CN109094561A - A kind of vehicle cruise system and method - Google Patents

A kind of vehicle cruise system and method Download PDF

Info

Publication number
CN109094561A
CN109094561A CN201810910364.1A CN201810910364A CN109094561A CN 109094561 A CN109094561 A CN 109094561A CN 201810910364 A CN201810910364 A CN 201810910364A CN 109094561 A CN109094561 A CN 109094561A
Authority
CN
China
Prior art keywords
vehicle
data
kinematic parameter
marker
distance
Prior art date
Application number
CN201810910364.1A
Other languages
Chinese (zh)
Inventor
易德威
易洪雷
Original Assignee
嘉兴学院
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 嘉兴学院 filed Critical 嘉兴学院
Priority to CN201810910364.1A priority Critical patent/CN109094561A/en
Publication of CN109094561A publication Critical patent/CN109094561A/en

Links

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
    • B60W30/14Adaptive cruise control
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
    • B60W30/08Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
    • B60W30/14Adaptive cruise control
    • B60W30/16Control of distance between vehicles, e.g. keeping a distance to preceding vehicle
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2520/00Input parameters relating to overall vehicle dynamics
    • B60W2520/10Longitudinal speed
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00Input parameters relating to objects

Abstract

The invention discloses a kind of vehicle cruise system and methods, are related to field of vehicle safety.A kind of vehicle cruise system provided by the invention and method, environment is perceived using multisensor, and the information of different sensors is characterized again to realize high-grade nonlinear data fusion, carry out aid decision using the data building part 3D map of fusion, system can be improved, depth perception is carried out to driving environment, realize it is good adaptively and decision;In addition, the present invention can identifying to barrier classification, more effective decision feedback can be made for the higher barrier of degree of danger, improve the decision real-time and flexibility of vehicle cruise system.Vehicle cruise system provided by the invention and method are stronger to the synthesis sensing capability of obstacles around the vehicle, stronger to the strain flexibility of barrier, and anti-collision technique function diversification is highly suitable for road conditions complicated and changeable.

Description

A kind of vehicle cruise system and method

Technical field

The present invention relates to field of vehicle safety, in particular to a kind of vehicle cruise system and method.

Background technique

As the rapid development and automobile volume of production and marketing of domestic automobile industry rise year by year, traffic pressure is growing day by day, road Transportation safety form is increasingly serious.It is reported according to " People's Republic of China's national economy and social development statistical communique in 2016 ", Road traffic accident every ten thousand vehicles death toll in China's is 2.1 people.Especially big pernicious traffic accident takes place frequently, and not only seriously endangers the people's The security of the lives and property, and easily cause complicated social concern, for this purpose, active safety technologies are as the weight for solving traffic problems Means are wanted, by the great attention of national governments, become in China major fields of long-term scientific and technological development planning and excellent The technological development direction first subsidized.

According to " People's Republic of China's road traffic accident in 2016 counts annual report ", traffic accident is mainly by driving The artificial fault of member causes.Since driver's ability all has certain limitation and unstability, for example, reaction time, reaction Speed, attention intensity, spiritual endurance and driving experience etc. easily cause driver behavior when driver is in a state of fatigue Fault, or even serious traffic accident is caused, to solve this problem, intelligent cruise system is come into being, and has been obtained each The positive regard of auto vendor and user.

In general, the intelligent cruise system of vehicle loading includes environmental perception module and collision prevention of vehicle module, collision prevention of vehicle mould Block is previously stored with vehicle development personnel or vehicle user according to the environmental data of the collision avoidance experience setting in vehicle operation With the corresponding relationship of mapping movement, wherein environmental data may include the position data of barrier, and mapping acts the side of may include Default rotation angle etc. is rotated along preset rotation direction to disk.Environmental perception module can have barrier in front of the traveling of vehicle When hindering object, the environmental data of vehicle current environment is obtained by trailer-mounted radar, collision prevention of vehicle module can be worked as according to vehicle Preceding environmental data matches the mapping of target corresponding to the environmental data from the corresponding relationship that environmental data and mapping act Movement maps action control vehicle driving according to the target, so that vehicle be avoided to collide with front obstacle.

In the implementation of the present invention, inventor find the relevant technologies the prior art has at least the following problems:

Intelligent cruise system in the related technology realizes the spy to the speed and distance of vehicle front barrier by radar mostly Survey, and can not effective cognitive disorders object type, and, intelligent cruise system can not perceive and obtain the ring of vehicle two sides and rear side Border data, therefore, intelligent cruise system are lower to the synthesis sensing capability of obstacles around the vehicle, flexible to the strain of barrier Property is poor, and anti-collision technique function is relatively simple.

Summary of the invention

In view of the above-mentioned problems existing in the prior art, the present invention provides a kind of vehicle cruise system and methods.

First aspect according to an embodiment of the present invention provides a kind of vehicle cruise system, the vehicle cruise system packet Include: environmental perception module, data fusion module, adaptively with decision-making module and execution module;

The environmental perception module is used for when vehicle is in driving status, is obtained the every of the vehicle side in real time and is driven ring Border data, and every driving environment data are pre-processed;

The data fusion module is used to carry out weight to by the pretreated every driving environment data of the environmental perception module Define data indicate, and by redefine data expression after every driving environment data nonlinear data is carried out by particle filter Fusion constructs part 3D map according to every driving environment data after non-linear fusion, and the part 3D map includes at least One marker;

Each mark in the part 3D map for being adaptively used to obtain the data fusion module building with decision-making module Show the position data and kinematic parameter of object;According to the position of the position data of the vehicle and kinematic parameter and each marker Data and kinematic parameter determine the barriers at different levels for meeting preset rules from each marker;According to barriers at different levels The position data and kinematic parameter of position data and kinematic parameter and the vehicle are obtained from experience driving data library and are moved Parameter adjusts data, and experience driving data library has recorded the pre- barrier phases at different levels for first passing through adaptive learning algorithm and determining To at least a pair of of the positional relationship of vehicle, the kinematic parameter of barrier, the kinematic parameter of vehicle and kinematic parameter adjustment data Corresponding relationship;

The execution module is used for according to the kinematic parameter adjustment data adaptively obtained with decision-making module to the vehicle Driving condition be adjusted.

Optionally, the kinematic parameter includes the average speed and directional velocity in preset time range, described adaptive It is also used to decision-making module:

Institute is calculated separately according to the position data of the vehicle and the position data of the marker for each marker State vehicle and current mapping distance L of the marker on traffic directiony1With perpendicular to the current mapping on traffic direction Distance Lx1

According to the current mapping distance Ly1, the current mapping distance Lx1, vehicle kinematic parameter and the marker Kinematic parameter calculates separately the expected mapping after preset duration between the vehicle and the marker on traffic direction Distance Ly2With perpendicular to the expection mapping distance L on traffic directionx2

When detecting the expected mapping distance Lx2Greater than preset exposure apart from when, determine the marker be non-barrier;When Detect the expected mapping distance Lx2No more than preset exposure distance, the expected mapping distance Ly2Greater than default safe police Guard against apart from when, determine the marker be non-barrier;When detecting the expected mapping distance Lx2No more than preset exposure away from From the expected mapping distance Ly2When no more than default Security alert distance and being greater than default safety threshold, described in determination Marker is level-one barrier;When detecting the expected mapping distance Lx2No more than preset exposure distance, the expected mapping Distance Ly2When no more than default safety threshold, determine that the marker is second level barrier.

