CN105046356B - A kind of electric car course continuation mileage optimization device and method - Google Patents

Abstract

The invention discloses a kind of electric car course continuation mileages to optimize device and method, and device includes data collection system, decision system and expert system；Data collection system includes mobile client, in-vehicle communication system and charging station communication module；Decision system includes the first wireless communication module and decision-making module；Expert system includes the second wireless communication module, database and path planning module；Expert system receives searching database after driving task attribute information, electric car vehicle condition information, charging pile work information, one group of existing programme path is returned if having matched data, otherwise path planning module is transferred to be handled, the wherein constraint conditions such as path planning module combination electric car vehicle condition, charging pile operating condition, traffic, driving task attribute, establish optimization problem model, the optimization problem is solved, a plurality of battery friendly routing plan is provided." mileage anxiety " problem can be effectively relieved in the present invention, greatly improve working efficiency, have deep realistic meaning.

Description

A kind of electric car course continuation mileage optimization device and method

Technical field

The invention belongs to automobile technical fields, are related to a kind of electric car course continuation mileage optimization device and method, especially It is related to a kind of electric car course continuation mileage optimization device and method of battery friendly.

Background technique

In recent years, fossil energy is exhausted, and environmental pollution is serious, research and development new-energy automobile become Resources for construction saving, The only way of friendly environment society.In this background, electric car comes into being.However electric car is sent out after many years Exhibition also fails to enter huge numbers of families, this is because its intrinsic limitation fails to break through always: energy storage is few, mileage travelled is short, fills Electric slow, cruising ability could be ensured by needing to charge by common charging device relay.Currently, in the continuation of the journey of countries in the world electric car Cheng Youhua correlative study is concentrated mainly on the fields such as influence of the automobile power cell modeling with electric car charging to electric network performance, However these researchs improve electric car course continuation mileage all without fundamentally solving how electric car is quickly found out charging pile The problems such as.

Under these contradictions urgently to be resolved, intelligent, information-based, networking electric car intelligent charge net is established Network optimizes electric car course continuation mileage by using the routing algorithm of battery sensitivity, plans most suitable driving for electric car Path, guarantee stroke in can relay charging be unlikely to cast anchor；Using driving task scheduling algorithm, intelligent Task sequence is mentioned It for the solution of optimization, realizes that the multi-goal path with task time real-time constraint is planned, increases substantially work effect Rate could solve charging problems in high quality, alleviate " the mileage anxiety " of automobile user significantly, push ev industry Development.

Summary of the invention

The present invention uses the routing algorithm of battery sensitivity, optimizes electric car course continuation mileage, most closes for electric car planning Suitable planning driving path, guarantee stroke in can relay charging be unlikely to cast anchor, alleviate " mileage anxiety " problem.

Technical solution used by the device of the invention is: a kind of electric car course continuation mileage optimization device, feature exist In: including data collection system, decision system and expert system；The data collection system includes mobile client, vehicle-mounted Communication system and charging station communication module；Decision system includes the first wireless communication module and decision-making module；Expert system includes Second wireless communication module, database and path planning module；Data collection system, decision system and the expert system it Between be connected with each other by the communication module, the first wireless communication module and the second wireless communication module that are arranged in mobile client it is logical Letter.

Technical solution used by method of the invention is: a kind of electric car course continuation mileage optimization method, feature exist In, comprising the following steps:

Step 1: user is acquired using mobile client and uploads driving task attribute information into expert system；

Step 2: in-vehicle communication system acquires and uploads electric car vehicle condition information into expert system；

Step 3: charging station communication module acquires and uploads charging pile work information into expert system；

Step 4: expert system receives searching database after the data uploaded in step 1, step 2 and step 3, if having One group of existing programme path is then returned with data, otherwise path planning module is transferred to be handled；

Step 5: path planning module combination constraint condition establishes optimization problem model, solves the optimization problem, carries out Multi-goal path planning, provides the routing plan of a plurality of feasible battery friendly；The constraint condition includes electric car Vehicle condition, charging pile operating condition, traffic, driving task attribute；

Step 6, after decision system receives the feasible one group of programme path of the condition of satisfaction, preceding k is taken according to probability feasibility Optimal route returns to mobile client, and user is transferred to select.

