Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of, and the vehicle based on high-precision map predicts cruise control method,This method makes full use of geographical advantage to optimize speed, can achieve the purpose that fuel-economizing and timeliness double shield.
In order to solve the above-mentioned technical problem, the vehicle prediction cruise control method of the invention based on high-precision map includesFollowing step:
Step 1: vehicle location
Vehicle present position is accurately positioned by GPS, and is obtained in X meters of front according to high-precision mapMap data information；
Step 2: map transmission
Map data information is transferred to controller according to ADASIS agreement；The map data information include it is current andNode location, the node gradient in front X meters, intersection location information, speed limiting position and mark；X=1000~3000 meter, XAdjacent node spacing is 10~100 meters in rice；
Step 3: map reconstructs
Map data information effectively reconstructs front map to controller based on the received, and obtaining, which can recognize, works as scarp slope roadThe reconstruct map data information of section, front slope road section information and front cross mouth position and speed limit identification information；
Step 4: prediction cruise speed planning
4.1, cruise speed does the limitation of most value according to following three kinds of situations:
A, there are speed limit mark, restricted speed V in a distance of frontlim1；
B, there are intersection information, restricted speed V in a distance of frontlim2；
C, the cruise speed that driver sets is Vref, the acceptable cruise speed upper deviation is Vinc, lower deviation isVdec；The cruise speed lower limit finally set is (Vref-Vdec), upper limit Vlim1、Vlim2、(Vref+Vinc) in minimum value；
4.2, cruise speed optimization
4.2.1 each node speed range in estimation front
Assuming that the maximum speed of node i is Vimax, the peak torque that vehicle is capable of providing is Tmax, complete vehicle quality m, sectionThe value of slope of point i+1 is Slope(i+1), the distance of node i to node i+1 is Δs(i_i+1), then according to dynamics of vehicle equation(1) it is calculated from node i and drives to the peak acceleration ACC that node i+1 can reach(i+1)max:
ACC(i+1)max=facc(Tmax, Vimax, Slope(i+1), m) and (1)
If Vimax2+2*Acc(i+1)max*Δs(i_i+1)< 0, then+1 the max speed V of node i(i+1)max=0.
If Vimax2+2*Acc(i+1)max*Δs(i_i+1)> 0, then+1 the max speed of node i
V0max=V0That is the speed of current vehicle position；
Assuming that the minimum speed of node i is Vimin, the maximum negative torque that vehicle is capable of providing is Tmin, vehicle is with maximum systemDynamic traveling, then be calculated from node i according to dynamics of vehicle equation (2) and drive to the maximum deceleration that node i+1 can reachSpend Dec(i+1)max:
Dec(i+1)max=fdec(Tmin, Vimin, Slope(i+1), m) and (2)
If Vimin2+2*Dec(i+1)max*Δs(i_i+1)< 0, then the minimum vehicle velocity V of node i+1(i+1)min=0；
If Vimin2+2*Dec(i+1)max*Δs(i_i+1)>=0, then the minimum speed of node i+1
V0min=V0That is the speed of current vehicle position；
4.2.2. by the optimal velocity V of end-node NNIt is planned to the cruise vehicle velocity V of vehicleref, i.e. VF=Vref；
4.2.3 other each node speed optimizations
(1) each node speed range predicted according to step 4.2.1, according to setting interval to each node speed modelEnclose carry out discretization；If obtaining n for -1 discretization of node i(i-1)A velocity nodeThe optimal velocity of node i is Vi；
(2) it when assuming that the maximum speed of node N-1 and minimum speed are equal to a certain velocity node of node N-1, calculatesThe velocity interval of obtained end-node N does not include the optimal vehicle velocity V of end-node NN, then it is assumed that the velocity node is ineligible；Exclude ineligible all velocity nodes of node N-1；
(3) speed with neutral gear hinged node N and node N-1 is attempted
Assuming that the gradient between node N and node N-1 is SlopeN-(N-1), distance is SN-(N-1)；For any eligibleVelocity node V(N-1)_x, calculate from velocity node V(N-1)_xIt sets out, S is travelled using neutral gearN-(N-1)The end speed obtained after distanceDegree；If the optimal vehicle velocity V of end speed and end-node NNSpeed differs 3km/h or more, then leaps to next step (4)；InsteadThen jump directly to step (6)；
