CN103679851B - A kind of multi-sensor data processes exempts from synchronous method and system - Google Patents
A kind of multi-sensor data processes exempts from synchronous method and system Download PDFInfo
- Publication number
- CN103679851B CN103679851B CN201310740531.XA CN201310740531A CN103679851B CN 103679851 B CN103679851 B CN 103679851B CN 201310740531 A CN201310740531 A CN 201310740531A CN 103679851 B CN103679851 B CN 103679851B
- Authority
- CN
- China
- Prior art keywords
- data
- storage area
- data storage
- sensor
- current time
- 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.)
- Active
Links
Abstract
The invention provides one realize intelligent vehicle multi-sensor data process exempt from synchronous method and system, arranging first, second data storage area being similar to double blackboard mechanism, the first data storage area record current time drives right of way radar map and the vehicle oneself state parameter of situation;The right of way radar map in a moment and vehicle oneself state parameter, the reading data within this decision-making period on second data storage area record.Before this control cycle carries out decision-making, the second data storage area resets and copies the data of the first data storage area, and the first data storage area continues right of way radar map information and the vehicle oneself state parameter of record current time, so moves in circles.The present invention utilizes what Double Data memory block mechanism realized intelligent vehicle perception data to exempt from synchronization process, solve the time synchronization problem between multi-sensor data, guaranteeing that the reading and writing data between different sensors is separate, the even temporarily loss of certain sensing data does not affects decision-making and the control of current time intelligent vehicle.
Description
Technical field
The invention belongs to intelligent vehicle technical field, particularly relate at a kind of multi-sensor data
Reason exempts from synchronous method and system.
Background technology
Intelligent driving technology relates to computer science, communication science, Cognitive Science, vehicle
Engineering, electric and electronic engineering, Control Science and Engineering, systematic science and technology, man-machine
Many science such as engineering science, artificial intelligence, intelligent vehicle is to weigh a national scientific research in fact
One of important symbol of power and industrial level.The appearance of automatic driving car, fundamentally changes
Become the vehicle drive mode in traditional " people-Che-road " closed loop system, by driver from
The most loaded down with trivial details, rudimentary, lasting driving-activity frees, utilizes advanced person's
Sensor and information technology control vehicle travel, allow conventional in driving-activity, persistently,
Rudimentary, repeat operation be automatically performed, it is possible to be greatly enhanced traffic system efficiency and
Safety, improves the quality of mankind's movable living, has social application widely and is worth.With
Time, the research of intelligent driving technology will greatly strengthen China in sides such as automobile active safety
The core competitiveness in face, to promoting China's automobile electronics and automobile industry autonomous innovation
Ability has great strategic importance.
Intelligent vehicle, on the basis of dynamics of vehicle, including environment sensing and self study, is led
The parts such as boat, network communication, intelligent decision, wagon control, cognitive simulation and calculating.
Wherein, intelligent vehicle relies on the various onboard sensor such as photographic head, radar to realize Zhou Bianhuan
The perception of environment information.The Physical Mechanism of many onboard sensors of intelligent vehicle, the sampling period,
Installation site, the data volume size etc. of acquisition have difference, will realize all multisensors
Synchronize then to need to ask for the greatest common divisor in its sampling period, or add timestamp, so
Cause treatment effeciency to reduce, and complicate the issue.
Summary of the invention
It is an object of the invention to provide the process of a kind of intelligent vehicle multi-sensor data to exempt to synchronize
Method and system, to solve in prior art between multiple onboard sensors due to the sampling period
Difference causes perception data time mismatch problem.
In order to complete the object of the invention, the invention provides a kind of intelligent vehicle multisensor number
Exempting from synchronous method according to process, described method comprises the steps:
Onboard sensor gathers the surrounding enviroment information of intelligent vehicle current time;By gather
The surrounding enviroment information of described current time writes the right of way radar map of current time in real time;
Vehicle oneself state ginseng by the right of way radar map of described current time with current time
Number is stored in the first data storage area;
Zeros data in second data storage area, replicates in described first data storage area
Data in the second data storage area, for the reading data in current decision cycle;
The driving situation map of described first data storage area real-time update current time and vehicle
Oneself state parameter;
After the current decision cycle completes, the zeros data in described second data storage area,
Replicate the data in described first data storage area after updating to the second data storage area
In, enter next decision-making period, so move in circles.