Preferably, driving task attribute information described in step 1 includes the geographical location of each task, deadline And precedence information.

Preferably, electric car vehicle condition information described in step 2 includes current vehicle position and remaining capacity information.

Preferably, charging pile work information described in step 3 includes the service condition and fault condition letter of charging pile Breath.

Preferably, one group of existing programme path, matching rule are returned described in step 4 if having matched data Are as follows: electric car vehicle condition information is consistent with driving task attribute information, i.e. current vehicle position, remaining capacity information, task Geographical location, deadline are consistent with precedence information.

Preferably, the specific implementation process of the planning of multi-goal path described in step 5 includes following sub-step:

Step 5.1: establishing optimization problem mathematical model；Equipped with N number of task node, c_ijIt indicates from node S_iTo node S_j's Transportation cost, including distance and electricity；P_ijIt indicates from node S_iTo node S_jConsuming electricity；ΔP_ijIt indicates from node S_iTo section Point S_jCharge capacity；P₀Indicate the initial value of electricity；s_iIt indicates to reach node S_iAt the time of and initial time difference；t_iIt indicates In node S_iThe time of completion task consumption；t_ijIt indicates from node S_iTo node S_jThe time of consumption；d_iIt is expressed as in node S_iAppoint The deadline of business and the difference of initial time；x_ijIndicate vehicle whether from node S_iTo node S_j；

Objective function:

Constraint:

Task restriction:

Decision variable:

x_ij=0or1；

Step 5.2, local paths planning, if f (i, j) is from node S_iTo node S_jCost valuation, g (i, j) is from section Point S_iTo node S_jActual cost, h (i, j) is from node S_iTo node S_jHeuristic information, estimate that each present node supports Up to the minimum cost of destination node, optimal path is obtained using A* algorithm is improved；Work as P_ij> P_curWhen, expression needs to charge, at this moment If meet task time real-time constraint, from node S_iTo node S_jIt can charge, otherwise return to a node and propose that charging is asked Seek simultaneously re-search route；Wherein, P_curIndicate that electric car is located at electricity when present node is point i；

Step 5.3: global path planning uses weight of the f (i, j) as each path in step 5.2 Christofides algorithm building access obtains optimal battery friendly multiple objective programming path.

Preferably, improvement A* algorithm described in step 5.2, if S_curTo be currently located node, S₀For initial time institute In node, S_pThe node where charging pile, S_tFor destination node, S_iFor arbitrary node in map, DEPTH is the depth of nesting, Δ P₀ For every kilometer of power consumption of electric car, L indicates distance between two points；Then improving A* algorithm specific implementation process includes following sub-step It is rapid:

Step 5.2.1: initialization enables S_cur=S₀；Construction Open table is for accessing node to be extended, and initial time For S₀；Construction Close table is for accessing the node extended, and initial time is stored in S₀；Construction PATH table is for accessing optimal road Node on diameter；

Step 5.2.2: it searches in Open table from S_curNearest node S_i, with f (S_cur,S_i) the smallest and meet task The node of constraint and Constraint condition is father node

If meeting the two conditions simultaneously, step 5.2.3 is gone to；

The two conditions can not be met simultaneously, go to step 5.2.5；

Step 5.2.3: if S_iFor destination node S_t, then the node collection stored in PATH table remembers path as optimal path Weight beTerminate algorithm；

Step 5.2.4: if S_iIt is not destination node S_t, then to S_iAll forerunner's state S_jIt is updated, if g_j< g_i+g (i, j), then by S_iIt is put into Close list；Otherwise g is enabled_j=g_i+ g (i, j), by S_iAs S₀The route of next step, enables S_cur=S_i And it is stored in PATH table；Return step 5.2.2；

Step 5.2.5: S is enabled_cur=S₀, DEPTH=DEPTH-1；

If DEPTH=0, step 5.2.6 is gone to；

Otherwise will will meetRange L in charging pile be set as special joint { S_p1,S_p2,…,S_pn, it uses Charge algorithm calculates optimal path；

If meeting task restriction and Constraint condition

Then the weight of outgoing route is f (S₀,S_t)=f (S₀,S_p)+f(S_p,S_t), terminate algorithm；

If task restriction and Constraint condition still can not be met simultaneously, step 5.2.6 is gone to；

Step 5.2.6: the weight of outgoing route is f (S₀,S_t)=∞ terminates algorithm.

Preferably, Charge algorithm described in step 5.2.5, specific implementation process include following sub-step:

Step 5.2.5.1: enabling DEPTH=0, maps { S_p1,S_p2,…,S_pnIt is { S₀₁,S₀₂,…,S_0n, utilize step 5.2.1 f (S is calculated to the principle of step 5.2.6_pi,S_t), i=1,2 ..., n, if f (S_pi,S_t)=∞, then give up p_it(p_iT table Show point p_iTo this route of point t) this route, otherwise calculate f (S₀,S_pi)；

Step 5.2.5.2: DEPTH=DEPTH is enabled_pi, (DEPTH_piIndicate corresponding S_piThe depth of nesting) mapping { S_p1, S_p2,…,S_pnIt is { S_t1,S_t2,…,S_tn, f (S is calculated using the principle of step 5.2.1 to step 5.2.6₀,S_pi), i=1, 2 ..., n, if f (S₀,S_pi)=∞, then give up p_iThis node；

Step 5.2.5.3: to reservation node p_iCalculate separately f (S₀,S_t)=f (S₀,S_pi)+f(S_pi,S_t), i=1,2 ..., N is minimized minf (S₀,S_t)=f (S₀,S_pi)+f(S_pi,S_t), i=1,2 ..., n are path weight value, and algorithm terminates.

Preferably, used described in step 5.3 Christofides algorithm building access realization process include with Lower sub-step:

Step 5.3.1, by task node { S each in city map_t1,S_t2,…,S_tn, retain special joint { S_pi, S_pi+1,…,S_pjAnd start node constitute point set construct minimum spanning tree；

Step 5.3.2: by weight of the f (i, j) as each path in step 5.2, if f (S_i,S_j)=∞, then delete Path ij finds degree of communication in minimum spanning tree and is the vertex of odd number, and creates minimal weight matching, constructs the circuit Eular；

Step 5.3.3: the repetition point in the removal circuit Eular and minimum spanning tree obtains access i.e. optimum programming route.

The present invention has the advantage that

1. optimizing electric car course continuation mileage using the routing of battery sensitivity, charging peak period and traffic congestion phase are avoided, Scientific dispatch drive a vehicle task, guarantee stroke in can relay charging be unlikely to cast anchor.In conjunction with electric car vehicle condition, charging pile work The constraint conditions such as condition, traffic, driving task attribute, provide a plurality of feasible routing plan；

2. intelligent Task sorts, the solution of optimization is provided, is alleviated " mileage anxiety "；

3. realizing that the multi-goal path with task time real-time constraint plans there is deep practical significance, significantly Improve working efficiency.

Detailed description of the invention

Fig. 1: being the device frame schematic diagram of the embodiment of the present invention.

Fig. 2: being the flow chart that A* algorithm is improved in the embodiment of the present invention.

Specific embodiment

Understand for the ease of those of ordinary skill in the art and implement the present invention, with reference to the accompanying drawings and embodiments to this hair It is bright to be described in further detail, it should be understood that implementation example described herein is merely to illustrate and explain the present invention, not For limiting the present invention.

