CN108733063A - A kind of autonomous collaboration traveling decision-making technique of automatic driving vehicle - Google Patents

Abstract

The invention proposes an autonomous cooperative driving decision-making method for an automatic driving vehicle, specifically: the vehicle collects driving state information through an on-board detection system; the vehicle applies fuzzy functions to realize fuzzy processing of driving state information parameters, and the driving state of the vehicle and its neighbor vehicles Estimate; the fuzzified driving state information parameters are used as the input parameters of the fuzzy inference engine, and the fuzzy inference engine infers the vehicle driving mode according to the set fuzzy rule base; the optimal driving mode decision result of the independent vehicle is output by the defuzzification process , to realize autonomous cooperative driving among cluster vehicles. Beneficial effects: through the information detection, calculation and communication capabilities of the self-driving vehicle, through the fuzzification and fuzzy logic reasoning of the vehicle's own driving state parameter set, the autonomous cooperative driving among cluster vehicles can be realized, and the best driving mode can be selected for independent vehicles Provide a reference to improve vehicle driving safety, reduce vehicle travel time, and reduce traffic energy consumption.

Description

An autonomous and cooperative driving decision-making method for autonomous vehicles

technical field

The invention relates to the Internet of Vehicles technology, and in particular to the autonomous cooperative driving technology among self-driving vehicles.

Background technique

The Internet of Vehicles (IoV) is the integration of wireless communication technology and transportation network. It is a typical application of the Internet of Things in the field of transportation, and it is the demand for the evolution to autonomous driving. Autonomous driving technology is a traffic intelligence and networking application with high mobility, complex road condition information, and high safety requirements. It is the development trend of road traffic and a necessary process for realizing unmanned driving. The autonomous cooperative driving of the vehicle is essentially a dynamic adjustment process of the driving trajectory according to the road traffic state, the vehicle's own driving state and the neighboring vehicles' driving state. In the Internet of Vehicles, vehicles generally travel in a cluster mode. Among them, the vehicles in the slow lane generally travel in the predetermined trajectory mode, the vehicles in the driving lane generally travel in the following mode, and the vehicles in the overtaking lane mainly travel in random modes such as speeding and lane changing. , to return to the driving lane after overtaking and drive in follow mode. As a node of the Internet of Vehicles, the self-driving vehicle has a complete on-board sensor detection, calculation processing and communication dissemination system, which can acquire, transmit and process vehicle driving status information in real time, realize autonomous and coordinated driving among clustered vehicles, and improve vehicle driving safety and road traffic efficiency.

At present, self-driving vehicles independently rely on their own additional sensor detection and computing processing systems to collect and process traffic environment status information in real time to ensure vehicle safety. The overall traffic efficiency of the transportation network. Therefore, using the powerful on-board detection system, information processing capabilities and communication capabilities of autonomous driving vehicles, the optimal mode of vehicle driving status can be calculated in real time to realize autonomous and cooperative driving among vehicles in the cluster, and promote large-scale and reliable commercial use of autonomous vehicles. Future autonomous driving provides operational data.

Contents of the invention

The purpose of the present invention is to overcome the deficiencies of the prior art, and provide an autonomous cooperative driving decision-making method among autonomous vehicles, which utilizes the real-time traffic state information acquisition capability of the autonomous vehicles, powerful information calculation and processing capabilities, and high reliability and low delay The communication ability of the vehicle is calculated by using the fuzzy function to calculate the membership degree of the vehicle's driving state information parameters, and the optimal driving mode of the vehicle is deduced by fuzzy logic, so as to realize the autonomous cooperative driving among the clustered vehicles in the Internet of Vehicles and the optimization of road traffic. The details are as follows: Realization of technical solutions:

The autonomous and cooperative driving decision-making method of the self-driving vehicle is equipped with a positioning system, a driving state information detection system, and a vehicle-mounted wireless communication system supporting vehicle-to-vehicle communication on the vehicle. Neighboring vehicles periodically exchange each other's driving state information through vehicle-to-vehicle communication technology, specifically content include:

The vehicle collects driving status information through the on-board detection system;

The vehicle applies the fuzzy function to realize the fuzzy processing of the driving state information parameters, and estimates the driving state of the vehicle and its neighbors;

The fuzzified driving state information parameters are used as the input parameters of the fuzzy inference machine, and the fuzzy inference machine infers the vehicle driving mode according to the set fuzzy rule base;

The optimal driving mode decision results of independent vehicles are output by defuzzification processing, and the autonomous cooperative driving among cluster vehicles is realized.

The further design of the autonomous cooperative driving decision-making method of the self-driving vehicle is to set the vehicle driving state information parameter set as θ={θ _V , θ _N , θ _L }, where θ _V is the vehicle driving speed, and θ _N is The number of neighbor nodes of the vehicle, θ _L is the current driving lane of the vehicle; according to θ _V , θ _N and θ _L , the corresponding fuzzy membership degrees μ(V), μ(N) and μ(L) are calculated respectively.

The further design of the autonomous cooperative driving decision-making method of the self-driving vehicle is to set the vehicle speed parameter θ _V ={V _L , V _M , V _H }, V _L , V _M , V _H correspondingly represent low speed, There are three driving speed states of medium speed and high speed. Set the upper limit of vehicle speed as V _U , the lower limit of vehicle speed as _{V LB} , and the average speed as V _A , then the fuzzy membership degree μ( V)={μ _H , μ _M , μ _L } is calculated according to formula (1), formula (2) and formula (3);

in,

The further design of the autonomous cooperative driving decision-making method of the self-driving vehicle is to set the vehicle neighbor node number parameter θ _N = { _NS , N _M , N _D }, where _NS , N _M , N _D represent sparse, normal As well as the dense distribution of three kinds of neighbor nodes, set the average number of neighbor nodes as N _N during normal traffic, the upper limit of the average number of neighbor nodes N _U = 2NN _N , the lower limit of the average number of neighbor nodes Then the fuzzy membership degree μ(N)={μ _S , μ _N , μ _D } of the distribution status parameter θ _N of sparse, normal and dense vehicle neighbor nodes according to formula (4), formula (5) and formula (6) calculate:

The further design of the autonomous cooperative driving decision-making method of the self-driving vehicle is to set the standardized single-lane width to 3.75 meters, and the distribution state of the three-lane roads from left to right is the passing lane, the driving lane and the slow lane, and the vehicle driving lane parameters θ _L ＝{L _O , L _C , L _S }, L _O , L _C , L _S respectively represent the three occupied lane states of overtaking lane, driving lane and slow lane, then the three types of overtaking lane, driving lane and slow lane The fuzzy membership degree μ(L)={μ _O , μ _C , μ _S } of the vehicle lane state parameter θ _L is calculated according to formula (7), formula (8) and formula (9):

The further design of the autonomous cooperative driving decision-making method of the self-driving vehicle is to set the vehicle driving state set as M={M _R , M _F , M _P }, and M _R , M _F and M _P respectively represent random patterns A driving state, a following mode driving state, and a predetermined track mode driving state.

The further design of the autonomous cooperative driving decision-making method of the self-driving vehicle is that the set fuzzy rule base is shown in Table 1,

Table 1

The advantages of the present invention are as follows:

The autonomous cooperative driving decision-making method among self-driving vehicles of the present invention realizes autonomous cooperation among clustered vehicles through the information detection, calculation and communication capabilities possessed by the self-driving vehicles, and through the fuzzification and fuzzy logic reasoning of the vehicle's own driving state parameter set Driving, provide reference for independent vehicles to choose the best driving mode, improve vehicle driving safety, reduce vehicle travel time, and reduce traffic energy consumption.

Description of drawings

Figure 1 is a cluster vehicle driving model.

Figure 2 is the vehicle speed membership function.

Figure 3 is the distribution membership function of the vehicle's neighbor nodes.

Figure 4 is the vehicle lane membership function.

Fig. 5 is the fuzzy inference system of vehicle driving mode.

Detailed ways

The technical scheme of the present invention is further described in conjunction with specific embodiments and accompanying drawings.

As shown in Figure 1, the autonomous and cooperative driving decision-making method for self-driving vehicles provided in this embodiment is equipped with a positioning system, a driving state information detection system, and a vehicle-mounted wireless communication system that supports vehicle-to-vehicle communication. Exchanging driving status information with each other, the specific content includes: the vehicle collects driving status information through the on-board detection system; when the vehicle intends to change its driving mode, it releases its own driving status information through active broadcasting;

The vehicle applies the fuzzy function to realize the fuzzy processing of the driving state information parameters, and estimates the driving state of the vehicle and its neighbors;

The fuzzified driving state information parameters are used as the input parameters of the fuzzy inference machine, and the fuzzy inference machine infers the vehicle driving mode according to the set fuzzy rule base;

The optimal driving mode decision results of independent vehicles are output by defuzzification processing, and the autonomous cooperative driving among cluster vehicles is realized.

In this embodiment, the vehicle driving state information parameter set is set as θ={θ _V , θ _N , θ _L }, where θ _V is the vehicle driving speed, θ _N is the number of neighbor nodes of the vehicle, and θ _L is the current driving lane of the vehicle ; Calculate the corresponding fuzzy membership degrees μ(V), μ(N) and μ(L) according to θ _V , θ _N and θ _L respectively.

Further, set the vehicle speed parameter θ _V ={V _L , V _M , V _H }, V _L , V _M , V _H correspond to the three speed states of low speed, medium speed and high speed respectively, set the vehicle The upper limit of the driving speed is V _U , the lower limit of the vehicle speed is V _LB , and the average driving speed is V _A , then the fuzzy membership degree of the vehicle speed parameter θ _V in the three speed states of high speed, medium speed and low speed μ(V)= {μ _H , μ _M , μ _L } are calculated according to formula (1), formula (2) and formula (3);

in,

In this embodiment, the normalized road driving safety index is set to be β=0.001H (H is hour), and under the premise of ensuring vehicle driving safety, the average number of neighbor nodes when the three lanes pass normally (R is the radial length of the communication area of the vehicle communication system), the upper limit of the average number of neighbor nodes The lower limit of the average number of neighbor nodes Then the fuzzy membership degree μ(N)={μ _S , μ _N , μ _D } of the distribution status parameter θ _N of the sparse, normal and dense three vehicle neighbor nodes is calculated as shown in formulas (4) to (6):