Optionally, described to be adaptively also used to decision-making module:

When detect not there is no barrier when, keep the driving condition of the vehicle;

When detecting there is only when level-one barrier, danger early warning signal is sent to the execution module, by the execution module Early warning feedback is made according to the danger early warning signal;

When detecting the presence of second level barrier, the position data and kinematic parameter according to barriers at different levels is executed, and The position data and kinematic parameter of the vehicle obtain the step of kinematic parameter adjusts data from experience driving data library.

Optionally, described to be adaptively also used to decision-making module:

The kinematic parameter that data calculate the vehicle after adjustment is adjusted according to the kinematic parameter;

For each marker, according to the current mapping distance Ly1, the current mapping distance Lx1, the vehicle after adjustment Kinematic parameter and the marker kinematic parameter, calculate separately the vehicle and the marker after preset duration Between expectation mapping distance L on traffic directiony3With perpendicular to the expectation mapping distance L on traffic directionx3

When detecting the expectation mapping distance Ly3Greater than the expected mapping distance Ly2When, the kinematic parameter is adjusted into number According to the execution module is sent to, data are adjusted to the driving shape of the vehicle according to the kinematic parameter by the execution module State is adjusted;

When detecting the expectation mapping distance Ly3No more than the expected mapping distance Ly2When, by preset order from the warp It tests driving data library and reacquires the kinematic parameter adjustment data for meeting corresponding relationship, execute described according to the kinematic parameter tune After entire data calculating adjustment the step of the kinematic parameter of the vehicle.

Optionally, the environmental perception module includes visual sensor, laser radar sensor and alignment sensor, described Environmental perception module is also used to:

The every driving environment data for obtaining the vehicle side in real time pass through vehicle described in the visual sensor captured in real-time The traveling image of front side carries out image segmentation and mark to the traveling image using depth convolution coder structure SegNet Object identification;

The every driving environment data for obtaining the vehicle side in real time obtain the vehicle by the laser radar sensor The number of scan points evidence of side object generates 3D cloud atlas by the location registration process to the number of scan points evidence;

The every driving environment data for obtaining the vehicle side in real time record two maintenance and operations of the vehicle by alignment sensor Dynamic rail mark is simultaneously shown in plane map;

At this point, the items driving environment data include that the traveling image of the vehicle front side, vehicle side object are swept Described point data, the two dimensional motion track of the vehicle and plane map.

The second aspect according to an embodiment of the present invention provides a kind of vehicle cruise method, which is characterized in that the method Include:

When vehicle is in driving status, every driving environment data of the vehicle side are obtained in real time, and drive to items Environmental data is pre-processed;

Redefine data to every driving environment data indicates, and will redefine every driving environment number after data expression Nonlinear data fusion is carried out according to by particle filter, constructs part 3D according to every driving environment data after non-linear fusion Map, the part 3D map includes at least one marker;

Obtain the position data and kinematic parameter of each marker in the part 3D map;According to the position data of the vehicle With the position data and kinematic parameter of kinematic parameter and each marker, determine to meet preset rules from each marker Barriers at different levels;According to the position data and movement of the position data of barriers at different levels and kinematic parameter and the vehicle Parameter obtains kinematic parameter from experience driving data library and adjusts data, and experience driving data library has recorded pre- first pass through certainly The movement ginseng of the positional relationship of barrier relative vehicles at different levels, the kinematic parameter of barrier, vehicle that adaptive learning algorithm determines Several at least a pair of of corresponding relationships with kinematic parameter adjustment data;

Data are adjusted according to the kinematic parameter to be adjusted the driving condition of the vehicle.

Optionally, the kinematic parameter includes the average speed and directional velocity in preset time range, described according to institute The position data of vehicle and the position data and kinematic parameter of kinematic parameter and each marker are stated, from each marker really Make the barriers at different levels for meeting preset rules, comprising:

Institute is calculated separately according to the position data of the vehicle and the position data of the marker for each marker State vehicle and current mapping distance L of the marker on traffic directiony1With perpendicular to the current mapping on traffic direction Distance Lx1

According to the current mapping distance Ly1, the current mapping distance Lx1, vehicle kinematic parameter and the marker Kinematic parameter calculates separately the expected mapping after preset duration between the vehicle and the marker on traffic direction Distance Ly2With perpendicular to the expection mapping distance L on traffic directionx2

When detecting the expected mapping distance Lx2Greater than preset exposure apart from when, determine the marker be non-barrier;When Detect the expected mapping distance Lx2No more than preset exposure distance, the expected mapping distance Ly2Greater than default safe police Guard against apart from when, determine the marker be non-barrier;When detecting the expected mapping distance Lx2No more than preset exposure away from From the expected mapping distance Ly2When no more than default Security alert distance and being greater than default safety threshold, described in determination Marker is level-one barrier;When detecting the expected mapping distance Lx2No more than preset exposure distance, the expected mapping Distance Ly2When no more than default safety threshold, determine that the marker is second level barrier.

Optionally, the method, further includes:

When detect not there is no barrier when, keep the driving condition of the vehicle;

When detecting there is only when level-one barrier, danger early warning signal is sent, early warning is made according to the danger early warning signal Feedback;

When detecting the presence of second level barrier, the position data and kinematic parameter according to barriers at different levels is executed, and The position data and kinematic parameter of the vehicle obtain the step of kinematic parameter adjusts data from experience driving data library.

Optionally, the method, further includes:

The kinematic parameter that data calculate the vehicle after adjustment is adjusted according to the kinematic parameter;

For each marker, according to the current mapping distance Ly1, the current mapping distance Lx1, the vehicle after adjustment Kinematic parameter and the marker kinematic parameter, calculate separately the vehicle and the marker after preset duration Between expectation mapping distance L on traffic directiony3With perpendicular to the expectation mapping distance L on traffic directionx3

When detecting the expectation mapping distance Ly3Greater than the expected mapping distance Ly2When, it is adjusted according to the kinematic parameter Data are adjusted the driving condition of the vehicle;

When detecting the expectation mapping distance Ly3No more than the expected mapping distance Ly2When, by preset order from the warp It tests driving data library and reacquires the kinematic parameter adjustment data for meeting corresponding relationship, execute described according to the kinematic parameter tune After entire data calculating adjustment the step of the kinematic parameter of the vehicle.

Optionally, the every driving environment data for obtaining the vehicle side in real time, and to every driving environment number According to being pre-processed, comprising:

In real time by the traveling image of vehicle front side described in visual sensor captured in real-time, depth convolution coder structure is used SegNet carries out image segmentation to the traveling image and marker identifies;

The every driving environment data for obtaining the vehicle side in real time obtain the vehicle side by laser radar sensor The number of scan points evidence of object generates 3D cloud atlas by the location registration process to the number of scan points evidence;

The every driving environment data for obtaining the vehicle side in real time record two maintenance and operations of the vehicle by alignment sensor Dynamic rail mark is simultaneously shown in plane map;

At this point, the items driving environment data include that the traveling image of the vehicle front side, vehicle side object are swept Described point data, the two dimensional motion track of the vehicle and plane map.

In conclusion a kind of vehicle cruise system provided by the invention and method, feel environment using multisensor Know, and the information of different sensors is characterized again to realize high-grade nonlinear data fusion, uses fusion Data building part 3D map carrys out auxiliary system decision, and system can be improved and carry out depth perception to driving environment, realize good Adaptively and decision;In addition, the present invention can identifying to barrier classification, can do for the higher barrier of degree of danger More effective decision feedback out improves the decision real-time and flexibility of vehicle cruise system.Vehicle cruise provided by the invention System and method is stronger to the synthesis sensing capability of obstacles around the vehicle, stronger to the strain flexibility of barrier, anticollision skill Art function diversification, is highly suitable for road conditions complicated and changeable.

