Multi-sensor fusion for stable autonomous flight in indoor and outdoor environments on a rotary airfoil (MAV)

摘要

The subject matter described here is different and integrates noisy measurements from multiple dissimilar sensors that generate absolute or relative observations at varying time intervals, in real time for autonomous flight, smooth Includes a modular and extensible approach that provides globally consistent assumptions. We have a new 1.9kg MAV platform with IMU, laser scanner, stereo camera, pressure altimeter, magnetometer and GPS receiver that performs state assumption and control on an onboard Intel NUC 3rd generation i3 processor Describe the development of algorithms and software architecture. We have our framework in a large-scale indoor-outdoor autonomous flight navigation experiment involving 440 meters of flight while entering and exiting an existing building under conditions of an average speed of 1.5 m / s and a wind speed of about 10 mph. To explain the stability.