(4) speed with positive-torque hinged node N and node N-1 is attempted
It calculates from velocity node V(N-1)_xIt sets out, from positive-torque T required for node N-1 to node N(N-1)_xIf obtainedPositive-torque T(N-1)_xGreater than vehicle motor peak torque, then step (5) are jumped directly to；It is on the contrary then jump directly to step(6)；
(5) speed with negative torque hinged node N and node N-1 is attempted
It calculates from velocity node V(N-1)_xIt sets out, available auxiliary braking torque and end-node N from node N-1 to node NSpeed be Vnext(N-1)_x；If Vnext(N-1)_xWith VNSpeed differs within 3km/h, then goes to step (6)；Otherwise give up the speedSpend node V(N-1)_x；
(6) speed and target function value of each enabled node are saved
When calculating all qualified velocity nodes as initial velocity, the objective function of node N is driven to from node N-1Value；For velocity node V(N-1)_x, target function value J(N-1)_x_NIt is as follows:
J(N-1)_x_N=Q(N-1)B*(B(N-1)_x_N/B(N-1)_Nref+Q(N-1)T*Time(N-1)_x_N/Time(N-1)_Nref (6)
Wherein, B(N-1)_x_N, Time(N-1)_x_NRespectively initial velocity V(N-1)_xDrive to amount of fuel consumed by node N andTime；Q(N-1)BAnd Q(N-1)TThe fuel economy of node N and the weight coefficient of real-time are respectively driven to from node N-1；B(N-1)_x_NIt is obtained by the vehicle startup engine oil consumption map figure demarcated in advance；Time(N-1)_x_N=S(N-1)_N/V(N-1)_x；B(N-1)_NrefIt is cruise vehicle velocity V for speed(N-1)_NrefWhen reference oil consumption from node N-1 to node N, Time(N-1)_NrefIt is to patrol for speedBoat vehicle velocity V(N-1)_NrefWhen from node N-1 to the reference time of node N；
Time(N-1)_Nref=SN-(N-1)/V(N-1)_Nref；
Save the state of each velocity node, i.e. speed and corresponding target function value；Selection is wherein compared with Small object letterThe corresponding velocity node of numerical value is as optimal vehicle velocity VN-1And it saves；
(7) it iterates, calculates optimal velocity
After the state that node N-1 has been calculated, each node N-2, N- below are successively calculated according to step (2)-(6) methodThe state of 3 ...；For any node, general objective when linking all nodes in X meters of front respectively with its each velocity node is calculatedFunctional value selects the wherein Speed Chain where lesser target function value, using the corresponding velocity node of the Speed Chain as the sectionThe optimal velocity of point；
(8) if the speed in any two non-first node can not be linked together by step (3), (4), (5),Then using latter node as final state point, then each node speed is optimized again according to step (1)-(7)；IfThe speed of first node 1 can not be linked together with the speed of node 2, then by the V of node 11Optimal velocity as node 2.
To sum up, by adopting the above-described technical solution, the beneficial effects of the present invention are:
1. the accurate positionin to vehicle, effective map number in front of real-time reception in a distance may be implemented in the present inventionAccording to；
2. the present invention by carrying out validity processing to received map datum, can carry out the road in front effectively heavyStructure identifies different kinds of roads operating condition；
3. can be prompted according to the front road conditions identified driver in driver's normal driving；
4. it is directed to cruising condition, comprehensive fuel-economizing and timeliness requirement, the adaptability in the range that driver is subjected toAdjust cruise speed, this advantage be even more on the route that landform is rugged and rough, road surface rises and falls performed to it is ultimate attainment；
5 present invention can identify speed limit mark and intersection in advance, and realize that effective speed controls according to different situations；
6. imitating the drive automatically for having experience driver the present invention is based on the understanding to road ahead, keeping as far as possibleLow fuel consumption.
