CN104867329B - Vehicle state prediction method of Internet of vehicles - Google Patents

Abstract

The invention relates to a vehicle state prediction method of Internet of vehicles. The method comprises the following steps: 1), obtaining vehicle environment information of a target vehicle and information of traffic lights in front of the target vehicle; 2), according to the vehicle environment information and historical state information of the target vehicle, predicting vehicle behavior of the target vehicle; and 3), according to the vehicle environment information, the information of the traffic lights in front of the target vehicle and the vehicle behavior obtained through prediction, through combination with the historical state information of the target vehicle, predicting the state of the vehicle. Compared to the prior art, the method provided by the invention has the following advantages: the vehicle behavior of the vehicle is predicted based on the vehicle environment information and the historical state of the vehicle, then the future state of the vehicle is predicted through the vehicle behavior, the prediction based on the vehicle behavior is more accurate, and the method is good in expandability, is simple and highly-efficient and is suitable for large-scale network conditions.

Description

A kind of car networking vehicle-state Forecasting Methodology

Technical field

The present invention relates to a kind of vehicle-state Predicting Technique, more particularly, to a kind of car networking vehicle-state Forecasting Methodology.

Background technology

Vehicle-state includes position, speed and the acceleration of vehicle, and the movement properties of these three vehicles are in car networking Communication between vehicle and vehicle, the communication between vehicle and infrastructure are most important, and it is to grind to study these three operation attributes Study carefully the important foundation of network connectivty in car networking.Whether position relationship is covered in mutual infinite network to two communication nodes In the range of, i.e., whether two communication nodes can be communicated.Communication node can be that vehicle or static roadside basis set Apply.Speed and acceleration have influence on duration of two communication nodes in communication range, i.e. communication keeps duration.Therefore vehicle Position, speed and acceleration are the key factors for affecting car networking network connectivty.Traditional network routing protocol is not accounted for Vehicle location, speed and acceleration attribute in car networking, it is difficult to the characteristics of adapting to height of car in car networking and run, including The classical Mobile Ad-hoc network network agreements such as DSDV routes, DSR routes and AODV routes, Karp et al. propose location-based first The stateless route GPSR on Routing Protocol greediness border, all vehicle nodes periodically update its position to location server and believe Breath, when source node needs to send data to destination node, to location server request destination node position, is then turned with greedy Send out the nearest next-hop node of each chosen distance destination node of pattern.Due to its find next-hop when only according to straight line away from From, the problem of crossing barrier is not accounted for, GPCR has done an improvement to this, that is, find that crossing is present then direct during node The node is transmitted to, but crossing is difficult to judge with the presence or absence of node, crossing is judged easily with the presence or absence of node by map, So current main study hotspot concentrates on the Routing Protocol based on map.Geographical source routing GSR is one and is typically based on The Routing Protocol of map, which is also required to from location server the position for obtaining destination node, then using electronic chart and position Information calculates source node to the shortest path of the node of purpose by dijkstra's algorithm.Above based on position and the route of map Agreement make use of vehicle location to carry out data forwarding well, but not account for the characteristic of vehicle high-speed movement, due to car Position frequently change, forward node and destination node shift in position are very big so that data forwarding be likely to failure.GPSR is assisted In view single-hop node forwarding probably due to the shift in position of adjacent node and the research of failure, Alsaqour R et al. shows not Accurately position is substantially reduced the efficiency of location-based agreement, and proposes a kind of NWLBP (Neignbor for the problem Wireless link break prediction, adjacent node failure prediction) predicting the position of adjacent node, but the party Method is only capable of solving the problems, such as wall scroll adjacent node shift in position, it is impossible to solve the problems, such as that destination node changes, and GPSR agreements are with mesh The position of node be oriented to for route, therefore the position of destination node is most important.

