1 | k-1)+BU (k)
P (k | k-1)=AP (k-1 | k-1) A+Q
Motion state X (k | k)=X (k | k-1)+Kg (k) (Z (k)-HX (k | k-1) of barrier block)
Kg (k) is Kalman gain
Covariance P (k | k)=(1-Kg (k) H) P (k | k-1) moved at present
Kg (k)=P (k | k-1) H/ (HP (k | k-1) H+R)
K is occupancy of the barrier to rectangle.
In practice, the depending on degree to flowing sexual dysfunction and confidence level of decision-making are determined in highest and lowest limit
's.Finally, confidence level of decision-making will be removed from the list lower than the block of fixed value in moving disorder list.Due to the present invention
Think that these obstacles disappear from the environment of non-driven vehicle, it is also possible to ensure to be arranged in moving disorder list
The quantity of obstacle will not increase over time.
Step 2 Ibeo exports moving obstacle data processing；
Step 2.1 Ibeo exports moving obstacle data format；
Four line laser sensors can provide the information about dyskinesia in environment, it models barrier using a rectangle
Hinder object, according to the profile of scanning barrier.
In the frame of macro-plan, environmental information is available, and on the basis of obstacle extracts, a few class situations are combined
Together, in the case where unmanned, basic exercise model has been carried out in the transportation environment of an On Fluctuations deep
Research, is the decision model based on decision tree on this basis to determine the condition of decision and the respective numbers of vehicle sport mode
Type.Finally, verifying feasibility by one On Fluctuations flow simulation environment of building.
The processing of step 2.2 Ibeo output data；
The present invention provides in the center of its directional velocity opposite direction BOX (x, y), as shown in Figure 4.It is flat in test
The coordinate of visual sensor on platform is defined as in the coordinate system of motor vehicles.Corresponding mathematical model is established, due to
The fragility of traffic environment picture quality and be difficult to the problem of extracting, filter algorithm is improved.And relative direction translation away from
Meet following formula from S:
S=λ (tV-tI)v
Wherein tv and tI is the time-consuming numerical value in acquisition process Velodyne data result and Ibeo data result, λ respectively
It is parameter.
Step 3 Data Fusion of Sensor；
Use the characterisitic parameter of the same object of Velodyne and Ibeo sensor measurement.In order to be passed in view of different sensors
The statistical property of defeated data judges that their validity is more section by using the relationship between the data of different sensors
It learns.Therefore, theoretical distance is trusted the success for being used to merge two data collectors and is merged.It is assumed that x1 and x2 action
It is that movement and Ibeo outflow are calculated by Velodyne, all of which all Gaussienne distribution.If obtained data are respectively
X1 and x2, then probability density function is shown in following formula:
xiTo xjConfidence distance
Establishing a crucial matrix is
R2To obtain one relational matrix.
Finally will meet to export supports the sensing data of sensor number 2 to merge according to following formula, finally obtains
The motion state x of barrier block, wherein l is the sensor number for meeting output and supporting sensor number 2.Handle the obstacle encountered and position
Set the safety of the unmanned intelligent automobile of obstacle at least in more than one classification and isolation.
Wherein l is number of probes.
After merging, it is contemplated that the occupancy of the time of data collection and processing, these obstacles are actually changing, this
Need to change these positions captured.It establishes corresponding numerical ciphers and represents absolute position, coding and measurement in beginning and end
This position it is related with measurement process.The environment road conditions obstacle of people and vehicular movement is, it is specified that fix the confidence of list and flowing
Value is greater than deadlocking for block, and the data of distance and speed merge action speed as shown in Fig. 5, and repairing for block is freezed in merging
Change and be updated, and be stored in moving disorder list, wherein letter S ' is translation distance, λ ' is undetermined coefficient.V ' is to update speed
Degree, tvFor the time.S ' reaches following size:
S '=λ ' tV·v'
As shown in barrier motion state detection result in table 1, in digital mobile 0, the error in speed and direction is small
In the obstructing objects of numerical value 2.It is detected in view of flowing environment only for four line laser radar lbeo in the area 0#, situation of wandering about as a refugee
Data be incorporated into Ibeo and Velodyne data, and vehicle real-time road condition information handles money from # 2 achievement Velodyne
Material.
