CN104021664B - The dynamic path planning method of forming into columns and travelling worked in coordination with by automobile - Google Patents

Abstract

The present invention relates to a dynamic route planning method for vehicles traveling in a coordinated formation, comprising: step 10, judging whether the destination has been reached, and if not, performing step 20; step 20, judging whether the leading vehicle has arrived at the intersection, and if the leading vehicle has arrived Intersection, execute step 30; step 30, use the SPFA algorithm to determine the initial shortest path; step 40, obtain the information of all cars at the current intersection; step 50, use the Fuzzy formation algorithm to judge whether the current intersection is suitable for formation, if suitable for formation, perform step 60, Otherwise, execute step 70; step 60, control the vehicles to form a formation on the target road section; step 70, update the source road section and the relevant information of the road section; step 80, execute step 10. The invention improves the safety of road traffic, the traffic capacity of the road, saves energy, reduces the labor intensity of the driver, and improves the driver's comfort, so that the road capacity can be increased as much as possible, and the driving can be smoothed as much as possible.

Description

A Dynamic Path Planning Method for Coordinated Formation Driving of Vehicles

technical field

The invention relates to the field of path planning, in particular to a dynamic path planning method for vehicles traveling in coordinated formation.

Background technique

As far as the current situation is concerned, the practical way to solve the traffic problem is how to improve the existing road traffic capacity and efficiency. For example, use advanced science and technology to improve vehicle performance; use automation technology to eliminate the interference caused by human factors such as driver response delay and inaccurate judgment; adopt more effective traffic management methods to rationally dispatch vehicles, etc. These methods can not only improve The road traffic flow can be reduced, and the occurrence of adverse phenomena such as traffic jams and traffic accidents can be reduced.

In recent years, various countries have carried out follow-up research on various fields of intelligent transportation, and we found that:

The optimal path planning module is a key module in the vehicle navigation system, and the real-time and optimality of the optimal path planning algorithm are important indicators to measure the performance of the navigation system. The optimal path planning is based on the topological relationship in the electronic map, considering the real-time traffic information, and plans the optimal path for the navigation system to guide the user. In this way, the user's travel expenses can be reduced, the user's unfamiliarity with unfamiliar urban roads can be solved, and even the city's car travel routes can be controlled in a coordinated manner, the utilization efficiency of urban roads can be improved, and the number of congested roads can be reduced.

The complexity of traffic restriction information and the changing characteristics of traffic conditions over time make the optimal path obtained by the general static path-finding algorithm based on the ideal network graph model very likely to be far from the actual optimal path. In this way, the navigation system is required to have the dynamic pathfinding capability considering the traffic restriction information and the dynamic changing characteristics of traffic conditions. Therefore, detailed analysis and research on the dynamic pathfinding technology of car navigation system, including dynamic traffic road network modeling, design and implementation of dynamic pathfinding algorithm, and platform structure design of dynamic navigation system, will have important practical application value.

Among many vehicle dispatching methods, the formation control of vehicles is an effective vehicle scheduling method. Its purpose is to improve the safety of land transportation, improve the traffic capacity of the road, save energy, reduce the labor intensity of the driver, and improve its comfort.

Through the dynamic optimal path selection and the control of vehicle formation, the road vehicle density can be increased, and the road capacity can be increased. At the same time, traffic congestion can be effectively alleviated, and traffic smoothness and safety can be enhanced. In addition, driving in formation can reduce the air resistance of the cars, reduce the fuel consumption of the cars and save energy. Therefore, it is necessary to study the control method of vehicle formation based on optimal path selection.

To sum up, the research on vehicle formation control based on dynamic routing has great theoretical and practical significance in improving traffic safety and road capacity.

Contents of the invention

The purpose of the present invention is to provide a dynamic path planning method for cars to travel in a coordinated formation, so as to improve the safety of road traffic, the traffic capacity of the road, save energy, reduce the labor intensity of the driver, and improve its comfort, so that the road can be The capacity is increased as much as possible to make the driving as smooth as possible.