Existing vehicle-state Predicting Technique is mainly used in vehicle and exempts from the system of hitting, according to track and the speed of surrounding vehicles Predict that its position, Hermes C et al. use position after trajectory predictions technological prediction vehicle 1-2s using the historical position information of vehicle Put.And use track sorting technique, extraction training is carried out to a large amount of vehicle running orbits, by different tracks be divided into left-hand rotation, Various behavior characteristicss such as right-hand rotation, then carry out pattern recognition classifier to existing track and are predicted.Dagdelen M,Reymond G,Kemeny A,et al.Model-based predictive motion cueing strategy for vehicle driving simulators[J].Control Engineering Practice,2009,17(9):995-1003 and Sorstedt J,Svensson L,Sandblom F,et al.A new vehicle motion model for improved predictions and situation assessment[J].Intelligent Transportation Systems,IEEE Transactions on,2011,12(4):Driver is turned left, turns right and become track etc. by 1209-1219 Drive subjective purpose and add forecast model.andita R,Caveney D S.Preceding vehicle state prediction:U.S.Patent Application 13/535,513 [P] .2012-6-28 proposes the car of some car compositions The vehicle-state Predicting Technique of team, is simulated using intelligent Driver Model and with garage is.Houenou A,Bonnifait P,Cherfaoui V,et al.Vehicle trajectory prediction based on motion model and maneuver recognition[C]//Intelligent Robots and Systems(IROS),2013IEEE/RSJ International Conference on.IEEE,2013:4363-4369 combines trajectory predictions technology and vehicle Activity recognition Realizing track of vehicle Predicting Technique, vehicle Activity recognition technology includes keeping track, changing Lane and steering etc. technology, can make With finite automaton state automata model or driving context identification technology.Such Predicting Technique only considers the vehicle of surrounding, number Amount is little, it is desirable to higher degree of accuracy.

The content of the invention

The purpose of the present invention is exactly to provide a kind of car networking vehicle shape to overcome the defect of above-mentioned prior art presence State Forecasting Methodology.

The purpose of the present invention can be achieved through the following technical solutions：

A kind of car networking vehicle-state Forecasting Methodology, including：

1) obtain the vehicle environmental information and the traffic lights information in front of target vehicle residing for target vehicle；

2) according to vehicle environmental information, and the vehicle behavior of the historic state information prediction target vehicle of target vehicle；

3) the vehicle behavior obtained according to the traffic lights information in front of vehicle environmental information, target vehicle and prediction, with reference to The state of the historic state information prediction vehicle of target vehicle.

Vehicle environmental information residing for the vehicle is specially：

Road=(long_c,inserction,vehicles)

Wherein：Road be vehicle environmental information, long_cFor the distance between target vehicle and junction ahead, Junction ahead information of the inserction for target vehicle, front vehicles information of the vehicles for target vehicle；

The junction ahead information is specially：

Inserction=(locations, lanes)

Wherein：Positional informationes of the locations for junction ahead, lanes are the track of each merging road at junction ahead Data,

The positional information of the junction ahead is specially：

Locations=[(x₁,y₁,θ₁),(x₂,y₂.θ₂),(x₃,y₃,θ₃),(x₄,y₄,θ₄)]

Wherein：x_iFor the longitude on i-th summit of junction ahead, y_iFor the latitude on i-th summit of junction ahead, θ_iFor front The angle of i-th of crossing, two merging roads of apex, i herein represent the sequence number on crossing summit,

At the junction ahead, the track data of each merging road are specially：

Lanes=[(forw₁,rev₁),(forw₂,rev₂),(forw₃,rev₃),(forw₄,rev₄)]

Wherein：forw_iFor the positive lane information of i-th road of junction ahead, rev_iFor i-th road of junction ahead Reverse lane information, i herein represent the sequence number of road；

The front vehicles information is specially：

Vehicles=[(S₁,lane_S1),(S₂,lane_S2),…(S_i,lane_Si),…(S_n,lane_Sn)]

Wherein：S_iFor the current state of i-th vehicle in front vehicles, lane_SiResiding for i-th vehicle in front vehicles Lane information；

The traffic lights information is specially：

Light=(sequence, long_L)

Wherein：Light be traffic lights information, long_LFor the distance between traffic lights and target vehicle, sequence is for it Traffic lights sequence in setting time Δ t afterwards, specially：

Sequence=[(color₁,time₁),…(color_i, time_i),…(color_m, time_m)]

Wherein：time_iFor color_iPersistent period, color_iFor the color state of traffic lights,

color_i∈{red,green,yellow}。

The step 2) specifically include step：

201) whether the time of target vehicle row to junction ahead is judged more than setting time Δ t, if it has, then performing step It is rapid 202), if it has not, then execution step is 205)；

