CN106227208B - A kind of intelligence control system and method based on multiple agent - Google Patents
A kind of intelligence control system and method based on multiple agent Download PDFInfo
- Publication number
- CN106227208B CN106227208B CN201610616961.4A CN201610616961A CN106227208B CN 106227208 B CN106227208 B CN 106227208B CN 201610616961 A CN201610616961 A CN 201610616961A CN 106227208 B CN106227208 B CN 106227208B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- escape
- server
- path
- stranded
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63G—MERRY-GO-ROUNDS; SWINGS; ROCKING-HORSES; CHUTES; SWITCHBACKS; SIMILAR DEVICES FOR PUBLIC AMUSEMENT
- A63G25/00—Autocar-like self-drivers; Runways therefor
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
Abstract
The invention discloses a kind of intelligence control system and method based on multiple agent, dodgem is completed along fixed route automatic Pilot, including positioning system, communication system, identifying system and server-side in the system coordination recreation ground；Communication system is made of the radio receiving transmitting module configured in each car, and radio receiving transmitting module includes the sending module for being placed in headstock and the receiving module for being placed in the tailstock；Positioning system includes the camera being placed in above athletic ground.Identifying system includes the binocular camera server-side reception positioning system for being placed in each car headstock and the live signal of radio receiving transmitting module transmission, does different disposal when receiving the automatic Pilot request or distress signal that vehicle issues.Excitement brought by rotation after present system and method allow dodgem joined the fresh experience of automatic Pilot and collide, and jam situation is efficiently solved by Flee Algorithm and automatic Pilot, allows player that can enjoy bigger enjoyment in finite time.
Description
Technical field
The present invention relates to a kind of amusement facility in recreation ground --- and dodgem, more particularly to one kind allow recreation ground
In dodgem " turning " intelligence control system and method got up.
Background technique
Dodgem is the amusement facility that recreation ground back warp can often be seen, but existing dodgem playing method cannot make it is each
Player obtains good experience, this is because it occur frequently that several vehicle congestions cannot dredge in time together in recreation ground
The case where opening；Furthermore existing playing method also has certain limitation, because existing dodgem is substantially manually, can be played
The technical restriction of family.
Summary of the invention
The present invention proposes the intelligence control system and method for a kind of recreation ground dodgem for the problems in background technique, can
Realize the automatic Pilot of dodgem.
The purpose of the present invention is achieved through the following technical solutions: a kind of intelligent control system based on multiple agent
System, the system coordination vehicle are completed along fixed route automatic Pilot, which includes positioning system, communication system, identifying system
And server-side；
The communication system is made of the radio receiving transmitting module configured in each car, and the radio receiving transmitting module includes setting
In headstock sending module and be placed in the receiving module of the tailstock, between all vehicles and between vehicle and server, pass through nothing
Line transceiver module is communicated.
The positioning system includes the camera being placed in above athletic ground, is believed by the position that camera obtains each car
Breath, and location information is sent to server-side.
The identifying system includes being placed in the binocular camera of each car headstock, obtains the vehicle and front vehicle by camera
Distance and adjacent vehicle, and the image Real-time Feedback for being taken camera by radio receiving transmitting module is to server-side.
The live signal that the server-side receives positioning system and radio receiving transmitting module is sent.When receive vehicle sending
When automatic Pilot is requested, the location information of image, positioning system transmission that server-side is sent according to identifying system and wireless receipts
The signal that module is sent is sent out, the specific location of controlled vehicle, the i.e. position of sending module and receiving module is obtained, is referred to according to the tailstock
To the vector of headstock, vehicle position information vectogram is obtained, it is vehicle body length, side that each car, which is abstracted into a length, on the diagram
It is driven to the vector for being directed toward headstock by the tailstock using the corresponding vector midpoint of each car as force analysis point according to automatic Pilot
Power algorithm obtains suffered resultant direction, and the current vector direction of controlled vehicle and resultant direction are projected to vector of position figure
On, the angle [alpha] that the angle between two directions less than 180 ° need to be turned over as stranded vehicle, to realize driving automatically for vehicle
It sails.When receiving the distress signal of vehicle sending, server-side determines the specific location and number of stranded vehicle in time, locks quilt
Tired region, and escape paths are determined by intelligent Flee Algorithm, so that the vehicle location and number on escape paths are obtained, and to
Vehicle on escape paths issues automatic Pilot instruction, realizes the escape of stranded vehicle.
Further, the automatic Pilot is carried out in the case where player does not select manual mode and gets congestion
, it can be achieved that controlled vehicle is allowed to knock the side for being arbitrarily designated target vehicle, obtain the experience of side crash；Target vehicle can pass through
Player is specified can also be specified with server-side；Wherein resultant force F and corner α is established a capital really using headstock towards path direction as target, F's
Size can arbitrarily be set.