In order to solve the above-mentioned technical problems, as an aspect of the present invention, a dynamic route planning method for vehicle cooperative formation driving is provided, including: step 10, judging whether the destination has been reached, and if the destination is not reached, then perform step 20; step 20 , to determine whether the leading vehicle has reached the intersection, if the leading vehicle arrives at the intersection, go to step 30; step 30, use the SPFA algorithm to determine the initial shortest path; step 40, obtain the information of all cars at the current intersection; step 50, use the Fuzzy formation algorithm to determine the current intersection Whether it is suitable for formation, if it is suitable for formation, execute step 60, otherwise execute step 70; step 60, control the car to form a formation on the target road section; step 70, update the relevant information of the source road section and road section; step 80, execute step 10.

Further, step 20 also includes: if the leading vehicle has not reached the intersection, then execute step 30'; step 30', obtain the information of the preceding vehicle; step 40', judge whether the formation is suitable according to the Fuzzy formation algorithm, and if so, execute step 50' , otherwise execute step 60'; step 50', control the vehicles to form a formation on the current road section; step 60', execute step 10.

Further, after step 10 and before step 20, it also includes: step 11, judge whether it is the leading vehicle, if it is the leading vehicle, perform step 20, otherwise perform step 10.

Further, before step 10, it also includes: step 1, loading map information; step 2, initializing vehicle information; and step 3, starting the vehicle.

Further, in step 10, before judging whether the vehicle has reached the destination, it also includes the step of processing changes in the vehicle's position and speed.

The invention carries out the research on the vehicle driving strategy under the condition of dynamic path planning combined with fuzzy control vehicle formation. Using the present invention, when the vehicle is driving, it advances according to the optimal path selected by the SPFA algorithm, and in the driving process, adopts the fuzzy control method and combines road vehicle information to drive in formation, so that the overall driving time of all vehicles on the road can be reduced. This increases car flow and increases road capacity.

At the same time, during the driving process of each car, consider the relationship with the position and path of the surrounding cars to judge whether to form a formation. In this way, the performance of the smart car is improved as a whole, and the time for all cars to reach their destinations is reduced as a whole, thereby alleviating the pressure on the road. More importantly, it has great value in the application, and can enable car novices to quickly reach their destinations. For example, in the case of a veteran with ten years of car driving experience, the novice can be safer. It saves more time and effectively enhances the safety of driving.

Description of drawings

Fig. 1 schematically shows the algorithm flow chart of the present invention.

detailed description

Embodiments of the invention are described in detail below, but the invention can be practiced in many different ways as defined and covered by the claims.

The number of automobiles in our country is increasing rapidly, and the road load is getting bigger and bigger, especially the traffic load of the urban road system. Aiming at this present situation, the present invention conducts in-depth research on the dynamic path planning and the fuzzy control algorithm for realizing vehicle formation, and obtains an efficient vehicle driving strategy by combining the two. For this reason, the present invention proposes a dynamic path planning method for coordinated formation driving of vehicles.

First, starting from the traditional perspective of road selection, the present invention realizes real-time change of road information by incorporating factors such as road length, width, and degree of congestion into changes in weights, and then achieves the purpose of dynamic path planning. After comprehensively considering the adaptability of the algorithm and map data storage structure, the time complexity of the algorithm, the space complexity of the algorithm, and the efficient response to traffic emergencies, the comprehensive superiority of the SPFA algorithm in dynamic route planning is determined.

Secondly, from the perspective of the intelligence of the automobile itself, the present invention deeply studies the Fuzzy algorithm for realizing the formation control of the automobile. In the intelligent "vehicle-road-vehicle-vehicle" collaborative environment, using the Fuzzy algorithm to realize vehicle formation driving can increase road vehicle density, increase road capacity, and at the same time simplify traffic control complexity, increase traffic controllability, and improve traffic safety. The combination of driving behaviors can greatly improve the flexibility and flexibility of the car platoon, and can fully exert the advantages of car platooning. Simultaneously, the present invention has also considered some specific situations (such as the situation that the number of cars is too short on a road section), and formation will be unfavorable for the driving of cars in these specific cases. Through consideration and analysis in all aspects, the present invention elaborates and demonstrates the vehicle selective formation model based on Fuzzy fuzzy control algorithm that can adapt to various situations. The invention has very good innovation and practicability.

