CN110146103B

CN110146103B - Unmanned equipment path planning method considering target trend and energy supply

Info

Publication number: CN110146103B
Application number: CN201910513985.0A
Authority: CN
Inventors: 李龙江; 范鹏辉; 梁昊阳
Original assignee: University of Electronic Science and Technology of China
Current assignee: University of Electronic Science and Technology of China
Priority date: 2019-06-14
Filing date: 2019-06-14
Publication date: 2021-03-02
Anticipated expiration: 2039-06-14
Also published as: CN110146103A

Abstract

The invention discloses an unmanned equipment path planning method considering target direction and energy supply, which is applied to the field of unmanned equipment path planning and aims to solve the problem that the energy supply of unmanned equipment in a midway energy station is not considered in the prior art.

Description

Unmanned equipment path planning method considering target trend and energy supply

Technical Field

The invention belongs to the field of path planning of unmanned equipment, and particularly relates to a path planning technology of the unmanned equipment, which considers the trend of a target and energy supply.

Background

The intelligent unmanned equipment is an important component of the internet of things era, and the path planning problem is one of the key problems to be solved in the development of the intelligent unmanned equipment. When the unmanned equipment completes a given task, the unmanned equipment needs to pass through a plurality of intermediate nodes, the selection of the passed paths is diversified and not unique, different paths have different execution times and different consumed energy sources, and therefore the unmanned equipment needs to consider the constraint conditions of the paths to find the optimal path under the condition that the requirements and conditions defined by a system are met, which is also the premise of intelligent equipment movement. For example, in emergency rescue, unmanned devices move among various target points due to high requirements on short time and low time delay, and how to reach a rescued target area with minimum cost and transmit data back to a control center in real time is a problem to be considered. Therefore, many scholars have also studied the problem of optimizing path planning.

According to the existing literature search, in the path planning problem, although there are many path planning algorithms, most of the path planning algorithms only consider the weight of the moving path edge, and do not consider the weight of the intermediate node, that is, the practical problem that the unmanned device performs energy supplement at the energy station midway is not considered, and the practical problems that the future target trend and the paths and the intermediate node are influenced are not considered, for example, due to different road conditions, different speeds through each path, different queuing times and energy prices of each energy supplement station node and the like, the practical problems can influence the path planning of the unmanned device, and meanwhile, the existing algorithms also have many practical problems of high model complexity, low algorithm convergence speed and the like.

Disclosure of Invention

In order to solve the technical problems, the invention provides an unmanned equipment lightweight path planning method considering energy supply, considers a vehicle moving model in reality, and provides a strategy for comprehensively considering node loss factors such as speed, energy supply time and energy price of a vehicle in a driving process to realize an optimal moving scheme.

The technical scheme adopted by the invention is as follows: the unmanned equipment lightweight path planning method considering energy replenishment comprises the following steps:

s1, establishing an energy supply node set according to the target trend sequence window, and establishing a first topological graph according to the energy supply node set;

s2, expanding the source nodes and all the target nodes in the target trend sequence window into the first topological graph to obtain a second topological graph;

s3, updating the weight of the second topological graph;

s4, calculating the optimal path from the source node to the current target node by adopting a trend reverse push algorithm according to the topological graph updated by the weight in the step S3;

s5, advancing to the current target node, and ending if the current target node is the global final target node; otherwise, taking the current target node as a new source node, and calculating the optimal path from the source node to the next target node.

Further, the weight values of step S3 are updated in consideration of the money cost:

the method comprises the steps of updating a money cost weight between two energy source nodes according to a distance cost unit price between the two energy source supplement nodes, updating a money cost weight between a source node and an adjacent energy source supplement node according to the distance cost unit price between the source node and the adjacent energy source supplement node, updating a money cost weight between a target node and the adjacent energy source supplement node according to the distance cost unit price between the target node and the adjacent energy source supplement node, and updating the money cost weight of the energy source supplement node according to the maximum capacity of equipment energy corresponding to unmanned equipment, the energy source unit price corresponding to the energy source supplement node and energy required to be supplemented at the energy source supplement node.

