Real-time path planning for Aerial robot in real environments

摘要

The problem of real-time path planning in actual environments is challenging. This paper presents a realtime planning algorithm using image processing techniques for navigating a Aerial robot in known irregular environments. It includes two steps. The proposed strategy, based on the generation of a novel search space, finds a suboptimal rough path as an initial condition by constructing and searching a graph in the first step. In the second step, the algorithm takes further advantage of techniques like Snake to optimize the initial path within the dynamic constraint of robot. Results from simulation show that the path planner was able to plan optimal path quickly and efficiently due to the simplicity of the search space.