For the solution method of the Shortest Path Problem (SPP) based on the abstract network graph, due to its various application requirements in communication, transportation, computer network, operation research, management and other disciplines, it has been fully developed for many years. Attention has been paid to and a large number of research results have been obtained. Among these many studies, most of them are based on static algorithms with constant network graph path weights, and dynamic shortest path search algorithms whose network path weights change with time. to receive more widespread attention.

Prior to this, there were some articles devoted to the study of navigation problems. Some of them have been used to study the dynamic shortest path. For example, Khasawneh, MA used the radiation avoidance algorithm to explore the search for the shortest path. These algorithms are indeed able to choose the appropriate path to reach the destination in an ideal time. But it simply studies how to choose the shortest path in a graph. For example, Tarantilis, CD uses a parameter meta-heuristic algorithm, but the heuristic algorithm does not guarantee the global optimization. Suppose there are 100 smart cars in a certain area, and the parameter meta-heuristic algorithm can only guarantee a certain number of 100 cars. The optimal number of vehicles does not guarantee that the sum of applying 100 vehicles to the destination is optimal. Thereby, the problem of how to run the shortest car is only solved in one aspect, that is, how a certain car goes the shortest, and does not relate to the problem of how the overall car runs the shortest. In order to solve this problem, we used the idea of formation again, that is, to form a team of vehicles that intend to go in the same direction. Therefore, the present invention considers the shortest path and the formation problem together, and can greatly reduce the sum of the time for all cars to reach the destination globally, thereby saving time. The present invention combines the SPFA algorithm with the idea of formation to deal with how to plan all the cars in a certain area so that the sum of the distances traveled by all the cars to the destination is the shortest The problem.

First of all, the present invention studies the map data storage structure based on the current navigation system, the dynamic path planning algorithm, and uses the structure of the adjacency list to store digital map information, including nodes, road sections and turning restrictions. In terms of dynamic path planning, a dynamic path planning model was established, and the dynamic path planning algorithms were compared. In the dynamic path planning experiment of a single car, the avoidance and detour of various algorithms for congested road sections and emergency road sections were verified, as well as for Dynamic real-time processing of dynamic traffic information. The superiority of the SPFA algorithm is proved by comparing the length of the user's travel and the time required to call the algorithm by various algorithms on the dynamic road.

For the formation module, the realization of vehicle formation driving in the intelligent "vehicle-road-vehicle-vehicle" collaborative environment can increase the density of road vehicles, increase road capacity, simplify traffic control complexity, increase traffic controllability, and improve traffic safety. Combining with driving behavior can greatly improve the flexibility and flexibility of car formation, and can make full use of the advantages of car formation driving, but in some specific cases (such as too short a road section and too many cars) Formation will be unfavorable to the driving of automobiles instead. The present invention is a fuzzy selective formation path planning method for automobiles under the intelligent "vehicle-road-vehicle-vehicle" collaborative environment, and its specific significance.

Please refer to Fig. 1, the present invention provides a kind of dynamic path planning method of vehicle cooperative formation driving, including:

Step 10, judging whether to reach the destination, if not, then perform step 20;

Step 20, judging whether the leading vehicle has arrived at the intersection, if the leading vehicle has arrived at the intersection, perform step 30;

Step 30, using the SPFA (ShortestPathFasterAlgorithm) algorithm to determine the initial shortest path; in the dynamic optimal path selection process, according to the topological relationship in the electronic map, considering real-time traffic information, plan the optimal path for the navigation system to guide the user. We adopted the SPFA algorithm, an improved algorithm of the Bellman-Ford algorithm, to select the shortest path on the dynamic traffic road, and then get the next node where the car is driving, and call this algorithm again on the next node until reaching the destination until. According to the optimization principle, in the total travel route of a smart car from the starting point to the end point, a sub-route of the optimal driving route, that is, from one intersection to the next intersection, is also optimal for the initial state and end state of the car.