Referring to Fig.1, a kind of electric car course continuation mileage provided by the invention optimizes device, including data collection system, certainly Plan system and expert system；The data collection system includes mobile client, in-vehicle communication system and charging station communication mould Block；Decision system includes the first wireless communication module and decision-making module；Expert system includes the second wireless communication module, database And path planning module；Communication between data collection system, decision system and expert system by being arranged in mobile client Module, the first wireless communication module and the second wireless communication module are connected with each other communication.

A kind of electric car course continuation mileage optimization method provided by the invention, comprising the following steps:

Step 1: user is acquired using mobile client and uploads the driving task attribute information (geography including each task The information such as position, deadline and priority) into expert system；

Step 2: in-vehicle communication system acquire and upload electric car vehicle condition information (including current vehicle position and residue electricity The information such as amount) into expert system；

Step 3: charging station communication module acquires and uploads charging pile work information (service condition and event including charging pile Hinder the information such as situation) into expert system；

Step 4: expert system receives searching database after the data uploaded in step 1, step 2 and step 3, if having One group of existing programme path is then returned with data, otherwise path planning module is transferred to be handled；

Its matching rule are as follows: electric car vehicle condition information is consistent with driving task attribute information, i.e. current vehicle position, surplus Remaining information about power, the geographical location of task, deadline are consistent with precedence information.

Step 5: path planning module combination constraint condition establishes optimization problem model, solves the optimization problem, carries out Multi-goal path planning, provides the routing plan of a plurality of feasible battery friendly；The constraint condition includes electric car Vehicle condition, charging pile operating condition, traffic, driving task attribute；

Wherein the specific implementation process of multi-goal path planning includes following sub-step:

Step 5.1: establishing optimization problem mathematical model；Equipped with N number of task node, c_ijIt indicates from node S_iTo node S_j's Transportation cost, including distance and electricity；P_ijIt indicates from node S_iTo node S_jConsuming electricity；ΔP_ijIt indicates from node S_iTo section Point S_jCharge capacity；P₀Indicate the initial value of electricity；s_iIt indicates to reach node S_iAt the time of and initial time difference；t_iIt indicates In node S_iThe time of completion task consumption；t_ijIt indicates from node S_iTo node S_jThe time of consumption；d_iIt is expressed as in node S_iAppoint The deadline of business and the difference of initial time；x_ijIndicate vehicle whether from node S_iTo node S_j；

Objective function:

Constraint:

Task restriction:

Decision variable:

x_ij=0or1；

Step 5.2, local paths planning, if f (i, j) is from node S_iTo node S_jThe cost valuation of j, g (i, j) be from Node S_iTo node S_jActual cost, h (i, j) is from node S_iTo node S_jHeuristic information, estimate each present node The minimum cost for arriving at destination node obtains optimal path using A* algorithm is improved；Work as P_ij> P_cur(P_curIndicate electric car position Electricity when present node is point i) when, expression needs to charge, if at this moment meet task time real-time constraint, from node S_iTo node S_jIt can charge, otherwise return to a node and propose charge request and re-search route；

See Fig. 2, A* algorithm is improved, if S_curTo be currently located node, S₀The node where initial time, S_pFor charging pile Place node, S_tFor destination node, S_iFor arbitrary node in map, DEPTH is the depth of nesting, Δ P₀It is every kilometer of electric car Power consumption, L indicate distance between two points；Then improving A* algorithm specific implementation process includes following sub-step:

Step 5.2.1: initialization enables S_cur=S₀；Construction Open table is for accessing node to be extended, and initial time For S₀；Construction Close table is for accessing the node extended, and initial time is stored in S₀；Construction PATH table is for accessing optimal road Node on diameter；

Step 5.2.2: it searches in Open table from S_curNearest node S_i, with f (S_cur,S_i) the smallest and meet task The node of constraint and Constraint condition is father node

If meeting the two conditions simultaneously, step 5.2.3 is gone to；

The two conditions can not be met simultaneously, go to step 5.2.5；

Step 5.2.3: if S_iFor destination node S_t, then the node collection stored in PATH table remembers path as optimal path Weight beTerminate algorithm；

