Dynamic collision prevention method for unmanned surface vehicle

Abstract

The invention discloses a dynamic collision prevention method for an unmanned surface vehicle based on a sparse iteration potential field algorithm. The method comprises the steps of firstly buildinga safe distance model and a potential function (a function of distances with a start and an end), then finding spare points by using a binary search thought and a sparse constraint method, and thus planning the shortest safe path at any angle from a grid map. Then aiming at the kinetic characteristic of the unmanned surface vehicle and the characteristic that the convention on the international regulations for preventing collisions needs to be conformed to during an unmanned surface vehicle planning process, maritime rules and kinematical constraints are considered while the unmanned surface vehicle avoids collision, so that the optimal path with quantized safe distance can be planed in real time for the unmanned surface vehicle. The dynamic collision prevention method provided by the invention can be applied to the field of unmanned surface vehicle path planning, and can quickly and dynamically prevent collisions in real time.

Description

technical field [0001] The invention relates to the field of unmanned boat path planning, in particular to a dynamic collision avoidance method for an unmanned boat. Background technique [0002] With the rapid development of unmanned systems and artificial intelligence technology, following drones and unmanned vehicles, unmanned surface vehicles (Unmanned Surface Vehicle, USV) are becoming more and more popular in the military and civilian fields. It plays an important role in various marine applications such as forensics, environmental monitoring and enemy detection. [0003] Path planning technology is one of the core technologies in the field of surface unmanned boats. It is a key step from manned to unmanned. It is the key development direction of intelligent unmanned boats. high and low. Therefore, the research on the path planning technology of UAV will help the further development of UAV. The path planning of the surface unmanned vehicle refers to how to find an a...

AI Technical Summary

Problems solved by technology

[0004] Most of the current path planning methods, such as the grid method, A* algorithm, dijkstra algorithm, etc., due to the defects caused by the discretization of the grid method, the above graph search algorithm cannot plan a path at any angle, which limits the path direction. The change can only be a multiple of π / 4, which will cause unnecessary movement and steering of the unmanned boat

[0005] However, the traditional artificial potential field method has the problem of local minima, which easily leads to path oscillation, the path cannot be identified in the group of similar obstacles, and the target is unreachable near the obstacle.

Images