Step 40, obtaining the information of all cars at the current intersection;

Step 50, use the Fuzzy formation algorithm to judge whether the current intersection is suitable for formation, if it is suitable for formation, execute step 60, otherwise execute step 70. During the driving process, consider road vehicle information, whether to carry out formation driving. In the part of car formation, we mainly use the fuzzy control algorithm to analyze the number of cars to be formed and the length of the road to control whether the cars are in formation. Firstly, we fuzzy the input of the fuzzy system (the number of vehicles to be formed and the length of the road) and the output (the formation suggestion factor); secondly, we determine the membership function of the fuzzy set and establish the fuzzy rules according to the conditions of the road; then we Carry out fuzzy reasoning on the obtained fuzzy relations; finally get our fuzzy control look-up table by defuzzification.

Step 60, controlling the cars to form a formation on the target road section;

Step 70, updating the source road segment and the relevant information of the road segment;

Step 80, execute step 10.

The invention carries out the research on the vehicle driving strategy under the condition of dynamic path planning combined with fuzzy control vehicle formation. The optimal path selected by the SPFA algorithm is used to advance when the vehicle is running, and the fuzzy control method is adopted in the driving process, combined with road vehicle information to carry out formation driving, so that the overall driving time of all vehicles on the road can be reduced. This increases car flow and increases road capacity.

At the same time, during the driving process of each car, consider the relationship with the position and path of the surrounding cars to judge whether to form a formation. In this way, the performance of the smart car is improved as a whole, and the time for all cars to reach their destinations is reduced as a whole, thereby alleviating the pressure on the road. More importantly, it has great value in the application, and can enable car novices to quickly reach their destinations. For example, in the case of a veteran with ten years of car driving experience, the novice can be safer. It saves more time and effectively enhances the safety of driving.

Preferably, said step 20 also includes: if the leading vehicle has not reached the intersection, then perform step 30';

Step 30', obtaining the information of the preceding vehicle;

Step 40', judge whether it is suitable for formation according to the Fuzzy formation algorithm, if suitable, execute step 50', otherwise execute step 60';

Step 50 ', control automobile to form formation in current section;

Step 60', execute step 10.

Preferably, after the step 10 and before the step 20, it also includes:

Step 11, judging whether it is the leading vehicle, if it is the leading vehicle, execute step 20, otherwise execute step 10.

Preferably, before said step 10, it also includes:

Step 1, load map information;

Step 2, initialize car information;

Step 3, start the car.

Preferably, in step 10, before judging whether the vehicle has reached the destination, it also includes the step of processing changes in the vehicle's position and speed.

Now, the above method in the present invention will be exemplified.

First, load the map information, such as the road network structure, the length information of each road section, the turning information of each intersection, the location information of the starting point and the location information of the destination, etc.

Then, initialize information such as the starting point, destination and speed of the car.

After that, the car starts and enters the driving state under the guidance of the "car driving processing module".

While the car is running, check whether it has reached the destination, if it arrives, the program ends; if it does not arrive at the destination, then check whether the car is the head car of the convoy, if not, then do not do any additional processing.

If the car is the lead car, check whether it has reached the intersection:

(1) If the intersection is not reached, the speed, destination and other information of the vehicle in front will be obtained, and the road section Fuzzy formation algorithm will be called to determine whether the formation condition is reached. If the formation condition is met, the vehicle formation of the road section will be formed. Then return to the "car driving processing module" and recycle.

(2) If the intersection is reached, call the SPFA algorithm, take the current intersection and the target intersection as input, and take the new next intersection as output. Cars that arrive at the intersection are processed in this way, so as to update the optimal route information of the car. Afterwards, the information of all the cars at the current intersection is used as input, and the "fuzzy formation algorithm at the intersection judges" whether it is suitable for formation, so as to obtain the vehicle formation results of each target road section.

Then, update the car number values of the source road segment and the target road segment to reflect the congestion degree of the road. Then return to the "car driving processing module" and recycle.