Detailed description of the invention

The drawings herein are incorporated into the specification and forms part of this specification, and shows and meets implementation of the invention Example, and in specification together principle for explaining the present invention.

Fig. 1 is a kind of structural schematic diagram of implementation environment involved in the drive manner of each embodiment offer of the present invention.

Fig. 2 is the system block diagram of vehicle cruise system provided by one embodiment of the present invention.

Fig. 3 is a kind of method flow diagram of vehicle cruise method provided by one embodiment of the present invention.

Fig. 4 is the method flow diagram of another vehicle cruise method provided by one embodiment of the present invention.

Fig. 5 is a kind of applicating flow chart of nonlinear data blending algorithm provided by one embodiment of the present invention.

Specific embodiment

To make the objectives, technical solutions, and advantages of the present invention clearer, (but unlimited below in conjunction with specific embodiment In illustrated embodiment) and attached drawing the present invention is described in detail, it is clear that described embodiment, which is only that the present invention is a part of, to be implemented Example, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creativeness All other embodiment obtained, shall fall within the protection scope of the present invention under the premise of labour.

Referring to FIG. 1, a kind of implementation environment involved in the drive manner provided it illustrates each embodiment of the present invention Structural schematic diagram, referring to Fig. 1, road is divided into tri- lanes A, B, C, and vehicle a is travelled on the B of lane, and vehicle b is in lane B row The front of vehicle a is sailed and is located at, vehicle c travels and be located at the left front of vehicle a in lane A, and vehicle d is in lane C traveling and position In the right front of vehicle a, vehicle e is travelled in lane B and is located at the rear of vehicle a.

During traveling, especially the high section of high speed carries out when driving, generally requiring to protect with front truck vehicle b vehicle a Pre-determined distance is held, when which is that emergency occurs for vehicle a, emergency braking measure is taken to carry out deceleration to static Without the safe distance to collide with front truck vehicle b.

Under normally travel, vehicle a driver only needs the holding safe distance of concern vehicle a and vehicle b, still, when When vehicle b and vehicle the e unexpected speed change of vehicle or lane B are inserted in the unexpected lane change of vehicle d of the vehicle c or lane C of lane A, vehicle a The distance between vehicle b, c, d, e are likely lower than safe distance in a short time, at this point, vehicle a be easy to vehicle b, c, D, e collides, and then causes serious traffic accident, brings serious harm to driver and its household.

In this implementation environment, vehicle a can be mounted with vehicle cruise system provided by the invention, and vehicle cruise system can be with The position data and kinematic parameter of real-time monitoring vehicle a surrounding vehicles b, c, d, e, and detect vehicle b, c, d, e have speed change or During person plugs in vehicle intention, vehicle cruise system detects vehicle b, c, d, e and the collision of vehicle a is possible, and is detecting vehicle A and vehicle b, c, d, e exist collision may when, deceleration, acceleration or brake measure are taken to vehicle a, to avoid traffic thing Therefore generation.

Referring to FIG. 2, it illustrates the system block diagram of vehicle cruise system 200 provided by one embodiment of the present invention, it should Vehicle cruise system 200 drives for vehicle cruise and collision avoidance.Referring to fig. 2, which includes: environment sensing Module 210, data fusion module 220, adaptively with decision-making module 230 and execution module 240.

The environmental perception module 210 is used for when vehicle is in driving status, obtains each of the vehicle side in real time Item driving environment data, and every driving environment data are pre-processed.

The data fusion module 220 is used for by the pretreated every driving environment of the environmental perception module 210 Data redefine data expression, and pass through particle filter progress for every driving environment data after data expression are redefined Nonlinear data fusion constructs part 3D map, the part 3D according to every driving environment data after non-linear fusion Figure includes at least one marker.

The part 3D for being adaptively used to obtain the building of data fusion module 220 with decision-making module 230 The position data and kinematic parameter of each marker in figure;According to the position data of the vehicle and kinematic parameter and each mark The position data and kinematic parameter for showing object determine the barriers at different levels for meeting preset rules from each marker;According to each The grade position data of barrier and the position data and kinematic parameter of kinematic parameter and the vehicle, from experience driving data Library obtain kinematic parameter adjust data, experience driving data library have recorded it is pre- first pass through adaptive learning algorithm determine it is each The grade positional relationship of barrier relative vehicle, the kinematic parameter of barrier, the kinematic parameter of vehicle and kinematic parameter adjust data At least a pair of of corresponding relationship.

The execution module 240 is used to adjust data according to the kinematic parameter adaptively obtained with decision-making module 230 The driving condition of the vehicle is adjusted.

Optionally, the kinematic parameter includes the average speed and directional velocity in preset time range, described adaptive It is also used to decision-making module 230:

Institute is calculated separately according to the position data of the vehicle and the position data of the marker for each marker State vehicle and current mapping distance L of the marker on traffic directiony1With perpendicular to the current mapping on traffic direction Distance Lx1

According to the current mapping distance Ly1, the current mapping distance Lx1, vehicle kinematic parameter and the marker Kinematic parameter calculates separately the expected mapping after preset duration between the vehicle and the marker on traffic direction Distance Ly2With perpendicular to the expection mapping distance L on traffic directionx2

When detecting the expected mapping distance Lx2Greater than preset exposure apart from when, determine the marker be non-barrier;When Detect the expected mapping distance Lx2No more than preset exposure distance, the expected mapping distance Ly2Greater than default safe police Guard against apart from when, determine the marker be non-barrier;When detecting the expected mapping distance Lx2No more than preset exposure away from From the expected mapping distance Ly2When no more than default Security alert distance and being greater than default safety threshold, described in determination Marker is level-one barrier;When detecting the expected mapping distance Lx2No more than preset exposure distance, the expected mapping Distance Ly2When no more than default safety threshold, determine that the marker is second level barrier.

Optionally, described to be adaptively also used to decision-making module 230:

When detect not there is no barrier when, keep the driving condition of the vehicle;

When detecting there is only when level-one barrier, danger early warning signal is sent to the execution module 240, by the execution mould Block 240 makes early warning feedback according to the danger early warning signal;

When detecting the presence of second level barrier, the position data and kinematic parameter according to barriers at different levels is executed, and The position data and kinematic parameter of the vehicle obtain the step of kinematic parameter adjusts data from experience driving data library.

Optionally, described to be adaptively also used to decision-making module 230:

The kinematic parameter that data calculate the vehicle after adjustment is adjusted according to the kinematic parameter;

For each marker, according to the current mapping distance Ly1, the current mapping distance Lx1, the vehicle after adjustment Kinematic parameter and the marker kinematic parameter, calculate separately the vehicle and the marker after preset duration Between expectation mapping distance L on traffic directiony3With perpendicular to the expectation mapping distance L on traffic directionx3

When detecting the expectation mapping distance Ly3Greater than the expected mapping distance Ly2When, the kinematic parameter is adjusted into number According to the execution module is sent to, data is adjusted according to the kinematic parameter by the execution module 240, the vehicle is driven The state of sailing is adjusted;

When detecting the expectation mapping distance Ly3No more than the expected mapping distance Ly2When, by preset order from the warp It tests driving data library and reacquires the kinematic parameter adjustment data for meeting corresponding relationship, execute described according to the kinematic parameter tune After entire data calculating adjustment the step of the kinematic parameter of the vehicle.