Further, the escape paths are ray escape paths, realize escape by obtaining shortest path, specifically include
Following two:
1) forward Flee Algorithm: current vehicle is set as A, the adjacent vehicle of A is obtained by identifying system, when A is besieged
Send an SOS, signal is transmitted to outer layer adjacent vehicle by inner layer vehicle, until outermost layer vehicle can not communication target, i.e., at it
Outside without vehicle.Recorded while transmitting signal each ray propagation path L1, L2 ..., pass through on LN
Vehicle number N1, N2 ..., NN and corresponding car number, wherein N1, N2 ..., NN be that signal by A reaches outermost layer vehicle
The vehicle number that is passed through and including A and outermost layer vehicle, takes Nmin=min { N1, N2 ..., NN }, NminCorresponding ray is propagated
Path is used as most short escape paths.This escape method needs to be arranged escape radius, if the maximum length of vehicle itself is d, then
Escape radius R need to meet R >=0.5 × Nmax×d.Wherein Nmax=max { N1, N2 ..., NN }.
2) backward Flee Algorithm: when stranded vehicle A sends an SOS, signal outer layers ray is propagated, and is being passed
Sowing time need to only be remembered the path of the vehicle number and place that pass through on current propagation path and previous propagation path, relatively to take later
Maximum value, until the last item propagation path participates in after comparing, obtained maximum value is in forward Flee Algorithm
Nmax, escape radius is thus calculated, is then starting point along propagation path backpropagation using radius of escaping, until passing emergency vehicle back
A similarly need to only remember the road of the vehicle number and place that pass through on current propagation path and previous propagation path when propagating
Diameter is relatively minimized later, and determines the car number on the path of minimum value place by system, until the last item propagates road
Diameter participates in after comparing, and the propagation path where obtained minimum value is most short escape paths.
Instruction is issued from server-side to the vehicle on escape paths when escape, outermost layer vehicle is first escaped, true by system
The resultant force that target vehicle and its surrounding obstacles vehicle of fixed one beyond escape radius are constituted is as the vehicle automatic Pilot
The angle [alpha] that driving force F and determination need to turn over after outermost layer vehicle scuttles away, is successively escaped along the vehicle on escape paths opposite direction,
Until A is escaped from other than escape radius, it is believed that escape successfully.
Further, the escape paths are Rotating Escape path, specific as follows: this method, which is used to work as, has vehicle to be trapped in
Situation in corner, it is assumed that vehicle C1 is stranded in corners by other vehicles, and C1 sends an SOS, and server-side confirms vehicle C1
It postpones, the approximation circle that the central point of this several vehicles is surrounded is specified clockwise or counterclockwise as rotating path as rotation
Turn direction, the headstock of each car is first gone into the position towards specified direction of rotation, further according to the side of approximate circle rotating path
To the F and α for determining automatic Pilot, then C1 is produced corner along the tangential direction in path by the direction F, and another vehicle entering angle
It falls.If executed in primary rear setting time, there is no the distress signal sendings in the region, it is considered that escaping successfully；It is no
Then, switch escape mode.
Further, the location information is coordinate of the vehicle in region.
Further, the automatic Pilot driving force algorithm are as follows: when vehicle first is in automatic Pilot state, be by identification
Currently selected target vehicle a is modeled as the effect between object there are gravitation to the effect of vehicle first by system, by vehicle week
It encloses remaining vehicle and effect between object there are repulsion is modeled as to the effect of vehicle first, according to conjunction suffered by current vehicle first
Power determines the current driving direction of vehicle first, speed, and wherein the direction of resultant force is the direction of traffic of vehicle, the travel speed of vehicle
Change with the variation of resultant force size, to realize the automatic Pilot of vehicle；It is specific as follows: assuming that target vehicle a is to vehicle
The gravitation F1 of first be one with the reduction of distance between vehicle first and vehicle a reduced numerical value；Other vehicles of surrounding are to vehicle first
Repulsion be respectively F2, F3 ..., Fn, the size of these repulsion reduces with the increase of the distance in two workshops, thus available
The direction of resultant force F and size suffered by vehicle first, size be one with the reduction of distance between vehicle first and vehicle a reduced number
Value, then by Newton's second law F=ma it is found that with resultant force size variation, vehicle first is closer to target vehicle a, acceleration
Smaller, final velocity reaches maximum value, car to car impact.
Further, the target vehicle is that the controlled vehicle that system or player specify under automatic driving mode will be hit
The vehicle hit.
Further, the fixed route includes the escape paths obtained by Flee Algorithm, the vehicle energy in escape radius
The route and system for enough avoiding obstacle vehicle are controlled the travel route that vehicle tends to target vehicle under automatic driving mode.?
Escape radius in avoid obstacle vehicle be in order to prevent jam situation time of origin at a distance of too short.
A kind of intelligent control method based on multiple agent, method includes the following steps:
Step 1: when player select manual mode when, as system monitor in real time trolley location information and its corresponding to
Number；When player selects automatic mode, then other than real time monitoring, adjacent vehicle information that system is also fed back to by trolley
It determines target vehicle, and allows controlled Vehicular automatic driving collision target vehicle, which can voluntarily be formulated by player；
Step 2: when trolley collides, after the vehicle meeting self-rotating system collided sets fixing turn M, it is further continued for driving,
Driving situation includes following two: manual mode voluntarily controls steering direction；Automatic mode, under being found at the end of rotation is fast
One target vehicle, and moved at the end of rotation；Wherein speed is slack-off again from low to fast when rotation.