Next, the SPFA (ShortestPathFasterAlgorithm) algorithm is exemplarily described.

First, explain the selection process of the shortest path in the dynamic graph. When selecting the shortest path, since the road changes during the operation of the smart car, the current road segment information must be updated continuously while the car is running on the road. At the same time, it must be continuously selected from the updated map. shortest path. Therefore, when a car starts from the starting point, it first uses the shortest path algorithm to select the shortest path, and then when it arrives at the next intersection, the traffic capacity of the road has changed due to the actual situation of the road and congestion. Therefore, the car needs to find the shortest path from this point to the destination again. This path is not necessarily the same as the above-mentioned path, and the above-mentioned process is repeated until the destination is reached.

In the process of the shortest path, the traffic road network is represented by G=(V,E), the road resistance coefficient w: E->D reflects the actual situation of the edge, and the road P=(V ₀ , V ₁ ,. . . ., Vi) reflects the sum of the weights of the edges passing through from the starting point to the destination point.

$\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; the\; s</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>$

Therefore, define the shortest path from point u to point v as:

$\mathrm{<span\; class="notranslate">\; \&chi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{c}\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; :</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; v</span>}\\ <span\; class="notranslate">\; \&infin;</span>\end{array}\right.$

From the above analysis, we can know that to find the shortest distance from the starting point to the end point, it is to sum the weights of the paths passed. However, because the road resistance coefficient is constantly changing, it is necessary to study the dynamic shortest path and formation.

For a weighted adjacency matrix, transform this matrix into the corresponding topology. Use this matrix to analyze the routing process of the entire car from the starting point to the ending point.

The used matrix is expressed as a topological graph, which is a simplified process of the actual traffic situation, and can explain the situation of vehicles running on the road.

The execution steps of the SPFA algorithm: first establish a queue D[n], where D[i] represents the distance from the starting point S to the i-th point. For D[i] at the beginning, D[i]=0, if i is the starting point S; D[i]=∞, if i≠S. pre[v] indicates a point before point v in the current shortest path from S to v. C(u, v) represents the cost from u to v, that is, the length of the path.

Considering that it is a multi-step decision-making process for a car to go from the starting point s to the end point t, when the car arrives at each traffic intersection, it must choose the most economical way to reach the next intersection according to the current road congestion. Similarly, the same operation will be performed after reaching the next intersection until the end point t is reached. In the actual running process, the road s->a->b->...->t selected based on the historical data may change at the beginning, so the actual road selection is needed to make up for this error. It can be seen that in the initial process, road 1 was selected. However, in the actual operation process, road 1 was found to be congested, resulting in a larger road resistance coefficient. At another intersection, road 2 was selected. , reached the destination.

The SPFA algorithm reduces the time spent by vehicles in the route selection process, thus reducing time overhead and increasing efficiency.

Below, the fuzzy control algorithm used in the present invention is illustrated with an example.

(1) Input and output of fuzzy control system

The length of the road and the number of cars related to the car formation are used as the input of the fuzzy control system, where the length of the road refers to the physical length of the next same road section, and the number of cars number refers to the cars waiting to enter the road section corresponding to the length at the intersection quantity.

Take the formation suggestion factor flag as the output.

(2) Fuzzification of input and output variables

Fuzzification of path length:

Let the fuzzy subsets of path length length be L ₁ , L ₂ , L ₃ , L ₄ , L ₅ , and their linguistic value sets and linguistic variables are:

l = {very short, short, medium, long, very long}

And the size of the length of the path is quantified into nine levels, which are respectively expressed as -3, -2, -1, 0, +1, +2, +3, and the domain of discussion is:

L={-3, -2, -1, 0, +1, +2, +3,}

Fuzzification of the car number number:

Assume that the fuzzy subsets of the number of cars are N ₁ , N ₂ , N ₃ , N ₄ , N ₅ , and their linguistic value sets and linguistic variables are:

n={few, few, general, many, many}

And the size of the car number is quantified into nine levels, which are respectively expressed as -3, -2, -1, 0, +1, +2, +3, and the domain of discussion is:

N = {-3, -2, -1, 0, +1, +2, +3}

Fuzzification of the formation suggestion factor flag (accurately, it is not fuzzy)

The fuzzy subsets of the formation suggestion factor flag are F ₁ , F ₂ , F ₃ , and F ₄ , and their linguistic value sets and linguistic variables are:

f＝{It is not recommended to form a formation, it is not recommended to form a formation, it is recommended to form a formation, it is very recommended to form a formation}

And quantify the size of the fuzziness of the formation result flag into two levels, which are respectively expressed as -1 and +1, and the domain of discussion is:

F={-1,+1}

flag is a variable showing whether formation is possible, the value is true or false, if true, formation is possible, if false, formation is not possible.

(3) Membership function of fuzzy set

The membership degree function of fuzzy linguistic variables determines the membership degree of elements in the universe to fuzzy linguistic variables, and its determination is based on existing successful applications and expert suggestions. According to the membership function, the vector form of the membership function can be calculated.

(4) Fuzzy control rules

Here L represents the length of the road, N represents the number of cars, and F represents the result of formation or not. Formulate fuzzy control rule table.

(5) Fuzzy relationship

The fuzzy relationship is calculated using the Mandani minimum algorithm.

(6) Fuzzy reasoning

After obtaining the fuzzy relationship of the designed fuzzy controller, the fuzzy value vector of the output control quantity can be solved by the synthesis reasoning method.

(7) Defuzzification

In the present invention, the weighted average method is adopted for defuzzification.

After the defuzzification process, the precise output is obtained, and then the scale is transformed according to the needs, and finally the calculation result is expressed in a matrix, which is the final fuzzy control look-up table (take two decimal places).

The reason for using the SPFA algorithm is not only because the route selection effect on the local map is good, but more importantly, the time complexity of SPFA is lower, which is more suitable for the real-time performance of the actual car route finding. Adding formation under fuzzy control on the basis of SPFA algorithm makes the overall driving efficiency higher, because the road conditions become better after the formation, so that the vehicle speed has an overall increase, and the road surface is also improved. Crowded situation, in the present invention, the formation in the road section is also added in the middle of the formation, it is suggested that the road section formation should be reduced as far as possible (because the road section formation will make the speed of the vehicle have a process of reducing and increasing), so that the formation can be further improved Effect.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (3)

1. the dynamic path planning method of forming into columns and travelling worked in coordination with by automobile, it is characterized in that, comprising:

Step 10, judges whether to arrive destination, if the object of arrival, then performs step 20;

Step 20, judges whether head car arrives crossing, if head car arrives crossing, performs step 30;

Step 30, adopts SPFA algorithm to determine initial shortest path;

Step 40, obtains the information of all automobiles in current crossing;

With Fuzzy formation algorithm, step 50, judges whether described current crossing is applicable to forming into columns, if be applicable to forming into columns, performs step 60, otherwise performs step 70;

Step 60, controls automobile and forms formation in target road section;

Step 70, upgrades the relevant information in section, source and section;

Step 80, performs step 10;

Described step 20 also comprises: if head car does not arrive crossing, then perform step 30 ';

Step 30 ', obtain front truck information;

Step 40 ', judge whether to be applicable to forming into columns according to Fuzzy formation algorithm, if be applicable to, perform step 50 ', otherwise perform step 60 ';

Step 50 ', control automobile and form formation at current road segment;

Step 60 ', perform step 10;

Also to comprise before described step 20 after described step 10:

Step 11, determines whether a car, if head car, then performs step 20, otherwise performs step 10.

2. according to the dynamic path planning method that claim 1 is stated, it is characterized in that, also comprised before described step 10:

Step 1, loads cartographic information;

Step 2, initialization automobile information;

Step 3, starts automobile.

3. according to the dynamic path planning method that claim 1 is stated, it is characterized in that, in described step 10, before judging whether to arrive destination, also comprise the step that automobile position and velocity variations are processed.