202) whether the density of target vehicle place lane preceding vehicle is judged more than setting density threshold ρ *, if it is, The then vehicle behavior of target vehicle predicts the outcome as lane change, and execution step is 3), if it has not, then execution step is 203)；

203) judge target vehicle with the speed difference of nearest vehicle in its front vehicles whether more than setting speed difference limen value δ V, if it has, then the vehicle behavior of target vehicle predicts the outcome as lane change, and execution step is 3), if it has not, then execution step 204)；

204) the acceleration and deceleration behavior according to target vehicle and the range prediction vehicle of fore-aft vehicle, and execution step is 3)；

205) turning behavior of the vehicle at crossing is judged according to the historic state information of target vehicle.

The step 204) specifically include step：

241) judge whether target vehicle is more than safe distance with the distance between nearest vehicle in front vehicles, if Be, then execution step 243), if it has not, then execution step is 242)；

242) judge whether the speed of target vehicle is more than the speed of nearest vehicle in front vehicles, if it has, then target The vehicle behavior of vehicle predicts the outcome to slow down, and execution step is 3), if it has not, then execution step is 243)；

243) judge whether target vehicle is more than safe distance with the distance between nearest vehicle in front vehicle, if It is that the then vehicle behavior of target vehicle predicts the outcome as at the uniform velocity, and execution step is 3), if it has not, then execution step is 244)；

244) judge whether the speed of target vehicle is more than the speed of nearest vehicle in front vehicle, if it has, then target The vehicle behavior of vehicle predicts the outcome as at the uniform velocity, and execution step is 3), if it has not, the then prediction of the vehicle behavior of target vehicle As a result to accelerate, and execution step is 3)；

The step 242) in vehicle behavior predict the outcome for slow down when acceleration magnitude be：

Wherein：a₁Acceleration magnitude during to predict the outcome to slow down, d_ijFor target vehicle and nearest car in front vehicles The distance between, σ is constant, v_iFor the present speed size of target vehicle, v_jFor the current speed of nearest vehicle in front vehicles Degree size, d_itFor target vehicle and the distance between nearest vehicle in front vehicle.

The step 244) in vehicle behavior predict the outcome for accelerate when acceleration magnitude be：

Wherein：a₂Acceleration magnitude during to predict the outcome to accelerate, d_ijFor target vehicle and nearest car in front vehicles The distance between, α, β be constant, v_iFor the present speed size of target vehicle, v_jFor in front vehicles nearest vehicle it is current Velocity magnitude, v_tFor the present speed size of nearest vehicle in front vehicle.

The step 205) specifically include step：

251) the historic state information of target vehicle is traveled through forward from the current state of target vehicle, and determines start offset State TurnPoint, specially：

TurnPoint=min (g | Offset_k＞ T＆Offset_k-1＞ T＆ ..., Offset_g＞ T })

Wherein：

Offset_i=(OffsetP_i,OffsetV_i,OffsetA_i)

T=(TP, TV, TA)

Wherein：

OffsetP_i,OffsetV_i,OffsetA_iRespectively the i-th -1 state is to position, speed in i-th state and adds The side-play amount of speed, TP, TV, TA are constant；

252) judge persistently to offset whether number of times exceedes set point number：And whether acceleration direction migration included angle meets circumference Kinetic characteristic, specially judgment formula：

K-TurnPoint ＞ TK

w(OffsetA_k) ＜ TW＆w (OffsetA_k-1) ＜ TW ..., w (OffsetA_TurnPoint+1) ＜ TW

Wherein：Persistently to offset number of times, TK is set point number to k-TurnPoint, and TW is deviation angle tolerable error,

If the determination result is YES, then predict that target vehicle occurs to turn, and turn direction is target vehicle current state position The offset direction put, if judged result is no, prediction target vehicle does not occur to turn.

Compared with prior art, the present invention has advantages below：

1) vehicle behavior of the present invention based on vehicle environmental information and vehicle historic state to vehicle is predicted, then The to-be of vehicle is predicted by vehicle behavior, the prediction based on vehicle behavior is more accurate, and is had good Ground autgmentability, it is simple efficient, it is suitable for large scale network situation.

