CN110530384A - A kind of vehicle path planning method using mixing potential field ant group algorithm - Google Patents
A kind of vehicle path planning method using mixing potential field ant group algorithm Download PDFInfo
- Publication number
- CN110530384A CN110530384A CN201910643608.9A CN201910643608A CN110530384A CN 110530384 A CN110530384 A CN 110530384A CN 201910643608 A CN201910643608 A CN 201910643608A CN 110530384 A CN110530384 A CN 110530384A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- potential field
- path
- speed
- road
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3453—Special cost functions, i.e. other than distance or default speed limit of road segments
- G01C21/3492—Special cost functions, i.e. other than distance or default speed limit of road segments employing speed data or traffic data, e.g. real-time or historical
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Abstract
The invention discloses a kind of using mixing potential field-ant group algorithm vehicle path planning method, step 1: obtaining vehicle itself and environmental information;Step 2: by analysis vehicle itself and environmental information, obtaining the current suffered virtual potential field power of vehicle, establish improved vehicle virtual potential field power model;Step 3: the relevant parameter of initialization with Ant colony algorithm;Step 4: path planning initial stage introduces improved virtual potential field resultant force in ant group algorithm and is used as heuristic information function;Introduce weight coefficient;Improve the transition function of ant group algorithm;Step 5: playing a role jointly with going deep into for obstacle-avoiding route planning, pheromone concentration and apart from inspiring;The effect for improving potential field resultant force is gradually decreased, carries out path planning using improved ant group algorithm;Step 6: being controlled by an electronic control unit speed and steering wheel angle according to the path that step 4 and 5 are planned.The present invention effectively avoids the stagnation of path search process, improves route searching efficiency.
Description
Technical field
The present invention relates to path planning and actuating mechanism controls decision Integrated Algorithms under a kind of complicated dynamic operation condition, belong to
Path Planning Technique field.
Background technique
With the continuous development of science and technology, the environment that Path Planning Technique is faced will be increasingly complex changeable.This will
Ask path planning algorithm that there is the ability for responding rapidly to complex environment variation.Ant group algorithm uses concurrent operation and positive feedback machine
System, it is the path planning algorithm of comparative maturity that search capability is strong, and robustness is good, but there are initial operating stage search randomness and
Blindness, easy the disadvantages of stagnating, the present invention propose mixing potential field-ant group algorithm, virtual potential field are introduced on the basis of ant group algorithm
Resultant force heuristic information function fully considers coefficient of road adhesion and car speed pair during ant group algorithm is used for path planning
The influence of path planning plays the advantage of two kinds of algorithms, effectively avoids the randomness searched for, blindness and easily stagnates scarce
Point.The characteristics of being constrained for path planning and control decision by intelligent vehicle dynamics, kinematics characteristic and traffic rules, the present invention
Using path planning and control method as research object, by improve virtual potential field and the planning of hybrid ant colony realizing route and
The unification of control decision.
Summary of the invention
The present invention introduces improved virtual potential field resultant force heuristic information on the basis of ant group algorithm is used for path planning
Function;Weight coefficient is introduced, the influence of the coefficient of road adhesion and Vehicle Speed of planning path to Path selection is considered, subtracts
The blindness and randomness at few path planning initial stage, effectively prevent the stagnation of path search process, improve route searching effect
Rate;And it tracks in driving process on the control of speed, steering wheel angle in path so that vehicle is not influencing riding comfort
Under the premise of be capable of the traveling of safety and stability.Concrete scheme is as follows:
It is a kind of to use mixing ant colony-improvement potential field algorithm vehicle path planning method, include the following steps:
Step 1: obtaining vehicle itself and environmental information;
Step 2: by analysis vehicle itself and environmental information, obtaining the current suffered virtual potential field power of vehicle, establish and improve
Vehicle virtual potential field power model afterwards;
Step 3: the relevant parameter of initialization with Ant colony algorithm;
Step 4: path planning initial stage introduces in ant group algorithm at this time since the pheromone concentration of route searching is insufficient
Improved virtual potential field resultant force is used as heuristic information function;Weight coefficient is introduced, considers the coefficient of road adhesion of planning path
Influence with Vehicle Speed to avoidance Path selection improves the transition function of ant group algorithm, make it by distance and
Improve the dual evocation of virtual potential field resultant force;Reduce the number of iterations of the improved ant group algorithm for route searching, In
It improves and guarantees path planning direction accuracy under the action of virtual potential field resultant force, reduce the blindness at path planning initial stage and random
Property.The stagnation for effectively preventing path search process improves avoidance route searching efficiency;
Step 5: playing a role jointly with going deep into for path planning, pheromone concentration and apart from inspiring;At this point, gradually dropping
The low effect for improving potential field resultant force carries out path planning using improved ant group algorithm;
Step 6: being controlled by an electronic control unit speed and steering wheel angle according to the path that step 4 and 5 are planned;
Step 7: when vehicle along planning path close to barrier and when starting avoidance, speed is controlled by electronic control unit
Within 40km/h, when far from barrier, speed restores normal;During path planning always by steering wheel angle control-
In 23 °~+23 ° (being positive clockwise), so as to improve vehicle safety and riding comfort.