The present invention is based on GPS and high-precision maps to predict front traffic information in real time, and to cruise vehicle under different operating conditionsSpeed carries out automatic adjusument, balances the relationship between low oil consumption and high timeliness, has saved transportation cost, improves conevying efficiency,Compared with common cruise, on the basis of guaranteeing vehicle timeliness, the fuel economy of vehicle is greatly improved.
Specific embodiment
The present invention is directed to the prediction cruise function of vehicle, identifies that vehicle is worked as according to global positioning system and high-precision mapFront position and front X meters of road information, and according to real-time road condition information on the basis of cruise, it is acceptable in driverThe best speed of optimization in range.
As shown in Figure 1, the vehicle prediction cruise control method of the invention based on high-precision map includes the following steps:
Step 1: vehicle location
Vehicle present position is accurately positioned by GPS (global positioning system) first, and according to high-precisionMap obtains the map data information in X meters of front.
Step 2: map transmission
It is effectively transmitted according to ADASIS agreement to map information and is sent to controller.Map datum includes currentRoad information, front X meters of road information specifically include the data such as front cross mouth, speed limit and the gradient；X=1000~3000Meter, adjacent node spacing is 10~100 meters in X meters；Here X=2000 meters are chosen, adjacent node spacing chooses 25 meters.
Step 3: map reconstructs
Front map is effectively reconstructed after controller real-time reception map datum.It mainly includes following several for wherein reconstructingA step:
The input of 3.1 signals
The road data that controller receives includes node location, the node gradient, intersection location information, speed limiting position andThe information such as mark.
3.2 signal processing
In order to guarantee the continuity and validity of data, for the data received, need to be handled by a series of singular points,Such as the grade point being mutated in data or a certain section of unreasonable data are removed, to guarantee the validity of subsequent road reconstruct.
The reconstruct of 3.3 roads
After step 3.2 signal processing, controller needs that front X meters of map datum is reconstructed, and is similar toNavigation is the same, by the road conditions induction-arrangement in front at the characteristics map that can be identified.It is divided into the following aspects:
3.3.1 identifying road ahead feature
Controller receives X meters of the road information in front, first the position to each node of road ahead and grade information intoRow storage.Then taxonomic revision is carried out to road information according to gradient difference, sets the range of grade in different typical sections, it is such as flatThe range of grade on road is S0—S1.According to current vehicle position and the front position Po Lu, road conditions locating for vehicle are divided into following severalSituation, including level road, upward slope, descending, mountain peak, mountain valley etc. open to driver can be used as front road condition advisory information, can also makeFor the foundation of speed control strategy.
3.3.2 current roadway characteristic is identified:
Regard the map data information in X meters of ranges of vehicle as known map, according to the current location information of vehicleIt is searched in map, parses current roadway characteristic, including current position, the information such as the gradient, and according to country to high speedThe construction specification of highway does the limitation of most value.
3.3.3 identifying intersection location information
Controller receives intersection location information in X meters of front, and is stored in array.
3.3.4 identifying speed limit identification information
Controller receives speed limit identification information in X meters of front, by speed limiting position and speed limit value storage into array, convenient for standardReally identification speed-limiting messages do corresponding limitation to speed and handle.
Step 4: prediction cruise speed planning
Basic ideas of the invention are to go to improve speed as far as possible before upward slope, reduce speed before descending, make full use of landformAdvantage reaches the dual optimization of running efficiency and cost.
Prediction cruise speed planing method according to the present invention including the following steps:
4.1. prediction cruise speed restriction strategy
Based on the above-mentioned identification to road, for cruising condition, controller first does most the cruise speed finally setValue limitation.Specifically there are following three kinds of speed limit situations:
(1) there are speed limit mark, restricted speed V in a distance of frontlim1；
(2) there are intersection information, restricted speed V in a distance of frontlim2(can self-calibration)；
(3) the cruise speed that driver sets is Vref, the acceptable cruise speed upper deviation is Vinc, lower deviation isVdec；The cruise speed lower limit finally set is (Vref-Vdec), upper limit Vlim1、Vlim2、(Vref+Vinc) in minimum value.