2) vehicle behaviour decision making, by conditional decision judge the acceleration of vehicle, deceleration, at the uniform velocity, change trains and the row such as turning To reach the purpose of the prediction of vehicle-state under the complexity road of predicted city with reference to current vehicle condition information and road information.

3) vehicle behaviour decision making distinguishes straight way and crossing is processed, and straight way considers the impact of surrounding vehicles, uses for reference artificial fish-swarm Method in algorithm to fish school behavior simulation, is simulated to vehicle behavior, vehicle's current condition and surrounding vehicles environment is entered Row modeling, derives effective conditional decision expression formula.Crossing situation is by the inclined of vehicle historical position, speed and acceleration Shifting is modeled, and derives the conditional decision whether vehicle turns with turn direction.

Description of the drawings

Fig. 1 is the key step schematic flow sheet of the present invention；

Fig. 2 is schematic diagram at crossing；

Fig. 3 is that crossing turning time calculates schematic diagram.

Specific embodiment

The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.The present embodiment is with technical solution of the present invention Premised on implemented, give detailed embodiment and specific operating process, but protection scope of the present invention be not limited to Following embodiments.

A kind of car networking vehicle-state Forecasting Methodology, as shown in figure 1, including：

1) obtain the vehicle environmental information and the traffic lights information in front of target vehicle residing for target vehicle；

Vehicle environmental information residing for vehicle is specially：

Road=(long_c,inserction,vehicles)

Wherein：Road be vehicle environmental information, long_cFor the distance between target vehicle and junction ahead, Junction ahead information of the inserction for target vehicle, front vehicles information of the vehicles for target vehicle；

As shown in Fig. 2 in the present embodiment, junction ahead is modeled as parallelogram, junction ahead information is specially：

Inserction=(locations, lanes)

Wherein：Positional informationes of the locations for junction ahead, lanes are the track of each merging road at junction ahead Data,

The positional information of junction ahead is specially：

Locations=[(x₁,y₁,θ₁),(x₂,y₂.θ₂),(x₃,y₃,θ₃),(x₄,y₄,θ₄)]

Wherein：x_iFor the longitude on i-th summit of junction ahead, y_iFor the latitude on i-th summit of junction ahead, θ_iFor front The angle of i-th of crossing, two merging roads of apex, i ∈ { 1,2,3,4 }, wherein θ₁As shown in Fig. 2

At junction ahead, the track data of each merging road are specially：

Lanes=[(forw₁,rev₁),(forw₂,rev₂),(forw₃,rev₃),(forw₄,rev₄)]

Wherein：forw_iFor the positive lane information of i-th road of junction ahead, rev_iFor i-th road of junction ahead Reversely lane information, i ∈ { 1,2,3,4 }, if the junction of three roads, can pass through to set the forward direction of a road in four roads Track and reverse number of lanes are 0, and the forward and reverse in track therein can be set according to the travel direction of target vehicle It is fixed, it is also possible to which that setting moved towards according to road；

Front vehicles information is specially：

Vehicles=[(S₁,lane_S1),(S₂,lane_S2),…(S_i,lane_Si),…(S_n,lane_Sn)]

Wherein：S_iFor the current state of i-th vehicle in front vehicles, lane_SiResiding for i-th vehicle in front vehicles Lane information, i is integer that span is 1～n；

Traffic lights information is specially：

Light=(sequence, long_L)

Wherein：Light be traffic lights information, long_LFor the distance between traffic lights and target vehicle, sequence is for it Traffic lights sequence in setting time Δ t afterwards, specially：

Sequence=[(color₁,time₁),…(color_i, time_i),…(color_m, time_m)]

Wherein：time_iFor color_iPersistent period, color_iFor the color state of traffic lights,

color_i∈ { red, green, yellow }, i are integer that span is 1～m.