Preferably, described vehicle oneself state parameter, all include: the position in track, place,
Driving model, speed, acceleration, angular velocity, angular acceleration.
Preferably, described onboard sensor includes video camera, radar, GPS device.
Preferably, described surrounding enviroment information uses number with vehicle oneself state parameter information
Group structure storage.
Preferably, described second data storage area can not be carried out except by the second data storage
Zeros data in district and the data in described first data storage area are copied to described
Other write operation in two data storage areas.
Preferably, described second data storage area does not consider within the current decision cycle described
The sensor senses data that first data storage area updates.
Preferably, described system includes: onboard sensor, right of way radar map memory element,
Vehicle oneself state parameter storage unit, the first data storage area, the second data storage area,
Described onboard sensor is for gathering the surrounding enviroment information of intelligent vehicle;
Described right of way radar map memory element is used for storing intelligent vehicle surrounding enviroment information;
Described vehicle oneself state parameter storage unit is for storing the vehicle obtained in real time certainly
Body state parameter;
Described first data storage area for store the described right of way radar map of current time with
The described vehicle oneself state parameter of current time;
Described second data storage area is replicated by described first data storage area for storage
The data arrived, and zeros data after the current decision cycle completes, replicate the institute after updating
State the data in the first data storage area, enter next decision-making period.
Preferably, described vehicle oneself state parameter, all include: position, track, place street
Put, driving model, speed, acceleration, direction, angular velocity, angular acceleration.
Preferably, described onboard sensor includes video camera, radar, GPS device.
Preferably, described surrounding enviroment information uses number with vehicle oneself state parameter information
Group structure storage.
Preferably, described second data storage area can not be carried out except by the second data storage
Zeros data in district and the data in described first data storage area are copied to described
Other write operation in two data storage areas.
Beneficial effects of the present invention is, compared with prior art, the present invention uses for reference the mankind and drives
The selective memory feature of the person's of sailing cognitive process, utilizes Double Data memory block mechanism to realize intelligence
Can car perception data exempt from synchronization process, solve between multiple onboard sensor due to sampling
Cycle difference causes perception data time mismatch problem, it is to avoid read/write conflict;Guarantee
Reading and writing data between different sensors is separate, is independent of each other, even certain sensing
Temporarily losing all without decision-making and the control affecting current time intelligent vehicle of device data.
Accompanying drawing explanation
Fig. 1 is the left blackboard schematic diagram of the present invention;
Fig. 2 is the right blackboard schematic diagram of the present invention.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and beneficial effect clearer,
Below in conjunction with drawings and Examples, the present invention is further elaborated.Should manage
Solve as specific embodiment described herein only in order to explain the present invention, be not used to limit
Protection scope of the present invention processed.
Human driver, in cognitive process, has the feature of selective attention: to
The driving-activity become history, the time is the most remote, forgets the most;Current driving situation,
It is that next step carries out the basis of decision-making, thus attention is placed on and drives current by driver
On the impermanent memory of situation.Meanwhile, to forgeing of the driving situation pass by, ability
Guarantee selective attention.Drive the extended residual of situation, be embodied in and there is statistical property
Among driving experience and behavior.
The present invention uses for reference the selective attention feature in human driver's cognitive process, proposes
The Double Data memory block using double blackboard mechanism realizes intelligent vehicle multi-sensor data and processes
Exempt from synchronous method, efficiently solve when all multi-sensor informations merge due to sampling week
The perception data time unmatched problem that phase difference is caused, carries out intelligence for intelligent vehicle
Decision-making provides reliable basis.
Being provided with blackboard in the classroom of school to impart knowledge to students for teachers' instruction, blackboard typically divides
For left and right (or upper and lower) two parts, left and right blackboard can push-and-pull.In office coaching, religion
When teacher needs writing on the blackboard, the most first write at left blackboard;After left blackboard is write completely, quilt
Push-and-pull to right side becomes right blackboard, and meanwhile, right blackboard is pulled or pushed to left side becomes left blackboard,
Continue for teacher's writing on the blackboard.
In actual application, the surrounding enviroment information storage that all kinds of onboard sensor perception obtain
In right of way radar map, Some vehicles oneself state parameter is obtained by sensors such as GPS,
Other parameters are obtained by automobile bus.
Current time t and the right of way radar map of a upper moment t-1 is represented respectively with left and right blackboard
With vehicle oneself state parameter, as shown in Figure 1 and Figure 2.
Left blackboard: deliver to the data in right of way radar map for recording various kinds of sensors, and
Comprise the driving condition parameter that intelligent vehicle is current, such as place street lane position, drive mould
Formula, speed, acceleration, direction, angular velocity, angular acceleration etc..Whom who comes is write, with
Time write at any time, information new on same grid covers old, sensor each other without
Need to synchronize.
Right blackboard: record a upper moment intelligent vehicle various kinds of sensors and deliver in right of way radar map
Data and vehicle driving state parameter, as place street lane position, driving model,
Speed, acceleration, direction, angular velocity, angular acceleration etc., within this decision-making period
Reading data, within this decision-making period, do not consider new sensing data.
Left and right blackboard differs one and controls the cycle.
Left and right blackboard push-and-pull: be responsible for the push-and-pull of left and right blackboard by decision-making device.Every time before decision-making
Once, right blackboard resets as left blackboard push-and-pull blackboard;Left blackboard becomes right blackboard, on it
Situation for the formation of decision parameters.If push-and-pull moment has sensor to send data,
Can lose.
Data message in the blackboard of left and right, available array structure correspondence therewith stores.
The present invention uses the technical scheme being similar to above-mentioned double blackboard mechanism to realize intelligent vehicle
Multi-sensor data exempt from synchronization process.
First data storage area is equivalent to above-mentioned left blackboard, and the second data storage area is suitable
In above-mentioned right blackboard.
Intelligent vehicle updates the driving behavior decision-making in each moment, control to control the cycle for frequency
Cycle processed can be A ms.The data of the first data storage area and the second data storage area
Data differ one and control the cycle.After every secondary control cycle starts, the second data storage area
Reset, replicate the data of the first data storage area in the second data storage area, now the
Data on two data storage areas are for the formation of intelligent vehicle driving behavior decision parameters.
The detailed description of the invention of the present invention is described in detail below in conjunction with preferred embodiment.
In intelligent vehicle driver's cabin, dead ahead is mounted with a PIKE100C video camera, is used for examining
Surveying and identify lane line, its recognition cycle is that (length of recognition cycle is with illumination etc. for 25-40ms
Environmental condition can change);Front bumper central authorities are mounted with a LUX4 laser radar,
In order to provide the obstacle distance in 110 degree of front angular field of view, 3.2 degree of elevation coverages and
Relative velocity, its recognition cycle is 80ms;Front bumper central authorities there is also mounted one
SICK291 laser radar, in order to the obstacle in providing front 40m, in the range of 180 degree
Thing distance and relative velocity, its recognition cycle is 80ms;GPS it is mounted with in the middle of roof
Equipment, in order to provide car speed and positional information, its process cycle is 50ms.It is visible,
The identification of each onboard sensor, process cycle are not quite similar.
In prior art, to realize the data syn-chronization of the sensor, technical sophistication,
Effect is poor;Utilize the method that the present invention provides, efficiently solve this problem.
The control cycle of vehicle is 100ms.
The right of way radar map information of the first data storage area real time record continuous renewal and intelligence
The oneself state parameter of car.These sensor senses data, whom who comes and writes, at any time with
Time write, the data that data cover new on same grid is old, between each sensor without with
Step.
After current control period starts, first by the zeros data in the second data storage area,
The data in described first data storage area of current time are copied to described second number again
According in memory block;Now record in the second data storage area is a upper moment intelligent vehicle
Data in sensor write right of way radar map and vehicle oneself state parameter, for this week
Reading data in phase, as the decision-making foundation of current intelligent vehicle driving behavior.
Store the data in described first data storage area being copied to described second data
During in district, data can not be write in described first data storage area, the most now pass
The data that sensor perception obtains can temporarily be lost.
The above, be only presently preferred embodiments of the present invention, not to the present invention's
Structure makees any pro forma restriction.Every technical spirit according to the present invention is to above real
Execute any simple modification, equivalent variations and modification that example is made, all still fall within the present invention's
In the range of technical scheme.For those skilled in the art, not
On the premise of departing from the principle of the invention, it is also possible to making some improvements and modifications, these change
Enter and retouch and also should be regarded as protection scope of the present invention.
The revised comment of preferred embodiment: understand to make to express the meaning the most burdensome, use one
The describing mode that those skilled in the art clearly understood that.
Claims (10)
1. a multi-sensor data processes and exempts from synchronous method, it is characterised in that described method includes following step
Rapid:
Onboard sensor gathers the surrounding enviroment information of intelligent vehicle current time;The described current time that will gather
Surrounding enviroment information write in real time in the right of way radar map of current time;
The vehicle oneself state parameter of the right of way radar map of described current time Yu current time is stored in first
In data storage area;
Data in described first data storage area are copied to institute by the zeros data in the second data storage area
State in the second data storage area, for the reading data in current decision cycle;
The right of way radar map of described first data storage area real-time update current time and vehicle oneself state ginseng
Number;
After the current decision cycle completes, the zeros data in described second data storage area, replicates after updating
Data in described first data storage area, to the second data storage area, enter next decision-making period.
Multi-sensor data the most according to claim 1 processes exempts from synchronous method, it is characterised in that institute
State the data real-time storage described in right of way radar map of current time in described first data storage area, institute
State the information that perception obtains by onboard sensor store in the right of way radar map of described current time without with
Step.
3. process according to the multi-sensor data described in claim 1 or 2 and exempt from synchronous method, its feature
It is, zeros data in described second data storage area, the data copied in described first data storage area
Time instant, can not write data, the information that now sensor senses obtains in described first data storage area
Lose.
Multi-sensor data the most according to claim 3 processes exempts from synchronous method, it is characterised in that the
In one data storage area, the formation time of data is the storages of adjacent twice duplication in the second data storage area the first data
The time interval of data in district.
Multi-sensor data the most according to claim 4 processes exempts from synchronous method, it is characterised in that institute
State the second data storage area can not carry out except by the zeros data in the second data storage area with by described first
Data in data storage area copy to other write operation in described second data storage area, do not consider working as
The sensor senses data that in front decision-making period, described first data storage area updates.
Multi-sensor data the most according to claim 5 processes exempts from synchronous method, it is characterised in that institute
State vehicle oneself state parameter, all include: place street lane position, driving model, speed, acceleration,
Direction, angular velocity, angular acceleration, described onboard sensor includes video camera, radar, GPS device.
Multi-sensor data the most according to claim 6 processes exempts from synchronous method, it is characterised in that institute
State right of way radar map and use structure of arrays storage with vehicle oneself state parameter information.
8. a multi-sensor data processes and exempts from synchronization system, it is characterised in that described system includes:
Onboard sensor, right of way radar map memory element, vehicle oneself state parameter storage unit, the first number
According to memory block, the second data storage area,
Described onboard sensor is for gathering the surrounding enviroment information of intelligent vehicle;
Described right of way radar map memory element is used for storing intelligent vehicle surrounding enviroment information;
Described vehicle oneself state parameter storage unit is for storing the vehicle oneself state parameter obtained in real time;
Described first data storage area is for storing the described right of way radar map of current time and the institute of current time
State vehicle oneself state parameter;
Described second data storage area is replicated, by described first data storage area, the data obtained for storage, and
Zeros data after the current decision cycle completes, replicates the data in described first data storage area after updating,
Enter next decision-making period.
A kind of multi-sensor data the most according to claim 8 processes exempts from synchronization system, it is characterised in that
Described vehicle oneself state parameter, including: place street lane position, driving model, speed, acceleration,
Direction, angular velocity, angular acceleration, described onboard sensor includes video camera, radar, GPS device.
The most according to Claim 8, a kind of multi-sensor data according to any one of or 9 process exempts to synchronize system
System, it is characterised in that described surrounding enviroment information and vehicle oneself state parameter information use structure of arrays to deposit
Storage, described second data storage area can not be carried out except by the zeros data in the second data storage area with by institute
State the data in the first data storage area and copy to other write operation in described second data storage area.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310740531.XA CN103679851B (en) | 2013-12-27 | 2013-12-27 | A kind of multi-sensor data processes exempts from synchronous method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310740531.XA CN103679851B (en) | 2013-12-27 | 2013-12-27 | A kind of multi-sensor data processes exempts from synchronous method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103679851A CN103679851A (en) | 2014-03-26 |
CN103679851B true CN103679851B (en) | 2016-08-24 |
Family
ID=50317287
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310740531.XA Active CN103679851B (en) | 2013-12-27 | 2013-12-27 | A kind of multi-sensor data processes exempts from synchronous method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103679851B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107819974B (en) * | 2016-09-13 | 2019-01-22 | 北京百度网讯科技有限公司 | Data capture method and device for automatic driving vehicle |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1209690C (en) * | 2000-10-16 | 2005-07-06 | 松下电器产业株式会社 | Automated guided vehicle, operation control system, and method for the same, and automotive vehicle |
US8355818B2 (en) * | 2009-09-03 | 2013-01-15 | Battelle Energy Alliance, Llc | Robots, systems, and methods for hazard evaluation and visualization |
US8126642B2 (en) * | 2008-10-24 | 2012-02-28 | Gray & Company, Inc. | Control and systems for autonomously driven vehicles |
JP5712097B2 (en) * | 2010-09-30 | 2015-05-07 | 本田技研工業株式会社 | Control device for autonomous vehicle |
CN103256939B (en) * | 2013-04-15 | 2015-09-23 | 李德毅 | Intelligent vehicle utilizes and becomes the method that granularity right of way radar map carries out information fusion |
-
2013
- 2013-12-27 CN CN201310740531.XA patent/CN103679851B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN103679851A (en) | 2014-03-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021212728A1 (en) | Unmanned vehicle lane changing decision-making method and system based on adversarial imitation learning | |
Chen et al. | Brain-inspired cognitive model with attention for self-driving cars | |
US11899748B2 (en) | System, method, and apparatus for a neural network model for a vehicle | |
CN108413975A (en) | Ground picture capturing method, system, cloud processor and vehicle | |
CN105939885B (en) | For making it possible to the device to Vehicular charging, method and computer readable storage medium | |
CN103853155B (en) | Intelligent vehicle road junction passing method and system | |
CN103699125B (en) | A kind of robot simulation air navigation aid based on the navigation of mouse cerebral hippocampal | |
CN109029463B (en) | Indoor multi-balance vehicle autonomous navigation and scheduling system for safe driving of vehicle | |
CN108646752A (en) | The control method and device of automated driving system | |
CN111752258A (en) | Operation test of autonomous vehicle | |
CN110532846A (en) | Automatic lane-change method, apparatus and storage medium | |
CN109976327A (en) | A kind of patrol robot | |
CN114282433A (en) | Automatic driving training method and system based on combination of simulation learning and reinforcement learning | |
CN106454108A (en) | Tracking shooting method and apparatus based on artificial intelligence, and electronic device | |
CN108052564A (en) | A kind of method and system for establishing unmanned database | |
CA3139421A1 (en) | Automatic annotation of object trajectories in multiple dimensions | |
CN103679851B (en) | A kind of multi-sensor data processes exempts from synchronous method and system | |
CN203338429U (en) | Driving recording device capable of performing lane shift and vehicle distance prompting | |
Gajjar et al. | A comprehensive study on lane detecting autonomous car using computer vision | |
CN110428693A (en) | User driving habits Training Methodology, training module, mobile unit and storage medium | |
Xia et al. | CyberC3: A prototype cybernetic transportation system for urban applications | |
CN115202479A (en) | Man-machine co-driving simulation system of man-in-loop and application thereof | |
CN114882759A (en) | Virtual-real hybrid integrated simulation intelligent ship multi-channel interactive simulation system and method | |
CN114120653A (en) | Centralized vehicle group decision control method and device and electronic equipment | |
Lam | A methodology for the optimization of autonomous public transportation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
CB03 | Change of inventor or designer information |
Inventor after: Li Deyi Inventor after: Zheng Siyi Inventor after: Jia Peng Inventor after: Liu Yuchao Inventor after: Guo Mu Inventor before: Li Deyi Inventor before: Jia Peng Inventor before: Fu Ying Inventor before: Huang Liwei |
|
COR | Change of bibliographic data | ||
GR01 | Patent grant |