Path Planning and Control of Vehicles in 3D Environment using Unstable Modes

摘要

The article addresses the problem of controlling a vehicle in an undetermined 3D environment with mobile and stationary obstacles. We propose a two-level system performing planning and path-following. The path planning problem is split into the problem of planning and following global paths-missions and the problem of local planning in an obstructed area. The global path is built based on a priory information. A local path corrects the global one if the system detects obstacles that are not presented on the map. At first, the system is proposed that follows the global path at the controller level. Secondly, the hybrid system is introduced. It avoids obstacles using unstable modes at the lower level together with an intelligent algorithm determining the desired direction of detour at the top level. The techniques of introducing unstable modes are analyzed. Numerical modeling results are presented using the example of a hexacopter control system.