4.2. prediction cruise speed control strategy
Introduced according to step 3.3.1, in front X meters position and the information such as the gradient store, be specifically described belowOnce how cruise speed to be optimized according to the cartographic information of storage.
4.2.1 each node speed range in estimation front
Under the premise of known current vehicle speed, the velocity interval of each node in front is predicted using needs.
It is introduced by taking node B as an example.
Maximum speed calculates
Assuming that the maximum speed of node B is VBmax, the peak torque that vehicle is capable of providing is Tmax, m is complete vehicle quality, sectionThe value of slope of point C is SlopeC, the distance of node B to node C is ΔsBC, then according to dynamics of vehicle side well known in the artJourney (1) drives to the peak acceleration ACC that node C can reach from node BCmaxAre as follows:
ACCCmax=facc(Tmax, VBmax, SlopeC, m) and (1)
If VBmax2+2*AccCmax*ΔsBC< 0, then node C the max speed VCmax=0.
If VBmax2+2*AccCmax*ΔsBC> 0, then node C the max speed
Note: according to above-mentioned rule VBmaxIt is to be calculated according to node A speed, VAmax=VAThat is current vehicle positionSpeed.
Minimum speed calculates
Assuming that the minimum speed of node B is VBmin, the maximum negative torque that vehicle is capable of providing is Tmin, then known in this fieldAccording to dynamics of vehicle equation (2), it is assumed that vehicle with maximum braking traveling, driving to node C from node B can reach mostBig retarding degree DecCmaxAre as follows:
DecCmax=fdec(Tmin, VBmin, SlopeC, m) and (2)
If VBmin2+2*DecCmax*ΔsBC< 0, then node C minimum vehicle velocity VCmin=0.
If VBmin2+2*DecCmax*ΔsBC>=0, then node C minimum speed
Note: according to above-mentioned rule VBminIt is to be calculated according to node A speed, VAmin=VAThat is current vehicle positionSpeed.Each node speed range is stored, specific schematic diagram is as shown in Figure 3.
4.2.2. end-node speed planning
After obtaining each node speed range, the speed of end-node is planned first, herein by minor details spot speedMetric divides the cruise speed of vehicle into, it is assumed that end-node F, then the optimal speed of end-node F is VF；
VF=Vref (3)
4.2.3 other each node speed optimizations
Node speed discretization chooses enabled node
The present invention is the thought based on Dynamic Programming for the planning of speed, from minor details spot speed forward impelling.According to stepThe optimal speed of rapid 4.2.2 elder generation plan node F is VF, then plan the speed of E point；Assuming that the section obtained according to step 4.2.1The maximum speed of point E is VEa, minimum speed VEc, the speed dispersion of node E is obtained by speed with setting speed interval delta vNode VEa、VEbAnd VEc.According to the method for step 4.2.1 calculating speed range, calculate node E initial velocity is respectively VEa、VEbAnd VEcThe velocity interval of Shi Jiedian F.It is assumed that the maximum speed and minimum speed as node E are equal to VEaWhen countThe velocity interval of obtained node F does not include the optimal vehicle velocity V of node FF, then it is assumed that speed VEaIt is ineligible.So sectionPoint E can plan that velocity node is VEbAnd VEc, below with velocity node VEbFor be illustrated.
(2) reference state is calculated
Vehicle in the EF of section is calculated first maintains cruise vehicle velocity VrefEach reference state amount when operation, such as:
According to the vehicle startup engine oil consumption map map analysis demarcated in advance, speed is cruise vehicle velocity VrefWhen from node E runTo the reference oil consumption B of node Feref, it is by Vref, SlopeF, it is fixed that the four-dimensional map of m composition decides by vote:
Beref=MAP (Vref, SlopeF, m) and (4)
Speed is cruise vehicle velocity VrefWhen reference time TimerefIt is as follows:
Timeref=SEF/Vref (5)
Wherein SEFDistance between EF, the value of slope of node F are SlopeF；
(3) trial links the speed of two nodes with neutral gear
Assuming that the gradient between EF is SlopeF, distance is SEF, it calculates from speed VEbIt sets out, S is travelled using neutral gearEFDistanceThe end speed V obtained afterwardsG0.If VFWith VG0Speed differs 3km/h or more, then it is assumed that two sections can not be connected using neutral position slidingPoint jumps directly to step (4).It is on the contrary then think to travel using neutral gear and from node E slide into node F, then omit step belowSuddenly (4) and (5) leap to step (6).
