Method and equipment for weighted iterative path planning of unmanned boat based on multi-tree rrt

Abstract

The invention belongs to the field of unmanned boats, and discloses a multi-tree RRT-based weighted iterative path planning method and equipment for unmanned boats. The method includes: 1. Constructing the first RRT random tree from the starting point to the target point; 2. Judging whether a path from the starting point to the target point is found, if yes, set it as the first path, otherwise re-plan the first random tree 3. Keep and record the node NodeList of the previous random tree as a guide node to build a new RRT random tree; 4. Determine whether the planning is successful, and if successful, continue to step 5; if unsuccessful, output the last planned path as the final optimized Path; 5. Calculate the weighted cost of the planned path. If it is higher than the preset expected cost, return to step 3; if it is not higher than the preset expected cost or reach the preset maximum number of iterations, then directly output that the current path is the final optimized path. The invention can quickly obtain the optimal solution or suboptimal solution, and the randomness and convergence are easy to control.

Description

technical field [0001] The invention belongs to the field of unmanned boat motion planning, and more specifically relates to a multi-tree RRT-based weighted iterative path planning method and equipment for unmanned boats. Background technique [0002] The path planning of an unmanned vehicle is an important guarantee for its autonomous movement. There are also many mature algorithms for this. Among them, the Rapidly-exploring Random Trees (RRT) algorithm has been widely developed and applied in the past ten years. The sampling-based motion planning algorithm has many advantages such as strong adaptability, complete probability, and fast generation speed. According to the sampling idea of ​​RRT, connecting a series of randomly sampled points from an unobstructed space is suitable for solving the path planning problem of multi-degree-of-freedom unmanned vehicles in complex and dynamic environments, so as to establish a path from the initial state to the target state. path of....

AI Technical Summary

Problems solved by technology

[0004] (1) RRT needs to randomly explore the entire space, the scope of random exploration is large, and the amount of calculation increases. The improved RRT, such as probability-based target preference expansion, changes with probability, and it is difficult to grasp the relationship between randomness and convergence;

[0005] (2) The RRT* algorithm can obtain a suboptimal solution close to the optimal one. Since it is also a global search extension, its optimization degree is at the cost of calculation and time, and the operating efficiency is low.

Images