In this embodiment, the standardized single-lane width is set to be 3.75 meters, and the distribution state of the three-lane roads from left to right is the overtaking lane, the driving lane and the slow lane, and the vehicle driving lane parameters θ _L = {L _O , L _C , L _S } , L _O , L _C , _{L S} respectively represent the three occupied lane states of overtaking lane, driving lane and slow lane, then the fuzzy membership degree μ( L)={μ _O , μ _C , μ _S } Calculated according to formula (7), formula (8) and formula (9):

In this embodiment, the set of vehicle driving states is set as M={M _R , M _F , M _P }, and M _R , _MF and M _P respectively represent the driving states of the random mode, the following mode and the predetermined trajectory mode. .

The fuzzy rule base of setting mentioned in the present embodiment is as table 1,

Table 1

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present invention can easily think of changes or Replacement should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (7)

1. An autonomous cooperative driving decision method for automatically driven vehicles features that a locating system, a driving state information detecting system and a vehicle-carried radio communication system supporting communication between adjacent vehicles are used to periodically exchange the driving state information between adjacent vehicles by communication technique between vehicles, and features that the autonomous cooperative driving decision method for automatically driven vehicles is used to automatically detect the driving state information of adjacent vehicles

The vehicle acquires driving state information through a vehicle-mounted detection system;

the vehicle uses a fuzzy function to realize fuzzification processing of the driving state information parameters, and estimates the driving states of the vehicle and the neighboring vehicles;

the running state information parameters after fuzzification are used as input parameters of a fuzzy inference machine, and the fuzzy inference machine infers a vehicle running mode according to a set fuzzy rule base;

and outputting an optimal running mode decision result of the independent vehicles through the fuzzy solution processing, and realizing the autonomous cooperative running among the clustered vehicles.

2. The autonomous cooperative travel decision method of an autonomous vehicle according to claim 1, characterized in that the vehicle travel state information parameter set is set to θ ═ { θ ═ θ_V,θ_N,θ_LIn which θ_VAs the vehicle running speed, theta_NNumber of neighbor nodes of vehicle, theta_LIs the current driving lane of the vehicle; according to theta_V、θ_NAnd theta_LCorresponding fuzzy membership degrees mu (V), mu (N) and mu (L) are calculated respectively.

3. The autonomous cooperative driving decision method of an autonomous-driven vehicle according to claim 2, characterized in that a vehicle driving speed parameter θ is set_V＝{V_L,V_M,V_H}，V_L,V_M,V_HRespectively correspondingly showing three running speed states of low speed, medium speed and high speed,

setting an upper limit of a vehicle running speed to V_UThe lower limit of the running speed of the vehicle is V_LBAverage running speed of V_AThen the vehicle running speed parameter theta_VThe fuzzy membership degree mu (V) of three running speed states of high speed, medium speed and low speed is { mu (V) }_H,μ_M,μ_LCalculating according to formula (1), formula (2) and formula (3);

wherein,

4. the autonomous cooperative driving decision method of an autonomous-driven vehicle according to claim 3, characterized in that a vehicle neighbor node number parameter θ is set_N＝{N_S,N_M,N_D}，N_S,N_M,N_DRespectively representing the distribution states of three neighbor nodes of sparse, normal and dense, and setting the average number of neighbor nodes to be N during normal traffic_NUpper limit of average neighbor node number N_U＝2N_NLower limit of average neighbor node numberThen the three kinds of vehicle neighbor nodes of sparse, normal and dense are distributed with the state parameter theta_NIs fuzzy degree of membership mu (N) { mu }_S,μ_N,μ_DIs calculated according to equation (4), equation (5), and equation (6):

where N represents the number of neighbor nodes.

5. The autonomous cooperative driving decision method of an autonomous vehicle according to claim 4, characterized in that a standardized lane width of 3 is set.75 m, the distribution states of the three-lane road are a passing lane, a traffic lane and a slow lane from left to right in sequence, and the vehicle driving lane parameter theta_L＝{L_O,L_C,L_S}，L_O,L_C,L_SRespectively and correspondingly representing three occupied lane states of a passing lane, a traffic lane and a slow lane, and then three vehicle driving lane state parameters theta of the passing lane, the traffic lane and the slow lane_LIs fuzzy degree of membership mu (L) { mu }_O,μ_C,μ_SIs calculated according to equation (7), equation (8), and equation (9):

where L represents the currently occupied lane.

6. The autonomous cooperative driving decision method of an autonomous vehicle according to claim 5, characterized in that the vehicle driving state set is set to M ═ { M ═ M_R,M_F,M_P}，M_R、M_FAnd M_PThe random mode running state, the follow-up mode running state, and the predetermined track mode running state are respectively expressed correspondingly.

7. The autonomous cooperative driving decision method of an autonomous vehicle according to claim 6, characterized in that the fuzzy rule base is set as table 1.

TABLE 1