Step 3: when only one vehicle sends an SOS, which is sent to server-side, and server-side receives the letter
Number differentiate whether the vehicle is really stranded, and determines corresponding number and location information, if not some corner for being located at region,
So signal will be sent to outer layer adjacent vehicle, and escape by ray escape paths, while calculating escape radius, and real
When monitor.Escape vehicle will hide obstacle vehicle within escape radius, when the vehicle to send an SOS escapes from escape radius
When in addition, escape process terminates, and system enters step normal condition described in 1 and step 2；If the vehicle to send an SOS
Positioned at some corner in region, then escaping according to Rotating Escape path, and if there is no emergency in setting time
Signal is issued from the region, then the process of escaping terminates, and system enters normal condition；Otherwise, it is escaped with ray escape paths,
Until escape terminates.
Step 4: if more vehicles are simultaneously emitted by distress signal, signal is sent to server-side first, judges these signals
Whether issued by same stranded region, and execute different instruction according to different situations:
Situation one: if distress signal comes from same stranded region, and the region is not located at corner, then by mostly intelligent
Body control system calculated separately according to ray escape paths algorithm using the vehicle 1 of sending an SOS, 2 ..., X as starting point most
Short escape paths L1, L2 ..., LX, then select L1, L2 ..., in LX shortest escape paths as final escape road
Diameter；
Situation two: if distress signal comes from same stranded region, and the region is located at corner, then being obtained by system monitoring
The location information figure arrived determines really tired stranded vehicle in corners, then completes to escape according to Rotating Escape path；
Situation three: if distress signal not in same stranded region, i.e., the distance between two stranded vehicle center points is greater than
The sum of the two escape radius R, then escape simultaneously respectively according to the method in step 3；
Situation four: if the distance being located between two stranded vehicle center points of different zones is less than or equal to the two escape half
The sum of diameter R, it is considered that distress signal in same stranded region, when the region is not located at corner, is carried out according to situation one
Escape；Otherwise, it escapes according to situation two.
The beneficial effects of the present invention are: present system and method allow dodgem joined the fresh experience of automatic Pilot with
And excitement brought by rotation after collision, and jam situation is efficiently solved by Flee Algorithm and automatic Pilot, it allows
Player can enjoy bigger enjoyment in finite time.
Detailed description of the invention
Fig. 1 is the work flow diagram of intelligence control system and method for the invention；
Fig. 2 is the schematic diagram of the embodiment of the present invention 1；
Fig. 3 is the schematic diagram of the embodiment of the present invention 2.
Specific embodiment
With reference to the accompanying drawing and specific embodiment the present invention is further elaborated.
A kind of intelligence control system based on multiple agent provided by the invention, the system coordination vehicle are completed along specified circuit
Line automatic Pilot, the system include positioning system, communication system, identifying system and server-side；
The communication system is made of the radio receiving transmitting module configured in each car, and the radio receiving transmitting module includes setting
In headstock sending module and be placed in the receiving module of the tailstock, between all vehicles and between vehicle and server, pass through nothing
Line transceiver module is communicated.
The positioning system includes the camera being placed in above athletic ground, is believed by the position that camera obtains each car
Breath, and location information is sent to server-side.
The identifying system includes being placed in the binocular camera of each car headstock, obtains the vehicle and front vehicle by camera
Distance and adjacent vehicle, and the image Real-time Feedback for being taken camera by radio receiving transmitting module is to server-side.
The live signal that the server-side receives positioning system and radio receiving transmitting module is sent.When receive vehicle sending
When automatic Pilot is requested, the location information of image, positioning system transmission that server-side is sent according to identifying system and wireless receipts
The signal that module is sent is sent out, the specific location of controlled vehicle, the i.e. position of sending module and receiving module is obtained, is referred to according to the tailstock
To the vector of headstock, vehicle position information vectogram is obtained, it is vehicle body length, side that each car, which is abstracted into a length, on the diagram
It is driven to the vector for being directed toward headstock by the tailstock using the corresponding vector midpoint of each car as force analysis point according to automatic Pilot
Power algorithm obtains suffered resultant direction, and the current vector direction of controlled vehicle and resultant direction are projected to vector of position figure
On, the angle [alpha] that the angle between two directions less than 180 ° need to be turned over as stranded vehicle, to realize driving automatically for vehicle
It sails.When receiving the distress signal of vehicle sending, server-side determines the specific location and number of stranded vehicle in time, locks quilt
Tired region, and escape paths are determined by intelligent Flee Algorithm, so that the vehicle location and number on escape paths are obtained, and to
Vehicle on escape paths issues automatic Pilot instruction, realizes the escape of stranded vehicle.
Further, the automatic Pilot is carried out in the case where player does not select manual mode and gets congestion
, it can be achieved that controlled vehicle is allowed to knock the side for being arbitrarily designated target vehicle, obtain the experience of side crash；Target vehicle can pass through
Player is specified can also be specified with server-side；Wherein resultant force F and corner α is established a capital really using headstock towards path direction as target, F's
Size can arbitrarily be set.
