100 results about "Travelling salesman problem" patented technology

The invention provides an unmanned aerial vehicle (UAV) flight path generation method and device, a storage medium and equipment. The method includes the steps of generating at least one ordered key angle of view subset based on a key angle of view set of landmarks; generating local flight paths of the landmarks using the ordered key angle of view subset; constructing and solving a generalized travel salesman problem based on the local flight paths to obtain at least one global flight path accessing at least one landmark; and generating an overall cost function based on a cost function of eachlocal flight path and a cost function of the global flight path, and selecting the local flight path and the global flight path to generate a UAV flight path according to the solving result of the overall cost function. According to the invention, to specify the order of key angles of view is unnecessary, and the difficulty of UAV control is reduced.

The invention discloses a method for planning logistics paths on the basis of bisectors of store point groups. The method includes steps of firstly, generating a clockwise initial path and an anticlockwise initial path; secondly, adding all passing points into the two initial paths to form a clockwise distribution path and an anticlockwise distribution path; thirdly, comprehensively considering unloading quantities and the distances of the paths and determining ultimate distribution paths. The method has the advantages that the angle bisectors of passing point groups are generated at first, the deterministic clockwise initial path and the deterministic anticlockwise initial path are obtained, then the passing points are added into the initial distribution paths according to given sequences and modes to obtain the complete distribution paths, and the unloading quantities and the distances of the paths are integrated with one another, so that the ultimate distribution paths can be determined; logistics distribution knowledge is effectively utilized as compared with ordinary optimal search algorithms, space distribution characteristics of stores are sufficiently utilized, accordingly, traveling salesman problem solving can be simplified, complicated search and comparative computation of intelligent optimization algorithms can be omitted, and the method is high in computation efficiency and good in stability.

Provided in the invention is a multi-robot path planning method based on an ant colony algorithm. The method comprises: calculating distances between all tasks according to position coordinates of different tasks; arranging M robots at positions of first M tasks and setting the first M tasks in a taboo table to indicate execution; calculating a transition probability of an ith robot from an ith task to a jth task not in the taboo table; selecting a largest probability value based on the transition probability and setting one corresponding task in the taboo table; and under the condition of allocating all tasks and adding a path, calculating the total length of the path and updating pheromone concentrations on paths between tasks based on the total length. According to the invention, on thebasis of the ant colony algorithm of transforming a multi-traveling-salesman problem into a traveling salesman problem without node adding, the effect of searching for a shortest path of multiple robots or a shortest path of the multi-traveling-salesman problem is good. The stability and optimized precision are high in path planning on a data set.