Further, the weight values of step S3 are updated in consideration of time consumption: the time weight between the two energy source nodes is updated according to the passing time of the distance between the two energy source supplement nodes, the time weight between the source node and the adjacent energy source supplement node is updated according to the passing time of the distance between the source node and the adjacent energy source supplement node, the time weight between the target node and the adjacent energy source supplement node is updated according to the passing time of the distance between the target node and the adjacent energy source supplement node, and the time weight of the energy source supplement node is updated according to the maximum energy capacity corresponding to the unmanned equipment, the energy supplement speed of the energy source supplement node and the energy needed to be supplemented at the energy source supplement node.

Further, the weight values in step S3 are updated by comprehensively considering the money cost and the time cost:

definition of R_MDefining R as a weight coefficient for money cost_TCost a weight coefficient for time, and R_M+R_T＝1；

According to the money cost weight, the time cost weight and R between the two energy supplement nodes_M、R_TUpdating the weight between the two energy nodes according to the money cost weight, the time cost weight and R between the source node and the adjacent energy supplement node_M、R_TUpdating the weight between the source node and the adjacent energy supplement node according to the money cost weight, the time cost weight, R between the target node and the adjacent energy supplement node_M、R_TUpdating the weight between the target node and the adjacent energy supplement node according to the money cost weight, the time cost weight and R of the energy supplement node_M、R_TAnd updating the weight of the energy supplement node.

Further, the step S1 includes the following sub-steps:

s11, establishing an energy supply node set according to the target trend sequence window;

and S12, if the distance between the two energy supply nodes in the step S11 set is smaller than the full energy movement distance of the unmanned equipment, adding a connecting line between the two energy supply nodes into the topological graph as a directed edge, wherein the weight of the directed edge is the distance between the two energy supply nodes, and traversing the pairwise combination of all the energy supply nodes to obtain a final first topological graph.

Further, step S2 includes:

s21, adding the source node to the first topological graph, specifically: if the distance from the source node to a certain energy source node is smaller than the maximum distance that the unmanned equipment can travel in the residual energy of the source node, adding a connecting line between the source node and the energy source supply node into the current topological graph as a directed edge, wherein the weight of the directed edge is the distance between the source node and the energy source supply node;

s22, adding the target node in the target heading sequence window to the first topological graph, which specifically includes: if the distance from a certain energy node to a target node is smaller than the difference between the full energy movement distance of the unmanned equipment and the distance from the target node to the nearest energy supply node, adding a connecting line between the energy supply node and the target node into the current topological graph as a directed edge, wherein the weight of the directed edge is the distance between the energy supply node and the target node;

s23, obtaining a second topological graph according to the step S21 and the step S22.

Further, the trend reverse algorithm in step S4 calculates the best path to the next target node, including the following sub-steps:

a1, calculating the energy needed to be reserved from the current target node to the next hop in the target moving sequence window;

a2, calculating the optimal path from the previous target node to the current target node by adopting Dijkstra algorithm according to the energy required to be reserved by the current target node and the second topological graph of the step S3;

or according to the energy required to be reserved by the current target node and the second topological graph of the step S3, calculating the optimal path from the source node to the current target node by adopting a Dijkstra algorithm.

Further, step a1 specifically includes:

if the current target node is the last target node in the target trend sequence window, taking the energy supply node closest to the target node as the next hop of the current target node, and calculating to obtain the energy required to be reserved from the current target node to the next hop;

otherwise, the next hop of the current target node is obtained according to the optimal path from the target node to the next target node, and the energy required to be reserved from the current target node to the next hop is calculated and obtained.

Further, the step S5 of calculating the best path from the source node to the next target node by taking the current target node as the new source node includes the following sub-steps:

b1, taking the current target node as a new source node, and updating the target moving sequence window;

b2, judging whether the current first topological graph covers the energy supply node set corresponding to the updated target trend sequence window, if so, returning to the step S2, and updating the current second topological graph; otherwise, returning to step S1, updating the current first topological graph according to the updated target trend sequence window.