Table 1
In order to distinguish both methods institute it is collected as a result, Fig. 7 and Fig. 8 be to an at the uniform velocity 8.3m/s, traveling to the south,
The automobile of zigzag traveling carries out Velodyne respectively and individually handles and the result of Velodyne, Ibeo fusion treatment.
As shown in fig. 6, the speed fluctuation range that Velodyne and Ibeo fusion treatment obtains is smaller, reliability is higher.
Fig. 7 is the results show that the directional velocity amplitude of variation crossed of Velodyne and Ibeo fusion treatment is smaller, and stability is more
It is high.It is effectively worked by establishing static static test and dynamic test in bulk and authentication module.Finally, it is verified that optimization is surveyed
The validity of trial and error procedure.The working frequency of each frame can reach in normal range (NR), it ensure that the regular real-time of system
And accuracy.
Combined monitoring of the Fig. 8 based on GPS positioning and laser radar detects laser to receive the obstacle of signal laser return
Reflection position detects obstacle signal by positioning, to plan intelligent route, unmanned vehicle moves avoidance obstacle and executes system function
It can be achieved.
Claims (5)
1. a kind of pilotless automobile automatic obstacle avoiding method, it is characterised in that: the following steps are included:
(1) data processing of the vehicle and barrier of the barrier of mobile environment；
(2) Ibeo exports moving obstacle data processing；
(3) Data Fusion of Sensor.
2. pilotless automobile automatic obstacle avoiding method according to claim 1, it is characterised in that: step (1) includes following
Step:
(1.1) Velodyne output data pre-processes；
(1.2) barrier block cluster and feature extraction；
(1.3) obstacle tracking.
3. pilotless automobile automatic obstacle avoiding method according to claim 1, it is characterised in that: step (2) includes following
Step:
(2.1) Ibeo exports moving obstacle data format；
(2.2) Ibeo output data is handled.
4. pilotless automobile automatic obstacle avoiding method according to claim 1, it is characterised in that: sensed using three-dimensional laser
Device Velodyne, for detecting a wide range of barrier of vehicle periphery.
5. pilotless automobile automatic obstacle avoiding method according to claim 4, it is characterised in that: and use Kalman's filter
To predict the motion state of barrier.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811391198.5A CN109683606A (en) | 2018-11-21 | 2018-11-21 | A kind of pilotless automobile automatic obstacle avoiding method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811391198.5A CN109683606A (en) | 2018-11-21 | 2018-11-21 | A kind of pilotless automobile automatic obstacle avoiding method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109683606A true CN109683606A (en) | 2019-04-26 |
Family
ID=66185746
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811391198.5A Pending CN109683606A (en) | 2018-11-21 | 2018-11-21 | A kind of pilotless automobile automatic obstacle avoiding method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109683606A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045376A (en) * | 2019-04-28 | 2019-07-23 | 森思泰克河北科技有限公司 | It can travel area obtaining method, computer readable storage medium and terminal device |
| CN110789287A (en) * | 2019-10-08 | 2020-02-14 | 江苏科技大学 | Adjustable additional air chamber air suspension system based on three-dimensional optical scanning and self-adaptive control method thereof |
| CN112132929A (en) * | 2020-09-01 | 2020-12-25 | 北京布科思科技有限公司 | Grid map marking method based on depth vision and single line laser radar |
| CN115600158A (en) * | 2022-12-08 | 2023-01-13 | 奥特贝睿（天津）科技有限公司（Cn） | Unmanned vehicle multi-sensor fusion method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107831777A (en) * | 2017-09-26 | 2018-03-23 | 中国科学院长春光学精密机械与物理研究所 | A kind of aircraft automatic obstacle avoiding system, method and aircraft |
| CN108536149A (en) * | 2018-04-25 | 2018-09-14 | 吉林大学 | A kind of automatic driving vehicle avoidance obstacle device and control method based on the paths Dubins |
