Planejamento de movimento para robôs móveis baseado na condição de horizonte continuado (CHC).

Abstract

Mobile robot navigation is the research field responsible for the investigation of the different approaches that may be used to define the movement actions of a robot in a certain environment. When optimal behaviors are needed, a deliberative reasoning must be performed to define the best possible action before its execution. Considering the mobile robot navigation field, the deliberative reasoning is usually performed by a motion planning strategy. In this sense, this work presents a path planner based on harmonic functions and on the contour definition called Continued Horizon Condition (CHC), proposed in this work. A broad bibliographic review about robot navigation and motion planning is presented, followed by a more specific one about harmonic function based planning strategies, main issue of this work. The proposed harmonic path planning strategy is inspired in the temperature propagation in a thermal conductive medium. Its main issue is to suppress the potencial distortion caused by the classical boundary conditions, usually applied to model the obstacles and the external limits of the environment map. Thereby, the thermal distribution in the conductive medium depends only on the temperature source of the goal, which results in the shortest path to the destination point. Several simulation results are presented to show the features of the proposed planning technique, comparing it with the classical harmonic path planning results. Additionally, the use of the proposed planner based on the CHC contour condition is considered to solve some extensions of the basic path planning problem thus showing its applicability.