The invention has the beneficial effects that: in the invention, a network topological graph is established in advance, and according to road conditions, energy supplementing time, price and other factors which need to be considered in reality, and the emphasis of different tasks on different weight factors are considered, the time and cost are jointly optimized, and a normalized weight network is established; the unmanned equipment acquires information of a weight network before executing the task, and an optimal path for executing the task is planned by using the target node energy constraint condition; meanwhile, the invention also provides an unmanned equipment path planning method considering the target trend and energy supply, the method is more suitable for the actual unmanned equipment path planning, the optimal path planning scheme can be obtained, and the operation cost of the unmanned equipment is reduced.

Drawings

Fig. 1 is a flow chart of the unmanned aerial vehicle lightweight path planning method considering energy replenishment.

Fig. 2 is a scene schematic diagram of the unmanned aerial vehicle lightweight path planning method considering energy replenishment.

Fig. 3 is an energy supply node network diagram of the unmanned aerial vehicle lightweight path planning method considering energy supply of the present invention.

Fig. 4 is an illustrative diagram of the unmanned aerial vehicle lightweight path planning method considering energy replenishment.

Detailed Description

In order to facilitate the understanding of the technical contents of the present invention by those skilled in the art, the present invention will be further explained with reference to the accompanying drawings.

In the invention, in the process that the unmanned equipment passes through a plurality of intermediate energy supply nodes from a source node to a target node, the optimal moving path of the unmanned equipment when executing a series of tasks is determined by considering the path planning of multi-factor joint optimization.

Fig. 1 is a flow chart of the method of the present invention, in which a coordinate map, paths and energy supply costs of all energy supply nodes are known, and fig. 2 is a schematic view of a scene provided in this embodiment, including a source node q_sA plurality of target nodes (Q ═<q₁,q₂,…,q_n>) And a plurality of energy supply nodes (P ═ P)₁,P₂,…,P_m]). Solving the following steps: current position s ═ q of vehicle₀To a series of target points (Q ═ Q)<q₁,q₂,…,q_n>) The optimal path of (2); the realization process of the invention is as follows:

step 1: and constructing an energy supply node topological graph G according to the trend of the unmanned equipment.

Step 1.1: based on the map, energy supply nodes near the target of the unmanned device are found to form an energy supply node set (P ═ P)₁,P₂,…,P_m]). The specific implementation steps are as follows:

step 1.2.1: the drone target course is represented by a sequence of source and target nodes as W ═ tone<q_i,q_i+1>I-s … s + w-1, indicating that the drone is to be routed by the source node q_sStarting from that, i.e. passing q_s+1,q_s+2,…,q_s+wAnd waiting for the target nodes, wherein w is the maximum number of the target nodes considered in the current calculation, and can be preconfigured or adaptively adjusted, and the value of w in the embodiment is 3, that is, the target moving sequence window at least comprises one source node and two target nodes.

Step 1.2.2: and setting P as an empty set, respectively checking the target trend sequence, and adding all energy complementing nodes near the trend sequence into P. The specific method is to<q_i,q_i+1>A circular area is defined on the map for the diameter, and all energy-complementing nodes located within the circular area are added to P, i ═ s … s + w-1.

Step 1.2.3: the complementary nodes in P are numbered as (P (1),,, P (m)).

Step 1.2: and (3) for the energy supply node sets (P (1),, P (m)), m represents the total number of the energy supply nodes, the distance between every two energy supply nodes is checked, and if a path with the distance smaller than the full energy moving distance L exists between P (i) and P (j), P (i) P (j) is used as a directed edge to be added into G. The specific implementation steps are as follows:

step 1.2.1: calculating the distance between P (1) and P (2), P (3) … P (m), respectively, assuming that the maximum distance that the unmanned device can move when fully powered is L, if d_(P(1),P(j))<L, if P (1) P (j) is called as the mutual neighbor point, then P (1) P (j) is added into the topological graph G, and the weight given to the edge is d_(P(1),P(j))；

Step 1.2.2: respectively calculating the distance between P (2) and P (1) P (3) … P (m), assuming that the maximum distance that the unmanned equipment can move when the unmanned equipment is full of energy is L, if d_(P(2),P(j))<L, adding P (2) P (j) into the edge set of the topological graph G, and giving the weight of the edge d_(P(2),P(j)). And by analogy, each node meeting the conditions is added into the network topology graph G. A schematic diagram of which is shown in fig. 3.