-
2018
- 2018-11-21 CN CN201811391198.5A patent/CN109683606A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107831777A (en) * | 2017-09-26 | 2018-03-23 | 中国科学院长春光学精密机械与物理研究所 | A kind of aircraft automatic obstacle avoiding system, method and aircraft |
| CN108536149A (en) * | 2018-04-25 | 2018-09-14 | 吉林大学 | A kind of automatic driving vehicle avoidance obstacle device and control method based on the paths Dubins |
Non-Patent Citations (1)
| Title |
|---|
| 辛煜等: "基于激光传感器的无人驾驶汽车动态障碍物检测及表示方法", 《机器人》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045376A (en) * | 2019-04-28 | 2019-07-23 | 森思泰克河北科技有限公司 | It can travel area obtaining method, computer readable storage medium and terminal device |
| CN110789287A (en) * | 2019-10-08 | 2020-02-14 | 江苏科技大学 | Adjustable additional air chamber air suspension system based on three-dimensional optical scanning and self-adaptive control method thereof |
| CN112132929A (en) * | 2020-09-01 | 2020-12-25 | 北京布科思科技有限公司 | Grid map marking method based on depth vision and single line laser radar |
| CN112132929B (en) * | 2020-09-01 | 2024-01-26 | 北京布科思科技有限公司 | Grid map marking method based on depth vision and single-line laser radar |
| CN115600158A (en) * | 2022-12-08 | 2023-01-13 | 奥特贝睿（天津）科技有限公司（Cn） | Unmanned vehicle multi-sensor fusion method |
| CN115600158B (en) * | 2022-12-08 | 2023-04-18 | 奥特贝睿(天津)科技有限公司 | Unmanned vehicle multi-sensor fusion method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10817731B2 (en) | Image-based pedestrian detection | |
| US11885910B2 (en) | Hybrid-view LIDAR-based object detection | |
| CN109683606A (en) | A kind of pilotless automobile automatic obstacle avoiding method | |
| US10860896B2 (en) | FPGA device for image classification | |
| CN108345822B (en) | Point cloud data processing method and device | |
| US20190310651A1 (en) | Object Detection and Determination of Motion Information Using Curve-Fitting in Autonomous Vehicle Applications | |
| US20190145765A1 (en) | Three Dimensional Object Detection | |
| Petrovskaya et al. | Model based vehicle detection and tracking for autonomous urban driving | |
| CN104520732B (en) | The method and relevant device of alignment sensor | |
| CN107609522A (en) | A kind of information fusion vehicle detecting system based on laser radar and machine vision | |
| CN109460709A (en) | The method of RTG dysopia analyte detection based on the fusion of RGB and D information | |
| CN107193011A (en) | A kind of method for being used to quickly calculate car speed in automatic driving car area-of-interest | |
| CN117441197A (en) | Laser radar point cloud dynamic segmentation and fusion method based on driving safety risk field | |
| CN111291697B (en) | Method and device for detecting obstacles | |
| CN114419152B (en) | Target detection and tracking method and system based on multi-dimensional point cloud characteristics | |
| US11657591B2 (en) | Autonomous vehicle system for intelligent on-board selection of data for building a remote machine learning model | |
| CN110211176A (en) | A kind of Camera extrinsic number correction System and method for | |
| CN113378741B (en) | Auxiliary sensing method and system for aircraft tractor based on multi-source sensor | |
| CN106683530A (en) | Computerized judging system and method based on three-dimensional laser vision and high-precision lane model | |
| CN113071518B (en) | Automatic unmanned driving method, minibus, electronic equipment and storage medium | |
| EP4016115A1 (en) | Vehicle localization based on radar detections | |
| CN115662166B (en) | Automatic driving data processing method and automatic driving traffic system | |
| US11820397B2 (en) | Localization with diverse dataset for autonomous vehicles | |
| CN110083099A (en) | One kind meeting automobile function safety standard automatic Pilot architecture system and working method | |
| CN113792598B (en) | Vehicle-mounted camera-based vehicle collision prediction system and method |
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 |
Application publication date: 20190426 |
|
| RJ01 | Rejection of invention patent application after publication |