Full-coverage confidence function path planning algorithm for mobile robots
Junliang Cui, Jiahao Yang · 2023
We propose a full-coverage path planning algorithm for mobile robots based on a raster confidence function. The algorithm aims to address the problem of guiding a mobile robot to traverse all reachable points in the working area while ensuring automatic obstacle avoidance. To achieve this, we assign a raster map based on environmental information and utilize different function values to represent obstacles, covered raster, and uncovered raster areas. This algorithm provides an effective solution for achieving full coverage in path planning for mobile robots, ensuring efficient exploration of the working area while avoiding obstacles; second, different directional confidence functions are introduced to determine whether the robot is caught in a dead zone, and the raster function values are adjusted. Finally, the robot generates the coverage path using the raster confidence function values. This algorithm not only enables the mobile robot to achieve complete coverage of the working area but also allows it to efficiently escape from dead zones and minimize path repetition. In the simulation experiments, it is demonstrated that the algorithm mentioned in this paper has higher coverage efficiency by comparing it with the bio-inspired neural network algorithm.