UAS Flight Path Planning and Collision Avoidance Based on Markov Decision Process

The growing interest and trend for deploying unmanned aircraft systems (UAS) in civil applications require robust traffic management approaches that can safely integrate the unmanned platforms into the airspace. Although there have been significant advances in autonomous navigation, especially in the ground vehicles domain, there are still challenges to address for navigation in a dynamic 3D environment that airspace presents. An integrated approach that facilitates semi-autonomous operations in dynamic environments and also allows for operators to stay in the loop for intervention may provide a workable and practical solution for safe UAS integration in the airspace. This thesis research proposes a new path planning method for UAS flying in a dynamic 3D environment shared by multiple aerial vehicles posing potential conflict risks. This capability is referred to as de-confliction in drone traffic management. It primarily targets applications such as UAM [1] where multiple flying manned and/or unmanned aircraft may be present. A new multi-staged algorithm is designed that combines AFP method and Harmonic functions with AKF and MDP for dynamic path planning. It starts with the prediction of aircraft traffic density in the area and then generates the UAS flight path in a way to minimize the risk of encounters and potential conflicts. Hardware-in-the-loop simulations of the algorithm in various scenarios are presented, with an RGB-D camera and Pixhawk Autopilot to track the target. Numerical simulations show satisfactory results in various scenarios for path planning that considerably reduces the risk of conflict with other static and dynamic obstacles. A comparison with the potential field is provided that illustrates the robust and fast of the MDP algorithm