(4) trial links the speed of two nodes with positive-torque
If the speed of two nodes can not be linked with neutral position sliding, attempt to calculate required for node E to node F justTorque TEbIf final torque TEb(it is greater than vehicle motor peak torque) not in reasonable range, then it is assumed that can notThe speed that two nodes are connected by positive-torque, jumps directly to step (5).It is on the contrary then think can by positive-torque link twoThe speed of node leaps to step (6).
(5) trial links the speed of two nodes with negative torque
If above-mentioned steps (3) (4) both of which can not link the speed of two nodes, attempt to use auxiliary braking chainIt connects, calculates and S is travelled using auxiliary brakingEFThe end speed V obtained after distancenextbIf VnextbWith VFSpeed is differed in 3km/hWithin, then it is assumed that auxiliary braking can be used up to the speed of two nodes of link, leap to step (6)；Otherwise give up the speedSpend node VEb。
(6) speed and target function value of each enabled node are saved
When calculating all qualified velocity nodes as initial velocity, the objective function of node N is driven to from node N-1Value；For example, being directed to velocity node VEb, utilize formula (6) calculating target function value JEb；Wherein formula (6) specifically considers vehicleFuel economy, the performances such as real-time can also be by user's self-defining.
JEb=QB*(BEbF/Beref)+QT*TimeEF/Timeref (6)
Wherein, BEbF, TimeEbFRespectively with initial velocity VEbNode F is driven to (by velocity node VEbAs initial velocity fromNode E neutral gear drives to node F, positive-torque drives to node F or negative torque drives to node F) consumed by amount of fuel andTime.QBAnd QTFor the weight coefficient of fuel economy and real-time, QBAnd QTIt is obtained using emulation or real train test debugging(test adjustment method is techniques well known)；BEbFIt can be obtained by the vehicle startup engine oil consumption map figure demarcated in advance；TimeEbF=SEF/VEb；BEFrefIt is cruise vehicle velocity V for speedEFrefWhen reference oil consumption from node E to node F, TimeEFrefForSpeed is cruise vehicle velocity VEFrefWhen from node E to the reference time of node F；TimeEFref=SEF/VEFref；
Save the state of velocity node Eb and Ec, including speed VEb, VEcWith target function value JEb, JEc, select wherein smallerThe corresponding velocity node of target function value is as optimal vehicle velocity VE, save the optimal vehicle velocity V of node EE；
If step (3)-(5) can not link the speed of two nodes, maximum value is set by target function value J(maximum value is setting, can be set to infinity).
(7) it iterates, calculates optimal velocity
Similarly, after the optimal velocity that node E has been calculated, the successively state of calculate node D, C, B, such as there are three D pointsVelocity node, needs to calculate by taking Da as an example the general objective functional value of Da-----Eb-----F and Da----Ec-----F, selection compared withSpeed Chain (assuming that being Da-----Eb-----F) where small catalogue scalar functions, then another chain can be rejected.Similarly successivelyThe optimal velocity chain where Db and Dc point is calculated, the optimal velocity and corresponding optimal catalogue scalar functions of each node are successively savedValue.
When calculating to second node B, as shown in Figure 4, Ba and Bd point can be rejected since velocity interval is not met, InB node can save two nodes Bb and Bc, select the smallest node of general objective functional value (by taking Bc as an example) in the two, then finally willVBcOptimal speed as optimization.When vehicle continues to move forward, repeat the above steps, real-time update subsequent time it is optimalSpeed.
(8) treatment on special problems
If the speed in any two non-first node can not be linked together by step (3) (4) (5), such as D pointWhen no matter velocity node can not all be linked to E point by which kind of mode, then using D point as final state point, then advise againIt draws.If the speed of first node A can not be linked together with the speed of B node, it is by the final speed planning of B nodeVA。