Step 8: after the unit time, repeating step 1 to 7 until vehicle reaches safety place.
Further, vehicle environmental information described in step 1 is obtained by millimetre-wave radar and CCD industrial camera, is obtained
The information taken includes: current driving road front obstacle information;Current sensor check frequency information;Current traffic signal lamp
Information;The lateral obstacle information of current driving road;Aiming spot information.The vehicle self information is sensed by vehicle body
Device and GPS are obtained, and acquired information includes: car speed, steering wheel angle, current vehicle position.
Further, improved vehicle virtual potential field power model described in step 2 includes:
(1) the virtual repulsion model of front obstacle:
In formula, freject-frontIt is front vehicles to the repulsion coefficient of rear car;μroadIt is attached for road ahead-tire of identification
Put forth effort characteristic information;viIt (t) is intelligent vehicle speed;β (t) is intelligent vehicle mass center deflection angle;For intelligent vehicle relative to
The speed of front obstacle;Relative distance for intelligent vehicle relative to front obstacle is that barrier and vehicle theory are pacified
Full distance, Ssafe-fronThe safe distance needed for colliding with front obstacle is avoided for intelligent vehicle.
(2) check frequency repulsion model:
In formula, freject-blindIt is check frequency to the repulsion coefficient of intelligent vehicle;It is intelligent vehicle relative to blind area
Relative velocity;Sm-s-oIt is that other vehicles avoid safe distance needed for colliding with intelligent vehicle;It is intelligent vehicle relative to blind area side
The relative distance on boundary;μroadFor road ahead-tire adhesion force characteristic information of identification;Ds-bIt is managed for intelligent vehicle and dead-zone boundary
By safe distance.
(3) traffic lights signal-virtual repulsion model:
In formula, fr-lightIt is traffic lights to the repulsion coefficient of intelligent vehicle;MlightFor traffic lights information, Mlight
=1 indicates that traffic lights are red light;Relative velocity for intelligent vehicle relative to traffic lights;SsafeFor intelligence
The minimum range that vehicle can stop before stop line;Relative distance for intelligent vehicle relative to traffic lights;
μroadFor road ahead-tire adhesion force characteristic information of identification.
(4) the lateral virtual repulsion model of intelligent vehicle:
In formula, freject-yIt is lateral barrier (guardrail, shrub, pedestrian etc.) to the repulsion coefficient of intelligent vehicle;For intelligence
Relative velocity of the vehicle relative to lateral barrier;viThe component that sin β (t) is intelligent vehicle speed on lateral;θ (t) is intelligent vehicle
Mass center deflection angle;It is the intelligent vehicle left and right sides at a distance from nearest object;μroadFor road ahead-tire attachment of identification
Force characteristic information;SsafSafe distance required for colliding with lateral barrier is avoided for intelligent vehicle.
Further, step 3 initialization with Ant colony algorithm relevant parameter, Basic Ant Group of Algorithm transition function are as follows:
In formula:Indicate node j pheromone concentration;α is the pheromone concentration factor;β is the heuristic information factor;
Allowed k (t) indicates to remove the node that can choose in next step after accessed node;Indicate node m to node n
Apart from heuristic information function;dmnIndicate the distance between node m to node n.
Ant completes one cycle, and the pheromone concentration in each path is updated using following formula:
τmn(t+ δ)=(1- ρ) τmn(t)+Δτmn
In formula: ρ (0 < ρ < 1) indicates pheromone concentration volatility coefficient;ΔτmnIt indicates to believe on path (m, n) in this circulation
Cease plain concentration increments;ΔτmnIndicate that kth ant stays in the pheromone concentration Q on path (m, n) in this circulation as pheromones
The factor;LkFor the path length of kth Ant Search.
Further, the pheromone concentration in planning path is first determined whether in step 4, if concentration is low, will be introduced after improving
Virtual potential field resultant force heuristic information function;Weight coefficient is introduced, considers the coefficient of road adhesion and vehicle driving of planning path
Influence of the speed to Path selection.
