Summary of the invention
The object of the present invention is to provide a kind of determination method on vehicle obstacle-avoidance lane-change opportunity and the control method of avoidance lane-change,To solve the problems, such as that automatic obstacle-avoiding or lane-change control strategy are unreasonable in intelligent automobile automatic Pilot.
To achieve the above object, the present invention provides a kind of determination method on vehicle obstacle-avoidance lane-change opportunity, including identified belowMethod and technology scheme:
Determine method scheme one: a kind of determination method on vehicle obstacle-avoidance lane-change opportunity, method includes the following steps:
1) whether detection vehicle front has barrier, if so, judging whether the distance of vehicle front barrier and speed are fullFoot first imposes a condition, if satisfied, detection vehicle right side or left-hand lane obstacle information；
2) judge whether the lateral distance of vehicle front barrier and its right side or left side barrier is greater than set distance；
3) if more than then judging whether there is barrier within the scope of vehicle right side or at least one lane width of left side；
If 4) judge whether the distance at vehicle right side rear or left side rear obstacle and speed meet without barrierSecond imposes a condition, if satisfied, then to the right or left side lane-change.
Beneficial effect is that this determination method scheme one is determined by obtaining obstacles around the vehicle information and traffic informationWhether vehicle carries out lane-change maneuver, by the intelligent environment sensing capability of vehicle, guarantees intelligence while vehicle safe drivingIt plans avoidance lane-change, solves the problems, such as that automatic obstacle-avoiding or lane-change control strategy are unreasonable in intelligent automobile automatic Pilot.
Determine method scheme two: on the basis of determining method scheme one, described first imposes a condition are as follows: vehicle front barrierThe speed of object is hindered to be less than the 1/2 of this vehicle speed, and front obstacle is more than or equal to two this vehicle vehicle bodies at a distance from this vehicle.
Determine method scheme three: on the basis of determining method scheme one or determining method scheme two, second settingCondition are as follows: vehicle right side rear or left side rear obstacle at a distance from this vehicle rear end not less than 1 this vehicle vehicle body and speed it is smallIn this vehicle speed；Or vehicle right side rear or left side rear obstacle speed are greater than this vehicle speed, and at a distance from this vehicle rear endGreater than 2 this vehicle vehicle bodies.
Determine method scheme four: on the basis of determining method scheme three, the set distance is 1 lane width.
The present invention provides a kind of control method of vehicle obstacle-avoidance lane-change, including following control method technical solution:
Control method scheme one: a kind of control method of vehicle obstacle-avoidance lane-change, method includes the following steps:
1) whether detection vehicle front has barrier, if so, judging whether the distance of vehicle front barrier and speed are fullFoot first imposes a condition, if satisfied, detection vehicle right side or left-hand lane obstacle information；
2) judge whether the lateral distance of vehicle front barrier and right side or left side barrier is greater than set distance；
3) if more than then judging whether there is barrier within the scope of vehicle right side or at least one lane width of left side；
If 4) judge whether the distance at vehicle right side rear or left side rear obstacle and speed meet without barrierSecond imposes a condition, if satisfied, then to the right or left side lane-change；
5) when vehicle is to the left or when the lane-change of right side, the starting point coordinate, coordinate of ground point and road course of lane-change are determinedAngle；
6) it is carried out curve fitting, is obtained from starting point coordinate according to starting point coordinate, coordinate of ground point and road course angleTo the path of coordinate of ground point, which is lane-change path.
Control method scheme two: on the basis of control method scheme one, described first imposes a condition are as follows: vehicle front barrierThe speed of object is hindered to be less than the 1/2 of this vehicle speed, and front obstacle is more than or equal to two this vehicle vehicle bodies at a distance from this vehicle.
Control method scheme three: on the basis of control method scheme one or control method scheme two, second settingCondition are as follows: vehicle right side rear or left side rear obstacle at a distance from this vehicle rear end not less than 1 this vehicle vehicle body and speed it is smallIn this vehicle speed；Or vehicle right side rear or left side rear obstacle speed are greater than this vehicle speed, and at a distance from this vehicle rear endGreater than 2 this vehicle vehicle bodies.
