A Path Planning Algorithm for Extended Duration Missions with Mobile Robots Open Access
Downloadable ContentDownload PDF
This thesis presents Energy Aware Path Planning (EAPP), a path planning algorithm for mobile robots deployed in extended missions. Battery constraints are often the limiting factor for the range and length of deployment in ground vehicles. In order to extend the lifetime of a mobile robot, solar panels or other energy transducers can be attached to the robot to harvest energy from the environment. The environment may not be well defined; this planning algorithm assumes the availability of satellite images which can be segmented into grids, but doesn't require explicit information about obstacles. EAPP defines a novel description of the environment consisting of two costs associated with each grid cell: an energy-based cost of traverse and a collision-based probability of traverse calculated from the model of the robot and the environment. The proposed algorithm takes a dynamic programming approach to solving the multi-objective optimization problem of finding the least-energy path through the environment with the highest probability of traverse. The global algorithm is combined with a local potential field method and sliding mode controller to present a complete navigation solution. Simulations show that the algorithm can successfully plan a minimum energy path through multiple environments while maximizing the probability of success.