Further, the escape paths are ray escape paths, realize escape by obtaining shortest path, specifically include
Following two:
1) forward Flee Algorithm: current vehicle is set as A, the adjacent vehicle of A is obtained by identifying system, when A is besieged
Send an SOS, signal is transmitted to outer layer adjacent vehicle by inner layer vehicle, until outermost layer vehicle can not communication target, i.e., at it
Outside without vehicle.Recorded while transmitting signal each ray propagation path L1, L2 ..., pass through on LN
Vehicle number N1, N2 ..., NN and corresponding car number, wherein N1, N2 ..., NN be that signal by A reaches outermost layer vehicle
The vehicle number that is passed through and including A and outermost layer vehicle, takes Nmin=min { N1, N2 ..., NN }, NminCorresponding ray is propagated
Path is used as most short escape paths.This escape method needs to be arranged escape radius, if the maximum length of vehicle itself is d, then
Escape radius R need to meet R >=0.5 × Nmax×d.Wherein Nmax=max { N1, N2 ..., NN }.
2) backward Flee Algorithm: when stranded vehicle A sends an SOS, signal outer layers ray is propagated, and is being passed
Sowing time need to only be remembered the path of the vehicle number and place that pass through on current propagation path and previous propagation path, relatively to take later
Maximum value, until the last item propagation path participates in after comparing, obtained maximum value is in forward Flee Algorithm
Nmax, escape radius is thus calculated, is then starting point along propagation path backpropagation using radius of escaping, until passing emergency vehicle back
A similarly need to only remember the road of the vehicle number and place that pass through on current propagation path and previous propagation path when propagating
Diameter is relatively minimized later, and determines the car number on the path of minimum value place by system, until the last item propagates road
Diameter participates in after comparing, and the propagation path where obtained minimum value is most short escape paths.
Instruction is issued from server-side to the vehicle on escape paths when escape, outermost layer vehicle is first escaped, true by system
The resultant force that target vehicle and its surrounding obstacles vehicle of fixed one beyond escape radius are constituted is as the vehicle automatic Pilot
The angle [alpha] that driving force F and determination need to turn over after outermost layer vehicle scuttles away, is successively escaped along the vehicle on escape paths opposite direction,
Until A is escaped from other than escape radius, it is believed that escape successfully.
Further, the escape paths are Rotating Escape path, specific as follows: this method, which is used to work as, has vehicle to be trapped in
Situation in corner, it is assumed that vehicle C1 is stranded in corners by other vehicles, and C1 sends an SOS, and server-side confirms vehicle C1
It postpones, the approximation circle that the central point of this several vehicles is surrounded is specified clockwise or counterclockwise as rotating path as rotation
Turn direction, the headstock of each car is first gone into the position towards specified direction of rotation, further according to the side of approximate circle rotating path
To the F and α for determining automatic Pilot, then C1 is produced corner along the tangential direction in path by the direction F, and another vehicle entering angle
It falls.If executed in primary rear setting time, there is no the distress signal sendings in the region, it is considered that escaping successfully；It is no
Then, switch escape mode.
Further, the location information is coordinate of the vehicle in region.
Further, the automatic Pilot driving force algorithm are as follows: when vehicle first is in automatic Pilot state, be by identification
Currently selected target vehicle a is modeled as the effect between object there are gravitation to the effect of vehicle first by system, by vehicle week
It encloses remaining vehicle and effect between object there are repulsion is modeled as to the effect of vehicle first, according to conjunction suffered by current vehicle first
Power determines the current driving direction of vehicle first, speed, and wherein the direction of resultant force is the direction of traffic of vehicle, the travel speed of vehicle
Change with the variation of resultant force size, to realize the automatic Pilot of vehicle；It is specific as follows: assuming that target vehicle a is to vehicle
The gravitation F1 of first be one with the reduction of distance between vehicle first and vehicle a reduced numerical value；Other vehicles of surrounding are to vehicle first
Repulsion be respectively F2, F3 ..., Fn, the size of these repulsion reduces with the increase of the distance in two workshops, thus available
The direction of resultant force F and size suffered by vehicle first, size be one with the reduction of distance between vehicle first and vehicle a reduced number
Value, then by Newton's second law F=ma it is found that with resultant force size variation, vehicle first is closer to target vehicle a, acceleration
Smaller, final velocity reaches maximum value, car to car impact.
Further, the target vehicle is that the controlled vehicle that system or player specify under automatic driving mode will be hit
The vehicle hit.
Further, the fixed route includes the escape paths obtained by Flee Algorithm, the vehicle energy in escape radius
The route and system for enough avoiding obstacle vehicle are controlled the travel route that vehicle tends to target vehicle under automatic driving mode.?
Escape radius in avoid obstacle vehicle be in order to prevent jam situation time of origin at a distance of too short.