2) according to vehicle environmental information, and the vehicle behavior of the historic state information prediction target vehicle of target vehicle, Specifically include step：

201) whether the time of target vehicle row to junction ahead is judged more than setting time Δ t, if it has, then judging car Straight way is travelled on, and execution step is 202), if it has not, then judge that vehicle travels on crossing, and execution step is 205)；

202) judge whether the density of the front vehicles in the visual range of target vehicle place track is more than setting density threshold Value ρ * (ρ * are specially set as needed, can be empirical value), if it has, then the vehicle behavior of target vehicle predicts the outcome For lane change, and execution step is 3), if it has not, then execution step is 203)；

203) judge target vehicle with the speed difference of nearest vehicle in its front vehicles whether more than setting speed difference limen value δ V, if it has, then the vehicle behavior of target vehicle predicts the outcome as lane change, and execution step is 3), if it has not, then execution step 204)；

204) the acceleration and deceleration behavior according to target vehicle and the range prediction vehicle of fore-aft vehicle, in the present embodiment, in order to Simplify prediction process, when carrying out acceleration and deceleration, the vehicle behavior of target vehicle predicts the outcome as at the uniform velocity, and execution step 3), step 204) specifically include step：

241) judge whether target vehicle is more than safe distance with the distance between nearest vehicle in front vehicles, if Be, then execution step 243), if it has not, then execution step is 242)；

242) judge whether the speed of target vehicle is more than the speed of nearest vehicle in front vehicles, if it has, then target The vehicle behavior of vehicle predicts the outcome to slow down, and execution step is 3), if it has not, then execution step is 243)；

Wherein predicting the outcome as acceleration magnitude during deceleration for vehicle behavior is：

Wherein：a₁Acceleration magnitude during to predict the outcome to slow down, d_ijFor target vehicle and nearest car in front vehicles The distance between, σ is constant, v_iFor the present speed size of target vehicle, v_jFor the current speed of nearest vehicle in front vehicles Degree size.

243) judge whether target vehicle is more than safe distance with the distance between nearest vehicle in front vehicle, if It is that the then vehicle behavior of target vehicle predicts the outcome as at the uniform velocity, and execution step is 3), if it has not, then execution step is 244)；

244) judge whether the speed of target vehicle is more than the speed of nearest vehicle in front vehicle, if it has, then target The vehicle behavior of vehicle predicts the outcome as at the uniform velocity, and execution step is 3), if it has not, the then prediction of the vehicle behavior of target vehicle As a result to accelerate, and execution step is 3)；

Wherein predicting the outcome as acceleration magnitude during acceleration for vehicle behavior is：

Wherein：a₂Acceleration magnitude during to predict the outcome to accelerate, d_ijFor target vehicle and nearest car in front vehicles The distance between, α, β be constant, v_iFor the present speed size of target vehicle, v_jFor in front vehicles nearest vehicle it is current Velocity magnitude, v_tFor the present speed size of nearest vehicle in front vehicle, d_itFor target vehicle and nearest car in front vehicle The distance between.

205) turning behavior of the vehicle at crossing is judged according to the historic state information of target vehicle, specifically includes step Suddenly, specifically include step：

251) the historic state information of target vehicle is traveled through forward from the current state of target vehicle, and determines start offset State TurnPoint, specially：

TurnPoint=min (g | Offset_k＞ T＆Offset_k-1＞ T＆ ..., Offset_g＞ T })

Wherein：

Offset_i=(OffsetP_i,OffsetV_i,OffsetA_i)

T=(TP, TV, TA)

Wherein：

OffsetP_i,OffsetV_i,OffsetA_iRespectively the i-th -1 state is to position, speed in i-th state and adds The side-play amount of speed, TP, TV, TA are constant；

252) judge persistently to offset whether number of times exceedes set point number：And whether acceleration direction migration included angle meets circumference Kinetic characteristic, specially judgment formula：

K-TurnPoint ＞ TK

w(OffsetA_k) ＜ TW＆w (OffsetA_k-1) ＜ TW ..., w (OffsetA_TurnPoint+1) ＜ TW

Wherein：Persistently to offset number of times, TK is set point number to k-TurnPoint, and TW is deviation angle tolerable error, k The sequence number of target vehicle current state is represented,

If the determination result is YES, then predict that target vehicle occurs to turn, and turn direction is target vehicle current state position The offset direction put, if judged result is no, prediction target vehicle does not occur to turn.