Optionally, the environmental perception module 210 includes visual sensor, laser radar sensor and alignment sensor, The environmental perception module 210 is also used to:

The every driving environment data for obtaining the vehicle side in real time pass through vehicle described in the visual sensor captured in real-time The traveling image of front side carries out image segmentation and mark to the traveling image using depth convolution coder structure SegNet Object identification;

The every driving environment data for obtaining the vehicle side in real time obtain the vehicle by the laser radar sensor The number of scan points evidence of side object generates 3D cloud atlas by the location registration process to the number of scan points evidence;

The every driving environment data for obtaining the vehicle side in real time record two maintenance and operations of the vehicle by alignment sensor Dynamic rail mark is simultaneously shown in plane map;

At this point, the items driving environment data include that the traveling image of the vehicle front side, vehicle side object are swept Described point data, the two dimensional motion track of the vehicle and plane map.

In conclusion vehicle cruise system provided in an embodiment of the present invention includes: environmental perception module, data fusion mould Block, adaptively with decision-making module and execution module, environment is perceived using multisensor, and to the letter of different sensors Breath is characterized again to realize high-grade nonlinear data fusion, is come using the data building part 3D map of fusion auxiliary Help decision, system can be improved, depth perception is carried out to driving environment, realize it is good adaptively and decision;In addition, the present invention can To identifying for barrier classification, more effective decision feedback can be made for the higher barrier of degree of danger, improves vehicle The decision real-time and flexibility of cruise system.Vehicle cruise system provided by the invention and method are to obstacles around the vehicle Synthesis sensing capability it is stronger, stronger to the strain flexibility of barrier, anti-collision technique function diversification is highly suitable for complexity Changeable road conditions.

It should be noted that vehicle cruise method provided in an embodiment of the present invention can be applied to following vehicle cruise sides Method, vehicle cruise method may refer to the description of hereafter each embodiment in the embodiment of the present invention.

Referring to FIG. 3, it illustrates a kind of method flow diagram of vehicle cruise method provided by one embodiment of the present invention, This method can vehicle cruise system as shown in Figure 2 execute, referring to Fig. 3, which may include walking as follows It is rapid:

In step 301, when vehicle is in driving status, every driving environment data of the vehicle side are obtained in real time, And every driving environment data are pre-processed.

In step 302, redefine data to every driving environment data indicates, and after redefining data expression Every driving environment data nonlinear data fusion is carried out by particle filter, ring is driven according to every after non-linear fusion Border data construct part 3D map, and the part 3D map includes at least one marker.

In step 303, the position data and kinematic parameter of each marker in the part 3D map are obtained.

In step 304, according to the position data of the position data of the vehicle and kinematic parameter and each marker And kinematic parameter, the barriers at different levels for meeting preset rules are determined from each marker.

In step 305, according to the position data of the position data of barriers at different levels and kinematic parameter and the vehicle And kinematic parameter, kinematic parameter, which is obtained, from experience driving data library adjusts data.

Experience driving data library has recorded the pre- determining barriers at different levels of adaptive learning algorithm that first pass through with respect to vehicle Positional relationship, kinematic parameter and the kinematic parameter adjustment data of the kinematic parameter of barrier, vehicle it is at least a pair of corresponding Relationship.

Within step 306, data are adjusted according to the kinematic parameter to be adjusted the driving condition of the vehicle.

In conclusion vehicle cruise method provided in an embodiment of the present invention, perceives environment using multisensor, and The information of different sensors is characterized to realize high-grade nonlinear data fusion again, uses the data structure of fusion Portion's 3D map of founding the bureau carrys out aid decision, can be improved system to driving environment carry out depth perception, realize it is good adaptively and certainly Plan;In addition, the present invention can identifying to barrier classification, can make more effectively for the higher barrier of degree of danger Decision feedback improves the decision real-time and flexibility of vehicle cruise system.Vehicle cruise system provided by the invention and method Stronger to the synthesis sensing capability of obstacles around the vehicle, stronger to the strain flexibility of barrier, anti-collision technique function is polynary Change, is highly suitable for road conditions complicated and changeable.

Referring to FIG. 4, it illustrates the method flows of another vehicle cruise method provided by one embodiment of the present invention Figure, this method can vehicle cruise system as shown in Figure 2 execute, referring to fig. 4, which may include as follows Step:

In step 401, when vehicle is in driving status, every driving environment data of vehicle side are obtained in real time, and right Every driving environment data are pre-processed.

Environmental perception module can obtain the every of the vehicle side in real time and drive ring when vehicle is in driving status Border data.In a kind of possible situation, environmental perception module may include visual sensor, laser radar sensor and positioning Sensor at this point, obtaining every driving environment data of the vehicle side in real time, and carries out every driving environment data pre- The method of processing may include:

(a) before obtaining every driving environment data of the vehicle side in real time by vehicle described in visual sensor captured in real-time The traveling image of side carries out image segmentation and marker to the traveling image using depth convolution coder structure SegNet Identification.

(b) the every driving environment data for obtaining the vehicle side in real time obtain the vehicle by laser radar sensor The number of scan points evidence of side object passes through the location registration process to the number of scan points evidence and generates 3D cloud atlas.

(c) the every driving environment data for obtaining the vehicle side in real time record the vehicle by alignment sensor Two dimensional motion track is simultaneously shown in plane map.

At this point, every driving environment data include that the traveling image of the vehicle front side, vehicle side object are swept Described point data, the two dimensional motion track of the vehicle and plane map.

Wherein, visual sensor can be vehicle-mounted camera, and alignment sensor can be GPS.

It should be noted that collaborative work of the present invention by multisensor, it can be within the scope of vehicle periphery pre-determined distance Object identified that the object includes vehicle, communal facility.

Further, the present invention can also identify the type of vehicle.

In step 402, redefine data to every driving environment data indicates, and after redefining data expression Every driving environment data nonlinear data fusion is carried out by particle filter, ring is driven according to every after non-linear fusion Border data construct part 3D map.

Part 3D map includes at least one marker.Each marker may include automobile, bicycle, electric vehicle Deng.

Since each sensor that environmental perception module includes is different the expression of same target, for example, laser thunder What it is up to sensor acquisition is three-dimensional point cloud chart, and what alignment sensor obtained is two dimensional motion track and map, and visual sensor is received Collection is image information, the driving environment data progress that therefore, it is necessary to be collected into using data fusion module to different sensors It characterizes again, so that unified to the same intermediate scheme get off that vehicle cruise system is assisted to be analyzed.Data fusion module is being incited somebody to action After the driving environment data of different sensors acquisition carry out unified characterization again, passes through the data after particle filter counterweight characterizes and carry out Data fusion simultaneously constructs accurately part 3D map.

In one possible implementation, the nonlinear data fusion method that the present invention uses is based on Markov and covers spy Carlow theoretical method (Markov Chain Monte Carlo) spreads point, adjusting particle weights and resampling by particle Mode realizes the non-linear fusion of multi-sensor data.

For the ease of explaining nonlinear data blending algorithm provided by the invention, it is referred to non-as shown in formula (1) Linear multisensor syste model:

Formula (1)

In the model,,It is the total number of sensor;It is time index, represents the time Index;It is sensor index, representsA sensor;For state variable;For measurand;It is process Noise andIt is to measure noise, mean value zero, and it is mutually indepedent;It is state transfer equation, passes throughState Variable derivesState variable;It is observational equation.According to all measured values,It can be come out by recursive estimation, it is such as public Shown in formula (2):

Formula (2)

In the nonlinear data blending algorithm, shown in posterior probability density function such as formula (3):

Formula (3)

It is with weight?A particle.Since multisensor measures the mutual independently working in actual measurement, Shown in likelihood function such as formula (4):

Formula (4)

In each round iteration, the weight of particle will be adjusted according to such as formula (5):

Formula (5)

In formula (5),It is probability density function, for generating particle.Because optional sampling is distributed It is difficult to obtain, priori density functionIt is taken as posterior probability density function, so that weight calculation formula Transform to the form as shown in formula (6):

Formula (6)

Wherein, above-mentioned nonlinear data blending algorithm is in vehicle cruise system provided by the invention and the applicating flow chart of method As shown in Figure 5.