The present invention also provides a kind of intelligent control methods based on multiple agent, as shown in Figure 1, this method includes following step
It is rapid:
Step 1: when player select manual mode when, by MAS control system monitor in real time the location information of trolley with
And its corresponding number；When player selects automatic mode, then other than real time monitoring, what system was also fed back to by trolley
Adjacent vehicle information determines target vehicle, and allows controlled Vehicular automatic driving collision target vehicle, which can also be by
Player voluntarily formulates；
Step 2: when trolley collides, after the vehicle meeting self-rotating system collided sets fixing turn M, it is further continued for driving,
Driving situation includes following two: manual mode voluntarily controls steering direction；Automatic mode, under being found at the end of rotation is fast
One target vehicle, and moved at the end of rotation.Wherein speed is slack-off again from low to fast when rotation；
Step 3: when only one vehicle sends an SOS, which is sent to server-side, and server-side receives the letter
Number differentiate whether the vehicle is really stranded, and determines corresponding number and location information, if not some corner for being located at region,
So signal will be sent to outer layer adjacent vehicle, and escape by ray escape paths, while calculating escape radius, and real
When monitor.Escape vehicle will hide obstacle vehicle within escape radius, when the vehicle to send an SOS escapes from escape radius
When in addition, escape process terminates, and system enters step normal condition described in 1 and step 2；If the vehicle to send an SOS
Positioned at some corner in region, then escaping according to Rotating Escape path, and if there is no ask in setting time 10s
It rescues signal to issue from the region, then the process of escaping terminates, and system enters normal condition；Otherwise, escaped with ray escape paths
Ease, until escape terminates.
Step 4: if more vehicles are simultaneously emitted by distress signal, signal is sent to server-side first, judges these signals
Whether issued by same stranded region, and execute different instruction according to different situations:
Situation one: if distress signal comes from same stranded region, and the region is not located at corner, then by mostly intelligent
Body control system calculated separately according to ray escape paths algorithm using the vehicle 1 of sending an SOS, 2 ..., X as starting point most
Short escape paths L1, L2 ..., LX, then select L1, L2 ..., in LX shortest escape paths as final escape road
Diameter；
Situation two: if distress signal comes from same stranded region, and the region is located at corner, then being obtained by system monitoring
The location information figure arrived determines really tired stranded vehicle in corners, then completes to escape according to Rotating Escape path；
Situation three: if distress signal not in same stranded region, i.e., the distance between two stranded vehicle center points is greater than
The sum of the two escape radius R, then escape simultaneously respectively according to the method in step 3；
Situation four: if the distance being located between two stranded vehicle center points of different zones is less than or equal to the two escape half
The sum of diameter R, it is considered that distress signal in same stranded region, when the region is not located at corner, is carried out according to situation one
Escape；Otherwise, it escapes according to situation two.
Embodiment 1
Here by taking forward Flee Algorithm as an example.If stranded vehicle A sends an SOS, system makes positioning to it, is in
Location information figure now as shown in Figure 2, discovery A not in corners, then by the forward Flee Algorithm in ray Flee Algorithm
Calculate most short escape paths, in calculating process, discovery have L1, L2 ..., L7 totally seven escape paths, and obviously L1 is
Most short escape paths, while obtaining NmaxIt is 3, then escape radius R >=0.5 × 3 × d, and general headquarters issue vehicle B and instruct later, and
Specifying the vehicle exceeded outside escape radius for it is to hit against target, the target vehicle and its surrounding obstacles determined by system
The resultant force that vehicle is constituted is as the driving force F and the angle [alpha] that need to turn over of determination of the vehicle automatic Pilot, and when B is scuttled away, A is with same
Method escape, until escaping from outside escape radius, escape terminates, and system restores normal condition.
Embodiment 2
If stranded vehicle C1 sends an SOS, control system real-time display is stranded zone position information, and discovery C1 is located at
One corner of whole region, so selection Rotating Escape algorithm, as shown in figure 3, with the central point of vehicle C1, C2, C3 and C4
The approximation circle surrounded is path, specified to rotate clockwise direction, the headstock of C2, C3 is first turned 180 °, then according to approximation
Along the tangential direction in path, α is used for for resultant force F that the clockwise direction of circle rotating path determines automatic Pilot and the corner direction α, F
The headstock direction of vehicle is finely tuned, C1 is then produced into corner, and C3 enters corner, it is assumed here that this four vehicles are along rotary road
Diameter turns over 180 °, can arbitrarily rotate angle by default.If executed in primary rear 10s, there is no the emergency in the region
Signal issues, it is considered that escaping successfully；Otherwise, it is escaped with ray escape paths, embodiment 1 is copied to carry out.
