This paper presents a path planning approach for Micro Aerial Vehicles (MAVs) in urban environments. We show that it is feasible to extend the endurance and range of an MAV by exploiting the orographic lifts from buildings in windy conditions. A sampling based rapidlyexploring random tree algorithm is implemented to generate feasible paths between the start and goal in a modelled urban environment. A cost function is proposed that leverages sampling based planner metric sensitivity to properly capture the aircraft's dynamics and utilize the environment for soaring. A three-degree of freedom aircraft model with a high fidelity control space is used to generate realistic reference trajectories. Multiple scenarios are setup to assess the behaviour of the planner and the effectiveness of the proposed metrics. The results obtained from the presented experiments validate the potential of utilising orographic lift for improving MAV endurance. This work also highlights the sensitive nature of sampling based planners to implementation parameters and emphasises the need for an adaptive parameter scheme.
↧