In step 403, the position data and kinematic parameter of each marker in the part 3D map are obtained.

Wherein, kinematic parameter includes the average speed and directional velocity in preset time range, the preset time range Value is preset by developer, and is stored in adaptively and decision-making module.For example, the value of preset time range is set as 0.1 Second.

It should be noted that the positional number of each marker is adaptively obtained by the part 3D map with decision-making module When according to kinematic parameter, the position data and kinematic parameter of vehicle itself are gone back while obtained.

Illustratively, in a kind of possible mode, the position data of the vehicle adaptively obtained with decision-making module and institute The position data for stating marker is indicated in a manner of coordinate, using user's vehicle a1 as origin, the operation of user's vehicle a1 Coordinate system is established as positive direction of the x-axis as positive direction of the y-axis, perpendicular to the right side method of user's vehicle a1 traffic direction in direction, Each unit distance is 1 meter corresponding, then adaptively obtains user's vehicle a1 with decision-making module and each mark article coordinate can be such as table one It is shown:

Vehicle code name Coordinate a1 (0,0) a2 (0,15) a3 (2,12) a4 (- 2,18)

Table one

Illustratively, in a kind of possible mode, the kinematic parameter and the mark of the vehicle adaptively obtained with decision-making module The kinematic parameter for showing object may include the average speed and directional velocity of vehicle or marker in 0.1 second, as shown in Table 2:

Vehicle code name Kinematic parameter (average speed, directional velocity) a1 (20.00m/s, front) a2 (15.00 m/s, front) a3 (20.10 m/s, just 5.71 degree to the left first) a4 (20.02m/s, just 2.86 degree to the right first)

Table two

In step 404, each marker is counted respectively according to the position data of vehicle and the position data of marker Calculate vehicle and current mapping distance L of the marker on traffic directiony1With perpendicular to the current mapping on traffic direction Distance Lx1

The traffic direction refers to the traffic direction of the vehicle.

When the position data of the vehicle adaptively obtained with decision-making module and the position data of the marker are with above-mentioned When the mode of coordinate is indicated, if the position data of vehicle is (x1, y1), the position data of marker is (x2, y2), then vehicle And current mapping distance L of the marker on traffic directiony1For the absolute value of y2 and y1 difference, i.e., | y2-y1 |, vehicle and mark Show object perpendicular to the current mapping distance L on traffic directionx1For the absolute value of x2 and x1 difference, i.e., | x2-x1 |.

It is illustrated with the data shown in one example of table, the position data that vehicle a1 is adaptively calculated with decision-making module is (0,0), the position data of vehicle a2 are (0,15), and the position data of vehicle a3 is (2,12), the position data of vehicle a4 be (- 2,18) vehicle a1 and current mapping distance L of the marker a2 on traffic direction, is then calculatedy1For 15m, perpendicular to fortune Current mapping distance L on line directionx1It is 0;Be calculated current on traffic direction of vehicle a1 and marker a3 map away from From Ly1For 12m, perpendicular to the current mapping distance L on traffic directionx1For 2m;Vehicle a1 is calculated and marker a4 exists Current mapping distance L on traffic directiony1For 18m, perpendicular to the current mapping distance L on traffic directionx1For 2m.

In step 405, according to current mapping distance Ly1, current mapping distance Lx1, vehicle kinematic parameter and mark The kinematic parameter of object is calculated separately by the expection mapping distance between preset duration rear vehicle and marker on traffic direction Ly2With perpendicular to the expection mapping distance L on traffic directionx2

It, adaptively can kinematic parameter according to vehicle and each mark with decision-making module under a kind of possible mode The speed difference v on traffic direction is calculated between vehicle and marker in the kinematic parameter of objectyAnd vehicle and marker Between perpendicular to the speed difference v on traffic directionx, then calculate speed difference v after preset duration tyOn traffic direction away from From change value LyAnd speed difference v after preset duration txPerpendicular on traffic direction apart from change value Lx, last root According to this apart from change value LyWith apart from change value LxIt is calculated after preset duration between the vehicle and the marker Expection mapping distance L on traffic directiony2With perpendicular to the expection mapping distance L on traffic directionx2

For example, be illustrated with the data shown in two example of table one and table, if preset duration is 1s, then adaptively and decision Module is calculated between vehicle a1 and marker a2 according to the kinematic parameter of vehicle a1 and the kinematic parameter of marker a2 Speed difference v on traffic directionyFor 20-15=5m/s, perpendicular to the speed difference v on traffic directionxFor 0-0=0m/s, in advance If speed difference v after duration 1syOn traffic direction apart from change value LyIt is then 5 × 1=5m, on perpendicular to traffic direction Apart from change value LxIt is then 0 × 1=0m, is calculated between preset duration 1s rear vehicle a1 and marker a2 in operation side Upward expection mapping distance Ly2For 15-5=10m, perpendicular to the expection mapping distance L on traffic directionx2For 0-0=0m.

Similarly, it is adaptively calculated with decision-making module between preset duration 1s rear vehicle a1 and marker a3 in traffic direction On speed difference vyFor 20-20=0m/s, perpendicular to the speed difference v on traffic directionxFor 2-0=2m/s, preset duration 1s Speed difference v afterwardsyOn traffic direction apart from change value LyIt is then 0 × 1=0m, changes perpendicular to the distance on traffic direction Variate LxIt is then 2 × 1=2m, is calculated pre- on traffic direction between preset duration 1s rear vehicle a1 and marker a3 Phase mapping distance Ly2For 12-0=12m, perpendicular to the expection mapping distance L on traffic directionx2For 2-2=0m;It calculates default Speed difference v between duration 1s rear vehicle a1 and marker a4 on traffic directionyFor 20-20=0m/s, perpendicular to operation Speed difference v on directionxFor 1-0=1m/s, speed difference v after preset duration 1syOn traffic direction apart from change value LyThen be 0 × 1=0m, perpendicular on traffic direction apart from change value LxIt is then 1 × 1=1m, preset duration 1s is calculated Expection mapping distance L between rear vehicle a1 and marker a4 on traffic directiony2For 18-0=18m, perpendicular to traffic direction On expection mapping distance Lx2For 2-1=1m.

In a step 406, when detecting expected mapping distance Lx2Greater than preset exposure apart from when, determine marker be non-barrier Hinder object;When detecting expected mapping distance Lx2No more than preset exposure distance, it is contemplated that mapping distance Ly2Greater than default Security alert Apart from when, determine marker be non-barrier;When detecting expected mapping distance Lx2No more than preset exposure distance, it is contemplated that mapping Distance Ly2When no more than default Security alert distance and being greater than default safety threshold, determine that marker is level-one barrier; When detecting expected mapping distance Lx2No more than preset exposure distance, it is contemplated that mapping distance Ly2No more than default safety critical away from From when, determine marker be second level barrier.

Wherein, the numerical value of preset exposure distance, default Security alert distance and default safety threshold can be by certainly It adapts to and decision-making module learns to obtain according to driver history operation data, or, being defaulted in adaptively and decision by developer Module.