Claims (9)
1. a kind of intelligence control system based on multiple agent, which is characterized in that dodgem is completed in the system coordination recreation ground
Along fixed route automatic Pilot, which includes positioning system, communication system, identifying system and server-side；
The communication system is made of the radio receiving transmitting module configured in each car, and the radio receiving transmitting module includes being placed in vehicle
The sending module of head and it is placed in the receiving module of the tailstock, between all vehicles and between vehicle and server, by wirelessly receiving
Hair module is communicated；
The positioning system includes the camera being placed in above athletic ground, obtains the location information of each car by camera,
And location information is sent to server-side；
The identifying system includes being placed in the binocular camera of each car headstock, obtains each car and front vehicles by camera
Distance and adjacent vehicle, and the image Real-time Feedback for being taken camera by radio receiving transmitting module is to server-side；
The live signal that the server-side receives positioning system and radio receiving transmitting module is sent；When receive vehicle sending it is automatic
When driving request, the location information and wireless receiving and dispatching mould of image, positioning system transmission that server-side is sent according to identifying system
The signal that block is sent, obtains the specific location of controlled vehicle, the i.e. position of sending module and receiving module, is directed toward vehicle according to the tailstock
Head vector, obtain vehicle position information vectogram, on the diagram each car be abstracted into a length be vehicle body length, direction by
The tailstock is directed toward the vector of headstock, using the corresponding vector midpoint of each car as force analysis point, is calculated according to automatic Pilot driving force
Method obtains suffered resultant direction, and the current vector direction of controlled vehicle and resultant direction are projected on vector of position figure, will
The angle [alpha] that angle between two directions less than 180 ° need to be turned over as stranded vehicle, to realize the automatic Pilot of vehicle；When connecing
When receiving the distress signal of vehicle sending, server-side determines the specific location and number of stranded vehicle in time, locks stranded region,
And escape paths are determined by intelligent Flee Algorithm, to obtain the vehicle location and number on escape paths, and to escape road
Vehicle on diameter issues automatic Pilot instruction, realizes the escape of stranded vehicle.
2. a kind of intelligence control system based on multiple agent according to claim 1, which is characterized in that described to drive automatically
Sail is carried out in the case where player does not select manual mode and gets congestion, it can be achieved that controlled vehicle is allowed to knock arbitrarily
The side of specified target vehicle, obtains the experience of side crash；Target vehicle can also be specified by the way that player is specified with server-side；Its
Middle resultant force F and corner α is established a capital really using headstock towards path direction as target, and the size of F can arbitrarily be set.
3. a kind of intelligence control system based on multiple agent according to claim 1, which is characterized in that the escape road
Diameter is ray escape paths, realizes escape by obtaining shortest path, specifically includes following two:
1) forward Flee Algorithm: current vehicle is set as A, the adjacent vehicle of A is obtained by identifying system, the sending when A is besieged
Distress signal, signal are transmitted to outer layer adjacent vehicle by inner layer vehicle, until outermost layer vehicle can not communication target, i.e., in the outer of it
Side is without vehicle；Recorded while transmitting signal each ray propagation path L1, L2 ..., the vehicle that passes through on LN
Number N1, N2 ..., NN and corresponding car number, wherein N1, N2 ..., NN reaches outermost layer vehicle by A by signal and passes through
The vehicle number crossed and including A and outermost layer vehicle, takes Nmin=min { N1, N2 ..., NN }, NminCorresponding ray propagates road
Diameter is used as most short escape paths；This escape method needs to be arranged escape radius and then escapes if the maximum length of vehicle itself is d
Ease radius R need to meet R >=0.5 × Nmax×d；Wherein Nmax=max { N1, N2 ..., NN }；
2) backward Flee Algorithm: when stranded vehicle A sends an SOS, signal outer layers ray is propagated, when propagating
It need to only remember the path of the vehicle number and place that pass through on current propagation path and previous propagation path, relatively to take maximum later
Value, until the last item propagation path participates in after comparing, obtained maximum value is the N in forward Flee Algorithmmax,
Thus escape radius is calculated, is then starting point along propagation path backpropagation using radius of escaping, until passing emergency vehicle A back, together
Reason need to only remember the path of the vehicle number and place that pass through on current propagation path and previous propagation path when propagating, than
It is minimized after relatively, and the car number on the path of minimum value place is determined by system, until the last item propagation path is joined
After compared with, the propagation path where obtained minimum value is most short escape paths；
Instruction is issued from server-side to the vehicle on escape paths when escape, outermost layer vehicle is first escaped, determined by system
Driving of the resultant force that one target vehicle beyond escape radius and its surrounding obstacles vehicle are constituted as Vehicular automatic driving
The angle [alpha] that power F and determination need to turn over after outermost layer vehicle scuttles away, is successively escaped along the vehicle on escape paths opposite direction, until
A is escaped from other than escape radius, it is believed that is escaped successfully.
4. a kind of intelligence control system based on multiple agent according to claim 1, which is characterized in that the escape road
Diameter is Rotating Escape path, specific as follows: having the case where vehicle is stranded in corners for working as, it is assumed that vehicle C1 is by other vehicles
Tired C1 sends an SOS in corners, after server-side confirms the position vehicle C1, the central point of this several vehicles is surrounded close
It is used as rotating path like circle, specify is direction of rotation clockwise or counterclockwise, and the headstock of each car is first gone to direction and is referred to
The position of fixed direction of rotation determines the F and α of automatic Pilot further according to the direction of approximate circle rotating path, and the direction F is along path
Then C1 is produced corner by tangential direction, and another vehicle enters corner；If executed in primary rear setting time, there is no this
Distress signal in region issues, it is considered that escaping successfully；Otherwise, switch escape mode.
5. a kind of intelligence control system based on multiple agent according to claim 1, which is characterized in that the position
Information is coordinate of the vehicle in region.