253) predict turning time of origin t_turn.Offset by acceleration angle, calculate turning rate ω, by crossing Angle calcu-lation deviation angle, so as to estimate the time for occurring of turning.As shown in figure 3, the angle occurred when turning from crossing is inclined Move and crossing angle is deducted for 180 degree, i.e.,

ω=(w (OffsetA_k)+w(OffsetA_k-1)+,....,w(OffsetA_TurnPoint+1))/(k-TurnPoint) Δt

t_turn=(180 ° of-θ)/ω

254) in turning time of origin t_turnInterior acceleration, speed and change in location, calculate according to circular motion rule , acceleration magnitude is constant, and direction at the uniform velocity offsets, and speed is obtained by acceleration and Time Calculation, and position is by speed and time It is calculated.

3) the vehicle behavior obtained according to the traffic lights information in front of vehicle environmental information, target vehicle and prediction, with reference to The state of the historic state information prediction vehicle of target vehicle, specially：

s_k+1=g (s_k,Road,Light,Behavior)

Wherein：s_k+1To predict the vehicle-state for obtaining, s_kFor the current vehicle-state of target vehicle, Behavior is pre- The vehicle behavior for measuring.

Claims (8)

1. a kind of car networking vehicle-state Forecasting Methodology, it is characterised in that include：

1) obtain the vehicle environmental information and the traffic lights information in front of target vehicle residing for target vehicle；

2) according to vehicle environmental information, and the vehicle behavior of the historic state information prediction target vehicle of target vehicle；

3) the vehicle behavior obtained according to the traffic lights information in front of vehicle environmental information, target vehicle and prediction, combining target The state of the historic state information prediction vehicle of vehicle.

2. a kind of car networking vehicle-state Forecasting Methodology according to claim 1, it is characterised in that residing for the vehicle Vehicle environmental information is specially：

Road=(long_c,inserction,vehicles)

Wherein：Road be vehicle environmental information, long_cFor the distance between target vehicle and junction ahead, inserction is mesh The junction ahead information of mark vehicle, front vehicles information of the vehicles for target vehicle；

The junction ahead information is specially：

Inserction=(locations, lanes)

Wherein：Positional informationes of the locations for junction ahead, lanes are the track data of each merging road at junction ahead,

The positional information of the junction ahead is specially：

Locations=[(x₁,y₁,θ₁),(x₂,y₂.θ₂),(x₃,y₃,θ₃),(x₄,y₄,θ₄)]

Wherein：x_iFor the longitude on i-th summit of junction ahead, y_iFor the latitude on i-th summit of junction ahead, θ_iFor junction ahead The angle of i-th apex, two merging roads,

At the junction ahead, the track data of each merging road are specially：

Lanes=[(forw₁,rev₁),(forw₂,rev₂),(forw₃,rev₃),(forw₄,rev₄)]

Wherein：forw_iFor the positive lane information of i-th road of junction ahead, rev_iFor the reverse of i-th road of junction ahead Lane information；

The front vehicles information is specially：

Vehicles=[(S₁,lane_S1),(S₂,lane_S2),…(S_i,lane_Si),…(S_n,lane_Sn)]

Wherein：S_iFor the current state of i-th vehicle in front vehicles, lane_SiCar residing for i-th vehicle in front vehicles Road information.

3. a kind of car networking vehicle-state Forecasting Methodology according to claim 2, it is characterised in that the traffic lights information Specially：

Light=(sequence, long_L)

Wherein：Light be traffic lights information, long_LFor the distance between traffic lights and target vehicle, sequence is to set afterwards The traffic lights sequence fixed time in Δ t, specially：

Sequence=[(color₁,time₁),…(color_i, time_i),…(color_m, time_m)]

Wherein：time_iFor color_iPersistent period, color_iFor the color state of traffic lights,

color_i∈{red,green,yellow}。

4. a kind of car networking vehicle-state Forecasting Methodology according to claim 3, it is characterised in that the step 2) it is concrete Including step：

201) whether the time of target vehicle row to junction ahead is judged more than setting time Δ t, if it has, then execution step 202), if it has not, then execution step is 205)；

202) whether the density of target vehicle place lane preceding vehicle is judged more than setting density threshold ρ *, if it has, then mesh The vehicle behavior of mark vehicle predicts the outcome as lane change, and execution step is 3), if it has not, then execution step is 203)；

203) judge whether target vehicle and the speed difference of nearest vehicle in its front vehicles are more than setting speed difference limen value δ v, if It is yes, then the vehicle behavior of target vehicle predicts the outcome as lane change, and execution step is 3), if it has not, then execution step is 204)；

204) the acceleration and deceleration behavior according to target vehicle and the range prediction vehicle of fore-aft vehicle, and execution step is 3)；

205) turning behavior of the vehicle at crossing is judged according to the historic state information of target vehicle.