Step 5.2.4: if S_iIt is not destination node S_t, then to S_iAll forerunner's state S_jIt is updated, if g_j< g_i+g (i, j), then by S_iIt is put into Close list；Otherwise g is enabled_j=g_i+ g (i, j), by S_iAs S₀The route of next step, enables S_cur=S_i And it is stored in PATH table；Return step 5.2.2；

Step 5.2.5: S is enabled_cur=S₀, DEPTH=DEPTH-1；

If DEPTH=0, step 5.2.6 is gone to；

Otherwise will will meetRange L in charging pile be set as special joint { S_p1,S_p2,…,S_pn, it uses Charge algorithm calculates optimal path；

If meeting task restriction and Constraint condition

Then the weight of outgoing route is f (S₀,S_t)=f (S₀,S_p)+f(S_p,S_t), terminate algorithm；

If task restriction and Constraint condition still can not be met simultaneously, step 5.2.6 is gone to；

Wherein the specific implementation process of Charge algorithm includes following sub-step:

Step 5.2.5.1: enabling DEPTH=0, maps { S_p1,S_p2,…,S_pnIt is { S₀₁,S₀₂,…,S_0n, utilize step 5.2.1 f (S is calculated to the principle of step 5.2.6_pi,S_t), i=1,2 ..., n, if f (S_pi,S_t)=∞, then give up p_it(p_iT table Show point p_iTo this route of point t) this route, otherwise calculate f (S₀,S_pi)；

Step 5.2.5.2: DEPTH=DEPTH is enabled_pi, (DEPTH_piIndicate corresponding S_piThe depth of nesting) mapping { S_p1, S_p2,…,S_pnIt is { S_t1,S_t2,…,S_tn, f (S is calculated using the principle of step 5.2.1 to step 5.2.6₀,S_pi), i=1, 2 ..., n, if f (S₀,S_pi)=∞, then give up p_iThis node；

Step 5.2.5.3: to reservation node p_iCalculate separately f (S₀,S_t)=f (S₀,S_pi)+f(S_pi,S_t), i=1,2 ..., N is minimized minf (S₀,S_t)=f (S₀,S_pi)+f(S_pi,S_t), i=1,2 ..., n are path weight value, and algorithm terminates；

Step 5.2.6: the weight of outgoing route is f (S₀,S_t)=∞ terminates algorithm.

Step 5.3: global path planning uses weight of the f (i, j) as each path in step 5.2 Christofides algorithm building access obtains optimal battery friendly multiple objective programming path；

Realization process using Christofides algorithm building access includes following sub-step:

Step 5.3.1, by task node { S each in city map_t1,S_t2,…,S_tn, retain special joint { S_pi, S_pi+1,…,S_pjAnd start node constitute point set construct minimum spanning tree；

Step 5.3.2: by weight of the f (i, j) as each path in step 5.2, if f (S_i,S_j)=∞, then delete Path ij finds degree of communication in minimum spanning tree and is the vertex of odd number, and creates minimal weight matching, constructs the circuit Eular；

Step 5.3.3: the repetition point in the removal circuit Eular and minimum spanning tree obtains access i.e. optimum programming route.

Step 6: after decision system receives the feasible one group of programme path of the condition of satisfaction, taking preceding k according to probability feasibility Optimal route returns to mobile client, and user is transferred to select.

It should be understood that the part that this specification does not elaborate belongs to the prior art.

It should be understood that the above-mentioned description for preferred embodiment is more detailed, can not therefore be considered to this The limitation of invention patent protection range, those skilled in the art under the inspiration of the present invention, are not departing from power of the present invention Benefit requires to make replacement or deformation under protected ambit, fall within the scope of protection of the present invention, this hair It is bright range is claimed to be determined by the appended claims.

Claims (7)

1. a kind of method for carrying out the optimization of electric car course continuation mileage using electric car course continuation mileage optimization device, feature exist In: the electric car course continuation mileage optimizes device, including data collection system, decision system and expert system；The number It include mobile client, in-vehicle communication system and charging station communication module according to acquisition system；Decision system includes the first channel radio Interrogate module and decision-making module；Expert system includes the second wireless communication module, database and path planning module；The data Communication module, the first wireless telecommunications mould between acquisition system, decision system and expert system by being arranged in mobile client Block and the second wireless communication module are connected with each other communication；

It the described method comprises the following steps:

Step 1: user is acquired using mobile client and uploads driving task attribute information into expert system；

Step 2: in-vehicle communication system acquires and uploads electric car vehicle condition information into expert system；

Step 3: charging station communication module acquires and uploads charging pile work information into expert system；

Step 4: expert system receives searching database after the data uploaded in step 1, step 2 and step 3, if there is coupling number According to one group of existing programme path is then returned, otherwise path planning module is transferred to be handled；

Step 5: path planning module combination constraint condition establishes optimization problem model, solves the optimization problem, carries out more mesh Path planning is marked, the routing plan of a plurality of feasible battery friendly is provided；The constraint condition include electric car vehicle condition, Charging pile operating condition, traffic, driving task attribute；

The specific implementation process of the multi-goal path planning includes following sub-step:

Step 5.1: establishing optimization problem mathematical model；Equipped with N number of task node, c_ijIt indicates from node S_iTo node S_jTransport Cost, including distance and electricity；P_ijIt indicates from node S_iTo node S_jConsuming electricity；ΔP_ijIt indicates from node S_iTo node S_j Charge capacity；P₀Indicate the initial value of electricity；s_iIt indicates to reach node S_iAt the time of and initial time difference；t_iIt indicates Node S_iThe time of completion task consumption；t_ijIt indicates from node S_iTo node S_jThe time of consumption；d_iIt is expressed as in node S_iTask Deadline and initial time difference；x_ijIndicate vehicle whether from node S_iTo node S_j；

Objective function:

Constraint:

Task restriction:

Decision variable:

x_ij=0or1；

Step 5.2, local paths planning, if f (i, j) is from node S_iTo node S_jCost valuation, g (i, j) is from node S_i To node S_jActual cost, h (i, j) is from node S_iTo node S_jHeuristic information, estimate each present node arrive at mesh The minimum cost for marking node obtains optimal path using A* algorithm is improved；Work as P_ij> P_curWhen, expression needs to charge, if at this moment full When sufficient task time real-time constraint, then from node S_iTo node S_jIt can charge, otherwise return to a node and propose charge request simultaneously Re-search route；Wherein, P_curIndicate that electric car is located at electricity when present node is point i；

The improvement A* algorithm, if S_curTo be currently located node, S₀The node where initial time, S_pIt is saved where charging pile Point, S_tFor destination node, S_iFor arbitrary node in map, DEPTH is the depth of nesting, Δ P₀For every kilometer of power consumption of electric car, L indicates distance between two points, then improving A* algorithm specific implementation process includes following sub-step:

Step 5.2.1: initialization enables S_cur=S₀；Construction Open table is for accessing node to be extended, and initial time is S₀； Construction Close table is for accessing the node extended, and initial time is stored in S₀；Construction PATH table is for accessing optimal path Node；

Step 5.2.2: it searches in Open table from S_curNearest node S_i, with f (S_cur,S_i) the smallest and meet task restriction Node with Constraint condition is father node

If meeting the two conditions simultaneously, step 5.2.3 is gone to；

The two conditions can not be met simultaneously, go to step 5.2.5；

Step 5.2.3: if S_iFor destination node S_t, then the node collection stored in PATH table remembers the power in path as optimal path Weight isTerminate algorithm；

Step 5.2.4: if S_iIt is not destination node S_t, then to S_iAll forerunner's state S_jIt is updated, if g_j< g_i+g(i, J), then by S_iIt is put into Close list；Otherwise g is enabled_j=g_i+ g (i, j), by S_iAs S₀The route of next step, enables S_cur=S_iAnd it deposits Enter in PATH table；Return step 5.2.2；

Step 5.2.5: S is enabled_cur=S₀, DEPTH=DEPTH-1；

If DEPTH=0, step 5.2.6 is gone to；

Otherwise will meetRange L in charging pile be set as special joint { S_p1,S_p2,…,S_pn, it is calculated using Charge Method calculates optimal path；

If meeting task restriction and Constraint condition

Then the weight of outgoing route is f (S₀,S_t)=f (S₀,S_p)+f(S_p,S_t), terminate algorithm；

If task restriction and Constraint condition still can not be met simultaneously, step 5.2.6 is gone to；

Step 5.2.6: the weight of outgoing route is f (S₀,S_t)=∞ terminates algorithm；

Step 5.3: global path planning uses weight of the f (i, j) as each path in step 5.2 Christofides algorithm building access obtains optimal battery friendly multiple objective programming path；

Step 6: after decision system receives the feasible one group of programme path of the condition of satisfaction, taking preceding k item most according to probability feasibility Major path returns to mobile client, and user is transferred to select.

2. according to the method described in claim 1, it is characterized by: driving task attribute information described in step 1 includes every Geographical location, deadline and the precedence information of item task.

3. according to the method described in claim 1, it is characterized by: electric car vehicle condition information described in step 2 includes vehicle Current location and remaining capacity information.

4. according to the method described in claim 1, it is characterized by: charging pile work information described in step 3 includes charging The service condition and fault condition information of stake.

5. according to the method described in claim 1, it is characterized by: one group is returned to described in step 4 if having matched data Existing programme path, matching rule are as follows: electric car vehicle condition information is consistent with driving task attribute information, i.e., vehicle is current Position, remaining capacity information, the geographical location of task, deadline are consistent with precedence information.

6. according to the method described in claim 1, it is characterized by: Charge algorithm described in step 5.2.5, specific implementation Process includes following sub-step:

Step 5.2.5.1: enabling DEPTH=0, maps { S_p1,S_p2,…,S_pnIt is { S₀₁,S₀₂,…,S_0n, extremely using step 5.2.1 The principle of step 5.2.6 calculates f (S_pi,S_t), i=1,2 ..., n, if f (S_pi,S_t)=∞, then give up p_iThis route of t, otherwise Calculate f (S₀,S_pi), wherein p_iT indicates point p_iTo this route of point t；

Step 5.2.5.2: DEPTH=DEPTH is enabled_pi, map { S_p1,S_p2,…,S_pnIt is { S_t1,S_t2,…,S_tn, utilize step 5.2.1 f (S is calculated to the principle of step 5.2.6₀,S_pi), i=1,2 ..., n, if f (S₀,S_pi)=∞, then give up p_iThis section Point；Wherein DEPTH_piIndicate corresponding S_piThe depth of nesting；

Step 5.2.5.3: to reservation node p_iCalculate separately f (S₀,S_t)=f (S₀,S_pi)+f(S_pi,S_t), i=1,2 ..., n take Minimum value minf (S₀,S_t)=f (S₀,S_pi)+f(S_pi,S_t), i=1,2 ..., n are path weight value, and algorithm terminates.

7. according to the method described in claim 1, it is characterized by: using Christofides algorithm described in step 5.3 The realization process for constructing access includes following sub-step:

Step 5.3.1, by task node { S each in city map_t1,S_t2,…,S_tn, retain special joint { S_pi,S_pi+1,…, S_pjAnd start node constitute point set construct minimum spanning tree；

Step 5.3.2: by weight of the f (i, j) as each path in step 5.2, if f (S_i,S_j)=∞, then delete path Ij finds degree of communication in minimum spanning tree and is the vertex of odd number, and creates minimal weight matching, constructs the circuit Eular；

Step 5.3.3: the repetition point in the removal circuit Eular and minimum spanning tree obtains access i.e. optimum programming route.