Comparisons of RRT and MCTS for Safe Assured Path Planning in Urban Air Mobility

Urban air mobility (UAM) using electrical vertical take-off and landing (eVTOL) aircraft is an emerging way of air transportation within metropolitan areas. However, the ability to navigate eVTOL aircraft through congested urban air environment poses a great challenge for path planning. For the successful operations of autonomous aerial vehicles in UAM, it is of great importance to take the collision-free requirement into consideration. It is also necessary to consider the presence of various forms of uncertainty in path planning. To address these issues, in this paper the path planning problem is first formulated as a Markov Decision Process (MDP) and then solved by an online algorithm of Monte Carlo Tree Search (MCTS) incorporating chance constraints of uncertainty. For the sake of illustration, a high-density free flight airspace simulator is created to test the performance of this proposed algorithm. Numerical simulation results demonstrate that this proposed algorithm outperforms the Chance Constrained Rapidly-exploring Random Tree (CCRRT) Algorithm, a path planning algorithm we proposed before, in terms of many aspects.