Control method scheme four: on the basis of control method scheme three, the set distance is 1 lane width.
Control method scheme five: on the basis of control method scheme one, control method scheme two or control method scheme fourOn, the method for the curve matching is B-spline method.
Control method scheme six: on the basis of control method scheme five, the B-spline method is B-spline Curve method.
Control method scheme seven: on the basis of control method scheme six, there is maximum in the B-spline curves planning processThe calculating of boundary's curvature is as follows: the shape of B-spline curves is determined by control point completely, for the determining Duan Sanci in three control pointsB-spline curves increase the midpoint of two line segments as new control point, and B-spline shape depends on lesser line in two line segmentsThe angle α of segment length L and two line segments obtain the Curvature varying of B-spline curves are as follows:
It enablesIt can be in the hope of as u=0.5, κ has maximum value；Curvature maximum substitute into above formula, thus obtain L andThe relationship of α between the two:
Wherein, any two adjacent control wires section meets above-mentioned relation, that is, can guarantee the curvature of B-spline curves.
Control method scheme eight: in control method scheme one, control method scheme two, control method scheme four, controlling partyOn the basis of method scheme six or control method scheme seven, before vehicle reaches target point, also judge whether vehicle meets return itemPart, if satisfied, then being returned according to setting return path control vehicle；If not satisfied, then controlling vehicle straight trip；
The return condition is specific as follows: 1, clear within the scope of vehicle returning direction side setting safe distance；2,The rear clear of vehicle returning direction side, or there are barrier but apart from this tailstock portion be greater than set it is safe when away from；3, the front clear of vehicle returning direction side, or there is barrier but speed to be greater than this vehicle speed.
Specific embodiment
The present invention will be further described in detail with reference to the accompanying drawing.
The present invention provides a kind of control method of vehicle obstacle-avoidance lane-change, can be applied in pilotless automobile, such as Fig. 1 instituteShow, which includes crosswise joint module, longitudinally controlled module, information Fusion Module and decision-making module, wherein letterIt ceases Fusion Module setting and is received for sampling connection ZigBee, a line laser radar, ultrasonic radar, red street lamp signal RFDevice, inertial navigation GPS, Lane detection video camera and millimetre-wave radar, and sampled signal is handled or parsed；Decision-making moduleInput terminal link information Fusion Module, for acquiring the output signal of information Fusion Module, while carrying logic judgment and outputSignal is controlled to crosswise joint module and longitudinally controlled module；Crosswise joint module controls the acceleration and deceleration of vehicle, longitudinal to controlMolding block controls steering wheel angle, to control vehicle obstacle-avoidance, lane-change.
The present invention provides a kind of control method of vehicle obstacle-avoidance lane-change, as shown in Fig. 2, when including a kind of vehicle obstacle-avoidance lane-changeThe determination method of machine, the specific steps are as follows:
1, obstacles around the vehicle and traffic information are obtained.
Vehicle front obstacle, rear obstacle, left and right front obstacle, left and right rear are acquired by radar or cameraThe information and traffic information of barrier, for example, with barrier spacing, barrier speed, lane width etc..
2, by the determination method on vehicle obstacle-avoidance lane-change opportunity, judge whether to meet lane-change condition.
Lane-change condition needs to guarantee vehicle safety and passenger comfort, and the generation of lane-changing intention needs to carry out multi-group dataComparison, judgment and detection repeatedly just carry out steering operation when thinking to meet lane-change condition, meanwhile, when the intention of lane-change generatesIntelligent vehicle is as human driver, and preset turn signal provides temporal guarantee before steering, keeps intelligent vehicle smoothGround incorporates normal traffic flow.But the width value for turning to intention time cannot be excessive, spends the opportunity that senior general misses lane-change；Also notCan be too small, it is too small, reduce the driving safety of vehicle.The determination method on vehicle obstacle-avoidance lane-change opportunity, specific steps include:
1) whether detection vehicle front has barrier, if so, judging whether the distance of vehicle front barrier and speed are fullFoot first imposes a condition, if satisfied, detection vehicle right side or left-hand lane obstacle information；
2) judge whether the lateral distance of vehicle front barrier and right side or left side barrier is greater than set distance；
3) if more than then judging whether there is barrier within the scope of vehicle right side or at least one lane width of left side；
If 4) judge whether the distance at vehicle right side rear or left side rear obstacle and speed meet without barrierSecond imposes a condition, if satisfied, then to the right or left side lane-change.
