Trajectory planning for mobile robots in a dynamic environment

Abstract

An algorithm is presented for path planning in a dynamic environment where not only obstacles are moving also the goal of the mobile robot. The basic idea of this algorithm is to introduce the Euclidean distance as a value for each cell in the discretized workspace. The moving obstacles are represented by a time-varying set of cells, whose values are multiplied by a factor greater than unity representing the virtual distance. Collision-free trajectory is calculated by finding the lowest value of the eight neighbor cells. This algorithm can easily work when the mobile robot objective is moving, even in the presence of a complex environment, and can also be extended to path planning in three dimensions. The usefulness of this algorithm is shown by simulation.