It should be noted that the value of default safety threshold is less than the value of default Security alert distance.

Illustratively, preset exposure distance is 0.5m, and presetting Security alert distance is 12m, and default safety threshold is 11m adaptively detects the expection mapping distance L of marker a4 then according to above-mentioned data with decision-making modulex21m is greater than default connect Touch distance 0.5m, it is determined that marker is the non-barrier of a4;The expected mapping of marker a2 is adaptively detected with decision-making module Distance Lx20m is less than preset exposure distance 0.5m, and expected mapping distance Ly2It is 11m that 10m, which is less than default safety threshold, then Determine that marker be a2 is second level barrier;The expection mapping distance L of marker a3 is adaptively detected with decision-making modulex20m is small In preset exposure distance 0.5m, and it is expected mapping distance Ly212m is equal to default safety and faces warning distance 12m, it is determined that marker It is level-one barrier for a3.

In step 407a, when detect not there is no barrier when, keep the driving condition of vehicle.

The driving condition of vehicle is kept to refer to and continues to drive with the kinematic parameter of current vehicle, adaptively and decision-making module The motion state of vehicle is not intervened.

It should be noted that environmental perception module obtains the every of vehicle side in real time and drives when executing step 407a Environmental data, and step 401 is repeated to step 406.

In step 407b, when detecting there is only when level-one barrier, danger early warning signal is sent, according to danger early warning Signal makes early warning feedback.

The transmission form of danger early warning signal can be the forms such as the tinkle of bells, voice, image, and vehicle driver can receive When the danger early warning signal, manual manipulation is carried out to the operating parameter of vehicle and avoids vehicle to change the operating parameter of vehicle With the shock of barrier.

In step 407c, when detecting the presence of second level barrier, according to the position data and movement of barriers at different levels The position data and kinematic parameter of parameter and vehicle obtain kinematic parameter from experience driving data library and adjust data.

Experience driving data library has recorded the pre- determining barriers at different levels of adaptive learning algorithm that first pass through with respect to vehicle Positional relationship, kinematic parameter and the kinematic parameter adjustment data of the kinematic parameter of barrier, vehicle it is at least a pair of corresponding Relationship.

Illustratively, the form of expression in experience driving data library can be as shown in Table 3, wherein is that experience is driven shown in table three Sail a part in database, it may include at least one scheme that kinematic parameter, which adjusts data, adaptively can be with decision-making module It is adjusted in data according to preset order or random manner from the kinematic parameter for meeting corresponding relationship and selects one group of kinematic parameter tune Entire data is adjusted the driving condition of vehicle:

Table three

According to above-mentioned data, illustratively, the part experience driving data library adaptively and shown in decision-making module from table three obtains two Corresponding kinematic parameter adjustment data " speed reduces to 17 m/s " of grade barrier a2.

It should be noted that adaptively a3 pairs of level-one barrier can also be obtained from experience driving data library with decision-making module Kinematic parameter adjustment data " speed reduces to 19 m/s " answered, and kinematic parameter adjustment data and other movements obtained are joined Number adjustment data are compared, and optimal kinematic parameter adjustment data are determined, for example, above-mentioned level-one barrier a3 and second level obstacle The corresponding kinematic parameter adjustment data of object a2 compare, and kinematic parameter adjustment data " speed reduces to 17 m/s " can be protected preferably Hinder vehicle not collide with barrier, therefore " speed reduces to 17 m/s " is selected to adjust data as final kinematic parameter.

It should be noted that every kinematic parameter adjustment data in experience driving data library can also include the acceleration of vehicle Degree adjustment data.

In a step 408, the kinematic parameter that data calculate adjustment rear vehicle is adjusted according to kinematic parameter.

It adaptively can be according to the fortune of the vehicle after the kinematic parameter adjustment data acquisition adjustment determined with decision-making module Dynamic parameter, for example, adaptively obtaining the corresponding kinematic parameter tune of second level barrier a2 from experience driving data library with decision-making module Entire data " speed reduces to 17 m/s ", then the kinematic parameter of the vehicle is " 17.00m/s, front " after adjusting.

In step 409, for each marker, according to current mapping distance Ly1, current mapping distance Lx1, after adjustment The kinematic parameter of vehicle and the kinematic parameter of marker are calculated separately by transporting between preset duration rear vehicle and marker Expectation mapping distance L on line directiony3With perpendicular to the expectation mapping distance L on traffic directionx3

Illustratively, if preset duration is 1s, adaptively vehicle a1 is calculated according to above-mentioned data and marks with decision-making module Show the expectation mapping distance L between object a2 on traffic directiony3For 12m, perpendicular to the expectation mapping distance on traffic direction Lx3For 0m.

In step 410, when detecting desired mapping distance Ly3Greater than expected mapping distance Ly2When, according to kinematic parameter Adjustment data are adjusted the driving condition of vehicle.

According to above-mentioned data, for second level barrier a2, a2 corresponding expectation mapping is adaptively detected with decision-making module Distance Ly312m is greater than expected mapping distance Ly210m, then according to the kinematic parameter of " speed reduces to 17 m/s " adjustment data to vehicle The driving condition of a1 is adjusted.

In step 411, when detecting desired mapping distance Ly3No more than expected mapping distance Ly2When, by preset order The kinematic parameter adjustment data for meeting corresponding relationship are reacquired from experience driving data library, execute and number is adjusted according to kinematic parameter The step of according to the kinematic parameter for calculating adjustment rear vehicle.

It should be noted that adaptively can be by the execution of step 411, to constantly adjust current vehicle with decision-making module Kinematic parameter to the primary condition that second level barrier is not present after preset duration is met, to ensure the life peace of driver Entirely.

Wherein, the present invention also provides the entrances that vehicle driver is used to enable vehicle cruise system, so that vehicle is driven The person of sailing can select to drive vehicle, or the driving by vehicle cruise system adapter tube vehicle by itself according to personal preference.

In conclusion vehicle cruise method provided in an embodiment of the present invention, perceives environment using multisensor, and The information of different sensors is characterized to realize high-grade nonlinear data fusion again, uses the data structure of fusion Portion's 3D map of founding the bureau carrys out aid decision, can be improved system to driving environment carry out depth perception, realize it is good adaptively and certainly Plan;In addition, the present invention can identifying to barrier classification, can make more effectively for the higher barrier of degree of danger Decision feedback improves the decision real-time and flexibility of vehicle cruise system.Vehicle cruise system provided by the invention and method Stronger to the synthesis sensing capability of obstacles around the vehicle, stronger to the strain flexibility of barrier, anti-collision technique function is polynary Change, is highly suitable for road conditions complicated and changeable.

A bit for needing to illustrate, data type, each data in such as experience driving data library shown in the embodiment of the present invention The exemplary only explanation such as calculation, meter of the present invention to such as data type in experience driving data library, each data Calculation mode etc. is simultaneously not construed as limiting.

Although having used general explanation, specific embodiment and test above, the present invention is described in detail, But on the basis of the present invention, it can be modified or be improved, this will be apparent to those skilled in the art.Cause This, these modifications or improvements, fall within the scope of the claimed invention without departing from theon the basis of the spirit of the present invention.

Those skilled in the art will readily occur to of the invention other after considering specification and practice invention here Embodiment.The present invention is directed to cover any variations, uses, or adaptations of the invention, these modifications, purposes or Adaptive change follow general principle of the invention and including the undocumented common knowledge in the art of the present invention or Conventional techniques.It should be understood that the invention is not limited to the accurate knots for being described above and being shown in the accompanying drawings Structure, and various modifications and changes may be made without departing from the scope thereof.