It wherein first imposes a condition are as follows: the speed of vehicle front barrier is less than the 1/2 of this vehicle speed, and preceding objectObject is more than or equal to two this vehicle vehicle bodies at a distance from this vehicle；Second imposes a condition are as follows: vehicle right side rear or left side rear barrierHinder object at a distance from this vehicle rear end not less than 1 this vehicle vehicle body and speed is less than this vehicle speed；Or vehicle right side rear or left sideRear obstacle speed is greater than this vehicle speed, and is greater than 2 this vehicle vehicle bodies at a distance from this vehicle rear end；Set distance is 1 vehicleRoad width.
Above-mentioned Rule of judgment can also optimize configuration according to experimental data:
For example, simulation mankind's driving habit, when vehicle front barrier speed be less than this vehicle speed 1/2, and with it is precedingThe distance of square barrier is more than two vehicle bodies；If lane-change to the right, width of the front obstacle apart from another barrier in right sideDegree is greater than 1 lane width, and its speed is greater than this vehicle speed, and similarly, if wanting Zuo Huandao, front obstacle distance is leftThe width of another barrier in side is also greater than a lane width, and its speed is greater than this vehicle speed；
It is to be not allow for any barrier within the scope of one lane width in right side meanwhile when vehicle determines lane-change to the rightInformation；It is the information for being not allow for any barrier within the scope of one lane width in left side when vehicle determines lane-change to the left；
Furthermore the same with mankind's driving vehicle, the mankind are after generating lane-changing intention, after also being looked over by rearview mirrorSquare barrier situation, judges whether lane-change, it is also necessary to meet distance of the rear obstacle apart from this vehicle rear end will 2 vehicle bodies withOn or distance of the rear obstacle apart from this vehicle rear end will be more than 1 vehicle body, and the speed of rear obstacle is less thanThis vehicle speed.
If 3, meeting above-mentioned condition, according to setting avoidance path clustering vehicle lane-changing.
When meeting lane-change condition, i.e., controllable vehicle carries out lane-change, and intelligent automobile needs to be carried out according to the path of settingLane-change, it is therefore desirable to which the path of vehicle obstacle-avoidance lane-change is planned.
When vehicle carries out lane-change, the starting point coordinate, coordinate of ground point and road course angle of lane-change are determined；According to startingPoint coordinate, coordinate of ground point and road course angle carry out curve fitting, and obtain the path from starting point coordinate to coordinate of ground point,The path is lane-change path.
Wherein, starting point coordinate is that the geodetic coordinates of current vehicle position or the earth after vehicle heading 0.2m are satMark；Coordinate of ground point be front obstacle endpoint to lane-change direction deviate half body width of this vehicle and set safe lateral distance itKnown with the information of the geodetic coordinates at place, front obstacle endpoint by laser radar；Road course angle is to be obtained by high definition map.
The essence of path planning is to calculate the curve of connection an initial position and final position, the requirement of path computingWith it is critical that curve wants continuous and derivable and curvature bounded.There are many curve is available at present: polynomial curve, shellfishSai Er curve and B-spline curves etc., wherein B-spline is relatively easy for the control and calculating of curvature, the method is as follows:
Pass through n+1 control pointAnd n+k+1 Parameter nodes vectorDetermine k rank,That is k-1 B-spline curves, expression formula are as follows:
Wherein Bi,k(u) it is known as Un,kUpper k rank B-spline basic function, the basic function are determined by deBoox-Cox recurrence relation, are closedIt is that formula is as follows:
And the property of B-spline curves according to the present invention has:
1) continuity: in r multiple knot uiC is at least at (k-1≤i≤n)k-1-r, the continuity of whole curve is not less thank-1-rmax, wherein rmaxIndicate node uiThe maximum value of tuple.
2) locality: Bi,k(u) only in section [ui,ui+1) on take positive value, be 0 on other sections, therefore B-spline curvesIn parameter section [ui,ui+1) part line segment on (k-1≤i≤n) only andTotal k control vertex is related.