5. a kind of car networking vehicle-state Forecasting Methodology according to claim 4, it is characterised in that the step 204) tool Body includes step：

241) judge whether target vehicle is more than safe distance with the distance between nearest vehicle in front vehicles, if it has, then Execution step 243), if it has not, then execution step is 242)；

242) judge whether the speed of target vehicle is more than the speed of nearest vehicle in front vehicles, if it has, then target vehicle Vehicle behavior predict the outcome to slow down, and execution step is 3), if it has not, then execution step is 243)；

243) judge whether target vehicle is more than safe distance with the distance between nearest vehicle in front vehicle, if it has, then The vehicle behavior of target vehicle predicts the outcome as at the uniform velocity, and execution step is 3), if it has not, then execution step is 244)；

244) judge whether the speed of target vehicle is more than the speed of nearest vehicle in front vehicle, if it has, then target vehicle Vehicle behavior predict the outcome as at the uniform velocity, and execution step is 3), if it has not, then the vehicle behavior of target vehicle predicts the outcome To accelerate, and execution step is 3).

6. a kind of car networking vehicle-state Forecasting Methodology according to claim 5, it is characterised in that the step 242) in Vehicle behavior predict the outcome for slow down when acceleration magnitude be：

$a_1=\mathrm{\σ}\frac{1}{d_{ij}(v_i-v_j)}$

Wherein：a₁Acceleration magnitude during to predict the outcome to slow down, d_ijFor in target vehicle and front vehicles recently vehicle it Between distance, σ is constant, v_iFor the present speed size of target vehicle, v_jPresent speed for nearest vehicle in front vehicles is big It is little.

7. a kind of car networking vehicle-state Forecasting Methodology according to claim 5, it is characterised in that the step 244) in Vehicle behavior predict the outcome for accelerate when acceleration magnitude be：

$a_2=-\mathrm{\α}\frac{1}{d_{ij}(v_i-v_j)}+\mathrm{\β}\frac{(v_t-v_i)}{d_{it}}$

Wherein：a₂Acceleration magnitude during to predict the outcome to accelerate, d_ijFor in target vehicle and front vehicles recently vehicle it Between distance, α, β be constant, v_iFor the present speed size of target vehicle, v_jFor the present speed of nearest vehicle in front vehicles Size, v_tFor the present speed size of nearest vehicle in front vehicle, d_itFor in target vehicle and front vehicle recently vehicle it Between distance.

8. a kind of car networking vehicle-state Forecasting Methodology according to claim 4, it is characterised in that the step 205) tool Body includes step：

251) the historic state information of target vehicle is traveled through forward from the current state of target vehicle, and determines start offset state TurnPoint, specially：

TurnPoint=min (g | Offset_k＞ T＆Offset_k-1＞ T＆ ..., Offset_g＞ T })

Wherein：

Offset_i=(OffsetP_i,OffsetV_i,OffsetA_i)

T=(TP, TV, TA)

Wherein：

OffsetP_i,OffsetV_i,OffsetA_iRespectively the i-th -1 state is to position, speed and acceleration in i-th state Side-play amount, TP, TV, TA are constant；

252) judge persistently to offset whether number of times exceedes set point number：And whether acceleration direction migration included angle meets circular motion Characteristic, specially judgment formula：

K-TurnPoint ＞ TK

w(OffsetA_k) ＜ TW＆w (OffsetA_k-1) ＜ TW ..., w (OffsetA_TurnPoint+1) ＜ TW

Wherein：Persistently to offset number of times, TK is set point number to k-TurnPoint, and TW is deviation angle tolerable error,

If the determination result is YES, then predict that target vehicle occurs to turn, and turn direction is target vehicle current state position Offset direction, if judged result is no, prediction target vehicle does not occur to turn.