Claims (10)

1. a kind of vehicle cruise system, which is characterized in that the vehicle cruise system includes: environmental perception module, data fusion Module, adaptively with decision-making module and execution module;
The environmental perception module is used for when vehicle is in driving status, is obtained the every of the vehicle side in real time and is driven ring Border data, and every driving environment data are pre-processed;
The data fusion module is used to carry out weight to by the pretreated every driving environment data of the environmental perception module Define data indicate, and by redefine data expression after every driving environment data nonlinear data is carried out by particle filter Fusion constructs part 3D map according to every driving environment data after non-linear fusion, and the part 3D map includes at least One marker;
Each mark in the part 3D map for being adaptively used to obtain the data fusion module building with decision-making module Show the position data and kinematic parameter of object;According to the position of the position data of the vehicle and kinematic parameter and each marker Data and kinematic parameter determine the barriers at different levels for meeting preset rules from each marker;According to barriers at different levels The position data and kinematic parameter of position data and kinematic parameter and the vehicle are obtained from experience driving data library and are moved Parameter adjusts data, and experience driving data library has recorded the pre- barrier phases at different levels for first passing through adaptive learning algorithm and determining To at least a pair of of the positional relationship of vehicle, the kinematic parameter of barrier, the kinematic parameter of vehicle and kinematic parameter adjustment data Corresponding relationship;
The execution module is used for according to the kinematic parameter adjustment data adaptively obtained with decision-making module to the vehicle Driving condition be adjusted.
2. vehicle cruise system according to claim 1, which is characterized in that the kinematic parameter includes preset time range Interior average speed and directional velocity, described to be adaptively also used to decision-making module:
Institute is calculated separately according to the position data of the vehicle and the position data of the marker for each marker State vehicle and current mapping distance L of the marker on traffic directiony1With perpendicular to the current mapping on traffic direction Distance Lx1
According to the current mapping distance Ly1, the current mapping distance Lx1, vehicle kinematic parameter and the marker Kinematic parameter calculates separately the expected mapping after preset duration between the vehicle and the marker on traffic direction Distance Ly2With perpendicular to the expection mapping distance L on traffic directionx2
When detecting the expected mapping distance Lx2Greater than preset exposure apart from when, determine the marker be non-barrier;When Detect the expected mapping distance Lx2No more than preset exposure distance, the expected mapping distance Ly2Greater than default safe police Guard against apart from when, determine the marker be non-barrier;When detecting the expected mapping distance Lx2No more than preset exposure away from From the expected mapping distance Ly2When no more than default Security alert distance and being greater than default safety threshold, described in determination Marker is level-one barrier;When detecting the expected mapping distance Lx2No more than preset exposure distance, the expected mapping Distance Ly2When no more than default safety threshold, determine that the marker is second level barrier.
3. vehicle cruise system according to claim 2, which is characterized in that described to be adaptively also used to decision-making module:
When detect not there is no barrier when, keep the driving condition of the vehicle;
When detecting there is only when level-one barrier, danger early warning signal is sent to the execution module, by the execution module Early warning feedback is made according to the danger early warning signal;
When detecting the presence of second level barrier, the position data and kinematic parameter according to barriers at different levels is executed, and The position data and kinematic parameter of the vehicle obtain the step of kinematic parameter adjusts data from experience driving data library.
4. vehicle cruise system according to claim 2, which is characterized in that described to be adaptively also used to decision-making module:
The kinematic parameter that data calculate the vehicle after adjustment is adjusted according to the kinematic parameter;
For each marker, according to the current mapping distance Ly1, the current mapping distance Lx1, the vehicle after adjustment The kinematic parameter of kinematic parameter and the marker, calculate separately after preset duration the vehicle and the marker it Between expectation mapping distance L on traffic directiony3With perpendicular to the expectation mapping distance L on traffic directionx3
When detecting the expectation mapping distance Ly3Greater than the expected mapping distance Ly2When, the kinematic parameter is adjusted into data It is sent to the execution module, data are adjusted to the driving condition of the vehicle according to the kinematic parameter by the execution module It is adjusted;
When detecting the expectation mapping distance Ly3No more than the expected mapping distance Ly2When, by preset order from the experience Driving data library reacquires the kinematic parameter adjustment data for meeting corresponding relationship, executes described according to kinematic parameter adjustment After data calculating adjustment the step of the kinematic parameter of the vehicle.
5. vehicle cruise system according to claim 1, which is characterized in that the environmental perception module includes visual sensing Device, laser radar sensor and alignment sensor, the environmental perception module are also used to:
In real time by the traveling image of vehicle front side described in the visual sensor captured in real-time, depth convolution coder knot is used Structure SegNet carries out image segmentation to the traveling image and marker identifies;
The number of scan points evidence for obtaining vehicle side object by the laser radar sensor in real time, by the scanning The location registration process of point data generates 3D cloud atlas;
The two dimensional motion track of the vehicle is recorded by alignment sensor and is shown in plane map in real time;
At this point, the items driving environment data include that the traveling image of the vehicle front side, vehicle side object are swept Described point data, the two dimensional motion track of the vehicle and plane map.
6. a kind of vehicle cruise method, which is characterized in that the described method includes:
When vehicle is in driving status, every driving environment data of the vehicle side are obtained in real time, and drive to items Environmental data is pre-processed;
Redefine data to every driving environment data indicates, and will redefine every driving environment number after data expression Nonlinear data fusion is carried out according to by particle filter, constructs part 3D according to every driving environment data after non-linear fusion Map, the part 3D map includes at least one marker;
Obtain the position data and kinematic parameter of each marker in the part 3D map;According to the position data of the vehicle With the position data and kinematic parameter of kinematic parameter and each marker, determine to meet preset rules from each marker Barriers at different levels;According to the position data and movement of the position data of barriers at different levels and kinematic parameter and the vehicle Parameter obtains kinematic parameter from experience driving data library and adjusts data, and experience driving data library has recorded pre- first pass through certainly The movement ginseng of the positional relationship of barrier relative vehicles at different levels, the kinematic parameter of barrier, vehicle that adaptive learning algorithm determines Several at least a pair of of corresponding relationships with kinematic parameter adjustment data;
Data are adjusted according to the kinematic parameter to be adjusted the driving condition of the vehicle.
7. according to the method described in claim 6, it is characterized in that, the kinematic parameter includes being averaged in preset time range The position data of speed and directional velocity, the position data and kinematic parameter according to the vehicle and each marker and Kinematic parameter determines the barriers at different levels for meeting preset rules from each marker, comprising:
Institute is calculated separately according to the position data of the vehicle and the position data of the marker for each marker State vehicle and current mapping distance L of the marker on traffic directiony1With perpendicular to the current mapping on traffic direction Distance Lx1
According to the current mapping distance Ly1, the current mapping distance Lx1, vehicle kinematic parameter and the marker Kinematic parameter calculates separately the expected mapping after preset duration between the vehicle and the marker on traffic direction Distance Ly2With perpendicular to the expection mapping distance L on traffic directionx2
When detecting the expected mapping distance Lx2Greater than preset exposure apart from when, determine the marker be non-barrier;When Detect the expected mapping distance Lx2No more than preset exposure distance, the expected mapping distance Ly2Greater than default safe police Guard against apart from when, determine the marker be non-barrier;When detecting the expected mapping distance Lx2No more than preset exposure away from From the expected mapping distance Ly2When no more than default Security alert distance and being greater than default safety threshold, described in determination Marker is level-one barrier;When detecting the expected mapping distance Lx2No more than preset exposure distance, the expected mapping Distance Ly2When no more than default safety threshold, determine that the marker is second level barrier.
8. the method according to the description of claim 7 is characterized in that the method, further includes:
When detect not there is no barrier when, keep the driving condition of the vehicle;
When detecting there is only when level-one barrier, danger early warning signal is sent, early warning is made according to the danger early warning signal Feedback;
When detecting the presence of second level barrier, the position data and kinematic parameter according to barriers at different levels is executed, and The position data and kinematic parameter of the vehicle obtain the step of kinematic parameter adjusts data from experience driving data library.
9. the method according to the description of claim 7 is characterized in that the method, further includes:
The kinematic parameter that data calculate the vehicle after adjustment is adjusted according to the kinematic parameter;
For each marker, according to the current mapping distance Ly1, the current mapping distance Lx1, the vehicle after adjustment The kinematic parameter of kinematic parameter and the marker, calculate separately after preset duration the vehicle and the marker it Between expectation mapping distance L on traffic directiony3With perpendicular to the expectation mapping distance L on traffic directionx3
When detecting the expectation mapping distance Ly3Greater than the expected mapping distance Ly2When, number is adjusted according to the kinematic parameter It is adjusted according to the driving condition to the vehicle;
When detecting the expectation mapping distance Ly3No more than the expected mapping distance Ly2When, by preset order from the experience Driving data library reacquires the kinematic parameter adjustment data for meeting corresponding relationship, executes described according to kinematic parameter adjustment After data calculating adjustment the step of the kinematic parameter of the vehicle.
10. according to the method described in claim 6, it is characterized in that, the every driving for obtaining the vehicle side in real time Environmental data, and every driving environment data are pre-processed, comprising:
In real time by the traveling image of vehicle front side described in visual sensor captured in real-time, depth convolution coder structure is used SegNet carries out image segmentation to the traveling image and marker identifies;
The number of scan points evidence for obtaining vehicle side object by laser radar sensor in real time, by the number of scan points According to location registration process generate 3D cloud atlas;
The two dimensional motion track of the vehicle is recorded by alignment sensor and is shown in plane map in real time;
At this point, the items driving environment data include that the traveling image of the vehicle front side, vehicle side object are swept Described point data, the two dimensional motion track of the vehicle and plane map.
CN201810910364.1A 2018-08-10 2018-08-10 A kind of vehicle cruise system and method CN109094561A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810910364.1A CN109094561A (en) 2018-08-10 2018-08-10 A kind of vehicle cruise system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810910364.1A CN109094561A (en) 2018-08-10 2018-08-10 A kind of vehicle cruise system and method