Step 2: and adding the nodes in the trend sequence W into the network topological graph to generate an expanded topological graph G'.

And copying the topology graph G to be used as an expanded topology graph G ', G ' is G, and adding nodes in the sequence W to G '.

Step 2.1: the first node in the sequence W, i.e. the source node q_sAdding to the topology map

If from q_sTo P (i) by a distance d₍q_s,P(i))Paths less than Ls, then (q) will be_sP (i)) adds G' as a directed edge. Here, Ls is the vehicle at q_sThe maximum distance that the remaining energy source can travel.

Step 2.2: the remaining target nodes { q ] in the sequence W_iWhere i + s +1, …, s + w, is added to the augmented topology graph G'

If from P (j) to q_iExists at a distance d_(P(j),qi)If the path is less than L-L (r), then (P (j), q)_i) G' is added as a directed edge. Where L (r) is from q_iThe closest distance to a nearby charging site. The establishment of the whole network topological graph is completed. A schematic diagram of which is shown in fig. 4.

And step 3: and calculating the weight on the extended topological graph G'.

Updating the weight of the edge set and the point set to complete the establishment of a weight network: the weight given to the edge set is only dependent on the length of the path, however, in an actual environment, the unmanned device needs to consider multiple comprehensive factors, such as road conditions and time for queuing to supplement energy, the road conditions and the road toll of each path are different, the time and the price of the supplement energy of each energy supplement node are different, and thus, the unmanned device needs to give weight again according to the actual situation. The weight calculation method is divided into the following 3 types of situations, and can be configured:

case 1. weight assignment in terms of money cost: the road conditions of each route are different, and the road cost unit price required by each route is different, and the road cost unit price between every two energy supplement nodes P (i) and P (j) is assumed to be M_p(i, j), then the required road cost for this trip is M_p(i,j)*d_(P(i),P(j))This value is assigned to the edge P (i) P (j) as its weight in terms of money costs, denoted W_M(i, j) between the source node and its adjacent energy supply nodeThe weights of (A) are similar and are denoted as W_M(q_sI) the weight between the target node and its neighboring nodes is denoted as W_M(j,q_i). The energy unit price of each energy supplement node is different and is set as M_p(i) And i is the number of each energy replenishment node. Assuming that each additional energy is full and the maximum capacity of the energy source of the device is F_maxSince the energy consumed between P (i) and P (j) is (d)_(P(i),P(j))*F_max) And L, the energy needed to be supplemented at the ith energy supplement node is L- (d)_(P(i),P(j))*F_max) L, the cost of supplemental energy requirement is [ L- (d)_(P(i),P(j))*F_max)/L]*M_p(i) This value is assigned to the node as its weight in terms of monetary cost, denoted W_M(i) In that respect Obviously, the weight of each edge is fixed, but the weight update of each energy supplement node is dynamic, which depends on the situation of the previous edge and also depends on the energy price of each energy supplement node. The updating of the weight of each edge and each energy supplement node in terms of cost is completed.

Case 2. weight assignment in time consumption: since the situation is different for each journey, the speed at which they pass is also different, which results in different times for passing through them. Assuming that P (i) P (j) the desired speed of the path is V (i, j), the time required to traverse this edge is d_(P(i),P(j))V (i, j), this value is denoted as W_T(i, j) and assigns it to this edge as a weight. And the weight between the source node and the energy supplement node adjacent to the source node is similar and is denoted as W_T(q_sI) the weight between the target node and its neighboring nodes is denoted as W_T(q_iJ). The energy supplementing speed of each energy supplementing node is different and is set as V_p(i) And i is the number of each energy replenishment node. Assuming that each additional energy is full and the maximum capacity of the energy source of the device is F_maxSince the energy consumed between P (i) and P (j) is (d)_(P(i),P(j))*F_max) And L, the energy needed to be supplemented at the ith energy supplement node is L- (d)_(P(i),P(j))*F_max) /L, supplementary energyThe required time is [ L- (d)_(P(i),P(j))*F_max)/L]/V_p(i) Considering the different crowdedness of each energy supplement node, suppose that the queuing waiting time of the node p (i) is expected to be t (i), then the total time spent at this node is [ L- (d)_(P(i),P(j))*F_max)/L]/V_p(i) + T (i), assigning this value to the node as its weight in time consumption, denoted W_T(i) In that respect Obviously, the weight of each edge is fixed, but the weight update of each energy supplement node is dynamic, depending on the situation of the previous edge, the shorter the previous edge (the smaller the parent edge weight), the larger the weight of the point is, and the longer the previous edge (the larger the parent edge weight), the smaller the weight of the point is. The weighting is also determined by the queue time expectation of each energy replenishment node. This completes the weight update for each edge and each energy replenishment node in terms of time spent.

Case 3. Integrated empowerment of money and time costs: the degree of emphasis on monetary cost and time consumption varies among tasks. For example, in disaster relief, the task is demanding a high expenditure of time, but in the course of transporting goods, the task is demanding a high expenditure of money. Based on this consideration, R is defined_MDefine R as the weight of money cost_TIs weighted for time, and R_M+R_TThese two values are freely defined according to the task requirements, 1. Then the total weight of each edge P (i) P (j) is W_all(i,j)＝R_M*W_M(i,j)+R_T*W_T(i, j) similarly, the total weight of each energy supplement node P (i) is W_all(i)＝R_M*W_M(i)+R_T*W_T(i) In that respect In particular, assume node P (I)₁+1) represents the target node q_iIts own weight is 0. This completes the point and edge entitlements.

Step 4, calculating the optimal path by using a trend reverse-thrust algorithm

An optimization path algorithm considering a node weight function and a target trend has the basic idea that the calculation is performed reversely along a trend sequence W, and the calculation is performed from q to q respectively_i-1To q_iI from s + w to s +1, and finally q_sGo to q_s+1The optimal path of (2). The trend reverse-push algorithm computes pseudo-code as follows:

step 4.1: calculating the slave q on G' by using Dijkstra algorithm_i-1To q_iThe shortest path of (2). The Dijkstra algorithm is a well-known algorithm in the computer field and the invention is not described in detail here.

Step 4.2: once q is found_i-1To q_iOf the unmanned device is in q_s+w-1The energy needed to be reserved is determined, and the unmanned equipment q is updated_i-1Energy sources requiring conservation at the site, i.e. Remainc (q)_i-1)＝L-dist(q_i-1；v_i-1(next))。v_i-1(next) is q_i-1To q_iMay be the target node or the complementary station.

5. After proceeding to the next target point, the optimal path is recalculated

When the target site q is reached along the optimal path_s+1And then judging whether the last target site is reached. If the last target point has been reached, the algorithm ends.

Otherwise, updating the current position to q by s-s +1_sAnd updating the trend sequence W, including a new target node, if the topological graph G already covers the range of the W, turning to the step 2, otherwise, turning to the step 1, and updating the topological graph G.

The invention is further illustrated by the following specific example.

As shown in fig. 4, a source node and a target node and an energy supplement node are given, and a network topology graph is established according to step 1, and each parameter of the edges and nodes has been labeled on the graph. Suppose unmanned equipment is full energy capacity F _max20, 100, weight of time spent and money spent R_M＝0.1，R_T0.9, then the following will be followedProceeding from source node arc to target node q_iFor the sake of illustration, the optimal path is planned by considering only the case of reaching one target node. Here, arc represents the source node q_sDst is the destination node q_i。

First, an array d [2 ] having a length of 4 is defined]Element d [ i-1 ] of]Represents the shortest distance from the source point arc to the node p (i). Array f [2 ] with a defined length of 4]. Its element f [ i-1 ]]If 1, it means that the shortest path from the source point arc to P (i) has been found, its element f [ i-1 ]]If 0, it means that the shortest path from the source point arc to P (i) has not been found. Defining an array p of length 5 for each node P (i)_i[]It stores the node numbers that are passed through in sequence from the source point to p (i), i.e. the shortest path.

And calculating the weights of the nodes and edges adjacent to the source node. Considering arc to P (1) first, the required road cost for this journey is W in terms of monetary cost_M(arc,1)＝M_p(arc,1)*d_(arc,1)3 x 50 x 150 x. The monetary cost at P (1) is W_M(1)＝[(d_(arc,P(1))*F_max)/L]*M_p(1) 50. Considering arc to P (1), the time required for this stretch is W in terms of time spent_T(arc,1)＝d_(arc,P(1))V (arc,1) ═ 1.67. The time spent at P (1) is W_T(1)＝[(d_(arc,P(1))*F_max)/L]/V_p(1) + T (1) ═ 2. Thus, the total weight W of edge arcP (1)_all(arc,1)＝R_M*W_M(arc,1)+R_T*W_T(arc,1) — 16.5, total weight W of point P (1)_all(1)＝R_M*W_M(1)+R_T*W_T(1) 6.8. The total weight of the path from arc directly to P (1) is then W_all(arc,1)+W_all(1)＝16.5+6.8＝23.3。

Then consider arc to P (3), the required road cost for this journey is W in terms of monetary cost_M(arc,3)＝M_p(arc,3)*d_(arc,3)40 x 0.4 x 16. The monetary cost at P (3) is W_M(3)＝[(d_(arc,P(3))*F_max)/L]*M_p(3) 32. Considering arc to P (3), the time required for this trip in terms of time costIs W_T(arc,3)＝d_(arc,P(3))V (arc,3) ═ 2. The time spent at P (3) is W_T(3)＝[(d_(arc,P(3))*F_max)/L]/V_p(3) + T (3) ═ 1. Thus, the total weight W of edge arcP (3)_all(arc,3)＝R_M*W_M(arc,3)+R_T*W_T(arc,3) ═ 3.4, total weight W of point P (3)_all(3)＝R_M*W_M(3)+R_T*W_T(3) 4.1. The total weight of the path from arc directly to P (3) is then W_all(arc,3)+W_all(3)＝3.4+4.1＝7.5。

Updating each array: d 0]And d 2]Updated to 23.3 and 7.5, respectively, and the other elements are ∞. Due to d [0 ]]>d[2]Then P (3) is the current closest point to the source point, updating the array f [2 ]]1, the other elements are 0, and the array p is updated₃[0]＝0，p₃[1]＝3，p₁[0]＝0，p₁[1]The first round of finding the minimum path is now completed by 1.

Taking P (3) as an intermediate path, and continuing the next round of operation. Considering P (3) to P (1), the required road toll for this route is W in terms of monetary cost_M(3,1)＝M_p(3,1)*d_(p(3),P(1))0.6 × 40 × 24. The monetary cost at P (1) is W_M(1)＝(d_(p(3),P(1))*F_max)/L*M_p(1) 40. Considering P (3) to P (1), the time required for this route in terms of time consumption is W_T(3,1)＝d_(p(3),P(1))And V (3,1) is 4. The time spent at P (1) is W_T(1)＝[(d_(p(3),P(1))*F_max)/L]/V_p(1) + T (1) ═ 1.8. Thus, the total weight W of the edge P (3) P (1)_all(3,1)＝R_M*W_M(3,1)+R_T*W_T(3,1) ═ 6, total weight W of point P (1)_all(1)＝R_M*W_M(1)+R_T*W_T(1) 5.6. The total weight of the path from P (3) directly to P (1) is then W_all(3,1)+W_all(1)＝6+5.6＝11.6。

Considering P (3) to P (2), the required road toll for this route is W in terms of monetary cost_M(3,2)＝M_p(3,2)*d_(p(3),P(2))＝1*9090. The monetary cost at P (2) is W_M(2)＝(d_(p(3),P(2))*F_max)/L*M_p(2) 108. Considering P (3) to P (2), the time required for this route in terms of time consumption is W_T(3,2)＝d_(p(3),P(2))V (3,2) ═ 4.5. The time spent at P (2) is W_T(2)＝[(d_(p(3),P(2))*F_max)/L]/V_p(2) + T (2) ═ 2.8. Thus, the total weight W of the edge P (3) P (2)_all(3,2)＝R_M*W_M(3,2)+R_T*W_T(3,2) 13, the total weight W of the point P (2)_all(2)＝R_M*W_M(2)+R_T*W_T(2) 13.3. The total weight of the path from P (3) directly to P (2) is then W_all(3,2)+W_all(2)＝13+13.3＝26.3。

Updating each array: d 0]＝23.3>d[2]+W_all(3,1)+W_all(1) So d [0 ] is updated as 19.1]Update d [1 ] 19.1]＝d[2]+W_all(3,2)+W_all(2) 33.8, the other elements are unchanged. Due to d [0 ]]<d[1]Then P (1) is the current closest point to the source point, and the array f [0 ] is updated]Other elements are 0. Because d [0 ]]With updates, so the array p is updated₁[]＝p₃[]And p is₁[2]1. At this time p₁[]＝[0,3,1,0,0]. The second round of finding the minimum path is now complete.

Following the above procedure, up to f 3]When p is equal to 1₄[]＝[0,3,1,2,4]The optimal paths are arc, P (3), P (1), P (2), q_iAs shown in fig. 4. Its path total weight value is stored in array d [3 ]]In (1).

It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. An unmanned equipment path planning method considering target trend and energy supply is characterized by comprising the following steps:

s1, establishing an energy supply node set according to the target trend sequence window, and establishing a first topological graph according to the energy supply node set; the step S1 includes the following sub-steps:

s12, if the distance between the two energy supply nodes in the step S11 set is smaller than the full energy movement distance of the unmanned equipment, a connecting line between the two energy supply nodes is used as a directed edge to be added into the topological graph, the weight of the directed edge is the distance between the two energy supply nodes, and the final first topological graph is obtained by traversing the pairwise combination of all the energy supply nodes;

s2, expanding the source nodes and all the target nodes in the target trend sequence window into the first topological graph to obtain a second topological graph; step S2 includes:

s22, adding the target node in the trend window into the first topological graph, specifically: if the distance from a certain energy node to a target node is smaller than the difference between the full energy movement distance of the unmanned equipment and the distance from the target node to the nearest energy supply node, adding a connecting line between the energy supply node and the target node into the current topological graph as a directed edge, wherein the weight of the directed edge is the distance between the energy supply node and the target node;

s23, obtaining a second topological graph according to the step S21 and the step S22;

s3, updating the weight of the second topological graph;

2. The method for unmanned aerial vehicle path planning considering target trend and energy replenishment according to claim 1, wherein the weight values of step S3 are updated in consideration of money cost:

3. The method for planning the path of the unmanned aerial vehicle considering the trend of the target and the energy supply as claimed in claim 1, wherein the weight values of step S3 are updated in consideration of time consumption: the time weight between the two energy source nodes is updated according to the passing time of the distance between the two energy source supplement nodes, the time weight between the source node and the adjacent energy source supplement node is updated according to the passing time of the distance between the source node and the adjacent energy source supplement node, the time weight between the target node and the adjacent energy source supplement node is updated according to the passing time of the distance between the target node and the adjacent energy source supplement node, and the time weight of the energy source supplement node is updated according to the maximum energy capacity corresponding to the unmanned equipment, the energy supplement speed of the energy source supplement node and the energy needed to be supplemented at the energy source supplement node.

4. The method for planning the path of the unmanned aerial vehicle considering the trend of the target and the energy supply as claimed in claim 2 or 3, wherein the weight values in step S3 are updated by comprehensively considering the cost of money and the cost of time:

5. The unmanned aerial vehicle path planning method considering target trend and energy replenishment as claimed in claim 1, wherein the trend back-stepping algorithm of step S4 calculates the optimal path to the next target node, comprising the following sub-steps:

6. The method for unmanned aerial vehicle path planning considering goal trend and energy supply according to claim 5, wherein the step A1 is specifically as follows:

7. The unmanned aerial vehicle path planning method considering target trend and energy replenishment as claimed in claim 6, wherein the step S5 of calculating the optimal path from the source node to the next target node with the current target node as the new source node comprises the following sub-steps: