Unmanned Aerial Vehicles (UAVs) are increasingly gain interest in many real-life applications ranging from agricultural farming to search&rescue missions or civil construction inspections. The movement of a UAV in an autonomous manner requires advanced processing capabilities from various sensors (3D cameras, lidards, etc.) and adaptive route decision including collision-avoidance (with moving or fixed objects), and proximity estimation. We propose to develop innovative algorithms for 3D path plannig for autonomous and adaptive UAV missions by controlling autopilots (e.g. ArduPilot or PX4) using MAVROS.