3) convex closure: B-spline curves are in parameter section [ui,ui+1) part on (k-1≤i≤n) is located at this k controlIn the convex closure on vertex, as shown in Figure 3.
According to the constraint of auto model, curve continuity and maximum curvature bounded are the requirements that path curve must satisfy.Other types curve is compared, the Boundary Condition for Solving of polynomial curve generally requires to find out analytic solutions using numerical method, solvesProcess is cumbersome；The control points and order of a curve number of Bezier correspond, if required path is longer, are not increasing songIn the case where line order, the control ability of dominating pair of vertices curve shape weakens；B-spline curves are general Beziers,It controls points and curve order does not have positive connection, and boundary condition and curvature limitation can be suitable by selectingControl point meets, and does not need to solve complicated numerical value and calculates, since cubic B-spline is sufficient for C2Continuity, to meetThe requirement of vehicle movement, therefore select B-spline Curve as path curve.The maximum bounded curvature estimation of B-spline is as follows:
As shown in figure 4, the shape of B-spline curves is determined by control point completely, one section three determined for three control pointsSecondary B-spline curves increase the midpoint of two line segments as new control point, and B-spline shape depends on lesser in two line segmentsThe angle α of line segment length L and two line segments, wherein it is assumed that two lines segment length is equal, be so that length is lesser when reality calculatesStandard is certain when wherein a line segment length increases if the line segment calculating with smaller length can satisfy curvature requirementAlso meet curvature requirement, obtain the Curvature varying of B-spline curves are as follows:
It enablesIt can be in the hope of as u=0.5, κ has maximum value；Curvature maximum substitutes into above formula, to obtain L and αRelationship between the two:
Wherein, any two adjacent control wires section meets above-mentioned relation, that is, can guarantee the curvature of B-spline curves.
The path that B-spline curves are cooked up is global path, as shown in figure 5, including avoidance path 2 and the setting of settingReturn path 4, wherein when vehicle obstacle-avoidance lane-change, travelled from avoidance origin coordinates point 1 to avoidance according to the avoidance path 2 of settingCoordinates of targets point 3；When vehicle returns, travelled from avoidance coordinates of targets point 3 to final position coordinate according to the return path 4 of settingPoint 6.
4, judge whether to meet return condition.
If satisfied, then being returned according to setting return path control vehicle；If not satisfied, then controlling vehicle straight trip.
It is specific as follows wherein to return to condition: 1, clear within the scope of vehicle returning direction side setting safe distance；2,The rear clear of vehicle returning direction side, or there are barrier but apart from this tailstock portion be greater than set it is safe when away from；3, the front clear of vehicle returning direction side, or there is barrier but speed to be greater than this vehicle speed.
In conclusion a kind of control method of vehicle obstacle-avoidance lane-change provided by the invention, as shown in figure 5, intelligent vehicle changesRoad is intended to after generating, and path planning, including avoidance origin coordinates point 1, avoidance coordinates of targets point 3 will be carried out according to B-spline curvesWith final position coordinate points 6 and road course angle；Vehicle reach avoidance origin coordinates point when start along setting avoidance path 2 intoRow form, vehicle reach avoidance coordinates of targets point 3 when, judged whether to meet return condition according to the instruction of decision-making module, ifMeet return condition, then controls setting return path 4 that vehicle is obtained according to B-spline curves from avoidance coordinates of targets 3 to end of pointStop bit is set coordinate points 6 and is returned, if being unsatisfactory for return condition, controls vehicle and keeps straight on from avoidance coordinates of targets point 3 along avoidancePath 5 is travelled.
Specific embodiment of the present invention is presented above, but the present invention is not limited to described embodiment.Under the thinking that the present invention provides, to the skill in above-described embodiment by the way of being readily apparent that those skilled in the artArt means are converted, are replaced, are modified, and play the role of with the present invention in relevant art means it is essentially identical, realizeGoal of the invention it is also essentially identical, the technical solution formed in this way is to be finely adjusted to be formed to above-described embodiment, this technologyScheme is still fallen in protection scope of the present invention.