6. a kind of intelligence control system based on multiple agent according to claim 1, which is characterized in that described to drive automatically
Sail driving force algorithm are as follows: when vehicle first is in automatic Pilot state, by identifying system by currently selected a pairs of target vehicle
The effect of vehicle first is modeled as the effect between object there are gravitation, remaining vehicle of vehicle periphery simulates the effect of vehicle first
There are the effect of repulsion between object, according to resultant force suffered by current vehicle first determine the current driving direction of vehicle first,
Speed, wherein the direction of resultant force is the direction of traffic of vehicle, and the travel speed of vehicle changes with the variation of resultant force size, from
And realize the automatic Pilot of vehicle；It is specific as follows: assuming that target vehicle a is one with vehicle first and vehicle to the gravitation F1 of vehicle first
The reduction of distance between a and reduced numerical value；Around other vehicles to the repulsion of vehicle first be respectively F2, F3 ..., Fn, this
The size of a little repulsion is reduced with the increase of the distance in two workshops, so that the direction of resultant force F and size suffered by vehicle first can be obtained,
Size be one with the reduction of distance between vehicle first and vehicle a reduced numerical value, then by Newton's second law F=ma it is found that
With the variation of resultant force size, for vehicle first closer to target vehicle a, acceleration is smaller, and final velocity reaches maximum value, two vehicle phases
It hits.
7. a kind of intelligence control system based on multiple agent according to claim 2, which is characterized in that the target
Vehicle is the controlled vehicle vehicle to be hit that system or player specify under automatic driving mode.
8. a kind of intelligence control system based on multiple agent according to claim 1, which is characterized in that the specified circuit
Line includes that the escape paths obtained by Flee Algorithm, vehicle can avoid the route and system of obstacle vehicle in escape radius
The travel route that vehicle tends to target vehicle is controlled under automatic driving mode；Escape radius in avoid obstacle vehicle be for
Prevent the jam situation time of origin at a distance of too short.
9. a kind of intelligent control method based on multiple agent, which is characterized in that method includes the following steps:
Step 1: when player selects manual mode, the location information of trolley and the volume corresponding to it being monitored in real time as system
Number；When player selects automatic mode, then other than real time monitoring, the adjacent vehicle information that system is also fed back to by trolley is true
Set the goal vehicle, and allows controlled Vehicular automatic driving collision target vehicle, which can voluntarily be formulated by player；
Step 2: when trolley collides, after the vehicle meeting self-rotating system collided sets fixing turn M, being further continued for driving, drive
Situation includes following two: manual mode voluntarily controls steering direction；Automatic mode finds next mesh at the end of rotation is fast
Vehicle is marked, and is moved at the end of rotation；
Step 3: when only one vehicle sends an SOS, which is sent to server-side, and server-side receives the signal and sentences
Whether the vehicle not sent an SOS is really stranded, and determines corresponding number and location information, if not positioned at region
Some corner, then signal will be sent to outer layer adjacent vehicle, and escape by ray escape paths, at the same calculate escape
Escape radius, and monitors in real time；Escape vehicle will hide obstacle vehicle within escape radius, when the vehicle to send an SOS is escaped
When out other than escape radius, escape process terminates, and system enters step the normal condition of 1 and step 2；If sent an SOS
Vehicle be located at some corner in region, then escaping according to Rotating Escape path, and if in setting time no longer
There is distress signal to issue from the region, then the process of escaping terminates, and system enters normal condition；Otherwise, with ray escape paths into
Row escape, until escape terminates；
Step 4: if more vehicles are simultaneously emitted by distress signal, signal is sent to server-side first, whether judges these signals
It is issued by same stranded region, and executes different instruction according to different situations:
Situation one: if distress signal comes from same stranded region, and the region is not located at corner, then by multiple agent control
System processed calculated separately according to ray escape paths algorithm using the vehicle 1 of sending an SOS, 2 ..., X escapes as the most short of starting point
Escape path L1, L2 ..., LX, then select L1, L2 ..., in LX shortest escape paths as final escape paths；
Situation two: if distress signal comes from same stranded region, and the region is located at corner, then obtained by system monitoring
Location information figure determines really tired stranded vehicle in corners, then completes to escape according to Rotating Escape path；
Situation three: if distress signal not in same stranded region, i.e., the distance between two stranded vehicle center points is greater than the two
Escape the sum of radius R, then escapes simultaneously respectively according to the method in step 3；
Situation four: if the distance being located between two stranded vehicle center points of different zones is less than or equal to the two escape radius R