Publications (1)

Publication Number Publication Date
CN109094561A true CN109094561A (en) 2018-12-28

Family

ID=64849387

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810910364.1A CN109094561A (en) 2018-08-10 2018-08-10 A kind of vehicle cruise system and method

Country Status (1)

Country Link
CN (1) CN109094561A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110606084A (en) * 2019-09-19 2019-12-24 中国第一汽车股份有限公司 Cruise control method, cruise control device, vehicle and storage medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07240000A (en) * 1994-03-01 1995-09-12 Nikon Corp Alarm device for distance between vehicles
US20090058677A1 (en) * 2007-08-30 2009-03-05 Industrial Technology Research Institute Method and apparatus for predicting/alarming the moving of hidden objects
CN102303606A (en) * 2011-06-17 2012-01-04 清华大学 Hazard assessment method based on retarded velocity required by vehicle
CN103895645A (en) * 2014-04-11 2014-07-02 安徽江淮汽车股份有限公司 Automobile anti-collision method and system
CN204845924U (en) * 2015-07-23 2015-12-09 江苏大学 Vertically keep away to hit and remind and follow integrated system automatically
CN106004879A (en) * 2015-03-27 2016-10-12 株式会社万都 Driving assistant system of vehicle and method for controlling same
CN106056972A (en) * 2016-06-29 2016-10-26 江苏科技大学 Security anti-collision early-warning method based on vehicle driving speed and position information fusion
CN206691107U (en) * 2017-03-08 2017-12-01 深圳市速腾聚创科技有限公司 Pilotless automobile system and automobile

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07240000A (en) * 1994-03-01 1995-09-12 Nikon Corp Alarm device for distance between vehicles
US20090058677A1 (en) * 2007-08-30 2009-03-05 Industrial Technology Research Institute Method and apparatus for predicting/alarming the moving of hidden objects
CN102303606A (en) * 2011-06-17 2012-01-04 清华大学 Hazard assessment method based on retarded velocity required by vehicle
CN103895645A (en) * 2014-04-11 2014-07-02 安徽江淮汽车股份有限公司 Automobile anti-collision method and system
CN106004879A (en) * 2015-03-27 2016-10-12 株式会社万都 Driving assistant system of vehicle and method for controlling same
CN204845924U (en) * 2015-07-23 2015-12-09 江苏大学 Vertically keep away to hit and remind and follow integrated system automatically
CN106056972A (en) * 2016-06-29 2016-10-26 江苏科技大学 Security anti-collision early-warning method based on vehicle driving speed and position information fusion
CN206691107U (en) * 2017-03-08 2017-12-01 深圳市速腾聚创科技有限公司 Pilotless automobile system and automobile

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110606084A (en) * 2019-09-19 2019-12-24 中国第一汽车股份有限公司 Cruise control method, cruise control device, vehicle and storage medium
CN110606084B (en) * 2019-09-19 2020-12-18 中国第一汽车股份有限公司 Cruise control method, cruise control device, vehicle and storage medium

Similar Documents

Publication Publication Date Title
CN105620489B (en) Driving assistance system and vehicle real-time early warning based reminding method
CN104573646B (en) Chinese herbaceous peony pedestrian detection method and system based on laser radar and binocular camera
US10137890B2 (en) Occluded obstacle classification for vehicles
CN103455144B (en) Vehicle-mounted man-machine interaction system and method
CN105892471B (en) Automatic driving method and apparatus
CN104793202B (en) The object emerging system of more radar imagery sensors
JP5939357B2 (en) Moving track prediction apparatus and moving track prediction method
CN104554258B (en) Using the path planning of the avoidance steering operation of virtual potential field technology
JP5971341B2 (en) Object detection device and driving support device
CN104217590B (en) Method for making the electronic controller in main vehicle determine traffic density
JP6329569B2 (en) System and method for evaluating abnormal driving behavior of a vehicle traveling on a road
EP3223260B1 (en) Drive control system
US8564657B2 (en) Object motion detection system based on combining 3D warping techniques and a proper object motion detection
CN103043057B (en) Abnormal driving based on vehicle position information judges and warning system
CN103176185B (en) Method and system for detecting road barrier
CN105009175B (en) The behavior of autonomous vehicle is changed based on sensor blind spot and limitation
US8781688B2 (en) Method and system for combining sensor data
CN102442248B (en) For warning the system and method for frontal collisions danger and using its vehicle
CN104933856B (en) Road conditions real-time evaluation system and method
Liebner et al. Driver intent inference at urban intersections using the intelligent driver model
CN102598083B (en) Driving support device
Göhring et al. Radar/lidar sensor fusion for car-following on highways
JP2013143134A (en) Vehicle having computer for monitoring and predicting object participating in traffic
KR100909741B1 (en) Monitoring device, monitoring method
JP2016001464A (en) Processor, processing system, processing program, and processing 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