Simplified generalized Voronoi diagram-based robot autonomous exploration method

Abstract

The invention belongs to automation technologies, and specifically relates to a simplified generalized Voronoi diagram-based robot autonomous exploration method. The method includes the following steps: constructing a simplified generalized Voronoi topology map based on a morphological method, finding an optimal cutting-edge point, planning a global path from a current position of a robot to the optimal cutting-edge point, and navigating the robot to the optimal cutting-edge point along the global path. The method makes full use of a feature of the simplified generalized Voronoi diagram, applies the simplified generalized Voronoi diagram to an indoor autonomous exploration task of the mobile robot, and converts an autonomous exploration problem from two-dimensional plane space to topologymap space, thereby greatly reducing a quantity of cutting-edge candidate points, reducing a calculation amount of path planning, and in addition, ensuring that the generated global path is an optimalcollision-free path. Compared with another method, the invention can faster guide the indoor robot to autonomous exploration, the simplified generalized Voronoi diagram is more flexible in a specificapplication, and therefore, a requirement for the autonomous exploration task of the mobile robot can be better met.

Description

technical field [0001] The invention belongs to the technical field of automation, in particular to a robot autonomous exploration method based on a simplified generalized Voronoi diagram. Background technique [0002] Due to the complexity and structural characteristics of indoor spaces, mobile robots can be widely used in tasks related to architectural surveying (such as indoor mapping and mapping and indoor search and rescue), so autonomous exploration and movement in unknown spaces is a research hotspot in robotics. one. Autonomous exploration technology can guide the robot to move in an unknown environment and use various sensors to perceive unknown things in the environment, so as to complete a given specific task. According to the existing research, the autonomous exploration technology of mobile robots mainly has difficulties in the following four aspects: (I) how to model the environment; (II) how to decide where the robot should go; (III) how to find connections i...

AI Technical Summary

Problems solved by technology

First, the number of candidate frontier points generated is large, and there is considerable redundancy, and it will waste a lot of computing time to calculate and compare their features; second, the path finding algorithm on the grid map is difficult to meet the real-time requirements, and Paths are often close to walls, creating a collision risk for mobile robots

Some scholars propose to use topological maps to directly model the environment, which can effectively reduce the number of candidate frontier points and speed up path planning. However, the environmental information carried by topological maps is very scarce, and it is difficult to effectively calculate and compare the characteristics of candidate frontier points.

Images