The sum of, it is considered that distress signal in same stranded region, when the region is not located at corner, is escaped according to situation one
Ease；Otherwise, it escapes according to situation two.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610616961.4A CN106227208B (en) | 2016-07-29 | 2016-07-29 | A kind of intelligence control system and method based on multiple agent |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610616961.4A CN106227208B (en) | 2016-07-29 | 2016-07-29 | A kind of intelligence control system and method based on multiple agent |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106227208A CN106227208A (en) | 2016-12-14 |
| CN106227208B true CN106227208B (en) | 2019-04-19 |
Family
ID=57534941
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610616961.4A Active CN106227208B (en) | 2016-07-29 | 2016-07-29 | A kind of intelligence control system and method based on multiple agent |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106227208B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107065700A (en) * | 2017-06-03 | 2017-08-18 | 合肥佳洋电子科技有限公司 | A kind of intelligence control system |
| CN109141449A (en) * | 2018-07-02 | 2019-01-04 | 中国计量大学 | Automated guided vehicle most becate shape method for path navigation and guide transport lorry |
| CN110898433B (en) * | 2019-11-28 | 2021-09-03 | 腾讯科技（深圳）有限公司 | Virtual object control method and device, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002068076A2 (en) * | 2001-02-21 | 2002-09-06 | Mattel, Inc. | Children's ride-on vehicle with electronic speed control |
| CN1750957A (en) * | 2003-02-24 | 2006-03-22 | 达恩·加维什 | Method and apparatus for a mobile ride-on toy |
| CN202028178U (en) * | 2011-01-04 | 2011-11-09 | 吕昌运 | Automatic driving dodgem |
| CN104477167A (en) * | 2014-11-26 | 2015-04-01 | 浙江大学 | Intelligent driving system and control method thereof |
| CN105377385A (en) * | 2013-03-15 | 2016-03-02 | 国际海洋工程公司 | Trackless vehicle and system for synchronous control of trackless vehicle |
| WO2016039989A1 (en) * | 2014-09-10 | 2016-03-17 | Universal City Studios Llc | Systems and methods for controlling the transportation of vehicles |
| CN205340106U (en) * | 2015-12-23 | 2016-06-29 | 好孩子儿童用品有限公司 | Bumper car |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150041230A1 (en) * | 2013-08-12 | 2015-02-12 | GKart Inc. | Amusement vehicle, amusement environment for a vehicle and method of using the same |
-
2016
- 2016-07-29 CN CN201610616961.4A patent/CN106227208B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002068076A2 (en) * | 2001-02-21 | 2002-09-06 | Mattel, Inc. | Children's ride-on vehicle with electronic speed control |
| CN1750957A (en) * | 2003-02-24 | 2006-03-22 | 达恩·加维什 | Method and apparatus for a mobile ride-on toy |
| CN202028178U (en) * | 2011-01-04 | 2011-11-09 | 吕昌运 | Automatic driving dodgem |
| CN105377385A (en) * | 2013-03-15 | 2016-03-02 | 国际海洋工程公司 | Trackless vehicle and system for synchronous control of trackless vehicle |
| WO2016039989A1 (en) * | 2014-09-10 | 2016-03-17 | Universal City Studios Llc | Systems and methods for controlling the transportation of vehicles |
| CN104477167A (en) * | 2014-11-26 | 2015-04-01 | 浙江大学 | Intelligent driving system and control method thereof |
| CN205340106U (en) * | 2015-12-23 | 2016-06-29 | 好孩子儿童用品有限公司 | Bumper car |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106227208A (en) | 2016-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101960237B1 (en) | Multi-passenger driving car | |
| CN106227208B (en) | A kind of intelligence control system and method based on multiple agent | |
| KR20210121161A (en) | Autonomous and user-controlled vehicle calls to cross-reference targets for related applications | |
| JP5471462B2 (en) | Automatic parking equipment | |
| CN106394545A (en) | Driving system, unmanned vehicle and vehicle remote control terminal | |
| EP3678919A1 (en) | Systems and methods for coordinated lane-change negotiations between vehicles | |
| CN106467106A (en) | System and method for driver assistance | |
| WO2018207778A1 (en) | Automated parking system, automated parking vehicle, and automated parking method | |
| JPH10105880A (en) | Mobile object control system | |
| CN106706342A (en) | Testing method for continuous crossroad passing ability of automatic vehicle and testing field | |
| JPWO2018235273A1 (en) | Parking control method and parking control device | |
| WO2018135508A1 (en) | Vehicle operation management system and vehicle operation management method | |
| CN108733026A (en) | Dynamic remote control relocation method based on the degree of approach with vehicle and equipment | |
| US20150041230A1 (en) | Amusement vehicle, amusement environment for a vehicle and method of using the same | |
| KR102256070B1 (en) | Golf cart system capable of changing traveling mode according to whether golf course field can enter or not and golf cart control method using the system | |
| CZ26196A3 (en) | System, consisting of a mobile vehicle, preferably a golf cart and a transmitter | |
| JP6869172B2 (en) | How to operate an autonomous mobile | |
| CN105984469A (en) | Vehicle and method of controlling the same | |
| KR20160050669A (en) | Stationary bicycle system for wireless toy car control | |
| US11724710B2 (en) | Systems and methods for implementing a hybrid control for an autonomous vehicle | |
| JP2021064205A (en) | Vehicle and remote control system | |
| US20110025542A1 (en) | Integration Interface of a Remote Control Toy and an Electronic Game | |
| JPH06142339A (en) | Game device | |
| WO2021229881A1 (en) | Travel route generation device, travel route generation method, and automatic driving system | |
| CN111123733B (en) | Automatic driving simulation method, device, equipment and computer readable medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |