Speed-up method for full-array path planning for multi-point return-to-source of tourist attractions

Abstract

The invention relates to an acceleration method for full-permutation path planning for scenic spot multi-point source returning. The acceleration method comprises the following steps: (1) setting thenumber of tourist attractions to be more than 3, recording coordinates of all the tourist attractions and shortest paths and distance values among all the tourist attractions, forming a path matrix, and establishing an xy two-dimensional coordinate system by taking a starting point as an origin of coordinates; (2) based on the xy two-dimensional coordinate system, establishing a two-dimensional coordinate system about the included angle value; (3) for each subarea, eliminating useless arrangement in a full-permutation path through a convex polygon jump point elimination method; and (4) arranging the paths among the partitions by using a full-permutation path planning method to obtain the shortest path value. The method has the advantages that on the basis of the full-permutation path planning method, aiming at the scenic spot distribution rule of a city, the calculation times of the full-permutation path planning method are reduced under the condition of ensuring that the shortest pathtraversing all passing points is found in practical application, so that the calculation speed is increased.

Description

technical field [0001] The invention relates to a speed-up method for full-arrangement path planning aiming at multi-point homing of tourist attractions, and relates to the field of path planning. Background technique [0002] The full-arrangement path planning method is to find out the shortest path for traversing all passing points (tourist attractions) when the starting point and the end point are the same. However, in the case of a large number of passing points, this method of path planning requires a huge amount of calculation, and the computer calculation takes a long time. [0003] 1. Existing theoretical methods: [0004] 1. Full-arrangement path planning method: List all the arrangements of the sequence of all passing points, and calculate their distances. The arrangement with the shortest distance is the shortest path planning. [0005] 2. The neural network algorithm of artificial intelligence: after people find the shortest path among multiple points, and then...

AI Technical Summary

Problems solved by technology

[0018] (1) Neural network algorithm of artificial intelligence: artificial intelligence neural network is suitable for fuzzy calculation, but path planning is a kind of precise calculation, so it is not suitable to use artificial intelligence neural network to solve the path planning problem

And the artificial intelligence neural network is to simulate the general ability of human beings. It has advantages in local optimal calculation, but has no advantage in global optimal calculation. However, path planning must consider both local optimal and global optimal

Moreover, the neural network of artificial intelligence needs to be trained by humans. It is necessary to train the neural network when humans can calculate the results. Calculate the results, let alone train the neural network by humans

[0019] (2) Ant colony algorithm, fish swarm algorithm, etc.: This kind of biological simulation algorithm has the same disadvantage as the artificial intelligence neural network-local optimal algorithm. Secondly, this algorithm is random and cannot guarantee to find the shortest path every time. , and simulating a large number of creatures to find paths requires a lot of computer resources. Once the search fails, it needs to be recalculated, so the calculation speed is also very slow

Other popular algorithms can only find a more reasonable path when the number of points is large, and it is difficult to guarantee that the shortest path can be found

[0022] 2. Existing similar software After investigation, so far, there is no travel route planning software that can dynamically calculate and traverse all passing points. In addition to travel, well-known route planning platforms include Baidu Maps, Gaode Maps, and Didi Express , Meituan Waimai, etc., none of them have realized the path planning method of multi-passage points, but it is easy to cause misunderstandings

[0024] 2. Didi Kuaiche’s ride-hailing function only provides the driver with a choice of points near the main route, and the driver decides the order of the points. The APP only provides the shortest path between every two points, but does not provide the shortest path for all points.

[0025] 3. For Meituan Takeaway, the rider’s navigation software only uses the shortest path planning algorithm between two points continuously, and does not plan the shortest path for all points at once

[0028] In addition, we have done experiments on the full-arrangement path planning program. The program written in the c# programming language, in the windows7 operating system, the hardware environment is that the CPU is 1.9G, the memory is 12G, and 10 points need to be calculated 3628800 times , it takes about an hour. If it is changed to 20 points, it needs to be calculated 2432902008176640000 times, and it takes 3 days to calculate it.

Images