From semiempirical road-tire mathematical model:
μ represents coefficient of road adhesion;S is slip rate;c1, c2, c3It is the constant obtained by experiment, see the table below:
Using wheel and road surface contact point as origin, the road surface on four direction (45 °, 135 °, 225 °, 315 °) is taken to adhere to system
Number, estimates current road attachment coefficient peak value are as follows:
In formula, λ1、λ2、λ3、λ4Indicate weight coefficient,Indicate current road attachment coefficient, μ1max、μ2max、μ3max、μ4max
Indicate the peak adhesion coefficient on the four direction of contact point between wheel and road surface contact point,It respectively indicates adjacent
The attachment coefficient on road surface, the above parameter meet following relationship:Vehicle Speed is classified,
Judge path planning process whether driving safety:
Wherein A: traveling is too fast;B: traveling is normal;C: low running speed is further
The transition function of ant group algorithm is improved are as follows:
In formula, dmnShow node m to node n distance;dngShow node n to target point distance;It indicates from node m to node
The variable quantity of n speed;It indicates apart from heuristic information function;It indicates to improve potential field resultant force heuristic information function;γ
For the heuristic information factor;KmaxFor maximum number of iterations;K is current iteration number;A > 1 is constant;FtotalIndicate potential field resultant force;
C1、C2Indicate weight coefficient;μmaxRepresent maximum coefficient of road adhesion;LvIndicate Vehicle Speed grade;θ indicates potential field resultant force
With the angle in optional path direction, angle is smaller,Bigger, transition probability is bigger, easier to be mobile towards the path direction.
Further, go deep into step 5 with path planning, play pheromone concentration and distance, speed heuristic information
Effect, gradually decreases the effect for improving virtual potential field resultant force, prevents intelligent vehicle from advancing all along potential field gradient direction.After improvement
Ant colony transition function so that intelligent vehicle is advanced with maximum probability to target point, reduce the blindness of route searching and random
Property, improve the efficiency of path planning.
Further, step 6 is controlled by an electronic control unit speed according to the path that step 4,5 are planned and steering wheel turns
Angle.
Further, in step 7 when intelligent vehicle along planning path close to barrier and when starting avoidance, by electronic control unit
By speed control within 30km/h, when far from barrier, speed restores normal;Steering wheel is turned always during path planning
Angle control in -23 °~+23 ° (being positive) clockwise, so as to improve vehicle safety and riding comfort.The control of speed
Simulation are as follows:
Do=Dse+Ds
In formula, λ is the constant coefficient less than 1, usually takes λ=0.4~0.6, DoFor automobile between barrier at a distance from;DseFor
The distance that intelligent vehicle travels in this section of braking process from initial velocity to safe speed;DsFor reserved safe distance;vsFor avoidance
Speed when braking starts;veSpeed when for end of braking;A (t) is braking deceleration, is the linear function of time;tdFor vapour
Vehicle reaches the time used in maximum value from starting to brake to braking acceleration.
From rate control module as can be seen that automobile is after barriers to entry object coverage, speed can with obstacle
The reduction of distance between object and reduce, when automobile cut-through object, speed increases with the increase of automobile and obstacle distance,
When automobile is driven out to the coverage of barrier, speed will be no longer by the control of model.
And if cause automobile to lose stabilization since side acceleration is excessive during turning to avoidance when automobile, tune can be passed through
Control of the steering wheel angle realization to body gesture is saved, guarantees safety and the comfort of vehicle driving.
Beneficial effects of the present invention:
(1) present invention introduces improvement potential field resultant forces to be used as route searching of the heuristic information factor for path planning initial stage,
The improvement ant group algorithm of proposition can be reduced the blindness and randomness at path planning initial stage for vehicle path planning, effectively avoid
The stagnation of path search process, improves route searching efficiency.
(2) present invention and consider that coefficient of road adhesion and speed change the influence to path planning, improve ant colony calculation
The probability transfer function of method makes vehicle advance with maximum probability towards target point;
(3) control during vehicle control proposed by the present invention to Vehicle Speed and steering wheel angle, effectively mentions
The safety and stability of vehicle driving during high path planning.
Detailed description of the invention
Fig. 1 is flow chart of the method for the present invention:
Fig. 2 is the schematic diagram of virtual potential field power suffered by vehicle;
Fig. 3 is ant group algorithm schematic diagram;
Fig. 4 is potential field resultant force and optional path relation schematic diagram in improved ant group algorithm transition function;
Fig. 5 is vehicle driving-cycle schematic diagram;
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.
The present invention provides a kind of use and mixes ant colony-improvement potential field algorithm vehicle path planning method, as shown in Figure 1,
Including the following steps:
Step 1: obtaining vehicle itself and environmental information.
Vehicle itself and environmental information are obtained by millimetre-wave radar and CCD industrial camera, and millimetre-wave radar is set as four
A, wherein 1 splits in front of the car bumper bar middle position, other two is individually positioned between the front door of two sides and back door
Middle position, the last one is placed on the tail portion of vehicle, for detecting obstacle information and transmission on vehicle four direction
Electron controls list ECU.CCD industrial camera is mounted on front windshield of vehicle over top, for by the situation of vehicle front
It is transferred to electronic control unit ECU.Acquired environmental information includes: current driving road front obstacle information;Work as forward pass
Sensor check frequency information;Current traffic signal information;The lateral obstacle information of current driving road;Aiming spot letter
Breath.The vehicle self information is obtained by vehicle body sensor and GPS, and acquired information includes: car speed, steering wheel turn
Angle, current vehicle position.
Step 2: establishing fictitious force model suffered in vehicle travel process, as shown in Figure 2, comprising:
(1) the virtual repulsion model of front obstacle:
In formula, freject-fronIt is front vehicles to the repulsion coefficient of rear car;μroadFor road ahead-tire attachment of identification
Force characteristic information;viIt (t) is intelligent vehicle speed;β (t) is intelligent vehicle mass center deflection angle;It is intelligent vehicle relative to preceding
The speed of square barrier;Relative distance for intelligent vehicle relative to front obstacle is barrier and vehicle theory α coefficient
Distance, Ssafe-fronThe safe distance needed for colliding with front obstacle is avoided for intelligent vehicle.
(2) check frequency repulsion model:
In formula, freject-blinIt is check frequency to the repulsion coefficient of intelligent vehicle;It is phase of the intelligent vehicle relative to blind area
To speed;Sm-s-oIt is that other vehicles avoid safe distance needed for colliding with intelligent vehicle;It is intelligent vehicle relative to dead-zone boundary
Relative distance;μroadFor road ahead-tire adhesion force characteristic information of identification;Ds-blindIt is managed for intelligent vehicle and dead-zone boundary
By safe distance.
(3) traffic lights signal-virtual repulsion model:
In formula, fr-ligIt is traffic lights to the repulsion coefficient of intelligent vehicle;MlightFor traffic lights information, Mlight=
1 indicates that traffic lights are red light;Relative velocity for intelligent vehicle relative to traffic lights;SsafeFor intelligent vehicle energy
Enough minimum ranges that can stop before stop line;Relative distance for intelligent vehicle relative to traffic lights;μroadFor
The road ahead of identification-tire adhesion force characteristic information.
(4) the lateral virtual repulsion model of intelligent vehicle:
In formula, freject-yIt is lateral barrier (guardrail, shrub, pedestrian etc.) to the repulsion coefficient of intelligent vehicle;For intelligence
Relative velocity of the vehicle relative to lateral barrier;ViThe component that sin β (t) is intelligent vehicle speed on lateral;θ (t) is intelligent vehicle
Mass center deflection angle;It is the intelligent vehicle left and right sides at a distance from nearest object;μroadFor road ahead-tire attachment of identification
Force characteristic information;Ssafe-Safe distance required for colliding with lateral barrier is avoided for intelligent vehicle.
Step 3: initialization improves potential field method and ant group algorithm relevant parameter, such as population quantity, cycle-index, heuristic information
Parameter, pheromone concentration, route searching taboo list etc..
Step 4: as shown in figure 3, path planning initial stage, pheromone concentration is insufficient, introduces improved virtual potential field resultant force and makees
For the heuristic information function of route searching, weight coefficient is introduced, considers the coefficient of road adhesion and vehicle row of avoidance planning path
Influence of the speed to Path selection is sailed, the transition function of ant group algorithm is improved, make it by distance and improves virtual potential field
The dual evocation of resultant force;The number of iterations of the improved ant group algorithm for route searching is reduced, in virtual potential field resultant force
Under the action of guarantee path planning direction accuracy, reduce path planning initial stage blindness and randomness.Improved ant colony
Potential field resultant force and the selection in path can be indicated with Fig. 4 in algorithm transition function.As seen from the figure, potential field resultant direction with can
Routing diameter angular separation is smaller,Bigger, transition probability is bigger, i.e., easier to be mobile towards the direction of path 1..
Go deep into step 5 with path planning, plays the effect of pheromone concentration and distance, speed heuristic information, by
The effect for gradually reducing virtual potential field resultant force prevents intelligent vehicle from advancing all along potential field gradient direction.
Step 6 is controlled by an electronic control unit speed and steering wheel angle according to the path that step 4,5 are planned.
Step 7: after the unit time, repeating step 1 to 6 until vehicle reaches safety place.
By taking the driving cycle in Fig. 5 as an example:
(1) vehicle itself and environmental information are obtained.
(2) by analysis vehicle itself and environmental information, virtual potential field power model is established, vehicle current driving environment is obtained
Lower suffered fictitious force.
(3) initialization of virtual potential field method and ant group algorithm relevant parameter.
(4) path planning initial stage introduces virtual potential field resultant force and is used as road since the pheromone concentration in planning path is insufficient
The heuristic information of path search improves the transition function of ant group algorithm, makes it by the dual of distance and virtual potential field resultant force
Evocation;Guarantee path planning direction accuracy under the action of virtual potential field resultant force, reduces the blindness at route searching initial stage
Property and randomness.
(5) virtual potential field power model is established, plans vehicle part driving path.
The virtual repulsion being subject in the process of moving from vehicle has: lateral barrier repulsion, lateral lane line repulsion, detection are blind
Area (pedestrian, vehicle etc.) repulsion, front vehicles repulsion and traffic lights repulsion;Under the collective effect of these fictitious forces, vehicle
The potential field established along fictitious force is moved from high potential energy point to low-potential energy point.
(6) whether setting speed threshold value and steering wheel angle threshold decision speed of operation and steering wheel angle are more than setting
Threshold value, if being more than the threshold value of setting, the unsafe conditions such as sideslip, whipping can occur for vehicle;Therefore it is controlled by an electronic control unit vehicle
In planning path driving process by speed and steering wheel angle control in the reasonable scope, guarantee the operation stabilization of vehicle
Property, improve driving safety and comfort.
(7) after the unit time, step (1) to (6) are repeated until vehicle reaches safety place.
Mixing potential field-ant group algorithm vehicle local paths planning method proposed by the present invention, with the rule of fictitious force model
Basis of the information as particle swarm algorithm optimizing is drawn, potential field resultant force is introduced and the part heuristic information of vehicle search path point is used as to close
Reason rapidly cooks up local driving path, and the blindness for reducing ant group algorithm at route searching initial stage improves path with this
Planning efficiency;The potential street accidents risks as brought by check frequency are reduced to improve the safety coefficient of driving;Introduce rule
The speed control in driving process and steering wheel angle control are drawn, vehicle handling stability in driving process, safety are improved
Property and comfort.
The series of detailed descriptions listed above only for feasible embodiment of the invention specifically
Protection scope bright, that they are not intended to limit the invention, it is all without departing from equivalent implementations made by technical spirit of the present invention
Or change should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of using mixing potential field-ant group algorithm vehicle path planning method, which comprises the steps of:
Step 1: obtaining vehicle itself and environmental information;
Step 2: by analysis vehicle itself and environmental information, obtaining the current suffered virtual potential field power of vehicle, establish improved
Vehicle virtual potential field power model;
Step 3: the parameter of initialization with Ant colony algorithm;
Step 4: path planning initial stage introduces improved virtual potential field resultant force in ant group algorithm and is used as heuristic information function;
The influence of coefficient of road adhesion and Vehicle Speed to Path selection for planning path introduces weight coefficient, improves ant
The transition function of group's algorithm makes it by distance and improves the dual evocation of virtual potential field resultant force;It reduces after improving
Ant group algorithm be used for route searching the number of iterations;
Step 5: playing a role jointly with going deep into for path planning, pheromone concentration and apart from inspiring;Change at this point, gradually decreasing
Into the effect of potential field resultant force, path planning is carried out using improved ant group algorithm;
Step 6: being controlled by an electronic control unit speed and steering wheel angle according to the path that step 4 and 5 are planned;
Step 7: when vehicle along planning path close to barrier and when starting avoidance, speed control is existed by electronic control unit
Within 40km/h, when far from barrier, speed restores normal;
Step 8: after the unit time, repeating step 1 to 7 until vehicle reaches safety place.
2. according to claim 1 a kind of using mixing potential field-ant group algorithm vehicle path planning method, feature exists
In environmental information described in step 1 includes: current driving road front obstacle information;Current sensor check frequency letter
Breath;Current traffic signal information;The lateral obstacle information of current driving road;Aiming spot information;The vehicle is certainly
Body information includes: car speed, steering wheel angle, current vehicle position.
3. according to claim 1 a kind of using mixing potential field-ant group algorithm vehicle path planning method, feature exists
In improved vehicle virtual potential field power model described in step 2 includes:
(1) the virtual repulsion model of front obstacle:
In formula, freject-frontIt is front vehicles to the repulsion coefficient of rear car;μroadIt is special for road ahead-tire adhesion force of identification
Property information;viIt (t) is intelligent vehicle speed;β (t) is intelligent vehicle mass center deflection angle;Hinder for intelligent vehicle relative to front
Hinder the speed of object;Relative distance for intelligent vehicle relative to front obstacle is barrier and vehicle theory α coefficient away from
From Ssafe-frontThe safe distance needed for colliding with front obstacle is avoided for intelligent vehicle;
(2) check frequency repulsion model:
In formula, freject-blinIt is check frequency to the repulsion coefficient of intelligent vehicle;It is intelligent vehicle relative to the relatively fast of blind area
Degree;Sm-s-oIt is that other vehicles avoid safe distance needed for colliding with intelligent vehicle;It is phase of the intelligent vehicle relative to dead-zone boundary
It adjusts the distance;μroadFor road ahead-tire adhesion force characteristic information of identification;Ds-blindPacify for intelligent vehicle and dead-zone boundary theory
Full distance;
(3) traffic lights signal-virtual repulsion model:
In formula, fr-lighT is repulsion coefficient of the traffic lights to intelligent vehicle;MlightFor traffic lights information, Mlight=1 table
Show that traffic lights are red light;Relative velocity for intelligent vehicle relative to traffic lights;SsafeIt can for intelligent vehicle
The minimum range that can stop before stop line;Relative distance for intelligent vehicle relative to traffic lights;μroadTo distinguish
The road ahead of knowledge-tire adhesion force characteristic information;
(4) the lateral virtual repulsion model of intelligent vehicle:
In formula, freject-yIt is lateral barrier to the repulsion coefficient of intelligent vehicle;Phase for intelligent vehicle relative to lateral barrier
To speed;viThe component that sin β (t) is intelligent vehicle speed on lateral;β (t) is intelligent vehicle mass center deflection angle;For intelligence
At left and right sides of vehicle at a distance from nearest object;μroadFor road ahead-tire adhesion force characteristic information of identification;SsafeFor intelligence
Vehicle avoids safe distance required for colliding with lateral barrier.
4. according to claim 1 a kind of using mixing potential field-ant group algorithm vehicle path planning method, feature exists
In in step 3, initialization with Ant colony algorithm relevant parameter includes:
Basic Ant Group of Algorithm transition function are as follows:
In formula:Indicate node j pheromone concentration;α is the pheromone concentration factor;β is the heuristic information factor;allowed
K (t) indicates to remove the node that can choose in next step after accessed node;Indicate that the distance of node m to node n open
Photos and sending messages function;dmnIndicate the distance between node m to node n.
Ant completes one cycle, and the pheromone concentration in each path is updated using following formula:
τmn(t+ δ)=(1- ρ) τmn(t)+Δτmn
In formula: ρ (0 < ρ < 1) indicates pheromone concentration volatility coefficient;ΔτmnIndicate in this circulation pheromones on path (m, n)
Concentration increments;ΔτmnIndicate that kth ant stays in the pheromone concentration Q on path (m, n) in this circulation as information prime factor;
LkFor the path length of kth Ant Search.
5. according to claim 1 a kind of using mixing potential field-ant group algorithm vehicle path planning method, feature exists
In the concrete methods of realizing of step 4 includes:
It first determines whether the pheromone concentration in planning path, if concentration is low, introduces improved virtual potential field resultant force and inspire letter
Cease function;Introduce weight coefficient, with consider planning path coefficient of road adhesion and Vehicle Speed to the shadow of Path selection
It rings;
From semiempirical road-tire mathematical model:
μ represents coefficient of road adhesion;S is slip rate;c1, c2, c3It is the constant obtained by experiment.
Using wheel and road surface contact point as origin, the coefficient of road adhesion on four direction (45 °, 135 °, 225 °, 315 °) is taken,
Estimate current road attachment coefficient peak value are as follows:
In formula, λ1、λ2、λ3、λ4Indicate weight coefficient,Indicate current road attachment coefficient, μ1max、μ2max、μ3max、μ4maxExpression connects
Peak adhesion coefficient on the four direction of contact between wheel and road surface contact point,Respectively indicate adjacent road surface
Attachment coefficient, the above parameter meet following relationship:
Vehicle Speed is classified, judge obstacle-avoiding route planning process whether driving safety:
Wherein
A: traveling is too fast;B: traveling is normal;C: low running speed.
6. according to claim 5 a kind of using mixing potential field-ant group algorithm vehicle path planning method, feature exists
In in the step 4, the transition function of ant group algorithm is improved are as follows:
In formula, dmnShow node m to node n distance;dngShow node n to target point distance;It indicates from node m to node n speed
Variable quantity;It indicates apart from heuristic information function;It indicates to improve potential field resultant force heuristic information function;γ is to inspire
The information factor;KmaxFor maximum number of iterations;K is current iteration number;A > 1 is constant;FtotalIndicate potential field resultant force;C1、C2
Indicate weight coefficient;μmaxRepresent maximum coefficient of road adhesion;LvIndicate Vehicle Speed grade;θ indicate potential field resultant force with can
The angle of path direction is selected, angle is smaller,Bigger, transition probability is bigger, easier to be mobile towards the path direction.
7. according to claim 1 a kind of using mixing potential field-ant group algorithm vehicle path planning method, feature exists
In in the step 7, always by steering wheel angle control in -23 °~+23 ° during path planning.
8. according to claim 1 a kind of using mixing potential field-ant group algorithm vehicle path planning method, feature exists
In, in the step 7, the Controlling model of speed are as follows:
Do=Dse+Ds
In formula, λ is the constant coefficient less than 1, usually takes λ=0.4~0.6, DoFor automobile between barrier at a distance from;DseFor from first
The beginning speed distance that intelligent vehicle travels into this section of braking process of safe speed;DsFor reserved safe distance;vsFor avoidance braking
Speed when beginning;veSpeed when for end of braking;A (t) is braking deceleration, is the linear function of time;tdFor automobile from
Start braking and reaches the time used in maximum value to braking acceleration.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910643608.9A CN110530384B (en) | 2019-07-17 | 2019-07-17 | Vehicle path planning method adopting hybrid potential field ant colony algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910643608.9A CN110530384B (en) | 2019-07-17 | 2019-07-17 | Vehicle path planning method adopting hybrid potential field ant colony algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110530384A true CN110530384A (en) | 2019-12-03 |
CN110530384B CN110530384B (en) | 2022-10-28 |
Family
ID=68661712
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910643608.9A Active CN110530384B (en) | 2019-07-17 | 2019-07-17 | Vehicle path planning method adopting hybrid potential field ant colony algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110530384B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111784079A (en) * | 2020-07-28 | 2020-10-16 | 上海海事大学 | Unmanned aerial vehicle path planning method based on artificial potential field and ant colony algorithm |
CN111914981A (en) * | 2020-05-31 | 2020-11-10 | 吉林大学 | Improved PI model identification method based on particle swarm-ant colony parallel cross algorithm |
CN113985896A (en) * | 2021-12-03 | 2022-01-28 | 中国重汽集团济南动力有限公司 | Obstacle avoidance path planning method for automatic driving vehicle, vehicle and readable storage medium |
CN117109625A (en) * | 2023-10-20 | 2023-11-24 | 湖南大学 | Unmanned path planning method based on improved PRM algorithm |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016045615A1 (en) * | 2014-09-25 | 2016-03-31 | 科沃斯机器人有限公司 | Robot static path planning method |
CN105549597A (en) * | 2016-02-04 | 2016-05-04 | 同济大学 | Unmanned vehicle dynamic path programming method based on environment uncertainty |
CN107917711A (en) * | 2017-11-14 | 2018-04-17 | 重庆邮电大学 | A kind of Robot Path Planning Algorithm based on optimization hybrid ant colony |
CN109282815A (en) * | 2018-09-13 | 2019-01-29 | 天津西青区瑞博生物科技有限公司 | Method for planning path for mobile robot based on ant group algorithm under a kind of dynamic environment |
CN109978286A (en) * | 2019-05-07 | 2019-07-05 | 中国民用航空飞行学院 | It is a kind of to be diversion thunderstorm Route planner based on the more aircrafts for improving ant group algorithm |
-
2019
- 2019-07-17 CN CN201910643608.9A patent/CN110530384B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016045615A1 (en) * | 2014-09-25 | 2016-03-31 | 科沃斯机器人有限公司 | Robot static path planning method |
CN105549597A (en) * | 2016-02-04 | 2016-05-04 | 同济大学 | Unmanned vehicle dynamic path programming method based on environment uncertainty |
CN107917711A (en) * | 2017-11-14 | 2018-04-17 | 重庆邮电大学 | A kind of Robot Path Planning Algorithm based on optimization hybrid ant colony |
CN109282815A (en) * | 2018-09-13 | 2019-01-29 | 天津西青区瑞博生物科技有限公司 | Method for planning path for mobile robot based on ant group algorithm under a kind of dynamic environment |
CN109978286A (en) * | 2019-05-07 | 2019-07-05 | 中国民用航空飞行学院 | It is a kind of to be diversion thunderstorm Route planner based on the more aircrafts for improving ant group algorithm |
Non-Patent Citations (3)
Title |
---|
HUI WANG等: "Ant colony optimization with improved potential field heuristic for robot path planning", 《PROCEEDINGS OF THE 37TH CHINESE CONTROL CONFERENCE》 * |
张强等: "基于改进势场蚁群算法的移动机器人最优路径规划", 《农业机械学报》 * |
王晓燕等: "基于改进势场蚁群算法的机器人路径规划", 《控制与决策》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111914981A (en) * | 2020-05-31 | 2020-11-10 | 吉林大学 | Improved PI model identification method based on particle swarm-ant colony parallel cross algorithm |
CN111914981B (en) * | 2020-05-31 | 2022-11-08 | 吉林大学 | Improved PI model identification method based on particle swarm optimization-ant swarm optimization parallel cross algorithm |
CN111784079A (en) * | 2020-07-28 | 2020-10-16 | 上海海事大学 | Unmanned aerial vehicle path planning method based on artificial potential field and ant colony algorithm |
CN113985896A (en) * | 2021-12-03 | 2022-01-28 | 中国重汽集团济南动力有限公司 | Obstacle avoidance path planning method for automatic driving vehicle, vehicle and readable storage medium |
CN113985896B (en) * | 2021-12-03 | 2023-12-08 | 中国重汽集团济南动力有限公司 | Automatic driving vehicle obstacle avoidance path planning method, vehicle and readable storage medium |
CN117109625A (en) * | 2023-10-20 | 2023-11-24 | 湖南大学 | Unmanned path planning method based on improved PRM algorithm |
CN117109625B (en) * | 2023-10-20 | 2024-01-16 | 湖南大学 | Unmanned path planning method based on improved PRM algorithm |
Also Published As
Publication number | Publication date |
---|---|
CN110530384B (en) | 2022-10-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110530384A (en) | A kind of vehicle path planning method using mixing potential field ant group algorithm | |
US11755012B2 (en) | Alerting predicted accidents between driverless cars | |
CN106926844B (en) | A kind of dynamic auto driving lane-change method for planning track based on real time environment information | |
CN106994969B (en) | A kind of fleet's formation control loop and method | |
CN109035862A (en) | A kind of more vehicles collaboration lane-change control method based on truck traffic | |
CN105722738B (en) | The drive dynamic control device of moving body | |
US10759432B2 (en) | Vehicle control apparatus, vehicle control method, and vehicle control program | |
US11072331B2 (en) | Vehicle control device, vehicle control method and vehicle control program | |
CN106515710B (en) | The automatic Pilot control device of vehicle | |
US20180201271A1 (en) | Vehicle control device, vehicle control method, and vehicle control program | |
US20170235307A1 (en) | Vehicle control system, vehicle control method, and vehicle control program | |
US20160288789A1 (en) | Adaptive cruise control system | |
KR101798178B1 (en) | Method and system for adaptive cruise control and vehicle | |
Chen et al. | Forward collision warning system considering both time-to-collision and safety braking distance | |
CN107226089A (en) | A kind of pilotless automobile collision avoidance strategy | |
US10246090B2 (en) | Vehicle collision severity mitigation | |
CN105160917A (en) | Vehicle and road collaborative information-based signalized intersection vehicle speed guidance system and method | |
CN107891860A (en) | System and method based on road curvature automatic adjusument speed | |
US9669831B2 (en) | Automatic driving controlling system and method | |
US20180043793A1 (en) | Smart cruise control and adas for range extension | |
CN110949378B (en) | Anti-collision rear-end collision method for automobile under high-speed working condition | |
CN109760678A (en) | A kind of method for limiting speed of automotive self-adaptive cruise system | |
CN107010064A (en) | A kind of fleet's formation drive manner and system | |
KR20180096114A (en) | Control Method of Autonomous Vehicle | |
US20150039201A1 (en) | Method of Controlling a Differential Lock |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |